MSc Political Science
In the Political Economy Track
Thesis Submitted as Part of the Research Project: Our changing Global Economic
Order: Growing Risk and Imbalances (30ECT)
U.S. Climate Politics after Withdrawing from the Paris
Agreement:
How and Why did Local Governments Respond?
By
Dick Spruitenburg
Student no: 10470913
Submitted: 22/06/2018
Wordcount: 21.025
Supervisor:
Second reader:
3
4
I.
Abstract
In 2017 the Federal U.S. Government retracted their support of the Paris Climate Agreement. In response, several other U.S. governments at the level of states and cities declared that they would still be working towards the targets set out in the Agreement. This thesis studies the United States Climate Alliance and the ‘We Are Still In’ initiative as two cases of local governments voluntarily pledging to implement local climate policy. A quantitative analysis of the states in the U.S. Climate Alliance found support for the heavily politicized view of such a decision. No significant effects of local physical risks of climate change was found. Six ‘outlier states’ are more qualitatively studied. Findings are in line with the literature on behavioral political polarization in the United States. This thesis contributes to the understanding of how the subject of climate change often serves as a medium for the signaling of ideological and party-political identities between politicians and their electorate. A better understanding of this process can contribute to a better communication of the scientific consensus on anthropogenic climate change to the general public, and help the realization of climate change policies.
5
II.
Acknowledgements
This thesis is the final product of a very interesting time of study and personal development, and I would like to thank all those that played a role in it. I would specifically like to express my gratitude to my supervisor Lukas Linsi for his constructive critique throughout the development of this project. I would also like to thank the members of the research project, and the ‘Feckless Free-Riders’, for their friendship and support. Collectively we can do anything!
6
III. Table of Contents
I.
Abstract . . . 4II.
Acknowledgments . . . 5III.
Table of Contents . . . 61.
Introduction . . . 72.
The Paris Agreement and U.S. Climate Politics . . . 92.1 The Paris Climate Agreement . . . . 9
2.2 US Climate Politics . . . 11
3
Literature Review . . . 17 3.1 Political dimension . . . 17 3.2 Economic dimension . . . 28 3.3 Physical dimension . . . 31 3.4 Literature conclusion . . . 324
Quantitative Analysis . . . 344.1 Methods and hypotheses . . . 34
4.2 Data . . . 37
4.3 Analysis and results . . . 43
4.4 Discussion of quantitative results . . . . 54
5
Qualitative Analysis . . . 595.1 Republican states in the U.S. Climate Alliance . 59
5.2 Democratic states not in the U.S. Climate Alliance . 63 5.3 Discussion on qualitative analysis . . . 69
6
Conclusion . . . 727
Limitations and future research . . . 76IV.
Appendix . . . 787
1. Introduction
On the 22nd of April 2016 the Paris Agreement was signed by nearly 200 countries. The agreement is said to be the most ambitious multilateral agreement aiming to combat climate change ever agreed upon. Despite this, the agreement has also received criticisms for not being ambitious enough and for being largely legally non-binding (Clemençon, 2016). Among the signees of the agreement was the U.S. federal government headed by former President Barack Obama. However, after the eventful times of the 58th presidential elections of 2016, 2017 saw the U.S. Federal government retracting their support of the agreement, arguing that the agreement “disadvantages the United states to the exclusive benefit of other countries” and stating that: “the United States will withdraw from the Paris Climate Accord […] but begin negotiations to reenter either the Paris Accord or a really entirely new transaction on terms that are fair to the United States, its businesses, its workers, its people, its taxpayers” (Trump, 2017).
In response, several lower U.S. governments at the level of states, counties, and cities and municipalities declared that they would still be adhering to the Paris Agreement despite the federal government’s decision. Two examples of such efforts are the “We Are Still In” initiative, supported by U.S. state, county, and municipal governments (WASI, 2018) which jointly represent around a third of the total U.S. population. The United states Climate Alliance is another bipartisan coalition of governors committed to reducing greenhouse gas emissions consistent with the goals of the Paris Agreement, signed by 17 U.S. state governors (USCA, 2018).
The series of events described above pose an interesting case for the study of collective action and multilevel climate governance, as it provides data across levels of government and geographic areas. This thesis adds a case study of the ‘We Are Still In’ initiative and the United States Climate Alliance to the growing literature on a local government’s decision to implement climate policies. The aim of this study is to identity the divers of this governance decision, and to see if it is driven solely by political motivations, or whether economic considerations and direct local physical risks of climate change also play a role. In this way, this study also adds to the growing literature on the politicization of climate change and political polarization in the United States.
The primary research question this thesis aims to answer is: How can we explain the observed different levels of stated compliance to the Paris Climate Agreement by local governments in the
8
U.S., in response to the U.S. Federal government’s retreat from the agreement. This question is first addressed with quantitative methods, using a logistic regression model to identify what variables have a significant effect on a government’s odds of joining the U.S. Climate Alliance and the ‘We Are Still In’ initiative. Furthermore, qualitative methods are used to study interesting cases selected from the quantitative analysis.
This thesis will proceed as follows. Chapter one provides the primary context of international climate politics, and U.S. climate politics in general. The first half of this chapter focusses on the details of the Paris Accord and how it relates to the other previous UN climate negotiations. The second half will focus on climate politics in the U.S. specifically, detailing the recent history of climate policy in the U.S. under the Clinton administration (1993-2001), G.W. Bush administration (2001-2009), Obama administration (2009-2017), and the Trump administration (2017-present). The second chapter provides the theoretical basis for this thesis, detailing the theoretical literature on collective action and multi-level governance, and discusses the relevant literature on the three dimensions of the case study: political, economic and physical. The fourth chapter presents the research question, hypotheses, and the quantitative analysis of this thesis. Specific states that deviate from what can be expected or that otherwise stand out in the analysis of the fourth chapter are qualitatively discusses in the fifth chapter.
9
2. The Paris climate agreement and U.S. climate politics
In this chapter I aim to provide the context on both international climate politics as well as the state of climate politics in the United States. This chapter consists of two sections. First, I provide a review of the general outcomes of the 2015 UN Climate Change Conference, specifically showing how the Paris Conference differed from previous conferences such as Kyoto (1997) or Copenhagen (2009). The second part of this chapter focusses on both the domestic and international climate politics of the United States, arguing that the turn-around regarding the Paris agreement after the 2016 election was not so much a deviation from ‘business as usual’ in U.S. climate policy, but instead a continuation of what appears as a cyclical trend. I argue that a key feature of this cyclical trend of implementation and consequent pushback against climate change regulations is a potential role local governments can play in protecting existing regulations, or implementing their own. These dynamics of domestic political competition on climate change policies will be discussed in context of the coalitions of governors and mayors that declared they will continue to support climate action to meet the Paris Agreement, such as the ‘We Are Still In’ initiative and the United States Climate Alliance.
The Paris Climate Agreement
Clémençon (2016) provides an extensive overview of the outcomes of the Paris Climate Agreement and provides the necessary historical context to show why the agreement can be seen both as an historic breakthrough and a failure at the same time. The Paris Climate Agreement had a different structure than the previous climate conferences. The conference was built entirely around voluntary pledges by participating nations. These pledges, or intended nationally determined contributions (INDCs), were submitted before the conference started with the goal of making each countries’ commitments known before going into negotiations. This is the crucial difference between the Paris Agreement and the Kyoto protocol; the INDCs do not constitute legally binding emission targets, instead focusing legal attention to the process of periodically reviewing such pledges and the transparent monitoring of national emissions. Clémençon (2016) argues that this was ultimately “the world accepting the domestic constraints in the United States as a feature of international climate talks” (p. 6), and a result of President Obama’s efforts to design the Paris Agreement in such a way that he would not have to get the agreement approved by the
10
Republican controlled U.S. Senate. Clémençon goes on to suggest that this concession to the position of the U.S. was made with the disparaging realization that, essentially, any agreement is better than no agreement at all and that the Paris Agreement could go on to serve as a historic event that could inspire more public and private sector action.
The Paris Agreement (UN, 2015) stipulates three crucial provisions that define the outcome of the conference. First, Article 2 (p.3) of the agreement specifies the most ambitious goal any climate agreement has set: “holding the increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C”. Secondly, Article 4 (p. 4) sets the general long-term objective of reaching peak emissions as soon as possible, and achieving a balance between anthropogenic emissions and removals in the second half of the century. Thirdly, Article 4 (p.5) continues to stipulate that countries will formulate and communicate long-term low greenhouse gas emission development strategies, to update actions every 5 years, making future steps at least as strong as current ones. Of such plans to reduce the emissions of greenhouse gasses the EU pledged the most ambitious INDC, aiming to reduce emissions by 40% from 1990 levels by 2030 (CPO, 2015). The U.S. pledged to reduce 26-28% of 2005 levels by 2025, and China made a more vague pledge to reduce its carbon intensity of the economy by 60 – 65% compared with 2005, by 2030.
Even though the goals set out in the Paris climate agreement are the most ambitious yet, they still rely on individual states to take action themselves without having an enforcement mechanism in place to ensure states will fulfill their pledges. In addition, it has been shown that the pledges made for the Paris agreement will not be sufficient to keep global temperatures from rising above the 1.5 or 2°C threshold, even if all countries would fulfill their commitments completely. Rogelj et al (2016) estimate that all INDCs (including the pledge made by the U.S. government under the Obama administration) will collectively lower greenhouse gas emissions to levels that imply a median warming of between 2.6 and 3.1 degrees Celsius by 2100. So even before the newly elected U.S. government withdrew from the agreement the goals set were not accompanied by the (non-binding) commitments necessary to achieve those goals. Any optimism over the prospect of reaching the targets seems misplaced, however, Rogeli et al. do state that: “substantial enhancements on current INDCs by additional national, sub-national and non-state action is required to maintain a reasonable chance of meeting the target” (p. 632), echoing Clémençon’s
11
argument that the agreement could prove to be an inspiration for more action in the private sector, or in other levels of the public sector.
U.S. Climate Politics
The second part of this chapter focusses on the domestic and international climate politics of the United States. Since climate change came on political agendas in the 1980s the United States have not followed a long-term strategy on climate policy. Instead, their climate policies have been highly political, with several presidents now implementing and repealing climate policies. In this section I will elaborate on this history to situate the events following the U.S. withdrawal from the Paris Agreement in its historical context, and elaborate on the implementation of climate policy in the U.S.. Subsequently I will present an overview the series of events following the withdrawal of the United States from the Paris Agreement, the central case study of this thesis.
Climate action under Clinton and G.W. Bush
The United States was a signatory to the Kyoto Protocol under Clinton Administration. Reluctantly agreeing to the binding targets and emission reduction timetables for individual countries the Protocol prescribed. For this concession, they introduced the emissions trading system. The Kyoto Protocol, however, was never ratified by the U.S.. In 1997 the U.S. Senate unanimously adopted the Byrd-Hagel Resolution which stated that it was not the sense of the Senate that the U.S. should be a signatory to the Kyoto Protocol (Byrd & Hagel, 1997), and in 2001 the Bush administration stated it would not be implementing the treaty, arguing it would disproportionately harm the U.S. economy (BBC, 2001). This sequence of events closely resembles the events following President Obama’s signing of the Paris Agreement.
Conversely, the development of climate policy did not stop with the rejection of the Kyoto Protocol. In 2006, Massachusetts, 11 other states, and several U.S. cities sued the federal Environmental Protection Agency (EPA), asking the EPA to regulate emissions of CO2 and other greenhouse gases from new motor vehicles under the Clean Air Act, which states that “Congress must regulate ‘any air pollutant’ that can ‘reasonably be anticipated to endanger public health or welfare’” (LII, 2018). The case was taken to the Supreme Court, where the court was asked to decide on two questions: may the EPA decline to issue emission standards for motor vehicles based on policy considerations not enumerated in the Clean Air Act? And, does the Clean Air Act give
12
the EPA authority to regulate carbon dioxide and other greenhouse gases? (SCOTUS, 2007). The court ruled against the first question, and in favor of the second, by a vote of 5-4. The majority ruled that if the EPA wanted to continue inaction on regulating CO2 emissions, the Clean Air Act required the agency base that decision in a consideration of "whether greenhouse gas emissions contribute to climate change." (LII, 2018).
Climate action under Obama
Clémençon (2016) has argued that President Obama successfully negotiated non-binding targets with the specific aim of not having to get the agreement approved by the Republican-held U.S. Senate. This meant President Obama was able to implement climate policies despite reluctance from the legislative branch of government to do so. This, however, also meant that subsequent Presidents would be able to start repealing these policies as soon as their first day in office. Obama’s Climate Action Plan, the set of policies that were set out to realize the U.S.’s INDC, was presented in 2015 (White House Office of the Press Secretary, 2015). The primary policy to reduce greenhouse gas emissions was the Clean Power Plan. The plan aimed to regulate the emissions by power plants, which are responsible for about 35% of total U.S. CO2 emissions (UCSUSA, 2017). The plan set out carbon emission reduction targets for each state and a flexible framework in which states could meet their target themselves by investing in renewable energy, energy efficiency, natural gas, nuclear power, and shifting away from coal-fired power. These targets differ across states due to each state’s unique profile of energy production and differing potentials for emission reductions. Other measures to reduce emissions included the introduction of standards for heavy-duty engines and vehicles, energy efficiency standards, and other economy-wide measures to reduce other greenhouse gases (White House Office of the Press Secretary, 2015).
Trump’s turn-around
On 28th of March, 2017, President Donald Trump issued an executive order ‘On Promoting Energy Independence and Economic Growth’ (Trump, 2017), ordering the revision of “existing regulations that potentially burden the development or use of domestically produced energy resources” and suspension of “[regulations] that unduly burden the development of domestic energy resources”. However, it also states that “necessary and appropriate environmental regulations” should “comply with the law, [be] of greater benefit than cost, when permissible,
13
achieve environmental improvements for the American people, and are developed through transparent processes that employ the best available peer-reviewed science and economics”. On the 16th of October the EPA announced its intention to repeal the Clean Power Plan, (EPA, 2017). The repeal of the Clean Power Plan is not the only channel through which the Trump administration is backpaddling on environmental protection regulation. In the proposed budget for 2019 the EPA’s enforcement budget has been lowered with 17% from the previous year’s spending levels, and the administration has set the goal of shrinking the total number of people employed by the EPA by 20% (Schaeffer & Pelton, 2018). This shift in priorities away from environmental protection can also be seen in the number of cases the EPA has lodged during the first year President Trump was in office. Table 1 compares the number of cases taken on by the EPA and the total amount in penalties collected, during the last four Presidents’ first year in office. It shows that the EPA, in the first year of the Trump administration, lodged 32% fewer cases than they did in the first year of the Obama presidency, and collected 63% less in penalties from polluters.
Administration Cases Lodged Penalties (millions USD, inflation adjusted)
Clinton 73 93
Bush 112 70
Obama 71 81
Trump 48 30
Table 1 - Civil Cases and Penalties Against Polluters in the First Year in Office (Schaeffer & Pelton, 2018)
Due to the recent nature of these developments little research has been done into the current developments at the EPA under the Trump administration. One study has been published, specifically stating their research agenda included “examining long-term changes to federal science and environmental policy, but, perhaps more unusually for a scholarly project, we have aimed to publicly release our findings in a timely manner” (Dillon et al, 2018, p. 90). The study reviews news articles, public documents, and rapid response interview study of current and retired EPA employees in order to critically assess the changes to the EPA in the first six months of the
14
Trump administration. The authors find that current changes to the EPA have broken with the Obama administration, but also with practices of all past administrations since the conception of the EPA in 1970. The authors find that the Agency seeks to “‘undo, delay or otherwise block’ at least 30 existing rules that seek to protect health and the environment” (p. 92). Along with the repeal of the Clean Power Plan and the budget cuts, these measures include the political appointments people with deep ties to regulated industries and restructuring of the agency’s science advisory boards as to allowing lobbyists on the scientific advisory boards. The authors conclude that the Trump Administration is actively enabling regulatory capture, stating the public and environmental health consequences of this practice to be several and far-reaching (p. 89). Response to the Federal Withdrawal
In response to the withdrawal of the Federal U.S. Government from the Paris Agreement two initiatives were started by lower level governments to continue to work towards the goals set in the Paris Agreement. This section briefly overviews the goals and structures of these initiatives. The primary case study of this thesis is the United States Climate Alliance (USCA, 2018), an alliance of State Governors pledging to work towards the Paris Agreement. The ‘We Are Still In’ initiative (WASI, 2018) is a similar initiative that includes lower level municipal governments. Focus was places largely the U.S. Climate Alliance due to constraints on the availability of data on the level of the municipality and the easy access to data at the state level. Therefore, the analysis at the municipal level serves as a check on the result at the state level and a starting point for future research.
The U.S. Climate Alliance is a subnational coalition of U.S. States, founded by the three Democratic Governors from New York, Washington, and California, on June 1st, 2017. Within the next week 9 other states joined, of which Massachusetts and Vermont were the first two Republican-led states to do so (AP, 2017). Since then, the Alliance has grown to a total of 16 State Governors, plus the Governor of the U.S. territory of Puerto Rico. Like the Paris Agreement, the Climate Alliance is not legally binding. Members of the Alliance submit to three core principles, and three commitments (USCA, 2018):
“States are continuing to lead on climate change: Alliance states recognize that climate change presents a serious threat to the environment and our residents, communities, and economy.
15
State-level climate action is benefiting our economies and strengthening our communities: Alliance members are growing our clean energy economies and creating new jobs, while reducing air pollution, improving public health, and building more resilient communities.
States are showing the nation and the world that ambitious climate action is achievable: Despite the U.S. federal government’s decision to withdraw from the Paris Agreement, Alliance members are committed to supporting the international agreement, and are pursuing aggressive climate action to make progress toward its goals.”
Specifically, the members of the Alliance commit to:
“Implement policies that advance the goals of the Paris Agreement, aiming to reduce greenhouse gas emission by at least 26-28 percent below 2005 levels by 2025
Track and report progress to the global community in appropriate settings, including when the world convenes to take stock of the Paris Agreement, and
Accelerate new and existing policies to reduce carbon pollution and promote clean energy deployment at the state and federal level.”
The ‘We Are Still In’ initiative is a broader initiative to which, at the moment of writing, 1905 businesses and investors, 265 cities and counties, 345 colleges and universities, 11 cultural institutions, 21 faith groups, 10 U.S. states, and 8 Native American tribes have subscribed (WASI, 2018). These various organizations have undersigned an open letter to the international community and parties to the Paris Agreement, declaring their continued support to climate action and the goals set in the agreement.
The open letter is less specific than the commitments made by the U.S. Climate Alliance. The signatories emphasize the importance of the voluntarily set emission reduction targets and states that it were the local, tribal, and state governments that were responsible for recent decreases in greenhouse gas emissions. The most concrete commitment in the letter is the declaration by the signatories that they will “continue to support climate action to meet the Paris Agreement” and that they “will remain actively engaged with the international community as part of the global
16
effort to hold warming to well below 2°C” (WASI, 2018). The choice for this vague formulation of commitments was likely made to accommodate the broad scope of signatories to the letter, all having different possibilities and responsibilities. This makes the initiative largely symbolic, but this is no peculiarity in the context of the voluntary targets of the Climate Agreement.
This chapter has provided a short overview of the volatile history of the politics of climate change in the United States. While the election of Donald Trump as President and the subsequent changes to environmental policies have marked a clear turning point in this history. They also clearly fit a larger narrative of the implementation and subsequent negative political response to climate change policies. The evidence presented serves as the historical context for the rest of this thesis, to which I refer throughout the following chapters. The next chapter reviews the scientific literature on the subject of climate change politics, implications of climate change and climate change policy for economic development and the role of physical risks on individual perceptions of climate change.
17
3. Literature review
In this chapter I discuss the theoretical frameworks of multilevel governance and rational choice institutionalism, as well as the empirical scientific literature on climate change and climate change governance. This chapter is divided by the three general dimensions of study regarding this topic: the political, the economic, and the physical dimension. This chapter will serve as the basis for the hypotheses of the empirical analysis done in chapter 4 and the case selection for the qualitative analysis of chapter 5.
Political dimension
Multilevel Governance of Global Climate Change
The concept of multilevel governance was first developed by in the widely cited article Unraveling the Central State, but How? Types of Multi-level Governance by Hooghe and Marks (2003). The authors observed a growing attention from scholars in the social sciences to the reallocation of authority away from central states and noted the lack of a clear consensus on how these reallocations can be organized. They propose that discussions on the jurisdictional design of decision-making diffused away from the central state can be conceptualized as two contrasting types of governance. They call these types of governance type I and type II governance. In this section I will first provide an overview of the core concepts of multilevel governance and how they have been applied. I start from the seminal paper by Hooghe and Marks (2003) using their definition of the two types of multilevel governance. I will draw on the extensive review of the multilevel governance literature by Stephenson (2013) for an overview of the uses of multilevel governance and a view on the direction of prospective research in the field. Lastly, I pay specific attention to the practical application of the multilevel governance approach in the context of climate change policies (Betsill & Bulkeley, 2006). I conclude this section with a short discussion on the applicability of the concept of multilevel governance to the central case study of this thesis. Hooghe and Marks (2003) provide the first binary typology multilevel governance. The two types of multilevel governance, type I and type II, are conceptualizations from four normative questions on the design of institutions of multilevel governance. The authors argue that empirically, the answers to these questions cluster together to form two distinct types of institutions of multilevel governance:
18
• Should jurisdictions be designed around particular communities, or should they be designed around particular policy problems?
• Should jurisdictions bundle competencies, or should they be functionally specific?
• Should jurisdictions be limited in number, or should they proliferate? • Should jurisdictions be designed to last, or should they be fluid?
(Hooghe & Marks, 2003, p. 236) Type I multilevel governance is designed around a particular community and has a limited number of jurisdictions in the range of international, national, regional, meso, and local. These jurisdictions do not serve one singular purpose, but instead bundle together multiple functions and policy responsibilities. Thus, they are more general-purpose institutions. Generally, these institutions are designed to last, are characterized by nonintersecting memberships, and adopt Montesquieu’s trias politicas. The foundations for type I multilevel governance is federalism with the sharing of power amongst a number of governmental institutions at different levels (p. 236).
Type II governance is more fragmented and designed around specific functionalities like the provision of a local service or governing a local common pool resource. This means that citizens are no longer served by a singular government even at a single level, but instead they are served by specific “public service industries” (p. 237). The authors note that this form of governance is widespread at the local level, calling on specific examples in Switzerland. In the United States, special districts in metropolitan areas are the closest functional equivalent. They have intersecting territorial boundaries and perform tasks specific to the metropolitan area they are operating in. Stephenson (2013) provides a comprehensive analysis of the study and development of multilevel governance, both as an empirically observable trend in governance and as concept in the political sciences. He identifies five main uses of multilevel governance. First, he acknowledges that Marks and Hooghe are “arguably the king and queen of multilevel governance” (p. 818), and notes how they first developed the concept of multilevel governance as a reaction to what they saw as an inadequate response of political science to the development of European regional cohesion policy. He argues that in this field the “established concepts [were being stretched] over the new phenomena” (p. 820), where political scientists that were studying federalism sought to explain
19
these developments within and between states, International Relation scholars extended and adapted their theories of international regimes to account for the newly observed diffuse authority within states.
These responses were attempts to explain more flexible arrangements of governance that were emerging. Stephenson calls these the original use of multilevel governance. The second use of multilevel governance is a functional use that emerged from 1997 onwards. As the concept was being used more to analyze institutional arrangements, the concept was soon picked up by policy makers themselves. Stephenson notes that multilevel governance, in this way, started to be picked up as a concept on institutional design and the process of decision making.
Stephenson calls the third use of multilevel governance ‘combined uses’, as the concept was now being challenged by other rival governance frameworks. Interestingly, Stephenson identifies rational choice institutionalism as the first serious alternative for the analysis of European cohesion policy (p. 825). Fourth, multilevel governance was beginning be picked up and criticized from a position on normative democracy, arguing the systems of multilevel governance in the EU were organized around self-organizing and self-regulating networks, with governance essentially about co-operation and co-ordination, and with traditional government systems of undermined or bypassed.
The 2006 study by Betsill and Bulkeley apply the multilevel governance approach to the study of global climate governance. They study the Cities for Climate Protection (CCP) program and how the program can be conceptualized. They argue that the older and established theories of International Relations, regime theory and transnational networks, are only partially useful for their task. Instead, the authors propose to use a multilevel governance perspective to describe the observed processes and institutions of governance that are operating and interacting at various scales.
The CCP program was established in 1993 as a program of the International Council for Local Environmental Initiatives (Betsill & Bulkeley, 2006, p. 143). The ICLEI is one of the largest transnational networks of subnational governments and launched the CCP program as a platform for local governments to develop strategies for the reduction of greenhouse gas emissions. Over 675 local governments from across the world are part of the program, with 159 U.S. governments representing around 20% of the total U.S. population joining the program by 2006 (ICLEI, 2006,
20
p. 2). Governments that join the program are asked to implement a five-milestone program, which ranges from conducting a baseline inventory and forecast of emissions to the monitoring and verification of the results (ICLEI, 2006, p. 5). The CCP network assists their signatories by offering technical support, quantification tools, best practice examples, policy templates, trainings and workshops, and communication with the other members that are working to combat climate change (ICLEI, 2006, p. 6).
Betsill and Bulkeley (2006) look at international regime theory to see how well it can be used for the conceptualization of an institution like the CCP. They argue that many scholars of international relations have studied global environmental governance by focusing on the interaction of nation-states through the formation of international regimes such as the UNFCCC and its Kyoto Protocol. This approach is closely connected to the view of climate change as a collective action problem, where these international regimes are formed as the institutions that facilitate cooperation and reduce transaction costs. The authors identify the weaknesses implied by the underlying assumption that equates political power with the nation-state (p. 146): the regime approach emphasizes that nation-states, with their monopoly on military and economic force, are the most significant actors on the global stage. This unitary view are more in line with a ‘government’ perspective than a ‘governance’ perspective. It treats nation-states as unitary agents and fails to account for intra-state political forces.
The authors also look at the more novel International Relations literature on transnational networks, to see how well it can be applied to institutions like the CCP. This literature challenges the hierarchical structure of authority of the regime theory approach, instead promoting a view of power as accumulating from multiple sources of authority that can include expertise and moral positions (p. 148). In this literature the concept of epistemic communities has been used to describe “networks of experts who share a common understanding of the scientific and political nature of a particular problem, […] gaining influence within international regimes by virtue of their authority” (Betsill & Bulkeley, 2006, p. 147). The concept of transnational advocacy networks have been used to describe “relevant actors working internationally […] bound by shared values, common discourse, and a dense exchange of information and services” (Betsill & Bulkeley, 2006, p. 147). The concept of a “global civil society” is the most radical reexamination of transnational networks, moving away from the state-centered analyses and instead arguing that “Governance occurs on a
21
global scale through both the co-ordination of states and the activities of a vast array of rule systems that exercise authority in the pursuit of goals that function outside normal national jurisdictions” (Betsill & Bulkeley, 2006, p. 148).
Betsill and Bulkeley argue that the transnational municipal network they study, the CCP, cannot be easily categorized using the concepts from the International Relations literature. While the authors agree that the development of the concept of transnational networks provides a necessary shift in focus away from hierarchical structures and towards more networked forms of organization, they argue that the literature focusses too much on nonstate-actors to be directly applicable to their study of governmental networks such as the CCP. Instead, they propose to take a multilevel governance approach, that views transnational networks as multifaceted, with features of both state and non-state organizations.
Betsill and Bulkeley (2006) contend that the international politics and governance of climate change is a complex and multilevel process, that transcends the traditional distinctions between state and non-state actors, local national and global scales, and international and domestic politics. Multilevel governance creates space for considering the role subnational governments play in the implementation of (inter)national policies and emphasizes the multiplicity of forms government and governance take in world politics. Leaving the traditional hierarchical view of regime theory, multilevel governance looks at international (climate) politics as an arrangement of overlapping and interconnected spheres of authority that operate semi-independent from each other. They argue that the CCP project is an example of such a governance structure where local authorities operate at international level.
The framework multilevel governance is useful for this thesis’ study of the U.S. Climate Alliance because of the analytic flexibility it provides in terms the classification of governance organizations. First, the federal structure of the United States plays an obvious crucial role in the disagreement between the Federal Government and the states regarding the implementation of climate policy. However, this disagreement is a domestic dispute taking place in an international context. The framework of multilevel governance provides a useful addition to the traditional framework of federalism to account for this. The members of the Climate Alliance are U.S. states that exist in the federal system. This arrangement, with jurisdictions formed around particular
22
communities, who are limited in number, ranging between federal, state and local, not serving one singular purpose, is a typical type I multilevel governance system.
This federal structure is nothing new and does not necessarily require the multilevel governance lingo. However, the primary subject of this thesis is the formation of the U.S. Climate Alliance as a political institution. States voluntarily formed a domestic network to cooperate and coordinate the implementation of policy aiming to meet the goals set in a large international multilateral agreement. The commitments these states made are very similar to those made by the nation states in the Paris Agreement: the states make a legally non-binding commitment to work towards the self-determined emission reduction targets, while tracking and reporting progress to the global community (USCA, 2018). It is this development, where sub-national governments form new networked institutions that work directly with the international community, independently from their Federal Government, that makes the framework of multilevel governance useful for this thesis’ analysis.
Local Climate Policies
Several reports on the potential of local governments in combatting climate change have been written. For example, the OECD published a report outlining the importance of cities in addressing climate change. Cities form the natural hubs of economic activity and therefore consume the vast majority of global energy and are major contributors of greenhouse gas emissions (Kamal-Chaoui & Roberts, 2009, p. 9). This sense of importance of lower level localities is echoed in the 2010 World Bank report Cities and Climate Change: An Urgent Agenda (World Bank, 2010). The report outlines how cities contribute to and are affected by climate change, how policy makers can use cities to change behaviors, improve technology related to climate change, and how cities can use these challenges as opportunities to improve their community (p. 1).
But why would local governments implement climate change policies? Engel and Orbach (2008) studied the micro-motifs of state and local governments in participating in climate change initiatives. They also observed that the majority of states and hundreds of cities were taking action to combat climate change, while the federal U.S. government has remained largely passive. They aim to answer a question closely related to this thesis’s research question, namely: “Why are citizens and politicians willing to shoulder the entire cost of local climate change initiatives when they must share the benefits with others planet-wide?” (p. 119). By studying potential incentives
23
for supporters of local climate change initiatives they propose several explanations for the support for local climate change action, based both in the traditional neoclassical model of an informed utility-maximizing decisions, and in specific cognitive biases that affect individual decision-making. They identify demand-sided motifs such a tendency to underestimate costs, overestimation of benefits and effectiveness of relatively small local reductions in emissions, or the view of local climate action as a second-best alternative to the first-best option of federal and international climate policies that seems to be unattainable. The authors also identify several supply-sides policy motifs, including political entrepreneurship over the ‘hot issue’ of climate change or the possibility of attracting new ‘green’ businesses to the local constituencies (p. 134). These mixed motifs bring up the question of the effectiveness of local climate policies. Bushnell, Peterman, and Wolfram (2008) study the possibilities local governments have and how effective these policies can be. They argue that ‘leakage’, the relocating of polluting activities to other jurisdictions with less stringent regulations is one of the main limitations of local climate policies. However, the upside of this mechanic is that subsidies for clean industries can attract new businesses. They conclude that local policies can make a contribution, but are also prone to evasion. This adds a significant symbolic element to local climate actions. However, the authors do note that local policies can function as a demonstration of various new technologies, and that “their success or failures could be important first steps in adopting effective low-carbon technologies on a more global scale”.
A similar argument is made by Nobel Memorial Prize laureate in Economics Elinor Ostrom (2010). She argues that local climate action can play a crucial role in the development of larger scale policies, and that the problem of climate change is too big to be waiting for an internationally binding agreement to start addressing it wherever possible.
The politicization of climate change
Engel and Orbach (2008) already touch upon the symbolic political dimension of local climate change policies, also proposing that local politicians can use the subject as an agenda to profile themselves (p. 134). In the U.S. specifically there is a large political dimension to the subject of climate change, and various studies of the political polarization on climate change have been done.
24
McCright and Dunlap have done multiple such studies. In their 2011 study, they find that the public’s views on climate change are largely corresponding to ideological and party positions (2011). In turn, they argue, that these positions have polarized over the subject since the Reagan administration labeled environmental regulations as a burden on the economy (2010) . In their 2011 study, the authors analyze combined data from 10 polls between 2001 and 2010 with between 1000 and 1060 respondents. They conclude that liberals and Democrats are more likely to express beliefs consistent with the scientific consensus on anthropogenic climate change and are more likely to express concern over the issue than are conservatives and Republicans. They also find that this divide has grown substantially over the past decade. They note that both sides of the political spectrum receive conflicting information on global warming, which helps to reinforce existing political differences. Moreover, they stress that the effect of educational attainment on personal concern about climate change are positive for liberals and Democrats, but are weak or negative for conservatives and Republicans. This challenges the common assumption that more education on the subject will communicate the scientific consensus to the general public.
The different stance on the issue of climate change between the two major parties in the U.S. is often attributed to the fact that the climate skeptical Republican party holds closer ties with the polluting corporate sector. However, Collomb (2014) argues such explanations should be complemented by two other important factors. The author argues that there are two ideological dimensions to the efforts to counter climate change action (p.2): the conservative ideology of laissez faire economic policies and a general dedication to an permanently expanding economic prosperity and consumption as the ‘American way of life’.
Firstly, the political conservative ideology that is traditionally dedicated to little government intervention in the market and free enterprise as a goal. Collomb (2014) argues that recognizing climate change for the global threat that the scientific community agrees it is, would mean a direct contradiction with the central ideological belief of the movement. This explanation also accounts for the observation of other forms of climate change denial; arguing the threat is not as large as people make it seem, or that the problem is not anthropogenic but caused by natural processes. These forms do not deny climate change outright, but fabricate an alternative explanation in order to defend the ideological stance of governmental non-intervention.
25
Figure 1. American’s beliefs about climate science from 2001-2010, by political ideology and party identification (McCright & Dunlap, 2011, p. 193)
In addition to the prevalence of the laissez fair doctrine in American politics, Collomb (2014) argues that there is a second ideological influence that can explain the observations of persistent climate change denial: The ‘American way of life’. The authors notes that the average voter values a clean environment highly (Stimson, 2009), but is not willing to accept the far-reaching changes in policy and lifestyle (e.g. higher prices of gasoline) to attain this goal. He attributes this to the ideology of the ‘American way of life’, that dictates a dedication to permanently expanding economic prosperity and consumption (p. 10). Dramatically reducing CO2 emissions implies a radical changes to the global economy, which will ultimately also affect consumption patterns and personal habits. The author notes that even the strong proponents of climate action like Al gore
26
have often times ignored this implication by focusing for example on the potential for “better jobs, new technologies, new opportunities for profit, and a higher quality of life” (Gore, 2007).
Polarization over the issue of climate change should also be seen in context of the debate in political science regarding political polarization more generally. The literature is divided over the issue of political polarization, with some authors arguing that the American public has polarized and is continuing to grow apart politically (Abramowitz, 2010; Jacobson, 2004; Hetherington, 2001), and others arguing that these observed trends reflect polarization of the American politicians and political parties, not polarization in the American electorate (Fiorina & Abrams, 2008; Levendusky & Pope, 2011)
Mason (2013) argues that these two positions stem from a lack of a common definition of the concept of polarization, and a lack of sufficient theoretical underpinnings (p. 141). She argues that the two camps address two different types of polarization: issue position polarization and behavioral polarization. Mason defines the former as an increase in the extremity of issue positions in the public, and the latter as an increase in partisan identification, strength, and bias. Looking at survey data from the American National Election Studies between 1972 and 2004, she finds that behavioral polarization has increased, but not issue position polarization. She explains this by referring to an increase in alignment between partisan and ideological identities in the United States, studied by several other authors (Levendusky, 2009; Abramowitz, 2010). Mason argues that this alignment between ideological identities and the political parties has increased intergroup bias (Hewstone, Rubin, & Willis, 2002), and caused political polarization in the American public despite there being no real shift in issue positions.
Mildenberger, Marlon, Howe and Leiserowitz (2017) study the spatial distribution of partisan climate and energy opinions. They find that Democratic party members more consistently support climate policies, finding only geographic variation in the intensity of these beliefs on climate. Amongst Republicans opinions on these issues vary greatly across state and congressional districts. Interestingly, the authors find that while large parts of Republicans do not believe climate change is anthropogenic, they do widely support renewable energy funding and carbon pollution regulation.
The study by Webern and Stern (2011) specifically addresses political polarization over climate change. They ask the question why U.S. public opinion has not undergone the same process of
27
solidifying opinions as has happened in the scientific community. They argue that several factors have played into this, including a mismatch between people’s usual modes of understanding and the one that is needed and a societal struggle surrounding the framing of the issue (p. 315). They emphasize the role of the media as an imperfect mediator between the public and the scientific community and point to political influences on this process. When one aims to improve public understanding of climate change, the authors note, it is important to understand that it is not a problem of a deficit of knowledge that can be filled through education. Instead, they argue that doubts over anthropogenic climate change are a different understanding of the same information. They emphasize the importance of framing, proposing that climate change responses will be more likely to convince people when presented as options for risk management, instead of self-evident responses to predictable events (p. 323).
Brulle, Carmichael, and Jenkins (2012) use data from 74 surveys over a 9-year period to construct quarterly measures of public concern over climate change. They study the effect of five factors on the level of public concern: extreme weather events, public access to accurate scientific information, media coverage, elite cues, and movement/countermovement advocacy. They find that information-based science advocacy has had a relatively small impact on public concern over climate change. Instead, they find that political mobilization by elites and advocacy groups are much stronger influential factors. They interpret this as an indication that individuals use media coverage largely to gauge the position of political elites, and interpret the information based on their own party and ideological identity. These findings at the aggregate level are in line with other research stressing the growing importance of ideology and elite cues as mediating factors in media communication with the public (Darmofal, 2009; Lenz, 2009; Guber, 2013).
These studies into the polarization of climate change all seems to point towards a conclusion in line with the central thesis presented by Achen and Bartels in their book (2016a) and their summarizing essay (2016b) titled Democracy for Realists. The authors assert that conventional accounts of the functioning of democracy are fundamentally flawed. One such common accounts is the traditional “populist ideal” theory of democracy (2016b, p. 270) that places the rationally calculating individual citizen as the central agent of democracy. The authors argue that this view is flawed and instead contend that: “party loyalties typically acquired in childhood […] are the primary drivers of political behavior (p. 270). The authors continue their argument aiming to build
28
a new “group theory” on the functioning of democracy (2016a, p. 297). Their model aims to account for the crucial role social identities play in shaping voting behavior. The primary implication of our analyses of retrospective voting is that election outcomes are mostly just erratic reflections of the current balance of partisan loyalties in a given political system” (Achen & Bartels, 2016b, p. 272).
Economic dimension
Rational Choice Institutionalism and the study of Collective Action
The theoretical question posed by the case study at hand is: how do the different political science literatures explain the observed behavior of government institutions when faced with the problem of anthropogenic climate change. The well-established political economy literature of rational choice institutionalism presents another starting point for an analysis of a government’s decision to either take action or refrain from doing so. The central question of this literature is: (how) do groups of individuals deal with collective problems? With potentially high individual costs and only widely diffused public benefits, anthropogenic climate change as an environmental byproduct of economic activity is a perfect example of a large scale collective action problem. In this section I will present an overview of the public choice and collective action literature, showing how this field of political economic theory sprang from neoclassical economics, and overviewing the theoretical explanations it can provide for the observed behavior in the case study. This will serve as the theoretical basis for the hypotheses on the economic dimension presented in the subsequent chapter.
The seminal work in the public choice literature on collective action is Olson’s Logic of Collective Action (1965). The central problem of this book is an explanation of when and how groups of individuals act collaboratively to promote their shared interests. Olson argues that individuals do not naturally act collectively to promote such a shared interest as a public good. Instead, individual actors have an incentive to not participate, shirking the individual costs, while still reaping the collective benefits when the group takes action without the individual. Olson concludes that this freerider problem will always arise in a group of individual actors, unless some form of punishment is put in place in the form of selective incentives (i.e. excluding freeriders from the public good). A more general conclusion Olson presents is that larger heterogeneous groups have higher costs of organizing than smaller more homogeneous groups due to their size and complexity. Hence,
29
smaller homogeneous groups will be more likely to effectively promote their shared interests than larger heterogeneous groups.
Rational choice theory has long been critiqued for their simplistic, unrealistic assumptions on the rationality of the atomistic individual from neoclassical economics (Bennett & Segerberg, 2012). However, already in her 1997 Presidential Address to the American Political Science Association, Ostrom (1998) posed that despite such critiques rational choice theory has remained one of the most powerful theories in the contemporary social sciences (p. 2). She stresses the strengths of the theory in the specific situations where it is directly applicable, like marginal behavior in competitive situations. She also notes how the theory can be adapted and extended to remain relevant and useful beyond those specific settings. Ostrom stresses the need for a behavioral theory of collective action; applying the works of boundedly rational and moral behavior to the logic of collective action. Efforts in this direction have since been undertaken (Ostrom, 2000; Kahan, 2003; Poteete, Janssen, & Ostrom, 2010), and have successfully incorporated concepts like social norms and reciprocal trust into the theory, and empirically tested the functioning of these models. Rational choice institutionalism is identified in the widely cited article by Hall & Taylor (1996) as one of the three ‘new institutionalisms’ recently developed in political science (p. 942), together with the other two ‘new institutionalisms’ of historical institutionalism (p. 937) and sociological institutionalism (p. 946) The concept developed as a response by political scientists in the United States to an apparent discrepancy between the postulates of conventional rational choice and empirically observed political behavior. The seminal work regarding this is the study of the workings and implications of majority rule in the United States Congress (Riker, 1980). As discussed earlier, the theory takes large parts of its analytic tools, such as utility maximization and the rational atomistic individual, from the neo-classical school of economics and the economics of organization (Buchanan, 1965; Coase, 1937) and uses them to study the political behavior. To conclude on the application of this field of theoretical literature to this thesis, it can be said that the theory of rational choice institutionalism uses the tools taken from economics to argue for a utility maximizing view on decision making. Specifically when concerning marginal behavior in competitive situations, it argues that institutional behavior can be best explained by a rational cost-benefit analysis. In the context of the central research question, the traditional rational institutional school of thought provides a theoretical explanation for local governments not taking action on
30
climate change, and seems to provide the rationale behind the Federal U.S. government withdrawing from the Paris Agreement. This will serve as the basis for the hypotheses aiming to test the influence of the economic dimension, as presented in the next chapter. The second half of this section will briefly review more empirical literature on the economic dimension of climate change.
The costs and opportunities of climate change
The seminal empirical study into the economic consequences of climate change is the Stern Review (Stern, 2006), a 700 page report commissioned by the Government of the United Kingdom. The report encompasses the many economic dimensions to the problem of climate change. Generally, it poses that the problem global problem of climate change, if not addressed quickly and adequately, is likely to be very damaging in the long run. It states that all evidence from both local, regional, sectoral, and global studies, show a serious impact on economic output and human life (p. VII). The report emphasizes that emission levels are currently largely driven by economic growth (p. VIII). More recent research confirms this, also pointing to a drop in global emissions during the 2008 financial crisis, and a strong growth of emissions in the recovery period (Peters, et al., 2012).
The report adds that despite the current correlation between economic growth and greenhouse gas emissions, the switch to a low carbon economy also offers new opportunities for growth in terms of the development of new low carbon industries.
The report also addresses the Environmental Kuznets Curve hypothesis (p. 181), the idea that total emissions follows a parabolic arch, first growing rapidly, and then dropping again as the economy adapts and reduces emissions. The report notes these observations are consistent with observations of specifically European economies, but that this is not a general rule. They offer an alternative explanation by pointing to the fact that emission-heavy industries did not disappear, but simply relocated to other regions. Furthermore, they even warn for the opposite effect; observing that lots of high-emission goods (such as air travels) have strong income-elasticity; where demand will grow as income levels rise.
31
Physical dimension
Some empirical studies have been done into the role of physical variables on individual’s support for climate change policies. In this section I provide a short overview of the most important studies in this field.
Zahran, Brody, Vedlitz, Grover and Miller (2008) study the motivation behind municipal governments voluntarily joining the Cities for Climate Protection campaign. They map a locality’s vulnerability to climate change at the county level, using expected temperature change, number of extreme weather events, and coastal proximity as risk indicators. They also include economic, demographic, and civic-participation variables to control for these factors. Using logistic regression, they find a significant effect on a locality’s odds of participation in the CCP campaign for the variables on the number of deaths and injuries from extreme weather events, projected temperature change, and coastal proximity. They also find a significant positive effect for the variable measuring the number of Democrats, and a negative effect for the percentage of the county’s labor force employed in carbon-intensive industries (p. 544).
Brody, Zahran, Vedlitz and Grover also studied the relation between physical vulnerability to the consequences of climate change and public perceptions of global climate change in the U.S. (2008). Using spatial data and multiple measures of climate change vulnerability, they find that there are specific types of physical conditions that correlate with an increased risk perception regarding climate change. These significant variables include a respondent’s distance to the coast, whether or not a respondent lived in the floodplain area, and the number of local fatalities due to natural hazards.
Spence, Poortinga, Butler and Pidgeon (2011) examine the relation between individual’s direct experience with flooding, perceptions of climate change, and preparedness to reduce energy use to address climate change. Using survey data from 1822 individuals in the UK, they show that respondents who report having experienced a flooding express significantly more concern over the problem of global climate change. Interestingly, these concerns also correlate with a higher stated willingness to reduce energy consumption to address the problem of climate change.
Egan and Mullin (2012) study how personal experiences affect political attitudes regarding climate change. In order to bypass the self-reporting bias they merge geocoded survey data on perceptions
32
of climate change with data on the local weather from local scientific weather authorities. They find that for each 3.1°F local temperatures have risen above normal in the week before the survey was taken is associated with a significant increase in respondents agreeing that there is “solid evidence” that the earth is getting warmer. The effect is strongest among those with a lower education, and those who do not report a strong affiliation with either the Democratic nor the Republican party.
Literature conclusion
The literature on the politics of climate change in the United States is closely linked to the subject of political polarization and party identification. Specifically in the context of the United States where ideological alignment between conservatives and the Republican party and liberals and the Democratic party has occurred (Levendusky, 2009), the subject of climate change has become strongly politized as one of the key issues over which the two parties and their supporters can signal their ideological position and identity. This also provides an explanatory argument for the proliferation of climate skepticism or outright climate change denial. This is the basis for the hypotheses on the political dimension presented in the next chapter.
The literature on multilevel governance has been used in the study of international climate policy networks. I aim to apply the same theoretical framework to the study of the U.S. Climate Alliance, and the ‘We Are Still Here’ initiative, and an interpretation of the results of the quantitative and qualitative chapters.
The economic literature on collective action and rational choice institutionalism provides a theoretical basis for the economic hypotheses presented in the next chapter. This theoretical literature forms the basis for two main interpretations of the effect of income: a ‘luxury good’ thesis, where demand for climate change policy is seen as a luxury good, for which demand rises as incomes do, and the Environmental Kuznets Curve hypothesis discussed in the Stern Review, that holds that economies tend to develop towards a low-emissions/high-GDP state.
The literature on the physical dimension includes studies with aims similar to this thesis’ central research question. The study of the Cities for Climate Protection campaign by Zahran, Brody, Vedlitz, Grover and Miller (2008) found significant results for certain measures of physical risk at
33
the city level. The quantitative chapter of this thesis will attempt to recreate these findings, and expand them to the level of the U.S. States.
34
4. Quantitative analysis
This chapter approaches the research question using quantitative methods, aiming to identify how much the political, economic, and physical dimensions influence a local government’s decision to implement climate change policies, using the United States Climate Alliance and the ‘We Are Still In’ initiative as case studies. This chapter consists of four subsections. First, I discuss methods and present the hypotheses. Secondly, I discuss the data, the operationalization of the variables, and sources. Thirdly, I present the analysis and results, which I will then discuss and interpret further in the last section.
Methods and Hypotheses
This thesis aims analyze the local governance responses to the decision to withdraw from the Paris Agreement by the federal U.S. government. The case study of the United States Climate Alliance, and the ‘We Are Still In’ initiative aims to identify the drivers of such a decision. The central research question of this thesis is:
How can we explain the different levels of voluntary stated compliance to the Paris Climate Agreement by local governments in the U.S.?
In order to answer this thesis’ central research question we need to closely examine and dissect the stated intentions and actual implementations of policies by the local state and municipal governments. The first step to answering the research question is by a quantitative analysis of the local government’s decision, and looking to identify the relevant correlations between the decision to pledge support for the Paris Agreement and hypothesized factors of interest. This thesis takes the decision to stick to the Paris Agreement as a binary categorical dependent variable in a logistic regression model. Logistic regression models is used on the level of the U.S. State and city level separately. These logistic regression models include independent variables from the three relevant dimensions: politics, economics, and the physical. On top of the data analysis, this section will also aim to identify outlier cases as candidates for the next chapter for a more qualitative analysis. As stated before, the relevant literature on the subject of climate change politics can be roughly divided into three dimensions. These three dimensions correspond to three sub-questions:
