The Predominant Role of Domestic Interests in Setting Ambition of Nationally Determined Contributions to Mitigation Efforts By Todd A. Eisenstadt, American University Department of Government Ward Building 244 School of Public Affairs American University 3600 Massachusetts Avenue NW Washington, DC 20016 Corresponding author eisensta@american.edu and Daniela Stevens Leon, American University Department of Government Ward Building 244 School of Public Affairs American University 3600 Massachusetts Avenue NW Washington, DC 20016 stevens.daniela@gmail.com Key Words: climate change, vulnerability, NDCs, Paris Agreement, mitigation, extreme weather events, climate change policy, energy subsidy, carbon intensity Abstract: In one of the first efforts in the social sciences to explore the political and economic as well as natural causes of climate change policy ambition, this article assesses nations’ climate change policies by using ratings of the Nationally Determined Contribution (NDC) pledges made at the 2015 Paris meeting of the United Nations. Our exploratory analysis shows that some domestic determinants of NDCs address political interests rather than strictly responding to mitigation needs. Key policy insights include the following: 1) nations subsidizing energy have less ambitious NDCs, implying that more effort may be needed to diminish direct effects of energy lobbies on climate policy; 2) vulnerability measured as economic damage and human loss caused by extreme weather events correlates only weakly with NDC ambition, meaning that greater government responsiveness to extreme weather vulnerability may be necessary in setting the scope of climate change policy; and 3) higher carbon emissions intensity is actually associated with less ambitious mitigation pledges. In other words, high emitter nations are actually doing less than low emitters. We conclude that if nations are left to their own devices, more ambitious NDCs are less likely to emerge from more fossil fuel-dominated, higher emitter nations and that true mitigation of climate change effects requires stronger and more enforceable national actions. The 2015 Paris Agreement was hailed as an overwhelming victory for climate change mitigation as the United Nations (UN) reached a consensus to invest heavily in solving this hitherto intractable problem after a decade of failed negotiations. In preparation for the Conference, over 180 countries sent “intended nationally determined contributions” (abbreviated as INDCs) presenting individual nation climate action plans. The use of this nation-by-nation “bottom up” structure of voluntary pledges to construct an international climate change policy was controversial, but nothing else had worked since the implementation of the Kyoto Protocol nearly two decades earlier. Some scholars, like Bodle et al. have argued that even beyond the legal text, the Paris Agreement established the commitment to long-term and increasingly stringent goals, evaluated through transparent accountability mechanisms, and signaling to all relevant actors, including business and financial sectors that “green investing” would become a worldwide force in deciding financial flows (Bodle et al., 2016:2). However, others (such as Rogelj et al. 2016 and Höhne et al., 2017) have questioned the ambition of the agreement, asking how the gap can possibly be closed between the agreement’s ambitious global objectives and humble national contribution pledges. This article seeks for the first time to systematically understand why some countries have more ambitious INDCs than others. While INDC are considered the building blocks of the post-Paris UN climate change governance structure, guidelines on their voluntary and un-regulated implementation have not yet been tightened, and these measures of national commitment have yet to be systematically studied and compared. We seek in this article to offer some initial patterns from statistical analysis of a cross- sectional study of 2015 INDCs from the world’s largest emitters. While we are limited to using the 56 NDCs coded for “levels of ambition” by the research organization Climate Action Tracker (CAT) Consortium, this admittedly preliminary assessment nonetheless identifies several important statistical patterns for a universe of countries accounting for over 80 percent of the world’s greenhouse gas (GHG) emissions. We find, contrary to expectations, that while a nation’s level of vulnerability does not affect the level of ambition of its INDC, nations with large oil subsidies are less likely to propose ambitious INDCs, and that larger emitters are also less likely to propose INDCs with higher levels of ambition. These findings have important policy implications which we consider in the article’s conclusion. The scientific consensus points that the Paris Agreement system of allowing nations to volunteer their levels of pledges is inadequate to the overall task of reducing climate change to the UN’s stated goal of an average temperature rise of 2ºC by the end of the 21st Century. The paper confirms skeptics’ admonition that voluntary approaches fall short for reducing global warming, and that national contributions to emissions reductions are weaker where current carbon emissions are higher. Using the CAT’s INDC Ambitiousness Assessment, we find a negative and statistically significant relationship between carbon emissions and INDC ambition. In other words, the nations doing the most to perpetuate the problem are the ones doing the least to solve it. The findings are preliminary because this is a cross sectional analysis, and because the sample is small. Even so, the results offer a strong warning. INDCs as a Compromise after an Era of Stalemate The signature of the United Nations Framework Convention on Climate Change Convention (UNFCCC) in 1992 represented a major step through which international parties agreed to stabilize greenhouse gas (GHG) atmospheric concentrations to prevent further interference with the climate. In Kyoto 1997, countries jointly signed an agreement setting specific targets to reduce emissions that would become binding only after ratification. The Kyoto Protocol (hereby Kyoto) established ambitious emission reductions on nations. However, Kyoto exempted some of the world’s biggest carbon emitters, like China and India. The United States signed but did not ratify the agreement arguing that it promoted an unfair adjudication of responsibilities between developed and developing countries, and Canada, another major emitter, withdrew from the treaty while it was in effect. As the first period of Kyoto ended in 2012, some countries refused to commit to a second round of commitments. The Kyoto protocol can be characterized as a traditional “top-down” agreement, because even though parties established their pledges and chose to ratify or not, the negotiations occurred jointly and countries were not given an opportunity to bring to the negotiating table a strategy that was feasible, ambitious and fair according to their own terms. Ultimately, Kyoto failed as only the European Union and a few other polluters (responsible for slightly more than a third of the world’s total carbon emissions) continued to submit themselves to the UN’s mandate. Hence, after nearly a decade of negotiations without an agreement, the UN settled on a new scheme based on voluntary reductions by nations. The advantage of INDCs was that nations stated themselves the climatic actions that they were willing and able to undertake, which was instrumental in reaching the Paris Agreement. The INDC mechanism, and the Paris process more broadly, were regarded by diplomats as practical and realistic tools to reach common goals with differentiated pledges. The nationally-determined contributions were summed to establish a starting goal for the Paris Agreement. That starting goal has been found inadequate (Rocha et al., 2016) and the institutional structures needed to monitor pledges are lacking (Aldy, 2015). Indeed, the level of ambition of INDCs has been questioned by scientists and policymakers alike. Critics argue that they are not stringent enough to slow down the warming of the world (Sachs et al., 2015, LeTreut and Beneviste, 2015, Boucher and Beneviste, 2015). Most highlight their shortsightedness and coincide in the need of longterm de-carbonization plans that commit beyond 2030 (Sachs et al., 2015). Political scientists have only recently approached the debate, and this article seeks to start a discussion on the inadequacy of the INDCs based on their lack of ambition, and, for the first time, by showing how a nation’s level of emissions is correlated –but negatively– with what nations offer. INDCs were the building blocks of the Paris Agreement and after its ratification they remain central for the pursuit and realization of the national commitments. The Agreement establishes that parties report successive Nationally Determined Contributions (NDCs) every 5 years, with the expectation of raising the bar of their domestically determined efforts. Hence NDCs are key to the future climate change tackling scenarios and systematic analysis of their determinants is overdue. Hypotheses Guided by theory (Adger, 2006) and other early social science empirical work (Eisenstadt and West 2017, and Eisenstadt et al., 2017), we contend that a) vulnerability to extreme weather events is associated with a higher level of ambition of NDCs. Second, we test the “fair attribution of responsibilities” argument, as made by developing nations under Kyoto (see Meyer and Sanklecha, 2017) and highlighted in the ongoing North-South Debate by scholars like Najam (in Axelrod and Vandeveer) that the larger CO2 emitters should take greater responsibility and offer more ambitious NDCs. We claim that b) the current major emitters are associated with more ambitious pledges. Finally, based on studies that empirically demonstrate that countries from the G20 have not followed their commitment to phase out fossil fuels because they give in to the pressure from energy intensive industries (Bast et al., 2015), we test that c) subsidies to electricity produced with fossil fuels have a negative association with ambitious emission pledges. We test these three hypotheses and provide empirical support using an Ordered Logit Model (OLM). Contrary to expectations, our results show that nations with higher levels of CO2 emissions offer less ambitious pledges and that vulnerability to extreme weather events does not provoke more ambitious pledges. As we did expect, energy subsidies, a proxy for the political influence of carbon-intensive sectors like oil, does correlate negatively with the ambition level of the NDCs. Vulnerability, Energy Subsidies and Current levels of emissions: Introducing Correlates of Nationally-Determined Contributions Vulnerability Clearly, climate change produces climate variation frequently expressed in the form of Extreme Weather Events (EWEs) like cyclones, hurricanes, floods, droughts, ocean current changes, tsunamis, and monsoons. The 2013 report of the Intergovernmental Panel on Climate Change (IPCC) found that over the past 50 years, temperatures have reached unprecedented extremes. Heat waves have become more frequent, and cold waves have become less frequent over time (IPCC, 2013:8). The report also describes an increased incidence of extreme high sea levels, tropical cyclone activity and other EWEs. Most or all of the global warming experienced over the past decades was anthropogenic or human-made (IPCC, 2013:10). The Bulletin of the American Meteorological Society (BAMS) identified a list of events attributable to climate change in 2015, finding that all heat-related event, without exception, were made more intense or likely due to humaninduced climate change. Furthermore, they show that “attribution science” has reached a sufficient level of development capable of separating anthropogenic causes from natural drivers (BAMS, 2015). The social sciences have been largely missing from the debate about vulnerability to climate change. Vulnerability assessments and sectoral exposure assessments abound in the natural and social sciences, but little attention (with the partial exception of Eisenstadt and West 2017) has been paid to vulnerability to EWEs as a factor eliciting preventative policy action rather than post hoc emergency responses. While studies do exist considering the role of governments in responding to such emergencies, theoretical positions linking vulnerability to EWEs and longer-term climate change policy actions is scarce. However, empirical studies that link EWEs to economic losses abound. Preston finds a trend of “increasing economic losses from extreme weather events” in the United States, attributed to an increased exposure to extremes (2012:719). Some authors have identified the oil and gas sectors as particularly threatened by EWEs, especially given that much of the sector’s infrastructure is located in low-lying areas or offshore (Cruz and Krausmann, 2013:41-42). Others have provided evidence of negative impacts of climate change on the energy sector (Wilbanks et al. 2007; Mabey and Mitchell 2010), on transportation (Savonis et al. 2008), and on small and medium businesses in the construction sector (Wiedatta et al., 2011:116). Admittedly, projected exposure to EWEs is expected to increase regardless of climate change (Preston, 2012:719). Similarly, some EWEs will continue to occur independent of climate. Overall, the relationship is not always linear and other elements influence EWEs’ economic impact, but scientists consistently agree that the role of climate change “cannot be excluded” (Visser et al., 2014:461). Definitions of vulnerable differ by nation, so it is necessary to use a single crosscutting measure. The NDCs do not indicate economic or human costs paid due to recent EWEs. Over 77% of low income countries, and 91% of e high-income countries do not report a concrete calculation of costs caused by EWEs, and only 16% of low-income countries (and no high-income countries) consider projected costs of climate-related hazards in their NDCs (Pauw et al., 2017). Role of energy subsidies Energy subsidies worldwide may be as high as $US 5.3 trillion per year, or 6.5 percent of the world’s GDP (Bridle et al., 2014). Most analysts oppose subsidies because of their warming effects on the environment and because they tend to have regressive distributional effects (Victor, 2009, Lehmann, 2007, Laan and Beaton, 2010, Beaton and Lontoh, 2010, Lin and Oyuang, 2014). In general, the empirical literature agrees on two broad points. The first is that there is a need to gradually reduce fossil fuel subsidies, but the second is that there are inherent difficulties in doing so because of the political economy of subsidization and the obstacles to changing existing policies (Victor, 2009:79). Several studies find that phasing out fossil fuel subsidies would bring an important environmental benefit (Ellis, 2010, Bridle et al., 2014). Burniaux and Château (2011) found that removing subsidies in non-OECD countries would aggregate to a seventh of the effort needed to maintain global warming below 2°C. Industrialized countries have made little progress in this regard (Bast et al., 2015), as OECD nations implement around 550 mechanisms that encourage the production or consumption of fossil fuels (OECD, 2013). Interest group theories support our argument, claiming that organized groups have the resources, organization, and interest to maintain subsidies in place. By this argument, policy-makers obtain electoral advantages, acquiring resources for their districts or campaign donations with the goal of staying in power (Baumgartner and Leech, 1998, Olson, 1984, Dür and De Bièvre, 2007, Victor, 2009). Multiple organizations claim a direct link between the fossil fuel industry and the continued subsidization of either energy consumption or production (Bast, et al., 2015, Coady, 2015). Here, we will use as a proxy the amount in US dollars that countries direct towards energy consumption subsidization, using data form the International Monetary Fund (IMF). The next section offers our third and final hypothesis on historical emissions. Then we describe our dependent variable, the level of ambition of INDCs. Afterwards we present our operationalization of variables, followed by our statistical analysis and discussion. Carbon Emission intensity The adjudication of responsibilities for emissions has been a constant source of tension and negotiation stalemates between the developing and the developed world. For the former, industrialized nations must reduce their contribution to global warming to a larger extent, given that they are the major historical emitters. Scholars such as Althor et al. (2016) have pointed out the inequity in “the burden of climate change,” and most agree that countries need to better differentiate responsibilities (Mbeva and Pauw, 2016). Emission intensity is the volume of emissions per unit of output, usually GDP or energy. While total emissions are a strong indicator of a country’s absolute contribution, and international efforts (including Kyoto) favor it, emissions intensity can be considered a lower bar, that is, a more permissive and conservative measure for intensive emitters. Granted, this permissiveness applies to the developed world as for the countries that currently undergo industrializing processes. We aim to test the effects of carbon dioxide (CO2) emission intensity on nations’ plans for reducing future emissions, assuming a higher intensity will be associated with more ambitious plans. For our measures, we use the ratio of CO2 per unit of energy. Therefore, intensity reflects not only a nation’s emissions but also its technological efficiency in energy production. Dependent Variable: Ambitiousness of the INDC The dependent variable of this study is the level of ambition of each country’s INDC mitigation pledge. Comparing NDCs has been challenging, as they do not follow a standard format. Countries choose their own targets, reporting standards, and timeframes. For example, China, the largest worldwide emitter, committed to reaching a peak of CO2 emissions “around 2030,” and to lowering its carbon intensity, measured as CO2 emissions per unit of GDP, by 60% to 65%, using 2005 as a baseline year (Chinese Department of Climate Change, 2015). In contrast, the United States, the second largest emitter, pledged a reduction of 26% to 28%, using also the 2005 baseline, but committing to do so by 2025. The European Union, consistently the lowest emitters among industrial nations, set a GHG emission reduction of “at least 40%” by 2030, but with a baseline comparison of 1990. Moreover, the EU explicitly stated that this commitment was binding (Latvian Presidency for the Council of the European Union, 2015). Given these important limitations, the Climate Action Tracker (CAT) effort sharing assessment of mitigation contributions to GHG reductions seems to be the best single measure of NDC ambition levels. Based on existing literature, including the IPCC report, the CAT assessment includes considerations of responsibility, equity, capability, and equality, considering only a compatibility with the 2°C limit under several scientific interpretations of fairness. As of early 2017, 163 countries had submitted NDC contributions to the United Nations, including the 28 European Union members, which submitted a joint NDC (i.e., one aggregate pledge rather than 28 separate ones). The CAT assessment evaluates 32 INDCs, including the EU member states’ contribution as one document. The CAT covers 56 countries when those in the EU are broken out individually. Although the coverage might seem small given that the UN has 190 member nations, the countries analyzed do represent 81.3% of 2010 global emissions. The CAT rating system evaluates the INDCs’: a) impact on emissions, b) effect of current policies on emissions, c) fairness of share of global effort to limit warming below 2°C.1 The system rates countries as follows: • Inadequate: Emission targets in this range are less ambitious that the 2ºC range. If all governments adopted this position, global warming would exceed 3ºC. • Medium: Emission targets in this range are in a 2ºC range, though in the least stringent part. If all governments adopted this position, global warming would likely exceed 2ºC. • Sufficient: Targets in this range are in the most stringent part of the 2ºC range. If all governments adopted this position, global warming would likely be below 2ºC. • Role Model: The emission targets in this range are more ambitious than the 2ºC limit. 1 The criteria are: 1) Responsibility: determined by the level of historical emissions of a country. See UNFCCC, 1997. 2) Capability: determined by the level of economic capability of a country (GDP/capita or the human development index). 3) Equality: Emissions per capita converge to the same level for all countries. See Chakravarty et al., 2009; GCI, 2005. 4) Equal cumulative per capita emissions: Emissions need to be reduced so that cumulative emissions per capita reach the same level. See Pan, Teng, & Wang, 2013; WBGU, 2009. 5) Responsibility/capability/need: A range of studies have explicitly used responsibility and capability as the basis for distributing emissions reductions. See Paul Baer, Athanasiou, Kartha, & Kemp-Benedict, 2009; Winkler, Jayaraman, et al., 2011. 6) Capability/cost: combination of mitigation potential and capability, using equal costs or welfare loss per GDP as a basis. 7) Staged: differentiation of commitments in various stages using many equity principles. See Michel G J den Elzen & Meinshausen, 2005; Höhne, Gardiner, Gilbert, Hagemann, & Moltmann, 2008. See all of this detailed methodology at the CAT “Comparability of effort” methods section at (http://climateactiontracker.org/methodology/85/Comparability-of-effort.html). Accessed on October 29, 2016. An NDC rated “medium” is one that commits to reductions that fall in the lower half of the emissions range of what could be considered as “fair.” If the pledge is in the upper half of the range, its rating is “sufficient.” The assessment categorizes 15 countries as inadequate, 40 as medium and 5 as sufficient, considering the European Union members individually. We operationalize inadequate with a 0, medium with a 1, and sufficient with a 2. Table 2 in the Appendix presents the descriptive statistics of all the available data in the dataset. The variables cat ordinal and cat dummy attribute to the individual members of the European Union the one value attributed as per the CAT assessment. For the ordinal variable, 66.6% (40) of the countries obtained a rating of inadequate, while 25% (15) were rated as medium, and only 8.3% (5) as sufficient. No country was categorized as role model. Independent Variables of Interest: Vulnerability, Subsidies to energy produced with fossil fuels and current levels of emissions We use two measures for our first independent variable of interest, vulnerability: material loss from EWEs as a percentage of GDP, and the yearly death toll, both averaged over the 20 years from 1995 to 2014. The data come from Germanwatch’s Global Climate Risk Index (CRI).2 The data considers weather events have caused damage to property or persons, and includes events such as storms, floods, extreme temperatures and heat and cold wave. It excludes geological factors, since they do not depend on the weather and 2 In turn, GermanWatch uses the data from the Munich Re NatCatSERVICE. For the complete Briefing Paper see “GermanWatch ratings of countries” by Kreft et al., 2016; accessed on August 9th, 2016. cannot be linked to climatic change. The CRI score ranks Honduras as the most vulnerable, and the least vulnerable are Hong Kong, Qatar, San Marino, and Sao Tome and Price. The two sub-indicators from the CRI are specifically: 1) Number of deaths per 100 000 inhabitants, and 2) Losses per unit of Gross Domestic Product (GDP) in million US dollars (original values, inflation adjusted). Both of these are 20-year averages. Although possibly correlated, fatalities and economic losses do represent different dimensions of vulnerability and capture differentiated effects of weather events. There are many missing data points within the period 1995-2014 for small or politically unstable countries. These cases are left out from the analysis. The descriptive statistics of the vulnerability-related explanatory variables show a wide range, with the death toll for Myanmar averaging 13.1 deaths over 20 years, while a few countries, such as Libya, Turkmenistan, Eritrea, and Equatorial Guinea had zero deaths due to extreme weather. The GDP losses variable had a minimum of 0% and a maximum of 9.06%. If we consider that the average was .45%, the maximum obtained by Grenada is very high. Above the average are also Bahamas (2.2%) and Kiribati (7.9%), while countries like Gabon and Qatar have a 0% loss. Regarding the possible influence of the fossil fuel interest groups on INDC ambition, we used energy subsidies as a percentage of the GDP as our proxy, with data from the IMF for 2013 (IMF, 2015). These data are consumer subsidies, which in principle aim to protect consumers, no producers. However, these subsidies also benefit producers. Not only do they provide relief to low-income households, but they also foster energy consumption in other sectors, for example, benefiting energy-intensive industries. Subsidies also contribute to budget deficits, and constitute a share of resources that a government could allocate to other areas of public spending, or to renewable energy (IMF, 2015, OECD, 2016) We consider a pre-tax subsidy to be a higher bar for the statistical associations we expect to find, as well as a more conservative estimate, and thus we use this estimate in our model. Reports present different figures of the amounts, but most agree that the pretax subsidies have declined progressively over the past couple of decades, and that developing nations are largely accountable for them (IEA, 2015, IMF, 2015). The mean for this variable is 2.4%, with 41 countries having values of zero. Iran possesses the highest value in the sample, 20.01%, while Zimbabwe is the second highest, with a subsidy of 19.22%. In earlier models, we also tested post-tax subsidies, but those have proven to be controversial and made little difference in the findings. The third independent variable we test is the CO2 emissions intensity for the year 2012. Here we test if the most intensive emitters are the ones taking action to mitigate climate change. As controls, we test GDP per capita, population growth rate, percentage of total of electricity production from renewable sources, a measure of the country’s regime type (whether democratic or authoritarian), the country’s past level of commitment to other international environmental treaties, and a dummy variable for EU membership. The summary statistics of our data are presented as Table 2 in the Appendix. We use an ordinal logit model (OLM), appropriate for dependent variables that have a natural ordering, as in this case, inadequate, medium and sufficient levels of NDC ambition. Given that only five cases qualify as possessing sufficient ambition, in Table 3 in the Appendix we include a logit model where we merged two categories into one. Our binary variable is 0 for inadequate, and the other is 1, the mix of medium and sufficient. Methodologically, it makes sense to add them because both include pledges whose ambition is appropriate to comply with the 2ºC scenario. The bulk of the observations (75%) fall into this collapsed category. We present the marginal effects of all models for a better interpretation. We consider the EU members as separate observations, as if they had submitted different NDCs, thereby increasing our total sample sizes to 60 and 56. Although EU members submitted a joint document, we utilize the variance on the independent and control variables to increase the sample, and attribute to each EU member the same ambition assessment of medium. Results and Discussion Table 1 presents the results of two OLM models. Model 1 only considers the vulnerability dimension operationalized with two independent variables, while Model 2 incorporates the rest of the IVs and controls. Given the small N, and that vulnerability represented our principal theoretical interest, for Model 1 we regressed only the two variables that operationalize vulnerability: economic losses and death toll caused by EWEs, on the CAT ambition assessment dependent variable. Model 1 shows that only the measure of economic losses is statistically significant in the expected direction; that is, an increase of one percentage point in economic losses a country experiences is associated with a decreased likelihood of 8 percent of having an inadequate INDC. An increase of a percentage point in economic losses corresponds to an increased likelihood of over 60 percent of having a medium NDC, and of over 40 percent of having a sufficient NDC. However, this model is clearly underspecified. Model 2 in Table 1 includes all the IVs of interest: the vulnerability measures, energy subsidies, and carbon emission intensity, plus the controls. The results show that vulnerability expressed as economic losses has a statistically significant association with the dependent variable at the .05 level, but only for one of the three outcomes, inadequate, and in the expected direction. The model shows that an increase of a percentage point in economic losses due to EWEs corresponds to a lower likelihood of an NDC being assessed as inadequate. For the most part, then, neither indicator of vulnerability correlates with national NDC ambition levels. In other words, the vulnerability of a nation’s population – in terms of economic loss and mortality – does not prompt nations to propose more ambitious NDCs. Surprisingly, contrary to expectation, nations are not responsive to EWEs faced by their populations via mitigation actions. Regarding emission intensity, the results in Model 2 are contrary to expectations: a higher intensity is associated with less ambitious NDCs. As the amount of carbon emitted (as a result of using one unit of energy in production) increases by a kilogram, the NDC is 17 percent more likely to be rated inadequate. That is, the proposed pledges will be more likely be so unambitious that if all countries adopted pledges within this category, the warming of the world would exceed 3ºC, according to the CAT. Model 2 also shows that as emissions intensity increases by a kilogram, the NDC is less likely to be rated as medium by 10 percent. The CAT reports that if all governments adopted this position, global warming would likely exceed 2ºC. Similarly, an increase of emissions intensity corresponds to a decreased likelihood of having a sufficient rating by 6.7 percent. A sufficient NDC is one, which, should all nations adopt a similar NDC, would limit global warming to less than 2ºC. It may be that the most intensive emitters do not make ambitious pledges because they consider such pledges disproportionately costly. The top emitters by emission intensity include countries that traditionally have used the argument of “growth first, the environment second,” such as Russia, China, India, and Brazil. Vulnerable small island nations can easily cut emissions by 50 percent or even more, especially if they are service- or tourist-based economies with negligible emissions. But any cut by China or India implies a sacrifice of industrial output or, at the very least, subsidy of solar energy or some other resource. The opportunity cost of emissions reduction is extremely high for CO2 polluter, by definition, and hence they have dragged their feet and, as we have shown, actually pledged much less in relative-term measures by the CAT. With the exception of the United States, Canada and the European Union, the most extensive polluters are developing nations. 3 Indeed, the most affluent nations are better enabled for initiating the process of “decoupling,” whereby their GDP grow s as their emissions decrease. This includes Austria, Denmark, Finland, Switzerland, and surprisingly, also the United States.4 For a complete chart of the top polluters, including also polluters per capita see Ge and coauthors’ “6 Graphs Explain the World’s Top 10 Emitters” at (https://wri.org/blog/2014/11/6-graphs-explain-world’s-top10-emitters). Accessed on June 7, 2017. 4 Visit Aden’s article “The Roads to Decoupling: 21 Countries Are Reducing Carbon Emissions While Growing GDP” for a full account of the countries that are de-linking growth from emissions at (https://wri.org/blog/2014/11/6-graphs-explain-world’s-top-10-emitters). Accessed on June 7, 2017. 3 Table 1. OLM Results. 1. OLM with vulnerability variables only Death toll GDP losses CO2 emission intensity Energy subsidies GDP Population growth rate GDP per capita Renewable Electricity Treaties index Polity (regime type) EU membership dummy n=60 r2= .44 2. OLM with all explanatory variables and controls Inadequate Medium Sufficient Inadequate Medium Sufficient -.08 .1 -1.08 .4** .049 .064 .64 .3* .03 .04 .44 .2* -.03 .07 -.908 .45** .17 .06** .099 .04* .08 .05 .02 .04 .55 .35 .10 .06* -.06 .03** .05 .03 .01 .029 .3 .2 -.06 .03** -.03 .02 .03 .02 4.32e-06 1.74e-06** -2.65e06 1.48e06* -1.66e-06 9.63e-07* -.005 .004 .03 .04 .023 .009** -.39 .11*** n= 56 r2= .29 .003 .003 -.019 .028 -.01 .008* .24 .11** .002 .002 -.012 .017 .009 .005** .15 .08* The variable of energy subsidies is also significant for the outcomes inadequate and medium. The model shows that if energy subsidies (as a percentage of GDP) increase a percentage point, the NDC is 17 percent more likely to be rated as inadequate. The substantive significance is relatively large for this category. The sign reverses direction to indicate that as energy subsidies increase a percentage point, the NDC is 6 percent less likely to be medium, meaning that the more countries subsidize energy consumption, the less likely they are to present NDCs that have a medium level of ambition. The dichotomous variable that captures European Union membership largely explains the high r2 of both models. If we remove this variable, the r2 lowers from .44 to .29, but the variables emission intensity and energy subsidies retain the statistical significance that they had in Model 1. The vulnerability variable capturing economic losses due to EWEs loses significance without the EU dummy. The logit model in the Appendix reinforces these results. An increase in carbon emission intensity is negatively associated with the probability of presenting an NDC rated as medium/sufficient, just as are energy subsidies to consumption. Both associations have significance at the .001 level. Similarly, GDP losses accrued after EWEs have a positive and statistically significant association with the DV; that is, as the losses increase by half of a percentage point, the likelihood of presenting an NDC rated as medium/sufficient increase by 20%. Conclusions: The Need for Increased Ambitiousness if NDCs are to be Foundation of Effective International Climate Change Mitigation In addition to highlighting broad critiques of NDCs, such as their non-standardized measurement, lack of ambition, and the failure to construct means of oversight or enforcement of these pledges even as they do exist, this exploratory analysis provides evidence that some domestic determinants of NDCs are problematic as they pertain to domestic constraints on constructing rational policies rather than on international mitigation demands. First, we find some initial results that show that the subsidies to energy have a negative influence on the level of ambition. This is normatively bad in that it demonstrates what many analysts have suspected, that domestic constraints related to political economy are diminishing the ambition levels of NDCs. The designers of the Paris Agreement system implicitly assumed that NDC ambition would be driven by the international demand for mitigation, but this is not the case. We found that two important dimensions of vulnerability, economic damage and human loss caused by EWEs, lose their correlation with NDC ambition when we consider important regressors such as levels of energy subsidy. Moreover, the implication that governments are not responsive to extreme weather vulnerability in setting the scope of their climate change policy also implies a failing on the part of national governments to design climate change policy consistent with national-level climate change demands. The third finding, also normatively pessimistic, is the most consistent result across models, showing that higher carbon emissions intensity is associated with less ambitious mitigation pledges. This finding is likely explained by the fact, mentioned above, that bigger emitters have a “longer way to go,” and thus are only willing to make smaller and much more incremental changes. This last finding is more consistent with developing country arguments for decades, and less surprising perhaps than the first two, which each demonstrate that NDC formulation has suffered from extreme domestic politics limitations. The upshot is that if nations are left to their own devices, more ambitious NDCs are less likely to emerge from domestically-hamstrung nations. Those who championed the Paris Agreement often claimed that while imperfect, the agreement, and its central foundation on NDC pledges, could succeed as they were but the beginning of a long process through which national contributions would be racheted up over time Even given the small size of our sample, our necessary cross-sectional rather than longitudinal design, and limited variance among three categories of NDC ambition, our findings are striking. They show that energy subsidization is negatively correlated with ambition of nations’ pledges. If the “stickiness” of interest groups and their ability to obstruct policy change in other issue areas across nations is any indicator, domestic change can occur very slowly. The results also show that pledges do not mirror an intention of higher emitter nations to reduce them. These implications do not auger well for reaching even the modest and existing Paris Agreement emissions reductions, much less more ambitious ones. Furthermore, pressures to “rachet up” the Paris terms will seemingly not come from domestic pressure, which seems to be pressing governments in the opposite direction, towards more lax standards and relaxation of their implementation standards and regulations. The long wait for effective climate change policy may continue if the existing NDC structure is the centerpiece of that international policy. References Aden, Nate, April 05, 2016, “The Roads to Decoupling: 21 Countries Are Reducing Carbon Emissions While Growing GDP,” World Resources Institute. Retrieved from: (https://wri.org/blog/2014/11/6-graphs-explain-world’s-top-10-emitters). Accessed on June 7, 2017. Adger, W.N. “Vulnerability,” in Global Environmental Change, Vol.16, 2006, No. 3, pp. 268-281. Aldy, Joseph, E., “Evaluating Mitigation Effort: Tools and Institutions for Assessing NDCs,” Harvard Kennedy School Faculty Research Working Papers, November 2015. Retrieved from: (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=0ahUK Ewjlspe3nMDTAhXFbSYKHb90DY4QFghJMAQ&url=https%3A%2F%2Fresearch.hks .harvard.edu%2Fpublications%2FgetFile.aspx%3FId%3D1277&usg=AFQjCNFcGQgbv VbpeglR9PD4JtKs2WMPHQ&sig2=f_WR4koNdK0eb1sk0X7YfQ). Accessed on April 25, 2017. Althor, Glenn, James E. M. Watson and Richard A. Fuller, “Global mismatch between greenhouse gas emissions and the burden of climate change,” in Scientific Reports, 2016, 6, pp. 20281. Bulletin of the American Meteorological Society (BAMS), “Explaining Extreme Events,” American Meteorological Society, December 2016. Retrieved from: (https://www.ametsoc.org/ams/index.cfm/publications/bulletin-of-the-americanmeteorological-society-bams/explaining-extreme-events-from-a-climate-perspective/). Accessed on January 25, 2017. Baumgartner, F.R. and Leech, B.L. Basic Interests: The Importance of Groups in Politics and Political Science, Princeton: Princeton University Press, 1998. Bast, Elizabeth, Alex Doukas, Sam Pickard, Laurie van der Burg and Shelagh Whitley, “Empty promises: G20 subsidies to oil, gas and coal production,” Overseas Development Institute and Oil Change International. Retrieved from: (http://priceofoil.org/content/uploads/2015/11/empty_promises_full_report_update.pdf). Accessed on January 15, 2017. Beaton, C. and L. Lontoh, “Lessons learned from Indonesia’s attempts to reform fossilfuel subsidies.” 2010, October. International Institute for Sustainable Development (IISD). Retrieved from: (http://www.iisd.org/pdf/2010/lessons_indonesia_fossil_fuel_reform.pdf). Accessed on January 15, 2017. Bodle, Ralph, Lena Donat, and Matthias Duwe, “The Paris Agreement: Analysis, Assessment and Outlook,” Background paper for the workshop Beyond COP21: what does Paris mean for future climate policy?, January 28, 2016, Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safety (BMUB), Berlin. Retrieved from: (http://ecologic.eu/sites/files/event/2016/ecologic_institute_2016_paris_agreement_asses sment_0.pdf). Accessed on January 25, 2017. Boucher, O., P. Criqui, H. Benveniste, “NDCs and 2°C trajectories,” Groupe Interdisciplinaire sur les Contributions Nationales, December 1st 2015. Retrieved from: (http://www.developpement-durable.gouv.fr/IMG/pdf/Presentation_NDC_-_GICN__COP21.pdf). Accessed on January 25, 2017. Burniaux, J., & Château, J., “Greenhouse gases mitigation potential and economic efficiency of phasing-out fossil fuel subsidies,” in International Economics, Vol. 140, pp. 71–88. CAT Consortium, “Tracking (I)NDCs. Assessment of mitigation contributions to the Paris Agreement,” 2016. Retrieved from: (http://climateactiontracker.org/NDCs.html). Accessed on January 25, 2017. Chinese Department of Climate Change, NDC, 2015. Retrieved from: (http://www4.unfccc.int/submissions/NDC/Published%20Documents/China/1/China's%2 0NDC%20-%20on%2030%20June%202015.pdf=). Accessed on December 2, 2017. Coady, Ian Parry, Louis Sears, and Baoping Shang, “How Large Are Global Energy Subsidies?” IMF Working Paper, WP/15/105, May 2015. Retrieved from: (www.imf.org/external/pubs/ft/wp/2015/wp15105.pdf). Accessed on December 2, 2016. Cruz, Ana Maria and Elisabeth Krausman, “Vulnerability of the oil and gas sector to climate change and extreme weather events,” in Climatic Change, Nov 2013, Vol. 121, Issue 1, pp.: 41–53. Dür Andreas and Dirk De Bièvre, “The Question of Interest Group Influence,” in Journal of Public Policy, Vol. 27, No. 1, Jan. - Apr., 2007, pp.: 1-12 Ellis, Jennifer, “The Effects of Fossil-Fuel Subsidy Reform: A Review of Modelling and Empirical Studies,” March 1, 2010, UNEP, The Global Subsidies Initiative (GSI), IISD. Retrieved from: (http://dx.doi.org/10.2139/ssrn.1572397). Accessed on April 25, 2017. Eisenstadt, Todd A. and Karleen Jones West, “Indigenous Belief Systems, Science, and Resource Extraction: Climate Change Attitudes in Ecuador,” in Global Environmental Politics, pp. 40-58. Online publication date: 22-Dec-2016. Retrieved from: (http://www.mitpressjournals.org/doi/pdf/10.1162/GLEP_a_00389). Accessed on January 5, 2017. Ge Mengpin, et al., November 25, 2014, “6 Graphs Explain the World’s Top 10 Emitters,” World Resources Institute. Retrieved from: (https://wri.org/blog/2014/11/6graphs-explain-world’s-top-10-emitters). Accessed on June 7, 2017. Höhne, Niklas, Takeshi Kuramochi, Carsten Warnecke, Frauke Röser, Hanna Fekete, Markus Hagemann, Thomas Day, Ritika Tewari, Marie Kurdziel, Sebastian Sterl, and Sofia Gonzales, “The Paris Agreement: resolving the inconsistency between global goals and national contributions,” in Climate Policy, Vol. 17 , Issue 1, 2017, pp. 16-32. International Energy Agency (IEA). “Energy and climate change,” World Energy Outlook Special Report, 2015, Paris: IEA. Intergovernmental Panel on Climate Change (IPCC), “2013: Summary for Policymakers.” In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.). Cambridge University Press, Cambridge, United Kingdom, and New York, NY, USA. Retrieved from: (https://www.ipcc.ch/pdf/assessment-report/ar5/wg1/WG1AR5_SPM_FINAL.pdf) Accessed on February 1, 2017. International Monetary Fund, “IMF Survey: Counting the Cost of Energy Subsidies,” July 17, 2015. Retrieved from: (http://www.imf.org/external/pubs/ft/survey/so/2015/NEW070215A.htm). Accessed on December 2, 2016. Laan, Tara and Beaton, Christopher, “Strategies for Reforming Fossil-Fuel Subsidies: Practical Lessons from Ghana, France and Senegal”, IISD, GSI, April 26, 2010. Retrieved from: (https://ssrn.com/abstract=1596033). Accessed on February 2, 2017. Latvian Presidency for the Council of the European Union, “Intended Nationally Determined Contribution of the EU and its Member States,” 2015. Retrieved from: (http://www4.unfccc.int/submissions/NDC/Published%20Documents/Latvia/1/LV-03-06EU%20NDC.pdf). Accessed on December 2, 2016. Lehmann, Johannes, “A handful of carbon,” in Nature, 10 May 2007, Vol. 447, pp.: 143144. Retrieved from: (http://www.nature.com/nature/journal/v447/n7141/full/447143a.html). Accessed on February 2, 2017. LeTreut, Herve and Helene Beneviste, “Les NDC : genese et evaluation en vue d’un accord a Paris,” Groupe Interdisciplinaire sur les Contributions Nationales, October 1st 2015. Retrieved from: (http://www.economie.gouv.fr/files/igpdedeficlimat3_01102015_h-letreut_h-benveniste_ipsl-cnrs.pdf). Accessed on January 20, 2017. Lin, B., and X. Ouyang, “A revisit of fossil-fuel subsidies in China: Challenges and opportunities for energy price reform,” in Energy Conversion and Management, 2014, Vol. 82, pp. 124–134. Mabey, Nick and John Mitchell, “Investing for an Uncertain Future: Priorities for UK Energy and Climate Securitu,” Chatham House Briefing Paper, July 2010. Retrieved from: (https://archive.org/details/fe_Investing_for_an_Uncertain_FuturePriorities_for_UK_Energy_and_Climate_Security). Accessed on January 22, 2017. Mbeva, Kennedy and Pieter Pauw, “Self-Differentiation of Countries’ Responsibilities: Adressing Climate Change through Intended Nationally Determined Contributions.” Discussion paper 4/2016, Deutsches Institute fur Entwicklungspolitik/German Development Institute. Meyer, Lukas and Pranay Sanklecha (eds.), Climate Justice and Historical Emissions, Cambridge: Cambridge University Press, 2017. Najam, Adil, “The view from the South: developing countries in global environmental politics,” in Axelrod, Regina and Stacy VanDeveer (eds.). The global environment: institutions, law, and policy, Los Angeles: CQ Press/Sage, 2015. Olson, Mancur, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities, New Haven: Yale University Press, 1984. Pauw, W.P, et al. “NDC Explorer.” German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE), Bonn and African Centre for Technology Studies (ACTS), Nairobi. 2016. Retrieved from: (http://klimalog.diegdi.de/ndc/#NDCExplorer/graph?NDC??climatechangeadaptation???cat25). Accessed on February 2, 2017. Preston, Bejamin, “Local path dependence of U.S. socioeconomic exposure to climate extremes and the vulnerability commitment,” in Global Environmental Change, Volume 23, Issue 4, August 2013, pp: 719–732. Retrieved from: (http://www.sciencedirect.com/science/article/pii/S0959378013000393). Accessed on February 2, 2017. Rocha, Marcia, et al., “Paris Agreement in force, but no increase in climate action. Climate Action Tracker Update,” 10 November, 2016, CAT Consortium. Retrieved from: (http://climateactiontracker.org/assets/publications/briefing_papers/CAT_temperature_up date_November_2016.pdf). Accessed on April 18, 2017. Rogelj, J., M. den Elzen, N. Höhne, T. Fransen, H. Fekete, H. Winkler, R. Schaeffer, F. Sha, K. Riahi, M. Meinshausen, “Paris Agreement climate proposals need a boost to keep warming well below 2°C,” in Nature, 2016, 534:631–639. Sachs, Jeffrey D., Guido Schmidt-Traub, Jim Williams, Laura Segafredo, Michel Colombier, Teresa Ribera, and Henri Waisman, “Why Climate Policy needs long-term Deep Decarbonization Pathways,” Working Paper, December 2015. Retrieved from: (http://unsdsn.org/wp-content/uploads/2015/11/151130-Why-Climate-Policy-requireslong-term-Deep-Decarbonization-Pathways_rev.pdf). Accessed on January 27, 2017. Savonis, Michael, Virginia Burkett, Joanne Potter, “Impacts of climate change and variability on transportation systems and infrastructure: Gulf Coast study, phase I,” Report by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research, March 18, 2008. Victor, David, “The Politics of Fossil-Fuel Subsidies,” GSI/ IISD, Geneva, Switzerland, October 2009. Retrieved from: (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1520984). Accessed on February 2, 2017. Visser, H., Petersen, A.C. & Ligtvoet, W., “On the relation between weather-related disaster impacts, vulnerability and climate change,” in Climatic Change, August 2014, Vol. 125, Issue 6, pp.: 461-477. Retrieved from: (http://link.springer.com.proxyau.wrlc.org/article/10.1007%2Fs10584-014-1179z#Sec19). Accessed on January 29, 2017. Wiedatta, Gayan, Bingunath Ingirige, Keith Jones, David Proverb “Extreme weather events and construction SMEs: Vulnerability, impacts, and responses,” in Structural Survey Vol. 29 No. 2, 2011, pp. 106-119. Retrieved from: www.emeraldinsight.com/0263-080X.htm ]). Accessed on January 26, 2017. Wilbanks, Tom and Patricia Romero, “Industry, Settlement and Society,” in Parry, Martin, Osvaldo Canziani, Jean Palutikof, Paul van der Linden and Clair Hanson (eds.). Climate Change 2007: Impacts, Adaptation and Vulnerability, Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge and New York, 2007, pp. 357-390. Appendix Table 1. Results (inserted in text). Table 2. Descriptive statistics. Variable cat_ordinal Obs 60 Mean .8333333 Std. Dev. .5574356 0 Min Max 2 Freq. of 0 15 (25%) Freq. of 1 40 (66.67%) cat dummy death_toll gdp_losses Carbon emission intensity EU member polity Pop growth rate gdp_percapita Renewable Electricity internat_treaties pretaxsubsid_gdp posttaxsubsid_gdp 60 174 174 1 13.51 9.065 45 (75%) 1.1757 1.136763 0 0 0 15 .415546 .4543828 182 164 181 173 170 .1593407 4.075 15.8717 13734.26 .0003699 .3670031 6.185361 1.871374 19504.82 .0008823 0 -10 9.940495 251.0145 0 1 10 21.02389 103924.8 .0079404 29 (15.93%) 182 153 154 3.641758 6.636732 2.018961 .9662328 8.626551 3.79377 0 0 0 5 51.08 20.76 (25%) Freq of 2 5 (8.33%) 153 (84.07%) Table 3. Logistic regression with binary dependent variable. Marginal effects. Logit Model CAT dummy Energy Subsidies % of GDP -1.016 .3*** -.008 .173 4.415 1.24*** -.06 .013*** .053 .172 -.0000175 5.11e-06** -0.23 -0.12 -0.14 -0.03*** EU membership omitted Renewable Electricity .03 .02 Carbon emission intensity Death toll GDP losses Regime type Population growth Gdp per capita Treaties index n= 29 r2= .58 The binary dependent variable is 0= inadequate, and 1=medium/sufficient).