Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Which Countries Send More Delegates to Climate Change Conferences? Analysis of UNFCCC COPs, 1995–2015

Which Countries Send More Delegates to Climate Change Conferences? Analysis of UNFCCC COPs,... Abstract The size of national delegations at the most critical intergovernmental climate change conferences—the annual gatherings of the Conference of the Parties (COPs) of the United Nations Framework Convention on Climate Change—vary greatly. The literature has emphasized the importance of national delegation size (NDS) for states’ formal and informal participation in climate change negotiations. To our knowledge, however, this is the first paper to comprehensively examine the determinants of NDS from 1995–2015. The findings highlight a country's resources and its interest in the mitigation of fossil fuel emissions as important determinants of its NDS. In contrast, the evidence for a connection between vulnerability to climate change and NDS is limited. Interest group politics appear more important than civil society or bureaucratic influence in determining NDS. In terms of policy implications, the distance between the country and the COP location is a robust deterrent of larger delegations, and there is a nonlinear relationship between NDS and financial capacity. Further, there are differences across Annex I and non-Annex I countries. Video Abstract Video Abstract Close The annual gatherings of the United Nations Framework Convention on Climate Change's (UNFCCC) Conference of the Parties (COP) provide the most important platform for multilateral negotiations and decisions on global climate change policy.1 COP observers have noted some countries attend these negotiations with hundreds of delegates, while others have tiny delegations. For instance, Eritrea and Trinidad and Tobago each sent only three delegates to the 2015 COP in Paris compared to Canada's 287 delegates. Countries also significantly alter their national delegation size (NDS) across different COPs. For example, the Chinese delegation ranged from 15 to 53 delegates at the first fourteen COPs but from 88 to 334 at the 2009–2015 COPs. NDS, which remains at the purview of member states,2 has been shown to significantly affect states’ voice at COPs, as countries with larger delegations can better participate in numerous simultaneous negotiations and have greater capacity to prepare and advocate for their positions (e.g., Yamin and Depledge 2004; Söderberg and Andersen 2008; Bailer 2011; Schroeder, Boyloff, and Spiers 2012). Why do some states have relatively larger delegations? Extant research underscores the importance of NDS, but it leaves the motivators of NDS underanalyzed.3 Surely, richer countries can send more delegates. But, after controlling for financial capacity, what factors matter? To what extent do pro-emissions interest group politics or the country's vulnerability to climate change predict its NDS? The answers to these questions comprise important pieces in explaining crucial outcomes and dynamics in climate change negotiations. For instance, if pro-emissions interest group politics boost delegation sizes, this helps explain the lack of more ambitious outcomes on mitigation. Unpacking the determinants of NDS also informs any reforms being contemplated to UNFCCC processes. Based on a dataset we have compiled of the sizes of all states’ delegations to the 1995–2015 COPs,4 we analyze which factors significantly affect NDS and explore the underlying mechanisms. Figure 1 shows that the average and median NDS have both increased over time, with the average consistently larger than the median, suggesting some countries are overrepresented. On average, Annex 1 countries send significantly more delegates than non-Annex 1 countries. Using the data underlying figure 1, this paper explores the variation in NDS. The data can be otherwise used—for instance, to track the shifts in specific countries’ (or groups of countries’) delegation sizes or to explore linkages between NDS and negotiation strategies over time.5 Figure 1. Open in new tabDownload slide Mean and median delegates by year and category type Figure 1. Open in new tabDownload slide Mean and median delegates by year and category type What are the main determinants of NDS? In addition to a country's general resources, the findings highlight the strength of the country's stake in the mitigation of fossil fuel emissions as the most important factor influencing NDS, suggesting the role of pro-emissions interest group politics. In contrast, the evidence for a connection between vulnerability to climate change and NDS is limited. These findings are qualitatively and quantitatively significant: they suggest that countries that need to undertake significant mitigation have a larger presence at the very negotiations that are supposed to deliver significant outcomes for curbing climate change. Some of our findings also inform potential policy interventions. There is a nonlinear relationship between NDS and financial capacity—money matters only up to a point. There is also strong evidence for the discouraging role played by distance between the attending country and the COP location. The paper advances analysis on a crucial yet underexamined issue. Its extensive data collection and analysis could inform future papers, and its conceptual discussions could be adapted to the context of other intergovernmental conferences, such as those on the Nuclear Non-Proliferation Treaty. After emphasizing the importance of NDS at COPs, we turn to developing a framework for the paper's empirical analysis. The conclusion discusses the findings’ policy relevance. Importance of Delegation Size at COPs Because the COP is the UNFCCC's “supreme body,” its yearly intergovernmental negotiations among 197 member states produce the institution's most critical decisions. These negotiations span two weeks and comprise “dozens of documents, a plethora of negotiating forums, late nights, and an array of activities on the side by NGOs, IGOs, and governments” (Yamin and Depledge 2004, 431; Dimitrov 2010; Gaventa 2010; Schroeder 2010; Rietig 2014). Even though the UNFCCC formally bestows a single vote on each country and relies on consensus decision-making, a country's NDS affects its ability to participate in formal and informal negotiations and side events at COPs for several reasons. The intensity and simultaneity of COP negotiations and events mean countries with more delegates can better prepare and advocate for their positions. Dubbed as “negotiation by exhaustion,” COPs include multiple, concurrent technical negotiations, often overnight (Ashton and Wang 2003; Yamin and Depledge 2004; Schroeder, Boyloff, and Spiers 2012, 835). Most of the key negotiations, including the manner in which parties’ proposals are distributed, occur in a time-crunch, which a larger NDS alleviates (Yamin and Depledge 2004). A larger delegation is also an asset in informal negotiations. The background work of the formal plenary takes place at working groups, which delegate their tasks to several informal working groups, resulting in a flurry of concurrent informal negotiations, informal consultations, and “contact groups” (Yamin and Depledge 2004; Nasiritousi and Linnér 2016). Further, COPs include an increasing number of “side events,” which provide opportunities for delegates to interact with one another and with non-state actors (Schroeder and Lovell 2012). These events can facilitate capacity-building and policy diffusion (Hjerpe and Linnér 2010; Rietig 2014). But having a “limited number of delegates curtails the possibility of attending all relevant parallel meetings and of taking part in all the side events” (Michealowa and Michealowa 2012, 586). Thus, countries that attend COPs with sizeable delegations can better network during the conference, prepare relatively well for the myriad negotiations and side events, and advocate more effectively for their positions. “Large delegations that have the ability to be many places at once wield tremendous agenda-setting power” (Roberts and Parks 2006, 17). Indeed, based on the Copenhagen COP, Bailer (2011) shows NDS statistically affects both the country's resources during the negotiations and the activities in which it participates. Hence asymmetries in NDS have raised questions regarding whether the poor and vulnerable nations with relatively small delegations have adequate voice (e.g., Roberts and Parks 2006; Söderberg and Andersen 2008; Schroeder, Boyloff, and Spiers 2012). Although NDS significantly affects representation, there is not a one-to-one correspondence between the size of the delegation and the country's voice. Small countries have multiple methods of overcoming the constraints delegation size imposes, including grouping together, as in the G77. Regardless, the size of individual delegations matters. For example, as the chief climate change negotiator for Tuvalu, an AOSIS country, explains, “An individual country can bring forward its views more strongly as an individual country. When it comes down to negotiations in the last few days then individual countries are not really considered” (Block 2010). Hence group representation, which can compensate for a country's small delegation size, is a highly imperfect substitute for individual voice, which reinforces the importance of analyzing NDS. Similarly, an examination of Least Developed Countries’ (LDC) participation in COPs notes: “Delegations from larger countries or blocs can often send a representative to each thematic session, something the LDCs can rarely do. As a result, they find it difficult to collect and share information, which can limit the group's impact on negotiations” (Monzani 2018). Not only can small NDS hinder individual participation, but individual delegation sizes can also affect group effectiveness. Overall, NDS indicates power and voice at COP negotiations (Bailer 2011; Weiler 2012). Conceptualizing the Determinants and Mechanisms of Delegation Sizes Why do some countries send relatively larger delegations to climate change conferences? Given the lack of an existing theory of delegation size, we commence the development of a conceptual framework. There are six reasons that could propel larger delegation sizes. While these six motivators of NDS are likely interrelated, they emphasize distinct rationales for wanting a large presence at climate change negotiations. The first four motivators discussed are non-issue-specific—that is, not particularly related to the environment and climate change—whereas the latter two are. First, evidently, the more resources, financial and otherwise, a country has, the greater its capacity to form and send a large delegation. Second, the general nature of country's domestic institutions could be determining NDS. For example, participation in international organizations is positively correlated with the country's level of democracy (Mansfield and Pevehouse 2006). Specific to environmental politics, Neumayer (2002) and Battig and Bernauer (2009) find democracies more ambitious on climate change policy. Böhmelt (2013) and Kruse (2014) find democracy to be significant in explaining, respectively, nonstate actors’ and women's participation in COPs. Regulatory quality, which indicates administrative quality, might also be important because sending a large COP delegation is an administrative effort. Regulatory capacity is highly correlated with the rule of law, low levels of corruption, and government effectiveness in the country (Wheeler 2011) but nonetheless indicates a different attribute than broad regime type. Third, the country's general level of visibility in international politics could be enhancing NDS. A country more active in international diplomacy is more likely to send a relatively larger delegation to COPs due to a general inclination toward diplomatic visibility. Fourth, nondiplomatic considerations may predict the country's NDS. The attractiveness of the locale of the negotiations translating into a larger NDS would be suggestive of officials using COP attendance as a perk of their office. Alternatively, logistical considerations may discourage NDS. Fifth and issue-specific, the country's existing level of international green political activity could be motivating a larger NDS. Countries that are already relatively highly attuned to global environmental governance may be naturally more inclined to send large delegations to COPs. Put differently, participation in other institutions of the climate change “regime complex” could also help predict NDS (Keohane and Victor 2011). Sixth, again issue-specific, the country's stake in climate change negotiations could be propelling NDS. A country that has more to gain or more to lose from these negotiations would want to send a large delegation with a view to influencing negotiation outcomes. Conceptually there is no reason to consider any of these six motivators to be inherently more important than others, especially in a diverse sample of countries. Yet, in the ensuing analysis, the non-issue-specific variables, which capture general country characteristics/behavior, constitute controls. This helps highlight the extent to which regime-specific factors explain the variation in NDS. These regime-specific variables are, in turn, significant pieces in linking NDS to outcomes in COPs. Indeed, we expect issue-specific factors to matter for NDS even after accounting for non-issue-specific considerations (hypothesis 1). Moreover, a focus on issue-specific factors allows for the examination of three prominent, plausible mechanisms: pro-emission interest group politics, civil society influence, and green bureaucratic politics. Regarding the country's stake in climate change negotiations, strong groups that stand to lose significantly could be boosting NDS by lobbying the government and demanding a voice.6 Indeed, the broad literature on climate change negotiations emphasizes that countries with higher costs of policy change take more recalcitrant positions in emissions reductions largely due to strong (pro-emission) interest groups (e.g., Sprinz and Vaahtoranta 1994; Bailer and Weiler 2015). As Weiler (2012, 556) explains, countries with high fossil fuel emissions are relatively more “politically vulnerable” to mitigation efforts, which activates their interest groups. Hence, we expect that economic interests at stake boost NDS through interest group politics (hypothesis 2). Yet, at the same time, more vulnerable countries also stand to gain or lose relatively more from COP negotiations, again increasing the country's stake in the negotiations. Particularly, we expect the relationship between vulnerability and NDS to exist for relatively more democratic countries, pointing to a civil society effect (hypothesis 3). Poor, vulnerable countries are generally underrepresented at COPs, given resource constraints (e.g., Söderberg and Andersen 2008; Schroeder, Boyloff, and Spiers 2012). Nonetheless, in relatively democratic countries, vulnerability concerns could overcome financial and other resource impediments given the societal salience of climate change. A civil society effect could happen either through direct pressure from nongovernmental and nonbusiness actors for representation at COPs or by virtue of governmental sensitivity to civic attention in the absence of direct pressure.7 It is not just through lobbying or civil society influence that a country's stake in climate change negotiations could be boosting its NDS—a bureaucratic mechanism is also plausible. Existing work finds that the presence of an environmental ministry affects environmental outcomes, such as the level of emitted pollutants (Aklin and Urpelainen 2014). This broadly suggests domestic environmental institutions may play a role in NDS as well. Plausibly, we expect a concerned bureaucracy to bolster NDS when the negotiations are of particular importance due both to vulnerability and pro-emissions interests (hypothesis 4). In the case of green international political activity, it is not possible to pinpoint a mechanism. A country with a relatively high level of international environmental activity likely has a greater NDS due to a myriad of factors that explain its higher levels of activity in the first place. Similarly, for the four non-issue-specific rationale, mechanisms seem not only multifaceted but also impossible to pin down. For instance, ascertaining the reasons for a positive correlation between NDS and democracy seems elusive. The more interesting questions there concern thresholds, such as the level beyond which the relationship between NDS and resources changes. Figure 2 schematically presents these discussions. Figure 2. Open in new tabDownload slide Rationales and mechanisms for a larger NDS Figure 2. Open in new tabDownload slide Rationales and mechanisms for a larger NDS Research Design Based on our dataset of all states’ delegation sizes to 1995–2015 COPs, the dependent variable is the log of the country's NDS to each COP, namely COPDelSize.8 The estimation model is a mixed effect model with fixed effects for time and a random intercept that varies by country. Appendix B further explains the model and our multiple imputation method for addressing missing data. It also discusses the alternative fixed effects and tobit models. As already noted, the two issue-specific rationale form the basis of our explanatory variables, and the four non-issue-specific factors act as controls, given their generic importance. To assess resources, we use the size of the country's population and its GDP per capita. GDP per capita captures financial resources. Population proxies human resources available for the delegation. We include the quadratic version of these variables to asses any nonlinear relationships. To examine domestic institutions, we use the relationship between NDS and the country's level of democracy, its regulatory capacity, and the ideology of the incumbent. The importance of democracy and regulatory capacity has already been noted. Including the ideology of the executive accounts for cases in which the official stance toward environmental politics varies significantly across political parties (e.g., US Republicans versus Democrats). To assess the country's general visibility in international politics, we include: whether the country sits on the United Nations Security Council (UNSC) and/or the World Bank's Executive Board (WBEB), G20 membership, and the number of international organizations to which the country belongs. Existing works find a country's temporary UNSC membership to boost that country's visibility significantly enough to affect its receipt of aid (Kuziemko and Werker 2006; Dreher, Sturm, and Vreeland 2009). WBEB may be relevant, since the World Bank is one of the largest environmental donors, though this variable does not pertain only to the environmental realm given the World Bank's broad development mission. The G20 includes rich and emerging economies of systemic political-economic significance. The more international organizations a country has joined, the more active it plausibly is in international relations. To explore nondiplomatic considerations, we include: the COP location's tourist arrivals (Neeff 2013) and the distance between the COP location and the country's capital city.9 To assess the country's level of green international political activity, we examine: whether the country is leading a constituency on the board of the Global Environmental Facility (GEF), which has been the UNFCCC's main financing mechanism during the examined period;10 the amount of bilateral aid the country dispenses annually; the amount of donations the country has made to GEF yearly to finance the organization; the number of environmental treaties the country has signed; the amount of bilateral environmental aid the country annually receives; the amount of annual multilateral aid from GEF to the country; and whether the country is a participant in the Clean Development Mechanism (CDM), either as an Annex I or a non-Annex I country. All these variables provide good proxies for understanding the country's level of activity in various facets of the regime complex. To assess the stakes a country has in climate change negotiations, we examine, first, the economic interests at stake by including the country's level of CO2 emissions, since higher levels of fossil fuel emission indicate greater sensitivity to mitigation efforts. Likewise, we include the country's membership in OPEC. Second, to assess stakes due to environmental vulnerability, we include: the country's Biodiversity Index; the change in its food production; its membership in AOSIS, which the UN recognizes to be at particularly high-risk due to climate change; the number of people affected by natural disasters in the country in the previous year; and the level of bilateral and GEF aid the country receives annually. Although the aid variables are not clean measurements of vulnerability, they are nonetheless good proxies, given that more vulnerable countries receive more aid (Hicks et al. 2010). To analyze the mechanisms of interest group pressure and civil society effect, we focus on the country's level of democracy (polity2). A positive interaction of polity2 and the country's economic interest plausibly suggests pro-emission interest group pressure enhancing NDS. This intuitive interpretation is also supported by the aforementioned literature (e.g., Weiler 2012). In contrast, when polity2 is interacted with vulnerability, positive values on the interaction variables reasonably suggest either direct civil society involvement or general governmental responsiveness to civil society created by the relatively democratic environment, increasing NDS. To analyze the bureaucratic politics mechanism, we use two variables. One variable indicates whether the country has an environmental ministry (Aklin and Urpelainen 2014). The other measures the strength of the country's domestic environmental institutions based on the International Development Association's Country Policy and Institutional Assessment (CPIA)’s “policies and institutions for environmental sustainability.” The CPIA variable, to our knowledge, provides the most extensive cross-country assessment of domestic environmental policy-making process and its effectiveness. Appendix A includes full descriptions of variables and their sources. Results Table 1 column (1) analyzes all six rationale together, though examining each of the categories separately in their own regressions does not alter the results. Column (1) suggests an inverted U-shaped relationship between GDPpercapita and COPDelSize and a U-shaped relationship between population and COPDelSize. Financial resources beyond a certain point do not increase NDS, and population up to a certain point does not boost COPDelSize. We further explore the effects of financial resources below. The country's regulatory quality and its level of democracy also significantly and positively relate to COPDelSize. Moreover, a thousand-mile increase between the COP location and the country's capital decreases NDS by about 3.1 percent, all else equal (p < 0.01). General visibility in international relations (IntOrg and UNSC) appears irrelevant, though WBEB drives up NDS. Out of the issue-specific factors, the economic interests at stake, CO2percapita and OPEC, seemingly matter the most for NDS (hypothesis 1). All else equal, a 1 percent increase in the country's CO2percapita boosts NDS by approximately 6.7 percent. There is no evidence vulnerability or the level of green international politics affect NDS, with the exception of the slightly significant finding on EnvTreaties. Here and elsewhere, the incumbent's ideology and the attractiveness of the COP location as a tourist destination are insignificant, so the coefficients for these two variables are not reported to save space. These primary findings withstand several robustness checks, which appendix B presents. Table 1. Determinants of COP delegation size, full sample (N = 4011) DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 Notes: All models include random country and fixed year effects and also control for tourism and executive ideology. ***p < 0.01, **p < 0.05, *p < 0.1; standard errors in parentheses. Open in new tab Table 1. Determinants of COP delegation size, full sample (N = 4011) DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 Notes: All models include random country and fixed year effects and also control for tourism and executive ideology. ***p < 0.01, **p < 0.05, *p < 0.1; standard errors in parentheses. Open in new tab Column (2) provides strong evidence that the strength of pro-emission interest groups are important drivers of COPDelSize. The interaction variable, polity2 with CO2percapita, is highly significant, suggesting a mechanism whereby groups with strong economic stakes in climate change increase NDS (hypothesis 2). The result on OPEC mirrors this finding. The other results from column (1) carry over.11 Column (3) examines whether the country's level of vulnerability translates into a larger NDS through a civil society mechanism by interacting all vulnerability variables with polity2. Despite significant findings, the interpretation of these results do not suggest this mechanism, with one exception. The finding on GEFAid, considering the effects of the interaction variable, suggests that the influence of GEFAid on NDS is lower for relatively more democratic countries. Similarly, the negative sign on the interaction of AOSIS with polity2 counters a civil society mechanism. The finding on biodiversity, however, suggests that only at relatively higher levels of democracy does the country's vulnerability relate to COPDelSize positively. The most plausible interpretation here is either a direct or indirect civil society effect (hypothesis 3). While not shown, we find no evidence for the bureaucracy mechanism (hypothesis 4). Although we find that having a higher CPIA score lowers the effects of GEFAid on COPDelSize, this mirrors the finding on polity2 and GEFAid. Moreover, we have no significant findings on any of the variables associated with green international political activity (not shown); interacting polity2 or the bureaucracy variables with any of the variables in this category produces no significant results. Subsample Analysis Table 2 splits the sample into Annex I and non-Annex I countries, given that UNFCCC processes, prominently the Kyoto Protocol, see these countries as having “common but differentiated” responsibilities, with most Annex I countries having agreed to mitigation targets within the protocol. Table 2. Subsample analysis . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 See table 1 notes. We control for all variables in table 1 but show only the significant results. Open in new tab Table 2. Subsample analysis . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 See table 1 notes. We control for all variables in table 1 but show only the significant results. Open in new tab In column (1) on Annex I countries, polity2 loses significance. Since the mean polity2 score for these countries is about 8.7 (max = 10), in contrast to 1.9 for the non-Annex I sample, the nonsignificant finding on polity2 makes sense—we are analyzing a sample of relatively democratic countries. For these countries, the size of the population seems to be divorced from NDS, though distance remains a discouraging factor. Importantly, CO2 emissions not only retains its high significance, but the coefficient on it is more than twice the coefficient in table 1 column (1), reinforcing the interest group mechanism. All else equal, a 1 percent increase in an Annex I country's CO2percapita boosts its NDS by about 15.7 percent. None of the three mechanisms are significant for any of the issue-specific variables in this subsample (not shown). Overall, financial resources (GDPpercapita) and interests seem to be the main drivers of NDS for Annex I countries. The rest of table 2 examines non-Annex I countries. Column (2) shows that in addition to those with high financial resources, non-Annex I countries that are relatively more democratic and have higher regulatory capacity have larger delegation sizes. This said, the effects of GDPpercapita and GDPpercapitasqaured differ across the two sub-samples, as shown in figure 3. Given their higher levels of average income, the peak for Annex I countries occurs at about USD 10,000 and for non-Annex I at around USD 2,100. Moreover, the highly significant result on OPEC provides quantitative evidence for the anecdotally well-known vocal participation of oil-producing countries in climate change negotiations. Figure 3. Open in new tabDownload slide Relationship of delegate size to GDP per capita by country type Figure 3. Open in new tabDownload slide Relationship of delegate size to GDP per capita by country type Column (3) interacts polity2 with the economic interest variables. The significant result on OPEC*polity2 is suggestive of governmental susceptibility to strong economic interests. There also appears to be a bureaucratic mechanism at work—column (4) shows the interaction variable between OPEC and CPIA to be significant. Reasonably, the relevant bureaucracy not only knows mitigation negotiations are key for OPEC countries but is also aware of its own role in implementing any decision. Further, we interact polity2 and the bureaucracy variables with all the variables associated with vulnerability and the level of green international political activity. As column (5) shows, some of the polity2 and vulnerability interactions are significant, but these variables do not strongly suggest a civil society mechanism. For example, for countries with higher levels of democracy, GEFAid has less of an impact on NDS compared to countries with relatively lower levels of democracy. Similarly, AOSIS membership has a greater impact on NDS for the relatively less democratic countries. However, column (5) shows that biodiversity influences NDS more for the relatively more democratic countries, potentially suggesting a civil society mechanism. Both columns also suggest the more democratic the country, the greater the effects of the level of bilateral environmental aid on NDS. In sum, there is weak evidence of vulnerability translating into a larger NDS through a civil society mechanism for the non-Annex I sample. Conclusions This paper presents the first extensive analysis of the determinants of UNFCCC members’ delegation sizes to the most important climate change conferences, the annual gatherings of the Conference of the Parties. It finds good evidence for the strength of domestic groups motivating larger delegation sizes, though evidence for the country's environmental vulnerability being positively related to its NDS is relatively weak. To the extent that pro-emission interest group politics dominate vulnerability considerations in explaining NDS, COP negotiations could be considered distorted in favor of those resisting ambitious mitigation efforts. Similarly, a country's existing level of activity in global environmental governance appears as a poor predictor of its NDS. Instead, resources and interests seem to be the primary drivers of larger delegations. The findings also offer policy implications. First, given the negative impact of distance, reform discussions should re-think COP locations. Choosing locations within easy reach of smaller and poorer nations would help these countries’ attendance at COPs, which is already hampered by low levels of financial and administrative capacity. Second, the funding from the UNFCCC Trust Fund for participation should focus on the poorest countries. Figure 3 showed that the relationship between average income and NDS turns negative after about USD 2,100, which corresponds to the lower range of the “lower middle income” category of the World Bank's 2015 country classifications. Hence, funds should go to low-income countries and the lowest income ones within the middle-income range. Third, given that regulatory quality significantly predicts NDS for non-Annex I countries, further attention to boosting the administrative capacity of smaller and poorer countries in COP negotiations is worthwhile. At the very least, existing efforts, such as the tradition of nongovernmental institutions providing support to developing countries (for example, the International Institute for Environment and Development supporting AOSIS countries) could be bolstered. Future research could investigate how NDS influences interstate interactions at other environmental platforms or link delegation size to outcomes at COPs. The determinants of delegation sizes and the variance in countries’ attendance at COPs are important to consider in discussions of reforms to the UNFCCC process. Reforms may ultimately hinge upon principles (Whose attendance should be boosted? Is limiting national delegation size at all desirable?), but analysis of the determinants of, and patterns in, national delegation sizes are critical to these policy debates. Footnotes 1 Herein, COPs. 2 UNFCCC Rules of Procedure (FCCC/CP/1996/2, Rule 17). 3 Neeff (2013) examines total attendance numbers. Schroeder et al. (2012) emphasizes increased attendance at COPs and notes that some small developing countries have downsized their delegations. They outline several plausible reasons for this but do not provide inferential statistics for these possibilities. Böhmelt (2013) analyzes nonstate actors within state delegations. Kruse (2014) examines women's attendance at COPs. Weiler (2012) explores the connection between NDS and negotiation positions at the Cancun COP; Bailer (2011) does the same for Copenhagen. 4 The UNFCCC releases only aggregate attendance numbers for each COP. We created the database of state-by-state delegation sizes from the UNFCCC's “lists of participants,” which lists each delegate by name. Böhmelt's (2013)supplementary files include delegation sizes until 2004 but does not analyze NDS as a variable. Our numbers resemble his. Most explanatory variables are not available beyond 2015. 5 Extant work does this for select COPs (e.g., Bailer 2011; Weiler 2012). 6 National delegations contain both state and nonstate (including business) actors. 7 Differentiating between direct and indirect pathways is not tenable here, and, regardless, both influences could simultaneously be at work. 8 We add one to each NDS to be able to consistently analyze countries that sent no delegates to some COPs but participated in others. We log some variables to dilute the effects of outliers. 9 In principle, COP venues rotate between the UN's five official regions; in practice, countries volunteer to host. 10 The directors on the GEF Council come from thirty-two “constituencies,” half of which belong to developing countries. The following countries have their own constituencies, hence directors: Canada, China, France, Germany, Italy, Iran, the Netherlands, the United Kingdom, and the United States. Most of these countries hold their own constituencies due to being the GEF's largest donors. In the remaining constituencies, directorship rotates. While no rule dictates this rotation, these directors tend to be from countries with relatively high voting power based on cumulative contributions to the organization (Perez del Castillo 2009; Lattanzio 2013). 11 Bolstering this interpretation, the results on the interaction of CO2percapita and OPEC with the regulatory quality variable are insignificant (not shown). Acknowledgements The authors are grateful to Katja Michaelowa, Axel Michaelowa, Thomas Bernauer, Christopher Kilby, Jennifer Peck, Alexandra Guisinger, and the many commentators at the 2016 Political Economy of International Organizations conference and the 2016 Temple Workshop on International Institutions and Global Governance for their feedback on this research. We additionally thank Axel Dreher and Christopher Kilby for sharing their data, respectively, on international organizations and World Bank Board membership. References Aklin Michael , Urpelainen Johannes . 2014 . “ The Global Spread of Environmental Ministries: Domestic-International Interactions .” International Studies Quarterly 58 ( 4 ): 764 – 80 . Google Scholar Crossref Search ADS WorldCat Ashton John , Wang Xueman . 2003 . “ Equity and Climate: In Principle and Practice .” In Beyond Kyoto: Advancing the International Effort against Climate Change , edited by Aldy J. E. et al. . , 61 – 85 . Arlington, VA : Pew Center on Global Climate Change . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bailer Stefanie . 2011 . “ Bargaining Resources and Strategies in Climate Change Negotiations .” Paper presented at the Workshop Negotiating Climate Change, Center for Comparative and International Studies (CIS), Zurich . OpenURL Placeholder Text WorldCat Bailer Stefanie , Weiler Florian . 2015 . “ A Political Economy of Positions in Climate Change Negotiations: Economic, Structural, Domestic, and Strategic Explanations .” Review of International Organizations 10 ( 1 ): 43 – 66 . Google Scholar Crossref Search ADS WorldCat Bättig Michele , Bernauer Thomas . 2009 . “ National Institutions and Global Public Goods: Are Democracies More Cooperative in Climate Change Policy? ” International Organization 63 ( 2 ): 281 – 308 . Google Scholar Crossref Search ADS WorldCat Block Ben . 2010 . “ Interview with Tuvalu Climate Negotiator Ian Fry .” WorldWatch Institute . January 13, 2010. http://www.worldwatch.org/node/6360 . OpenURL Placeholder Text WorldCat Böhmelt Tobias . 2013 . “ A Closer Look at the Information Provision Rationale: Civil Society Participation in States’ Delegations at the UNFCCC .” Review of International Organizations 8 ( 1 ): 55 – 80 . Google Scholar Crossref Search ADS WorldCat Dimitrov Radoslav S. 2010 . “ Inside UN Climate Change Negotiations: The Copenhagen Conference .” Review of Policy Research 27 ( 6 ): 795 – 821 . Google Scholar Crossref Search ADS WorldCat Dreher A. , Sturm Jan-Egbert , Vreeland J. . 2009 . “ Development Aid and International Politics: Does Membership on the UN Security Council Influence World Bank Decisions? ” Journal of Development Economics 88 : 1 – 18 . Google Scholar Crossref Search ADS WorldCat Gaventa Jonathan . 2010 . “ Environmental Mega-Conferences and Climate Governance beyond the Nation-State: A Bali Case Study .” St Antony's International Review 5 ( 2 ): 29 – 45 . OpenURL Placeholder Text WorldCat Hjerpe Mattias , Linnér Björn-Ola . 2010 . “ Functions of COP Side-Events in Climate-Change Governance .” Climate Policy 10 ( 2 ): 167 – 80 . Google Scholar Crossref Search ADS WorldCat Hicks Robert L. , Parks Bradley C. , Roberts J. Timmons , Tierney Michael J. . 2010 . Greening Aid? Understanding the Environmental Impact of Development Assistance . Oxford, UK : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Keohane Robert , David Victor . 2011 . “ The Regime Complex for Climate Change .” Perspectives on Politics 9 ( 1 ): 7 – 23 . Google Scholar Crossref Search ADS WorldCat Kruse Johannes . 2014 . “ Women's Representation in the UN Climate Change Negotiations: A Quantitative Analysis of State Delegations, 1995–2011 .” International Environmental Agreements 14 ( 4 ): 349 – 70 . Google Scholar Crossref Search ADS WorldCat Kuziemko Ilyana , Werker Eric D. . 2006 . “ How Much Is a Seat on the Security Council Worth? Foreign Aid and Bribery at the United Nations .” Journal of Political Economy 114 ( 5 ): 905 – 30 . Google Scholar Crossref Search ADS WorldCat Lattanzio Richard K. 2013 . “ International Environmental Financing: Global Environmental Facility (GEF) .” Congressional Research Service Report , R41165 . OpenURL Placeholder Text WorldCat Mansfield E. , Pevehouse J. . 2006 . “ Democratization and International Organizations .” International Organization 60 ( 1 ): 137 – 67 . Google Scholar Crossref Search ADS WorldCat Michaelowa Katharina , Michaelowa Axel . 2012 . “ Negotiating Climate Change .” Climate Policy 1295 ( 5 ): 527 – 33 . Google Scholar Crossref Search ADS WorldCat Monzani Bernardo . 2018 . “ IIED Support to the Least Developed Countries Group. Influencing Global Climate Change Negotiations .” Evaluation Case Study, IIED: London . OpenURL Placeholder Text WorldCat Nasiritousi Nagmeh , Linnér Björn-Ola . 2016 . “ Open or Closed Meetings? Explaining Nonstate Actor Involvement in the International Climate Change Negotiations .” International Environmental Agreements 16 ( 1 ): 127 – 44 . Google Scholar Crossref Search ADS WorldCat Neeff Till . 2013 . “ How Many Will Attend Paris? UNFCCC COP Participation Patterns 1995–2015 .” Environmental Science and Policy 31 : 157 – 59 . Google Scholar Crossref Search ADS WorldCat Neumayer Eric . 2002 . “ Do Democracies Exhibit Stronger International Environmental Commitment? A Cross-country Analysis .” Journal of Peace Research 39 ( 2 ): 139 – 64 . Google Scholar Crossref Search ADS WorldCat Perez del Castillo Carlos . 2009 . “ Governance of the GEF .” OPS Technical Document #4, GEF Independent Evaluation Office: Washington, DC . OpenURL Placeholder Text WorldCat Rietig Katharina . 2014 . “ Reinforcement of Multilevel Governance Dynamics: Creating Momentum for Increasing Ambitions in International Climate Negotiations .” International Environmental Agreements 14 ( 4 ): 371 – 89 . Google Scholar Crossref Search ADS WorldCat Roberts Timmons J. , Parks Bradley . 2006 . A Climate of Injustice: Global Inequality, North-South Politics, Climate Policy . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schroeder Heike . 2010 . “ Agency in International Climate Negotiations: The Case of Indigenous Peoples and Avoided Deforestation .” International Environmental Agreements: Politics, Law and Economics 10 ( 4 ): 317 – 32 . Google Scholar Crossref Search ADS WorldCat Schroeder Heike , Boyloff Maxwell T. , Spiers Laura . 2012 . “ Equity and State Representations in Climate Negotiations .” Climate Change 2 ( 12 ): 834 – 36 . OpenURL Placeholder Text WorldCat Schroeder Heike , Lovell Heather . 2012 . “ The Role of Non-Nation-State Actors and Side Events in the International Climate Negotiations .” Climate Policy 12 ( 1 ): 23 – 37 . Google Scholar Crossref Search ADS WorldCat Söderberg Mattias , Andersen Sigrid Bjerre . 2008 . “ Unbalanced Participation .” Dan Church Aid Research Paper 8, August . OpenURL Placeholder Text WorldCat Sprinz Detlef , Tapani , Vaahtoranta . 1994 . “ The Interest-Based Explanation of International Environmental Policy .” International Organization 48 ( 1 ): 77 – 105 . Google Scholar Crossref Search ADS WorldCat United Nations Conference of the Parties , Second Session . “ Framework Convention on Climate Change .” FCCC/CP/1996/2, July 8–19, 1996 . OpenURL Placeholder Text WorldCat Weiler Florian . 2012 . “ Determinants of Bargaining Success in the Climate Change Negotiations .” Climate Policy 12 ( 5 ): 552 – 74 . Google Scholar Crossref Search ADS WorldCat Wheeler David . 2011 . “ Quantifying Vulnerability to Climate Change: Implications for Adaptation Assistance .” Working Paper 240. Center for Global Development, Washington, DC . OpenURL Placeholder Text WorldCat Yamin Farhana , Depledge Joanna . 2004 . The International Climate Change Regime: A Guide to Rules, Institutions and Procedures . Cambridge : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC © The Author(s) (2020). Published by Oxford University Press on behalf of the International Studies Association. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Foreign Policy Analysis Oxford University Press

Which Countries Send More Delegates to Climate Change Conferences? Analysis of UNFCCC COPs, 1995–2015

Download PDF
14 pages

Loading next page...
 
/lp/oxford-university-press/which-countries-send-more-delegates-to-climate-change-conferences-dnWmJ0pmLz

References (37)

Publisher
Oxford University Press
Copyright
© The Author(s) (2020). Published by Oxford University Press on behalf of the International Studies Association.
ISSN
1743-8586
eISSN
1743-8594
DOI
10.1093/fpa/orz031
Publisher site
See Article on Publisher Site

Abstract

Abstract The size of national delegations at the most critical intergovernmental climate change conferences—the annual gatherings of the Conference of the Parties (COPs) of the United Nations Framework Convention on Climate Change—vary greatly. The literature has emphasized the importance of national delegation size (NDS) for states’ formal and informal participation in climate change negotiations. To our knowledge, however, this is the first paper to comprehensively examine the determinants of NDS from 1995–2015. The findings highlight a country's resources and its interest in the mitigation of fossil fuel emissions as important determinants of its NDS. In contrast, the evidence for a connection between vulnerability to climate change and NDS is limited. Interest group politics appear more important than civil society or bureaucratic influence in determining NDS. In terms of policy implications, the distance between the country and the COP location is a robust deterrent of larger delegations, and there is a nonlinear relationship between NDS and financial capacity. Further, there are differences across Annex I and non-Annex I countries. Video Abstract Video Abstract Close The annual gatherings of the United Nations Framework Convention on Climate Change's (UNFCCC) Conference of the Parties (COP) provide the most important platform for multilateral negotiations and decisions on global climate change policy.1 COP observers have noted some countries attend these negotiations with hundreds of delegates, while others have tiny delegations. For instance, Eritrea and Trinidad and Tobago each sent only three delegates to the 2015 COP in Paris compared to Canada's 287 delegates. Countries also significantly alter their national delegation size (NDS) across different COPs. For example, the Chinese delegation ranged from 15 to 53 delegates at the first fourteen COPs but from 88 to 334 at the 2009–2015 COPs. NDS, which remains at the purview of member states,2 has been shown to significantly affect states’ voice at COPs, as countries with larger delegations can better participate in numerous simultaneous negotiations and have greater capacity to prepare and advocate for their positions (e.g., Yamin and Depledge 2004; Söderberg and Andersen 2008; Bailer 2011; Schroeder, Boyloff, and Spiers 2012). Why do some states have relatively larger delegations? Extant research underscores the importance of NDS, but it leaves the motivators of NDS underanalyzed.3 Surely, richer countries can send more delegates. But, after controlling for financial capacity, what factors matter? To what extent do pro-emissions interest group politics or the country's vulnerability to climate change predict its NDS? The answers to these questions comprise important pieces in explaining crucial outcomes and dynamics in climate change negotiations. For instance, if pro-emissions interest group politics boost delegation sizes, this helps explain the lack of more ambitious outcomes on mitigation. Unpacking the determinants of NDS also informs any reforms being contemplated to UNFCCC processes. Based on a dataset we have compiled of the sizes of all states’ delegations to the 1995–2015 COPs,4 we analyze which factors significantly affect NDS and explore the underlying mechanisms. Figure 1 shows that the average and median NDS have both increased over time, with the average consistently larger than the median, suggesting some countries are overrepresented. On average, Annex 1 countries send significantly more delegates than non-Annex 1 countries. Using the data underlying figure 1, this paper explores the variation in NDS. The data can be otherwise used—for instance, to track the shifts in specific countries’ (or groups of countries’) delegation sizes or to explore linkages between NDS and negotiation strategies over time.5 Figure 1. Open in new tabDownload slide Mean and median delegates by year and category type Figure 1. Open in new tabDownload slide Mean and median delegates by year and category type What are the main determinants of NDS? In addition to a country's general resources, the findings highlight the strength of the country's stake in the mitigation of fossil fuel emissions as the most important factor influencing NDS, suggesting the role of pro-emissions interest group politics. In contrast, the evidence for a connection between vulnerability to climate change and NDS is limited. These findings are qualitatively and quantitatively significant: they suggest that countries that need to undertake significant mitigation have a larger presence at the very negotiations that are supposed to deliver significant outcomes for curbing climate change. Some of our findings also inform potential policy interventions. There is a nonlinear relationship between NDS and financial capacity—money matters only up to a point. There is also strong evidence for the discouraging role played by distance between the attending country and the COP location. The paper advances analysis on a crucial yet underexamined issue. Its extensive data collection and analysis could inform future papers, and its conceptual discussions could be adapted to the context of other intergovernmental conferences, such as those on the Nuclear Non-Proliferation Treaty. After emphasizing the importance of NDS at COPs, we turn to developing a framework for the paper's empirical analysis. The conclusion discusses the findings’ policy relevance. Importance of Delegation Size at COPs Because the COP is the UNFCCC's “supreme body,” its yearly intergovernmental negotiations among 197 member states produce the institution's most critical decisions. These negotiations span two weeks and comprise “dozens of documents, a plethora of negotiating forums, late nights, and an array of activities on the side by NGOs, IGOs, and governments” (Yamin and Depledge 2004, 431; Dimitrov 2010; Gaventa 2010; Schroeder 2010; Rietig 2014). Even though the UNFCCC formally bestows a single vote on each country and relies on consensus decision-making, a country's NDS affects its ability to participate in formal and informal negotiations and side events at COPs for several reasons. The intensity and simultaneity of COP negotiations and events mean countries with more delegates can better prepare and advocate for their positions. Dubbed as “negotiation by exhaustion,” COPs include multiple, concurrent technical negotiations, often overnight (Ashton and Wang 2003; Yamin and Depledge 2004; Schroeder, Boyloff, and Spiers 2012, 835). Most of the key negotiations, including the manner in which parties’ proposals are distributed, occur in a time-crunch, which a larger NDS alleviates (Yamin and Depledge 2004). A larger delegation is also an asset in informal negotiations. The background work of the formal plenary takes place at working groups, which delegate their tasks to several informal working groups, resulting in a flurry of concurrent informal negotiations, informal consultations, and “contact groups” (Yamin and Depledge 2004; Nasiritousi and Linnér 2016). Further, COPs include an increasing number of “side events,” which provide opportunities for delegates to interact with one another and with non-state actors (Schroeder and Lovell 2012). These events can facilitate capacity-building and policy diffusion (Hjerpe and Linnér 2010; Rietig 2014). But having a “limited number of delegates curtails the possibility of attending all relevant parallel meetings and of taking part in all the side events” (Michealowa and Michealowa 2012, 586). Thus, countries that attend COPs with sizeable delegations can better network during the conference, prepare relatively well for the myriad negotiations and side events, and advocate more effectively for their positions. “Large delegations that have the ability to be many places at once wield tremendous agenda-setting power” (Roberts and Parks 2006, 17). Indeed, based on the Copenhagen COP, Bailer (2011) shows NDS statistically affects both the country's resources during the negotiations and the activities in which it participates. Hence asymmetries in NDS have raised questions regarding whether the poor and vulnerable nations with relatively small delegations have adequate voice (e.g., Roberts and Parks 2006; Söderberg and Andersen 2008; Schroeder, Boyloff, and Spiers 2012). Although NDS significantly affects representation, there is not a one-to-one correspondence between the size of the delegation and the country's voice. Small countries have multiple methods of overcoming the constraints delegation size imposes, including grouping together, as in the G77. Regardless, the size of individual delegations matters. For example, as the chief climate change negotiator for Tuvalu, an AOSIS country, explains, “An individual country can bring forward its views more strongly as an individual country. When it comes down to negotiations in the last few days then individual countries are not really considered” (Block 2010). Hence group representation, which can compensate for a country's small delegation size, is a highly imperfect substitute for individual voice, which reinforces the importance of analyzing NDS. Similarly, an examination of Least Developed Countries’ (LDC) participation in COPs notes: “Delegations from larger countries or blocs can often send a representative to each thematic session, something the LDCs can rarely do. As a result, they find it difficult to collect and share information, which can limit the group's impact on negotiations” (Monzani 2018). Not only can small NDS hinder individual participation, but individual delegation sizes can also affect group effectiveness. Overall, NDS indicates power and voice at COP negotiations (Bailer 2011; Weiler 2012). Conceptualizing the Determinants and Mechanisms of Delegation Sizes Why do some countries send relatively larger delegations to climate change conferences? Given the lack of an existing theory of delegation size, we commence the development of a conceptual framework. There are six reasons that could propel larger delegation sizes. While these six motivators of NDS are likely interrelated, they emphasize distinct rationales for wanting a large presence at climate change negotiations. The first four motivators discussed are non-issue-specific—that is, not particularly related to the environment and climate change—whereas the latter two are. First, evidently, the more resources, financial and otherwise, a country has, the greater its capacity to form and send a large delegation. Second, the general nature of country's domestic institutions could be determining NDS. For example, participation in international organizations is positively correlated with the country's level of democracy (Mansfield and Pevehouse 2006). Specific to environmental politics, Neumayer (2002) and Battig and Bernauer (2009) find democracies more ambitious on climate change policy. Böhmelt (2013) and Kruse (2014) find democracy to be significant in explaining, respectively, nonstate actors’ and women's participation in COPs. Regulatory quality, which indicates administrative quality, might also be important because sending a large COP delegation is an administrative effort. Regulatory capacity is highly correlated with the rule of law, low levels of corruption, and government effectiveness in the country (Wheeler 2011) but nonetheless indicates a different attribute than broad regime type. Third, the country's general level of visibility in international politics could be enhancing NDS. A country more active in international diplomacy is more likely to send a relatively larger delegation to COPs due to a general inclination toward diplomatic visibility. Fourth, nondiplomatic considerations may predict the country's NDS. The attractiveness of the locale of the negotiations translating into a larger NDS would be suggestive of officials using COP attendance as a perk of their office. Alternatively, logistical considerations may discourage NDS. Fifth and issue-specific, the country's existing level of international green political activity could be motivating a larger NDS. Countries that are already relatively highly attuned to global environmental governance may be naturally more inclined to send large delegations to COPs. Put differently, participation in other institutions of the climate change “regime complex” could also help predict NDS (Keohane and Victor 2011). Sixth, again issue-specific, the country's stake in climate change negotiations could be propelling NDS. A country that has more to gain or more to lose from these negotiations would want to send a large delegation with a view to influencing negotiation outcomes. Conceptually there is no reason to consider any of these six motivators to be inherently more important than others, especially in a diverse sample of countries. Yet, in the ensuing analysis, the non-issue-specific variables, which capture general country characteristics/behavior, constitute controls. This helps highlight the extent to which regime-specific factors explain the variation in NDS. These regime-specific variables are, in turn, significant pieces in linking NDS to outcomes in COPs. Indeed, we expect issue-specific factors to matter for NDS even after accounting for non-issue-specific considerations (hypothesis 1). Moreover, a focus on issue-specific factors allows for the examination of three prominent, plausible mechanisms: pro-emission interest group politics, civil society influence, and green bureaucratic politics. Regarding the country's stake in climate change negotiations, strong groups that stand to lose significantly could be boosting NDS by lobbying the government and demanding a voice.6 Indeed, the broad literature on climate change negotiations emphasizes that countries with higher costs of policy change take more recalcitrant positions in emissions reductions largely due to strong (pro-emission) interest groups (e.g., Sprinz and Vaahtoranta 1994; Bailer and Weiler 2015). As Weiler (2012, 556) explains, countries with high fossil fuel emissions are relatively more “politically vulnerable” to mitigation efforts, which activates their interest groups. Hence, we expect that economic interests at stake boost NDS through interest group politics (hypothesis 2). Yet, at the same time, more vulnerable countries also stand to gain or lose relatively more from COP negotiations, again increasing the country's stake in the negotiations. Particularly, we expect the relationship between vulnerability and NDS to exist for relatively more democratic countries, pointing to a civil society effect (hypothesis 3). Poor, vulnerable countries are generally underrepresented at COPs, given resource constraints (e.g., Söderberg and Andersen 2008; Schroeder, Boyloff, and Spiers 2012). Nonetheless, in relatively democratic countries, vulnerability concerns could overcome financial and other resource impediments given the societal salience of climate change. A civil society effect could happen either through direct pressure from nongovernmental and nonbusiness actors for representation at COPs or by virtue of governmental sensitivity to civic attention in the absence of direct pressure.7 It is not just through lobbying or civil society influence that a country's stake in climate change negotiations could be boosting its NDS—a bureaucratic mechanism is also plausible. Existing work finds that the presence of an environmental ministry affects environmental outcomes, such as the level of emitted pollutants (Aklin and Urpelainen 2014). This broadly suggests domestic environmental institutions may play a role in NDS as well. Plausibly, we expect a concerned bureaucracy to bolster NDS when the negotiations are of particular importance due both to vulnerability and pro-emissions interests (hypothesis 4). In the case of green international political activity, it is not possible to pinpoint a mechanism. A country with a relatively high level of international environmental activity likely has a greater NDS due to a myriad of factors that explain its higher levels of activity in the first place. Similarly, for the four non-issue-specific rationale, mechanisms seem not only multifaceted but also impossible to pin down. For instance, ascertaining the reasons for a positive correlation between NDS and democracy seems elusive. The more interesting questions there concern thresholds, such as the level beyond which the relationship between NDS and resources changes. Figure 2 schematically presents these discussions. Figure 2. Open in new tabDownload slide Rationales and mechanisms for a larger NDS Figure 2. Open in new tabDownload slide Rationales and mechanisms for a larger NDS Research Design Based on our dataset of all states’ delegation sizes to 1995–2015 COPs, the dependent variable is the log of the country's NDS to each COP, namely COPDelSize.8 The estimation model is a mixed effect model with fixed effects for time and a random intercept that varies by country. Appendix B further explains the model and our multiple imputation method for addressing missing data. It also discusses the alternative fixed effects and tobit models. As already noted, the two issue-specific rationale form the basis of our explanatory variables, and the four non-issue-specific factors act as controls, given their generic importance. To assess resources, we use the size of the country's population and its GDP per capita. GDP per capita captures financial resources. Population proxies human resources available for the delegation. We include the quadratic version of these variables to asses any nonlinear relationships. To examine domestic institutions, we use the relationship between NDS and the country's level of democracy, its regulatory capacity, and the ideology of the incumbent. The importance of democracy and regulatory capacity has already been noted. Including the ideology of the executive accounts for cases in which the official stance toward environmental politics varies significantly across political parties (e.g., US Republicans versus Democrats). To assess the country's general visibility in international politics, we include: whether the country sits on the United Nations Security Council (UNSC) and/or the World Bank's Executive Board (WBEB), G20 membership, and the number of international organizations to which the country belongs. Existing works find a country's temporary UNSC membership to boost that country's visibility significantly enough to affect its receipt of aid (Kuziemko and Werker 2006; Dreher, Sturm, and Vreeland 2009). WBEB may be relevant, since the World Bank is one of the largest environmental donors, though this variable does not pertain only to the environmental realm given the World Bank's broad development mission. The G20 includes rich and emerging economies of systemic political-economic significance. The more international organizations a country has joined, the more active it plausibly is in international relations. To explore nondiplomatic considerations, we include: the COP location's tourist arrivals (Neeff 2013) and the distance between the COP location and the country's capital city.9 To assess the country's level of green international political activity, we examine: whether the country is leading a constituency on the board of the Global Environmental Facility (GEF), which has been the UNFCCC's main financing mechanism during the examined period;10 the amount of bilateral aid the country dispenses annually; the amount of donations the country has made to GEF yearly to finance the organization; the number of environmental treaties the country has signed; the amount of bilateral environmental aid the country annually receives; the amount of annual multilateral aid from GEF to the country; and whether the country is a participant in the Clean Development Mechanism (CDM), either as an Annex I or a non-Annex I country. All these variables provide good proxies for understanding the country's level of activity in various facets of the regime complex. To assess the stakes a country has in climate change negotiations, we examine, first, the economic interests at stake by including the country's level of CO2 emissions, since higher levels of fossil fuel emission indicate greater sensitivity to mitigation efforts. Likewise, we include the country's membership in OPEC. Second, to assess stakes due to environmental vulnerability, we include: the country's Biodiversity Index; the change in its food production; its membership in AOSIS, which the UN recognizes to be at particularly high-risk due to climate change; the number of people affected by natural disasters in the country in the previous year; and the level of bilateral and GEF aid the country receives annually. Although the aid variables are not clean measurements of vulnerability, they are nonetheless good proxies, given that more vulnerable countries receive more aid (Hicks et al. 2010). To analyze the mechanisms of interest group pressure and civil society effect, we focus on the country's level of democracy (polity2). A positive interaction of polity2 and the country's economic interest plausibly suggests pro-emission interest group pressure enhancing NDS. This intuitive interpretation is also supported by the aforementioned literature (e.g., Weiler 2012). In contrast, when polity2 is interacted with vulnerability, positive values on the interaction variables reasonably suggest either direct civil society involvement or general governmental responsiveness to civil society created by the relatively democratic environment, increasing NDS. To analyze the bureaucratic politics mechanism, we use two variables. One variable indicates whether the country has an environmental ministry (Aklin and Urpelainen 2014). The other measures the strength of the country's domestic environmental institutions based on the International Development Association's Country Policy and Institutional Assessment (CPIA)’s “policies and institutions for environmental sustainability.” The CPIA variable, to our knowledge, provides the most extensive cross-country assessment of domestic environmental policy-making process and its effectiveness. Appendix A includes full descriptions of variables and their sources. Results Table 1 column (1) analyzes all six rationale together, though examining each of the categories separately in their own regressions does not alter the results. Column (1) suggests an inverted U-shaped relationship between GDPpercapita and COPDelSize and a U-shaped relationship between population and COPDelSize. Financial resources beyond a certain point do not increase NDS, and population up to a certain point does not boost COPDelSize. We further explore the effects of financial resources below. The country's regulatory quality and its level of democracy also significantly and positively relate to COPDelSize. Moreover, a thousand-mile increase between the COP location and the country's capital decreases NDS by about 3.1 percent, all else equal (p < 0.01). General visibility in international relations (IntOrg and UNSC) appears irrelevant, though WBEB drives up NDS. Out of the issue-specific factors, the economic interests at stake, CO2percapita and OPEC, seemingly matter the most for NDS (hypothesis 1). All else equal, a 1 percent increase in the country's CO2percapita boosts NDS by approximately 6.7 percent. There is no evidence vulnerability or the level of green international politics affect NDS, with the exception of the slightly significant finding on EnvTreaties. Here and elsewhere, the incumbent's ideology and the attractiveness of the COP location as a tourist destination are insignificant, so the coefficients for these two variables are not reported to save space. These primary findings withstand several robustness checks, which appendix B presents. Table 1. Determinants of COP delegation size, full sample (N = 4011) DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 Notes: All models include random country and fixed year effects and also control for tourism and executive ideology. ***p < 0.01, **p < 0.05, *p < 0.1; standard errors in parentheses. Open in new tab Table 1. Determinants of COP delegation size, full sample (N = 4011) DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 DV: COPDelSize (log) . Full model . Economic interests and interest group mechanism . Vulnerability and civil society mechanism . . (1) . (2) . (3) . GDPpercap (log) 1.540*** 1.521*** 1.508*** (0.288) (0.283) (0.294) GDPpercapsq (log) −0.185*** −0.182*** −0.182*** (0.042) (0.040) (0.042) Population (log) −1.874*** −1.791*** −1.866*** (0.424) (0.435) (0.427) PopSquared (log) 0.181*** 0.173*** 0.181*** (0.033) (0.034) (0.034) Polity2 0.012** 0.014*** 0.005 (0.005) (0.005) (0.014) RegQual 0.074* 0.074* 0.074* (0.044) (0.043) (0.045) G20 0.186 0.237 0.137 (0.178) (0.180) (0.174) UNSC −0.00001 0.003 0.003 (0.047) (0.047) (0.047) IntOrg 0.004 0.003 0.003 (0.003) (0.003) (0.003) WBEB 0.100** 0.103** 0.091** (0.046) (0.046) (0.046) Distance −0.031*** −0.031*** −0.031*** (0.003) (0.003) (0.003) GEFCouncil 0.056 0.059 0.055 (0.039) (0.039) (0.039) EnvAidDonor (log) −0.015 −0.013 −0.014 (0.021) (0.021) (0.021) GEFDonor (log) 0.008 0.008 0.009 (0.008) (0.008) (0.008) EnvTreaties 0.002* 0.002* 0.002* (0.001) (0.001) (0.001) EnvAidRecip (log) 0.017 0.017 0.007 (0.015) (0.015) (0.017) CDM Donor −0.108 −0.105 −0.121* (0.066) (0.067) (0.067) CDM Host −0.020 −0.019 −0.022 (0.064) (0.063) (0.063) GEFAid 0.024 0.022 0.042 (0.034) (0.034) (0.036) CO2percap (log) 0.067*** 0.053** 0.067*** (0.026) (0.026) (0.025) OPEC 0.412*** 0.434*** 0.450*** (0.118) (0.117) (0.117) Biodiversity (log) 0.052 0.053 0.024 (0.055) (0.055) (0.055) FoodProd 0.099 0.102 0.087 (0.109) (0.110) (0.124) AOSIS 0.040 0.022 0.145 (0.130) (0.131) (0.131) NatDistaster (log) 0.006 0.006 0.007 (0.008) (0.008) (0.009) CPIA 0.011 0.011 0.012 (0.019) (0.019) (0.018) EnvMinistry −0.019 −0.025 −0.019 (0.038) (0.036) (0.037) CO2percap * Polity2 0.004*** (0.002) OPEC * Polity2 0.028*** (0.011) EnvAidRecip * Polity2 0.003 (0.02) GEFAid * Polity2 −0.007* (0.004) Biodiversity * Polity2 0.008* (0.004) FoodProd * Polity2 0.006 (0.020) AOSIS * Polity2 −0.022*** (0.008) NatDistaster * Polity2 −0.0002 (0.001) AIC 7955 7957 7999 Notes: All models include random country and fixed year effects and also control for tourism and executive ideology. ***p < 0.01, **p < 0.05, *p < 0.1; standard errors in parentheses. Open in new tab Column (2) provides strong evidence that the strength of pro-emission interest groups are important drivers of COPDelSize. The interaction variable, polity2 with CO2percapita, is highly significant, suggesting a mechanism whereby groups with strong economic stakes in climate change increase NDS (hypothesis 2). The result on OPEC mirrors this finding. The other results from column (1) carry over.11 Column (3) examines whether the country's level of vulnerability translates into a larger NDS through a civil society mechanism by interacting all vulnerability variables with polity2. Despite significant findings, the interpretation of these results do not suggest this mechanism, with one exception. The finding on GEFAid, considering the effects of the interaction variable, suggests that the influence of GEFAid on NDS is lower for relatively more democratic countries. Similarly, the negative sign on the interaction of AOSIS with polity2 counters a civil society mechanism. The finding on biodiversity, however, suggests that only at relatively higher levels of democracy does the country's vulnerability relate to COPDelSize positively. The most plausible interpretation here is either a direct or indirect civil society effect (hypothesis 3). While not shown, we find no evidence for the bureaucracy mechanism (hypothesis 4). Although we find that having a higher CPIA score lowers the effects of GEFAid on COPDelSize, this mirrors the finding on polity2 and GEFAid. Moreover, we have no significant findings on any of the variables associated with green international political activity (not shown); interacting polity2 or the bureaucracy variables with any of the variables in this category produces no significant results. Subsample Analysis Table 2 splits the sample into Annex I and non-Annex I countries, given that UNFCCC processes, prominently the Kyoto Protocol, see these countries as having “common but differentiated” responsibilities, with most Annex I countries having agreed to mitigation targets within the protocol. Table 2. Subsample analysis . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 See table 1 notes. We control for all variables in table 1 but show only the significant results. Open in new tab Table 2. Subsample analysis . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 . Annex 1 countries (N = 840) . Non-Annex 1 countries (N = 3171) . DV: COPDelSize (log) . Full model . Full model . Pro-emission interest groups . Bureaucracy . Civil society & vulnerability . Civil society & green activity . . (1) . (2) . (3) . (4) . (5) . (6) . GDPpercap (log) 4.350*** 1.374*** 1.349*** 1.392*** 1.323*** 1.385*** (1.005) (0.360) (0.356) (0.349) (0.364) (0.363) GDPpercapsq (log) −0.429*** −0.193*** −0.188*** −0.197*** −0.186*** −0.193*** (0.127) (0.052) (0.051) (0.050) (0.051) (0.051) Population (log) 0.045 −1.928*** −1.870*** −1.962*** −1.930*** −1.955*** (0.988) (0.448) (0.456) (0.448) (0.454) (0.456) PopSquared (log) 0.038 0.179*** 0.174*** 0.181*** 0.180*** 0.182*** (0.077) (0.036) (0.037) (0.036) (0.036) (0.037) Polity2 −0.011 0.011** 0.011* 0.012** 0.007 0.023 (0.012) (0.005) (0.006) (0.005) (0.014) (0.014) RegQual 0.029 0.101** 0.100** 0.102** 0.103** 0.102** (0.089) (0.044) (0.043) (0.042) (0.044) (0.044) WBEB 0.229** 0.023 0.025 0.023 0.014 0.024 (0.089) (0.052) (0.052) (0.053) (0.052) (0.052) Distance −0.021** −0.034*** −0.034*** −0.034*** −0.034*** −0.034*** (0.008) (0.003) (0.003) (0.003) (0.003) (0.003) GEFCouncil 0.071 0.061 0.062 0.061 0.061 0.090* (0.068) (0.047) (0.047) (0.048) (0.046) (0.054) CO2percap (log) 0.157*** 0.035 0.038 0.111 0.031 0.028 (0.042) (0.032) (0.032) (0.090) (0.032) (0.032) OPEC 0.537*** 0.545*** 0.153 0.571*** 0.551*** (0.123) (0.123) (0.254) (0.124) (0.125) CO2percap * Polity2 0.002 (0.002) OPEC * Polity2 0.028** (0.011) OPEC * CPIA 0.128* (0.067) EnvAidRecip * Polity2 0.005** 0.005** (0.002) (0.002) GEFAid * Polity2 −0.009* −0.006 (0.005) (0.005) Biodiversity * Polity2 0.009** (0.005) AOSIS * Polity2 −0.019** (0.008) AIC 1609 6384 6371 6379 6400 6431 See table 1 notes. We control for all variables in table 1 but show only the significant results. Open in new tab In column (1) on Annex I countries, polity2 loses significance. Since the mean polity2 score for these countries is about 8.7 (max = 10), in contrast to 1.9 for the non-Annex I sample, the nonsignificant finding on polity2 makes sense—we are analyzing a sample of relatively democratic countries. For these countries, the size of the population seems to be divorced from NDS, though distance remains a discouraging factor. Importantly, CO2 emissions not only retains its high significance, but the coefficient on it is more than twice the coefficient in table 1 column (1), reinforcing the interest group mechanism. All else equal, a 1 percent increase in an Annex I country's CO2percapita boosts its NDS by about 15.7 percent. None of the three mechanisms are significant for any of the issue-specific variables in this subsample (not shown). Overall, financial resources (GDPpercapita) and interests seem to be the main drivers of NDS for Annex I countries. The rest of table 2 examines non-Annex I countries. Column (2) shows that in addition to those with high financial resources, non-Annex I countries that are relatively more democratic and have higher regulatory capacity have larger delegation sizes. This said, the effects of GDPpercapita and GDPpercapitasqaured differ across the two sub-samples, as shown in figure 3. Given their higher levels of average income, the peak for Annex I countries occurs at about USD 10,000 and for non-Annex I at around USD 2,100. Moreover, the highly significant result on OPEC provides quantitative evidence for the anecdotally well-known vocal participation of oil-producing countries in climate change negotiations. Figure 3. Open in new tabDownload slide Relationship of delegate size to GDP per capita by country type Figure 3. Open in new tabDownload slide Relationship of delegate size to GDP per capita by country type Column (3) interacts polity2 with the economic interest variables. The significant result on OPEC*polity2 is suggestive of governmental susceptibility to strong economic interests. There also appears to be a bureaucratic mechanism at work—column (4) shows the interaction variable between OPEC and CPIA to be significant. Reasonably, the relevant bureaucracy not only knows mitigation negotiations are key for OPEC countries but is also aware of its own role in implementing any decision. Further, we interact polity2 and the bureaucracy variables with all the variables associated with vulnerability and the level of green international political activity. As column (5) shows, some of the polity2 and vulnerability interactions are significant, but these variables do not strongly suggest a civil society mechanism. For example, for countries with higher levels of democracy, GEFAid has less of an impact on NDS compared to countries with relatively lower levels of democracy. Similarly, AOSIS membership has a greater impact on NDS for the relatively less democratic countries. However, column (5) shows that biodiversity influences NDS more for the relatively more democratic countries, potentially suggesting a civil society mechanism. Both columns also suggest the more democratic the country, the greater the effects of the level of bilateral environmental aid on NDS. In sum, there is weak evidence of vulnerability translating into a larger NDS through a civil society mechanism for the non-Annex I sample. Conclusions This paper presents the first extensive analysis of the determinants of UNFCCC members’ delegation sizes to the most important climate change conferences, the annual gatherings of the Conference of the Parties. It finds good evidence for the strength of domestic groups motivating larger delegation sizes, though evidence for the country's environmental vulnerability being positively related to its NDS is relatively weak. To the extent that pro-emission interest group politics dominate vulnerability considerations in explaining NDS, COP negotiations could be considered distorted in favor of those resisting ambitious mitigation efforts. Similarly, a country's existing level of activity in global environmental governance appears as a poor predictor of its NDS. Instead, resources and interests seem to be the primary drivers of larger delegations. The findings also offer policy implications. First, given the negative impact of distance, reform discussions should re-think COP locations. Choosing locations within easy reach of smaller and poorer nations would help these countries’ attendance at COPs, which is already hampered by low levels of financial and administrative capacity. Second, the funding from the UNFCCC Trust Fund for participation should focus on the poorest countries. Figure 3 showed that the relationship between average income and NDS turns negative after about USD 2,100, which corresponds to the lower range of the “lower middle income” category of the World Bank's 2015 country classifications. Hence, funds should go to low-income countries and the lowest income ones within the middle-income range. Third, given that regulatory quality significantly predicts NDS for non-Annex I countries, further attention to boosting the administrative capacity of smaller and poorer countries in COP negotiations is worthwhile. At the very least, existing efforts, such as the tradition of nongovernmental institutions providing support to developing countries (for example, the International Institute for Environment and Development supporting AOSIS countries) could be bolstered. Future research could investigate how NDS influences interstate interactions at other environmental platforms or link delegation size to outcomes at COPs. The determinants of delegation sizes and the variance in countries’ attendance at COPs are important to consider in discussions of reforms to the UNFCCC process. Reforms may ultimately hinge upon principles (Whose attendance should be boosted? Is limiting national delegation size at all desirable?), but analysis of the determinants of, and patterns in, national delegation sizes are critical to these policy debates. Footnotes 1 Herein, COPs. 2 UNFCCC Rules of Procedure (FCCC/CP/1996/2, Rule 17). 3 Neeff (2013) examines total attendance numbers. Schroeder et al. (2012) emphasizes increased attendance at COPs and notes that some small developing countries have downsized their delegations. They outline several plausible reasons for this but do not provide inferential statistics for these possibilities. Böhmelt (2013) analyzes nonstate actors within state delegations. Kruse (2014) examines women's attendance at COPs. Weiler (2012) explores the connection between NDS and negotiation positions at the Cancun COP; Bailer (2011) does the same for Copenhagen. 4 The UNFCCC releases only aggregate attendance numbers for each COP. We created the database of state-by-state delegation sizes from the UNFCCC's “lists of participants,” which lists each delegate by name. Böhmelt's (2013)supplementary files include delegation sizes until 2004 but does not analyze NDS as a variable. Our numbers resemble his. Most explanatory variables are not available beyond 2015. 5 Extant work does this for select COPs (e.g., Bailer 2011; Weiler 2012). 6 National delegations contain both state and nonstate (including business) actors. 7 Differentiating between direct and indirect pathways is not tenable here, and, regardless, both influences could simultaneously be at work. 8 We add one to each NDS to be able to consistently analyze countries that sent no delegates to some COPs but participated in others. We log some variables to dilute the effects of outliers. 9 In principle, COP venues rotate between the UN's five official regions; in practice, countries volunteer to host. 10 The directors on the GEF Council come from thirty-two “constituencies,” half of which belong to developing countries. The following countries have their own constituencies, hence directors: Canada, China, France, Germany, Italy, Iran, the Netherlands, the United Kingdom, and the United States. Most of these countries hold their own constituencies due to being the GEF's largest donors. In the remaining constituencies, directorship rotates. While no rule dictates this rotation, these directors tend to be from countries with relatively high voting power based on cumulative contributions to the organization (Perez del Castillo 2009; Lattanzio 2013). 11 Bolstering this interpretation, the results on the interaction of CO2percapita and OPEC with the regulatory quality variable are insignificant (not shown). Acknowledgements The authors are grateful to Katja Michaelowa, Axel Michaelowa, Thomas Bernauer, Christopher Kilby, Jennifer Peck, Alexandra Guisinger, and the many commentators at the 2016 Political Economy of International Organizations conference and the 2016 Temple Workshop on International Institutions and Global Governance for their feedback on this research. We additionally thank Axel Dreher and Christopher Kilby for sharing their data, respectively, on international organizations and World Bank Board membership. References Aklin Michael , Urpelainen Johannes . 2014 . “ The Global Spread of Environmental Ministries: Domestic-International Interactions .” International Studies Quarterly 58 ( 4 ): 764 – 80 . Google Scholar Crossref Search ADS WorldCat Ashton John , Wang Xueman . 2003 . “ Equity and Climate: In Principle and Practice .” In Beyond Kyoto: Advancing the International Effort against Climate Change , edited by Aldy J. E. et al. . , 61 – 85 . Arlington, VA : Pew Center on Global Climate Change . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bailer Stefanie . 2011 . “ Bargaining Resources and Strategies in Climate Change Negotiations .” Paper presented at the Workshop Negotiating Climate Change, Center for Comparative and International Studies (CIS), Zurich . OpenURL Placeholder Text WorldCat Bailer Stefanie , Weiler Florian . 2015 . “ A Political Economy of Positions in Climate Change Negotiations: Economic, Structural, Domestic, and Strategic Explanations .” Review of International Organizations 10 ( 1 ): 43 – 66 . Google Scholar Crossref Search ADS WorldCat Bättig Michele , Bernauer Thomas . 2009 . “ National Institutions and Global Public Goods: Are Democracies More Cooperative in Climate Change Policy? ” International Organization 63 ( 2 ): 281 – 308 . Google Scholar Crossref Search ADS WorldCat Block Ben . 2010 . “ Interview with Tuvalu Climate Negotiator Ian Fry .” WorldWatch Institute . January 13, 2010. http://www.worldwatch.org/node/6360 . OpenURL Placeholder Text WorldCat Böhmelt Tobias . 2013 . “ A Closer Look at the Information Provision Rationale: Civil Society Participation in States’ Delegations at the UNFCCC .” Review of International Organizations 8 ( 1 ): 55 – 80 . Google Scholar Crossref Search ADS WorldCat Dimitrov Radoslav S. 2010 . “ Inside UN Climate Change Negotiations: The Copenhagen Conference .” Review of Policy Research 27 ( 6 ): 795 – 821 . Google Scholar Crossref Search ADS WorldCat Dreher A. , Sturm Jan-Egbert , Vreeland J. . 2009 . “ Development Aid and International Politics: Does Membership on the UN Security Council Influence World Bank Decisions? ” Journal of Development Economics 88 : 1 – 18 . Google Scholar Crossref Search ADS WorldCat Gaventa Jonathan . 2010 . “ Environmental Mega-Conferences and Climate Governance beyond the Nation-State: A Bali Case Study .” St Antony's International Review 5 ( 2 ): 29 – 45 . OpenURL Placeholder Text WorldCat Hjerpe Mattias , Linnér Björn-Ola . 2010 . “ Functions of COP Side-Events in Climate-Change Governance .” Climate Policy 10 ( 2 ): 167 – 80 . Google Scholar Crossref Search ADS WorldCat Hicks Robert L. , Parks Bradley C. , Roberts J. Timmons , Tierney Michael J. . 2010 . Greening Aid? Understanding the Environmental Impact of Development Assistance . Oxford, UK : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Keohane Robert , David Victor . 2011 . “ The Regime Complex for Climate Change .” Perspectives on Politics 9 ( 1 ): 7 – 23 . Google Scholar Crossref Search ADS WorldCat Kruse Johannes . 2014 . “ Women's Representation in the UN Climate Change Negotiations: A Quantitative Analysis of State Delegations, 1995–2011 .” International Environmental Agreements 14 ( 4 ): 349 – 70 . Google Scholar Crossref Search ADS WorldCat Kuziemko Ilyana , Werker Eric D. . 2006 . “ How Much Is a Seat on the Security Council Worth? Foreign Aid and Bribery at the United Nations .” Journal of Political Economy 114 ( 5 ): 905 – 30 . Google Scholar Crossref Search ADS WorldCat Lattanzio Richard K. 2013 . “ International Environmental Financing: Global Environmental Facility (GEF) .” Congressional Research Service Report , R41165 . OpenURL Placeholder Text WorldCat Mansfield E. , Pevehouse J. . 2006 . “ Democratization and International Organizations .” International Organization 60 ( 1 ): 137 – 67 . Google Scholar Crossref Search ADS WorldCat Michaelowa Katharina , Michaelowa Axel . 2012 . “ Negotiating Climate Change .” Climate Policy 1295 ( 5 ): 527 – 33 . Google Scholar Crossref Search ADS WorldCat Monzani Bernardo . 2018 . “ IIED Support to the Least Developed Countries Group. Influencing Global Climate Change Negotiations .” Evaluation Case Study, IIED: London . OpenURL Placeholder Text WorldCat Nasiritousi Nagmeh , Linnér Björn-Ola . 2016 . “ Open or Closed Meetings? Explaining Nonstate Actor Involvement in the International Climate Change Negotiations .” International Environmental Agreements 16 ( 1 ): 127 – 44 . Google Scholar Crossref Search ADS WorldCat Neeff Till . 2013 . “ How Many Will Attend Paris? UNFCCC COP Participation Patterns 1995–2015 .” Environmental Science and Policy 31 : 157 – 59 . Google Scholar Crossref Search ADS WorldCat Neumayer Eric . 2002 . “ Do Democracies Exhibit Stronger International Environmental Commitment? A Cross-country Analysis .” Journal of Peace Research 39 ( 2 ): 139 – 64 . Google Scholar Crossref Search ADS WorldCat Perez del Castillo Carlos . 2009 . “ Governance of the GEF .” OPS Technical Document #4, GEF Independent Evaluation Office: Washington, DC . OpenURL Placeholder Text WorldCat Rietig Katharina . 2014 . “ Reinforcement of Multilevel Governance Dynamics: Creating Momentum for Increasing Ambitions in International Climate Negotiations .” International Environmental Agreements 14 ( 4 ): 371 – 89 . Google Scholar Crossref Search ADS WorldCat Roberts Timmons J. , Parks Bradley . 2006 . A Climate of Injustice: Global Inequality, North-South Politics, Climate Policy . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schroeder Heike . 2010 . “ Agency in International Climate Negotiations: The Case of Indigenous Peoples and Avoided Deforestation .” International Environmental Agreements: Politics, Law and Economics 10 ( 4 ): 317 – 32 . Google Scholar Crossref Search ADS WorldCat Schroeder Heike , Boyloff Maxwell T. , Spiers Laura . 2012 . “ Equity and State Representations in Climate Negotiations .” Climate Change 2 ( 12 ): 834 – 36 . OpenURL Placeholder Text WorldCat Schroeder Heike , Lovell Heather . 2012 . “ The Role of Non-Nation-State Actors and Side Events in the International Climate Negotiations .” Climate Policy 12 ( 1 ): 23 – 37 . Google Scholar Crossref Search ADS WorldCat Söderberg Mattias , Andersen Sigrid Bjerre . 2008 . “ Unbalanced Participation .” Dan Church Aid Research Paper 8, August . OpenURL Placeholder Text WorldCat Sprinz Detlef , Tapani , Vaahtoranta . 1994 . “ The Interest-Based Explanation of International Environmental Policy .” International Organization 48 ( 1 ): 77 – 105 . Google Scholar Crossref Search ADS WorldCat United Nations Conference of the Parties , Second Session . “ Framework Convention on Climate Change .” FCCC/CP/1996/2, July 8–19, 1996 . OpenURL Placeholder Text WorldCat Weiler Florian . 2012 . “ Determinants of Bargaining Success in the Climate Change Negotiations .” Climate Policy 12 ( 5 ): 552 – 74 . Google Scholar Crossref Search ADS WorldCat Wheeler David . 2011 . “ Quantifying Vulnerability to Climate Change: Implications for Adaptation Assistance .” Working Paper 240. Center for Global Development, Washington, DC . OpenURL Placeholder Text WorldCat Yamin Farhana , Depledge Joanna . 2004 . The International Climate Change Regime: A Guide to Rules, Institutions and Procedures . Cambridge : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC © The Author(s) (2020). Published by Oxford University Press on behalf of the International Studies Association. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Foreign Policy Analysis – Oxford University Press

Published: Jun 1, 2020

There are no references for this article.