專制體制與環境表現 - 政大學術集成

全文

(1)International Master’s Program of International Studies National Chengchi University 國立政治大學國際研究英語碩士學位學程. 政治大. ‧. ‧ 國. 學. 立 Regime Types and Authoritarian Environmental Performance 專制體制與環境表現. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Morgan Stemler Advisor: Wen-yang Chang. May 2020. DOI:10.6814/NCCU202000806.

(2) Abstract Climate change has become an ever-increasingly important topic in a multitude of academic fields. A lot of the research on regime type and environmental degradation has focused on democracies versus non-democracies, and concluded that because democracies outperform in this sector, then democratization could be the solution. However, viewpoints seem to be shifting in terms of whether this research is as accurate in the current environment. This paper attempts to reanalyze this topic with a specific focus on whether certain authoritarian regime types, if any, have an effect on carbon dioxide levels within a regime. I argue that not only will authoritarian regime types have an effect, but that party regimes will have the greatest effect in limiting or decreasing carbon. 政治大. dioxide emissions from 1997-2010. To provide empirical support, I include a vast literature review. 立. on the topic and utilize two different quantitative methods of analysis: time-series cross-sectional. ‧ 國. 學. and spatial analysis using Geographic Information Systems. Statistical results mostly confirm that authoritarian regime type can play a role in determining carbon dioxide emission levels, but on the. ‧. contrary, it is military regimes and not party regimes that have the greatest effect on CO 2 levels. However, after robustness checks, it seems to be that limitations skew the results due to a short. y. Nat. sit. time-frame and small sample size. This results in the secondary analysis concluding that, while. n. GDP per capita has a greater effect.a. er. io. positive, regime type is not significant in determining carbon dioxide emissions, but rather that. iv l C n hengchi U. Keywords: Authoritarian regime, Environmental performance, Regime type, Carbon dioxide, Party regime, Degradation. 2 DOI:10.6814/NCCU202000806.

(3) Contents 1. 2. Introduction 1.1. Research Motivation . . . . . . . . . . . . . . . . . . . . . . . . 6. 1.2. Propose of Research . . . . . . . . . . . . . . . . . . . . . . . . 9. 1.3. Thesis Structure . . . . . . . . . . . . . . . . . . . . . . . . . . 9. The Selectorate and Democratic Institutions: Possible Explanations 2.1 2.2. 3. The Selectorate . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Democratic Institutions . . . . . . . . . . . . . . . . . . . . . . . 12. Authoritarianism and the Environment: Understanding the Possible Explanations 3.1. The Role of Authoritarian Regimes . . . . . . . . . . . . . . . . . . 14. 3.2. Interdisciplinary Approach to Authoritarian Rule . . . . . . . . . . . . 18 Public Goods Provision. 3.2.2. Miscellaneous Fields. 學. . . . . . . . . . . . . . . . . . . . . .. 21. Authoritarian Environmentalism . . . . . . . . . . . . . . . . . . . 24. ‧. Research Design. Dependent Variable: Carbon Dioxide (CO2) Emissions . . . . . . . . . .. 4.2. Independent Variable: Authoritarian Regime Types. 4.3. Control Variables . . . . . . . . . . . . . . . . . . . . . . . . . 27. 4.4. Model Selection. er. io. sit. . . . . . . . . . . . 26. . .a. . . . . . . . . . . . . . . . . . . . . . .. n. iv l n C Results: Time-Series Cross-Sectional & Geographic Information h e n gAnalysis chi U Systems. 28. Time-Series Cross-Sectional Analysis . . . . . . . . . . . . . . . . . 31 5.1.1 Statistical Summary . . . . . . . . . . . . . . . . . . . . . . 31 5.1.2. 5.2. 26. y. 4.1. 5.1. Results . . . . . . .. . . . . . . . . . . . . . . . . . . . .. 37. Geographic Information Systems Analysis . . . . . . . . . . . . . . . 41 5.2.1 Statistical Summary . . . . . . . . . . . . . . . . . . . . . . 41 5.2.2. 6. . . . . . . . . . . . . . . . . . . . . 19. Nat. 5. ‧ 國. 3.2.1 3.3. 4. 立. 政治大. Results . . . . . . .. . . . . . . . . . . . . . . . . . . . .. 47. Conclusion 6.1 6.2 6.3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Reasoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Suggestions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60. Appendix. 3 DOI:10.6814/NCCU202000806.

(4) List of Figures 5.1: Top Five Carbon Dioxide (per Capita) Emitters (1997-2010) . . . . . . . . . . .. 32. 5.2: Average CO2 (Metric Tons per Capita) Emissions in the Four Different Authoritarian Regime Types (1997-2010) . . . . . . . . . . . . . . . . . . . . . . . . . . 33 5.3: Box Map (Hinge = 1.5) to Determine Outliers in CO2 Emissions (Metric Tons per Capita) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 41. 5.4: Parallel Coordinate Map Representing the Change of CO2 Emissions (Metric Tons per Capita) From 1997 to 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . 42. 5.5: Map for Carbon Dioxide Emissions in Authoritarian Regimes (1997-2010) . . . . . . 44 . 治政 5.7: Map of Trade Levels in Authoritarian Regimes (1997-2010) 大 . .... 立 in Authoritarian Regimes (1997-2010) 5.8: Map of Population Density Levels 5.6: Map of GDP per Capita Levels in Authoritarian Regimes (1997-2010). . . . . . . 44 . . .. . . . 45. . . . . . . 45. ‧ 國. 學. 5.9: Map of Urban Population Levels in Authoritarian Regimes (1997-2010) . . . . . . . 46 5.10: Map of Industry Levels in Authoritarian Regimes (1997-2010) . . . . . . .. . . . 46. ‧. 5.11: Local G* Cluster Map with the Distance Weight Set to the 5 Nearest Neighbors . . . . 48. y. Nat. 5.12: Local G* Cluster Map with the Distance Weight Set to the 10 Nearest Neighbors . . . 49. io. sit. 5.13: Bubble Chart Showing the Relationship Between GDP per Capita and Regime Type on. n. al. er. CO2 Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53. Ch. engchi. i n U. v. 4 DOI:10.6814/NCCU202000806.

(5) List of Tables Table 1: Summary of Countries Per Regime Type . . . . . . . . . . . . . . . . . . 34 Table 2: Summary Statistics .. . . . . . . . . . . . . . . . . . . . . . . . . . 36. Table 3: Authoritarian Regime Type and Carbon Dioxide Emissions: Models with Random Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Table 4: Authoritarian Regime Type and Carbon Dioxide Emissions: GIS Results. . . . . Table 5: Diagnostics for Spatial Dependence . . . . . . . . . . . . . . . . . .. 立. 51. . 53. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 5 DOI:10.6814/NCCU202000806.

(6) 1. Introduction. 1.1 Research Motivation The efforts around the globe to curb environmental degradation have become increasingly large; people from all walks of life are now making an effort to live a "greener" lifestyle either through recycling, buying secondhand clothing, reducing their use of plastics, or by using many other ecofriendly methods. However, these efforts still have a minimal impact on the amount of degradation we are still causing to our environment. With an increasing global population and levels of consumerism, there is a rapid increase in the change of the global climate. According to the National Oceanic and Atmospheric Administration (NOAA), the amount of atmospheric carbon. 政治大. dioxide has increased since the 1960s from 0.6  0.1 parts per million (ppm) growth rate per year. 立. to 2.3 ppm per year from 2016-2018 (Lindsey 2020).. The United Nations Environment. ‧ 國. 學. Programme (UNEP) has also verified this and deems that governments should make efforts to decrease their CO2 emissions so that there is only an increase of 1.5 degrees Celsius by the end of. ‧. the century (UNEP 2019). Therefore, while civil society must be active in fighting climate change, deforestation, and many other negative impacts on the environment, the fight for a cleaner. y. Nat. sit. environment needs to be fought on a larger scale. Therefore, in order to have more of an effect on. er. io. curbing environmental degradation, governments around the world need to implement more robust. n. a environmental policies in order to combat environmental degradation i v effectively.. l C n U h i e h n When thinking of the environment, most peoplegdocnot make an association with authoritarian regimes; however, with a rise in authoritarianism across the world, these types of regimes are becoming more important and should be further researched, notably, in the realm of environmental degradation. According to Freedom House, several factors help indicate the rise in authoritarianism: elections, term limits for executives, freedom of expression, rights of migrants, the safety of expatriates, and ethnic cleansing (Freedom House 2019). For all of these factors, there was a decline (besides an increase in cases of ethnic cleansing), and the availability of political rights and civil liberties are questionable (Freedom House 2019). In regards to term limits for executives, for the past 13 years, term limits in 34 different countries have attempted to revise these with 31 of them being successful (Freedom House 2019). All of these factors are a sure sign of the increasing interest of authoritarian regimes. With the rise of China, one of the most. 6 DOI:10.6814/NCCU202000806.

(7) successful authoritarian regimes, promoting and conveying its model of censorship and surveillance, and also by having one of the world's largest and most influential economies, a growing military, and a growing presence in the global world order, it is highly likely that authoritarian regimes are here to stay and will continue to spread (Freedom House 2019). Therefore, it is imperative to the world and our environment that we better understand how this affects our ecosystem and how we can solve it. With this in mind, it is essential to look at what types of governments tend to have better environmental policies in order to have the most significant impact. After reviewing the literature, one can see that most of the literature does take heed of authoritarian regimes, and it mainly focuses on how democracies are more prone to implement environmental policies. However, some of the. 政治大 authoritarian regimes can implement effective long-term policies. While most of these policies 立 have focused on the economy or the financial sector, this literature is essential in order to see literature – Gat (2007), Rosato (2003), Geddes (1999b) – explains that different types of. ‧ 國. 學. whether these types of authoritarian regimes can have any effect in decreasing environmental degradation. Therefore, by understanding the possible explanations that causes environmental. ‧. degradation, future researchers or policymakers can have a better understanding of how to combat. sit. y. Nat. this issue.. er. io. By utilizing Geddes's, Wright's, and Frantz's 2014 dataset on authoritarian regimes, we can identify that there are four types of regimes:aparty, military, personal, and monarchy (Geddes, Wright, and. n. iv l C n Frantz 2014). The literature on authoritarian is limited in terms of the environment and h eregimes ngchi U focuses mainly on other vital issues such as the economy, security, regime durability, and public goods provisions. However, the literature also outlines a newer topic known as authoritarian environmentalism, which is a model that allows for a stable, uncorrupt authoritarian regime to implement policies in order to improve environmental outcomes (Gilley 2012). Two important possible explanations that allow authoritarian regimes to improve environmental outcomes depend on the regime's selectorate and democratic institutions. The "selectorate" consists of two groups, one of which being the "selectorate" which is a group of people who play a role in choosing leaders and receiving benefits and exclusive privileges, and a second group known as the "winning coalition" who help the leader maintain their rule and receive benefits from that (Bueno de Mesquita et al. 2003, xvi). Democratic institutions are things like elections or legislatures that are. 7 DOI:10.6814/NCCU202000806.

(8) generally found in democracies but have been implemented into authoritarian regimes to better connect the regime with its population. With party regimes being the example, China, one of the most well-known single-party authoritarian regimes, has been doling out environmental projects and policies to help improve environmental outcomes in recent years. According to the World Economic Forum, China has been implementing policies that will allow it to fund a greener future (Song 2018). Certain things like dismantling coal-power plants over a certain length of time, innovating their environmental agency, and transitioning three major industrial cities to be turned into sustainable development zones that will implement the UN's Sustainable Development Goals for 2030. Also, China has been collecting an environmental tax in order to help fund a greener future and a less. 政治大 to improving environmental outcomes, China's CO levels are still rapidly increasing with an 立 output of 0.26GtCO , and have increased global CO emissions (Hausfather 2019). Therefore, environmentally degrading economy (Song 2018). While all of these steps are important in regards 2. 2. 2. ‧ 國. 學. since greenhouse gas emissions (GHG) are the most detrimental form of pollution (EPA 2019), yet the most difficult to detect by regular citizens, the question remains whether regimes are using. ‧. their efforts to gain international recognition and appease its citizens or are trying to actually. sit. y. Nat. improve environmental conditions.. er. io. Overall, the issue of environmental degradation has been a huge topic around the world, especially with the younger generation. College a students or other youth have spurred most environmental. n. iv l C n movements with one of the most pronounced activists of today being a teenager h e nenvironmental hi U c g from Sweden, Greta Thunberg. However, with much of civil society disengaged in public life (Carpini 2000, 341), the rest are limited with their abilities to make a change or make other parts of society care about the environment. Therefore, the role of government is crucial in regards to environmental policy implementation. With this in mind, by looking at different institutional settings, one can have a better idea as to which and what authoritarian regimes can have an impact in the fight for environmental protection. The environment is a critical issue, because not only could further environmental degradation lead to detrimental effects on the environment and humanity now, but it would pose grave risks for our future generations. By adding to the current literature, policymakers and civil society could have a better idea of which or what type of authoritarian regime(s) would best help combat environmental degradation.. 8 DOI:10.6814/NCCU202000806.

(9) 1.2 Propose of Research This thesis aims to fill a gap in the literature by contributing to the unwritten literature on regime type and the environment. This thesis will answer two questions. The first is, "What are the effects of authoritarianism on environmental governance?" This question is essential to this research because it is crucial to look at all types of authoritarian regimes and discover whether the structures of these regimes have a direct influence on the environment. It is crucial to determine whether some or any explanation allow for authoritarian regimes to pursue better environmental policies. Some of the explanations in this research include the selectorate (Bueno de Mesquita et al. 2003) and democratic institutions. Once the explanations are determined, then it will be much easier to determine the significance and results of the second question.. 政治大 Not only are the possible explanations important, but reviewing which types of authoritarian 立 regimes have a more significant impact on environmental policies can help add to the current. ‧ 國. 學. research on the issue at hand. Therefore, the second question asked in this research is, "What types of authoritarian regimes tend to have a better environmental performance?" There is quite a bit of. ‧. research completed on how the conventional wisdom that democracies tend to outperform non-. sit. y. Nat. democracies in many different sectors (economics, security, peace, public goods distribution, etc.). io. er. is not holding to be consistently accurate. Therefore, if certain types of authoritarian regimes tend to outperform others and compete at the same level as democracies in these sectors, then one can. n. a. look at environmental protection in al new C light.. hengchi. i n U. v. 1.3 Thesis Structure In this chapter, I have given a brief background on the topic and discussed my motivation for this research and what questions I would like to answer. The latter part of the paper will give a more thorough depiction of the research questions, why the questions were asked, and how to find and what the outcome is. In chapter 2 and 3, there will be an extensive literature review covering many different topics related to this issue, such as regime type and the environment, the impact of the economy, security, public goods, and regime durability on the environment, authoritarian environmentalism, and the possible explanations surrounding authoritarian regime types and. 9 DOI:10.6814/NCCU202000806.

(10) environmental degradation. Chapter 4 will contain the research design and the plans on proving the hypotheses. Chapter 5, Authoritarianism and Environmental Performance: Time-Series CrossSectional Analysis and Geographic Information Systems (GIS), will include a detailed analysis on the authoritarian regimes’ environmental performances using time-series cross-sectional analysis. It will also use spatial analysis to map out the levels of GHG emissions in different authoritarian regimes to determine whether there is a pattern in emission levels. Lastly, chapter 6 will be the conclusion, which will include a discussion on the findings of this research and also suggestions for future research.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 10 DOI:10.6814/NCCU202000806.

(11) 2. The Selectorate and Democratic Institutions: Possible Explanations. This section will provide and explain the possible explanations that could be responsible for how authoritarian regimes respond and interact with the environment. Two possible explanations will help theorize how authoritarian regime types and environmental performance are correlated; these two possible explanations are the selectorate and democratic institutions. The size of the "selectorate" is an important explanation because it determines how susceptible a government is to its selectorate. Once one knows a government's selectorate, one can better understand how decisions are made and who are making them. Democratic institutions are institutions that are. 政治大 appease the population and validate立 the ruling party. Collective regimes, especially party regimes, fundamental to democracy, but when implemented into authoritarian regimes, they utilize it to. ‧ 國. 學. are more likely to implement democratic institutions, and once implemented, tend to be more productive. This is because personalistic regimes tend to focus on private rather than public goods. ‧. (Bonvecchi, Simison 2017). Democratic institutions are a possible explanation for environmental performance because when regimes implement democratic institutions, they are more likely to. y. Nat. sit. provide more public goods, such as environmental concessions, which can, in turn, improve. n. al. er. io. environmental outcomes for these types of regimes.. 2.1 The Selectorate. Ch. engchi. i n U. v. A common mechanism in the realm of international political economy is "the Selectorate Theory," found in Bruce Bueno de Mesquita's The Logic of Political Survival (Bueno de Mesquita et al. 2003). In this work, Bueno de Mesquita et al. examine how political selection institutions, or the "selectorate," influence the likelihood that a leader will survive or not. They also look at how the "selectorate" influences a leader's behavior and decision-making practices. While this author's main point is to argue for economic demands, it can easily relate to environmental demands. Since political leaders need to remain in office for both democracies and autocracies, if the people are demanding environmental action from the government, then the government should make concessions. However, this is not always the case, as pointed out by Bueno de Mesquita, because. 11 DOI:10.6814/NCCU202000806.

(12) there is always some group, known as the winning coalition, which helps the leader maintain their power. The winning coalition is essential because, in order to remain in power, one must have the winning coalition on its side. As stated in the reading, “This group controls the essential features that constitute political power in the system. In democracies, the winning coalition is the group of voters who elect the leader; in other systems, it is the set of people who control sufficient other instruments of power to keep the leader in office. If the leader loses the loyalty of a sufficient number of members of the winning coalition, a challenger can remove and replace the leader from office” (Bueno de Mesquita 2003: 10). These members come from the "selectorate," and in democracies, the winning coalition tends to be larger if the "selectorate" is larger, but in autocracies,. 政治大 the winning coalition will be more loyal to the leader because there is no opposition, and there is 立 a higher risk of exclusion if one defects (Bueno de Mesquita et al. 2003, 11). the winning coalition is small with a larger "selectorate." Therefore, in autocracies, members of. ‧ 國. 學. The "selectorate" can be seen as an essential explanation because it outlines how regimes act in. ‧. order to keep power and appease their selectorate, or more importantly, their winning coalition. This author believes that regardless of regime type, political actors have the common goal to. y. Nat. sit. remain in power, and therefore will do its best to maintain its winning coalition. However, the. er. io. author also points out that if there is a small winning coalition but a large selectorate, then it is. n. a be challenged (Bueno devMesquita et al. 2003). However, more likely that the leader will eventually i l C n in authoritarian regimes, this is not alwayshthe authoritarian regimes not only have a e ncase. hi U g cSome small winning coalition and a larger "selectorate," but also have managed to maintain power for. decades. Therefore, it is vital to determine which types of autocracies stay in power longer and why. By determining this, one would possibly be able to understand which type can implement better environmental policies.. 2.2 Democratic Institutions The rise of authoritarianism has been seen as an increasing threat to the liberal world order for several decades. Many western liberal-minded scholars are intrigued by how authoritarian regimes maintain their power, but also why young democracies were and are failing. Because of this, there has been much research and literature on the topic of authoritarianism. In regards to this thesis, in 12 DOI:10.6814/NCCU202000806.

(13) order to have a better understanding as to why specific authoritarian regimes make concessions and provide more public goods to their population can depend on an essential factor, democratic institutions. As David Art cites in his 2012 article when reviewing Jennifer Gandhi's article, Political Institutions under Dictatorship, "…these democratic-looking institutions are not simply 'window dressings,' but that they play a central role in the construction, policymaking, economic performance, and durability of authoritarian regimes (Art 2012, 359).” There are several different examples of democratic institutions, but the main two reviewed in this thesis will be elections and legislatures. Democratic institutions are institutions that make up the basis of democracy, its inner-workings, and fundamental structure. However, these institutions have been implemented in authoritarian regimes over the years in order to placate the population's. 政治大 concede to specific demands of their population, but it also allows them to allow for more public 立. demands and extend the regime's rule. By implementing these institutions, regimes cannot only. goods provisions on the local level. These institutions are essential because, once implemented,. ‧ 國. 學. can help authoritarian rulers maintain stability at the local level, control political opposition, and also maintain their regime's durability (Art 2012).. ‧. y. Nat. The importance of democratic institutions in regards to authoritarian regime type and the. io. sit. environment is not only that it can provide for more public goods, such as environmental protection.. er. In regards to regime type, it is emphasized by Bonvecchi & Simison (2017) that these institutions,. n. a l usually are more frequenti vand more effective in collective specifically legislatures for their case, n. C. U h epersonalistic regimes (party and military) rather than and monarchy). The first n g c h i (personal democratic institution I will explain, elections, especially in a smaller electorate, can provide. competition with local officials that can make them appeal to the majority and provide more public goods provisions to help them win re-election (Martinez-Bravo et al. 2012). The other democratic institution I utilize in this thesis, legislatures, is an institution that provides more public rather than private gain by constraining a leader's ability to utilize their power for their gain. Overall, democratic institutions are costly to implement in regards to a leader's legitimacy coming into question, but if not implemented or if ignored when implemented, they can have detrimental effects on the survival of the regime.. 13 DOI:10.6814/NCCU202000806.

(14) 3. Authoritarianism & the Environment: Understanding the Possible Explanations. 3.1 The Role of Authoritarian Regimes The role of government is essential for many things in a society; the government does not only set up the authority and the way a state is run or governed, but it is also responsible for anything that encompasses the state, such as establishing policies and institutions in order for the state to define its territory, population, sovereignty, and international recognition. According to Kirlin (1996), governments must develop effective policies for their public, make decisions that protect civil. 政治大 can be different based on the form of government established in each state. In authoritarian regimes, 立 the government holds power in establishing laws and policies; however, in different authoritarian society, and deliver social services. However, the way that a government is organized or structured. ‧ 國. 學. regimes, there can be an influence on decision-making due to the size of the "selectorate" and the level of democratic institutions in the regime. Regardless of this fact, authoritarian regimes'. ‧. primary decision maker in regards to laws and policies come from the government, which makes. sit. y. Nat. it very important in regards to the environment.. er. io. However, it is essential not to overlook the effect on how government implemented policies have. n. a l These policies can be especially a significant effect on civil society. crucial in authoritarian iv C. n. regimes where only a select few have thehopportunity i Uability to influence policy. Cavatorta e n hand. gc. (2012) explains that the idea of civil society and civic activism typically relates to democracy/democratization; however, in authoritarian regimes, civil activism does not look the same. Civil activism is different in authoritarian regimes because activism occurs in small groups or individuals rather than large, organized groups. Also, non-political issues are sometimes remaining non-political and not always forming into political issues on democratization and human rights. Moreover, activism in authoritarian regimes does not take place in the public sphere but rather must co-exist with state-sponsored civil activism. An example here was a Vietnamese woman, famous for her activism, imprisoned for distributing propaganda against the state. She was not only using the internet as her platform, but she was also pursuing activism at the individual level (Nguyen, Datzberger 2018). This example is essential to this research because it shows that. 14 DOI:10.6814/NCCU202000806.

(15) it is not just the government that decides what actions to take, but instead that the demands from civil society, while typically constrained by the government, also affect policy implementation. There are several different types of autocratic regimes; these include party regimes, personalist dictatorships, monarchies, military regimes (Geddes, Wright, & Frantz 2014). All of these regimes classify under the same category of an authoritarian regime, but these regimes can be quite different in terms of how they rule, the policies they will enact, the state relations they create, and many other factors. Recently, there has been an increase in research on which types of autocratic regimes are more prone to openness in terms of economic, security, education, or other types of policies, such as Weeks (2008), Chandra & Rudra (2013), Geddes (1999b). In the field of international political economy, there has been an increasing consensus that autocratic regimes. 政治大 context, the selectorate are the ones who are choosing the policies and having an influence on 立. with a larger selectorate tend to incorporate more open economic policies into their agenda. In this. anything else that needs addressing (Bueno de Mesquita et al. 2003). Therefore, in political-. ‧ 國. 學. economic thought, the idea is that certain autocratic regime types are more likely to have a stable, long-term rule and can, therefore, implement more trade openness and grow their economies. ‧. (Kailitz, Stockemer 2017). One of the most cited and successful examples in this research is China.. y. Nat. io. sit. China's success is critical to consider because, in recent years, there has been an increase in the. er. return to authoritarianism (Freedom House 2019), and with this, we need to understand how civil. n. al society can have influences in authoritarian regimes' decisions. That i v is to say that certain types of n. C. authoritarian regimes can incorporate morehdemocratic i U and institutions into their system e n g c hprinciples of government, which in turn, allows them to incorporate their civilians' wants and needs better, suppress dissent, and control political opposition. Competitive authoritarianism, a regime that holds democratic institutions but violates the rules often enough that it cannot be considered a democracy (Levitsky, Way 2002, 52), is essential here. According to Wetenschappelijke and the Scientific Council for Government Policy (2007), “Citizens are not just voters (or legal subjects), but participate in all manner of social relationships and structures (Wetenschappelijke, Scientific Council for Government Policy 2007, 119)." The same is true for specific authoritarian regimes. By incorporating democratic institutions into their governmental frameworks, such as elections and legislatures, regimes are more likely to have support from their population because they are more likely to increase public goods provisions (Martinez-Bravo et al. 2012). Also, Geddes (2005). 15 DOI:10.6814/NCCU202000806.

(16) explains that elections are a safer and more predictable for authoritarian regimes to gage their popular support and diminish the threat of a military or civilian coup. This example, therefore, emphasizes that the role of government in authoritarian regimes, regardless if it is personalistic, a military regime, a monarchy, or a party regime, plays a significant role in policymaking. Moreover, contrary to conventional thought, authoritarian regimes must eventually take into consideration what civil society demands. This understanding can open up the possibility for authoritarian regimes to consider the importance of the environment. Elections are an essential type of democratic institution. Not only do they allow leaders to gauge their public support, but they also allow the regime to pacify their population by providing them options to express their demands. In turn, this can help the regime with proper decision making. 政治大 can better control the actions of their citizens and officials, especially at the local level. By doing 立 while increasing the durability of their rule. By implementing elections, authoritarian governments. this, local officials are more likely to change their incentives to favor the wants and needs of their. ‧ 國. 學. local population, which eventually leads to more funding for public goods provision. In China in the 1980s, they implemented village-level elections in order to maintain control in rural areas of. ‧. the country, which allowed the government to appease voters based on the fact that local officials. sit. y. Nat. wanted to appeal to the majority in order to obtain re-election (Martinez-Bravo et al. 2012). Also,. io. er. by implementing this in a smaller electorate rather than a larger one, the regime itself is not threatened by opposition or the chance of being voted out (Takeuchi 2013). While this case only. n. al. iv. highlights single-party regimes, the caseCof Tanzania also shows n that dominant-party regimes who. i U. he. n gpublic c h goods, particularly in more competitive also implement elections will also provide more regions, in order to gain support and maintain their rule (Rosenzweig 2015, 83).. Another essential form of democratic institutions is legislatures. A common misconception about authoritarian regimes is that legislatures in authoritarian regimes are unimportant because personalistic regimes still hold onto their power and promote their policies. However, according to Bonvecchi & Simison (2017), legislatures, once established in authoritarian regimes, can constrain the power of dictators because they are costly to ignore, and by looking at the foundations of the legislature, one can determine how powerful it will be. In addition to this, Gandhi & Przeworski (2007) explain that these legislatures foster cooperation and undermine dictators in order to lengthen the regime’s survival. For example, when threats arise in an authoritarian regime,. 16 DOI:10.6814/NCCU202000806.

(17) and a leader must make concessions, this takes the form of policy compromises where legislature or institutions are needed in order for agreements to be agreed upon (Art 2012, 359). By having legislatures, they can allow leaders to answer and concede to these threats without appearing weak (Art 2012, 360). However, while dictatorships and personalistic regimes are essential, Bonvecchi & Simison (2017) also emphasize that legislatures are more effective and influential when the regime’s power is collective rather than personalistic. Reuter and Robertson (2015) also explain that authoritarian legislatures help leaders reduce social protest; however, for these authors, the reduction mainly had to do with opposition elites demobilizing their supporters for monetary gain. In addition to democratic institutions, the other important mechanism that can define which types of authoritarian regimes can improve environmental outcomes depends on what Bueno de. 政治大 selecting a leader and keeping the leader in office (Bueno de Mesquita et al. 2003). Hankla and 立 Mesquita et al. calls the selectorate. As mentioned in chapter one, the selectorate is responsible for. Kuthy (2013) also note that states with larger selectorates can adopt better policies that will help. ‧ 國. 學. it extend and strengthen their regime. The authors state, "First, we argue that autocratic regimes with a larger ''selectorate'' should have more significant incentives to provide public rather than. ‧. private goods. As a result, we expect that multiparty, and to a lesser extent, single-party autocracies. sit. y. Nat. will tend to prefer more open trade policies than non-party (often personalistic) dictatorships,. io. er. monarchies, and military juntas. Second, we contend that more stable autocratic regimes will have longer time horizons and, therefore, more significant incentives to adopt policies, such as trade. n. al. iv. openness, that may strengthen long-runCeconomic performance n (Hankla, Kuthy 2013, 492)." As. i U. he. g c hcoalition and selectorate should be more one can see, autocratic regimes with a larger n winning open to different economic and environmental policies.. For further emphasis, Cao & Ward (2015) concluded that while some authoritarian regimes can concede to some issues, these mostly occur in regimes that are looking long-term and are willing to make investments into different public goods or sectors. They argue this because they believe that regimes with shorter time horizons are not as likely to invest in public goods, like health and education (Cao, Ward 2015). However, for regimes that have a larger winning coalition, a larger state capacity, and more durability, then these types of goods will be provided in the long-term. Therefore, this shows that certain types of authoritarian regimes, such as non-personalist regimes are capable of investing in society. Other authors who negate the mainstream idea of democracies. 17 DOI:10.6814/NCCU202000806.

(18) and their role of environmental protection are claiming that because autocratic leader(s) focus on maintaining their rule for as long as possible, they may be more willing to concede to some environmental protection in order to extend their length of rule (Klick 2002). Klick (2002) explains that by providing some environmental growth, the ruler can extend its rule by pacifying their people. The author states, "High environmental standards represent one tool the autocrat could use to placate his people without providing them with any revolutionary resources (Klick 2002, 1)." This argument is valid in the sense that all authoritarian leaders want to maintain their hold on power and will be willing to give something minute up in order to extend their regime's duration. Therefore, the idea that only democracies can provide better environmental policies and protection has become increasingly contradictory in recent years.. 治. 政 to Authoritarian 3.2 Interdisciplinary Approach Rule 大. 立. ‧ 國. 學. As previously stated, there has been an increase in the importance of authoritarian regimes. Because of this, the literature over the years has expanded to fit many fields. In this section, there. ‧. will be a discussion on how researchers have linked authoritarian regime types to many other fields. These fields include public goods provision, conflicts and peace, economic development, and. y. Nat. sit. regime durability. Because authoritarian regime types have not been researched in regards to the. er. io. environment, by linking the results of these sections, one can better determine better which type. n. of authoritarian regime outperformsathe others. Therefore, for this section, the most important field, v. i l C n and the one linked to the environment, is public h e ngoods h i U As will be stated in section 3.2.1, g c provision. environmental protection is a public good that should be preserved and provided to the public.. Therefore, by better understanding how authoritarian regimes promote and provide public goods, one can better understand which type of authoritarian regime will be more likely to improve environmental outcomes. The next section, 3.2.2, will combine the three other fields (conflict and peace, economic development, and regime durability) that discuss which authoritarian regime type outperforms the other in each respected discipline.. 18 DOI:10.6814/NCCU202000806.

(19) 3.2.1 Public Goods Provision As explained by E. Wesley F. Peterson (2000), “Public goods are characterized by some degree of nonrivalry in consumption as well as difficulty in excluding noncontributors from consuming the good (Peterson 2000, 355).” Nonrivalry is that if one person uses that good, it will still be available for use by others and not prevent others from using the good. For noncontributors or non-exclusion, it means that one cannot exclude individuals from using this good. There are many prime examples of public goods, such as education, national defense, and clean air. All regimes should provide these goods. However, this is not always the case, especially in regards to environmental protection in non-democratic countries.. 政治大 more and more public goods. In Martinez-Bravo et al. (2012), she and her co-authors discovered 立 that by implementing elections into smaller electorates in rural China, they were able to increase On the other hand, recently, there has been an increase in specific authoritarian regimes providing. ‧ 國. 學. public goods provision, reduce income inequality, and reduce land leasing to outsiders for farming. This result occurred because, in order to remain in power and gain re-election, officials must bend. ‧. to public will. In addition to this, Tsai (2007, 355) also notes that even when democratic institutions. sit. y. Nat. are weak, local officials can still increase public goods provision due to "unofficial rules and norms. io. er. that establish and enforce their public obligations." Therefore, the types of authoritarian regimes that can hold elections, and in turn, incorporate democratic institutions into the regime, will be. n. a. v. more likely to increase public goodsl expenditures. As noted by n i Hadenius and Teorell (2006, 6), C. i U. he. n gelections c h for parliament or the executive office, electoral regimes, such as ones that hold popular. can be defined as party regimes that either has no-party regimes (elections are held, but all parties and candidates are prohibited), one-party regimes (all but one party are prohibited from elections), and limited multiparty regimes (candidates can participate in elections outside of the ruling regime). Moreover, in addition to democratic institutions, another mechanism, the selectorate, also emphasizes the state's ability to provide public goods. Hankla and Kuthy (2013) explain that regimes with a larger selectorate, specifically multi and single-party regimes, should provide more public goods, and because of this, they will be more stable, which will allow them to have longer time horizons and therefore adopt more long-term policies. As noted by these authors, in regimes with a higher number of constituents, it is difficult for them to pay off all of the actors that would. 19 DOI:10.6814/NCCU202000806.

(20) help them maintain their power. Therefore, they must increase their interest and funds into public goods provision based on the fact that public goods tend to affect society at large (Hankla, Kuthy 2013, 492). Hankla and Kuthy (2013, 494) note that these regimes with larger selectorates (normally party regimes) will also have to rely on a larger winning coalition, and their support can usually be bought by public and even, sometimes, private goods. Now that we have determined the type of authoritarian regime that will be able to outperform others in terms of public goods provision, it is also essential to better understand the importance of environmental protection as a vital public good that every government provides. According to Matthew J. Kotchen (2014), many environmental qualities classify as a public good, such as air pollution. Kotchen (2014) states, "Under most circumstances, one person’s breathing of fresh air. 政治大 air (Kotchen 2014, 271)." In addition to this, he also outlines that other environmental factors 立 does not reduce air quality for others to enjoy, and one cannot prevent people from breathing the. included as public goods are water quality, open space, biodiversity, and a stable climate (Kotchen. ‧ 國. 學. 2014, 272). Recognizing environmental protection as a public good is important because by understanding that environmental protection is a public good and which type of authoritarian. ‧. regime, if any, can provide more public goods, then we can better understand how to move forward. io. sit. y. Nat. in tackling climate change.. er. As has been mentioned above, environmental protection is a vital public good that should be. n. a l of regime type. However,i vit is also relevant (especially to provided by all governments, regardless n. C. this thesis) to note that party regimes willhbe to provide public goods, and in turn, e nmore hi U g clikely provide environmental protection, which will result in lower levels of environmental degradation. It is also important to note that the two possible explanations, the selectorate and democratic institutions, have an essential role to play in environmental protection. Therefore, for emphasis, regimes such as party regimes that have larger selectorates and democratic institutions will be more likely to provide more public goods, such as environmental protection.. 20 DOI:10.6814/NCCU202000806.

(21) 3.2.2 Miscellaneous Fields As noted in section 3.2, there has been an expansion of knowledge on authoritarian regime types in different fields over the years. The three fields researched most in-depth will consist of conflict and peace, economic development, and regime durability. These fields are necessary to highlight because while they do not focus on environmental performance, they can provide an idea of which type of authoritarian regime will outperform others and be more likely to provide more public goods for their population. Another reason why it is essential to look at these fields is that there is no research on the relationship between authoritarian regime type and environmental performance. Therefore, one must utilize the existing available literature. In regards to conflict and peace, dictatorial or authoritarian peace has been the counterargument to. 政治大 scientists to promote the idea that different 立 types of authoritarian regimes react differently beyond. the well-cited Democratic Peace Theory (DPT). Geddes (1999a) was one of the first political. ‧ 國. 學. the status quo. According to Geddes (1999a), "To summarize that argument, military regimes tend to split when challenged, personalist regimes to circle the wagons, and party regimes to coopt their. ‧. challengers (Geddes 1999a, 1).” Here, Geddes is explaining again how party regimes will tend to maintain and hold power longer than the other two regimes even when facing an economic. Nat. sit. y. downturn. This is because personalist regimes tend to end in uprisings, and military regimes tend. er. io. to end in negotiations because they usually are suffering from internal issues in their own. n. selectorate. Other works like Peceny, a Beer, & Sanchez-Terry. (2002), Ishiyama, Conway, Haggans. l. iv. n have all expanded on Geddes C Peceny (2008), Letkzian and Souva (2009), and & Butler U (2004) h engchi. work and determined that party regimes can produce similar results as democracies in terms of peace and security. In addition to this, whether states enter or do not enter into conflict can be linked to audience costs. According to Weeks (2008), audience costs are the domestic punishment that leaders would incur for backing down from public threats or threats from the audience. Democracies, as noted through the literature, have higher audience costs for backing out of a conflict. However, Weeks (2008) concludes that single-party authoritarian regimes can also have very high audience costs and generate credible threats that are up to par with that of democracies. Weeks also determines that personalist regimes, new democracies, and unstable democracies will have a more difficult time maintaining their threats. This idea is consistent with Peceny, Beer, & Sanchez-Terry (2002). 21 DOI:10.6814/NCCU202000806.

(22) argument. For if audience costs are high, then conflict is more likely to occur for a significant amount of time, and therefore, it would be discouraged to enter into conflict with another singleparty authoritarian regime rather than a mixed dyad conflict. This argument, as well as the ones mentioned in this section, defend the argument that authoritarian regimes, primarily single-party regimes, can uphold dictatorial peace (Weeks 2008). In terms of economic development, the world has seen a rise in different authoritarian regimes developing faster than many democracies. Some have related this to decision making, such as Sah (1991, 70), who explained that a more centralized society, or one with few preceptors such as an authoritarian regime, is more likely to make decisions unilaterally. However, others have focused on coalition size, or in other words, the selectorate. Bueno de Mesquita et al. (2003) argument. 政治大 argue that an authoritarian regime will tend to use the income from economic growth primarily 立 amongst the elites and those in power. However, for this instance, coalition size matters. Those expands on how an authoritarian regime will behave once economic growth is occurring. They. ‧ 國. 學. with a larger coalition size will have to focus primarily on public welfare, while those with a smaller coalition size will have to focus on the party members and the elites. Therefore, more. ‧. sit. Nat. because it will help them to maintain their place in power.. y. party-affiliated authoritarian regimes will be able to promote economic growth for the public. er. io. On the other hand, many other authors believe that increased economic growth has to deal with either the level of repression in a regime a or level of public deliberation. Weede (1996) notes that. n. iv l C n a lot of this growth has to do with how repressive the regimeUis. In terms of the level of repression, heng chi Weede (1996) uses the seven-point scale of democracy, so that ordinary or semi repressive authoritarian regimes as he labels them, would be labeled at partly free. A partly free regime would be considered as ordinary or semi repressive regimes and would allow for higher levels of. economic growth compared to other authoritarian regimes. Chandra & Rudra (2013) also work off of this idea but extend this idea to incorporate public deliberation and how it can allow for economic development in authoritarian regimes. According to Chandra & Rudra (2013, 253), it is not a regime type that allows for more or less economic development, but rather it is the level of public deliberation. Decentralized decision-making can help explain why there is a relationship between political regimes and economic growth. The authors' key point is that high levels of public. 22 DOI:10.6814/NCCU202000806.

(23) deliberation allow for higher economic growth rates and that this growth can only be experienced by negotiation and participation. In order for a regime to have a high level of environmental performance, it has to have stability. Therefore, the durability of a regime to maintain its power over its civil society is of the utmost importance. When looking at regime transition, it can take many forms, so if a regime is overthrown or a civil war breaks out, then not only will the government, its economy, and its society breakdown, but it can cause significant destruction to its environment as well. For this reason, several authors address the issue of regime transition (Hadenius, Teorell 2007; Geddes 1999a). Regime transition is an important topic because it is imperative to understand which regimes will have the most peaceful transitions in order to gauge which authoritarian regime type will have the least negative impact on the environment.. 政治大 In terms of the nature of the authoritarian regime transition, according to Geddes (1999a) different 立 types of authoritarian regimes have different transition outcomes. For personalist regimes, they. ‧ 國. 學. will tend to have late transitions because they will attempt to hold onto their power for as long as possible. Late transitions will possibly result in an uprising or attempted coup to overthrow the. ‧. regime. As for military regimes, they tend to end in negotiation or bargaining due to most of their. sit. y. Nat. transitioning occurring from internal disagreements. As for single-party regimes, it is unclear what. io. er. the transition outcome is because few transitions have occurred. According to Geddes (1999a), most single-party regime officials believe that there will not be significant changes in the system,. n. a. v. l to negotiate. In terms nofi the outcome of the authoritarian so therefore, they will be more inclined. Ch. U. e explain n g c hthei outcomes for monarchies, single-party regime transition, Hadenius & Teorell (2007) regimes, military regimes, and multiparty regimes. For monarchies (personalist regimes), they explain that these tend to result in a restricted form of electoral monarchism. For single-party regimes transition outcome, as stated before, is more complicated, but tend to result in dominant or nondominant multiparty systems or military regimes. For military regimes, they mostly result in limited multiparty systems, and more multiparty regimes, they mainly result in democratization. In conclusion, when looking at these diverse fields, one can notice a pattern occurring. For economic development, conflict and peace, and regime durability, party regimes are most likely to outperform in all of these categories. This can also be related to the explanations, such as a larger selectorate or incorporating institutions. Regardless, by better understanding these fields,. 23 DOI:10.6814/NCCU202000806.

(24) one can have an idea of which type of authoritarian regime will outperform. As already stated, while these fields do not directly relate to the environment, they can give a better understanding of what should happen in that field based on what is occurring in the ones studied in this section.. 3.3 Authoritarian Environmentalism Authoritarian environmentalism is a concept established to counter the idea of democratic environmentalism and the idea that only democracies were able to obtain high levels of environmental performance. Heilbroner (1974) first established this concept in his work, An Inquiry into the Human Prospect. He wrote this during a time not only of conflicting ideologies, where part of the world pushed for authoritarianism, and the other pushed for democracy but also. 政治大 environmentalism was that most democratic regimes would not be able to fix their economic 立 during a time where income inequality was beginning to rise. His main idea behind authoritarian. ‧ 國. 學. structures to limit the number of inequalities that are inherently built into capitalism. He explains that this issue will pin classes against one another and that the current political institutions in. ‧. democracies will not be able to transform itself fast enough and reduce the amount of income inequality. Therefore, in this type of desperate situation, Heilbroner (1974) believes that states will. Nat. io. sit. y. begin to resort to authoritarianism.. a. er. Gilley (2012) extends this by explaining that there are many alternative concepts and models. n. v l environmental degradation, related to an increase in awareness of n i and one of the most popular Ch. U. e n g(2012) models is authoritarian environmentalism. Gilley c h i defines authoritarian environmentalism by stating: "Thus, authoritarian environmentalism can be provisionally defined as a public policy model that concentrates authority in a few executive agencies manned by capable and uncorrupt elites seeking to improve environmental outcomes. Public participation is limited to a narrow cadre of scientific and technocratic elites, while others are expected to participate only in state-led mobilization for implementation. The policy outputs that result include rapid and comprehensive response to the issue and usually some limits on individual freedoms (Gilley 2012, 288)." Several characteristics develop due to environmental issues that can be fixed with the non-participatory nature of authoritarian regimes: public ignorance, public irrationality, free-riding, the need for immediate action, the lack of availability heuristics to motivate social action, and multistakeholder veto players (Gilley 2012, 292). However, while all of these characteristics are. 24 DOI:10.6814/NCCU202000806.

(25) important, Gilley's key point is that while authoritarian environmentalism will bring a necessary push towards environmental change, the issues of lacking social concern due to the separation and prohibition of representatives outside the government will not only make authoritarian approaches to better environmental performance more critical, but it will also make them more difficult. In addition to Gilley's work, Beeson (2010) also establishes the importance of the structure of authoritarian regimes and their ability to produce effective environmental outcomes. In Beeson's article, he focuses solely on East and Southeast Asia and their historical legacy with authoritarianism. He also emphasizes that there has been a rise in illiberal, authoritarian regimes (Diamond 2008), and especially with the prominent rise of China, the possibility of authoritarianism increasing, especially in Asia, is highly possible. His main argument is that. 政治大 (2010) believes that authoritarian regimes will be more capable than other forms of government at 立 regardless of the adverse effects on civil and human rights that come with authoritarianism, Beeson. confronting environmental challenges. His reasoning is due to the centrality of power. It states, "In. ‧ 國. 學. such circumstances, forms of ‘good’ authoritarianism, in which environmentally unsustainable forms of behavior are simply forbidden, may become not only justifiable but essential for the. ‧. survival of humanity in anything approaching a civilized form (Beeson 2010, 289).” Because of. sit. y. Nat. this reasoning and the fact that regimes such as China have experienced economic growth by. io. er. prioritizing it over political issues, Beeson (2010) believes that governments will choose to implement authoritarianism in order to combat the environmental issues of the future.. n. al. Hypotheses. Ch. engchi. i n U. v. Based on the theoretical arguments made in the three proceeding sections, I derive the following hypotheses:. H1: Authoritarian regime types will have an effect on environmental performance.. H2: Out of all other authoritarian regimes, party states will outperform other types of authoritarian regimes and have a positive impact on environmental performance.. 25 DOI:10.6814/NCCU202000806.

(26) 4. Research Design. 4.1 Dependent Variable: Carbon Dioxide (CO2) Emissions The dependent variable examined in this paper is the amount of carbon dioxide emissions emitted around the globe measured in metric tons per capita. The data retrieved for this variable is from the World Bank's World Development Indicators (WDI) compilation, and it covers the 1997-2010 period (World Bank 2019). I chose this response variable over others based on the starting date from the Kyoto Protocol: 1997. This agreement was put into effect to reduce the levels of greenhouse gas emissions internationally. One hundred ninety-two parties in the United Nations. 政治大. are a part of the Kyoto Protocol, and 84 countries have signed onto the agreement (UNFCCC).. 立. The World Bank identifies carbon dioxide emissions as a greenhouse gas emitted from the burning. ‧ 國. 學. of fossil fuels and the manufacturing of cement. They state that these emissions include CO 2 produced during the consumption of solid, liquid, gas fuels, and gas flaring (World Bank 2019).. ‧. Greenhouse gas emissions are the most detrimental form of environmental degradation because. sit. y. Nat. not only does it lead to temperature increases, which can affect other parts of the ecosystem, but it. io. er. also can stay in the atmosphere for up to thousands of years (EPA 2019). Other forms of degradation, such as deforestation, air pollution, and water pollution may be more apparent in. n. a. v. i only to affect the area where l Ceasily manageable andntend everyday life, but they tend to be more the degradation is occurring.. hengchi U. 4.2 Independent Variable: Authoritarian Regime Type To test the impact of regime type on carbon dioxide emissions, I apply the data collected by Geddes, Wright, and Frantz in 2014 on authoritarian regimes from their dataset, "Autocratic Breakdown and Regime Transitions: A New Data Set." In their dataset, authoritarian regime types are broken down into four main categories, which include party, military, personal, and monarchy. However, several subcategories are included to make up these four main categories: party-based, partymilitary, party-personal, party-personal-military, oligarchy, indirect military, and militarypersonal (Geddes, Wright & Frantz 2014).. 26 DOI:10.6814/NCCU202000806.

(27) For Geddes, Wright, and Frantz (2014), several main occurrences need to happen in order for a regime to be considered autocratic: a leader comes to power undemocratically, military intervention in competitive elections, or informal or formal rules inhibit competitive elections. The authors also note that an autocratic regime can switch when there is a successful competitive election when a government is ousted and replaced by a different regime, and when policies are changed and implemented that effectively alter the way leaders are chosen. Geddes, Wright, and Frantz (2014) also note that the selectorate is essential when determining the type of authoritarian regime. For party regimes, it is the dominant-party dictatorships, for monarchies, it is the royal family, for military regimes it is the military institution, for personalist, it is the individual dictator, and for oligarchies (classified under party regimes) it is competitive elections but with most of the population still powerless (Geddes, Wright, and Frantz 2014, 317-18).. 政治大. Using Geddes, Wright, and Frantz's 2014 dataset, I was able to identify 81 authoritarian regimes. 立. and 51 authoritarian regimes that did not switch to another type of regime during 1997-2010. While. ‧ 國. 學. most research on regime type and the environment focuses on how democracies outperform autocracies, and while this is still majorly true, research and results are beginning to show some. ‧. authoritarian regimes to make efforts to control CO 2 emissions. A party regime such as China, which is also the most significant greenhouse gas emitter in the world, has recently been showing. y. Nat. sit. signs of slowing CO2 emissions in 2018 (Korsbakken, Andrew, Peters 2019). While this is nearly. n. could possibly be a good indicator a of CO2 emission reduction.. 4.3 Control Variables. er. io. a decade after the end of the dataset's timeline, it is still showing that authoritarian regime types. iv l C n hengchi U. In addition to this, I will also be adding several control variables to test how these variables affect carbon dioxide emissions and authoritarian regimes type. These control variables will cover economic, demographic, and industrialization variables. I will briefly describe each of these variables. All of the control variables are retrieved from the World Bank's World Development Indicators database to attribute to whether economic, demographic, and industrialization factors contribute to specific authoritarian regimes having higher CO2 emissions. In regards to economic factors, having higher rates of economic growth can be linked to higher CO2 levels due to a preference for economic growth over environmental protection. Poorer countries that have low levels of. 27 DOI:10.6814/NCCU202000806.

(28) economic growth are said to have lower levels of CO2 emissions, according to Azomahou, Laisney, and Van (2006). Azomahou, Laisney, and Van (2006) also point out that developed nations, such as Western liberal democracies, are more likely to have higher CO2 emissions due to financial, technological, or other vital resources. However, when looking at the World Bank's data, CO 2 emissions in places like western Europe and even the US had either began to level off or decrease around 2014 (World Bank 2019). Therefore, trade as a percentage of GDP is used to refer to trade openness (to the world), and GDP per capita is measured in current US Dollars (US$). In regards to demographics, having a higher rate of population growth and a more densely populated area can significantly affect the environment and also the number of resources needed to provide the population with daily means. In addition to this, urban areas tend to exacerbate climate change in the city and surrounding areas, according to Grimm et al. (2008). The. 政治大. demographic variables are population density and urban population growth, both provided by the. 立. World Bank (2019). Population density is measured by the number of people per square kilometer. ‧ 國. 學. of land area, and urban population growth is measured by the annual percentage of growth in urban areas (World Bank 2019). In regards to the industrialization variable, industrialization is an. ‧. essential factor for both CO2 emissions and authoritarian regimes because most Western liberal democracies have already undergone industrialization. In contrast, in many autocracies in other. y. Nat. sit. parts of the world, industrialization is ongoing or just beginning. Industrialization is also a central. er. io. agent in the reduction of environmental quality in a state, according to Cherniwchan (2012).. n. a as a percentage of GDPi tov represent the industrialization Therefore, I selected industry measured variable (World Bank 2019).. l C hengchi Un. 4.4 Model Selection The central research method for this research will be conducted through quantitative analysis. Primarily, I will be using time-series cross-sectional analysis in order to examine how autocratic regimes test from 1997-2010 in regards to their CO2 emissions. In addition to completing the timeseries cross-sectional analysis, I will also be utilizing Geographic Information Systems (GIS) by mapping out the levels and fluctuation of CO2 emissions in the four different authoritarian regime types. I will be utilizing different databases and sources, with the central database being the World Bank, where data will be pulled for the dependent and control variables. I will also be utilizing the. 28 DOI:10.6814/NCCU202000806.

(29) database prepared by Geddes, Wright, and Frantz (2014) as my primary dataset in order to outline the types of authoritarian regimes. By applying time-series cross-sectional analysis with random effects (RE), I will be able to test the relationship between authoritarian regime type and the level of carbon dioxide emissions within each regime. There are two main benefits when testing variables using the random effects model. Both of these benefits are due to the nature of the data utilized in this thesis. First, because I use country-level data, by using the RE model, the data can be measured properly and at the right level (Bell, Jones 2015, 135-136 as cited in Chang, Wei 2019, 330). Secondly, because the sample size in this thesis is small and includes climate and geographical variables which are time-invariant, the RE model allows for estimations to be made and out-of-sample observations to be predicted, which makes this model the most preferable (Clark, Linzer 2014; Baltagi 2008; Bell, Jones 2015, 136 as cited in Chang, Wei 2019, 330).. 立. 政治大. In addition to this, I will take into consideration the problem of reverse causality and endogeneity.. ‧ 國. 學. In order to avoid these issues, I will lag all dependent variables by one year. All things combined, the RE model in this paper is:. ‧. 𝐶𝑂2𝑐,𝑡+1 = 𝛼 + 𝛽𝐴𝑢𝑡ℎ𝑜𝑟𝑖𝑡𝑎𝑟𝑖𝑎𝑛𝑐,𝑡−1 + 𝐺𝐷𝑃𝑐𝑎𝑝𝑐,𝑡−1 + 𝑡𝑟𝑎𝑑𝑒𝑐,𝑡−1 + 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑐,𝑡−1. sit. y. Nat. + 𝑢𝑟𝑏𝑎𝑛𝑐,𝑡−1 + 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝑐,𝑡−1 + 𝛼𝑐 + 𝛿𝑡 + 𝜇𝑐,𝑡−1. io. er. where 𝑖 stands for country and 𝑡 stands for time, and I test authoritarian regime types (Authoritarian) against my y-variable, CO2, which controls for individual and time effects with 𝑖𝑡.. n. al. iv. The control variables and x-variable represented in this equation, n in the case for removing reverse C. i U. he. n y-variable, g c h represented by 𝑡 − 1. As for 𝛼𝑐 + 𝛿𝑡, causality, are started at one year before that of the this represents the controls for individual fixed effects and time fixed effects, and 𝜇𝑐,𝑡 represents the idiosyncratic error term. In addition to this, in order to increase the robustness of the research and make sure that the hypothesis and results are valid, I will also be utilizing a second method, spatial analysis using Geographic Information Systems (GIS). This form of analysis will allow me to not only include visual results through mapping, but it will also allow me to include some other figures such as a boxplot and a parallel coordinate map to better understand the data and look for any possible outliers. In order to obtain more efficient results, several regressions will be run and we will also test for clustering through the Local G* clustering map to see if there is any correlation in certain. 29 DOI:10.6814/NCCU202000806.

(30) areas between regime type and carbon dioxide emissions. I will also include a bubble chart to see the relationship between regime type and CO2 emissions and the relationship between GDP per capita and CO2 emissions. Lastly, the main form of analysis in this section will be based on three regression models: a classical regression model, a spatial lag regression model, and a spatial error model. The reasoning for this is that after testing several distance weighted variables with different distance bandwidths and different numbers of nearest neighbors, the results concluded that all three of these models should be tested. For example, the bandwidths set to the automatic setting (6,639.41) showed positive results to test for spatial lag and error, but 10,000 km and 20,000 km did not show any significant results.. 政治大 feasibility and data transparency. In立 terms of authoritarian regimes' data transparency, because of. To conclude, when looking at the methodologies and data used, it is also imperative to look at. ‧ 國. 學. these regimes' secretive and confidential nature, it usually is challenging to obtain data from these types of regimes. Because of this, there is the possibility of the data being skewed or incorrect in. ‧. order to best present the leader or party's best interests. However, while keeping this in mind, caution was taken when reviewing databases and data sources. While there are some issues of. Nat. sit. y. missing or unavailable data for specific authoritarian regimes, after reviewing the data, I believe. er. io. that there is enough to run a comprehensive analysis. Therefore, while I believe that there will be some issues of feasibility that willabe uncovered during this research, I believe that I can still. n. iv l C n provide a comprehensive analysis on this relevant we are still dealing with today. h e n gissue chi U. 30 DOI:10.6814/NCCU202000806.

(31) 5. Results: Time-Series Cross-Sectional Analysis & Geographic Information Systems. 5.1. Time-Series Cross-Sectional Analysis. Before discussing the statistical results of this research, we will first provide summary statistics through several tables provided in this section below. In the first portion of this section, we will cover the summary statistics of the main analysis: time-series cross-sectional analysis. These statistics will include number of observations, frequency of the variables, standard deviation, maximal values, minimal values, and skewness. We will also include some graphs depicting the. 政治大 second portion, the regression results will be shown. 立. relationship between authoritarian regime type and CO2 emissions over the chosen timeline. In the. ‧ 國. Summary Statistics: Time-Series Cross-Sectional. ‧ sit. y. Nat. Analysis. 學. 5.1.1. er. io. In order to better understand the data and what the relationship was between the variables carbon. n. dioxide emissions (metric tons per capita) a and authoritarian regime type, we first reviewed the data. l. iv. n and determined which authoritarian countries of CO2 in each regime type. C are the largestUemitters hen chi. g gap between the top emitter for monarchies When looking at figure 5.1, the results show a large. and the other top emitters in other regimes. For party regimes, Singapore is the top emitter, for military regimes, it is Algeria, for personal regimes, it is Russia, and for monarchies, it is Kuwait. While it is not shown in this graph, in general, it is monarchies that can be considered the top producers of per capita carbon dioxide emissions. However, when looking at overall emissions measured in kilotons, the results show a much different estimate with China being the top CO 2 emitter by a long shot with over 800,000 kt by 2010. The other four top emitters (Russia, Saudi Arabia, Iran, and Kazakhstan) rest under 200,000 kt. However, in this research, we will be utilizing per capita CO2 emissions in order to have a better variation.. 31 DOI:10.6814/NCCU202000806.

(32) 立. 政治大. ‧. ‧ 國. 學. io. sit. y. Nat. n. a. er. In addition to this, one can also look at Figure 5.2 to gain a better understanding of the relationship. v. l carbon dioxide emissions. between authoritarian regime type and n i This figure shows on average. Ch. U. h i performed in carbon dioxide emissions n g chave how the four different types of authoritarian e regimes from 1997 to 2010. The solid line, representing monarchies, are shown to perform the worst, while the other three regime types (party, military, and personal) are averaging between 2-3 metric tons per capita. Based on the literature, the hypothesis we chose focused on party regimes outperforming any other type of authoritarian regime type. However, based on this preliminary figure, one can notice that military regimes are emitting the lowest amount of CO 2 (metric tons per capita) emissions. This could be due to several reasons, such as fewer observations to pull from or low levels of development; however, this could possibly have no effect on the final results.. 32 DOI:10.6814/NCCU202000806.

(33) 立. 政治大. ‧. ‧ 國. 學. Figure 5. 2: Average CO2 (Metric Tons per Capita) Emissions in the Four Different Authoritarian Regime Types (1997-2010). sit. y. Nat. er. io. After seeing the results in Figure 5.1 and 5.2, it is important to understand which countries are in. n. what authoritarian regime type. Therefore, in Table 1, one can better a v understand out of all fifty-. l. i. C h ones are categorized one regimes utilized in this research, which U n into which authoritarian regime engchi. type. For monarchies, one can see that out of the seven countries considered as a monarchy, four of them are ranked as the top four carbon dioxide (per capita) emitters. The other three monarchies (Eswatini (Swaziland), Jordan, and Morocco), all have lower levels of CO 2 emissions (per capita), which show that these other four monarchies are skewing the results for the other three monarchies. This is more than likely the result of these four monarchies being large oil producers, including Russia which is labeled as a personal authoritarian regime.. 33 DOI:10.6814/NCCU202000806.

(34) Table 1: Summary of Countries Per Regime Type Regime Type. Country Military. Monarchy. Party. Personal. Algeria. 1. 0. 0. 0. Angola. 0. 0. 1. 0. Armenia. 0. 0. 0. 1. Azerbaijan. 0. 0. 0. 1. Belarus. 0. 0. 0. 1. Botswana. 0. 0. 1. 0. Burkina Faso. 0. 0. 0. 1. Cambodia. 0. 0. 1. 0. 0. 1. 0. 1. 1. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 0. 1. 0. 1. 0. 0. 1. Cuba. Eswatini. Gambia. al. n. Gabon. io. Ethiopia. Nat. Eritrea. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. ‧. Egypt. 0. 學. Congo, Dem. Rep.. ‧ 國. China. y. 立. sit. Chad. 0 政00 治大 0. er. Cameroon. i n Ch 0 e n g c h i U0 0. 0. v. Guinea. 0. 0. 0. 1. Iran. 0. 0. 1. 0. Jordan. 0. 1. 0. 0. Kazakhstan. 0. 0. 0. 1. Korea, North. 0. 0. 1. 0. Kuwait. 0. 1. 0. 0. Kyrgyz Republic. 0. 0. 0. 1. Lao PDR. 0. 0. 1. 0. Libya. 0. 0. 0. 1. Malaysia. 0. 0. 1. 0. 34 DOI:10.6814/NCCU202000806.