CLIMATE GOVERNANCE PAPERS Adapting Fiscal Decentralization Design to Combat Climate Change Jorge Martinez-Vazquez Adapting Fiscal Decentralization Design to Combat Climate Change © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO), http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Jorge Martinez-Vazquez. 2021. Adapting Fiscal Decentralization Design to Combat Climate Change. Climate Governance Papers Series. Washington, DC: World Bank. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third- party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Graphic Designer: Maria Lopez / lopez.ten@gmail.com Abstract There are still many countries around the world that have not effectively engaged their subnational governments in their climate change strategies and policy frameworks. Where subnational levels are involved, generally they still play a relatively small role. This paper examines how the principles of fiscal decentralization design (in expenditure and revenue assignments, transfers, and borrowing) can be adapted to successfully engage subnational governments in fighting climate change. In addition, the paper critically reviews ongoing international practices, both effective and less effective, that involve subnational governments in climate change mitigation and adaptation. Shared responsibility for policy and program design and implementation, fee- or charge-funded adaptation activities, objective-targeted intergovernmental transfers, and the use of green bonds are some of the most promising approaches analyzed. Clearly, there is ample space ahead for the further involvement of subnational governments across the world in combating climate change. Adapting Fiscal Decentralization Design to Combat Climate Change 3 Contents 6 List of Figures 6 List of Tables 7 List of Abbreviations Acknowledgements 10 1. Introduction 11 2. Climate Change and the Need for Government Action: The Stylized Facts Relevant to Fiscal Decentralization Design 13 Measuring the Current Involvement of Subnational Governments in Climate Action 18 3. Reassessing the Assignment of Functional Expenditure Responsibilities for More Effective Action on Climate Change 23 Basic Principles and Most Common Problems with Expenditure Assignments 24 Applying Basic Principles and Best Practices to the Assignment of Responsibilities for Climate Change Policy Action 26 Other Potential Obstacles to the Devolution of Climate Change Responsibilities to Subnational Governments: Political Economy Considerations, Free Riding, and 29 Common Pool Problems What Are Current Effective and Less Effective Practices in Expenditure Assignments in Different Countries? 29 4. Improving the Assignment of Revenue and Tax Authority for Enhancing Action on Climate Change 39 Basic Principles of Revenue Assignments 40 Applying Basic Principles and Best Practices of Revenue Assignments for Climate Change Policy Action 41 What Are Current Effective and Less Effective Practices in Revenue Assignments in Different Countries? 44 5. Redesign and Innovations in Transfer Systems to Combat Climate Change 47 Basic Principles of Transfer Design 47 Applying Basic Principles and Best Practices of Transfer Design for Climate Change Policy Action 50 What Are Current Effective and Less Effective Practices in Transfer Design in Different Countries? 52 6. Adapting Subnational Borrowing and Other Fiscal Rules to Facilitate Effective Policies for Climate Change 55 Basic Principles of Subnational Borrowing and Other Fiscal Rules 55 Applying Basic Principles and Best Practices in Borrowing and Rules for Climate Change Policy Action 56 What Are Current Effective and Less Effective Practices in Borrowing Design in Different Countries? 57 7. Summary and Conclusions 59 References 64 Figures Figure 1. Sources of Emissions 14 Figure 2. Current Polluting-Fuels Taxes versus Ideal Level 45 Figure 3. Total Sovereign Green Bond Issuance in Millions of Dollars USD 58 Tables Table 1. SNG Spending on Environmental Protection: Breakdown by Region/Country 19 Table 2. Example Territorial Climate-Energy Plans (TCEP) 32 Adapting Fiscal Decentralization Design to Combat Climate Change 6 List of Abbrevations 14th FC 14th Finance Commission ADEME Agency of the Environment and Energy Management (L'Agence de l'Environnement et de la Maîtrise de l'Energie) AFLRA Association of Finnish Local and Regional Authorities BANS New Style Intergovernmental Agreement (Bestuursakkoord Nieuwe Stijl) BAU Business as Usual BNDES Brazilian Development Bank (Banco Nacional de Desenvolvimento Econômico e Social) BNPB National Disaster Management Authority (Badan Nasional Penanggulangan Bencana) C2ES Center for Climate and Energy Solutions CAR Regional Autonomous Corporation (Corporación Autónoma Regional) CBI Climate Board Initiative CCD Climate Change Directorate CCP Cities for Climate Protection CCS Carbon Capture and Sequestration/Carbon Capture and Storage CDR Regional Development Councils (Consejos de Desarrollo Regional) CDS Corporations for Sustainable Development CESA Clean Energy States Alliance CIDPs County Integrated Development Plans CO2 Carbon Dioxide COP15 15th Conference of the Parties to the United Nations Framework Convention on Climate Change in Copenhagen COVID / Coronavirus disease 2019 (and the associated pandemic) COVID-19 EBRD European Bank for Reconstruction and Development EGCA European Green Capital Award EFTs Ecological Fiscal Transfers EIs Economic Instruments EIAs Environmental Impact Assessments Adapting Fiscal Decentralization Design to Combat Climate Change 7 EPA Environmental Protection Agency ETS Emission Trading System EU European Union FEMA Federal Emergency Management Agency FONDEN Natural Disasters Fund (El Fondo de Desastres Naturales) GDP Gross Domestic Product GHG Greenhouse Gases GIZ German Corporation for International Cooperation (Deutsche Gesellschaft für Internationale Zusammenarbeit) GNCC National Climate Change Cabinet (Gabinete Nacional de Cambio Climático) GST Goods and Services Tax HDI Human Development Index Hinku Towards Carbon Neutral Municipalities (Kohti Hiilineutraalia Kuntaa) ICLEI Local Governments for Sustainability (International Council for Local Environ- mental Initiatives) ICMS Tax on the Circulation of Goods and Services (Imposto sobre Circulação de Mercadorias e Serviços) IGA-FFR Intergovernmental Agreement on Federal Financial Relations IMCE Inter-Ministerial Conference on the Environment INDC Intended Nationally Determined Contribution IPCC Intergovernmental Panel on Climate Change LFL Local Finances Law LGFAs Local Government Funding Agencies MAPTAM Modernization of Territorial Public Action and Assertion of Metropolitan Areas (Modernisation de l'Action Publique Territoriale et d'Affirmation des Métropoles) NAFCC National Adaptation Fund on Climate Change NAP-ETH National Adaptation Strategy of Ethiopia NAPCC National Action Plan on Climate Change NCCAP National Climate Change Action Plan NCCC National Climate Change Council NDCs Nationally Determined Contributions NDRC National Development and Reform Commission NOTRe New Organization of the Republic (Nouvelle Organisation Territoriale de la République) OECD Organisation for Economic Co-operation and Development OEFA Agency for Environmental Assessment and. Enforcement (Organismo de Evaluación y Fiscalización Ambiental) PNACC National Climate Change Policy (Plan Nacional de Adaptación al Cambio Climático) REDD+ Reducing Emissions from Deforestation and Forest Degradation Adapting Fiscal Decentralization Design to Combat Climate Change 8 RGGI Regional Greenhouse Gas Initiative SAPCCs State Action Plan on Climate Change SEPA State Environment Protection Agency SERNA Natural Resources and Environment Secretariat (Secretaria de Recursos Natu- rales y Ambiente) SISCLIMA National System of Climate Change (Sistema Nacional de Cambio Climático) SNG Subnational Government SNG-WOFI World Observatory of Subnational Government Finance and Investment TCEP Territorial Climate-Energy Plans UCLG United Cities and Local Governments UCM U.S. Conference of Mayors UNCDF United Nations Capital Development Fund UNDP United Nations Development Programme UNEP United Nations Environment Programme UNFCCC United Nations Framework Convention on Climate Change URBACT European Exchange and Learning Programme Promoting Sustainable Urban Development US United States USAID United States Agency for International Development VAT Value Added Tax Adapting Fiscal Decentralization Design to Combat Climate Change 9 Acknowledgements This paper was prepared under the auspices of the World Bank Mainstreaming Climate Change in Governance Program. I am thankful to Nicholas Menzies and Serdar Yilmaz for multiple discussions and to Adrian Fozzard, Stephane Hallegatte, James Newman, and Roland White for additional very useful inputs. I am also indebted to Abdulaziz Almuzaini, Amakoe Alognon, Yasin Civelek, Kshitiz Shrestha, and Chris Thayer for excellent research assistance and inputs.  Adapting Fiscal Decentralization Design to Combat Climate Change 10 01 Introduction There is little doubt that one of the most preeminent issues facing policy makers around the world today is that of taking action to address the causes and consequences of climate change.1 At the same time, in many countries, one of the most salient institutional transformations over the past several decades in how governments make decisions has been the devolution and decentralization of fiscal, administrative, and political power to subnational governments (SNGs). Despite their individual prominence, these two thrusts of policy reforms have, with a few notable exceptions, largely ignored each other. Not so, however, with research. As seen throughout this paper, there have been numerous academic contributions to the design of environmental policy in fiscal decentralization settings, especially environmental research, and notable, though fewer, contributions from the fiscal decentralization literature. Yet, even with the important role of SNGs in public expenditures—in some countries hovering around two-thirds of the total aggregate expenditure—and the importance and urgency of climate change issues, there has been little discussion of how intergovernmental fiscal design should adapt to this new reality.2 In addition, there has been little explicit work in the literature on intergovernmental relations to inform the appropriate design of and balance between regulations, fiscal transfers, and devolved tax powers as incentives for climate action and tools to address moral hazard or incentives to assume additional risks because the full costs of those actions are borne elsewhere. 1. The COVID-19 pandemic has reset governments’ priorities, at least in the shorter run. The hope is that the longer-term issue of climate change will not lose political support and funding, both of which can be better assured when legislation has been passed demanding action. An extensively widespread shared hope is that stimulus packages in response to the COVID crisis will provide an opportunity for heightening aggressive decarbonization and mitigation policies worldwide. 2. Some notable exceptions, which will be further discussed below, include UNCDF, UNDP, and UNEP (2013). Adapting Fiscal Decentralization Design to Combat Climate Change 11 The main goal of this paper is to attempt to integrate the The third aim of this paper is to provide in-depth need for climate change policy action into the existing recommendations and guidance that will inform the best practice framework of intergovernmental fiscal World Bank’s work, as well as that of other organizations, design. It does so by bringing closer together climate on the design of intergovernmental fiscal relations in all change policy and intergovernmental design, focusing its dimensions—functional assignments, devolved tax on how the four pillars of fiscal decentralization—the powers, the design of transfer regimes, and borrowing— assignment of functional expenditure responsibilities, to strengthen SNGs’ capabilities to take climate change the assignments of taxes and other revenue sources, action on their own and to enable them to collaborate the design of fiscal transfers, and subnational credit and more effectively with central authorities on these matters.4 borrowing—could be adapted for fiscal federalism and decentralized fiscal structures to become more effectively The remainder of this paper is organized as follows. supportive of climate change action.3 In section two, it outlines the stylized facts defining the climate change issues that are most salient to the design of Beyond the conceptual redesign of the intergovernmental intergovernmental fiscal relations. Section three explores fiscal architecture, a second objective of this paper is how the assignment of expenditure responsibilities to review and highlight how different countries, both among different levels of government must be adapted federations and other fiscally decentralized states, have to the most effective multilevel government response to been responding to the prospect of implementing effective climate challenges. Section four covers the same issue in policies to address the challenges of climate change. The relation to the assignment of taxing powers; for example, important question is whether decentralized governance if a carbon tax is one of the good responses to climate has been a way to leverage the effectiveness of the change, what level of government should implement climate change policies of central governments, or if it has that tax, or should it be shared? Section five explores hindered (or been neutral to) their implementation. This the role of the different types of transfers, subventions, can be answered only by identifying both effective and and agreements federal and central authorities can take less-effective international practices in the assignment to stimulate action by SNGs against climate change. of functions to fight climate change, approaches to tax Section six analyzes whether the need for climate duties and the design of intergovernmental transfers change action at the subnational level justifies redrafting to enable climate change action, and the redesign and relaxing borrowing and other fiscal rules that many and adaptation of borrowing and other fiscal rules to countries have introduced in recent years with the facilitate SNG engagement in climate decarbonization objective of ensuring overall fiscal sustainability. Finally, and adaptation. section seven summarizes the most important findings and options for reform and presents a conclusion. 3. It is well accepted that for overall decentralization to work, fiscal decentralization needs to be accompanied by administrative and political decentralization. Although it is likely relevant for adapting country policies to combat climate change, the redesign of administrative and political decentralization falls outside the scope of this paper. See Smoke and Cook (2021) for an analysis of adapting administrative decentralization to combat climate change. 4. A separate guidance note is being prepared on administrative issues. Adapting Fiscal Decentralization Design to Combat Climate Change 12 02 Climate Change and the Need The consensus in the scientific literature is that the stock of greenhouse gases (GHGs) in the earth’s atmosphere for Government Action: is rising, trapping heat and causing global warming and other changes in the climate. Climate change, in turn, The Stylized Facts Relevant to affects nations and individuals directly through violent Fiscal Decentralization Design storms, floods, droughts, rising sea levels, melting ice caps, and other severe changes (Stern 2008) and indirectly through a number of consequences from these disruptions. The economic impact of climate change has been estimated to be quite critical, but there are differences as to how it should be aggregated. The Stern Review (Stern 2006) estimates that under “business as usual (BAU),” global GDP would fall at least by 5 percent every year. Similarly, the cost of extreme weather events could reach 0.5–1 percent of world GDP by the mid-century. However, not everyone agrees with the dire predictions of the Stern Review. For example, some researchers have qualms about the low discount rate it uses (Weitzman 2007; Nordhaus 2007). Others have emphasized the significant and uneven distributional impacts of climate change, with effects that depend on the location of countries, some of which could actually experience increased income or reduced expenditures, thus offsetting the losses in income and increased related expenditures in other countries (Tol 2005).5 For example, Leppanen, Solanko, and Kosonen (2017) find that higher temperatures from climate change have led to lower regional government expenditures in Russia, although the positive effects have been smaller in warmer regions. 5. There is also uncertainty regarding the science behind the economic models of climate change. In a recent paper, Dietz et al. (2020) claim that most economic models of climate change assume a much too lengthy delay of the impact of CO2 emissions on global warming. These models also ignore the decreasing returns from policy actions as the stock of emissions and temperatures grow. Adapting Fiscal Decentralization Design to Combat Climate Change 13 More recently, the World Bank Outlook 2050 (Mukhi et al. (responsible for 14 percent of emissions) and buildings 2020) highlights the further critical implications of climate (8 percent) typically tend to be regulated and taxed by change on poverty and other development challenges, regional and local governments, although of course, such as health and food security. For example, there are differences across decentralized systems. conditions related to climate change have led to the Also, in terms of non-energy emissions, land use (18 recent desert locust swarms in the Horn of Africa, the percent of emissions) and waste (3 percent) tend to a Arabian Peninsula, and southwest Asia (UNEP 2020). large extent to be regulated or influenced in different As shown in figure 1, reproduced from the Stern (2006) ways by SNGs, although again, there are differences report, GHG emissions come from a variety of production across countries.6 Other significant sources of GHGs, and consumption activities involving both direct energy such as power generation, industry, or agriculture, tend emissions (e.g., power generation) and non-energy to be more the purview of central government authorities, emissions (agricultural activities). The power of figure 1 is although in some decentralized countries these activities that it can help to relate those sources to the conventional may also be regulated and taxed by SNGs. In sum, assignment of functional expenditure responsibilities in over 40 percent of GHG emissions may come from fiscally decentralized systems of government, especially activities over which SNGs typically exert regulatory regarding the decarbonization question. But the role of and taxing powers. This means prima facie that SNGs SNGs, as seen below, becomes even stronger in terms in decentralized countries have an important role to play of adaptation policies, generally because of the smaller in the implementation of policies to fight climate change, role played by externalities and economies of scale such as regulation, emission controls and energy- issues in this dimension of fighting climate change. In efficiency standards, promotion of alternative (“green”) terms of direct energy emissions, transport activities energy, carbon taxation, and so on.7 Figure 1. Sources of Emissions Power Industry 24% 14% Buildings Transport Energy Emissions Non-Energy Emissions 8% 14% Land Use 18% Waste 3% Other energy related 5% Agriculture 14% Total emissions in 2000: 42 GtCO2e Source: Stern (2006, 196). 6. SNGs also manage land-use zoning, which has consequences on climate change; for example, urban sprawl has contributed to emissions by increasing daily commute distances and reducing green spaces. 7. Note that there is no complete agreement that these conventional policies to combat climate change will be ultimately effective. An important skeptic, Sinn (2012) suggests that by regulating and reducing the demand for fossil fuels, suppliers of carbon resources will respond by trying to preempt the impact of future regulation via the acceleration of the production of fossil energy. This is what Sinn calls the “Green Paradox”: pushing for the future reduction of carbon consumption leads to the acceleration of climate change in the present. The alternative policy suggested by Sinn is supply-side policies based on inducing suppliers to leave more of the carbon resources underground by levying source taxes on capital income. There are also disagreements on what conventional policies are most effective in combating climate change. For example, McKibbin and Wilcoxen (2002) are highly critical of the Kyoto Protocol of 1997 as an agreement with no real chance of reducing GHG emissions because it focuses on rigid targets and timetables for emissions reductions instead of the supposedly more efficient and practical approaches that rely on emissions taxes and tradable permits. Adapting Fiscal Decentralization Design to Combat Climate Change 14 14 An alternate way to consider the potential role of SNGs is Thus, it appears that the most important consideration in offered by the World Bank “Decarbonizing Development” the question of what level of government should assume report (Fay et al. 2015). This report highlights that responsibility for the different sources of climate action is achieving zero net emissions by 2050 will require action how the presence of significant externalities are handled on four fronts:8 (i) the decarbonization of electricity in decentralized environments. As is discussed in more generation (i.e., generation using renewable, nuclear, and/ detail in the next section, the theory and principles of fiscal or carbon capture and sequestration [CCS] methods); (ii) federalism, enshrined in Oates’ (1972) “decentralization electrification (to increase reliance on clean electricity), theorem,” point out that where significant externalities and where that is not possible, a switch to cleaner fuels; exist, subnational decision making may be less optimal (iii) improved efficiency and reduced waste in all sectors, than national decisions, unless there is a great deal of including building, transport, and agriculture; and (iv) coordination and incentives for SNGs to consider the the preservation and increase of carbon sinks, such as externality effects.10 On the other hand, climate change forests and other vegetation and soils, to increase the policies in practice take many forms, and many of those removal of carbon from the atmosphere. The role that can activities may yet be implemented by SNGs. This is be played by SNGs in those four pillars is not as directly so because in climate change situations, significant obvious. Nevertheless, thinking of ways in which SNGs externalities are not always present. For example, SNGs can contribute policy measures according to those pillars could be better implementers of medium-term sectoral can be a useful way to operationalize the role of fiscal targets regarding building or public transit. In the next decentralization in fighting climate change. For example, sections below, with a more granular approach to the SNGs have increasingly been playing, as discussed relevant policy questions, it becomes clear that the below, roles in improving efficiency and reducing waste roles of SNGs in combating climate change are multiple in the building and transport sectors, as well as in and important. preserving forests and other means of removing carbon from the atmosphere. When significant externalities are present, even with coordination and incentives, the case for decentralized The recognized major advantage of the decentralized provision gets somewhat further complicated because the provision of climate-related functions, as well as other nature of the externalities generated from GHG emissions public goods, is the ability to cater to differences in tend to differ from other externalities in several relevant needs and preferences, which may promote much faster ways (Stern 2008). This means that: (i) they tend to be action in some places. It can also provide a laboratory for global in their origins and impacts; (ii) some of the effects incubating innovations and testing the effectiveness of are very long term and governed by a complex flow-stock different approaches that can be subsequently imitated process; (iii) there is a great deal of uncertainty in most in other places. The big disadvantage of decentralization steps of the scientific chains; and (iv) the effects tend is that it may be much less effective in dealing with to be quite substantial and may be irreversible. Given externalities or spillover effects across government the gravity of these special features and the fact that boundaries and in taking advantage of economies of expenditure assignments and taxing powers for some scales in the production and delivery of services. It is far of the key activities are indeed decentralized in many from clear which of these attributes, on balance, is more countries, intergovernmental coordination in any country dominant, but given the high importance of externalities also needs to be extended to intercountry or international in climate action (in comparison to many other public coordination. Therefore, the relevant questions are: will goods), there may be an initial presumption that decentralized governance institutions in any country be centralized provision would be preferable to decentralized able to coordinate internationally more effectively than provision when it comes to decarbonization policies, but centralized ones? Or what may be the most effective not necessarily so for adaptation interventions.9 division of responsibilities for domestic and international responses to the challenges posed by climate change? 8. These fronts are based on recent reports by the Intergovernmental Panel on Climate Change (IPCC), which analyze 30-plus climate change models based in turn on a variety of assumptions on decarbonization and mitigation activities and technological possibilities simulating a wide range of future scenarios. 9. For example, Banzhaf and Chupp (2012) simulate decentralized and centralized models of air pollution control for the U.S. electricity sector and show that a centralized policy may outperform a decentralized policy because interstate externalities may be more important than interstate heterogeneity. 10. For example, Hankla, Martinez-Vazquez, and Rodriguez (2019) argue for a “strong decentralization theorem”: They suggest that decentralized provision will still be efficient in the presence of externalities if fiscal decentralization is supported by political decentralization institutions, such as integrated national parties, where nominated electoral candidates need to internalize externality effects across jurisdictions. Adapting Fiscal Decentralization Design to Combat Climate Change 15 The issue of roles and responsibilities is a matter to be Does fiscal decentralization result in higher standards assessed and discussed in the following sections of this and lower levels of emissions or in lower standards and paper. However, it is clear and generally accepted that higher emissions? SNGs need to be involved in the two main categories of issues pertaining to climate change: (i) decarbonization, The theoretical models analyzing the effects of whereby efforts are made to reduce GHG emissions decentralized environmental policy making predict a through limiting energy consumption, such as in burning variety of often contradictory results, such as both a “race of fossil fuels, solid waste generation, and so on, and (ii) to the top” and a “race to the bottom” (Garcia-Valiñas adaptation, which involves coping with the consequences 2004). Several researchers emphasize the virtues of of climate change through the management of the decentralized governance and positive competition localized impact of natural disasters resulting from or among units to conclude that decentralization will yield heightened by climate change, such as hurricanes, sea- less pollution, unless local governments solely pursue level increases, droughts, flooding, bushfires, disease maximizing their tax base and revenues (Oates and outbreaks, and the like.11 Schwab 1988, 1991, 1996; List and Mason 2001). On the other hand, Markusen, Morey, and Olewiler (1993, 1995), In reality, SNGs are important for both mitigation and assuming firms operate under increasing returns to scale, adaptation policy actions. Indeed, in some countries (as find that competition will generally lead to higher levels of seen in the following sections), SNGs have taken the lead pollution. Fredriksson and Gaston (2000) conclude that in climate action when national governments have lagged both centralized and decentralized governments may in their policies. On the side of decarbonization actions, have similar effects, depending on whether firms are SNGs are commonly responsible for land-use planning able to form lobbying groups affecting policy decisions. and enforcement, water and power utilities, waste Empirically, the findings are more hopeful but still mixed disposal, transport systems, and urban management. and, not surprisingly, country-dependent. Therefore, On the side of adaptation actions, climate change implies no clear picture emerges with respect to the impact of that infrastructure investments on energy, transport, and decentralization on the level of emissions. water and sanitation systems will have to be built to be resilient to extreme events (Fay et al. 2017).12 13 In reality, More specifically, in the United States, a number of as seen in other sections of the paper, SNGs are on the empirical studies have examined whether a “race to the frontline of disaster response and charged with investing bottom” effect exists in setting environmental standards in climate adaptation. at the subnational level. Some studies do not find evidence of this but instead, in some cases, a “race to Before turning to examining what may be the best the top” with spatial correlation and mimicking (Dinan, design of the fiscal decentralization pillars for effective Cropper, and Portney 1999; List and Gerking 2000; decarbonization and adaptation policies, it is useful Fredriksson and Millimet 2002; Millimet and List 2003; to review the evidence so far on the impact of fiscal Fomby and Lin 2006). For example, Millimet (2003) decentralization on pollution and global warming and examined the impact of decentralized environmental whether fiscal decentralization tends to aggravate or policy making under President Ronald Reagan. In terms ameliorate the impact of natural disasters associated with of pollution abatement expenditures, he found that climate change. Does fiscal decentralization lead to more Reagan’s decentralization had no significant impact prior or less spending on decarbonization and adaptation? to the mid-1980s, but afterwards, the results support a 11. Even though they are different tools and most of the time are carried out by different agencies, mitigation and adaptation are policy substitutes both working to reduce the impacts of climate change. Adaptation and mitigation can therefore be considered simultaneously, allowing for the trade-offs between them in terms of budget and other dimensions (Tol 2005). 12. The impact of natural disasters can complicate SNGs’ ability to address climate adaptation needs. For example, Jerch, Kahn, and Lin (2020) report that since 1980, over 2,000 local governments in U.S. Atlantic and Gulf states have been hit by a hurricane, and that in the case of major hurricanes, local revenue losses of up to 7 percent persist at least 10 years after the hurricane event, leading to cuts in important public goods and services and a significant increase in the risk of default on municipal debt. Localities rated as riskier by ratings agencies face higher borrowing costs and therefore lower ability to invest in climate change adaptation. 13. The needs associated with adaptation policies can interact with other trends, such as urban sprawl and slum creation, leading to significant increases in related costs. For example, Fay et al. (2017) note that urban density has declined in many large cities in Latin America driven by transport, land use, and housing policies. If expansion patterns continue unchanged, built-up urban areas could double in the region by 2035, pushing up infrastructure and climate adaptation costs. Denser urbanization may reduce infrastructure investments and maintenance costs by half, while slum infrastructure upgrading can cost two to eight times more than regular land development. Adapting Fiscal Decentralization Design to Combat Climate Change 16 race to the top process. However, he finds no statistically Jr. (2000) shows that coordination problems arising from significant effects on actual emissions. However, decentralization cause fiscal deficits. Woods (2006) finds supportive evidence of a race to the bottom among U.S. states involving surface-mining Given the severity of air pollution in China, it is not regulations. In summary, a race to the bottom may not surprising that a good number of studies have examined always be expected, but neither can it be excluded from the relationship between fiscal decentralization and consideration as an outcome of any scenario. In this carbon emissions (and other forms of pollution) in that respect, Kunce and Shogren (2005) explore conditions country. The findings are mixed, with negative, positive, under which a race to the bottom may or may not happen and even neutral effects within the same overall country in the context of inter-jurisdictional competition and context. Cheng et al. (2020) find that the impact of fiscal environmental federalism. decentralization on CO2 emissions is nonlinear, and that higher per capita levels of decentralized expenditure are Several papers have also highlighted some drawbacks more effective in reducing carbon emissions, especially of fiscal decentralization when it comes to environmental when emissions reduction policies are conducted at the pollution, namely, the free riding of states and intermediate (province and city) levels of government. coordination failures. For instance, water pollution Ran, Zhang, Hao (2020) find a strong positive spatial is greater when the rivers cross subnational borders correlation in regional carbon emissions in China, and (Sigman 2004; Lipscomb and Mobarak 2017) and that the measure of environmental decentralization international borders (Sigman 2007). Moreover, SNGs actually promotes carbon emissions, suggesting that are inclined to spend less on the preservation of species China’s environmental decentralization system may that are directly in conflict with development (List, Bulte, not be conducive to carbon emission control. However, and Shogren 2002). When such governments have they also find that the decentralization of environmental a different level of environmental regulation, bodies supervision activities dampens emissions. Even with stricter environmental policies tend to strategically though they use similar data and approaches, Zhang locate the polluting firms near the boundaries (Monogan, et al. (2019) are far less conclusive, finding that there Konisky, and Woods 2017; Helland and Whitford 2003).14 is considerable heterogeneity in the impact of fiscal De Mello Jr. (2000) highlights the coordination failures decentralization across regions. Lastly, He (2015) finds in intergovernmental fiscal relations as a result of fiscal that fiscal decentralization has no significant effect on decentralization. These failures, he describes, might arise environmental pollution (wastewater, gas emissions, from agency problems due to information asymmetry or solid waste), although fiscal decentralization has a between the central and local governments regarding significant positive effect on the pollution abatement the costs and benefits of government spending. Although spending and pollutant discharge fees charged. decentralization brings government closer to people and allows local jurisdictions to identify the preferences of China is far from the only country to be studied in relation the local community, it also increases the informational to this issue. Additional evidence from the Philippines distance between the central government and SNGs. shows that decentralized subsystems are more likely For instance, a central government may be unable to solve collective action problems, such as free riding, to monitor the local government’s revenue utilization, conflict resolution, and rule enforcement in large-scale spending, and delivery of services. Another cause for irrigation efforts (Araral 2011). Local governments are coordination failure is the “common pool” problem, which more responsive to complaints about water pollution in arises when a large share of SNG revenues comes from Indonesia (Bedner 2009). However, the lack of resources revenue sharing with the central government. When to carry out the environmental tasks remains. For this is the case, SNGs might underutilize their own tax cross-country findings, Farzanegan and Mennel (2012) bases. Indeed, using a sample of 30 countries, De Mello analyzed the impact of fiscal decentralization on different 14. Internationally, research has been conducted on whether stricter environmental standards in some countries would lead to the creation of pollution havens in others, a proposition that does not appear to be supported by the empirical evidence (Brunel and Levinson 2021). In a related manner, interestingly, Shapiro and Walker (2021) do not find breaches of locational environmental justice across jurisdictions in U.S. air pollution offset markets; specifically, pollution is not relocated to or from low- income or minority communities. Adapting Fiscal Decentralization Design to Combat Climate Change 17 indicators of pollution for a large panel of countries over better when natural disasters strike in terms of the effects three decades and found that it does tend to increase on the population, effects that are much more damaging pollution, but the effects are mitigated by the quality in developing countries. of institutions and central and local preferences for environmental quality. Their findings would also seem to Measuring the Current Involvement weakly support the prediction of a “race to the bottom” under decentralization. of Subnational Governments in Climate Action A possible avenue for a causal relationship between fiscal decentralization and environmental pollution is through economic growth. Several papers have provided If there are good reasons for SNGs to be active agents a theoretical framework for an inverse U-shaped in climate change action, an important question is how relationship between environmental pollution and involved they actually are. Measuring this involvement economic growth (Lopez 1994; Stokey 1998; Hartman poses significant difficulties with the information that is and Kwon 2005; Copeland and Taylor 2004). Recent available. This paper approaches the question from two papers have shown that there is also an inverted complementary angles. First, it looks at the quantitative U-shaped relationship between fiscal decentralization evidence on the relative share of environmental and environmental pollution (Liu and Li 2019; Hao et al. expenditures of SNGs across developed and developing 2020; Liu, Ding, and He 2019). countries. This clearly provides only a limited view, since SNGs are likely involved in other related decarbonization Another set of recent papers explore the relationship activities (transportation, housing, sanitization, etc.) and between fiscal decentralization and (renewable) the category “environmental expenditures” quite likely energy consumption. Elheddad et al. (2020) find a non- does not include expenditures in adaptation projects.17 But linear relationship between fiscal decentralization and the functional budget classifications currently in use do not renewable energy consumption in Chinese provinces. Su, separately identify those other types of expenditures.18 Umar, and Khan (2020) also find evidence of a similarly Second, the paper also looks at the evidence on SNG positive relationship in Organisation for Economic Co- involvement in the Nationally Determined Contributions operation and Development (OECD) countries. (NDCs) under the Paris Agreement. There is less empirical evidence so far on whether fiscal On the first issue, Gilmore and St. Clair (2018) find decentralization aggravates the damage of natural that climate decarbonization and adaptation activities disasters associated with climate change. The few represent less than one percent of total public spending empirical studies that have been conducted on this in most states in the United States. This is sobering question seem to indicate that it is not a catalyst for information, since state governments in the United States increased damages.15 Skidmore and Toya (2013), using are well placed to identify opportunities for mitigation and data from a large panel of countries, find that fiscal needs for adaptation to climate change and further have decentralization leads to a smaller number of disaster- a very high degree of budgetary autonomy. The evidence related deaths.16 Similarly, Escaleras and Register (2012) reported by the OECD (2018) for a sample of 30 countries find that relatively more decentralized countries fare 15. A different question is whether there are long-term damages associated with natural disasters independent of the level of fiscal decentralization. Here, again, the evidence is not very robust. For example, Kocornik-Mina et al. (2020) studied the question of permanent damage in the context of floods, which they found have displaced more than 650 million people worldwide over the past 35 years; they also found that damages are not permanent and that flooded areas tend to recover rather quickly. In a similar vein, Gignoux and Menéndez (2016) have examined the long-term effects of earthquakes in rural Indonesia since 1985 and found that the stocks of productive assets, notably in farms, are reconstituted and public infrastructures are actually improved, seemingly partly through external aid. 16. Note that Skidmore and Toya (2013) consider all types of natural disasters, not just those linked to climate change. 17. On the other hand, not all environmental expenditures might address climate change. 18. Admittedly, as just pointed out above, SNGs undertake substantial expenditures in areas, such as urban transit and waste management, that may not count as expenditure on a “climate action” but still could have relatively strong climate impacts. If those expenditures were added in, the picture of SNG involvement in climate change action in many countries would likely change rather significantly but measuring what SNGs do in these areas and how they contribute (or not) to mitigation and adaptation in tangible ways poses significant difficulties. How much of the spending, for example, on sanitation or transportation could be linked to climate change adaptation or mitigation? To what extent should the existing deficiencies in the provision of those services otherwise linked to climate change be accounted for as contributing to the aggravation of the problem? Such fundamental complications make nuanced measurement very challenging. Adapting Fiscal Decentralization Design to Combat Climate Change 18 over the 2000–2016 period is not much more positive 2016 on average in the 30-country sample, both in real (Plouin and Allain-DuPre 2018). In these countries, the terms and as a share of GDP. As a third piece of evidence, majority of environmental and climate-related spending table 1 summarizes the most recent information available occurs at the subnational level (55 percent for total in the World Observatory of Subnational Government spending and 64 percent for capital investments). Finance and Investment (SNG-WOFI) on SNG spending However, the share of subnational environmental and on environmental matters across developing and climate-related spending and investment is very low developed countries (OECD and UCLG 2019). For with respect to GDP at 1.3 and 0.4 percent on average, the vast majority of countries included in that work, the respectively, over 2000–16. Perhaps more of a concern share of environmental expenditures in total subnational is that both environmental and climate-related spending expenditures or GDP is still very small. and investment saw minimal change between 2000 and Table 1. SNG Spending on Environmental Protection: Breakdown by Region/Country Region Country SNG Spending on Environmental Protection Africa Mauritius Environmental protection was ranked third and made up 17% of total SNG expenditure. Rwanda Environmental protection is the least funded category (0.2%). Uganda Noteworthy is the low expenditure in economic affairs/transport, housing/ community affairs, and environmental protection due to the lack of capital finance for local governments. Zimbabwe Environmental protection makes up less than 1% of SNG spending. Eurasia Azerbaijan Environmental protection makes up 0.8% of total SNG expenditure. Belarus 0.1% is allocated to environmental spending. Georgia Environmental spending makes up 4.2% of SNG expenditure. Kazakhstan Environmental protection makes up 0.3% of total SNG expenditure. Kyrgyzstan Expenditure items, such as environmental protection, have a negligible impact on subnational budgets. Moldova Environmental protection makes up 0.3% of total SNG expenditure. Russia Environmental protection is not among the primary areas of SNGs and 98.7% of environmental protection expenditures are undertaken by the federal government. Tajikistan Environmental protection makes up 0.1% of total SNG expenditure. Ukraine Environmental protection makes up 0.4% of total SNG expenditure. Uzbekistan Environmental protection makes up less than 1% of total SNG expenditure. Latin America Brazil Environmental protection makes up 0.7% of total SNG spending. Colombia Environmental protection makes up 1.9% of total SNG spending. Georgia Environmental spending makes up 4.2% of SNG expenditure. Dominican Local expenditure is primarily devoted to general public services (42.5%), Republic followed by environmental protection (24.7%) and housing and community amenities. Adapting Fiscal Decentralization Design to Combat Climate Change 19 Region Country SNG Spending on Environmental Protection Mauritius There is very little intervention by municipalities in environment-related issues (3.1%). Guatemala Local government expenditure is concentrated in four sectors: public administration (civil registration), economic affairs (roads), environmental protection (7.4%) (waste management and sewerage), and cultural and recreation activities and infrastructures. Nicaragua Environmental protection makes up 2.6% of total SNG spending. Peru Other significant items, after “general services expenditure,” are health (regional responsibility) and environmental protection (10.9%). Uruguay Environmental protection makes up 11.4% of total SNG expenditure. Asia-Pacific Australia Environmental protection makes up 3% of total SNG expenditure. Local government areas are responsible for the majority of public spending (52%). They also play a key role in environmental protection and management of green areas, being responsible for 61% of SNG spending in this sector, amounting to 42% of total public spending. China Environmental protection makes up 3.7% of total SNG expenditure. India Environmental protection makes up 0% of total SNG expenditure in India. Indonesia Environmental protection makes up only 2% of total SNG expenditure. However, regions and cities are also heavily involved in housing and community amenities and environmental protection (respectively 65.5% and 55.2% of public expenditure). Japan Environmental protection makes up 6.8% of total SNG expenditure. Korea Environmental protection makes up 4.4% of total SNG expenditure. New Zealand Major categories of SNG spending include general public services, environmental protection (16.9%) (including waste and water), housing and community amenities (including water distribution), and recreation and culture. Philippines Environmental spending makes up less than 2% of total SNG expenditure. Middle East and Israel Environmental protection makes up 9% of total SNG expenditure. West Asia Israeli local governments are responsible for the vast majority of overall public spending in the areas of environmental protection (87% of public spending) and housing and community amenities. Turkey The two most important SNG expenditure items (excluding general public services) are community amenities (mainly drinking water, housing and community development) and economic affairs and transport. They are followed by environmental protection (9.5%), recreation, and culture. Europe Albania Environmental protection makes up less than 0.1% of total SNG expenditure. Austria Environmental protection makes up 1.2% of total SNG expenditure. Belgium Environmental protection makes up 3% of total SNG spending. Bulgaria Environmental protection makes up 8% of total SNG expenditure. Adapting Fiscal Decentralization Design to Combat Climate Change 20 Region Country SNG Spending on Environmental Protection Croatia Environmental protection makes up less than 0.2% of total SNG expenditure Cyprus The second most important item is housing and community amenities (mainly urban lighting, water supply), recreation and culture, and environmental protection (14.2%). Czech Republic Environmental protection makes up 6.3% of SNG expenditure. SNGs are responsible for 82% of total public expenditure in environmental protection policies, as well as for 68% in the housing and community amenities sectors Denmark Environmental protection makes up 0.6% of total SNG expenditure. Estonia Environmental protection makes up 2.9% of total SNG expenditure. Finland Environmental protection makes up 0.4% of total SNG expenditure. France Environmental protection accounts for 7.4% of total SNG spending. Germany Environmental protection makes up 2.1% of total SNG expenditure. Greece The main categories of SNG spending include general public services, economic affairs and transport, environmental protection (14.9%), and social protection. Hungary The shares of SNG spending in education, health, housing and community amenities, recreation and culture, and environmental protection have sharply decreased since 2011. Environmental protection makes up 2.1% of total SNG spending. Iceland Environmental protection only makes up 1.7% of total SNG spending. Ireland Environmental protection makes up 10.2% of total SNG spending. Italy Environmental protection makes up 5.7% of total SNG expenditure. SNGs are also responsible for the large majority of overall public spending in the areas of environmental protection and housing and community amenities. Latvia Environmental protection makes up 1.3% of total SNG spending. Lithuania Environmental protection makes up 3% of total SNG spending. Luxembourg Economic affairs/transport, recreation and culture, and environmental protection account for around 15% each, while education and social protection account around 10–12%. Malta Administrative tasks represent almost 37% of expenditure compared to 16.5% in the EU28 in 2016. This is followed by environmental protection (33.7%) (primarily waste management, landscaping, and maintenance of parks and gardens) and economic affairs and transport. Netherlands SNGs are responsible for the vast majority of total public spending in the areas of environmental protection (9.4%), housing and community amenities (water distribution, street lighting, etc.), and recreation and culture, especially since they include infrastructure projects. Norway Environmental protection makes up 4.2% of total SNG spending. Poland Environmental protection makes up 2.3% of total SNG expenditure. Adapting Fiscal Decentralization Design to Combat Climate Change 21 Region Country SNG Spending on Environmental Protection Portugal Environmental protection makes up 8.1% of total SNG expenditure. Romania Environmental protection makes up 4.9% of total SNG expenditure. Serbia Environmental protection makes up 3.4% of total SNG expenditure. Slovak Republic Environmental protection makes up 5.5% of total SNG spending. SNGs are responsible for more than half of public expenditure in the area of environmental protection and housing and community amenities, especially since their tasks include infrastructure related to water supply, sewage, heating, parks and open spaces, and public lightning. Slovenia Environmental protection makes up 3.5% of total SNG expenditure. Spain In health, education, environmental protection, and housing and community amenities, Spanish SNGs are responsible for almost all public spending nationally. For environmental protection and housing and community amenities, funding responsibilities are shared between regional and local governments, with the latter being responsible for the larger share, respectively, 70.3% and 66.5% of SNG total spending in each category. Sweden Environmental protection makes up 0.6% of total SNG expenditure. Switzerland Environmental protection makes up 2.5% of total SNG expenditure. Municipalities are responsible for a large share of total public spending on housing and community amenities (they are responsible for 83% of total public spending in this sector), environmental protection (62% of total public spending), and recreation and culture (58% of total public spending). United Kingdom Environmental protection makes up 4% of total SNG expenditure. Source: Adapted from OECD and UCLG (2019). The second measure of SNG involvement to consider specifically requested support for subnational action is the NDCs submitted by countries participating in (NDC Partnership).20 The most common requests the United Nations Framework Convention for Climate include institutional capacity building to integrate climate Change (UNFCCC) adoption of the Paris Agreement. change policies and planning at both the national and These NDCs are not implementation plans or blueprints subnational levels.21 for implementation, but they offer the general frameworks for action against climate change by member countries. In summary, up to now and despite its potential, SNG Sarmiento et al. (2018) report that in 2015, of the involvement in climate-related actions is generally NDCs submitted by 60 “REDD+ countries,” only 14 modest in most countries. This finding provides further explicitly mentioned a role for SNGs in mitigation and justification for this report, seeking to lay out the only four of these latter gave SNGs a decision-making general venues through which SNGs could become role.19 Overall, the NDC Partnership up to 2020 has leading agents in the planning and implementation of received 802 requests for support from a wide range decarbonization and adaptation policies. of countries, of which 105 (from just 20 countries) have 19. REDD stands for “reducing emissions from deforestation and forest degradation.” Each REDD+ Country Participant is a developing country located in a subtropical or tropical area that has signed a Participation Agreement to participate in the Readiness Fund of the UNFCCC. 20. NDC Support Facility (worldbank.org). 21. By region, the countries requesting subnational action support are: (i) East Asia and Pacific (9 requests: 1 from Mongolia, 4 the Philippines, 1 Marshall Islands, and 3 Vietnam); (ii) Europe and Central Asia (3 requests: 2 from Albania, 1 Georgia); (iii) Latin America and the Caribbean (8 requests: 5 from Colombia, 1 Mexico, and 2 Peru); (iv) Middle East and North Africa (21 requests: all from Jordan); (v) South Asia (2 requests: both from Nepal); and (vi) Sub-Saharan Africa (62 requests: 2 from Burkina Faso, 15 Ethiopia, 5 Kenya, 15 Mozambique, 3 Namibia, 6 Nigeria, 1 Rwanda, 3 Uganda, and 12 Zimbabwe). Adapting Fiscal Decentralization Design to Combat Climate Change 22 03 Reassessing the Assignment One fundamental reason for assigning some climate change actions in general, and in particular environmental of Functional Expenditure responsibilities, to SNGs is the fact that, as for any other types of public services, they are expected to be more Responsibilities for More responsive and accountable to residents.22 However, Effective Action on as in the case of other services, given the scope of associated externalities or the economies of scale in Climate Change the production and delivery of services, analysts should expect that there are environmental activities that are logically better assigned to central governments. Thus, the basic question is: what level of government is best equipped to address the different policy challenges in decarbonization and adaptation posed by climate change? How should expenditure responsibilities be assigned between central, regional, and local- level governments to maximize the effectiveness of decarbonization and adaptation activities? Regardless of the specific assignment of expenditure responsibilities arrived at, vertical and horizontal coordination between government units can lead to improved outcomes.23 However, even if free rider concerns are set aside, effective coordination still demands an assignment of functional responsibilities that is as clear as possible, bringing balance between central, top-down control and subnational, bottom-up discretion in climate change action. To proceed, next is a review of the general theoretical principles for expenditure assignments as well as the most common errors and problems with assignments 22. For example, Bedner (2009) finds that local governments are more responsive to complaints about water pollution in Indonesia. For empirical evidence on the responsiveness of SNGs in other general areas of public services, see Martinez-Vazquez, Lago-Peñas, and Sachi (2017). 23. There has been a lack of consensus among theorists on the need for coordinated action among governments in the presence of significant externalities, such as those associated with climate change. Some claim that the presence of pollution spillovers in decentralized settings does not always have to result in inefficient allocation of public resources to combat pollution. For example, in a well-known paper, Ogawa and Wildasin (2009) analyze a situation in which jurisdictions compete with each other in taxing mobile capital and find that uncoordinated policy making leads to an efficient allocation of resources, even without any central government intervention or subnational coordination. However, others have shown that under alternative assumptions those results do not hold. For example, Yamagishi (2019) shows that if the level of environmental regulation is endogenized, a likely outcome is a race to the bottom, and thus environmental policy coordination across jurisdictions is needed. Adapting Fiscal Decentralization Design to Combat Climate Change 23 in actual practice. The goal is to find how best to use size of the “benefit area” associated with those those principles and lessons to inform decisions and services.24 This also works as a default clause for assess actual practices in the assignment of functional the assignment of residual (not explicitly assigned) responsibilities in the area of climate change. and any new functions. Third, efficient service provision is generally Basic Principles and Most Common enhanced by making use of the “benefit” principle Problems with Expenditure of service finance: wherever possible, utilizing fee Assignments structures and issuing service charges to service users. This general “pricing” mechanism not only contributes to recovering the costs of provision, but Expenditure assignments tor SNGs are usually guided it also helps reveal users’ demands and preferences by several specific principles, as follow: for services and reduce congestion by rationing demand.25 First, the “correspondence” principle simply states that the geographical dimension of the Fourth, the principle of “affordability” considers benefits received from a service should match the whether subnational authorities have enough geographical dimension of the level of government resources to adequately deliver the service and responsible for its provision. Thus, services expenditure responsibilities assigned to them. spreading throughout the national territory (for Although a basic question, it is difficult to answer example, air traffic control) should be assigned at well, as it is a complex issue that depends on how the central level, services with benefits with regional needs and standards are defined and on the design dimensions (for example, tertiary specialized of the entire SNG financing system (own revenues, hospitals) at the regional level, and those with a local transfers, and so on). dimension (for example, street cleaning) at the local level. The correspondence principle implies that Fifth, the principle of “capacity” puts the focus on given the nationwide externalities associated with whether SNGs are administratively and technically income distribution and macro stabilization policies, capable of delivering the service function assigned. these activities should be mostly reserved for the Wider behavioral notions of “capacity” are central government. Local and regional governments sometimes used to include subnational political may have difficulties sustaining independent (self- responsibility for duly prioritizing functions perceived financed) redistribution programs because they risk to be of national priority. This latter tends to be a attracting the needy from other areas while more thorny issue, especially in newly decentralized heavily taxing their (potentially mobile) residents. For systems where there is central mistrust about stabilization policies, such as massive investment the capacity and intentions of newly appointed or unemployment compensation, SNGs may be subnational authorities. Some countries address relatively too small to be effective and their efforts these issues by differentiating between obligatory may be quickly diluted within the open borders with versus discretionary or voluntary responsibilities or the rest of the national economy. by imposing compulsory minimum standards and other forms of expenditure rules or mandates (for Second, the “subsidiarity” principle states that the example, some minimum percentage of budget responsibility for services that can be provided at resources should be spent on education, or no more different levels should be assigned at the lowest than a certain percentage may be spent on public possible level of government compatible with the employee salaries). 24. The European Charter of Subnational Governments uses the “subsidiarity” principle as the most prominent guide for expenditure assignments. In academia, a prominent defender of the application of the principle of subsidiarity to the assignment of responsibilities in environmental issues is Wallace Oates and his coauthors (Oates 1998, 2002; Oates and Schwab 1996; Oates and Portney 2001). Assigning responsibility to the lowest level of government that spatially encompasses most of the benefits and costs associated with the service allows the service to be tailored to the preferences and needs of residents and to the costs of production and other local conditions, and therefore yields higher social welfare (Dalmazzone 2006). 25. In practice, however, fees and charges are not as frequently used because of fairness considerations regarding lower-income users, especially when there are no offsetting alternative income compensation and redistribution mechanisms. Adapting Fiscal Decentralization Design to Combat Climate Change 24 Last, and as already pointed out above, for any (federal versus unitary), available technologies, citizen system of expenditure assignments to work preferences, and traditions and history, may affect the smoothly, there is always a need for the principle final shape of assignments. What may be considered of “coordination” among the different levels of a good assignment is likely to transform over time with government. No expenditure assignment can changes in the costs of provision as well as technological ever be detailed enough to preempt the need for innovations. More important than the “best” assignment coordination and dialog among the different levels is to achieve an explicit, clear, and stable assignment of of government, especially in areas of concurrent expenditure responsibilities. responsibilities. Ultimately, disagreements must be left for interpretation and rulings by the courts. The assignment of responsibilities may also need to take into account the presence of corruption and other political The application of the above principles facilitates economy constraints. For example, it is not uncommon the assignment of expenditure responsibilities. For for zoning and city planning to be exposed to corruption certain functions, it is possible to arrive at “exclusive” pressures, which may require the reassignment of assignments. All attributes of that function, including responsibilities or monitoring and third-party oversight by regulation, financing, and implementation, are to be central government officials. carried out by one single level of government. Exclusive assignments are desirable because they more easily The international experience provides some other lead to transparency and accountability. For other valuable lessons on what to avoid in expenditure functions, some of the attributes, such as regulation, assignments. First, a commonly found problem is the lack may be assigned to one level while other attributes, like of a formal or explicit assignment of responsibilities in the financing or implementation, are assigned to other levels laws. In some cases, that may not be a problem if there of government. In some cases, some attributes may be is a long tradition and stability of de facto assignments; the responsibility of several levels of government. All of however, these arrangements can take a long time to get these are concurrent or shared responsibilities. But even established and are exposed to more institutional friction in these cases, clarity and accountability may (to some and inefficiencies (including duplications or preemption). extent) be preserved if the main attributes imbedded Second, there is the risk of inefficient or wrong in any service responsibility (regulation or norming, assignments that openly break with the correspondence financing, and implementation) are explicitly assigned. principle. Third, it is important not to establish unstable and constantly shifting assignments, such as those that One of the most useful lessons from the theory and change according to annual budgets. The final pitfall to practice of expenditure assignments is that it is not avoid is the extensive use of concurrent responsibilities, meaningful to talk about “the single best” assignment. preventing clarity on what level of government is Other relevant factors, such as the nature of the state responsible for the different “attributes” of the service. Adapting Fiscal Decentralization Design to Combat Climate Change 25 Applying Basic Principles and Best Third, one that lies between the two previous polar cases, a local type with regional externalities to other Practices to the Assignment of neighboring jurisdictions, but without adding up to an Responsibilities for Climate Change “aggregate” national or global externality; examples include shared water basins, acid rain, and so on. Policy Action The question of which level should be in charge can be answered by applying the correspondence principle How can those principles and lessons on general to the geographical scope of the externality (Alm and functional assignments be applied to the area of Banzhaf 2012). In the case of a national or global type of climate change? What level of government is better externality, the central government should be in charge, equipped to address each of the potential challenges since subnational levels of government likely will not from climate change and maximize the effectiveness consider the full geographical effects of the emissions. of decarbonization and adaptation activities? From Of course, what being in charge means depends on the discussion above, the strong argument for the the policy choices actually made, for example, setting decentralization of services weakens as the costs of the overall regulatory context and emissions targets, decentralized provision increase because of economies determining the role of local governments or leaving it to of scale or scope in the delivery of services or because the market with an emission trading system (ETS) and of the presence of substantial spillovers or externalities. carbon tax, and so on. For the second type of externality, Thus, an important component of determining suitable the local type, local governments should be responsible, responsibility assignments is how economies of scale and according to the correspondence and subsidiarity externalities manifest themselves in different activities principles. For the third type with the presence of regional and policies related to climate change. For example, externalities, regional- or intermediate-level governments when one asks the question of which level of government would be the most indicated by again applying the should be assigned the responsibility for “emissions or correspondence and subsidiarity principles. pollution control,” the answer is that it depends on which subfunction or aspect of pollution control or climate Indeed, the “correspondence” principle is at the change policy the term refers. Generally, three different heart of getting the expenditure assignments right dimensions of polluting activities have been discussed for environmental and climate issues in a variety of in the economics literature (Oates, 2002; Alm and other ways. For example, it is of standard use with the Banzhaf 2012):26 establishment of water resource management districts because water basins are often different from the First, a national or global type, where environmental geographical boundaries of administrative units.27 This is quality in any location depends on the sum or also why, for example, the Netherlands created separate aggregate level of emissions from all local areas, administrative units to manage floods, different from other even though conditions may change locally administrative units. Getting the right correspondence of with weather patterns, and so forth. This type of the service provision with the right geographical area for externality is at the core of climate change caused by things such as flood zoning regulations is affected by GHGs as well as pollutants that deplete the earth’s the high level of fragmentation and competition among ozone layer. small municipalities in large metropolises or loose urban areas.28 However, the useful application of the Second, a local type where pollution in each locality correspondence principle is contingent on the fulfillment depends exclusively on the level of emissions of the other principles of expenditure assignments, such there, with no externalities or spillovers beyond as capacity and affordability. So, will local governments the boundaries of the local government itself, for have the capacity to monitor and sanction polluters of the example, the solid waste generated in the area. 26. The same applies to other services. For example, conservation activities, such as sustainable water management or biodiversity, are associated with regional, national, or even global public goods. And again, when large spatial externalities are present and no mechanisms are in place to sufficiently compensate for them, the likely result is an under-provision of the services. Typically, central governments utilize conditional grants to make SNGs internalize some of the externalities that otherwise would be ignored in their decisions. That use of transfers is discussed in another section of the paper below. 27. See, for example, Van den Brandeler, Gupta, and Hordijk (2019). 28. For example, Przyluski, and Hallegatte (2012) describe how many of the mayors of municipalities affected by the storm Cynthia on the coast of France mentioned the intense competition to attract population and tax resources and the risk of losing them to neighboring municipalities if they were to take flood zoning too seriously. Similar difficulties are common in large metropolitan areas with many fragmented jurisdictions. Adapting Fiscal Decentralization Design to Combat Climate Change 26 local type or perform other needed activities? Capacity However, not all decarbonization activities need to be has become an everyday problem for many climate centrally assigned. There are a good number of activities issues. As mitigation increasingly relies on new, fast- that have a local dimension, such as urban transport, evolving technologies, capacity constraints become a housing construction codes, and land-use regulations, bigger issue.29 In addition, many SNGs cannot afford to which also have a related impact on GHG emissions.30 hire experts on disaster management, and only when the In most of these cases, given the local dimension of the funds are available can they rely on private contractors activity, it would not make a lot of sense to transfer the that are more or less reliable. In this sense, capacity can responsibility to upper-level government. Nevertheless, be a binding constraint for allocating responsibilities for even for these decidedly subnational activities, there climate change activities. may be a role for the central authorities in developing regulations, establishing minimum standards, helping Even if the answer is that both capacity and affordability with financing, or information dissemination, monitoring, are present, things can go wrong if SNGs adopt harmful and oversight. strategic behavior by competing among themselves, selecting low levels of service (with correspondingly In some cases, subnational authorities may be able to deal lower charges and taxes) in order to attract more efficiently with activities representing large externalities if economic activity. Thus, whether SNGs will make efficient strong institutions for policy coordination are present. The decisions on the level of the service (for example, control intergovernmental fiscal system should incentivize local of local-level pollution)—that is, whether the functional officials to consider all types of externalities in making assignment is right—will also depend on the absence their decisions. Fiscal incentives could be in the form of of strategic behavior. Exposure to corrupt practices conditional transfers or subsidies to particular actions and is another consideration to keep in mind in the final programs.31 Political decentralization institutions, such as assignment of responsibilities. centralized political party arrangements, can also better ensure that externalities and the national good will be More generally, the assignment of responsibilities among internalized by subnational officials.32 different levels of government is likely to differ for the broad categories of decarbonization and adaptation The considerations above indicate that a good number activities. For decarbonization activities, in the presence of responsibilities involving decarbonization policies will of extensive negative externalities from consumption and call for concurrent assignments involving two or more production activities that lead to the emission of GHGs, levels of government. In this case, as described in the the assignment of responsibility should first consider the discussion of principles of expenditure assignment, need for the involvement of higher-level government. what becomes important is how to design these The assignment at higher levels of government would concurrent responsibilities to ensure maximum clarity increase the chances that, to some extent, those and coordination.33 The key for this is clearly identifying extensive externalities would be sufficiently taken into what level of government is responsible for the different account in policy decisions. But even that may not be attributes of the function (regulation, financing, and the case; when the negative externalities’ effects go actual implementation).34 beyond national boundaries, there might be justification for international action, thus the reason for the existence The two most common non-tax economic instruments of the Paris Agreement and other international forums. (EIs) used to correct externalities such as those 29. The problem seems to be widespread. For Japan, Takao (2012) finds that only prefectures and the largest municipalities in the country have the technical capacity to develop comprehensive plans for climate adaptation. In a similar vein, Næss et al. (2005) find deficient capacity for flood adaptation and prevention policies at the local level in Norway. USAID (2017) reports on similar capacity constraints at the local level for climate change adaptation in Mali. 30. A more complete list of how SNGs engage in mitigation activities includes: maintaining building codes and infrastructure standards to increase energy efficiency; establishing land-use and urban planning standards to promote and preserve green areas and forests; reducing water and energy use in the provision of public services; promoting green activities, such as soil and water conservation, rain water harvesting, or crop diversification and agricultural practices; and improving public transport, vehicle emissions standards, and vehicle energy mix (Danida 2009). 31. For more on this, see the discussion of actual country practices in the intergovernmental transfers section below. 32. See, for example, Hankla, Martinez-Vazquez, and Rodriguez (2019). In this regard, Estache, Garsous, and Seroa da Motta (2016) find that the hierarchy within a political party allowed the governor of São Paulo, Brazil to provide aligned mayors with tailored incentives to significantly improve local sanitation infrastructure. 33. When there is shared responsibility for the production of some services, the lack of clarity in regulation and responsibility powers at the different levels of government can lead to inefficiencies and the under-provision of services. For example, Kresch (2020) documents how regulatory uncertainty between state and local governments in Brazil’s water and sanitation sector led to the undersupply of services, and how a legal reform that clarified responsibilities led to the almost doubling of total system investment by municipal companies, resulting in significant increases in system access and decreases in child mortality. 34. The assignment to SNGs may involve the attribute of financing. One argument often raised against giving SNGs more responsibility on environmental and climate change issues is their lack of resources to finance these tasks. This issue of financing climate action is addressed separately in the next section. Adapting Fiscal Decentralization Design to Combat Climate Change 27 associated with GHG emissions are a “cap and trade” However, the incidence and costs of extreme climate framework—whereby trades or offsets are allowed events also imply the need for redistributive or among polluters and the total level of pollution does not compensatory actions from upper-level governments.36 exceed the cap—and a “command and control” system, In this regard, most decentralized systems contain used when pollution is not easily measurable (Cointrau an important element of national insurance covering and Hornig 2003; Alm and Banzhaf 2012).35 Because individual jurisdictions for the effects of extreme climate “cap and trade” applications require a relatively broad events. Thus, it will be also the case that many adaptation geographical market, it is most likely that this type of activities will call for the assignment of concurrent instrument will be applied by either central governments responsibilities. Although SNGs have responsibility for or very large regional or state governments. All levels prevention measures, such as building the necessary of government may apply the “command and control” infrastructure, and remedial actions, such as cleaning instrument as long as they have monitoring capacity, and and rebuilding in the aftermath of those climate events, in some cases, SNGs may have better information, which higher levels of government, including central or federal is also a traditional justification for decentralization. authorities, have responsibility for contributing financing to the recovery efforts as well for regulating general In the case of adaptation activities, unlike decarbonization prevention measures.37 But this typical division of policies, the risks and consequences of climate change– responsibilities between central government and SNG related disasters tend to be more localized, covering creates a thorny moral hazard issue. Although SNGs the extent of some localities or even regions or states, typically rely on federal/central support funds when but with few cases in which significant externalities are natural disasters occur, these are the jurisdictions that present affecting other jurisdictions. However, slow onset are responsible for the capital infrastructure necessary adaptation needs (e.g., changes in agricultural zones) to reduce the impact of those events.38 Private insurance may require more of a national perspective. Thus, using markets can play a positive role in reducing adaptation the correspondence and subsidiarity principles, the costs for SNGs.39 assignment of responsibilities for adaptation policies would appear generally to fall more logically with the It is also questionable whether many SNGs would have potentially affected subnational jurisdictions. Typical the capability to conduct their own disaster and climate SNG adaptation activities will include issuing regulations risk assessments. This is especially true in the case of on land-use planning and zoning that avoid construction SNGs in developing countries, which may lack awareness or use of areas at high risk of flooding, developing and as well as personnel and other resources to conduct risk preparing contingent response plans for climate disaster assessments for land-use planning, and so on. This is emergencies, and improving infrastructure to better clearly another area in adaptation activities where central handle flooding or other climate change–related risks. governments may need to support SNGs.40 35. It must be also recalled that there are certain potential equivalences between tax and non-tax EIs, such as quotas. See, for example, Li (2020). 36. Another form of risk for SNGs is that of “stranded assets,” that is, current infrastructure that is not appropriate for the new natural disaster environment. There are other variations of the stranded assets’ implications. See, for example, Morris, Kaufmann, and Doshi (2020) who study the significant tax revenue implications of future carbon constraining policies for counties in the United States that are heavily reliant on coal-based revenues. 37. Nevertheless, central disaster relief is not always assured to work properly. Sobel and Leeson (2006) examine the failure of the Federal Emergency Management Agency (FEMA) and other government agencies in the United States to provide effective relief support in the wake of Hurricane Katrina. 38. For example, Goodspeed and Haughwout (2012) show that when central authorities commit to full insurance against natural disasters, SNGs have an incentive to under-spend on costly adaptation measures. Lower levels of insurance transfers may result in larger preventive investment spending, but a good balance is probably hard to achieve. Mcleman and Smith (2006) review possible examples from Canada, New Zealand, and the United States where central government actions have encouraged forms of development that put more people and property at risk. Similarly, Bagstad, Stapleton, and D’Agostino (2007) highlight how fragmented federal programs for coastal disaster protection in the United States have promoted growth in areas of increased risk. 39. In general, there may be little awareness of the risks associated with natural disasters (Edwards et al. 2019 and ICLEI 2018). When private insurance companies withdraw from certain markets, that may signal less development and smaller damages from natural disasters. Insurers may also incentivize local governments to undertake appropriate preventive measures (GIZ, UN-Habitat, and ICLEI 2017). Linnerooth-Bayer and Mechler (2006) discuss possible strategies for supporting developing country adaptation to the risks of climate change via insurance, building on the intent of Article 4.8 of the UNFCCC. The authors describe two potential international examples. First, the Turkish Catastrophe Insurance Pool, which was launched in 2000 with support from the World Bank to cover earthquake risks, but which could be expanded to other natural disasters. Second is Mexico’s taxpayer-supported national catastrophe fund (Natural Disasters Fund [El Fondo de Desastres Naturales, or FONDEN]), in combination with a mixed catastrophe bond issued as part of an insurance risk-transfer strategy to protect FONDEN (investors receives an above-market return when a specific catastrophe does not occur in a specified time but sacrifice interest or part of the principal following the event). 40. Colombia and the Philippines are country examples in building SNG indices for disaster risk. See, respectively, Mejía 2018 and World Bank 2017. Adapting Fiscal Decentralization Design to Combat Climate Change 28 Other Potential Obstacles to the that states applying for “authorization” tended to adopt stricter environmental norms compared to the federal Devolution of Climate Change standards. Relatedly, Araral (2011) found that in the Responsibilities to Subnational Philippines, decentralized subsystems were more likely to solve collective action problems, such as free riding, Governments: Political Economy conflict resolution, and rule enforcement on large-scale Considerations, Free Riding, and irrigation programs. Common Pool Problems On the other hand, Sigman (2002) found that pollution in international rivers is worse than in domestic rivers, Central government authorities might not want to except for European Union (EU) countries where devolve decision making to SNGs even when it is cooperation is greater. Also, in a well-known paper, List, more efficient to do so. For example, the central Bulte, and Shogren (2002) found that from 1993 to 1998, government has a political incentive to be seen as the the devolution by the U.S. Environmental Protection primary responder to emergencies and to ensure that Agency (EPA) to the states (where delegation increased it is recognized for providing disaster assistance (Friis- by 75 percent) led to widespread free riding by the Hansen, Bashaasha, and Aben 2013). In addition, some states; more specifically, states were less inclined to arguments frequently made against decentralizing spend money on the preservation of species that would responsibility for environmental and climate change require a large habitat area or were directly in conflict issues involve the presence of free riding, “common pool” with development goals. problems, coordination failures, and strategic behavior among SNGs (de Mello Jr. 2000). Coordination failures What Are Current Effective and Less in intergovernmental fiscal relations could arise from agency problems due to information asymmetry between Effective Practices in Expenditure the central and local governments regarding the costs Assignments in Different Countries? and benefits of government spending. Coordination failure may also follow from common pool problems, often present when a large share of SNG revenues International practice with expenditure assignments for come from revenue sharing with the central government, climate-related policies is varied and rich and therefore leading SNGs to underutilize their own tax sources and difficult to categorize.41 Expenditure assignments are inflate their spending needs. multidimensional, involving the attributes of norming or regulation, financing, and implementation. A complete The empirical evidence on the presence of strategic taxonomy of expenditure assignments for combating behavior and competition among subnational units climate change would therefore require fully listing regarding pollution standards and other environment- all policies involved and identifying which level of related issues is somewhat mixed. For example, List government is responsible for regulation, financing, and and Gerking (2000) found that the devolution of authority implementation of each such policy. That task is beyond from the federal to state governments on environmental the scope of this paper; however, it is useful to look at quality standards during President Reagan’s new what general approaches countries have followed in the federalism in the 1980s did not lead to a reduction of division of responsibilities for climate change policies. those standards. Similarly, Konisky and Woods (2012) The most distinctive attribute to differentiate among did not find any supporting evidence of environmental country approaches is the level of government that is free-riding behaviors across U.S. states regarding responsible for norming or regulating climate-related enforcement of the federal Clean Water Act. Under most standards and economy-wide targets. In the experience federal environmental laws in the United States, states of most countries, both financing and implementation are may apply for “primacy,” that is, authority to implement generally concurrent attributes implemented by all levels and enforce federal law, through a process known as of government. From that perspective, international “authorization.” There have been fears that states may practice shows three clearly distinct options: a first use that process to put into place laxer regulations. group of countries where climate-related standards But, in fact, Chang, Sigman, and Traub (2007) found and economy-wide targets are exclusively centrally 41. There is also little precedent in the literature. See, for example, Breton et al. (2007) for a collection of country essays on multilevel coordination of environmental policies. Adapting Fiscal Decentralization Design to Combat Climate Change 29 mandated/determined; a second group with a mixed Three Main Approaches to Expenditure approach, where both central governments and SNGs Assignments contribute to the determination of targets and where national governments work as floor targets and SNGs (i) First, in some countries, national target setting by the are allowed to innovate and improve those minimum central government is followed by further allocation targets; and a third group, where standard setting is of targets to SNGs for implementation. totally decentralized, with no central mandates or targets. What follows in terms of financing and implementation, Two examples of these practices are Belgium then, is that under the first approach the implementation and China. may be only national or also delegated to SNGs; with the second and third approaches, SNGs are always also In the case of Belgium, this process is more involved in financing and/or implementation. conceptual than practical. Belgium’s Kyoto- agreed targets with the EU were reallocated to the The section below identifies and describes in some regional level through a process of negotiation, and detail several countries that operate in each of the three regional climate plans were prepared to implement categories. Later, there will be a more exhaustive look at those regional targets (Happaerts, Schunz, and how many other countries may possibly fit into each of Bruyninckx 2011).42 Most of the climate change those categories, but for now, some general reflections policies in Belgium (including GHG reductions and on the classification process and the advantages and carbon emission trades) have been redesigned and disadvantages of each are possible. First, each country implemented at the subnational level (of Flanders, case is complex and distinct, and thus some countries Wallonia, and Brussels). So, in effect, in the case of may not neatly fall into any of the three categories. Belgium, there is no national policy per se but rather Second, of the three approaches, the first (the a national framework, and in practice the regional centralized approach) sets minimum standards but only governments design and implement their own utilizes SNGs as implementers, while in the third (the policies within that framework. For this framework, decentralized approach), SNGs are free to innovate and there is an Inter-Ministerial Conference on the compete, though with no minimum national standards so Environment (IMCE), composed of the federal and there is also a risk that competition will turn into a race subnational environment ministers, that is in charge to the bottom. The advantage of the second approach is of coordinating intergovernmental cooperation on that in addition to offering minimum national standards, it climate change policies. also allows for innovation and initiatives among SNGs in what possibly may become a race to the top. 42. The process of agreements is to a large extent anchored in the particular institutions of Belgian federalism, where the principle of non-hierarchy means that the federal government cannot impose on the SNGs. Adapting Fiscal Decentralization Design to Combat Climate Change 30 In the case of China, the tradition of strong hierarchical (ii) There are other countries that more explicitly share and vertical organization of government facilitates and coordinate responsibilities for the design and the approach of a vertical division of climate change implementation of climate change polices among responsibilities and targets. First, the national action different levels of government. Countries like plan and targets are set and monitored by the central India, France, Germany, and Colombia are thus government, frequently in consultation with the somewhere between the model of central control provincial governments. Then, the implementation of of design and models of much more decentralized environmental, energy, and climate change policies assignment of responsibilities. and programs is delegated to the provinces, which in turn organize and supervise the further delegation of In the case of India, policy targets are set in a mix policies and targets to the other three levels of SNG, of national and subnational regulations. Functional from large cities and prefectures to townships.43 The responsibilities are shared between the central existing institutional arrangements for the promotion government and the state governments on the basis and supervision of subnational officials potentially of the “National Action Plan on Climate Change facilitate the effectiveness of national target settings (NAPCC)” at the Union level and the states’ “State for GHG reduction. At the national level, the powerful Action Plan on Climate Change (SAPCCs),” which is National Development and Reform Commission developed and implemented separately by each state (NDRC) and the State Environment Protection government (Dubash et al. 2013). In 2018, the central Agency (SEPA) are the central agencies responsible government asked the states to revise the SAPCCs, for environmental planning and energy policies.44 In and most states appear to have done so. Although contrast, it is the inter-ministerial National Climate predominantly focusing on the implementation of Change Policy Coordinating Committee, established national-level targets, the assignment arrangements in 1990, that makes climate policies; unlike the case allow for sufficient flexibility to pursue state-level of energy and environment policies, for climate goals, such as in the cases of the Climate Change policies there is no clear division of responsibilities department in the state of Gujarat or the Green with lower-level governments nor apparent incentives Fund in the state of Kerala (Atteridge et al. 2012).45 or penalties. Some provinces have adopted a However, there have been questions about how climate change policy and others have not; between effective the executive branches have been in 2007 and 2008, eight provinces, province-level implementing policies to fight climate change. For municipalities, and autonomous regions (of 34 such example, Mandal and Rao (2007) highlight how bodies) established Leading Groups on Climate the failure of the executive to regulate and monitor Change based on the central government’s model pollution and forest cover has led to Supreme Court (Teng and Gu 2007; Held, Nag, and Roger 2011). intervention, which issued several rulings on the In addition, the NDRC approved seven provincial implementation of environmental regulation, taking pilot cap and trade schemes for carbon emissions in over the role of the executive branch. 2011. The participating provinces were intentionally selected at varying stages of development, and Another good example of the mixed approach although the schemes have some common features is France. Even though France is a rather fiscally (all six GHGs covered under the Kyoto Protocol centralized country, its approach to the assignment are subject to the system and output levels require of responsibilities for the environment and climate third-party verification), the provinces were given change is a mix of national standards and SNGs’ own considerable discretion to design their own schemes initiatives. France’s national legislature, the Assemble (Zhang, Z. 2020). Nationale, adopted the EU objectives as its national 43. See Held, Nag, and Roger (2011) and Kostka and Hobbs (2012). 44. The 11th five-year plan (2006–10) included a 20 percent reduction in energy intensity and a 10 percent reduction in major pollutants, and these two targets were translated into differentiated targets at the provincial level. Before 2007, no SNG (from the provincial to the township level) had expressed serious interest in working on climate change issues (Qi et al. 2008). Different capacity to finance environmental policies has also affected the performance of different provinces. Teng and Gu (2007) note the lack of an intergovernmental transfer mechanism to improve energy efficiency and environmental standards in the poorer western China provinces. 45. Reportedly, prior to the NAPCC and the prime minister’s urging in August 2009 that all states develop a state action plan consistent with the strategies of the NAPCC, there was little initiative at the state level. There is still significant variation across states in recognizing climate change as a priority (Atteridge et al. 2012). Adapting Fiscal Decentralization Design to Combat Climate Change 31 goals in 2010; that same law required all urban law of 2014 assigned expenditure responsibilities agglomerations of over 50,000 people to prepare to the different levels of SNGs (region, county, and “Climate and Energy Territorial Plans” (TCEPs) to municipality), with most of the responsibilities for meet these goals and to address adaptation needs. climate change (biodiversity protection, climate, This mainstreaming of climate change objectives energy, and air quality) assigned to the regional into ongoing local public planning had the advantage level. In addition, the New Organization of the of making sure that all other local planning related Republic (NOTRe) Law, passed in 2015, clarifies to policies for transport, building, urban planning, the responsibilities of the Metropolitan Area of and energy had to conform and support the national Paris (Grand Paris) and strengthens the role of objectives for climate change (Vidalenc 2017). regional governments on energy transition issues. A Later, the Modernization of Territorial Public Action summary with highlights of the TCEP for the city of and Assertion of Metropolitan Areas (MAPTAM) Paris is shown in table 2. Table 2. Example Territorial Climate-Energy Plans (TCEP) The City of Paris and Its TCEP First adopted in 2007 and then updated in 2012, then Paris climate-energy plan calls for 75% reduction in emissions by 2050: Areas of Action Examples Land Use Planning Renewal of 10% of the city’s area (10km2) to attain high energy efficiency and low carbon standards. Housing/Public 15% reduction in emissions (900,000 t CO2 eq) obtained by building 25,000 public Buildings housing units to high energy efficiency standards, and by renovation of 25,000 units. Mobility • Opening of new tramway lines, large-scale implantation of the Vélib bike-share scheme and the Autolib shared electric vehicle rental service. • Certain metro lines extended, longer service hours on weekends, 15 BRT lines created, 700 km of bike lanes. Adaptation • First full audit conducted to assess the robustness and vulnerabilities of Paris in the face of climate change. • New green areas planted on over 100 hectares in 10 years. • Stronger preventive measures under the national scheme to care for persons at risk in the event of a heat wave. Consumption • Action to prevent and abate energy and fuel poverty: “100,000 Low-Energy Light Bulbs” campaign for low-income residents in Paris, 15,000 water-saver kits installed in public housing by water utility Eau de Paris. • Sustainable food sources: 30% organic foods in city food services (schools, employee canteens), representing 30 million meals/year, and 25,000 t of avoided CO2 emissions/year. Outreach in the The Parisian Climate Agency was created to proactively support the Climate Plan. Territory A “serious game”, Clim’Way Paris, has been launched to give all Paris residents the opportunity to complate their own climate plan by 2050. Source: ADEME (2015). Adapting Fiscal Decentralization Design to Combat Climate Change 32 In the case of Germany, the federal government, and sectoral guidelines for the strategy of low carbon which has also adopted the EU objectives as development, the national adaptation plan, the its national goals, has primary responsibility for national strategy for the reduction of emissions due to determining climate change objectives and it deforestation, forest degradation, and other causes. initiates most climate actions (Weidner and Mez All 32 departments—regional-level governments— 2008). Even though the federal states (the Länder) have included climate change management in their have few implementation responsibilities, they can development plans, and a good number of territorial take additional climate change actions, such as plans have also been formulated. The Financial establishing their own supplementary measures Management Committee of the SISCLIMA is the affecting GHG emissions or setting other energy interinstitutional coordination mechanism for climate efficiency and renewable energy targets. In financing; its National Climate Financing Strategy addition, each state has a comprehensive climate estimates that approximately 0.87 percent of the protection plan. annual national GDP would have to be spent to accomplish Colombia’s emission reduction targets The states also have something like veto power on up to 2030, and that 62 percent of the resources central initiatives through their representation in the would have to come from private sector polluting upper house of parliament, the Bundesrat. However, activities (according to their actual GHG emissions) despite this multilevel governance, conflict is rare and 38 percent from public investment. The in Germany’s cooperative federalism regarding this financing framework proposes the creation of a tariff and other expenditure assignment matters. As an for air emissions (equivalent to a carbon tax), the example, the federal and state governments jointly issuance of green bonds, and soft loans to SNGs for drew up the Dusseldorf Declaration, which proclaims decarbonization and adaptation projects. the EU targets as the national climate targets, and the federal and state governments hold a biannual In other countries, like the Netherlands, Brazil, (iii) Conference of Environmental Ministers. Recently, the the United States, and Canada, regional and local “Climate Action Plan 2050,” outlining the principles governments play a leading role in the design and and processes for national policy, was developed implementation of climate change polices. The lack in consultation with the Länder, local governments, of an enforced national policy risks not meeting some associations, and citizens. The main message in minimum standards, but it also allows for subnational the plan is that success in climate action will require innovation and initiative. In particular, the United that it be considered and implemented at all levels States and Canada provide good examples for how of government and by all stakeholders (GIZ, UN- SNG carbon emission policies can lead and improve Habitat, and ICLEI 2017). Finally, municipalities in on federal/central government actions. Germany are active in climate change activities, but with significant variations; some of those activities In the case of the Netherlands, the central include retrofitting public buildings and schools, street government has a long-established subsidy lighting, and public transport and promoting green program to municipalities to combat climate public procurement and “zero-energy consumption” change. As reported by Gupta, Lasage, and Stam (Weidner and Mez 2008). (2007), a national policy agreement on climate change, Bestuursaccord Nieuwe Stijl (BANS), was Although Colombia contributes little to worldwide negotiated in 1999 with about half of the country’s GHG emissions, it has the associated pressing 487 municipalities and all 12 provinces. The central problem of the deforestation of its tropical forests. government provided an initial subsidy of €37 Colombia adopted a vertically integrated climate million for this scheme in 2002 and has provided policy framework in the Intended Nationally an additional €6 million annually since 2007. BANS Determined Contribution (INDC) and the National covers 50 percent of the costs incurred by local System of Climate Change (SISCLIMA) in 2016 government, while the other 50 percent must come (GIZ et al. 2017). This latter organization is charged from other sources, such as provincial-municipal with the coordination of central, regional, and local funds, EU funds, or even private sources. climate change efforts and is organized into nine regional climate change hubs. The National Climate Financed by these subsidies, the Netherlands’ Change Policy (PNACC) of 2017 defines territorial local governments get involved in a wide range of Adapting Fiscal Decentralization Design to Combat Climate Change 33 climate change activities. The BANS program has (to some extent) by some states and also local seven themes: municipal buildings and installations; governments, as allowed in the Constitution.47 As housing; business and business parks; the part of the Conference of Mayors’ Climate Protection agricultural sector; traffic and transport; sustainable Agreement (2007), more than 1,000 U.S. cities energy; and international cooperation. SNGs can and municipalities signed an agreement to commit select from three groups of policy options: active, to reaching the targets signed onto by the United front runner, and innovative, with almost one-third States under the Kyoto Agreement. And even with falling into the last category. An indication of the the backtracking by the Trump administration, more breadth of involvement is that about two-thirds of than 700 cities continued to be committed to their the municipalities have a permanent budget for local climate change policies (Mehling and Frenkil 2013). climate policy, and about one-fourth calculate local Individual cities have adopted a variety of policies GHG emissions. Most of the cities in the Netherlands (McGarvery and Morsch 2016). For example, have developed energy-efficiency plans for the Chicago enacted the Building Energy Use and housing and transport sector and quantitative targets Benchmarking Ordinance in 2013 to improve the for GHG emissions; those with adaptation measures energy performance of the city’s building stock; all are fewer but, importantly, participation in the commercial, institutional, and multi-family properties BANS program cannot be enforced, and therefore 50,000 square feet or greater are covered by the mechanisms for monitoring are limited (Gupta et ordinance. Seattle enacted legislation in 2016 to al. 2007). reduce emissions from the building sector’s largest and least efficient commercial and multi-family In the case of Brazil, several important cities have buildings, requiring periodic retro-commissioning taken specific actions for climate change control (a comprehensive audit of equipment for improved going beyond Brazil’s national policy targets. For performance followed by low- to zero-cost instance, in 2009, São Paulo committed to undertake operational and maintenance improvements). New a significant GHG emission reduction target of 20 York City is implementing a large, mandatory building percent by 2020 compared to its 2005 levels (Lucon retrofitting program for energy efficiency, which is and Goldemberg 2010).46 The first Brazilian city part of its target to achieve an 80 percent reduction to pass legislation related to climate change was in emissions below 2005 levels by 2050. Other cities, Palmas, in the state of Tocantins, in 2003. Three like Detroit, Los Angeles, and Salt Lake City, have other cities that issued municipal climate policy similar programs involving transportation, renewable legislation in 2011 were Belo Horizonte (in the state energy, and building energy consumption efficiency. of Minas Gerais), Feira de Santana (Bahia), and However, as Victor and Muro (2020) report, as Rio de Janeiro (Rio de Janeiro) (Barbi and Ferreria of 2017, only 45 of the 100 largest cities had any 2013). As in the case of São Paulo, the cities of serious climate pledge at all, and about two-thirds Belo Horizonte and Rio de Janeiro have set specific of those are lagging in their targeted emissions cuts, deadlines and targets for reducing GHG emissions, while 13 others do not appear to have emissions plus Rio de Janeiro and São Paulo also include tracking in place. adaptation plans for the impact of climate change. Many states in the United States have been also In the case of the United States, federal government pursuing aggressive policies to fight climate change. policy to reduce carbon emissions has varied The lead was taken by California in 2006 with the markedly, both in intensity and direction, over the Global Warming Solutions Act, followed by nine different recent presidential mandates. For example, northeast states that associated into the Regional the Trump administration returned to the polices Greenhouse Gas Initiative (RGGI) to introduce a of President George Bush Jr. of opposing controls region-wide cap and trade system for emissions (Kolstad 2020). The vacuum created by federal from electricity generation. As reported by the Center inaction to control carbon emissions was filled for Climate and Energy Solutions (C2ES) (C2ES 46. Law No. 13.798 of November 2009 was enacted prior to the 15th Conference of the Parties to the UNFCCC in Copenhagen (COP15) following an elaborate public consultation process. The establishment of mandatory GHG reduction targets in a region of a non–Annex I country (within the UNFCCC framework) was a matter of political controversy involving repercussions for economic development and was quite innovative among developing countries, bringing comparisons to the role of São Paulo in Brazil with that of California in the United States (Lucon and Goldemberg 2010). Note that in the case of São Paulo, climate change actions were initiated by the state government and not just the city; in all other cases in Brazil, the policies were city government initiatives. 47. As Kolstad (2020) also indicates, there is a tradition for this regarding other environmental issues in the United States. For example, the 1963 Clean Air Act sets federal standards but delegates the regulation of pollution sources to the states. Adapting Fiscal Decentralization Design to Combat Climate Change 34 2020), 23 states plus the District of Columbia have subject of public interest,” and grants inherent rights adopted specific targets for GHG reduction involving to Mother Earth and its constituents (for example, carbon pricing, emission limits, energy efficiency the right to clean air and water). The implementation mandates and incentives, and the promotion of of some functions is delegated to SNGs. Besides cleaner transportation. the usual delegation of water, sanitation, and other services, municipal governments are active in the In Canada, there are broad-based emissions pricing forest sector, responsible for enforcing national policies across the entire country, but following other forestry laws and assisting local forest users in aspects of Canadian federalism, those policies show their compliance. a high degree of diversity in both policy design and implementation (Winter 2020).48 In terms of design, In Chad, the Ministry of Environment oversees the besides the differing exemptions, there is one design of policies related to energy, the environment, province using a broad-based carbon tax and two mitigation, and adaptation, while regional with cap and trade systems, and most have both governments are charged with the execution and a carbon tax plus an output-based subsidy system management of the national programs. (Canada has 13 province-level governments in total). Six provinces and one territory arrived at their own In Colombia, the national government, through system in a voluntary manner, while two territories the Ministry of Environment and Sustainable adopted the federal scheme, also voluntarily. One Development, is mainly in charge of implementing province experienced partially voluntary adoption and and enforcing environmental laws. At the subnational partially federally imposed policy, while the federal level, the Regional Autonomous Corporations (CARs) policy was fully imposed on three other provinces. and Corporations for Sustainable Development Earlier policy action by several provinces (Alberta, (CDS) are public environmental authorities British Columbia, and Quebec first and Ontario responsible for overseeing all the environmental later) led to federal action. Although the federal matters and natural resources within their jurisdiction government signed the Kyoto Protocol in 1997— and promoting sustainable development at the committing to a 6 percent reduction in emissions regional level. The CARs/CDS also have the power by 2012, relative to the 1990 baseline—in reality, to issue environmental licenses for small- and it moved slowly by focusing on sector-by-sector medium-sized projects. The Departments, as the regulations (Hoberg 2016). The already-existing political-administrative entity of the regional level, provincial policies on GHG emissions became the have residual competencies over the administration foundation and inspiration for the federal policy that of the environment. The Urban Environmental was later imposed on several provinces. Authorities have the same powers as the CARs, with the difference that they discharge them within urban A More Exhaustive Review of Expenditure municipalities, districts, or metropolitan areas with Assignments in Other Countries a population of at least 1,000,000 inhabitants. The municipalities, as the political-administrative entity at The practice with expenditure assignments in other the local level, also have residual competencies over countries is not easily summarized. The following the administration of the environment. paragraphs highlight how a long list of countries may fit into the three different models of expenditure assignment Ethiopia faces highly erratic hydration patterns of modalities reviewed above. intense rainfall and prolonged drought. The National Adaptation Strategy of Ethiopia (NAP-ETH), approved First, a good number of countries fit in the centralized in 2019, identifies the roles and responsibilities of mode, with leadership at the central level delegating regional and woreda (district) committees. The Inter- implementing responsibility to the subnational level: Ministerial Steering Committee is responsible for formulating and monitoring NAP-ETH at the highest In Bolivia, the Constitution sanctions the Law of Mother level. The regional bureaus are responsible for the Earth, which defines Mother Earth as “a collective implementation of the strategic adaptation options, 48. Expenditure assignments in Canada between the federal and provincial governments are stated in the Constitution, with exclusive and shared assignments, while residual (not assigned) responsibilities rest with the federal government. This latter is important because the Constitution is silent on environment responsibilities. However, following several Supreme Court decisions, de facto, environmental responsibilities are concurrent or shared. Adapting Fiscal Decentralization Design to Combat Climate Change 35 monitoring and evaluating regional-, zonal-, and In Austria, the federal government and the Länder woreda-level performance. The woreda offices (states) have some exclusive responsibilities are responsible for implementing, monitoring, and regarding the environment and climate change evaluating the woreda-level strategic options. policy and also share others. Following the national strategy, the states have introduced regional In Pakistan, the federal government’s EPA is the adaptation policies to fight the effects of climate main government body in charge of implementing change, including guidelines and the responsibilities and enforcing environmental laws. The provincial assigned to the municipalities. governments have their own EPAs that help enforce environmental laws in their jurisdictions. In Bulgaria, SNGs have substantial responsibilities regarding climate change action, including land In Peru, the national government, through the development, waste management, and programs Ministry of Environment, is the main government for the improvement of energy efficiency and the body in charge of implementing and enforcing reduction of GHG emissions. They also share environmental policies. The Environmental responsibility with the central government for Assessment and Control Agency (OEFA), part of the afforestation policies. Ministry of Environment, is in charge of overseeing the fulfillment of environmental commitments under In Finland, local governments implement nationally the Environmental Impact Assessments (EIAs) agreed emission reduction targets, but municipalities and regulations. However, regional, and municipal also network and work together (the Carbon Neutral governments enforce these laws and regulations in Municipalities [Hinku] network) to reduce GHG their respective jurisdictions. emissions more rapidly than the EU targets adopted at the national level. Other countries that also mainly follow a centralized approach include Algeria, Chile, Czech Republic, In Honduras, the Natural Resources and Estonia, Democratic Republic of the Congo, Egypt, Environmental Secretariat (SERNA) at the central Hungary, Ireland, Israel, Italy, Japan, Latvia, Lithuania, level leads climate change–related policy making New Zealand, Nicaragua,49 Namibia, Niger, Norway, in the country, and since 2012, the Regional Paraguay, Poland, Portugal, Russia, Slovakia, Slovenia, Development Councils (CDRs) have been charged South Korea, Sweden, Tanzania, Thailand, Turkey, with developing climate change programs, though it and Ukraine. is uncertain how effective they have been. Second, a smaller number of countries better fit the In Kenya, the National Climate Change Council model of mixed central and subnational leadership: (NCCC), chaired by the president, has oversight of the mainstreaming of climate change–related In Argentina, many of the environment-related functions at the national and county levels. At the responsibilities are distributed among several central national level, several ministries and departments ministries but coordinated through the Argentinean have established climate change units and climate National Climate Change Cabinet (GNCC, for its change–related plans and policies. All county Spanish acronym), which organizes the duties governments are required by the Climate Change among several ministries and all inter-jurisdictional Act of 2016 to develop climate change policies matters regarding different levels of governments. in their County Integrated Development Plans In addition, several provinces have established laws (CIDPs). County governments are also expected to for environmental protection in a variety of areas, establish Climate Change Units that will oversee the such as soil conservation, water, and forestry. More implementation of climate change actions. Counties generally, since there are no national minimum are responsible for: disaster risk management standards regulating waste management, each (response measures to address drought, floods, province has its own law establishing minimum and other climate-driven disasters); food and conditions and requirements for the storage of nutrition security; water and the “blue economy” hazardous waste. (water management and implementation of national 49. In the case of Nicaragua, citizens between 6 and 65 years of age have the duty to plant 10 trees a year in areas defined and prepared for that purpose by the municipalities Adapting Fiscal Decentralization Design to Combat Climate Change 36 government policies on water conservation); The Importance of Intergovernmental forestry, wildlife, and tourism; health, sanitation, and Coordination and Cooperation: Country human settlements; manufacturing; and energy and and International Examples transport. Counties have formed six economic blocs. The counties in each of the blocs were consulted The discussion of the assignment of functional while preparing the National Climate Change Action responsibilities for climate change policies in the previous Plan (NCCAP 2018–2022). The Climate Change section heightens the importance of intergovernmental Directorate (CCD) in the Ministry of Environment coordination and cooperation for effective policy making and Forestry is responsible for coordinating the as it has also been highlighted in the literature (see, for implementation of national climate change action example, Nelson et al. 2014).50 This section underscores plans. The county governments work closely with the utilization of different forums for intergovernmental the CCD. County governments designate a County coordination on climate work intra-nationally and Executive Committee member to coordinate climate internationally. In some cases, the mechanisms change affairs and report to the County Assembly. integrate vertical coordination with horizontal subnational coordination (i.e., among SNGs), and in some others, the In Mali, the Ministry of Environment, Sanitation, emphasis is exclusively on horizontal coordination. and Sustainable Development at the central government is mainly in charge of implementing and A good example of single-country horizontal cooperation enforcing environmental laws. However, each local is Finland. One form of this cooperation is the Association government is also responsible for the management, of Finnish Local and Regional Authorities (AFLRA), planning, conservation, and safeguarding of the which coordinates the Cities for Climate Protection ecological balance of its area. (CCP) campaign (Niinistö et al. 2017). The purpose of this campaign is to encourage municipalities to plan In Nigeria, the federal government is the main and initiate their own actions for reducing local GHG governmental organization in charge of establishing emissions. Another instance is the Hinku network, which environmental policies and enforcing them. However, brings together municipalities, as well as businesses and each state and local government in the country may citizens, with the goal of achieving reductions in GHG set up its own environmental protection unit, and emissions more rapidly than EU targets require (by 80 the states are also empowered to make their own percent by 2030) (Hinku 2020). In addition, and as an environmental laws. example of both vertical and horizontal coordination, Finland is one of the few European countries where In South Africa, the central government’s Department voluntary means, such as the voluntary Energy Efficiency of Environmental Affairs has exclusive competence Agreement scheme, have proven to work effectively and over natural resources but shares responsibility yield results. The Energy Efficiency Agreement for the with the provincial governments over most Municipal Sector is an agreement between the Ministry of environmental matters involving legislation, Economic Affairs and Employment, the Energy Authority, implementation, and enforcement. and the AFLRA on increasing the efficient use of energy in the municipal sector. Municipalities, cities, and joint Other countries that would seem to fit into the intermediate municipal authorities sign their own Energy Efficiency mixed governance model include Iceland, Romania, Agreement, in which they commit themselves to the Spain, Switzerland, and the United Kingdom. actions and targets specified in the Energy Efficiency Agreement for the Municipal Sector. Finally, no other countries, other than those already discussed above, would seem to fit into the Another example of vertical and horizontal cooperation decentralized model with bottom-up leadership in is the United States’ Clean Energy States Alliance climate change policies. (CESA), which created the “100% Clean Energy 50. These authors also argue that cooperative policies (like central transfers to SNGs) and coordination policies (like enabling and funding mechanisms) facilitate effectiveness and promote vertical interdependence, thereby increasing effectiveness, in contrast to the so-called competitive policies (like federal performance standards and price mechanisms), which are prone to increasing political conflict and decreasing effectiveness. Adapting Fiscal Decentralization Design to Combat Climate Change 37 Collaborative” in 2002. CESA is a national, nonprofit 2020). The program helps cities customize solutions coalition of public agencies and organizations working to decarbonization and their environmental needs; on together to advance clean energy. CESA works with joining the program, each city adopts its own Green state leaders, federal agencies, industry representatives, City Action Plan, including institutional capacity for and other stakeholders to develop clean energy implementation, monitoring, sustainable transport, “blue- programs and inclusive renewable energy (Leon 2020). green” infrastructure, and sustainable and efficient energy The Collaborative provides and shares information and and waste management.52 Recently, the World Bank technical assistance to states, and so far 14 states plus published its Outlook 2050, which presents strategies the District of Columbia have set 100 percent clean for supporting client countries in meeting their long-term energy goals for electricity, either through legislation goals of decarbonization, involving technical assistance, or governors’ actions. Another example of horizontal lending, country programs, and technical assistance, in cooperation in the United States is the U.S. Conference order to make climate action and country development of Mayors and the C2ES, which have built an Alliance policies compatible (see Mukhi et al. 2020). The EU also for a Sustainable Low-Carbon Future, involving the encourages municipalities to become greener through membership of all U.S. cities with over 30,000 inhabitants initiatives such as the Green Capital Award, the Green (UCM and C2ES 2020). Leaf Award, and the Green City Tool (EGCA 2020). In addition, SNGs are increasingly engaging with each Several other international associations have been other across national boundaries on the international recently active. The C40 Clean Air Cities Declaration stage, which is likely to have positive demonstration effects was launched in October 2019 with membership of on domestic action against climate change. One such cities across the world that have agreed to set pollution example is URBACT, a European exchange and learning standards that meet or exceed national targets by program promoting sustainable urban development. It 2025 and publicly report on their progress through helps share develop sustainable solutions for social and engagement with nation-states and businesses.53 The environmental issues. It is financed by EU funds and Under2 Coalition is a global association of SNGs aiming national and local contributions (URBACT 2020).51 to achieve GHG emissions mitigation.54 As of 2018, the list of signatories had grown to over 220 jurisdictions, Another example is the European Bank for which combined, encompassed over 1.3 billion people Reconstruction and Development (EBRD) Green Cities and 43 percent of the world economy. Another example program. Funded with €1.5 billion in 2016, the Green of an international association is the Compact of States Cities program has grown to cover 41 cities (EBRD and Regions. 51. See URBACT’s webpage, https://urbact.eu/who-we-are. 52. In addition, EBRD recently announced that it is scaling up its climate and environmental programs, devoting over 50 percent of its annual investments to the green economy by 2025, as well as targeting specific emission reductions over the next five years and setting a date for a decision on when all of its projects are aligned to the Paris Climate Agreement (Bennett 2020). 53. The cities so far that have signed up to the C40 Clean Air Cities Declaration are: Amman, Austin, Bengaluru, Barcelona, Berlin, Bogotá, Buenos Aires, Copenhagen, Delhi, Dubai, Durban (eThekwini), Guadalajara, Heidelberg, Houston, Jakarta, Los Angeles, Lima, Lisbon, London, Madrid, Medellin, Mexico City, Milan, Oslo, Paris, Portland, Quezon City, Quito, Rio de Janeiro, Rotterdam, Seoul, Stockholm, Sydney, Tel Aviv-Yafo, Tokyo, Warsaw, and Washington, DC. 54. Originally, it was created as a partnership between the governments of California and Baden-Wurttemberg, with the Climate Group acting as secretariat. Adapting Fiscal Decentralization Design to Combat Climate Change 38 04 Improving the Assignment In theory, SNGs can use their own revenue authority— where it exists—in order to penalize or reward, as the of Revenue and Tax Authority case may be, actions by businesses and households related to climate change.55 As shown later in this section, for Enhancing Action on there are many examples among OECD countries of Climate Change such policies. From the conventional theory of revenue assignments, one of the most important objectives is to be able to raise revenues as efficiently as possible, or in other words, with the lowest levels of “excess burden” losses. That goal is met when the behavioral response—or the elasticities—of taxpayers is relatively small. That is the one constraint SNGs have to raising revenues efficiently: the mobility of businesses or even households. Economic agents will consider relocation to another jurisdiction if, all other things being equal, they can manage to reduce their tax burdens. All these considerations play quite differently in the case of using taxes to combat climate change. As discussed further below, to correct for the externalities associated with climate change, the main type of EI utilized is to tax polluting activities with “Pigouvian taxes.” Although these taxes typically yield a “double dividend” in that they discourage the polluting activity and raise revenues, their main purpose is to influence behavior and not to raise revenues. Therefore, a good revenue-raising tax and a good Pigouvian tax have different properties. In particular, a general good (subnational) tax falls on goods and activities with limited elasticity, thus being less distortive and raising more revenues. In a Pigouvian setting, however, the goal is to maximize the elasticity of response, so that with small tax levels, for example in the case of carbon, emissions can be reduced as much as possible.56 55. The basic question asked in this section is how revenue assignments in fiscally decentralized systems can be best adapted to combat climate change. However, it is also very likely that there is some sort of reverse causation. That is, climate change events are likely to affect the revenue structure of SNGs by changing economic activity patterns, development, and tax bases. For example, Sanoh (2015) studies the impact of rainfall shocks on local government financing in Mali. 56. In this case, the reduction in economic activity is not an “excess burden” but rather a desired benefit. See, for example, Avner, Rentschler, and Hallegatte (2014) for a discussion of these points. Adapting Fiscal Decentralization Design to Combat Climate Change 39 Note that with both “good” subnational revenue-raising knowledgeable, responsive, and accountable to citizens’ taxes and also with Pigouvian taxes, economic agents needs and preferences. At the same time, there is a large (especially in the case of households) are likely to weigh consensus among experts in decentralization that the the benefits of additional general public services and increased accountability associated with decentralization also environmental services (such as a cleaner and safer can be assured only when SNGs have an adequate environment) against the burdens of any added taxes. level of autonomy and discretion in raising their own However, when the benefits are not localized, the quid pro revenues. This responds to a simple yet powerful logic: quo of involving local taxes is more likely to dissipate. In subnational authorities will spend resources differently those cases, the potential mobility of economic agents— (more responsibly and efficiently) when they need to and therefore of potential tax bases—is likely to impose incur the political costs of taxing their constituents, in significant constraints on well-intentioned subnational comparison to what they would do if the money comes to tax measures for climate change. Raising taxes in one them, practically cost free, from the central government location may just move the polluting activities to another in various forms of transfers. As part of a “common pool location. These situations call for using taxes at a higher problem,” when subnational expenditures are being level of government, such as regionally or nationally, financed by other taxpayers outside the jurisdiction, there or indeed, if possible, via internationally coordinated will be perverse incentives to overspend and/or to spend taxation across national boundaries. less efficiently because of the lower scrutiny over and accountability for spending funds that come from outside A second constraint for using taxation at the subnational than for funds that have to be raised in the jurisdiction level to combat climate change, and one of a more with higher political costs for decision makers. practical nature, is that SNGs in many countries, and especially in developing countries, often have very There are multiple trade-offs in designing the tax structure low levels of general taxation authority. This is a more at the subnational level. Given the different criteria that general problem that leads to substantial vertical subnational taxes should meet, the choice of particular fiscal imbalances, as SNGs often have much larger taxes comes down to a comparison of the advantages expenditure obligations than authority to raise their own and disadvantages that each of those taxes offer. Among revenues and thus have to rely on central government the desirable properties are: good revenue potential, with transfers to finance those expenditure needs. This low high elasticity or buoyancy (meaning that revenues will revenue or tax authority is likely to be more limiting grow proportionally with the economy); high adaptability for SNGs’ decarbonization actions, which, generally, to the benefit principle (so that those paying the tax are require some type of Pigouvian taxes, and less so for those using the public services); an even geographical adaptation policies, which can also be supported with distribution of tax bases (so that there is less disparity intergovernmental transfers. An additional obstacle for in per capita revenues across jurisdictions); fairness SNG action is the perception (or reality) that charges and equity in the distribution of tax burdens (those with for desirable household actions (e.g., improved housing higher incomes pay proportionally more); lower costs and energy consumption standards) will tend to have a of administration and compliance by taxpayers; higher regressive incidence, penalizing lower-income groups, acceptance and better perception about the tax among while SNGs have less ability to offset any regressive officials and the public; and high visibility of the taxes to effects via other tax or expenditure policies. facilitate greater accountability by public officials. Basic Principles of Revenue Among those properties that are negative, the most significant to be avoided is the inter-jurisdictional mobility Assignments of the tax bases, as highly mobile bases make it much harder for SNGs to imprint their own policies and may be conducive to inter-jurisdictional tax competition, potentially Subnational revenue autonomy is sometimes dismissed creating scenarios with competing jurisdictions engaging as an unimportant peripheral issue, but it is at the core of in rounds of lowering tax rates in order to attract large tax effective fiscal decentralization. The fundamental premise bases. Other negative properties to be avoided include of fiscal decentralization is that it has the potential to the inducement of large inefficiencies in the allocation increase the overall efficiency of the public sector and of economic resources, the high sensitivity of revenue lead to improved social and economic welfare. One collections to the fluctuations of economic business critical assumption behind those expected outcomes is cycles (since SNGs have less ability to finance deficits), that decentralized governments will generally be more the risk of incentivizing or facilitating corrupt practices Adapting Fiscal Decentralization Design to Combat Climate Change 40 among tax administrators and taxpayers, and the risk fiscal policy. At the same time, SNGs are reluctant to that tax burdens may be exported to other taxpayers take on the responsibility of making politically unpopular in other jurisdictions (causing less accountability and decisions to raise their own taxes, preferring central fiscal responsibility). government transfers as the mode of financing their expenditure needs. A situation of transfer dependence by In addition, there are some practical considerations in SNGs with large vertical fiscal imbalances is a common designing revenue assignments: outcome in many decentralized countries. (i) Prior establishment of expenditure needs is required Applying Basic Principles and Best since “finance should follow function” in an adequate manner, and because different services may call Practices of Revenue Assignments for for different forms of financing (fees or charges Climate Change Policy Action versus taxes).57 (ii) Providing SNGs with a closed list of subnational taxes is preferable because it avoids the introduction What do the theory of revenue assignments and the best of nuisance or even highly distortionary taxes. practices just reviewed say about the level of government (iii) Although central governments may prefer the that is better equipped to impose taxes, fees, and charges assignment of exclusive tax bases at different levels, for climate action? How should the different “green” the co-habitation of tax bases is generally preferable. taxes be assigned between central-, regional-, and This allows meaningful subnational revenue local-level governments to maximize decarbonization autonomy by providing access to some sizable and and adaptation activities? An important factor to keep in buoyant tax bases; if negative vertical externalities mind, as already discussed above, is that a good general (tax decisions at one level affect the tax bases and revenue-raising subnational tax and a good Pigouvian revenues of other levels of government), these can tax have different properties. In order to correct for the be addressed by intergovernmental transfers. externalities associated with GHGs, the main policy is (iv) Revenue autonomy for SNGs is best provided by to discourage the polluting activity via a Pigouvian tax, allowing them to set tax rates (between minimum and to the extent that pollution is measurable, the goal and maximum rates) and not by allowing them to is to set said tax equal to the marginal damage of the determine tax bases. Harmonized tax bases across external cost caused, thus creating the proper incentives jurisdictions reduce otherwise potentially large for the polluters (Cointrau and Hornig 2003; Alm and compliance and enforcement costs, and the decision Banzhaf 2012).58 Again, the main objective is not to raise of which level should oversee administering each revenues but to reduce the polluting activity. Of course, tax should be based on pragmatic considerations using green taxes begs the question of what level of regarding information and scale. In particular, government should be responsible for these taxes within some subnational taxes can be best administered the general legislation on revenue assignments.59 If the by the central tax administration and still remain a externality is confined to a local area, the responsibility subnationally determined tax. for the tax should be local. As the area of the externality increases to the regional or national level, then the proper Despite the benefits of subnational tax autonomy, the assignment of tax powers should be regional or national. observed international practices, with some exceptions, have low levels of tax independence. This would appear As already discussed in the introduction, potentially the to respond to powerful political economy incentives biggest constraint on subnational taxation is the mobility resulting in a perverse equilibrium between the two main of tax bases—firms and households may move out of players involved. Central governments are reluctant to jurisdictions with taxes above the average. Thus, SNGs devolve taxing powers for fear of having to compete with need to consider the strategic responses and interactions SNGs for the same tax bases and of losing control over with neighboring jurisdictions in competing for mobile tax 58. The alternative is a subsidy on pollution abatement that acts as a “carrot” (as opposed to the “stick” represented by the tax). Even though it is politically more attractive, it has the significant negative side of incentivizing more pollution. There is also the issue of where the funds for the subsidy will come from. Thus, clean-up subsidies are not an attractive policy strategy. Note that there are other (non-tax) government revenue-generating instruments, such as charges, auctioned permits, or even reductions in subsidies. 59. However, within the traditionally locally assigned taxes, there may be room for some “green surcharges.” For example, under the property tax, buildings that do not meet certain environmental conditions could be taxed at higher rates, or alternatively, those that do meet the restrictions may be taxed at a lower rate or be granted credits. The same applies for local motor vehicle taxes. Adapting Fiscal Decentralization Design to Combat Climate Change 41 bases.60 Again, this is not a certain outcome, especially if The next paragraphs more closely cover the specific there is a balance in the fiscal exchange and the benefits forms of taxation that can be used for decarbonization are perceived to offset the higher taxation costs. As green activities and also potentially help with the financing of taxes yield a double dividend (a cleaner environment and adaptation activities. increased budget revenues), these additional revenues can be used to produce more and higher-quality public Decarbonization activities. Carbon taxes are among the services or to invest in adaptation investment measures most popular decarbonization tax measures, but would that may reduce the incentive to out-migrate. they work as decentralized government instruments?62 There are some considerations in favor of their There are some other additional factors conditioning decentralization. Carbon taxes not only offer incentives green revenue assignments. Following the principles of to keep the cost of decarbonization down, but a certain taxation discussed above, many countries will provide level of flexibility may be desirable to enable SNGs SNGs with a closed list of taxes. This means that SNGs to design suitable policy frameworks specific to their must be a willing to use the “green” taxes, and the local circumstances, which may be harder for national national authorities must be willing to grant those powers authorities to do because they lack information about the in the closed list of taxes for SNGs. Both the willingness different types of polluting activities and so on. It is in by SNGs to use the taxes and the willingness by central this context that Metcalf and Weisback (2009) argue that governments to devolve that authority are generally carbon taxes can be fully or partially delegated to SNGs scarce in actual international practice.61 so that tax rates can be set to fit the differing costs of carbon emissions and abatement. Some technical issues may also act as binding constraints. Even though emission taxes may be more On the other hand, there are several considerations that administratively efficient to collect and harder to avoid work against the decentralization of carbon taxes or other than, say, income taxes, some forms of environmental similar instruments, such as cap and trade. As already taxation are quite complex to administer. These difficulties discussed above and emphasized early on by Hahn can be compounded when taxpayers are sophisticated and Noll (1981), subnational authorities would need to global multi-unit businesses that practice in different understand the pre-regulation pattern of emissions and venues for tax avoidance, such as through transfer the abatement opportunities available and also have the pricing and profit shifting. It is likely that many SNGs ability to monitor emissions and pollution levels—again, may not have the required high levels of sophistication requirements that are unlikely to be met, especially in in tax administration to overcome such behaviors by the developing countries that lack that type of administrative tax base. capacity. There is also the threat of subnational tax competition yielding a suboptimal equilibrium with low Another important technical constraint is related to the “green taxes,” and the resulting outcome may be a ability to measure and monitor pollution emissions levels. suboptimal, perverse equilibrium in which a large number Again, it is likely that many SNGs may lack this capacity of SNGs introduce lower than optimal tax rates. even in developed countries. However, in a number of countries, large, sophisticated subnational agencies do, Beyond the reasons just listed, and the generally large in fact, undertake emission monitoring (for example, in level of externalities involved with emissions, there South Korea). In addition, the capacity constraint is most may be other important reasons why the federal or relevant to emissions taxes and not to carbon taxes on central governments should be the level administering goods and services that are GHG intensive (such as gas carbon taxes in a federal or decentralized system of taxes). In the latter case, there would seem to be more finance. From a tax administration perspective, a central significant scope globally for the further assignment of government tax would be much easier for tax officials such taxes to SNGs, including in developing countries. to administer and enforce and easier for taxpayers to 60. There is some empirical evidence on the mobility of factors of production and households in response to taxes, but the extent to which this is a constraint for SNGs in using taxation to control pollution, GHG emissions, and so on is more of an open question. The evidence of subnational tax competition and “races to the bottom” is reviewed in Wilson (1999). 61. Typically, however, in the case of non-tax policies, such as licenses and fees and charges for forestry, small mining activities, and so on, SNGs are granted much more freedom 62. More specifically, for example, Ahmad and Stern (2009) discuss, in the contexts of India and Pakistan, the possibilities of implementing a carbon tax, assigning it to the federal/central level, and levying it as an excise tax at the production or import stage (as a specific, as opposed to ad valorem, excise tax), based on quantities imported and produced domestically as against a state/regional tax at the final point of sale. Adapting Fiscal Decentralization Design to Combat Climate Change 42 comply with than a multitude of potentially different But even if carbon taxes may not be the optimal instrument subnational taxes (Ahmad and Stern 2009). Also, carbon for subnational revenues, there are many other non- taxation is likely to generate an uneven distribution tax revenue sources that can be used to encourage of costs among SNGs, firms, and households.63 If climate-friendly actions by citizens and businesses. distributional considerations matter, and they always These involve fees and charges for the preservation of do, federal/central collection of revenues would make it the environment, involving construction, transportation, simpler to introduce compensatory measures for SNGs forestry, and so on.65 that would be relative losers—for low-income groups or for selected businesses. The introduction of carbon A particularly relevant issue in subnational “green” taxes may have to be accompanied by other changes in taxation is whether the revenue from green taxes and redistribution policies, including direct transfer programs fees should be earmarked to action and programs related or through changes in direct taxes (Poterba 1991; to the decarbonization of climate change. For example, Callan et al. 2008; Ahmad and Stern 2009; Gago et al. the state of Georgia in the United States has debated in 2020). The potentially uneven distribution of the costs recent years whether to introduce a (state) constitutional of carbon taxes among SNGs can be an obstacle to amendment to force the state to spend “green” fees (for their introduction that could be addressed by changes tire disposal and landfills) on environmental cleanup in the current system of intergovernmental transfers programs. Earmarking, in general, is not considered (Tang 2011).64 There may also be market considerations to be a good budgetary practice, because all spending requiring government intervention, because carbon taxes programs should be considered simultaneously at budget and tradable emission permits could have a different time and compete for the scarce resources available. effect on firms, for example, benefiting larger firms with However, there are exceptions to this general principle, monopoly powers (Noll 1982). such as gasoline taxes used for highway maintenance, which work as a user fee substitute that is easier to collect. 63. See, for example, Bureau (2011) for an estimation of the distributional effects of a carbon tax in France. 64. Tang (2011) argues that uneven distribution of carbon taxes across provinces in Canada has stalled progress on this reform. 65. However, SNGs might and do levy related fees and charges more with revenue collection and financial benefits in mind rather than focusing on the reduction of emissions or the preservation of ecosystems. See, for example, Nurfatriani et al. (2015) for a critique of Indonesia’s local government forestry fees. In this regard, more generally, the International Monetary Fund recently recommended that policy makers use “feebates,” which are defined as sliding scales of fees on activities or outputs (such as transportation, power generation, agriculture, or forestry) with above average emission rates, and subsidies or rebates for those with below average emissions (Zhang, T. 2020). Adapting Fiscal Decentralization Design to Combat Climate Change 43 The purpose of taxation would also seem to be a valid by using tax increment financing and other land value argument for keeping the uses of revenues from green capture measures. Indeed, land value capture taxes taxes and fees for environmental and climate change and fees are increasingly being used as revenue raising purposes. However, the main effectiveness of carbon instruments to implement many types of adaptation taxes, like other green taxes and fees, in discouraging infrastructure developments (see, for example, Ingram activities that produce GHGs is realized independently of and Hong 2012; Smolka 2013; Grafakos et al. 2018; how those revenues are later utilized. Kozak et al. 2020). This is one of the promising trends the next subsection will touch upon. Adaptation activities. These activities may include the building of certain types of infrastructure that can limit What Are Some Current Effective and economic damages down the road due to climate change. They can be financed via transfers, especially when Less Effective Practices in Revenue geographical externalities are present, or through own Assignments in Different Countries? revenues. In addition, taxes and other charges can be used by SNGs to deter certain decisions and behaviors that will tend to increase the costs of adaptation. For One of the most conspicuous features of the current example, higher taxes and fees can be imposed on state of taxation policy for climate change is its significant home builders in flood plains if outright forbidding them underutilization. Revenues from environmental taxes, via zoning does not work. In general, with adaptation including carbon taxes and all other levies on products activities, the question is not so much one of deterrence and activities harming the environment, are generally or creating disincentives to particular activities, but low even in OECD countries. As of 2016, on average, instead how to generally help finance adaptation work in OECD countries were raising 1.63 percent of GDP from response to natural disasters and so on. To the extent that environmental taxes, representing only 5.3 percent of total the consequences of climate change–related disasters revenues and US$621 per capita (OECD 2019). Further, tend to be more localized, there is more room for using taxes on polluting fuels currently appear to be nowhere the benefit principle for subnational financing. That would close to the levels needed to encourage a significant shift call for the use of fees and charges as well as traditional toward clean energy. As reported by the OECD (2019) subnational instruments, such as real estate property in figure 2, 70 percent of energy-related CO2 emissions taxes, betterment levies and other land value capture from advanced and emerging economies are entirely measures, and property transfer taxes. One issue to take untaxed, and some of the most polluting fuels remain into account is the incidence and distributional impact of among the least taxed. Thus, before describing some these measures. For example, it tends to be the case that current practices in the assignment of tax authority to fight the areas most prone to natural disasters in metropolitan climate change, it is important to notice that, generally, areas are more likely to be slums inhabited by low- regardless of the actual assignment, there is ample room income households with little access to more desirably for both central governments and SNGs to make much placed housing. Applying the benefit principles under more extensive use of green taxation. These overall low those conditions certainly would be questionable and levels of green taxation nationally, in combination with the likely hardly effective, since the ability of those residents generally large vertical imbalances most decentralized to pay would be low. countries experience—that is, the asymmetry between the much larger expenditure decentralization versus the A more recent innovation in the financing of adaptation much smaller revenue decentralization or tax autonomy policies merits special mention. SNGs can use taxes of SNGs—leads us to anticipate generally low levels of and fees that capture the value created by installing green taxation by subnational decentralized units. green infrastructure, such as storm water management, Adapting Fiscal Decentralization Design to Combat Climate Change 44 Figure 2. Current Taxes on Polluting Fuels versus Ideal Level Average effective carbon tax rates on non-road emissions* fof selected countries 0 10 20 30 40 Brazil low-end carbon price benchmark US Chile Australia Germany Japan France Italy Netherlands * Electricity generation, manufacturing, heating, aviation, and maritime transport. Source: Saint-Amans (2020). In a relatively large number of countries, just about all In Belgium, the Flemish regionally established green tax decisions and sources are centralized, and the climate fund draws its revenues from the auction of overall level of green taxation varies quite considerably.66 emission allowances (under the EU’s ETS). In another group of countries, SNGs have been assigned traditional levies that have a potential bearing on green In Brazil, most states and large municipalities taxation, such as taxes on motor vehicles, road taxes, operate environmental funds partly fed by revenue and water supply, sewage, and waste charges, but these from environmental fines and oil and gas revenues. generally are not linked to emissions or environmental performance, and SNGs have not been granted any In Canada, the Pan-Canadian Framework authority over revenue-yielding non-traditional green Implementation is supported by investments from levies.67 In addition, many of these countries have some the federal government to meet the 2030 emissions form of revenue sharing of green national taxes with reductions target and consists of several pillars, SNGs and here again, in this group, the overall level of the main one being the Pricing Carbon Pollution green taxation varies considerably. arrangement. In October 2016, the government of Canada introduced a carbon-pricing benchmark that The text that follows identifies green revenue assignment allows provinces and territories to implement the practices in different countries that perhaps could be pricing system of their choice (a carbon tax, a carbon imitated elsewhere to strengthen the role of SNGs in levy and performance-based emission system, or a fighting climate change from the revenue side of the cap-and-trade system). The revenues from these budget. As mentioned above, these and all green taxes pricing systems are to remain in the jurisdiction of provide the double dividend of local budget financing and origin. For example, British Columbia’s “Sustainable the decarbonization of local economic activity. Environment Fund” gets financed by a tax on batteries and tires. 66. These include Argentina, Estonia, Chile Czech Republic, Finland, Greece, Hungary, Iceland, Ireland, Japan, Mali, Mexico, the Netherlands, New Zealand, Portugal, Serbia, and the United Kingdom. 67. Among these countries are Bolivia, Bulgaria, Colombia, India, Honduras, Kenya, Romania, Russia, and Switzerland. Adapting Fiscal Decentralization Design to Combat Climate Change 45 In China, provincial governments have the authority In Slovakia, there is a local air pollution tax. to set the tax rate for coal production activities within a range of 2 to 10 percent, which is legislated by the In South Korea, local governments can levy an central government. ecosystem conservation charge (with half of the revenues shared with the central government) In France, both municipalities and departments can on development projects that have a substantial charge a tax on electricity consumption: €0.250 per environmental impact. megawatt hour, multiplied by a factor between 0 and 8 for the municipal part of the tax and between 2 and In Spain, regional governments levy taxes on 4 for the departmental part. air pollution emissions, disposable plastic bags, industrial waste, environmental damage caused by In Indonesia, provinces and local governments have large commercial establishments, and the storage of been assigned a list of environment-related taxes. spent nuclear and radioactive waste. They also levy a fee paid by enterprises applying for the granting, In Italy, municipalities and provinces can levy an modification, extension, and renovation of the additional tax on electrical energy. European Ecolabel. In Latvia, local governments can levy an air pollution In Sweden, local governments can levy an excavation tax (and are able to keep 60 percent of the revenues) charge and a license fee for the exploitation of peat. and a tax on the imports of radioactive substances (in this case keeping 30 percent of the revenues). In the United States, many states have introduced their own green taxes. For example, California has In Lithuania, local governments receive 70 percent introduced an electronic waste recycling fee and of the air emission non-compliance fees charged on hazardous substances taxes; the state of Illinois emission amounts exceeding standard limits. has a dry-cleaning license tax and an electricity excise tax; and in Indiana, there is an aircraft license In Norway, the local government of Svalbard excise tax, a hazardous chemicals inventory fee, charges an environmental fee to all tourists visiting an underground storage tank fee, and a waste the jurisdiction. tire management fee. In addition, many of the federally levied green taxes are shared with state In Poland, there is a local charge for bush and tree governments. removals, and the revenues from the tax on air pollution are shared by all levels of government. Adapting Fiscal Decentralization Design to Combat Climate Change 46 05 Redesign and Innovations in From a first look, intergovernmental fiscal transfers appear to be the most suitable public finance instrument for Transfer Systems to Combat internalizing spatial externalities associated with climate change including for financing adaptation programs, Climate Change especially those showing inter-jurisdictional externalities. Transfers, subsidies, and direct performance contracts and agreements between the central government and SNGs represent the most versatile and direct set of tools to influence the subnational policies to fight climate change. As in the case of expenditure and revenue assignments, in order to assess the potential role that transfers, subvention, and agreements between different levels of government may play in devising more effective policies for climate change, it is helpful to briefly review the general theoretical principles for transfer design as well as the most common challenges observed in practice. Basic Principles of Transfer Design Transfers and grants are important components of the revenue architecture for SNGs. They are necessary to help close vertical fiscal gaps due to the inadequacy of own tax sources in revenue assignments and horizontal fiscal imbalances stemming from the different fiscal capacities and expenditure needs across SNGs at any level. Transfers are also important for pursuing national priorities through SNG actions, addressing different types of externalities that SNGs would otherwise not take into account, financing certain expenditure programs on behalf of the central government and specific investment programs, incentivizing better subnational expenditure performance, and so on. Adapting Fiscal Decentralization Design to Combat Climate Change 47 The most common mistake made in transfer design is to revenue sources and tax sharing result in horizontal pursue multiple—sometimes contradictory—objectives fiscal imbalances. These horizontal imbalances can with a single instrument. This results in confusion about be enlarged when demographic composition and other the objectives pursued and makes monitoring actual circumstances create further disparities in expenditure performance of the transfer difficult or impossible. The needs. Horizontal fiscal disparities are undesirable on best international practice is to follow the general rule both equity and efficiency grounds. If not addressed, of using one separate instrument (transfer) for each citizens in different locations would have different access distinct objective. to basic services, which may lead to undesirable fiscal migration (as opposed to economic migrations for job The most typically used types of intergovernmental opportunities, etc.). transfers include revenue sharing, equalization transfers, conditional block and specific grants (including capital The commonly accepted objective of equalization grants), conditional block and specific subventions, and grants is to allow all SNGs to provide similar access performance-based grants. The latter utilize ex post type to all residents to a standard package of public conditionalities, as opposed to the ex-ante conditionalities services, exercising an average level of tax effort. used in the case of conventional conditional grants. Next The basic architecture of equalization grants includes is a review of the basic structure and objectives of each the determination of the pool of funds and an of those main categories. allocation formula. The former determines the actual degree of equalization to be achieved, and the latter Revenue sharing arrangements provide SNGs with typically incorporates differences in fiscal capacity a percentage of certain central tax revenues. Their and expenditure needs across SNGs. Equalization allocation is generally implemented on a derivation basis funds are also generally unconditional and part of (shared with the jurisdiction where the central revenues general subnational revenues. This, and the general are collected), although other allocation formulas are principle of one separate instrument for each separate also used, such as by population or per a weighted index objective, also makes equalization transfers generally with several variables. The two most important objectives unsuitable for the pursuit of environmental and climate pursued with revenue sharing are that of helping to close change objectives.70 existing vertical fiscal imbalances and that of “devolution” (i.e., to provide incentives to develop regional economies Conditional transfers are generally used to incentivize or and a sense of national balance and cohesion rewarding mandate certain types of actions from, or spending by, economic activity).68 Revenue sharing funds are typically SNGs.71 These transfers come in many different shapes provided unconditionally, so that recipient SNGs can use and forms and because of their design flexibility are them as part of their general discretionary budgets—it is ideal instruments to pursue environmental and climate in this sense that they are expected to help close existing change objectives at the subnational level. The main vertical gaps.69 This feature, the unconditional use of types of conditional grants include block grants (with funds, reduces the attractiveness of revenue sharing general ex ante conditions that the funds must be spent as an instrument to pursue environmental and climate in some general functional area, for example, improving change objectives. the environment) and specific grants (with much more detailed ex ante conditions for how the funds must Equalization transfers are generally needed because the be spent, for example, a sewage treatment plant in a level of economic activity and wealth differs markedly particular location). Both types may or may not include a across subnational units, and the assignment of own matching clause, requiring the simultaneous contribution 68. Revenue sharing following devolution objectives clearly benefits the most economically dynamic regions of the country. But the inequality or horizontal fiscal imbalances created can be addressed by a strong equalization grant that recognizes the revenue-sharing funds as part of the fiscal capacity of SNGs, in turn reducing, if not eliminating, the equalization funds eventually received by those areas favored by the revenue-sharing arrangements. 69. Nevertheless, there are countries (like Argentina, Brazil, or India), that attach multiple objectives, including equalization, to revenue-sharing grants. The multiple objectives are pursued through different variables and factors used in the sharing allocation formulas. 70. Still, adequate transferred equalization funds can be quite crucial to facilitating the financing of adaptation policies by less well-off SNGs with lower fiscal capacity. 71. They may also be known as conditional grants, subsidies, or subventions. Agreements between the central government and SNGs are also generally financed by conditional transfers. Conditionalities typically specify the function for which and how the funds are to be spent. These may involve—non-exclusively—financial conditions (like matching requirements), administrative conditions (like production and service delivery conditions), and informational conditions (like data on impact and effectiveness). Adapting Fiscal Decentralization Design to Combat Climate Change 48 of funds by the recipient SNG. Finally, the type of SNGs to address externalities. In terms of design, capital conditionality attached can be ex ante or ex post—the grants can either be specific project-based grants or latter as in the case of performance-based grants.72 categorical or block grants, which give more discretion on how the funds can be spent.73 And in terms of their Conditional grant design should always start with distribution, capital grants can be allocated by ad hoc the fundamental question of what type of instrument decisions and negotiations by using pre-established would best achieve the goal being sought. From this formulas or competitive processes with defined perspective, it is important to recognize that there is no application procedures. Finally, as in the case of most absolute dominance of one type of conditional grant over other transfers, incentives matter in the design of capital others, for example, of block grants over specific grants grants. One overriding common concern is to achieve or performance-based grants over ex ante conditional “additionality,” ensuring that the grant does not simply grants. Because conditional grants limit the (spending) substitute for SNG funds. Matching fund arrangements autonomy of SNGs, general block grants are generally are used for this purpose. But the concept of additionality preferable. However, specific purpose grants (with ex should not be erroneously interpreted to mean that ante conditionality or ex post performance-oriented transfer funds would be made available only if the project conditionality) may be what is needed, depending on were not completely or entirely viable or could not be the policy objectives being pursued. For example, financed by any other means.74 when central governments try to address externalities (negative and positive) across SNGs, a specific grant Performance-based grants use ex post conditionality may be appropriate since it offers greater assurance that and are increasingly employed by central authorities to the central governments’ preferences will be respected incentivize improvements in service delivery and overall during implementation. SNG performance. The amount of funding is linked to performance according to predetermined criteria. It is worth highlighting two types of conditional transfers Incentives can be provided by: 1) granting access to the with relevance to climate change policies: capital grant scheme based on performance; 2) increasing or transfers and performance-based transfers. decreasing the grant amount based on performance; and 3) allowing more discretion in the use of the grant Regarding capital grants, the international practice resources, also based on performance. The most crucial is generally to use distinctive separate capital grants design area in this type of grant lies in the specification or transfers in support of SNGs’ needs to build public of the performance criteria. Often, the performance infrastructure, including climate change adaptation measures—which should always be linked exclusively to efforts. Even though all budget priorities should be actions for which SNGs are genuinely responsible—are considered simultaneously in the allocation of scarce related to institutional dimensions, such as in revenue fiscal resources, given their “lumpiness” and non- collection efforts, planning, financial management and recurrent nature, the needs for capital infrastructure reporting, or governance issues. Performance can typically require specialized instruments, different from also relate to service delivery as measured by outputs (recurrent) equalization transfers or ordinary conditional and outcomes. Typically, access to the program is grants. Capital grants generally complement operating linked to meeting some minimum conditions. Many surpluses and borrowing as the financing sources of the intergovernmental agreements or contracts, as used in capital infrastructure needs of SNGs. countries like Australia and the United Kingdom, can be interpreted as a variation on performance-based grants, In terms of objectives, capital transfers may seek to close since the payment under contract is typically tied to differences in infrastructure stocks across SNGs, support delivery or performance criteria. central governments’ specific sectoral objectives, or help 72. Conditional grants are also used to finance delegated functions (central authorities are still responsible for the service but use the SNG as an agent for its delivery). These are known as “pass through,” “agency,” or “cost-reimbursement” grants. Overall, given the many forms of fund transfers, the types of distribution approaches, and the sorts of control that can be imposed, there may be over 60 types of possible transfers. As a reaction to the overuse of specific grants (often small, costly to administer, and with high compliance costs and often contradictory objectives), there has been an international trend over the past several decades toward the simplification of transfer systems by increasing the share of general-purpose block grants over specific grants. 73. Two common design biases are to believe that capital expenditures are always more efficient than recurrent expenditures and to ignore the future maintenance requirements of new infrastructure. Matching arrangements are frequently used to induce subnational ownership and proper maintenance of existing infrastructure. 74, See World Bank (2018) for a discussion of this issue. Adapting Fiscal Decentralization Design to Combat Climate Change 49 Applying Basic Principles and Best of SNGs in many countries in areas such as housing and transport, it would seem that designing those existing Practices of Transfer Design for transfers in climate-sensitive ways may have significant Climate Change Policy Action net climate impacts. The “one objective per transfer” principle of transfer design can still be largely respected if transfers are designed to also encourage or further The generally desirable unconditional use of funds in incentivize the use of green means of transportation equalization grants and revenue sharing makes these or house building—in other words, altering how the types of transfers generally not well suited to coordinated objective is achieved, not the objective itself. On the other climate change policy action among central governments hand, there could be cases in which embedding climate and SNGs.75 This does not mean that it cannot be considerations in the transfer design may go too far done, but rather that it is much more desirable to utilize and pervert the main objective of the transfer. This may separate instruments for each objective, so that the have been the case in India’s equalization grant design effectiveness of each instrument as well as the policy (discussed further below) when it included the “forest tree impact are more transparent and easier to evaluate. coverage” as an important additional expenditure need In contrast, conditional grants, for recurrent or capital in the equalization formula. In this case, preserving the purposes and with ex ante or ex post conditionality, are wholeness of “true” expenditure needs and fiscal capacity generally well equipped for designing and implementing calculations would be primordial for equity and fairness climate-specific transfers to SNGs.76 in the design of the equalization grant. Introducing “forest tree coverage”—a worthwhile objective in itself—as Despite the general desirability of the design principle an expenditure need of the state governments likely of “one objective per transfer,” the embedding of climate distorted the true fiscal needs balance and led to unfair considerations into transfers that are targeted primarily results for poorer states. That policy also may have at other purposes may not always be a bad idea. Under had less merit, because using other conditional grants what circumstances might embedding climate issues be outside the equalization grant was certainly possible in desirable? Given the extensive expenditure assignments that country. 75. An entirely different issue, somewhat addressed above, is whether the expenditure needs of SNGs for mitigation and adaptation policies related to climate change should be considered in the formulas utilized for the distribution of existing equalization grants. The answer to this question is decidedly affirmative if those responsibilities have been assigned—as is commonly the case—to SNGs. For further discussion, see Ring (2002). 76. One question to ask is when to use transfers versus when to use taxes to address climate change issues at the subnational level. Although for simple political economy reasons, subnational jurisdictions may prefer transfers, levying green taxes and the like has the additional benefit of generating higher SNG accountability. Adapting Fiscal Decentralization Design to Combat Climate Change 50 Perhaps the general rule is the following: Embedding promote green procurement practices.79 In Germany, the climate change considerations in the design of fiscal federal funding scheme “Municipal Directive for Climate transfers is desirable when there is complementarity and Protection in Social, Cultural and Public Institutions” no distortion in the goals of the transfer (as, for example, targets local governments and institutions under their in the case of the urban transportation conditional jurisdiction, such as schools, to improve energy efficiency grant discussed above), but it is not desirable when the in municipal buildings, transport and mobility, water, goals or objectives of the transfers may be distorted sewage, and the management of municipal enterprises.80 (as in the case of the equalization grant in India and the insertion of “forest tree coverage” as an additional Adaptation activities. How to incentivize SNGs to spend expenditure need). on climate adaption? Conditional transfers can be used to compensate local governments for introducing Decarbonization activities. How to incentivize SNGs environmentally cautious zoning. Capital block grants to spend on climate decarbonization? Conditional can be designed to promote other climate adaptation transfers can be used to encourage SNGs to introduce policies, for example, investing in large projects, protective land-use restrictions and to incentivize such as seawalls and flood prevention landscaping different conservation activities within their jurisdictions. and infrastructure. Conditional transfers are also suitable to compensate local jurisdictions for internalizing different types of In the United States, the federal government has a externalities, such as the United Nations Capital subsidy program to help local governments (implemented Development Fund (UNCDF) Local Climate Adaptive through a federal agency, the Army Corps of Engineers) Living Facility conditional grant program. combat and prevent the damaging impacts of climate change. The Army Corps subsidizes local governments In particular, matching transfers can be used by central to purchase homes highly prone to flooding and have governments to incentivize SNGs to provide more them torn down. In recent years, the federal government adequate levels of environmental protection, even in has imposed a condition that local governments must situations where inter-jurisdictional competition for use “eminent domain” to force people out of flood- geographically mobile resources may coerce the lower prone homes or otherwise forfeit the funds the federal governments to ignore emissions and keep their green government would allocate locally to combat climate levies low (Oates 1999; Alm and Banzhaf 2012).77 change. Some local governments have agreed to these new terms and others have not.81 Federal transfers Subsidies from the central government to incentivize appear to play an important role in how SNGs respond specific actions by SNGs are just another form of to natural disasters. Miao, Hou, and Abrigo (2018) conditional transfer and are used, for example, in the analyzed data covering over 40 years and found that the Netherlands to improve municipal buildings, increase response by U.S. states following natural disasters is sustainable energy usage, smooth traffic flows, and so fundamentally determined by the size of the federal relief on.78 Performance-based grants can be designed for transfers received.82 a variety of decarbonization activities, for example, to 77. See also Silva and Caplan (1997) and Nagase and Silva (2000) for formal discussions of how central governments can incentivize SNGs to take inter-jurisdictional externalities into account even when central authorities have limited information on the pollution processes and abatement costs but can observe pollution outcomes. 78. Some types of green subsidies, especially to businesses, have the potential of running afoul of rules of the World Trade Organization, as discussed by Charnovitz (2014). 79. See, for example, Kollner, Schelske, and Seidl (2002) for a discussion of how conservation performance measures regarding biodiversity can be inserted into the distribution formula of already existing transfers in Switzerland. 80. Up to now, more than 11,500 projects in roughly 3,000 German municipalities (out of approximately 12,000 in total) have been subsidized with around €525 million from the National Climate Initiative (GIZ, UN-Habitat, and ICLEI 2017). 81. Part of the official mission of the Army Corps of Engineers is to offer protection from flooding and coastal storms by building levees, seawalls, and so on, and typically it takes responsibility for two-thirds of the costs, while local governments have to assume one-third. In order to decide what houses to condemn for demolition, the Army Corps of Engineers compares the estimated cost of flood damage over the next five decades to the cost of tearing the house down, plus other expenses, such as moving the owners (Flavelle 2020). 82. FEMA in the United States coordinates the response to disasters that overwhelm the resources of local and state authorities. The governor of the state in which the disaster occurs must declare a state of emergency and formally request federal assistance. Other countries have similar arrangements. For example, if an event in Indonesia is declared a national disaster, there will be direct funds from the central government to the affected region, and the funds are managed by the National Disaster Relief Agency (BNPB). Adapting Fiscal Decentralization Design to Combat Climate Change 51 What Are Current Effective and Less Overall, international experience offers a wide array of other types of intergovernmental transfers pursuing, for Effective Practices in Transfer Design in the most part, decarbonization objectives and, perhaps Different Countries? surprisingly, much less so in terms of established transfer programs in support of adaptation objectives, as opposed to ad hoc aids following catastrophic events. This discussion begins with the example of the inclusion of environmental measures in the design of India’s Conditional grants are being used in some countries equalization grant.83 The current weighted index formula for forest conservation programs and, complementarily, for this grant uses a couple of variables (population and considerable research has been conducted analyzing the land area) that can be interpreted as approximating role of intergovernmental transfers for local deforestation the states’ expenditure needs: one variable estimating and forest degradation management. Much less work, fiscal capacity (the deviation from the highest per however, appears to have been conducted on the role capita income) and another encouraging environmental of conditional transfers regarding other dimensions of conservation (the extent of forest areas in the states). climate change policy. In Brazil, the ecological “ICMS”87 This last criterion of “forest cover” was introduced by was introduced in the 1990s by several states as the 14th Finance Commission in 2015, because the part of their revenue-sharing formula with their local Union government mandated that it give consideration governments based on ecological indicators. States in to the ecology, environment, and climate change in Brazil are under a federal mandate to share 25 percent the sharing of federal revenues with the states.84 The of the ICMS collections with their municipalities, with issue in this approach, however, is that even though three-fourths shared on a derivation basis and the other distributing resources to encourage forest conservation one-fourth according to any specific state rules. In 1989, is clearly a positive measure against climate change, it is the state of Paraná introduced the ecological ICMS by doubtful that it really belongs in an equalization grant as distributing that one-fourth of resources according to opposed to a separate conditional transfer.85 In this last the percentage of preserved land that the municipality respect, India’s Finance Commission already designs had set aside. Other states followed suit: São Paulo grants for disaster management from the National (1993), Minas Gerais (1995), Amapá (1996), and so on. Adaptation Fund on Climate Change (NAFCC) and Currently, 18 of the 26 Brazilian states have adopted could design an additional one for forest conservation. the ecological ICMS.88 One good practice aspect of the However, from a political economy angle, the attraction ecological ICMS is that it keeps the different objectives— in India and elsewhere of using the equalization grant devolution via destination basis and land preservation— or other large existing transfers as a vehicle to pursue in different transfer instruments.89 environmental and climate change objectives is that the pool of funds involved is indeed large and probably Portugal introduced an ecological fiscal transfer for land much larger than those funds potentially available for conservation in its 2007 Portuguese Local Finances Law climate-oriented transfers.86 (LFL). The transfer provides significant incentives for 83. See Martinez-Vazquez (2019). 84. The weight given to forest area in the formula was 7.5 percent. This measure had been recommended at different times in India (see, for example, Kumar and Managi 2009). Similar proposals have been made for other countries. For example, for Indonesia, Mumbunan, Ring, and Lenk (2012) argue that changing the current general- purpose fiscal equalization transfers by incorporating “protected areas” as a criterion would benefit one-third of Indonesian provinces with an increase in transfers from 4.2 to 12.6 percent (the transfer is weighted by the protected area relative to the general area). 85. The fundamental question is how equalizing this variable (forest cover) may be in comparison to other variables that, for example and as reported by the 14th Finance Commission, some states had been pushing to include in the devolution formula, such as the Human Development Index (HDI), poverty measures, an index of social and economic backwardness, lack of infrastructure and communication facilities, and so on. Those states with greater ability to implement forest conservation may not be the ones that need more equalization in fiscal capacity and general expenditures needs. 86. And is there any evidence of this policy change in India? As Busch and Mukherjee (2018) note, India’s equalization grant or tax revenue distribution reform in 2014 created the world’s first ecological fiscal transfers (EFTs) for forest cover. In a preliminary analysis, these authors did not find any as yet increased forest cover across Indian states, probably due to the short period of time the measure has been in effect. 87. The ICMS (Imposto sobre Circulação de Mercadorias e Serviços) works in Brazil as a state level value added tax (VAT) (Faria 2020). 88. See Ring (2008) for further discussion of Brazil’s ecological ICMS. 89. Not many other developing countries appear to have introduced land conservation transfers despite the availability of international funds. In this regard, Irawan, Tacconi, and Ring (2014) analyze the design of intergovernmental fiscal transfers to channel REDD+ international funds to local governments in countries where this level of government has authority over forest management. Two design approaches are considered: cost-reimbursement and derivation. With the former, SNGs with more degraded forests—which have higher opportunity costs—would receive a higher payment per unit of carbon emission reduction; the latter approach sets a fixed percentage and rate to distribute the funds. Adapting Fiscal Decentralization Design to Combat Climate Change 52 those local governments that set aside a large proportion In the Democratic Republic of Congo, a large share of their land under protected status (Santos et al. 2012). (40 percent) of forest area fees is transferred to Other European countries (such as Germany and Poland) SNGs (provinces and territories). have discussed the introduction of similar transfers but only France has to some extent also implemented them In Mexico, many states implement decarbonization (Schroter-Schlaak et al. 2014). projects, and some assist their municipalities in doing the same. For example, the state of As a different modality of assistance, in some Jalisco has created a framework to provide funds circumstances it can be useful for higher-level to municipalities as well as to associations of governments to enter into contracts with local municipalities to implement climate protection governments to undertake “climate services.” Most projects, often in matching arrangements (GIZ, UN- often, contracts between governments are based on Habitat, and ICLEI 2017). performance indicators, and as such, contracts have large similarities with performance-based grants. Several In Niger, the National Environment Fund allocates countries implement this type of contract. resources to support the actions of local communities in protecting and improving the environment. In Australia, the federal government uses four types of intergovernmental agreements for the pursuit of In the United Kingdom, performance-based climate change objectives: (i) the Intergovernmental grants have been used in “Local Public Service Agreement on Federal Financial Relations (IGA- Agreements,” with significant additional funds FFR) is an envelope agreement that includes the provided for good performance on previously agreed distribution of federal transfers, including goods indicators related to climate decarbonization targets and services tax (GST) revenues, among the and adaptation outcomes. states as well as funding for natural disasters; (ii) National Agreements that set out policy objectives In the United States, as in many other countries, and performance measures in many sectors but the federal and state governments have a plethora through which no funding is provided;90 (iii) National of green subsidies to businesses and households Partnerships with time-limited funding agreements for reducing emissions, saving energy, and in a variety of sectors, including the environment introducing green types of energy production, but (examples include disaster risk reduction; grants the intergovernmental subset of transfers is the assistance to primary producers impacted by the one relevant to the current work. In this regard, the north Queensland floods; and the National Water federal government grants environmental subsidies Infrastructure Development Fund); and (iv) Project to state and local governments in various forms, Agreements with simpler time-limited funding for low- including to rural communities experiencing high cost and/or low-risk service or projects, including the energy costs. Also, with the Low-Income Home environment (for example, Improving Great Artesian Energy Assistance Program, federal funds are Basin Drought Resilience). At the state-local level, distributed to the state governments to implement New South Wales has established by itself a Climate assistance to low-income households. Among Change Fund that supports programs by local the states, California, for example, provides local councils, schools, and community groups to plant governments with grants and low-interest loans for 5 million trees by 2030. Similarly, the Queensland drainage water projects that, if discharged untreated, government provides grants to local government/ would pollute or threaten to pollute the waters of the councils over 12 months to monitor, act on, and state. Eligible applicants include cities, counties, prevent illegal dumping activities. districts, and others engaged in water management. 90. One such example is the Reef 2050 plan, which is a joint effort of the federal government and the state of Queensland and was jointly funded by the two in 2016. An independent panel consisting of scientists and other experts and a Reef 2050 Advisory Committee have been established to advise the two governments on implementation and to review the Reef 2050 plan. Another example is the federal 20 Million Trees Program, which pursued the goal of planting 20 million trees by 2020 by involving local councils and private organizations. Adapting Fiscal Decentralization Design to Combat Climate Change 53 The program involves, for example, grants to local control, monitoring, and inspection; sustainable forest governments for oil recycling, subsidies to all levels management; ecological and economic zoning; territorial of governments for safe household hazardous waste arrangement and agricultural regulation; preservation management, and subsidies to school districts for and sustainable use of biodiversity; and recovery of lower-emission school buses. As another example, deforested areas. Although the Amazon Fund was created the state of Illinois allocates funds to local recycling by the government, it is managed by a public bank, the grants, school greening grants, subsidies for Brazilian Development Bank (BNDES), and it functions as ecosystems protection and land preservation, and a private fund. As part of the same objectives, the federal zero-waste schools grants. Ministry of Environment periodically publishes a list that identifies municipalities responsible for the greatest share There are also cases of international transfers pursuing of deforestation in the region. Once included on the list, environmental and climate-related goals. One example municipalities become prioritized for environmental and is the Amazon Fund in Brazil. It was created to raise land control actions executed by the state and federal donations (international and domestic) to prevent, governments. In addition to becoming subject to intense monitor, and combat deforestation and to promote the surveillance, these municipalities are prohibited from conservation and sustainable use of forests in the Amazon obtaining new land clearing authorizations, apart from Biome. Activities supported include: management of special cases described by decree (Amazon Fund 2013; public forests and protected areas; environmental Ferreira Levy 2019). Adapting Fiscal Decentralization Design to Combat Climate Change 54 06 Adapting Subnational Borrowing represents a powerful resource for enabling SNGs to invest in infrastructure projects for Borrowing and Other Fiscal decarbonization and adaptation activities. However, subnational borrowing also carries potential threats to Rules to Facilitate Effective macro stability and fiscal sustainability that need to be Policies for Climate Change heeded. As in the case of the other three pillars of fiscal decentralization, in order to assess the potential role that borrowing may play in devising more effective policies for climate change in a multilevel government context, it is helpful to briefly review the general principles for subnational borrowing and how best to regulate it. Basic Principles of Subnational Borrowing and Other Fiscal Rules The last pillar of the fiscal decentralization architecture is subnational borrowing. All decentralized systems need to address the issue of long-term financing for the capital infrastructure expenditure needs of SNGs. This is so because typically, a considerable share of capital infrastructure investment responsibilities must be assigned to the subnational level.91 But how is this financed? Beyond the use of their current savings and capital transfers received from the central government, the best practical solution is for SNGs to borrow the necessary funds. Borrowing is efficient because it allows SNGs to make large lump-sum payments to acquire the necessary infrastructure in a way that is not generally possible by using current savings. Borrowing is also equitable because it allows SNGs to match the timing of consumption of the services with periodic debt service payments. Having one generation of taxpayers pay for 91. For example, in OECD countries, SNGs represent about two-thirds of all public expenditure on capital infrastructure. Adapting Fiscal Decentralization Design to Combat Climate Change 55 capital equipment, with later generations consuming the follow regarding debt ceilings or limits on total borrowing services that equipment provides, would not be fair. or ceilings on debt service expenditure (for principal and interest); the “golden rule” (funds can be used only to The two general common forms of credit-based finance investment on infrastructure); restrictions on who financing are bond issuance and borrowing from can be a lender or on whether lenders can be foreign financial intermediaries. There is no intrinsic superiority entities; and so on. In practice, the “rule-based approach” to either of these approaches to financing SNGs’ long- is the most used. term capital investment needs. Direct borrowing from private commercial banks and other financial institutions, Applying Basic Principles and Best international lending programs, bond issues, and the creation of local government development funds are all Practices in Borrowing and Rules for valid and desirable alternatives. Climate Change Policy Action On the other hand, subnational borrowing is often seen as an activity full of fiscal risks. A number of factors create As just shown, borrowing to finance subnational capital adverse incentives for SNGs not only to overspend infrastructure is desirable from both efficiency and and under-tax but also to borrow excessively, which equity perspectives. Generally, those conclusions apply can all lead to fiscal imbalances and macroeconomic equally to infrastructure investment for decarbonization instability.92 Such factors include the common pool and adaptation to climate change, but two additional problem, moral hazard risks, and the vulnerabilities of questions must be asked regarding climate investment. weak decentralization structures with low tax autonomy First, is there a case for subsidizing or facilitating and large vertical imbalances. There are also simple but subnational borrowing for climate change? Second, is powerful political economy incentives at work: borrowing there a case for specially structuring subnational access allows subnational officials to increase spending and to national and international climate finance? engage in ribbon cutting, while it postpones most payments to future officials and generations of taxpayers. Regarding the first question, the most important innovation has been the introduction of “green bonds.” Green bonds Thus, ensuring disciplined and responsible subnational (also known as climate bonds) are a special type of bond borrowing behavior generally requires more than good or fixed-income instrument for which the raised funds decentralization design. It also requires explicit rules, are earmarked for environmental and climate change monitoring, and oversight regarding subnational debt. investment projects.93 Very often green bonds offer tax Most countries adopt or issue fiscal rules that can be advantages, such as exemptions and credits, making understood as institutional constraints on policy makers’ them more attractive to investors relative to other types decision-making discretion. The objective is to allow of bonds, although the price advantage logically depends SNGs flexibility with responsible borrowing and at the on demand. These bonds are backed by the full faith and same time prevent undisciplined or irresponsible behavior. credit of the issuing government, as in the case of other Borrowing and fiscal rules used internationally range from bonds, and thus carry the same credit ratings. Even the most restrictive, as in the “administrative approach”— though green bonds are typically linked to investment in with central authorities controlling subnational borrowing specific assets, the investments themselves do not aim through debt ceilings, prohibitions on external borrowing, to generate earmarked revenue streams, as in the case, or prior approval of conditions for any new debt—to the for example, of “revenue bonds.” The green bond status least restrictive, the “market-based approach,” where may be granted by an independent board, such as the central authorities rely on financial markets to regulate Climate Bond Standard Board, part of the Climate Board and restrict subnational borrowing. In between those two Initiative (CBI), which verifies that the funds will be used extremes, there are several other approaches, including: for green projects and certifies that the bond will fund the “cooperative approach,” where rules are determined projects that include benefits to the environment.94 The through negotiations and dialog between the central EU’s Technical Expert Group on Sustainable Finance government and SNGs; the “rule-based approach,” which has recently published a report on the issuance of green relies on clearly stated ex ante rules that SNGs need to bonds to fund climate change–mitigating projects, the 92. See, for example, Ter-Minassian (2007). 93. See, for example, World Bank (2015). Adapting Fiscal Decentralization Design to Combat Climate Change 56 EU Green Bond Standard, giving the public, investors, International agencies like the World Bank also frequently and industry a definition of what is “green.” issue green bonds, and there is increasing international interest in promoting this type of instrument. For example, OECD (2018) has recently recommended that What Are Current Effective and Less the regulatory framework be adjusted to provide SNGs Effective Practices in Borrowing Design with the ability to mobilize external funding, including the in Different Countries? ability to borrow and access capital markets by issuing green and climate bonds. A number of developed countries have been experimenting In the case of developing countries, green bonds are with the use of green bonds by SNGs. For example, over much less utilized. Perhaps the big exception here is the past five years, states in Australia have been issuing China, where many provincial and local governments green bonds to finance projects with environmental and have issued green bonds. The Chinese government anti-climate change objectives, with certification by the strongly promotes “green finance,” meaning that many CBI. In Australia, state governments have the authority to provincial and local governments have issued green generate funds by issuing green bonds. The government finance guidance documents. In 2018 alone, Chinese of Victoria issued AU$300 million in green bonds in July local government entities issued at least US$5.9 billion 2016. Green bonds are used to finance new and existing in green bonds (Sandalow 2019). Other countries that projects that offer climate and environmental benefits. have used green bonds at the subnational level include The CBI, a London-based not-for-profit organization, Colombia, Estonia (Tallinn), Mexico (Mexico City), and verifies all projects funded by Victoria’s green bonds. In South Africa (Johannesburg and Cape Town). The less France, SNGs (regions, departments, and municipalities) frequent use of green bonds by SNGs in developing also have the authority to raise revenue by issuing green countries is largely explained by the many factors that bonds (Frandon-Martinez and Filkova 2018). So far, the severely constrain their access to debt finance in general; Ile-de-France region, Hauts-de-France region, Essonne after all, in order to issue a green bond, a SNG has first to department, and city of Paris have issued green bonds. be able to issue a bond at all.96 In the cases of Denmark, Finland, Norway, and Sweden, green bonds are intermediated through local government In closing, green bonds still remain a small source of funding agencies (LGFAs). These are well-established finance. As most recently reported by the OECD (OECD financial institutions specialized in financing local 2020), up to 16 countries have issued sovereign green governments of the region. For example, KommuneKredit bonds to finance green projects in governments’ budgets, accounts for almost all subnational public sector lending exceeding US$80 billion (see figure 3). However, despite in Denmark (Filkova and Frandon-Martinez 2018). its recent growth, the size of the sovereign green bond Similar arrangements exist in Finland. SNGs are also market is still quite small; for example, among OECD using green bonds in other developed countries, such as countries, sovereign green bonds account for only 0.1 Canada, Iceland, and the United States.95 percent of all government debt securities. 94. The Climate Bond Standard Board reports to the Governors of the CBI, which specializes in the verification of the green credentials of proposed projects. 95. Similar to the green bonds idea, in the United States, the federal government exempts interest on state and local bonds issued to support water, sewer, and hazardous waste facilities. 96. See White and Wahba (2019) for a recent discussion of why green bond financing still poses very limited opportunities for climate financing in most developing countries and the factors that should be addressed if the scope for subnational borrowing (for climate purposes or otherwise) is to be significantly improved. Adapting Fiscal Decentralization Design to Combat Climate Change 57 Figure 3. . Total Sovereign Green Bond Issuance in Millions of Dollars USD $28,000 $26,000 $24,000 $22,000 $20,000 $18,000 $16,000 $14,000 $12,000 $10,000 $8,000 $6,000 $4,000 $2,000 $0 2016 2017 2018 2019 2020 Belgium Chile France Germany Ireland Netherlands Poland Others Note: Data as of July 2020. “Others” include Fiji (2017), Hong Kong (China) (2019), Hungary (2020), Indonesia (2018, 2019, and 2020), Lithuania (2018), Korea (2019), Nigeria (2017), Seychelles (2018), and Sweden (2020). Source: OECD (2020), Medcalf et al. (2020). Adapting Fiscal Decentralization Design to Combat Climate Change 58 07 Summary and Conclusions The main objectives of this paper have been to re- examine the basic principles of fiscal decentralization from the perspective of fighting climate change and to critically review the international practices for engaging SNGs in addressing climate change. Even though the theory of fiscal decentralization provides clear guidelines on redesigning fiscal decentralization architecture from the perspective of fighting climate change, and international good practices are available for replication across many other countries, many countries around the world so far have not effectively engaged their SNGs in their climate change policy frameworks. This frequent failure to explicitly account for the role that SNGs could and should play in the design and implementation of national strategies to combat climate change is the most significant finding of this review paper and also a clear call for action. The scientific consensus is clear that the stock of GHGs in the earth’s atmosphere is causing global warming and other changes in the climate. Even though there are differences of opinion on how to quantify and aggregate the damages, it is widely accepted that the economic impact of climate change has been estimated to be quite severe in terms of losses to GDP, health, and food security and increases in poverty. From the main sources of GHG emissions identified in the Stern (2006) report, an estimated 40 percent come from activities over which SNGs typically exert regulatory and taxing powers, meaning that these entities potentially have a big role to play in fighting climate change. The role of SNGs becomes even stronger in terms of adaptation policies, given the smaller importance of externalities and economies of scale in this dimension of fighting climate change. Adapting Fiscal Decentralization Design to Combat Climate Change 59 Still, the role to be played by SNGs has some natural the world will work to strengthen their policy frameworks limits. Although decentralized governments can better by adapting their fiscal decentralization design to the cater to differences in needs and preferences and provide best principles and practices, as discussed in this paper. a laboratory for incubating innovations and testing However, realistically speaking, those changes will not effectiveness, their big disadvantages are their potentially be easy to execute. Not only is institutional reform often lower effectiveness in dealing with large externalities and slow and difficult to implement, but fiscal decentralization in taking advantage of economies of scale. Thus, it is design, as also shown in this paper, pursues many other mainly the former, the presence of significant externalities significant worthwhile objectives, which in the eyes and how they can be handled in decentralized settings, of policy makers may be as or more important than that becomes the central consideration for the role that fighting climate change. Thus, a significant rebalancing SNGs can play in decarbonization policies. Nevertheless, of government objectives regarding decentralization will climate change policies in practice take many forms, and need to take place over time. many of those activities indeed can be implemented by SNGs. On the practice of expenditure assignments, it is clear that the advantage of involving SNGs in climate Thus, in many countries, SNGs do play significant roles change policies is that in a decentralized setting, these on certain types of decarbonization actions, such as governments are expected to be more responsive managing land use, housing codes, water and power and accountable to residents. Working against this utilities, waste disposal, or transport systems. On the involvement is the commonly large scope of the side of adaptation actions, SNGs have also been, in externalities associated with activities causing climate many countries, on the front line of disaster response change. Thus, how externalities manifest themselves and climate adaptation programs. However, measuring in the different activities and policies related to climate actual SNG involvement has proven difficult with the change—the geographical scope of the externalities— information that is currently available. becomes a critical linchpin in deciding the assignment of expenditure responsibilities for climate change policies. One way to measure such involvement is to look at the share of environmental expenditures of SNGs in each For a decentralized response to be operational and country. Clearly, this provides only a limited view, since effective, it is also necessary that SNGs have the it disregards other subnational decarbonization activities, required administrative capacity and funding, and that such as in transportation or sanitization, and ignores they do not get involved in negative strategic behaviors expenditures in adaptation projects. Overall, this type of (for example, competing with low standards in order measure provides very low numbers, even when SNGs to attract more economic activity). Furthermore, the have more of a presence than central governments, assignment of responsibilities among different levels of mainly because the overall share of environmental and government is likely to differ for decarbonization and climate-related spending and investment is very low adaptation activities, with SNGs playing a more important with respect to GDP in most countries, rich and poor. role in the latter. Even in the case of decarbonization An alternative measure, SNG involvement in the NDCs policies, however, optimal assignments are likely to call under the Paris Agreement, has also yielded modest for concurrent assignments involving two or more levels indicators thus far. These preliminary findings provide of government. In this case, clarity in the assignment of further motivation for seeking venues through which functional responsibilities requires identifying what level SNGs can become more engaged in the planning of government is responsible for the different attributes of and implementation of decarbonization and adaptation the function (regulation, financing, and implementation). policies. This discussion also highlights the importance of intergovernmental coordination and cooperation for This paper has been structured around the four pillars effective policy making. of fiscal decentralization architecture: the assignment of functional expenditure responsibilities, the assignment In practice, there are significant variations in how of tax and other revenue sources, the systems of countries organize the division of responsibilities intergovernmental transfers, and subnational credit and regarding policies and programs to fight climate change. borrowing. One of its foci has been to identify the best Although some countries provide the central government tools available within fiscal decentralization design to with exclusive responsibility for these tasks and make effectively engage SNGs at the national level in the fight SNGs just implementation arms, other countries—the against climate change. The hope is that countries around largest number—assign joint responsibility to the central Adapting Fiscal Decentralization Design to Combat Climate Change 60 government and SNGs and at the same time provide policies with explicit minimum national standards can SNGs with varying degrees of flexibility to expand on ensure that geographical externalities are recognized national policies and exceed the targets programmed and can avoid any possibility of a “race to the bottom.” at the central level. Yet other countries, perhaps the The first or centralized approach fails to take advantage smallest number but important for the size of their of innovation and experimentation at the subnational economies, practice a completely decentralized model, level, while the third or decentralized approach risks not with much less coordination among the different tiers addressing externalities or allowing “race to the bottom” of government and sometimes with subnational units and free-riding SNG behaviors. playing a leadership role in the absence of explicitly coordinated national policies and programs. In terms of revenue assignments, it is clear that a good revenue-raising tax and a good Pigouvian tax have On balance, it appears that the second model, with different properties. While the former raises revenue with shared responsibilities between central and subnational a low impact on economic activity, thus reducing excess authorities allowing subnational units to exceed national burdens, in a Pigouvian setting, the goal is to maximize targets, is a superior approach because it is best the elasticity of response so that with small tax levels, adapted to the peculiarities of climate change policies polluting activities are reduced as much as possible. in a multilevel government setting. Given the enormous Given the mobility of businesses and households and the economic externalities present in most climate change fact that the benefits of climate change action are in many issues, it is important that the central authorities design cases not localized, the most effective use of Pigouvian national programs and coordinate actions throughout the taxes may be at higher levels of government, such as entire national territory. However, the mixed or shared regionally, nationally, or even (if feasible) internationally. approach also takes advantage of the laboratory setting that fiscal decentralization provides for experimenting with An important constraint to using green taxes at the innovative subnational methods. In this type of setting, subnational level is that in practice, for many reasons SNGs can innovate and compete in “a race to the top” not related to climate change, SNGs in many countries in in fighting climate change, while nationally coordinated fact have very low levels of taxation authority. In addition, Adapting Fiscal Decentralization Design to Combat Climate Change 61 some other factors may contribute to the low powers of This review of country practices with green subnational green taxation at the subnational level. First, there is often levies uncovers a rich collection of different tax the perception that many related subnational charges instruments, charges, and levies that SNGs use in little have a regressive incidence, penalizing lower income more than a handful of decentralized countries. The groups and so tying the hands of SNGs, which lack many other decentralized countries that so far have the ability to offset that regressivity with other policies. not introduced subnational green taxation—other than Second, there are also technical or administrative issues. the usual charges for waste collection, and so forth— Some green taxes can be quite complex to administer, could replicate some of those types of tax assignments especially when dealing with global multi-unit businesses. to strengthen the role of their SNGs in fighting climate In addition, SNGs may have lower ability to measure change from the revenue side of the budget. and monitor pollution emissions and other activities that constitute the tax bases of green levies; this is part of the Intergovernmental fiscal transfers are likely the most capacity issue. suitable financing instrument for internalizing spatial externalities related to decarbonization and adaptation From the theory of revenue assignments, the question programs. They are also a versatile and adaptable of what level of government is better equipped to levy instrument. Beyond conventional conditional grants, different green taxes is answered by determining central government funds can be also transferred via the extent of the externality present. If the externality different types of subsidies and direct performance is confined to a local area, the responsibility for the contracts or agreements between the central government green tax should be local. As the area of the externality and SNGs to fight climate change. increases to the regional or national level, the proper assignment of tax powers should be regional or national. In general, these transfers are designed separately An important advantage of green taxes emphasized from other transfer policies, such as unconditional throughout the paper is that they yield a double dividend, equalization grants and revenue sharing on a derivation the decarbonization of activities and increased budget basis pursuing the objectives of equalization and revenues, which could derive into more public services, devolution to SNGs. In contrast, in cases where the including climate adaptation investment measures. main objective of the conditional transfers is not diluted These benefits will tend to offset any incentives to out- or confused, it may be feasible and desirable to imbed migrate because of the green taxes. climate change objectives into transfers. More generally, conditional grants, for recurrent and capital purposes In the case of adaptation activities, to the extent that and with ex ante or ex post conditionality, are generally the consequences of climate change–related disasters well equipped for designing and implementing climate- tend to be more localized, there is more room for using specific transfers to SNGs. Conditional transfers can be the benefit principle for subnational financing. This calls used to compensate local jurisdictions for internalizing for the use of fees and charges as well as traditional different types of externalities, to incentivize different subnational instruments, such as real estate property conservation activities, to promote energy efficiency, taxes, property transfer taxes, betterment levies, and or to introduce land-use restrictions to prevent other land value capture measures. These latter are flooding damages. especially attractive because most adaptation activities will result in land value increments. This paper has further identified a large variety of country practices in using intergovernmental transfers to engage In practice, regardless of the assignment of taxing powers SNGs in the fight against climate change. Conditional to different levels of government, the most significant grants are being used in some countries for forest fact is the overall very low levels of green taxation in conservation programs, as in the case of Brazilian states’ most countries around the world, which provides ample ecological “ICMS” or Portugal’s ecological fiscal transfer space for both central governments and SNGs to make for land conservation. In other countries like Australia much more extensive use of this type of levy. This fiscal and the United Kingdom, higher-level governments space is even larger at the subnational level because of are entering into agreements and contracts with local the large vertical fiscal imbalances most decentralized governments to undertake determined “climate services.” countries experience. Providing SNGs with greater green tax autonomy becomes an additionally desirable way to Finally, this work highlights the ways that prudential close those vertical fiscal imbalances. borrowing can be a powerful financing complement Adapting Fiscal Decentralization Design to Combat Climate Change 62 enabling SNGs to invest in infrastructure projects for in large part because most SNGs still do not have access decarbonization and adaptation activities. Provided that to capital markets. Despite their recent growth, the size borrowing is prudent and within bounds—as outlined, of the sovereign green bond market is still quite small, for example, by the country’s fiscal rules—this form of even in developed countries. financing subnational capital infrastructure is in general desirable from both efficiency and equity perspectives, There is much work to do at all levels to reduce the and these conclusions hold even more so for such emission of GHGs and implement climate-adaptive borrowing done to fund green infrastructure investments. infrastructure. Currently, the relative GDP share of environment-targeted spending and effort even on the Whether subnational borrowing rules can be justifiably national level is low in many countries. The subnational relaxed to some extent to further incentivize SNGs’ levels, in federal or decentralized nations, generally still infrastructure spending on green projects is a more play a relatively small role. Nevertheless, there have complex issue, since it is necessary to avoid moral been many innovative and exciting developments in this hazards and keep subnational debt on a sustainable space, whether in the form of expenditure assignments, basis. The safest approach is to facilitate and subsidize tax authority, funding transfers, or borrowing. Shared green borrowing, an area that has seen some recent responsibility for policy and program design and innovations. The most important development has been implementation, fee- or charge-funded adaptation the introduction of green bonds (also known as climate activities, objective-targeted intergovernmental transfers bonds); these instruments often offer tax advantages, or conditional grants, and the use of green bonds are such as exemptions and credits, making them more some of the most promising approaches profiled. attractive to investors relative to other types of bonds, Regardless of the methods chosen, however, what is while they are backed by the full faith and credit of the clear is that there is ample space ahead for the further issuing government. An increasing number of developed involvement of SNGs in combating climate change countries are making use of green bonds by SNGs, but across the world. their adoption has been slower in developing countries, Adapting Fiscal Decentralization Design to Combat Climate Change 63 References ADEME (French Environment and Energy Management Agency). 2015. “Addressing Climate Change – French Know How.” ADEME, Angers. Ahmad, Ehtisham, and Nicholas Stern. 2009. “Effective Carbon Taxes and Public Policy Options: Insights from India and Pakistan.” Working Paper 28, Asia Research Centre, London School of Economics and Political Science, London. Alm, James, and H. Spencer Banzhaf. 2012. “Designing Economic Instruments for the Environment in a Decentralized Fiscal System.” Journal of Economic Surveys 26 (2): 177–202. Amazon Fund. 2013. “Amazon Fund Project Document.” Amazon Fund, National Bank for Economic and Social Development, Rio de Janeiro. Araral Jr., Eduardo K. 2011. “The Impact of Decentralization on Large Scale Irrigation: Evidence from the Philippines.” Water Alternatives 4 (2): 110–23. Atteridge Aaron, Manish Kumar Shrivastava, Neha Pahuja, and Himani Upadhyay. 2012. “Climate Policy in India: What Shapes International, National and State Policy?” Ambio 41 (Suppl 1): 68–77. Avner, Paolo, Jun Rentschler, and Stéphane Hallegatte. 2014. “Carbon Price Efficiency: Lock-in and Path Dependence in Urban Forms and Transport Infrastructure.” Policy Research Working Paper 6941, World Bank, Washington, DC. Bagstad, Kenneth J., Kevin Stapleton, and John R. D’Agostino. 2007. “Taxes, Subsidies, and Insurance as Drivers of United States Coastal Development.” Ecological Economics 63 (2-3): 285–98. Banzhaf, H. Spencer, and B. Andrew Chupp. 2012. “Fiscal Federalism and Interjurisdictional Externalities: New Results and an Application to US Air Pollution.” Journal of Public Economics 96 (5-6): 449–64. Barbi, Fabiana, and Leila da Costa Ferreira. 2013. “Climate Change in Brazilian Cities: Policy Strategies and Responses to Global Warming.” International Journal of Environmental Science and Development 4 (1): 49–51. Adapting Fiscal Decentralization Design to Combat Climate Change 64 Baumol, William J. 1972. “On Taxation and the Control of Externalities.” The American Economic Review 62 (3): 307–22. Bedner, Adriaan. 2009. “Consequences of Decentralization: Environmental Impact Assessment and Water Pollution Control in Indonesia.” Law & Policy 32 (1): 38–60. Bennett, Vanora. 2020. “EBRD Unveils Proposal to be Majority Green Bank by 2025” (online). European Bank for Reconstruction and Development, London, July 8, https://www. ebrd.com/news/2020/ebrd-unveils-proposal-to-be-majority-green-bank-by-2025.html. Breton, Albert, Giorgio Brosio, Silvana Dalmazzone, and Giovorra Garrone, eds. 2007. Environmental Governance and Decentralisation. New Horizons in Environmental Economics. Cheltenham, UK: Edward Elgar. Brunel, Claire, and Arik Levinson. 2021. “Globalization and Greenhouse Gas Emissions: Evidence from the United States.” NBER Working Paper 28372, National Bureau of Economic Research, Cambridge, MA. Bureau, Benjamin. 2011. “Distributional Effects of a Carbon Tax on Car Fuels in France.” Energy Economics 33 (1): 121–30. Busch, Jonah, and Anit Mukherjee. 2018. “Encouraging State Governments to Protect and Restore Forests Using Ecological Fiscal Transfers: India’s Tax Revenue Distribution Reform.” Conservation Letters 11 (2): 1–10. C2ES (Center for Climate and Energy Solutions). 2020. “State Climate Policy Maps” (online). C2ES, Arlington, VA, https://www.c2es.org/content/state-climate-policy/. Callan, Tim, Sean Lyons, Susan Scott, Richard Tol, and Stefano Verde. 2009. “The Distributional Implications of a Carbon Tax in Ireland.” Energy Policy 37 (2): 407–12. Chang, Howard, Hilary Sigman, and Leah Traub. 2007. “Endogenous Decentralization in Federal Environmental Policies.” NBER Working Paper 13238, National Bureau of Economic Research, Cambridge, MA. Charnovitz, Steve. 2014. “Green Subsidies and the WTO.” Policy Research Working Paper 7060, World Bank, Washington, DC. Cheng, Shulei, Wei Fan, Jiandong Chen, Fanxin Meng, Gengyuan Liu, Malin Song, and Zhifeng Yang. 2020. “The Impact of Fiscal Decentralization on CO2 Emissions in China.” Energy 192: 116685. Cointreau, Sandra, and Constance Hornig. 2003. “The Application of Economic Instruments in Water and Solid Waste Management.” Regional Policy Dialogue, Inter-American Development Bank, Washington, DC. Copeland, Brian, and M. Scott Taylor. 2004. “Trade, Growth, and the Environment. Journal of Economic Literature 42(1): 7-71. Dalmazzone, Silvana. 2006. “Decentralization and the Environment.” In Handbook of Fiscal Federalism, Ehtisham Ahmad and Giorgio Brosio, eds., 459–77. Cheltenham, UK: Edward Elgar. Adapting Fiscal Decentralization Design to Combat Climate Change 65 Danida (Danish International Development Agency). 2009. “Thematic Review of Special Interventions within Environment (2004–2008), Final Report May 2009.” Danida, Ministry of Foreign Affairs, Copenhagen. De Mello Jr., Luiz. R. 2000. “Fiscal Decentralization and Intergovernmental Fiscal Relations: A Cross-Country Analysis.” World Development 28 (2): 365–80. Dietz, Simon, Frederick van der Ploeg, Armon Rezai, and Frank Venmas. 2020. “Are Economists Getting Climate Dynamics Right and Does it Matter?” Cesifo Working Paper 8122-2020, Munich Society for the Promotion of Economic Research, Munich. Dinan, Terry M., Maureen L. Cropper, and Paul R. Portney. 1999. “Environmental Federalism: Welfare Losses from Uniform National Drinking Water Standards.” In Environmental and Public Economics: Essays in Honor of Wallace E. Oates, Arvind Panagariya, Paul Portney, and Robert M. Schwab, eds., 13–31. Cheltenham, UK: Edward Elgar. Dubash Navroz K., Markus Hagemann, Niklas Höhne, and Prabhat Upadhyaya. 2013. “Developments in National Climate Change Mitigation Legislation and Strategy.” Climate Policy 13 (6): 649–64. EBRD (European Bank for Reconstruction and Development). 2020. “EBRD Green Cities” (online). EBRD, London. https://www.ebrdgreencities.com/. Edwards, Ian, Johanna Nalau, Donovan Burton, and Brendan Mackey. 2019. “Implications of Emergent Risk for Application of Risk Transfer Mechanisms by Local Governments in Queensland.” Environmental Science & Policy 96 (June): 1–8. EGCA (European Green Capital Award). 2020. “About EGCA” (online). European Commission, Brussels, https://ec.europa.eu/environment/europeangreencapital/about- the-award/. Elheddad, Mohamed, Nassima Djellouli, Aviral Tiwari, and Shawkat Hammoudeh. 2020. “The Relationship between Energy Consumption and Fiscal Decentralization and the Importance of Urbanization: Evidence from Chinese Provinces.” Journal of Environmental Management 264: 110474. Escaleras, Monica, and Charles A. Register. 2012. “Fiscal Decentralization and Natural Hazard Risks.” Public Choice 151: 165–83. Estache, Antonio, Grégoire Garsous, and Ronaldo Seroa da Motta. 2016. “Shared Mandates, Moral Hazard, and Political (Mis)alignment in a Decentralized Economy.” World Development 83 (July): 98–110. Faria, Caroline. 2020. “O Que é o ICMS (“What is ICMS”).” ICMS Ecológico, InfoEscola, https://www.infoescola.com/ecologia/icms-ecologico. Farzanegan, Mohammad Reza and Tim Mennel. 2012. “Fiscal Decentralization and Pollution: Institutions Matter.” MAGKS Joint Discussion Paper Series in Economics, No. 22-2012 Adapting Fiscal Decentralization Design to Combat Climate Change 66 Fay, Marianne; Hallegatte, Stephane; Vogt-Schilb, Adrien; Rozenberg, Julie; Narloch, Ulf; Kerr, Tom. 2015. Decarbonizing Development : Three Steps to a Zero-Carbon Future. Climate Change and Development;. Washington, DC: World Bank. https://openknowledge. worldbank.org/handle/10986/21842 Fay, Marianne, Luis Alberto Andres, Charles Fox, Ulf Narloch, Stephane Straub, and Michael Slawson. 2017. Rethinking Infrastructure in Latin America and the Caribbean: Spending Better to Achieve More. Washington, DC: World Bank. Ferreira Levy, Joaquim. 2019. “Amazon Fund 10 Years: Activity Report 2018.” Amazon Fund, Naional Bank for Economic and Social Development, Rio de Janeiro. Filkova, Monica, and Camille Frandon-Martinez. 2018. “Nordic and Baltic Public Sector Green Bonds.” Foreign and Commonwealth Office, London. Flavelle, Christopher. 2020. “Trump Administration Presses Cities to Evict Homeowners from Flood Zones.” The New York Times. March 11, https://www.nytimes.com/2020/03/11/ climate/government-land-eviction-floods.html. Fomby, Thomas, and Limin Lin. 2006. “A Change Point Analysis of the Impact of Environmental Federalism on Aggregate Air Quality in the United States: 1940–1998.” Economic Inquiry 44(1):109-20. Frandon-Martinez, Camille, and Monica Filkova. 2018. “Green Bond Opportunities in France.” Country Briefing, Climate Bonds Initiative, April. Fredriksson, Per, and Noel Gaston. 2000. “Environmental Governance in Federal Systems: The Effects of Capital Competition and Lobby Groups.” Economic Inquiry 38 (3): 501–14. Fredriksson, Per, and Daniel L. Millimet. 2002. “Strategic Interaction and the Determination of Environmental Policy Across U.S. States.” Journal of Urban Economics 51 (1): 101–22. Friis-Hansen, Esbern, Bernard Bashaasha, and Charles Aben. 2013. “Decentralization and Implementation of Climate Change Policy in Uganda.” DIIS working paper 2013:17, Danish Institute for Internation Studies, Copenhagen. Gago, Alberto, Xavier Labandeira, José M. Labeaga, and Xiral López-Otero. 2020. “Transport Taxes and Decarbonization in Spain: Distributional Impacts and Compensation.” Hacienda Publica Española/Review of Public Economics, https://labandeira.eu/ publicacions/hpe2020.pdf. Garcia-Valiñas, Maria. 2004. “Environmental Federalism: A Proposal of Decentralization.” European Regional Science Association, conference paper 492. Gignoux, Jérémie, and Marta Menéndez. 2016. “Benefit in the Wake of Disaster: Long-Run Effects of Earthquakes on Welfare in Rural Indonesia.” Journal of Development Economics 118: 26–44. Gilmore, Elisabeth. A., and Travis St. Clair. 2018. “Budgeting for Climate Change: Obstacles and Opportunities at the US State Level.” Climate Policy 18 (6): 729–41. Adapting Fiscal Decentralization Design to Combat Climate Change 67 GIZ (Deutsche Gesellschaft für Internationale Zusammenarbeit), UN-Habitat, and ICLEI (Local Governments for Sustainability). 2017. “Enabling Subnational Climate Action through Multi-Level Governance.” GIZ, Bonn, November. Goodspeed, Timothy J., and Andrew F. Haughwout. 2012. “On the Optimal Design of Disaster Insurance in a Federation.” Economics of Governance 13: 1–27. Grafakos, Stelios, Alexandra Tsatsou, Luca D’Acci, James Kostaras, Adriana Lopez, Nohemi Ramirez, and Barbara Summers. 2019. “Exploring the Use of Land Value Capture Instruments for Green Resilient Infrastructure Benefits: a Framework Applied in Cali, Colombia.” Working Paper WP19SG1, Lincoln Institute of Land Policy, Cambridge, MA. Gupta, Joyeeta, Ralph Lasage, and Tjeerd Stam. 2007. “National Efforts to Enhance Local Climate Policy in the Netherlands.” Environmental Sciences 4 (3): 171–82. Hahn, Robert, and Roger Noll.1981. “Designing a Market for Tradable Emissions Permits.” California Institute of Technology working paper 398. Hankla, Charles R., Jorge Martinez-Vazquez, and Raul Ponce Rodriguez. 2019. Local Accountability and National Coordination in Fiscal Federalism: A Fine Balance. Studies in Fiscal Federalism and State-Local Finance. Cheltenham, UK: Edward Elgar. Hao, Yu, Yu-Fu Chen, Hua Liao, and Yi-Ming Wei. 2020. “China’s Fiscal Decentralization and Environmental Quality: Theory and an Empirical Study.” Environment and Development Economics 25 (2): 159–81. Happaerts Sander, Simon Schunz, and Hans Bruyninckx. 2011. “Federalism and Intergovernmental Relations: the Multi-Level Politics of Climate Change Policy in Belgium.” Working Paper 58, Leuven Centre for Global Governance Studies, Leuven. Hartman, Richard, and O-Sung Kwon. 2005. “Sustainable Growth and the Environmental Kuznets Curve.” Journal of Economic Dynamics and Control 29 (10): 1701–36. He, Qichun. 2015. “Fiscal decentralization and environmental pollution: Evidence from Chinese panel data.” China Economic Review 36(C): 86-100. Held, David, Eva-Maria Nag, and Charles Roger. 2011. “The Governance of Climate Change in China.” LSE Global Governance Working Papers, London School of Economics, London. Helland, Eric, and Andrew B. Whitford. 2003. “Pollution Incidence and Political Jurisdiction: Evidenec from the TRI.” Journal of Environmental Economics and Management 46 (3): 403–24. Hinku. 2020. “Hinku Network - Towards Carbon Neutral Municipalities” (online). Carbon Neutral Finland, June 23, https://www.hiilineutraalisuomi.fi/en-US/Hinku. Hoberg, George. 2016. “Unsustainable Development: Energy and Environment in the Harper Decade.” In The Harper Factor: Assessing a Prime Minister’s Policy Legacy, Jennifer Ditchburn and Graham Fox, eds. Montreal, QC: McGill-Queen’s University Press, Montreal, 253–63. Adapting Fiscal Decentralization Design to Combat Climate Change 68 ICLEI (International Council for Local Environmental Initiatives). 2018. “Resilient Cities Report 2018: Tracking Local Progress on the Resilience Targets of SDG 11.” ICLEI, Bonn. Ingram, Gregory K., and Yu-Hung Hong, eds. 2012. Value Capture and Land Policies. Cambridge, MA: Lincoln Institute of Land Policy. Irawan, Silvia, Luca Tacconi, and Irene Ring. 2014. “Designing Intergovernmental Fiscal Transfers for Conservation: The Case of REDD+ Revenue Distribution to Local Governments in Indonesia.” Land Use Policy 36 (January): 47–59. Jerch, Rhiannon, Matthew E. Kahn, and Gary C. Lin. 2020. “Local Public Finance Dynamics and Hurricane Shocks.” NBER Working Paper 28050, National Bureau of Economic Research, Cambridge, MA. Kocornik-Mina, Adriana, Thomas K. J. McDermott, Guy Michaels, and Ferdinand Rauch. 2020. “Flooded Cities.” American Economic Journal: Applied Economics 12 ( 2): 35–66. Kollner, Thomas, Oliver Schelske, and Irmi Seidl. 2002. “Integrating Biodiversity into Intergovernmental Fiscal Transfers Based on Cantonal Benchmarking: a Swiss Case Study.” Basic and Applied Ecology 3 (4): 381–91. Kolstad, Charles D. 2020. “Subnational Carbon Pricing in the US.” ifo DICE Report, ifo Institute, Leibnitz Institute for Economic Research, Munich. Konisky, David M., and Neal D. Woods. 2012. “Environmental Free Riding in State Water Pollution Enforcement.” State Politics and Policy Quarterly 12 (3): 227–51. Kostka, Genia, and William Hobbs. 2012. “Local Energy Efficiency Policy Implementation in China: Bridging the Gap between National Priorities and Local Interests.” The China Quarterly 211: 765–85. Kozak, Daniel, Alejandro de Castro Mazarro, Hayley Henderson, Demián Rotbart, Rodolfo Aradas, and Gabriel Olivares. 2020. “Implementación de Infraestructura Azul-Verde (IAV) a través de Mecanismos de Captación de Plusvalía en la Región Metropolitana de Buenos Aires. El Caso del Arroyo Medrano.” Lincoln Institute of Land Policy, Cambridge MA. Kresch, Evan Plous. 2020. “The Buck Stops Where? Federalism, Uncertainty, and Investment in the Brazilian Water and Sanitation Sector.” American Economic Journal: Economic Policy 12 (3): 374–401. Kumar, Surender, and Shunsuke Managi. 2009. “Compensation for Environmental Services and Intergovernmental Fiscal Transfers: The Case of India.” Ecological Economics 68 (12): 3052–9. Kunce, Mitch, and Jason F. Shogren. 2005. “On Interjurisdictional Competition and Environmental Federalism.” Journal of Environmental Economics and Management 50 (1): 212–24. Leon, Warren. 2020. “100% Clean Energy Collaborative” (online). Clean Energy States Alliance, https://www.cesa.org/projects/100-clean-energy-collaborative/. Adapting Fiscal Decentralization Design to Combat Climate Change 69 Leppänen, Simo, Laura Solanko, and Riitta Kosonen. 2017. “The Impact of Climate Change on Regional Government Expenditures: Evidence from Russia.” Environmental and Resource Economics 67: 67–92. Li, Gang. 2020. “Trade and the Equivalence between Environmental Tax and Quota.” MPRA Paper 103463, Munich Personal RePEc Archive, Munich. Linnerooth-Bayer, Joanne, and Reinhard Mechler. 2006. “Insurance for Assisting Adaptation to Climate Change in Developing Countries: a Proposed Strategy.” Climate Policy 6 (6): 621–36. Lipscomb, Molly, and Ahmed Mobarak. 2017. “Decentralization and Pollution Spillovers: Evidence from the Re-drawing of County Borders in Brazil.” Review of Economic Studies 84 (1): 464–502. List, John A, and Shelby Gerking. 2000. “Regulatory Federalism and Environmental Protection in the United States.” Journal of Regional Science 40 (3): 453–71. List, John A. and Charles Mason. 2001. “Optimal Institutional Arrangements for Transboundary Pollutants in a Second-Best World: Evidence from a Differential Game with Asymmetric Players.” Journal of Environmental Economics and Management 42 (3): 277–96. List, John A., Erwin H. Bulte, and Jason F. Shogren. 2002. “‘Beggar thy Neighbor:’ Testing for Free Riding in State-Level Endangered Species Expenditures.” Public Choice 111 (3/4): 303–15. Liu, Liangliang, and Linmu Li. 2019. “Effects of Fiscal Decentralisation on the Environment: New Evidence from China.” Environmental Science and Pollution Research 26 (3-4): 36878–86. Liu, Liangliang, Donghong Ding, and Jun He. 2019. “Fiscal Decentralization, Economic Growth, and Haze Pollution Decoupling Effects: A Simple Model and Evidence from China.” Computational Economics 54 (4): 1423–41. Lopez, Ramon. 1994. “The Environment as a Factor of Production: The Effects of Economic Growth and Trade Liberalization.” Journal of Environmental Economics and Management 27 (2): 163–84. Lucon, Oswaldo, and Jose Goldemberg. 2010. “São Paulo – The ‘Other’ Brazil: Different Pathways on Climate Change for State and Federal Governments.” The Journal of Environment and Development 19 (3): 335–57. Mandal, Subrata, and M. Govinda Rao. 2007. “Overlapping Fiscal Domains and Effectiveness of Environmental Policy in India.” In Environmental Governance and Decentralization, Albert Breton, Giorgio Brosio, Silvana Dalmazzone, and Giovorra Garrone, eds. Cheltenham, UK: Edward Elgar. Markusen, James, Edward Morey, and Nancy Olewiler. 1993. “Environmental Policy When Market Structure and Plant Locations are Endogenous.” Journal of Environmental Economics and Management 24 (1): 69–86. Adapting Fiscal Decentralization Design to Combat Climate Change 70 ———. (1995), “Competition in Regional Environmental Policies when Plant Locations are Endogenous.” Journal of Public Economics 56 (1): 55–77. Martinez-Vazquez, Jorge. 2020. “Emerging Trends in Fiscal Transfer Systems in Selected Federations: Implications for India.” In Intergovernmental Transfers in Federations, Serdar Yilmaz and Farah Zahir, eds. Cheltenham, UK: Edward Elgar. Martinez-Vazquez, Jorge, Santiago Lago-Peñas, and Agnese Sachi. 2017. “The Impact of Fiscal Decentralization: A Survey.” Journal of Economic Surveys 31 (4): 1095–129. McGarvey, Todd, and Amy Morsch. 2016. “Local Climate Action: Cities Tackle Emissions of Commercial Buildings.” Center for Climate and Energy Solutions, Arlington, VA. McKibbin, Warwick, J., and Peter J. Wilcoxen. 2002. “The Role of Economics in Climate Change Policy.” Journal of Economic Perspectives 16 (2): 107–29. Mcleman, Robert, and Barry Smith. 2006. “Vulnerability to Climate Change Hazards and Risks: Crop and Flood Insurance.” Canadian Geographer / Le Géographe Canadien 50 (2): 217–26. Medcraft, Greg, et al. 2020. OECD Business and Finance Outlook 2020. OECD. https:// www.oecd.org/daf/oecd-business-and-finance-outlook-26172577.htm Mehling Michael, and David J. Frenkil. 2013. “Climate Law in the United States: Facing Structural and Procedural Limitations.” In Climate Change and the Law, Erkki Hollo, Kati Kulovesi, and Michael Mehling, eds., 473–87. Dordrecht, Netherlands: Springer. Mejía, Luis, 2018. “Índice Municipal de Gestión del Riesgo de Desastres en Colombia.” Unidad Nacional para la Gestión del Riesgo de Desastres, Bogotá. Metcalf, Gilbert. E., and David Weisbach. 2009. “The Design of a Carbon Tax.” Chicago Unbound, University of Chicago Law School, Chicago. Miao, Qing, Yilin Hou, and Michael Abrigo. 2018. “Measuring the Financial Shocks of Natural Disasters: A Panel Study of U.S. States.” National Tax Journal 71 (1): 11–44. Millimet, Daniel. L. 2003. “Assessing the Empirical Impact of Environmental Federalism.” Journal of Regional Science 43 (4): 711–33. Millimet, Daniel. L., and John A. List. 2003. “A Natural Experiment on the ‘Race to the Bottom’ Hypothesis: Testing for Stochastic Dominance in Temporal Pollution Trends.” Oxford Bulletin of Economics and Statistics 65 (4): 395–420. Monogan III, James E., David Konisky, and Neal Woods. 2017. “Gone with the Wind: Federalism and the Strategic Location of Air Polluters.” American Journal of Political Science 61 (2): 257–70. Morris, Adele, Noah Kaufmann, and Siddhi Doshi. 2020. “Revenue at Risk in Coal- Reliant Counties.” NBER Working Paper 27307, National Bureau of Economic Research, Cambridge, MA. Adapting Fiscal Decentralization Design to Combat Climate Change 71 Mukhi, Neha, Suneira Rana, Sara Mills-Knapp, and Eskedar Gessesse. 2020. “World Bank Outlook 2050 Strategic Directions Note. Supporting Countries to Meet Long-Term Goals of Decarbonization.” World Bank, Washington, DC. Mumbunan, Sonny, Irene Ring, and Thomas Lenk. 2012. “Ecological Fiscal Transfers at the Provincial Level in Indonesia.” UFZ Discussion Paper 06/2012, Helmholtz Centre for Environmental Research, Leipzig. Murty, M. Narsimha. 1996. “Fiscal Federalism Approach for Controlling Global Environmental Pollution.” Environmental and Resource Economics 8 (4): 449–59. Nagase, Yoko, and Emilson C. D. Silva. 2000. “Optimal Control of Acid Rain in a Federation with Decentralized Leadership and Information.” Journal of Environmental Economics and Management 40 (2): 164–80. Næss, Lars Otto, Guri Bang, Siri Eriksen, and Jonas Vevatne. 2005. “Institutional Adaptation to Climate Change: Flood Responses at the Municipal Level in Norway.” Global Environmental Change 15 (2): 125–38. Nelson, Hal T., Adam Rose, Dan Wei, and Thomas Peterson. 2015. “Intergovernmental Climate Change Mitigation Policies: Theory and Outcomes.” Journal of Public Policy 35 (1): 97–113. Niinistö, Sini, Paavo-Petri Ahonen, Laura Aho, Veli Auvinen, Eleonoora Eilittä, Matti Kahra, Jaana Kaipainen et al. 2017. Finland’s Seventh National Communication under the United Nations Framework Convention on Climate Change. Helsinki: Ministry of the Environment and Statistics. Noll, Roger G. 1982. “Implementing Marketable Emissions Permits.” The American Economic Review 72 (2): 120–4. Nordhaus, William. D. 2007. “A Review of the Stern Review on the Economics of Climate Change.” Journal of Economic Literature 45: 686–702. Nurfatriani, Fitri, Dudung Darusman, Dodik Ridho Nurrochmat, Ahmad Erani Yustika, and Muhammad Zahrul Muttaqin. 2015. “Redesigning Indonesian Forest Fiscal Policy to Support Forest Conservation.” Forest Policy and Economics 61 (December): 39–50. Oates, Wallace E. 1972. Fiscal Federalism. New York: Harcourt, Brace, Jovanovich. ———. 1998. “Thinking about Environmental Federalism.” Resources 130. ———. 1999. “An Essay on Fiscal Federalism.” The Journal of Economic Literature 37 (4): 1120–49. ———. 2002. “A Reconsideration of Environmental Federalism.” In Recent Advances in Environmental Economics, John A. List and Aart de Zeeuw, eds., 1–32. Cheltenham, UK: Edward Elgar. Oates, Wallace E., and Paul R. Portney. 2001. “The Political Economy of Environmental Policy.” Discussion Paper 01-55, Resources for the Future, Washington, DC. Adapting Fiscal Decentralization Design to Combat Climate Change 72 Oates, Wallace E., and Robert M. Schwab. 1988. “Economic Competition among Jurisdictions: Efficiency Enhancing or Distortion Inducing.” Journal of Public Economics 35 (3): 333–54. ———. 1991. “The Allocative and Distributive Implications of Local Fiscal Competition.” In Competition Among States and Local Governments: Efficiency and Equity in American Federalism, Daphne Kenyon and John Kincaid, eds., 127–45. Washington, DC: Urban Institute. ———. 1996. “The Theory of Regulatory Federalism: the Case of Environmental Management.” In The Economics of Environmental Regulation, Wallace E. Oates, ed., 319–31. Cheltenham, UK: Edward Elgar. OECD (Organisation for Economic Co-operation and Development). 2018. Financing Climate Futures: Rethinking Infrastructure. Paris: OECD. ———. 2019. “Green Growth Challenge: Shifting the Tax Burden in Favour of Environmentally Related Taxation” (online). OECD, Paris, http://www.oecd.org/env/tools- evaluation/environmentaltaxation.htm. ———. 2020. “Growing Momentum for Sovereign Green Bonds” (online). OECD, Paris, http://www.oecd.org/coronavirus/en/data-insights/growing-momentum-for-sovereign-green- bonds. OECD and UCLG (United Cities and Local Government). 2019. “2019 Report of the World Observatory on Subnational Government Finance and Investment – Country Profiles.” OECD, Paris. Ogawa, Hikaru, and David E. Wildasin. 2009. “Think Locally, Act Locally: Spillovers, Spillbacks, and Efficient Decentralized Policymaking.” American Economic Review 99 (4): 1206–17. Plouin, Marissa, and Dorothee Allain-DuPre. 2018. Financing Climate Objectives in Cities and Regions to Deliver Sustainable and Inclusive Growth.” OECD Case Study Key Findings. OECD, Paris. Postula, Marta, and Katarzyna Radecka-Moroz. 2020. “Fiscal Policy Instruments in Environmental Protection.” Environmental Impact Assessment Review 84 (September): 106435. Poterba, James M. 1991. “Tax Policy to Combat Global Warming: On Designing a Carbon Tax.” NBER Working Paper 3649, National Bureau of Economic Research, Cambridge, MA. Przyluski, Valentin, and Stéphane Hallegatte. 2012. Gestion des Risques Naturels: Leçons de la Tempête Xynthia. Versailles: Éditions Quæ. Qi, Ye, Li Ma, Huanbo Zhang, and Huimin Li. 2008. “Translating a Global Issue into Local Priority: China’s Local Government Response to Climate Change.” The Journal of Environment & Development 17: 379–400. Ran, Qiying, Jinning Zhang, and Yu Hao. 2020. “Does Environmental Decentralization Exacerbate China’s Carbon Emissions? Evidence Based on Dynamic Threshold Effect Analysis.” Science of the Total Environment 721: 137656. Adapting Fiscal Decentralization Design to Combat Climate Change 73 Ring, Irene. 2002. “Ecological Public Functions and Fiscal Equalisation at the Local Level in Germany.” Ecological Economics 42 (3): 415–27. ———. 2008. “Integrating Local Ecological Services into Intergovernmental Fiscal Transfers: The Case of the Ecological ICMS in Brazil.” Land Use Policy 25 (4): 485–97. Saint-Amans, Pascal. 2020. “Supporting the Global Economy: What Role for Tax Systems in Responding to COVID-19?” (online). Organisation for Economic Co-operation and Development, September 3, https://www.oecd-forum.org/posts/supporting-the-global- economy-what-role-for-tax-systems-in-responding-to-covid-19. Sandalow, David. 2019. “Guide to Chinese Climate Policy 2019.” Columbia Center on Energy Policy, School of Public and International Affairs, Columbia University, New York. Sanoh, Aly. 2015. “Rainfall Shocks, Local Revenues, and Intergovernmental Transfer in Mali.” World Development 66 (February): 359–70. Santos, Rui, Irene Ring, Paula Antunes, and Pedro Clemente. 2012. “Fiscal Transfers for Biodiversity Conservation: The Portuguese Local Finances Law.” Land Use Policy 29 (2): 261–73. Sarmiento Barletti, Juan Pablo, Anne M. Larson, and Natalia Cisneros. 2018. “What Roles do Sub-National Governments Play in Nationally Determined Contributions? Between Rhetoric and Practice in REDD+ Countries.” Info Brief, Center for International Forestry Research, Bogor, Indonesia. Schröter-Schlaack, Christoph, Irene Ring, Thomas Koellner, Małgorzata Grodzińska- Jurczak, Rui Ferreira dos Santos, Paula Antunes, Pedro Clemente, Raphael Mathevet, and Maud Borie. 2014. “Intergovernmental Fiscal Transfers to Support Local Conservation Action in Europe.” Zeitschrift für Wirtschaftsgeographie 58 (2-3): 98–114. Shapiro, Joseph S., and Reed Walker. 2021. “Where is Pollution Moving? Environmental Markets and Environmental Justice.” NBER Working Paper 28389, National Bureau of Economic Research, Cambridge, MA. Sigman, Hilary. 2002. “International Spillovers and Water Quality in Rivers: Do Countries Free Ride?” American Economic Review 92 (4): 1152–9. ———. 2004. “Transboundary Spillovers and Decentralization of Environmental Policies.” Working Paper 10717, National Bureau of Economic Research, Cambridge, MA. ———. 2007. “Decentralization and Environmental Quality: An International Analysis of Water Pollution.” Working Paper 13098, National Bureau of Economic Research, Cambridge, MA. Silva, Emilson C. D., and Arthur J. Caplan. 1997. “Transboundary Pollution Control in Federal Systems.” Journal of Environmental Economics and Management 34 (2): 173–86. Sinn, Hans-Werner. 2012. The Green Paradox: A Supply-Side Approach to Global Warming. Cambridge, MA: MIT Press. Adapting Fiscal Decentralization Design to Combat Climate Change 74 Skidmore, Mark, and Hideki Toya. 2013. “Natural Disaster Impacts and Fiscal Decentralization.” Land Economics 89 (1): 101–17. Smoke, Paul, and Mitchell Cook. 2021. “Guidance Note on Administrative Decentralization and Climate Action.” Forthcoming. Smolka. Martim O. 2013. Implementing Value Capture in Latin America: Policies and Tools for Urban Development. Cambridge, MA: Lincoln Institute of Land Policy. Sobel, Russell S., and Peter T. Leeson. 2006. “Government’s Response to Hurricane Katrina: A Public Choice Analysis.” Public Choice 127: 55–73. Stern, Nicholas. 2006. Stern Review: The Economics of Climate Change. Her Majestry’s Treasury, London, http://mudancasclimaticas.cptec.inpe.br/~rmclima/pdfs/destaques/ sternreview_report_complete.pdf. ———. 2008. “The Economics of Climate Change.” American Economic Review 98 (2): 1–37. Stokey, Nancy. 1998. “Are There Limits to Growth?” International Economic Review 39 (1): 1–31. Su, Chi–Wei, Muhammad Umar, and Zeeshan Khan. 2020. “Does Fiscal Decentralization and Eco-Innovation Promote Renewable Energy Consumption? Analyzing the Role of Political Risk.” Science of the Total Environment 751. Takao, Yasuo. 2012. “Making Climate Change Policy Work at the Local Level: Capacity- Building for Decentralized Policy Making in Japan.” Pacific Affairs 85 (4): 767–88. Tang, Monica. 2011. “Carbon Pricing, Fairness, and Fiscal Federalism: A Cost-Sharing Proposal for Canada.” Queen’s Policy Review 2 (2): 17–36. Teng, Fei, and Alun Gu. 2007. “Climate Change: National and Local Policy Opportunities in China.” Environmental Sciences 4 (3): 183–94. Ter-Minassian, Teresa. 2007. “Fiscal Rules for Subnational Governments: Can They Promote Fiscal Discipline?” OECD Journal on Budgeting 6 (3): 1–11. Tiebout, Charles. 1956. “A Pure Theory of Local Expenditures.” The Journal of Political Economy 64 (5): 416–24. Tol, Richard S. J. 2005. “Adaptation and Mitigation: Trade-offs in Substance and Methods.” Environmental Science & Policy 8 (6): 572–78. UCM (United States Conference of Mayors) and C2ES (the Center for Climate and Energy Solutions). 2020. “American Mayors and Businesses: Building Partnerships for a Low- Carbon Future.” Volume 3, Alliance for a Sustainable Future, UCM and C2ES, Washington, DC and Arlington, VA. Adapting Fiscal Decentralization Design to Combat Climate Change 75 UNCDF (United Nations Capital Development Fund), UNDP (United Nations Development Programme), and UNEP (United Nations Environment Programme). 2013. “Financing Local Responses to Climate Change: Implications of Decentralisation on Responses to Climate Change.” UNCDF, UNDP, UNEP, New York and Nairobi. UNEP (United Nations Environment Programme). 2020. “Locust Swarms and Climate Change. Interview with Richard Munang, United Nations Environment Programme Expert on Climate and Africa” (online). UNEP, Nairobi, February 6, http://www.unenvironment.org/ news-and-stories/story/locust-swarms-and-climate-change. USAID (United States Agency for International Development). 2017. “Decentralized Governance and Climate Change Adaptation: A Case Study on Mali.” Technical Report, USAID, Washington, DC. Van den Brandeler, Francine, Joyeeta Gupta, and Michaela Hordijk. 2019. “Megacities and Rivers: Scalar Mismatches between Urban Water Management and River Basin Management.” Journal of Hydrology 573 (June): 1067–74. Victor, David G., and Mark Muro. 2020. “Cities are Pledging to Confront Climate Change, but are their Actions Working?” The Avenue. Cross-Brookings Initiative on Energy and Climate Paper Series, Brookings Institution, Washington, DC. Vidalenc, Eric. 2017. “Local Energy Planning in France: A Way to Reinforce Energy Transition.” Summer Study Proceedings, European Council for an Energy Efficient Economy, Stockholm. Weidner, Helmut, and Lut Mez. 2008. “German Climate Change Policy: A Success Story With Some Flaws.” The Journal of Environment & Development 17(4): Weitzman, Martin. 2007. “A Review of the Stern Review on the Economics of Climate Change.” Journal of Economic Literature 45 (3): 703–24. White, Roland, and Sameh Wahba. 2019. “Addressing Constraints to Private Financing of Urban (Climate) Infrastructure in Developing Countries.” International Journal of Urban Sustainable Development 11 (3): 245–56. Wilson, John D. 1999. “Theories of Tax Competition.” National Tax Journal 52 (2): 269–304. Winter, Jennifer. 2020. “Carbon Pricing in a Federal State: The Case of Canada,” ifo Institute DICE Report, Leibnitz Institute for Economic Research, Munich. Woods, Neal D. 2006. “Interstate Competition and Environmental Regulation: A Test of the Race-to-the-Bottom Thesis.” Social Sciences Quarterly 87 (1):174–89. World Bank. 2015. What Are Green Bonds? http://documents1.worldbank.org/curated/en/ 400251468187810398/pdf/99662-REVISED-WB-Green-Bond-Box393208B-PUBLIC.pdf ———. 2017. “The Philippines. Disaster Risk Management Development Policy Loan with a Catastrophe Deferred Drawdown Option.” Project Performance Assessment Report 114746, World Bank, Washington, DC. Adapting Fiscal Decentralization Design to Combat Climate Change 76 ———. 2018. “Strategic Use of Climate Finance to Maximize Climate Action: A Guiding Framework.” World Bank, Washington, DC. World Bank and PPIAF (Public-Private Infrastructure Advisory Facility). 2015. “What Are Green Bonds?” World Bank, Washington, DC. Yamagishi, Atsushi. 2019. “Transboundary Pollution, Tax Competition and the Efficiency of Uncoordinated Environmental Regulation.” Canadian Journal of Economics/Revue Canadienne d`Economique 52 (3): 1165–94. Zhang, Tao. 2020. “Sustainable Transformation of Societies – A Green Consensus for Macro-Fiscal Policies?” Remarks by Mr. Tao Zhang, Deputy Managing Director, International Monetary Fund, October 9, https://www.imf.org/en/News/Articles/2020/10/09/ sp100920-sustaincable-transformation-of-societies?cid=em-COM-123-42119. Zhang, Zhong Xiang. 2020. “Regional Pilots and Carbon Pricing in China.” ifo Institute DICE Report, Leibnitz Institute for Economic Research, Munich. Zhang, Jie, Yinxiao Qu, Yun Zhang, Xiuzhen Li 4, and Xiao Miao 5. 2019. “Effects of FDI on the Efficiency of Government Expenditure on Environmental Protection Under Fiscal Decentralization: A Spatial Econometric Analysis for China.” International Journal of Environmental Research and Public Health 16: 2496. Adapting Fiscal Decentralization Design to Combat Climate Change 77 A Adapting Fiscal Decentralization Design to Combat Climate Change