Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2019 Jun 21;33(2):287–304. doi: 10.1111/gove.12427

Should I stay or should I go? Explaining variation in nonstate actor advocacy over time in global governance

Marcel Hanegraaff 1, Jorik Vergauwen 2, Jan Beyers 2
PMCID: PMC7161925  PMID: 32322136

Abstract

The past decades have been characterized by a growing number of nonstate actors (NSAs) involved in global governance. However, despite this growth, only a small number of NSAs have been able to maintain a prolonged global presence over a substantial period of time. To explain why some NSAs are more active, we rely on resource dependence theory. We demonstrate that sustained advocacy over time can be explained by a density dependence mechanism, namely the more NSAs mobilize, the lower the chance that individual NSAs will prolong their global advocacy efforts. Analysis of data stemming from a unique data set of 5,627 NSAs active at the global climate conferences demonstrates that much advocacy in this field is indeed of an incidental nature, namely a large number of groups attend once and never return.

1. INTRODUCTION

As the Cold War ended, international organizations (IOs) increasingly began opening up, resulting in a substantial increase in the number of nonstate actors (NSAs) becoming involved in global governance (Beckfield, 2003; Hanegraaff, 2015a; Nordang‐Uhre, 2014; O'Brien, Goetz, Scholte, & Williams, 2000; Schroeder, Boykoff, & Spiers, 2012; Smith, 2005; Smith & Wiest, 2005; Steffek, Kissling, & Nanz, 2007; Tallberg, Sommerer, Squatrito, & Jönsson, 2014). To illustrate, while in 1990 approximately 20,000 transnational NSAs were registered in the Yearbook of International Organizations, by 2016, approximately 68,000 registered transnational NSAs were reported, an increase of 340%. Attempts to map patterns of transnational advocacy are an important step forward in qualifying the role of NSAs in global governance. Yet simply tallying the number of mobilized transnational interests can also be misleading, because aggregate measures discount variation in types of transnational advocacy (Berkhout & Lowery, 2011; Gray & Lowery, 1996; Halpin & Jordan, 2012). This could have important consequences. For instance, if we were to evaluate the influence of NSAs merely on the basis of the numbers that are active in IOs, this would imply that NSAs should be influential in almost every issue area. Yet such a conclusion is empirically unwarranted, as many scholars have demonstrated that global governance is still largely state driven, with limited opportunities for NSAs to effectively influence global policymaking ( Betsill & Corell, 2001, 2008; O'Brien et al., 2000; Tallberg, Sommerer, Squatrito, & Jönsson, 2013).

A useful conceptual tool in this regard is to identify a “core” and a “periphery” in interest communities (Halpin & Fraussen, 2017; Heinz, Laumann, Nelson, & Salisbury, 1993; LaPira, Herschel, & Baumgartner, 2014). This distinction acknowledges that not all NSAs are equal and that some are more “important” in the political process than others. Halpin and Fraussen (2017, p. 727) have argued, for instance, that the notion of prominence captures the idea that “groups vary with respect to how preeminent they are as voices for a particular constituency among political elites, and thus refers to the situation whereby some groups are simply assumed to be relevant to the issue at hand” (see also Betsill & Corell, 2001, p. 77). Yet studies have differed in the way they operationalize groups that are at the core and those on the periphery of NSA communities. For instance, one can look at who gains more consistent access to policymakers (Betsill & Corell, 2001; Rasmussen & Carroll, 2014); one can analyze the number of issues in respect of which organizations are active (Halpin, 2011; Halpin & Thomas, 2012); or one can assess the number of political venues NSAs target (e.g., Binderkrantz, Christiansen, & Pederson, 2015).

This article takes a novel approach and makes a distinction based on the level of activity of NSAs over time. This is important, because repeated engagement increases the chances of generating policy influence, and in the long term may positively affect the organizational maintenance of NSAs (Fraussen, 2014; Gray & Lowery, 1996; Halpin & Jordan, 2012; Heinz et al., 1993). Yet despite its importance, to date no study has taken into account the volatile nature of NSA participation in global governance. Our overall argument is premised on resource dependence theory. The central idea is that NSAs need resources—both financial means and expert information—for their maintenance and political activities (Binderkrantz et al., 2015; Bouwen, 2002; Lowery, 2007; Rasmussen & Gross, 2015). Our expectation is that NSAs that possess, or are more readily able to acquire, resources most valuable for global advocacy are more likely to sustain their political activities over a longer period of time.

To test our hypotheses, we mapped all NSAs that have been active at the Conferences of the Parties of the United Nations Framework on Climate Change Conventions (COPs and UNFCCC hereafter) between 1997 and 2011. This was followed by an event history analysis to explain variation in advocacy activity over time. The results demonstrate that much advocacy in this field is of an incidental nature, namely many NSAs attend once and never return to later conferences. Moreover, the analysis indicates that increased competition, measured as the number of advocates attending global climate conferences, corresponds with higher levels of volatility and less sustained political advocacy.

2. FROM ENTERING GLOBAL POLITICAL ARENAS TO STAYING ACTIVE

The late 1990s and early 2000s witnessed much debate on whether or not IOs should become more open to the input of NSAs. The proponents of greater openness argued that IOs should grant more access to a wide range of NSAs as this would contribute to accountability and legitimacy. The normative assumption underlying an inclusive model of global democracy is that the actors affected by some political decision should be given the opportunity to participate and make their voices heard in the decision‐making process (Bexell, Tallberg, & Uhlin, 2010; Dellmuth & Tallberg, 2015; Scholte, 2004; Steffek et al., 2007). As Agné et al. (2015, p. 7) put it, different variants of stakeholder theory are unified by the assumption that strengthened opportunities for involvement of self‐organized stakeholders in political procedures hold significant promise for making those procedures more democratic. As such, increased participation by NSAs holds the distinct promise to “open up means of influence for groups in society that often are marginalized in representative bodies” (Bexell et al., 2010, p. 87).

Other scholars have criticized these optimistic views and have argued that IOs should remain state‐driven institutions in which NSAs have only a limited role, as NSA advocacy tends to be unevenly biased in favor of wealthy interests (Tallberg et al., 2013; Zürn, 2014). Recent studies have also indicated that the processes of collective interest representation are not equally distributed among citizens and do not necessarily benefit marginalized groups (Skocpol, 2004; Strolovitch, 2007). Translating this to global politics, this would mean that interests from developed countries and/or business groups would profit more from increased access opportunities compared to advocates who represent interests located in developing countries (Bexell et al., 2010; Steffek et al., 2007; Tallberg et al., 2014). Hence, opening up IOs would not weaken the position of developed countries or the business community, but it might just reinforce their position (see Bexell et al., 2010).

Ultimately, however, whether or not the participation of NSAs has led to a more level playing field is an empirical question. Therefore, many scholars have begun to map global NSA communities to see how they are constructed and which types of interests are more intensely mobilized. This has taught us a great deal. Most prominently, the results have highlighted a rather biased distribution of NSAs active in IOs, more in line with the second, more pessimistic, account described earlier (see Zürn, 2014, p. 62). For instance, most of these studies have highlighted how NSAs from developed countries vastly outnumber NSAs from developing countries or the least developed countries active in IOs and in global governance more broadly (Beckfield, 2003; Hanegraaff, Braun, De Bièvre, & Beyers, 2015; Nordang‐Uhre, 2014; Schroeder et al., 2012; Smith, 2005; Smith & Wiest, 2005; Steffek et al., 2007; Tallberg et al., 2013). Moreover, other mapping studies have highlighted how NSA participation in various IOs is largely a matter of business lobbying (Berkhout, Beyers, Braun, Hanegraaff, & Lowery, 2018; Berkhout, Hanegraaff, & Braun, 2017; Nordang‐Uhre, 2014; Tallberg et al., 2013).

Overall, mapping NSA communities at the global level has taught us a great deal about transnational advocacy. Yet simply tallying the number of mobilized actors has certain limitations. Most importantly, the findings can be misleading, because aggregate measures discount the unequal levels of mobilization within these populations (Halpin & Jordan, 2012, p. 245). Simply put, some NSAs are central players in the policy process, while others are mostly looking from the outside in and have no noticeable impact on political processes (Betsill & Corell, 2001; Halpin & Fraussen, 2017; Heinz et al., 1993; LaPira et al., 2014). A simple map of this variation indicates that all actors are considered equal and no core–periphery structure is detected. As said before, there are various ways in which one could further structure NSA populations, yet in this article, we take a novel approach by differentiating core NSAs from peripheral NSAs based on their level of activity over time. We therefore differentiate between, on the one hand, “core players” that are NSAs which, due to their persistent presence over time, are more central players in the policymaking process. On the other hand, NSAs on the periphery can be labeled as “peripheral players” because they quickly exit the transnational domain once mobilized for the first time.

The reason we structure a global NSA community based on activity rates is that studies on domestic advocacy have taught us that being a repeat player has considerable benefits. That is, sustained activity increases the chances of generating policy influence, and in the long term tends to positively affect the organizational maintenance of organized interests (Fraussen, 2014; Gray & Lowery, 1996; Halpin & Jordan, 2012; Heinz et al., 1993). Indeed, it is not hard to imagine that groups, which invest in long‐term relations with policymakers, are more likely to have an impact on policy outcomes. We expect this to apply to global politics as well, and perhaps even more so. Global policymaking is generally slow, which makes transnational advocacy a costly enterprise, because covering such time spans means that NSAs have to invest substantial resources over the course of many years (Tarrow, 2005; Zürn, 2014). Groups that have carefully monitored the policy process and kept close contacts with policymakers should experience considerable benefits when aiming to influence policy outcomes. As a consequence, it is relevant to know who the core and peripheral players are, and what explains why some are better able to sustain their political activities.

3. ANALYZING NSA ACTIVITY OVER TIME

In explaining who are core and who are more peripheral players and why, our argument is premised on resource dependence theory (e.g., Binderkrantz et al., 2015; Bouwen, 2002; Rasmussen & Gross, 2015; Tallberg, Dellmuth, Agné, & Duit, 2018). Resource dependence theory states that organizational “success,” in our case, sustained advocacy at climate summits, is affected by a combination of NSA's resources and how these resources fit into the context in which the NSA operates.

About the first, the resources an NSA possesses, analytically one can make a distinction between political and financial resources. With regard to “political resources,” the overall idea is that policymakers will provide more access to NSAs that are able to supply information they are in need of. Over time, we expect that NSAs that are able to supply valuable information to policymakers will demonstrate the best odds to sustain their global political activities. Next to informational exchanges with policymakers, NSAs, however, also need “financial resources” to bear the costs of sustained advocacy (Gray & Lowery, 1997, p. 28; Lowery, 2007; Halpin & Thomas, 2012). Simply put, if NSAs do not possess sufficient resources, they cannot maintain their advocacy efforts for long and will exit prematurely. Furthermore, resource dependence theory states that the resources NSAs have at their disposal are not equally valuable in every context. It is therefore important to also consider how the context(s) in which NSAs operate affects the chances that NSAs stay active at the climate summits (e.g., Nordang‐Uhre, 2014; Schroeder et al., 2012; Steffek et al., 2007; Tallberg et al., 2014). In global politics we need to take stock of the fact that NSAs are multilevel players, operating in various contexts situated at the “global” and “national” levels.

Combined, this thus means we need to analyze how the national and global contexts facilitate NSAs to exchange information with policymakers and/or acquire financial support to sustain their lobbying activities. This leads us to formulate four hypotheses. Two hypotheses focus on the type of resources NSAs have at their disposal, political as well as financial resources, and how these resources affect sustained global advocacy (Hypothesis 3 and Hypothesis 4). The other two hypotheses deal with the context in which groups operate and how this context affects NSAs' ability to exchange political resources with policymakers (Hypothesis 1) and to acquire financial resources to sustain their global activities (Hypothesis 2). Table 1 summarizes the hypotheses.

Table 1.

A resource dependence perspective on why NSAs stay active at global venues

Type of resources needed for sustained global advocacy
Political resources Financial resources
Is the context beneficial for acquiring resources? Competition among organizations for access to policymakers (Hypothesis 1) Ability to acquire financial resources at the national level (Hypothesis 2)
Does the organization possess valuable resources themselves? Match between type of information organizations have and policymakers demand (Hypothesis 4) Financial resources organizations already have at their disposal (Hypothesis 3)
Open in a new tab

We start with the context. A number of contextual forces can be considered to directly affect NSA's ability to secure the resources that are essential for maintaining political activities (Lowery, 2007). First, we expect that the perception of impacting negotiation outcomes is strongly shaped by the amount of competition NSAs face, namely how many other NSAs are seeking attention from and access to policymakers. More precisely, organizational ecology scholars understand the level of competition as a density‐dependent process, namely the more interests are mobilized, the greater the competition for resources and the more difficult it is to maintain a lobbying presence (Gray & Lowery, 1996; also Hannan & Freeman, 1989; for one of the first applications of this perspective to global environmental advocacy see Hadden, 2015). Density should thus affect the access opportunities NSAs enjoy. Who should we expect to stay active and who is more likely to exit as a result of this aspect? Following the former logic, we expect that NSAs that entered the community for the first time when hardly any competitors were active are more likely to remain active (and vice versa) (Abbott, Green, & Keohane, 2016; Gray & Lowery, 1996; Halpin & Jordan, 2012; Hanegraaff, 2015a). That is, these groups can establish contacts with policymakers as well as other NSAs, or they may gain expertise in policy issues and develop experience in how to lobby effectively. Moreover, gaining sustained access is not only important for NSAs in their search for impact; they also need to do this to legitimize their actions to their members (Gray & Lowery, 1997, p. 28; Halpin & Thomas, 2012; Bouwen, 2002).

Key to gaining access is the amount of competition groups face, namely how many other NSAs are actively seeking access (see Salisbury, 1990, p. 203). If there are many NSAs active at a particular venue, policymakers need to be much more selective about whom they talk to. If there are only a few competitors, NSAs are better able to gain access to policymakers. Yet, as the attention span of policymakers is limited, the more NSAs reach out to them, the more NSAs will be excluded from the policy process. We thus presume that those who enter a dense environment face more competition for access, while those who enter when a few others are active experience less competition (Abbott et al., 2016; Gray & Lowery, 1996; Halpin & Jordan, 2012; Hanegraaff, 2015a). As a consequence, NSAs that operate in a less dense community are expected to develop a steadier presence, while those who attend more crowded events are less likely to become repeat players.

While the competition NSAs face at the global level affects their political opportunities, other resources such as finances and constituency support are mostly affected by the national context. Our expectation is that NSAs that can more easily attract these critical resources are more likely to sustain global activity. Indeed, effective advocacy requires sustained monitoring of the policy process at a foreign location, developing global policy networks, gathering evidence across different political entities, and rallying constituencies from various parts of the globe (Tarrow, 2005; Zürn, 2014, p. 62). Sustaining these activities over long periods of time should exhaust NSAs' resources, meaning that NSAs that are less able to mobilize resources demonstrate a lower propensity to sustain these activities. Importantly, the funding of NSAs stems mostly from a national context, such as from a supportive constituency and the potential membership of NSAs, the extent to which such support is available, and/or patronage by political entrepreneurs. This means that the capacity to remain active in global politics is crucially influenced by the ability of groups to obtain resources from a national context (Lowery, Gray, & Fellowes, 2005; Tarrow, 2005). We therefore expect the maintenance of an advocacy presence over time to covary with the overall resource base that is potentially available at the national level (see also Lowery & Gray, 1996; Lowery & Gray, 2004; Nownes, 2010). Empirically, we expect the resource base to be larger in developed countries compared to developing countries (Hanegraaff et al., 2015; Smith & Wiest, 2005; Tallberg et al., 2014; Zürn, 2014). Nonstate actors in developed countries should find it easier to mobilize constituents, find donors, or attract sponsors financing global advocacy campaigns. As such, organizations rooted in developed countries are more likely to stay active at global political venues.

Combined, we propose two hypotheses related to the context in which NSAs operate:

Hypothesis 1

Nonstate actors that start lobbying in a less dense environment develop a steadier lobbying presence than NSAs that start lobbying in a denser context.

Hypothesis 2

The more resources available in the direct environment, the more likely NSAs are to develop a steady lobbying presence.

Besides contextual factors, as stated, resource dependence theory also includes expectations on organizational traits and how these fit with the environments (Abbott et al., 2016; Gray & Lowery, 1997, 1998). The question is which NSAs possess relevant resources for global advocacy. First, building on the former hypothesis, a defining feature of global policymaking is its inertial nature, making global advocacy costly (Tarrow, 2005; Zürn, 2014, p. 62; see earlier). Following a resource dependence perspective, we would therefore expect NSAs better equipped in terms of resources to be better able to sustain their presence at climate summits. In other words, next to the ability of NSAs to acquire resources (see Hypothesis 2), the resources already at the disposal of NSAs should also matter in explaining why some NSAs are more active at climate summits than others. We therefore expect that NSAs with more resources will develop a more sustained presence at climate conferences compared to groups that have less immanent resources at hand. Empirically, this leads us to expect that especially business organizations will develop a more sustained presence, as these groups are broadly understood to have more relevant financial and political resources at their disposal compared to nonbusiness interests (for instance, see Baumgartner, Berry, Hojnacki, Kimball, & Leech, 2009; Lowery, 2013; Dür & Mateo, 2016).

Second, next to resources, we expect the scope of representation to matter as well. Some NSAs specialize in particular strategies and issues, while others cast their nets wide, aiming to represent broader interests (Bouwen, 2002; Browne, 1985). Advocacy is a reciprocal process, whereby NSAs seek access, but policymakers grant access to some NSAs and not to others. Hence, the selective demands of policymakers matter a great deal and who gains access depends on how valuable the supplied information is (Bouwen, 2002). As the climate conferences are characterized by large‐scale negotiations, in which a multiple set of issues are negotiated simultaneously (Falkner, Stephan, & Vogler, 2010; Jinnah, 2011), we expect policymakers to seek the input of NSAs that can provide information about broad issue linkages among multiple more specialized interests, or at least information that is not only representative of a rather narrow interest (see Davis, 2004; Miller & Dolšak, 2007). Therefore, more encompassing groups that represent broad and diverse constituencies are helpful, as these groups aggregate a diverse set of more specialized views. This logic resembles traditional modes of corporatist policymaking (Molina & Rhodes, 2002). As it is impossible for policymakers to interact with all relevant actors individually, the intermediation capacity of a small number of stakeholders who represent a broad constituency is crucial (Willetts, 2006). Hence, NSAs representing encompassing interests are best able to provide this type of information, and policymakers will grant these NSAs more access than NSAs representing highly specialized interests (see Bouwen, 2002; Dür & Mateo, 2016). Combined, we propose two hypotheses related to NSA characteristics:

Hypothesis 3

Nonstate actors with more resources—more specifically business interests—will develop a more sustained advocacy presence compared to NSAs with fewer resources.

Hypothesis 4

Encompassing NSAs—more precisely cross‐sectoral business associations and NGOs—will develop a more sustained advocacy presence compared to more specialized NSAs.

4. DATA AND METHODS

We analyzed the exit and entry rates of all NSAs that attended the annual UNFCCC COPs between 1997 and 2011. Given our aims—to map an entire NSA community over time—our options were limited, because these data are usually not readily accessible for most IOs. Helpfully, the UNFCCC has listed all NSAs present at each COP since 1997. Another advantage is that, over time, the UNFCCC has consistently demonstrated a high level of openness, which means a wide range of active NSAs hailing from many countries. This is especially important for testing Hypothesis 3 and Hypothesis 4. We mapped all NSAs appearing at each of the 16 COPs under study. Doing so indicated the volatility of the entire population during this period and allowed us to detect “core players” and “peripheral players.” The total number of groups that attended these conferences is 6,634. After identifying them, we checked all the organizational websites. Supervised by a PhD researcher, several student coders collected the data, working in small teams of four and regularly meeting to cross‐verify the treatment of specific cases. For most NSAs, a Web site offered elaborate data; for about 20% we found no organizational Web site, but information stored elsewhere (for example on LinkedIn and Wikipedia) enabled us to code enough variables. For a very small number (n = 21), no information was found. In short, our data set provides a comprehensive overview of the organization types, national origins, areas of interest, constituency bases, as well as how the NSAs are organized.

We used event history analysis to model the likelihood of exit after an NSA attended a COP for the first time. In a nutshell, we modeled immediate exit, that is, the first exit or instance of not attending a COP after the first occasion of attending a COP, and compared that to the number of successive times an NSA had attended COPs before that exit. This allowed us to observe why organizations exit quickly (for instance, after one appearance) or stay active over longer time spans (for instance, three COPs). Only for the 534 organizations that participated for the first time at the last analyzed COP in our data set (Durban in 2011), exit rates could not be calculated and these were thus omitted from the analysis. Moreover, 473 organizations (7.7%) were excluded because of a missing value on at least one of the independent variables. This left us with 5,627 NSAs for which we analyzed the advocacy patterns over time.

Furthermore, consecutive conference participation and exit were considered repeatable. In other words, organizations may participate for a certain period, followed by an interruption, and subsequently participate again. Each successive participation was defined as an episode, in other words, a continuous period during which the NSA is at risk of experiencing exit. This led to a data set of 11,931 episodes (see Figure 1 for a distribution of the episodes). The event history analysis models exit rates after each episode, considering the length of this episode. The longer an episode, the better this NSA had sustained its participation in the conferences, in line with our theoretical model. It is important to add that the event history analysis pooled all episodes together. That is, we did not consider returning NSAs as new, because they had been active at conferences before. Therefore, a random effect was included at the organizational level. Put differently, the multilevel model controls for the fact that some NSAs reentered the population at a later period, yet are still treated as the same NSA in the analysis (see Data S1 for more technical details on how we modeled exit).

Figure 1.

Figure 1

Open in a new tab

Number of organizations by episodes (N = 6,634)

For an overview of dependent, independent, and control variables, see Table S1. More specifically, we tested four hypotheses. For Hypothesis 1 (global context), we developed a measure of the size of each policy domain‐related advocacy community at the previous COP the NSA had attended (that is, this index varies by COP for every NSA). We constructed an index by counting the number of NSAs in a similar domain (see Table S1). For instance, if 100 agricultural organizations were active at the 1997 COP, we coded the 1998 density measure for an agricultural organization as 100. This relates the decision of (non)participation in 1998 to the experienced density of the previous attended COP (1997). Due to its skewed distribution, we log‐transformed this variable. For 2 (national context), we assumed that an overall assessment of economic development provides the best approximation of the resources that are potentially available in the direct environment of an NSA. For this purpose, we used the country GDP per capita in the year prior to the COP; the data were retrieved from the World Bank Statistical Division.

For organizational traits, that is Hypothesis 3 and Hypothesis 4, we distinguished between three types of NSAs. Specialized business groups defend the interests of product‐level economic sectors (for example, dairy farming), while encompassing business groups focus on a broader set of products, sectors or multiple sectors, such as agriculture. To differentiate between the two types, we rely on the International Standard Industrial Classification of All Economic Activities (ISIC). NSAs representing issue areas coded ISIC levels 3 or 4 were categorized as specialized business organizations; organizations coded ISIC levels 1 or 2 were categorized as encompassing business associations. Next, NGOs were coded by considering their activity areas (for example, human rights, environment, etc.). Finally, the data set includes research organizations—NSAs mostly funded by governments and which portray themselves predominantly as creators or disseminators of expert knowledge—and labor unions representing employee interests. For Hypothesis 3 (resources), we relied on the distinction between business groups and NGOs as a proxy for variance in resource endowment. A more accurate measure would be the actual resources NSAs have at their disposal. However, this would require the collection of detailed information for each NSA for a time span of almost 15 years, which is impossible for over 6,000 organizations. Assuming that business interests typically have more resources is a common presumption in the literature (for similar arguments see Baumgartner et al., 2009; Lowery, 2013; Dür & Mateo, 2016). For Hypothesis 4 (scope), we distinguished between specialized business groups and encompassing NSAs (NGOs and business associations).

Our models contain several control variables. First, we included a measure for the extent to which the policy portfolio of an NSA aligns with the agenda of the COP, as we expected NSAs demonstrating a closer alignment to develop a steadier lobbying presence. For this purpose, we compared domains that have a direct link to COP meetings, such as the environment, with fields that are less directly connected to the COPs, such as human rights (see Table S2).

Second, based on the GeoDist database, we added a control for the distance of conferences to the countries of origin of the organizations (Nordang‐Uhre, 2014). It is expected that a COP taking place closer to NSA's country of origin would increase the likelihood of repeated COP participation.

Third, we controlled for the level of democracy of the country where the NSA hails from (Rohrschneider & Dalton, 2002, p. 528; Barry, Bell, Clay, Flynn, & Murdie, 2015). For this we relied on the Polity score one year prior to the COP; the scale ranges from −10 to 10, where a high score refers to a full democracy and a low score to an authoritarian regime (Marshall, Gurr, & Jaggers, 2018). As most NSAs originate from democratic countries, the measure for the level of democracy of the country of origin has a strongly skewed distribution (70% of the NSAs scored 10). Therefore we recoded the Polity score into three categories: fully democratic (scores 9 and 10), partially democratic (scores 6 to 8), and limited democracy or nondemocratic (scores 5 or lower). The expectation was that NSAs hailing from democratic countries would demonstrate a higher level of sustained advocacy compared to the other two categories.

Fourth, we included the jurisdictional level from which the NSAs retrieve their resources. Each NSA was coded as being a national, regional (if it scoped multiple countries), or global actor depending on the jurisdictional level at which its supportive constituency (members, sponsors, donors, etc.) is located. We expected that global NSAs would be more actively monitoring global negotiations, as this is their natural habitat.

Fifth, we controlled for public salience, as advocacy might be more robust if climate conferences are more salient. Here we made a distinction between domestic salience (for domestic groups) and global salience (for global groups). We relied on the online news engine Factiva in order to code media hits by searching for the words “climate conference,” “climate negotiations,” “UNFCCC,” and “climate change.” For global salience, we used a special filter in Factiva that selects all major global news sources (such as Associated Press, Xinhua News Agency, Agence France‐Presse, BBC Monitoring, Deutsche Presse‐Agentur, and Russian News Agency TASS; see https://global.factiva.com/). For domestic salience, we could not find newspapers in all countries, and therefore coded public salience only for those countries that demonstrated the highest attendance rates at COPs (mostly OECD countries). If possible, we coded public visibility for the five most read newspapers per country. For both global and domestic salience, we calculated the relative attention for each COP, that is, the number of articles about a particular COP as a proportion of the total number of articles per country. This allowed us to compare media attention over time across countries.

5. DATA ANALYSIS

We first present descriptive results illustrating the volatility of the COP advocacy community. Figure 2 plots the number of times all NSAs attended the conferences between 1997 and 2011, clearly demonstrating this strong volatility. To give a few examples, 64% of the organizations had attended the conferences only once, whereas only 0.3% of NSAs (or, n = 19) had attended all COPs (see Table S3 for the top 100 most active NSAs). The latter NSAs include the usual suspects: broadly organized business associations such as BusinessEurope, the International Chamber of Commerce (ICC), the World Business Council for Sustainable Development (WBCSD), the International Gas Union, and prominent NGOs such as Greenpeace, the US Climate Action Network (CAN), the World Council of Churches, and The Nature Conservancy. Other well‐known core players are the Edison Electrical Institute, the Center for Clean Air Policy, and the Citizens' Alliance for Saving the Atmosphere and the Earth. There are no firms among these top 100 attendees. Yet firms appear frequently among the most frequent attendees (more than 10 appearances) and include firms for which there is much at stake, such as Shell, ExxonMobil, E.ON UK, DuPont, The British Petroleum Company, Chevron Corporation, and the Dow Chemical Company.

Figure 2.

Figure 2

Open in a new tab

Number of appearances at all COPs (not necessarily in a row) (N = 6,634)

Those NSAs that had attended a large number of conferences constitute a fraction of the entire community. For instance, NSAs that had attended more than 10 conferences constitute only 3% of the entire community. By contrast, groups that were present at only one or two conferences in total constitute no less than 78%, or 5,160 organizations, of the entire NSA community. These organizations include a highly diverse array of actors, ranging from the Indigenous Peoples' Alliance of South Sulawesi to the Japan Cement Association, and from the Montreal West United Church to the Kenya Electrical Trade and Allied Workers' Union. Generally, it seems that most NSAs appear only sporadically at COPs and that many of them disappear after their first attendance. Given these differences, it is crucial to learn who the repeat players and who the irregular attendees are.

For our multilevel discrete‐time event history models, we considered different episodes of NSA mobilization together (see research design). Table 2 presents three models. We distinguish between one global model and two national models, namely one for all UNFCCC member states and another model that focuses on a narrower set of OECD countries. What are the results? First, for Hypothesis 1, the density hypothesis, we expected that if NSAs entered a conference for the first time when many NSAs were active in a similar domain, the chance that they would return to the next COP was significantly lower compared to those NSAs that entered a less crowded environment. The results support the proposed hypothesis, that is, when NSAs lobby for the first time at a conference attended by many other similar‐minded stakeholders, we see a drastic increase in the likelihood of exit, or at least of not repeating their advocacy efforts by attending the next COP. By contrast, NSAs that mobilize for the first time at less crowded conferences have a significantly higher chance of attending successive COPs. This confirms Hypothesis 1, supporting our assertion that NSAs entering a crowded community give up their advocacy efforts more readily than those entering for the first time when there is less competition for access.

Table 2.

Predicting exit rates or transnational advocates at UNFCCC COPs (odds ratios)

Global National—all National—OECD
Coef. SE Coef. SE Coef. SE
Constant 0.098*** (0.04) 2.317 (1.41) 0.239*** (0.10)
Hazard function
Spline 1 2.678*** (0.29) 2.923*** (0.28) 2.986*** (0.34)
Spline 2 Ref. Ref. Ref.
Spline 3 0.618*** (0.09) 0.775** (0.10) 0.795 (0.12)
Spline 4 0.323*** (0.12) 0.428** (0.16) 0.364** (016)
Independent variables
Density‐sector 1.108** (0.05) 1.150*** (0.46) 1.157*** (0.05)
Group type
Business specialized Ref. Ref. Ref.
Business association 0.612*** (0.10) 0.564*** (0.10) 0.493*** (0.10)
NGO 0.651*** (0.10) 0.450*** (0.08) 0.474*** (0.10)
Labor 1.012 (0.44) 0.449* (0.18) 0.594 (0.32)
Research 0.736* (0.12) 0.575*** (0.09) 0.612** (0.10)
GDP/capita 0.787*** (0.04)
Controls
Alignment 0.848 (0.11) 0.826 (0.11) 0.875 (0.14)
Distance—log 1.295*** (0.42) 1.274*** (0.04) 1.217*** (0.04)
Polity
Full democracy Ref.
Partial democracy 0.824 (0.18)
No democracy 1.274 (0.21)
Level
Global Ref.
Regional 0.579*** (0.08)
Global salience 0.969 (0.03)
Domestic salience 0.972 (0.04)
Random effects
Organization variance 0.623 (0.22) 0.705 (0.17) 0.763 (0.22)
Country variance 0.028 (0.06) 0.287 (0.08) 0.242 (0.09)
Sector variance 0.024 (0.06) 0.137 (0.10) 0.200 (0.14)
Model diagnostics
N observations 4,534 7,397 4,922
N organizations 1,680 3,947 1,390
BIC 5,426.48 8,128.02 5,634.67
AIC 5,323.77 8,010.57 5,537.15
Open in a new tab

Note: The coefficients are estimated using multilevel regression.

Abbreviation: Ref, Reference category.

*p ≤ .1; **p ≤ .05; ***p ≤ .01.

More broadly, this indicates that more NSA activity at climate conferences does not necessarily increase the likelihood of NSAs actively participating in future COPs. This is a significant finding, because it means that if venues become more crowded, newcomers are more inclined to exit and discontinue attending subsequent COPs. As a result, the set of core players remains mostly intact, while much volatility takes place on the periphery of the NSA community. To put it differently, the policy participants at COPs resemble a network of a small and stable core of active participants, with a broad crowd looking on from the outside.

As a robustness check, we also tested models with an alternative operationalization of density (Table S4). That is, one might question why we consider sector‐related density and not overall density or the total number of NSAs mobilizing at a particular COP. Our density variable presumes that competition is mostly among those operating in the same area. Perhaps this approach is too narrow and a substantial part of the competition comes from actors active in other sectors. For instance, farming interests face competition from consumer organizations, the food industry, and environmental NGOs. In view of this, we tested an alternative model with an overall density measure. Table S4 illustrates that the results are highly similar and further supports our hypothesis that mobilization density at earlier COPs is a robust predictor of future attendance rates. Overall, this means that NSAs also respond to the overall density at a COP: when conferences become increasingly crowded, as they have done over the years, more NSAs decide to discontinue their transnational advocacy efforts. This confirms Hypothesis 1, namely that the denser NSA communities become, the more likely it is that newcomers remain policy tourists and do not attend subsequent conferences.

Our second contextual variable is the impact of the level of development of the country of origin (Hypothesis 2; see Model 2). The results clearly indicate that wealth affects the steadiness of advocacy positively, or reduces the chances that advocates will exit. This confirms the importance of resource availability in an NSA's national environment. In the core group, we find NSAs with their headquarters in countries such as the United States, Japan, Great Britain, Switzerland, Germany, and the Netherlands, while NSAs from developing countries—also large countries such as India, Brazil, and South Africa—are much less visible in the core group. The finding is in line with research observing a gap between the ability of groups from developed and developing countries to get involved in transnational advocacy (Beckfield, 2003; Hanegraaff et al., 2015; Nordang‐Uhre, 2014; Schroeder et al., 2012; Smith, 2005; Smith & Wiest, 2005; Steffek et al., 2007; Tallberg et al., 2013). The results indicate that NSAs from developed countries are not only more numerous, but also much more likely to develop a sustained advocacy presence at COPs.

Coming to the organizational characteristics, we hypothesized that encompassing interests are more likely to belong to the set of repeat players (Hypothesis 4), but that business interests have an advantage compared to nonbusiness interests (Hypothesis 3). All three models demonstrate that NGOs are more likely to sustain their advocacy efforts over longer periods of time compared to specialized business groups. NGOs have a greater chance of between 35% (global) and 55% (national) to remain active at COPs compared to specialized business associations. By contrast, however, there is no significant difference between NGOs and business associations. Both organizational types have an almost equal chance of exit (see almost identical odds ratios across all models compared to specialized business interests). It seems that the long time span of COP negotiations, combined with the need for issue linkages, “fits” better with encompassing NSAs.

This result is in line with the assertion that global negotiation forums are better suited to encompassing groups. Hence, it is not so much the nature of the represented interest, in terms of economic versus social interests, which determines whether or not NSAs stay active at COPs, but rather the encompassing nature of interest representation. Overall, we thus reject Hypothesis 3, which stated that business interests develop a steadier lobbying presence, but confirm Hypothesis 4 on the scope of representation. In light of research emphasizing the dominance of business groups in global governance (e.g., Berkhout et al., 2018; Berkhout & Lowery, 2011; Nordang‐Uhre, 2014), this is an important finding. It indicates that, although business groups are on an aggregate level more active at the global level (compared to NGOs), they are also more likely to discontinue their efforts. Instead, we observe a clear difference between specialized NSAs on the one hand, and NSAs with a more encompassing orientation (such as business associations and NGOs) on the other hand. Here, one can mention the core players active at most COPs, such as the ICC, BusinessEurope, the WBCSD, Greenpeace, and CAN. Typical is that these groups, especially combined, represent the demands of large societal segments. Even groups such as the International Gas Union, while more specific than the ICC or BusinessEurope, have a more representative voice than the individual gas companies. Individual gas companies such as the California Gas Company, Unigas (the Netherlands), the Italian Gas Association, and Metrogas S.A. (Argentina) each attended only one conference. Of course, some multinational corporations, such as Shell, ExxonMobil, and Statoil are regular attendees, but these companies represent a rather small subset of the core players, that is, the regular participants who are mostly business associations and NGOs.

As for the control variables, distance is an important predictor of the observed exit rates. The farther away from an NSA's country of origin the next COP is held, the less likely it is that the NSA will be present at that COP (Nordang‐Uhre, 2014). However, there is no link between the alignment of issues on the agendas of the NSAs and the agenda of the COP or the democratic nature of the country of origin. Especially the latter is somewhat surprising, as it means that NSAs from autocratic states are as likely to remain active as organizations from democratic states (see also Rohrschneider & Dalton, 2002, p. 528; Barry et al., 2015; Smith & Wiest, 2005; Tallberg et al., 2014). Finally, our results indicate that media salience (both with regard to the global sources and the more specific country sources) has no effect on sustained activity. Whether there is much or limited attention in the media for a particular COP, the chance that NSAs will exit at the next COP remains the same.

6. CONCLUSION

The opening up of IOs to NSAs is sometimes considered one of the most significant global developments of the past 20 years (Bexell et al., 2010; Tallberg et al., 2014). The magnitude of this phenomenon has led several scholars to ascribe considerable influence to NSAs. However, when we analyze the extent to which NSAs maintain a steady lobbying presence, we observe that the advocacy community at COPs is highly volatile. While our evidence prevents us from making strong claims about power and influence, the observed volatility suggests that the magnitude of transnational advocacy should not be overstated. Almost 65% of the identified NSAs attended only one conference and over three quarters attended no more than two conferences. A mere 0.3% of NSAs attended all conferences, while less than 4% attended more than nine out of the 16 climate conferences. This finding further challenges the inclusive nature of global governance. It is true, as various studies have demonstrated, that a large and diverse array of NSAs are active at the global level (Beckfield, 2003; Hanegraaff, 2015b; Nordang‐Uhre, 2014; Schroeder et al., 2012; Smith, 2005; Smith & Wiest, 2005; Steffek et al., 2007), but in many instances NSA participation is highly unstable and volatile.

At the very least, these observations justify the starting point for our analysis: aggregate numbers can be misleading, providing a distorted assessment of transnational advocacy. Instead, we have demonstrated that transnational NSA communities consist of a small core of active policy participants, while most other NSAs are peripheral players, indicated by a lack of repeat appearances over the course of 15 years. Hence, our results reflect an interesting paradox. If international arenas become more densely populated with transnational advocates, the awkward effect is that the volatile nature of their presence increases and that the number of repeat players remains relatively low, while the number of policy tourists increases. This is very much in line with what Salisbury observed a long time ago with respect to lobbying in Washington: “[A]s the overall number of groups in Washington increases, the chances of each individual organization diminishes by the increased competition over political attention” (1990, p. 203). We observe a similar trend with respect to transnational climate advocacy, highlighting that the openness of the UNFCCC toward NSAs does not necessarily lead to a larger and stable set of mobilized stakeholders.

These findings are highly relevant given some ongoing debates in the literature. One concern associated with the opening up of IOs has always been who would benefit the most from these developments. Does it empower deprived actors to (somewhat) balance the power in global politics, or does it “reproduce[s] existing inequalities” in which developed nations and specialized economic interests still benefit the most (Zürn, 2014, p. 62)? Initially there was some optimism surrounding the inclusion of NSAs in global affairs, as this would strengthen the position of deprived actors and countries in global politics (for an overview of this argument, see Scholte, 2004, p. 233; Bexell et al., 2010, p. 87; Tallberg et al., 2018). Yet, since then several empirical studies have demonstrated that this might be more wishful thinking than reality, as NSAs from developed countries and business interests increasingly outnumber NSAs from developing nations (Beckfield, 2003; Hanegraaff, 2015a; Nordang‐Uhre, 2014; Schroeder et al., 2012; Smith, 2005; Smith & Wiest, 2005; Steffek et al., 2007; Tallberg et al., 2013).

Our findings, on the one hand, confirm the findings of the aforementioned studies. That is, most organizations from less developed countries seem to face considerable constraints in developing long‐lasting patterns of advocacy. Yet, on the other hand, our results also provide a more optimistic view, because NGOs are among the most likely to become repeat players at COPs. This is a relevant finding, because it means that NGOs apparently compensate for their lack of numerical activity by repeat attendance of a large number of COPs. As stated before, this should provide them with substantial opportunities to interact with policymakers and, as such, become key players in global climate policymaking.

Obviously, we need to remain careful and avoid inferring too broad implications based on this study, especially because we focused on only one venue. Yet there are several reasons to suspect that density is an important driver for the dynamics in NSA communities. For instance, studies in the United States (Lowery & Gray, 1996) and in the European Union (Berkhout & Lowery, 2011) have indicated that it is harder for some groups to sustain their advocacy efforts in a dense community compared to environments with little competition for resources and political attention. The fact that our results are generally consistent with these studies indicates the robustness of our conclusions. For other explanatory factors, such as group type or national wealth of the country of origin, it is less clear how these travel to global settings in other areas. The climate conferences are one of the most open global venues for NSAs. Moreover, there are few formal procedures structuring the interactions between NSAs and policymakers. This is one of the reasons that we see such a densely populated venue with so much volatility. Other international venues are far less populated, with possibly different levels of volatility (Hanegraaff, 2015a). Another potential difference relates to the actor types that are allowed to attend global diplomatic conferences. At other venues, for instance the WTO, it is much more difficult for firms to attend the global conferences, which may affect the dynamic of the competition between business and nonbusiness interests. Indeed, previous research has indicated mixed results as to whether or not business groups or NGOs have better chances to maintain their advocacy activities (see Halpin & Jordan, 2012, for discussions on this issue). An interesting implication might be that increased openness of global political venues could lead to more inequality than mobilization at venues where access is more exclusive and managed by the IO staff (see also Beyers & Arras, 2019). While we cannot test this hypothesis with the data at hand, this is certainly a relevant avenue for future comparative research.

Our main takeaway message is that we must critically assess the perceived openness of IOs as well as the mobilization of NSAs. While most analysts—including us—have found that global governance is densely populated with NSAs (Beckfield, 2003; Steffek et al., 2007; Tallberg et al., 2013, 2014), this observation could lead to the misguided conclusion that NSAs play a vital role in global policymaking. Our study provides a cautionary note against leaping to such far‐reaching conclusions by illustrating how, despite the plethora of political activity that surrounds climate conferences, many NSAs have not been repeatedly involved in the policymaking process. As such, we hope that our study leads to more research highlighting variation in transnational advocacy. While the many mapping studies are valuable to demonstrate the extensive nature of transnational advocacy, as Halpin and Jordan (2012, p. 246) have pointed out, advocacy “is not basketball”; not every group is equally important, equally heard, equally active, and certainly not equally influential. In this study, we structured an interest community by differentiating advocacy activity over time, yet other options are possible. For instance, one may analyze the access NSAs gain to policymakers; one may compare variations across multiple venues; or one may assess the extent to which NSAs work on multiple issues at once. Key is that we analyze NSA communities as vibrant communities with some hierarchical order, in which core and peripheral NSAs play different roles in and have varying effects on global policymaking.

Supporting information

Appendix S1: Supplementary Information

Click here for additional data file. (61.1KB, docx)

ACKNOWLEDGMENTS

We are grateful to Jonas Tallberg, David Lowery, and the anonymous reviewers for their helpful suggestions. Earlier versions of the article were presented at the “ECPR Joints Sessions Conference”, 2015, Warsaw, and the “Pressure Groups Conference,” 2017, Princeton University. We thank all the participants of these conferences for their valuable comments. Finally, we would also like to thank the Research Foundation‐Flanders (Odysseus Program, project number G.0908.09) and the European Research Council (ERC‐2013‐CoG 616702‐iBias) for their financial contributions.

Hanegraaff M, Vergauwen J, Beyers J. Should I stay or should I go? Explaining variation in nonstate actor advocacy over time in global governance. Governance. 2020;33:287–304. 10.1111/gove.12427

Funding information H2020 European Research Council, Grant/Award Number: ERC‐2013‐CoG 616702‐iBias; Research Foundation‐Flanders, Grant/Award Number: Odysseus Programme project number G.0908.09

DATA ACCESSIBILITY

Replication materials are available in the dataverse‐account of the corresponding author at https://dataverse.harvard.edu/.

REFERENCES

  1. Abbott, K. , Green, J. , & Keohane, R. (2016). Organizational ecology and institutional change in global governance. International Organization, 70, 247–277. [Google Scholar]
  2. Agné, H. , Delmuth, L. M. , & Tallberg, J. (2015). Does stakeholder involvement foster democratic legitimacy in international organizations? An empirical assessment of a normative theory. Review of International Organizations, 10, 465–488. [Google Scholar]
  3. Barry, C. M. , Bell, S. R. , Clay, K. C. , Flynn, M. E. , & Murdie, A. (2015). Choosing the best house in a bad neighborhood: Location strategies of human rights INGOs in the non‐Western world. International Studies Quarterly, 59, 86–98. [Google Scholar]
  4. Baumgartner, F. R. , Berry, J. M. , Hojnacki, M. , Kimball, D. C. , & Leech, B. L. (2009). Lobbying and policy change: Who wins, who loses, and why. Chicago: University of Chicago Press. [Google Scholar]
  5. Beckfield, J. (2003). Inequality in the world polity: The structure of international organization. American Sociological Review, 68, 401–424. [Google Scholar]
  6. Berkhout, J. , Beyers, J. , Braun, C. , Hanegraaff, M. , & Lowery, D. (2018). Making inference across mobilisation and influence research: Comparing top‐down and bottom‐up mapping of interest systems. Political Studies, 66(1), 43–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Berkhout, J. , Hanegraaff, M. , & Braun, C. (2017). Is the EU different? Comparing the diversity of national and EU‐level systems of interest Organisations. West European Politics, 40(5), 1109–1131. [Google Scholar]
  8. Berkhout, J. , & Lowery, D. (2011). Short‐term volatility in the EU interest community. Journal of European Public Policy, 18, 1–16. [Google Scholar]
  9. Betsill, M. M. , & Corell, E. (2001). NGO influence in international environmental negotiations: A framework for analysis. Global Environmental Politics, 1, 65–85. [Google Scholar]
  10. Betsill, M. M. , & Corell, E. (2008). NGO diplomacy: the influence of nongovernmental organizations ininternational environmental negotiations. Cambridge, MA: Mit Press. [Google Scholar]
  11. Bexell, M. , Tallberg, J. , & Uhlin, A. (2010). Democracy in global governance: The promises and pitfalls of transnational actors. Global Governance, 16, 81–101. [Google Scholar]
  12. Beyers, J. , & Arras, S. (2019). Who feeds information to regulators? Stakeholder diversity in European Union regulatory agency consultations. Journal of Public Policy. 10.1017/S0143814X19000126 [DOI] [Google Scholar]
  13. Binderkrantz, A. , Christiansen, P. M. , & Pederson, H. H. (2015). Interest group access to the bureaucracy, parliament, and the media. Governance, 28, 95–112. [Google Scholar]
  14. Bouwen, P. (2002). Corporate lobbying in the European Union: The logic of access. Journal of European Public Policy, 9, 365–390. [Google Scholar]
  15. Browne, W. P. (1985). Variations in the behavior and style of state lobbyists and interest groups. The Journal of Politics, 47, 450–468. [Google Scholar]
  16. Davis, C. (2004). International institutions and issue linkage: Building support for agricultural trade liberalization. American Political Science Review, 98, 153–169. [Google Scholar]
  17. Dellmuth, L. M. , & Tallberg, J. (2015). The social legitimacy of international organizations: Interest representation, institutional performance, and confidence extrapolation in the United Nations. Review of International Studies, 41, 451–475. [Google Scholar]
  18. Dür, A. , & Mateo, G. (2016). Insiders versus outsiders: Interest group politics in multilevel Europe. Oxford: Oxford University Press. [Google Scholar]
  19. Falkner, R. , Stephan, H. , & Vogler, J. (2010). International climate policy after Copenhagen: Towards a “building blocks” approach. Global Policy, 1, 252–262. [Google Scholar]
  20. Fraussen, B. (2014). The visible hand of the state: On the organizational development of interest groups. Public Administration, 92, 406–421. [Google Scholar]
  21. Gray, V. , & Lowery, D. (1996). A niche theory of interest representation. The Journal of Politics, 58, 91–111. [Google Scholar]
  22. Gray, V. , & Lowery, D. (1997). Life in a niche: Mortality anxiety among organized interests in the American states. Political Research Quarterly, 50, 25–47. [Google Scholar]
  23. Gray, V. , & Lowery, D. (1998). To lobby alone or in a flock: Foraging behavior among organized interests. American Politics Quarterly, 26(1), 5–34. [Google Scholar]
  24. Hadden, J. (2015). Networks in contention: The divisive politics of climate change. Cambridge, MA: Cambridge University Press. [Google Scholar]
  25. Halpin, D. (2011). Organized interests and cascades of attention: Unpacking policy bandwagon dynamics. Governance, 24, 205–230. [Google Scholar]
  26. Halpin, D. , & Fraussen, B. (2017). Conceptualizing the policy engagement of interest groups: Involvement, access and prominence. European Journal of Political Research, 56, 723–732. [Google Scholar]
  27. Halpin, D. , & Jordan, G. (2012). The scale of interest Organization in Democratic Politics: Data and research Methods. New York: Palgrave Macmillan. [Google Scholar]
  28. Halpin, D. , & Thomas, H. (2012). Interest group survival: Explaining sources of mortality anxiety. Interest Groups & Advocacy, 1, 215–238. [Google Scholar]
  29. Hanegraaff, M. (2015a). Transnational advocacy over time: Business and NGO mobilization at UN climate summits. Global Environmental Politics, 15, 83–104. [Google Scholar]
  30. Hanegraaff, M. (2015b). Interest groups at transnational negotiation conferences: Goals, strategies, interactions, and influence. Global Governance: A Review of Multilateralism and International Organizations, 21(4), 599–620. [Google Scholar]
  31. Hanegraaff, M. , Braun, C. , De Bièvre, D. , & Beyers, J. (2015). The domestic and global origins of transnational advocacy: Explaining lobbying presence during WTO ministerial conferences. Comparative Political Studies, 48, 1591–1621. [Google Scholar]
  32. Hannan, M. , & Freeman, J. (1989). Organizational ecology. Cambridge, MA: Harvard University Press. [Google Scholar]
  33. Heinz, J. , Laumann, E. O. , Nelson, R. , & Salisbury, R. (1993). The hollow core: Private interests in national policy making. Cambridge, MA: Harvard University Press. [Google Scholar]
  34. Jinnah, S. (2011). Climate change bandwagoning: The impacts of strategic linkages on regime design, maintenance, and death. Global Environmental Politics, 11, 1–9. [Google Scholar]
  35. LaPira, T. , Herschel, T. , & Baumgartner, F. (2014). The two worlds of lobbying: Washington lobbyists in the core and on the periphery. Interest Groups & Advocacy, 3, 219–245. [Google Scholar]
  36. Lowery, D. , & Gray, V. (1996). The population ecology of interest representation: Lobbying communities in the American states. Ann Arbor: University of Michigan Press. [Google Scholar]
  37. Lowery, D. (2007). Why do organized interests lobby? A multi‐goal, multi‐context theory of lobbying. Polity, 39, 29–54. [Google Scholar]
  38. Lowery, D. (2013). Lobbying influence: Meaning, measurement and missing. Interest Groups & Advocacy, 2, 1–26. [Google Scholar]
  39. Lowery, D. , & Gray, V. (2004). Bias in the heavenly chorus: Interests in society and before government. Journal of Theoretical Politics, 16, 5–30. [Google Scholar]
  40. Lowery, D. , Gray, V. , & Fellowes, M. (2005). Sisyphus meets the Borg economic scale and inequalities in interest representation. Journal of Theoretical Politics, 17, 41–74. [Google Scholar]
  41. Marshall, M.G. , Gurr, T.R. & Jaggers, K. (2018). Polity IV: Political Regime Characteristics and Transitions, 1800–2017. Retrieved from http://www.systemicpeace.org/inscrdata.html.
  42. Miller, A. R. , & Dolšak, N. (2007). Issue linkages in international environmental policy: The international whaling commission and Japanese development aid. Global Environmental Politics, 7, 69–96. [Google Scholar]
  43. Molina, O. , & Rhodes, M. (2002). Corporatism: The past, present, and future of a concept. Annual Review of Political Science, 5, 305–331. [Google Scholar]
  44. Nordang‐Uhre, A. (2014). Exploring the diversity of transnational actors in global environmental governance. Interest Groups & Advocacy, 3, 59–78. [Google Scholar]
  45. Nownes, A. (2010). Density dependent dynamics in the population of transgender interest groups in the United States, 1964‐2005. Social Science Quarterly, 91, 689–703. [Google Scholar]
  46. O'Brien, R. , Goetz, A. M. , Scholte, J.‐A. , & Williams, M. (2000). Contesting global governance: Multi‐lateral economic institutions and global social movements. New York: Cambridge University Press. [Google Scholar]
  47. Rasmussen, A. , & Carroll, B. J. (2014). Determinants of upper‐class dominance in the heavenly chorus: Lessons from European Union online consultations. British Journal of Political Science, 44, 445–459. [Google Scholar]
  48. Rasmussen, A. , & Gross, V. (2015). Biased access? Exploring selection to advisory committees. European Political Science Review, 7, 343–372. [Google Scholar]
  49. Rohrschneider, R. , & Dalton, R. (2002). A global network? Transnational cooperation among environmental groups. Journal of Politics, 64, 510–533. [Google Scholar]
  50. Salisbury, R. (1990). The paradox of interest groups in Washington: More groups, less clout In King A. (Ed.), The new American political system (pp. 203–230). Washington, DC: American Enterprise Institute. [Google Scholar]
  51. Scholte, J.‐A. (2004). Civil society and democratically accountable global governance. Government and Opposition, 39, 211–233. [Google Scholar]
  52. Schroeder, H. , Boykoff, M. , & Spiers, L. (2012). Equity and state representations in climate negotiations. Nature Climate Change, 2, 834–836. [Google Scholar]
  53. Skocpol, T. (2004). Voice and inequality: The transformation of American civic democracy. Perspectives on Politics, 2, 3–20. [Google Scholar]
  54. Smith, J. (2005). Globalization and transnational social movement organizations. New York: Cambridge University Press. [Google Scholar]
  55. Smith, J. , & Wiest, D. (2005). The uneven geography of global civil society: National and global influences on transnational association. Social Forces, 84, 621–652. [Google Scholar]
  56. Steffek, J. , Kissling, C. , & Nanz, P. (2007). Civil society participation in European and global governance. New York: Palgrave Macmillan. [Google Scholar]
  57. Strolovitch, D. Z. (2007). Affirmative advocacy: Race, class, and gender in interest group politics. London: Chicago University Press. [Google Scholar]
  58. Tallberg, J. , Dellmuth, L. M. , Agné, H. , & Duit, A. (2018). NGO influence in international organizations: Information, access and exchange. British Journal of Political Science, 48, 213–238. [Google Scholar]
  59. Tallberg, J. , Sommerer, T. , Squatrito, T. , & Jönsson, C. (2013). The opening up of international organizations: Transnational access in global governance. Cambridge, MA: Cambridge University Press. [Google Scholar]
  60. Tallberg, J. , Sommerer, T. , Squatrito, T. , & Jönsson, C. (2014). Explaining the transnational design of International Organizations. International Organization, 68, 741–774. [Google Scholar]
  61. Tarrow, S. (2005). The new transnational activism. Cambridge, MA: Cambridge University Press. [Google Scholar]
  62. Willetts, P. (2006). The Cardoso report on the UN and civil society: Functionalism, global corporatism, or global democracy? Global Governance, 12, 305–324. [Google Scholar]
  63. Zürn, M. (2014). The politicization of world politics and its effects: Eight propositions. European Political Science Review, 6, 47–71. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1: Supplementary Information

Click here for additional data file. (61.1KB, docx)

Data Availability Statement

Replication materials are available in the dataverse‐account of the corresponding author at https://dataverse.harvard.edu/.

Articles from Governance (Oxford, England) are provided here courtesy of Wiley

RESOURCES