Assessing the interactions of interest organizations and coalitions : mobilization, participation and balance : a case study of Climate Action Network Europe

(1)

Graduate School of Social Sciences MSc Public Policy & Governance

Master Thesis Academic Year 2016-2017

Assessing the Interactions of Interest Organizations and

Coalitions: Mobilization, Participation and Balance

A case study of Climate Action Network Europe

Word Count: 21000

Supervisor: Conny Roggeband Second reader: Joost Berkhout

Submission date: 23 June 2017

By Tariq RAMI 11243023

(2)

“Unity makes strength

1

”

1_{Interview CAN Europe, 2017}

(3)

Acknowledgements

The research presented in this thesis was made possible thanks to my supervisor Dr. Conny Roggeband. I am grateful for the guidance, the support and the feedback she provided me throughout the research project. In the same vein, I express my gratitude to Dr. Joost Berkhout for his teachings and for accepting to read my thesis.

Moreover, I would like to thank my respondents for their willingness to take part in the study and for the insights they provided me during my research. More specifically, Wendel Trio for allowing me to attend the general assembly of CAN Europe; Pierre-Jean Brasier for helping me to better understand the internal dynamics of advocacy coalitions.

I cannot leave without mentioning my family, without whom I would not be where I am today. I deeply appreciate the trust they have always given to me, that has contributed to my personal fulfilment.

Particular thanks to my brother Alexandre Marcou with whom I underwent this long journey, and who backed me during the best and worst moments.

Finally, I would like to gratefully acknowledge Dad, Anne Van Doesburg and the NCs who have always offered me with a haven of peace and relaxation.

(4)

Abstract

This paper explores the internal interactions between interest organizations and coalitions. First, it seeks to explain the process of mobilization and maintenance for interest organizations joining coalitions. The literature of interest groups and social movements provides relevant arguments to explain this first phase of collective action for individuals joining organizations. I use these arguments to explain the concept of “coalition building”. Moreover, the study considers the role of member organizations in framing collective issues and reaching the consensus within coalitions. These two processes are intimately related since the participation of members in framing issues impacts the process of consensus building. Finally, I provide arguments explaining how coalition members balance between their individual and collective actions. All combined, these elements contribute to explaining the internal dynamics between coalitions and their members. Indeed, while scholarship recognizes the advantages of advocating collectively, the interactions between coalitions and their members have not been studied through the lens of the collective action literature. I fill this gap. Empirically, the analysis mainly draws on interview data with eight interest organization representatives that are members of the Climate Action Network, Europe’s largest coalition working on climate and energy issues. The study demonstrates that selective benefits perceived by members—notably purposive and solidary incentives—are key determinants to explain coalition building. Moreover, when framing issues and strategies, coalitions involve their members’ positions and strive to reach consensus regarding their common interests, strategies and priorities. In turn, members balance their participation in these processes according to their resources and capacities.

Keywords: Interest organizations; coalitions; transnational advocacy; coalition building; issue framing; consensus building; balance

(5)

List of Abbreviations

CAN — Climate Action Network CMW—Carbon Market Watch EC—European Commission EP—European Parliament EU—European Union

ETS—Emission Trading Scheme ESD—Effort Sharing Decision GA – General Assembly

NGO—Non-governmental organization WWF—Worldwide Fund for Nature

(7)

1. Introduction

Interest organizations have been studied for decades, and scholars have considered their environment, institutional structure, and organizational characteristics (see Lowery & Brasher, 2003; Baroni et al., 2014). Generally, interest groups are viewed as strategic organizations competing with other groups for their survival and to push their policy preferences onto the political landscape (Sorurbakhsh, 2016). The European Union (EU) offers a broad and favourable political opportunity structure for interest organizations (Ibid.). Undeniably, the multiple layers of governance and the relatively high fragmentation of the EU institutions and policy-making processes provide a plurality of access points and venues (Klüver, 2013). Despite this, interest groups have not always been able to equally exploit their access, and empirical studies have tried to use this element to explain a recognized imbalance in influence (see Lowery & Gray, 2004). In addition, scholars generally draw on two main aspects to study interest groups’ influence: interest groups properties and issue-specific features (Baroni et al. 2014). The former refer to explanatory factors such as the type of interests represented and the level of resources (Ibid.). These characteristics are useful to distinguish different types of interest organizations. On the other hand, issue-specific factors—such as salience, degree of conflict and complexity of issues—have also been used to provide arguments explaining the variation of access and influence of interest groups. However, empirical studies tend to focus on specific group properties or issue-specific contexts while not tackling the bigger picture (see Dür and De Bièvre, 2007; Dür, 2008).

Meanwhile, scholars acknowledge advocacy coalitions as actors affecting the success of advocacy strategies of organizations in function of the issues at stake (e.g. Baumgartner et al., 2009). Indeed, lobbying is often described as a collective enterprise (see Baumgartner and Leech, 1998; Klüver, 2013). Interest groups regularly interact with other parties through institutionalized entities such as networks, umbrella organizations, and coalitions. In fact, they coalesce into coalitions in order to advocate collectively and achieve their common interests. In the EU, much lobbying takes place in coalitions of organizations composed of large and resourceful groups, and smaller—less resourceful ones (Baroni et al., 2014). In the literature, the characteristics and strategies of advocacy coalitions have been sufficiently documented (see Yanacopulos, 2005; Levi & Murphy, 2006; Klüver, 2013). However, while scholars encompass the advantages of collective advocacy, less had been said regarding the internal dynamics

(8)

between interest organizations and coalitions (for exceptions, see Sorurbakhsh, 2016; Sabatier, 2016).

Consequently, the purpose of this research is to analyze the interactions between interest organizations and advocacy coalitions through the scope of the collective action literature. Despite the fact that interest organizations and coalitions have been studied extensively although separately, it seems that focusing on coalitions’ internal dynamics have both theoretical and practical relevance. Theoretically, it seems relevant to study coalitions using the theories of interest groups and social movements. Indeed, coalitions composed by individual interest organizations can be studied in the same fashion as for interest groups composed by rational and self-interested individuals. Further, I provide arguments supporting the importance of this research from a theoretical viewpoint. Practically, the empirical study aims to highlight organizational challenges and opportunities for interest organizations having limited knowledge regarding the advantages of advocating in coalitions. Also, the findings provide coalitions with additional awareness regarding their internal dynamics. Ultimately, my contribution to the literature is to provide a comprehensive answer to the following research question: “How interest organizations join, frame issues and build consensus as part of advocacy coalitions? How do they balance their individual and collective actions?”

This question encompasses several concepts that I briefly introduce for more clarity. First, this research focuses on the interactions between the individual and collective actors. That is, I focus on interest organizations that are part of lobbying coalitions. Furthermore, a division of this research question is required in order to recognize the links between the different concepts. Accordingly, I first consider the selective benefits perceived by members as key determinants to explain the concept of “coalition building”. Theoretically, I use the theories explaining the mobilization and maintenance of interest groups and transpose them to coalitions. After having identified the different selective incentives perceived by members, I specify their impact in the process of joining coalitions and acting collectively. In a second step, I discuss the role of coalition members in shaping collective frames and strategies—the concept of “issue framing”. According to the literature, one of the main objectives for coalitions is to develop common interests and strategies to pursue their goals. Subsequently, I address the process of building consensus in coalitions, which is directly linked to the framing of issues and strategies. Indeed, this phase consists of reaching agreements on the collective frames previously developed. Finally, I aim to provide insights explicating the balance between individual and collective actions undertaken by coalition members.

(9)

The empirical study relies on a single case study of CAN Europe—a coalition acting at the EU level on climate and energy issues. To collect data, I conducted eight semi-structured expert interviews with non-governmental organizations (NGOs) that are part of this coalition. Moreover, I participated in the general assembly of CAN Europe in Brussels. Lastly, the findings draw on documents collected from the members and the coalition. In sum, this qualitative research design is relevant to assess the interactions between member organizations and coalitions.

This paper is organized into three main sections. In the theoretical framework, I present the main concepts used throughout my research based on previous research. Subsequently, I develop my theoretical expectations. In a second part, I present the characteristics of the research design. This section aims at detailing and justifying the methods and data used to answer the research question. Next, I proceed to the actual analysis of the findings. This section represents the cornerstone of my research—it intends to assess the different components under study. Finally, the final section provides a meaningful answer of the research question and my contribution to the literature. I also address the shortcomings of this study and provide my projection for the direction of future research, which I hope will benefit from further theoretical and empirical work.

(10)

2. Theoretical Framework

In this first section, I present the theoretical aspects encompassed in this research. In a first step, I detail the relevant elements defining interest organizations and coalitions. Subsequently, I link these two actors using the theories of collective action. This provides theoretical arguments regarding the interactions and common characteristics between these two actors. Notably, I find support to explain the internal dynamics of coalitions through the scope of collective action theories. Then, I offer an overview of the literature with regards to the concepts of issue framing and consensus building. Finally, I consider how the literature explains the balance between individual and collective actions for coalition members. I find no clear theoretical support regarding this aspect of my research. Therefore, I rely on a conceptual framework explaining the balance of strategies and extrapolate. This represents the theoretical contributions of my research. Finally, I articulate the different assumptions drew from the literature and that I will test and confront to the empirical data collected.

2.1 Definitions of the actors: interest organizations and coalitions

Interest groups

Interest groups and social movement studies generally cover a large variety of topics and scholars use different definitions and concepts to talk about the same thing. I herein offer a selective overview of the literature surrounding the definition of interest organizations and coalitions. This matters because while

“different approaches to conceptualization may be motivated by differences in research focus, they often make it difficult to draw comparisons between the findings generated from different studies of interest representation” (Baroni et al., 2014, p. 142).

First, a part of the literature defines interest organizations according their behavioural aspect and organizational characteristics (Baroni et al., 2014). The first aspect points to the policy-related activities of interest groups which have a long history in the interest groups literature. Accordingly, an interest group can be defined as any group “acting, or tending toward action” (Bentley, 1908, p. 211); “making certain claims upon other groups in society” (Truman, 1951, p. 37); “seeking to influence policy” (Lindblom, 1977, p. 85) or “the formulation and implementation of public policy” (Grant, 1989, p. 9). Other authors have

(11)

stressed the challenge of delimiting the boundaries of this behavioural definition (see Wilson, 1990; Gray and Lowery, 1996; Baumgartner et al., 2009). For instance, Wilson (1990) has outlined the difficulty of measuring the political activity necessary for an organization to be considered as an interest group (Ibid.). Nevertheless, scholars generally see interest groups as organizations acting to achieve political influence, which highlight the political objectives and activities of such organizations (see Binderkrantz et al., 2015; Baumgartner et al., 2009).

The second aspect refers to organizational characteristics as key defining features (Baroni et al., 2014). This facet corresponds to how interest groups are composed based on criteria such as the “membership structure, level of mobilization, staff and financial resources” (Ibid. pp. 143–144). Within this approach, the term “interest group” had been reserved for membership-based organizations (Ibid.). While certain studies have focused on the dynamics of group membership and mobilization (see Jordan & Maloney, 1998; De Bruycker et al. 2016), others have concentrated on the lobbying strategies and their outcomes (for example, Binderkrantz and Krøyer, 2012; Hanegraff et al. 2016). Nonetheless, this approach has also shown some limitations, specifically potential normative biases because of the focus on groups presented as “voluntary, democratically accountable and individual-based organizations” (Baroni et al., 2014, p. 144). Consequently, it seems necessary to prevent potential biases of selection by specifying my focus on organizations representing a certain type of interest, regardless of their organizational characteristics.

As far as interest organizations are concerned, another aspect of the definition must be highlighted. That is, there is a distinction in the nature of the interest being represented. Indeed, a contrast is generally acknowledged by scholars between specific and diffuse interests (for example, Beyers, 2004; Klüver, 2013; De Bruycker et al., 2016). While diffuse interest organizations represent and promote the interests of broad segments of society and issues of general public concern, specific interests defend and represent well-circumscribed constituencies (Beyers, 2004). Similarly, Klüver (2013) speaks of narrow “sectional groups” and broad ‘cause groups”. Yet, empirical studies often fail to agree upon the list of interest organizations falling under these categories. Scholars tend to build typologies based on the different aspects presented above, depending on their specific focus of research. Notably, Baumgartner et al. (2009) add coalitions, think tanks and governmental associations to the list. Therefore, the problem with such an approach is a lack of theoretical justification for the inclusion or exclusion of groups.

In their study, Binderkrantz, Christiansen and Pedersen (2015) define interest groups as membership organizations ‘working to obtain political influence’ (Ibid. p. 97). Following

(12)

this approach, Lowery and Brasher (2003) argue that ‘voluntary’ organizations can be defined by their membership. They divide this feature into two subtypes of organizations, membership groups that count on individual citizens as members, and associations that enroll institutions as members (Ibid. p. 5). Accordingly, these organizations are organized to represent their constituents. That is, indifferently whom they represent—general public, institutions or companies—voluntary organizations rely on members for their own existence (Ibid.). This has important implication in my research. Indeed, when it comes to the mobilization and maintenance of interest organizations, members are considered as rational and self-interested. This aspect is considered in the next subsection of this theoretical framework.

A last element of definition is offered by authors such as Lowery and Brasher (2003) and Maloney (2008). They propose three conditions to encompass the term interest organizations. To be considered as such, groups must encompass the following features: organization, political interests, and private status (Ibid.). First, interest groups must draw on some sort of organization. This feature relates to the nature of the group and excludes unorganized broad movements such as social movements. The second feature relates to the political interest. Actors must pursue the objective of influencing political decision-making and shaping policy outcomes. Finally, the last characteristic that must be present to talk about interest organization is private status. It refers to the fact that interest groups are not seeking public office and are not public institutions. Therefore, this last characteristic differentiates interest groups from political parties (Klüver, 2013; Lowery & Brasher, 2003).

In brief, these defining elements are a necessary to understand the organizational features of interest organizations. Next, I consider the definitions of advocacy coalitions. To do so, I first introduce the transnational advocacy network framework before identifying the specific characteristics of advocacy coalitions.

Coalitions

The fact that non-state actors interact and form networks received a great deal of attention during the last decade. The influential work of Keck and Sikkink (1998) provided a framework for exploring transnational networks in many issue areas, including the environment, the women’s movement and human rights. Considering how interactions in international politics are structured in networks, the authors distinguish between: (1) networks composed by economic actors and firms; (2) networks of scientists and experts whose professional ties and ideas underpin their efforts to influence policy; (3) networks of activists

(13)

sharing principled ideas and values motivating their formation (Keck & Sikkink, 1998). In that sense, their concept of transnational advocacy network (TAN) is used to describe a network that

‘transcends national boundaries and that consists of members who are bound by shared values, a common discourse, and dense exchanges of information and services’ (Ibid., p. 89).

By building new links among actors in civil societies, states, and international organizations, transnational advocacy networks ‘multiply the opportunities for dialogue and exchange’ (Ibid. p. 89). Further, they describe networks as forms of organization that are characterized by ‘voluntary, reciprocal and horizontal patterns of communication and exchange’ (Ibid. p. 91). Therefore, the concept of network emphasizes the fluid and open relations among ‘committed and knowledgeable actors working in specialized issue areas’ (Ibid.). Moreover, they justify the usage of ‘advocacy networks’ because it is already used by the actors themselves:

‘over the last two decades, individuals and organizations have consciously formed and named networks, developed and shared networking strategies and techniques, and assessed the advantages and limits of this kind of activity’ (Ibid. p. 90).

Another point explaining why the authors refer to advocacy networks is because advocates, ‘plead the causes of others or defend a cause or proposition; they are stand-ins for persons and ideas’ (Keck & Sikkink, 1998, p. 91). In that sense, ‘advocacy’ underlines the distinctiveness of these transnational networks

‘they are organized to promote causes, principled ideas and norms, and often involve individuals advocating policy changes that cannot be easily linked to their interests’ (Ibid.).

Last but not least, they argue that advocacy networks have been particularly important in value-laden debates over human rights, the environment, women, infant health and indigenous peoples (Ibid.). Nonetheless, they stress that researchers came late into the debate: ‘scholars have been slow to recognize either the rationality or the significance of activist networks’ (Ibid. p. 89).

Regarding the composition of these TANs, Keck and Sikkink (1998) support that advocacy networks may include a plethora of actors such as: international and domestic NGOs, local social movements, foundations, the media, churches, parts of regional and international intergovernmental organizations and part of governments. While not all these different actors have to be present in each advocacy network, emphasis is made on the central role of

(14)

international and domestic non-governmental organizations (NGOs) in most advocacy networks. In effect, they usually initiate actions, introduce new ideas, provide information, lobby for policy changes, and pressure more powerful actors to take positions (Ibid.). Tarrow (1994) adds that organizations and individuals within advocacy networks are political entrepreneurs which mobilize resources like information and membership, and show a sophisticated awareness of the political opportunity structures within which they operate (Ibid). While the seminal TAN framework developed by Keck and Sikkink has served as a starting point for many researchers, I hereafter present a distinction between ‘network’ and ‘coalition’. I do so in order to clarify my use of the term ‘coalition’. In effect, the case study of my research—CAN Europe—is a network of NGOs that describes itself as Europe’s largest

coalition working on climate and energy issues. Accordingly, throughout my thesis I will

emphasize on the concept of coalition. According to Yanacopulos (2005), the differences between these two organizational forms are ‘both subtle and important’ (Ibid. p. 95).

Notably, Keck and Sikkink (1998) specify their usage of the term ‘networks’—rather than coalitions, movements or civil society—to evoke the structured and structuring dimension in the actions of these actors (Ibid.). Alternatively, Yanacopulos (2005) claim that coalitions usually form more permanent links than transnational advocacy networks. They generally have ‘permanent staff members, a more permanent membership base, a headquarters or secretariat, and are organizations in and of themselves’ (Ibid. p. 95). Most importantly, they have broader strategic aims than single-issue focused networks. What’s more, while these coalitions are organizational entities in and of themselves, they and their members also frequently belong to transnational advocacy networks (Ibid.).

Egan (1995), describes the fundamental difference between networks and coalitions when he talks about the ‘value-added’ in working with others through collaboration rather than mere exchange. Moreover, the author maintains that coalitions create a greater value and commitment. In brief, ‘while networking is an important part of the coalition building, networks can exist without coalitions’ (Yanacopulos, 2005, p. 95).

Coalitions require a high level of commitment coming from member organizations; generally, they are created to deal with broad goals. This organizational structure reflects the multi-level functioning of a coalition in which members have connections with the grassroots while the coalition functions at the national and international levels. Coalition structures reveals that these organizations have moved beyond information sharing, particularly as one of their primary objectives is to influence policies (Yanacopulos, 2005).

(15)

Advocating is often described as a collective venture (Baumgartner and Leech, 1998; Greenwood, 2007; Klüver, 2013). As a matter of fact, advocacy coalitions can be seen as “alliances between group of individual actors who share the same policy goal” (Baumgartner et al., 2009, p. 6). These advocacy coalitions are by definition issue-specific since the policy preferences of interest organizations determine whether they pull decision-makers in the same direction (Klüver, 2013). For instance, several case studies have shown coalitions of actors that collaborate to attain joined goals (for example, Coen, 2005; Yanacopulos, 2005). By doing so, interest organizations composing these coalitions may profit of each other from sharing resources and expertise, decreasing costs through group specialization, and increasing legitimacy and power by speaking with one voice.

A last aspect deserves to be highlighted. It concerns the temporal aspect of coalitions. According to Fowler (1997, p. 115) Coalitions can formed for both short or long terms. Following this, Tarrow (2005) describe factors that affect the duration of the coalition. Among these elements, he argues that coalitions form around short-term sets of threats and opportunities but disperse when the occasion for collaboration passes. Accordingly, long-term coalitions are those who succeeded to develop strong identities and common interests (Ibid.). In turn, Levi and Murphy (2006) detail important explanatory factors that are likely to influence the formation and duration of coalitions. Among these determinants, they include: framing, trust, credible commitments, management of difference and selective incentives. In my research, I take into consideration the framing and selective incentives factors.

To put it in a nutshell, I have introduced some key elements present in the debate on interest organizations, transnational advocacy networks and coalitions. This overview will be of particular importance for the next section in which I draw on the collective action literature to link the mobilization of interest organizations and the coalition building. Put differently, I intend to explain why organizations join and mobilize in coalitions by linking it with why people join and mobilize in interest organizations. An important condition is the consideration of interest organizations as rational and self-interest entities.

2.2 Linking Interest Group Mobilization and Coalition Building

The study of mobilization of interest groups has a long tradition in social science. Perhaps one of the most influential study in this field of research is Mancur Olson’s logic of collective action (1965). According to Olson (1965), groups are composed of individuals having common preferences and interests. Similarly, Arthur Bentley (1949) founder of the

(16)

“group theory” of modern political science confirms that “there is no group without its interest” (Ibid., p. 211). Moreover, as Olson (1965) states “the view that groups act to serve their interests is based on the assumption that individuals in groups act out of self-interest” (Ibid. p. 23). In other words, the idea that groups tend to act in support of their group interests is based on a general premise of rational and self-interested behaviour (Ibid.). Following this conception, it has long been thought that individuals in a group would act to achieve collective objectives. Yet, Olson challenges this assumption when he claims that “rational, self-interested individuals will not act to achieve their common or group interests” (Ibid. p. 24). Still, he indicates that his argument is valid for large groups and specifies exceptions for small groups and groups that can coerce or provide “some other special device to make individuals act in their common interest” (Ibid.).

Olson’s logic of collective action (1965) is the reaction from the transaction school to the pluralist school notably represented by Truman (1951). The pluralist approach suggests that individuals instinctively and collectively act on the basis of shared interests and problems— such as political, social or economic (Lowery & Brasher, 2003). This implies that groups exist following a natural order shaped by the perception of the common interest, which Olson countered by claiming that mobilization is not a natural and spontaneous process. Instead,

“if the members of a large group rationally seek to maximize their personal welfare, so they will not act to advance their common or group objectives unless there is coercion to force them to do so, or unless some separate incentive, distinct from the achievement of the common or group interest, is offered to the members of the group individually on the condition that they help bear the costs or burdens involved in the achievement of the group objectives”. (Olson, 1965, p. 24)

Furthermore, he extrapolates his argument to organizations when he claims that they will follow the same principle: “Nor will such large groups form organizations to further their common goals in the absence of the coercion or the separate incentives” (Ibid.). Olson focuses on organizations that are expected to further the common interest of their members. Among these organizations, there are labour unions, farm organizations, cartels, firms and states. Detailing the different interests that these organizations strive to provide to their members, he separates two notions of goods: (1) the collective benefits indiscriminately provided from the achievement of the common interest to the members as a whole; (2) the incentives being selectively provided to individuals in the group in function of their contribution to the group’s interest. Hence, the notion of separate or selective incentives constitutes for Olson an important stimulant for rational individuals to join groups and act in a group-oriented way. Lastly, his

(17)

study distinguish between negative and positive incentives (Ibid.). In both cases, he calls “mobilized latent groups” a group that has succeeded to act in the group interest either by coercion or by rewards. Conversely, a “latent” group has a latent capacity for action and potential power can only be realized through the provision of “selective incentives” (Ibid.).

Following the definition of coalitions presented above, I maintain that interest groups can be seen as strategic organizations forming networks and coalitions in order to influence other actors. By analogy, I claim that the common interest of such interest organizations is the successful influence of political decision-making and policy outcomes. Moreover, as Yanacopulos (2005) argues, members belonging to the coalitions are independent and they choose to work together in the coalition. Since members are independent, one can assume that they are not constrained by coercion from the coalition. Additionally, they choose to be part of the coalition which underpins a process of mobilizing comparable to the one described in the logic of collective action. Furthermore, when coalescing, interest groups come together to promote mutual interests, creating a distinct entity for such a purpose. Another important point discussed by Yanacopulos (2005) regards the strategy as one of the main drivers of coalition building. In the words of the author, coalitions develop shared visions or desires that motivate organizations to work together (Ibid.). In that sense, “collective strategies involve reaching agreement about how to implement a shared vision” (Gray, 1996, p. 59). These arguments are in line with my objective of making an analogy between the mobilization of groups and the concept of coalition building. Indeed, by looking at the theoretical arguments explaining how groups and organizations form, I identify the similarities of the mechanism and link it to the concept of coalition building.

As regards to the internal dynamics of interest organizations, I recognize the existing relationships with their members, and the underlying implications this can have on their behaviour. However, I consider interest organizations as coherent groups. That is, the interactions with their constituents are constant and do not affect groups’ strategies. I do know that the empirical veracity of such an assumption might be challenged, but I purposely exclude this second layer of dynamics, being beyond my scope.

In sum, as I connect the mechanisms of mobilization of interest groups and coalition building, I take for granted that the collective action problems described for interest organizations occur in a similar fashion for coalitions. This is the main assumption on which I rely on to develop the following sub-sections. I now elaborate on the collective action problems that occur within such organizations. While it constitutes a major insight of my research, this analogy has not been addressed by the literature.

(18)

2.3 The role of selective incentives in the mobilization and maintenance of

organizations.

As described by Olson (1965), collective benefits are indiscriminately distributed with the achievement of the common interest (Ibid.). In this context, De Bruycker et al. (2016) describe “free-riding” as a collective action problem that arises when “individuals benefit from the provision of a certain collective good regardless of their contribution in the provision of it” (Ibid. p. 4).

The underlying implication of this problem is that to act in a group-oriented way, rational individuals need to be stimulated by selective incentives2 (Olson, 1965). I have described them above, either positive (rewards) or negative (coercion). As a matter of fact, mobilization and maintenance strategies of interest organizations are commonly framed in terms of the types of selective benefits provided. Maintenance strategies can also be framed in accord to the types of patronage received (private, public or corporate). For practical matters, I only focus on the types of selective benefits received by interest organizations as maintenance strategies pursued by coalitions.

While the “material” nature of selective benefits is the classic argument used by scholars to explain the mobilization and maintenance of interest groups and social movement, an important addition to Olson’s idea is the identification of “non-material” incentives (see Wilson, 1974; Salisbury, 1969). These non-material benefits can be of purposive and solidary nature. Several authors including Salisbury (1969) and De Bruycker et al. (2015) acknowledge the “relevance and distinct, mutually exclusive qualities of non-material, expressive and solidary incentives” (Ibid. p. 7). Hereafter, I offer an overview of the characteristics of the above-mentioned incentives.

Expressive3 benefits are mechanisms allowing individuals to publicly express the values relevant for them (Salisbury, 1969). These range from demonstrating, transferring money and forwarding messages on social media (De Bruycker et al. 2016). The bottom line of expressive incentives is that the realization of these values does not necessarily bring material benefits to individuals. On the other hand, solidary benefits are explicitly independent of the goals of the organization because they refer to the process of coming together. These include rewards such as “socializing, the sense of group membership, identification and status

2_{I use indifferently selective incentives and selective benefits.} 3_{Also called purposive by Clark & Wilson (1961)}

(19)

resulting from membership, fun and conviviality and the maintenance of social distinctions” (Clark & Wilson, 1961, pp. 134–135). As a result, the combination of different types of incentives is important to maintain continuity as an organization and successful groups are those who are able to offer an “attractive mix of these incentives” (De Bruycker et al., 2016, p. 7). However, it is admitted that this task is arduous in the sense that the management of these incentives puts different demands on the organization (Ibid.). I follow the argument posed by De Bruycker et al. (2016) which states that the incentives brought to the members by group leaders aim to maintain their organizations. In turn, they affect the potential of groups to voice political interests, which is the second-phase of collective action by interest groups.

Empirically, I intend to study how coalitions maintain themselves using the same approach as for interest organizations. The objective is to test whether the coalitions behave in a comparable way as interest organizations. I argue that since the mobilization of groups and coalition building can be theoretically linked, they are subject to the same internal dynamics. This argument is based on the assumption that organizations are likely to act rationally. Consequently, the first sub-question arising from this assumption is: What does interest

organizations perceive as selective benefits of acting collectively in a coalition?

2.4 Second Phase of Collective Action: Consensus Building and Issue Framing

During this phase, existing interest organizations decide on the positions they take on political issues. More specifically, “issue framing” and “consensus building” are the two mechanisms that have to be taken into account to understand this phase of collective action. On the behalf of De Bruycker et al. (2016), consensus formation—which I label “consensus building”—is the “internal management of heterogeneity of policy interest of members” (Ibid. p. 9). With regards to “issue framing”, McAdam et al. (1996) describe it as “conscious strategic efforts by groups of people to fashion shared understandings of the world and of themselves that legitimate and motivate collective action” (Ibid. p. 6).

De Bruycker et coll. (2016) postulate that the “nature of the selective incentives offered by interest group leaders to their members will limit or ease organizational action in the policy process” (Ibid. p. 2). That is, the authors argue that such differences arising during the phase of mobilization and maintenance affects the second, which concerns the process of interest aggregation. Accordingly, they purposefully link these two phases of collective action with the types of incentives offered by interest organizations to their members (figure 2). To do so, they distinguish between diffuse and specific groups.

(20)

On the one hand, they assert a strong focus on selective expressive benefits for diffuse interests. In fact, this can be explained by the non-divisible nature of the values expressed by diffuse interests which make it possibly available for anybody to benefit from the realization of these values. Moreover, “these benefits are also diffusely spread over the population” (Ibid. p. 9). On the other hand, selective material benefits are unavailable for diffuse interests because they are relatively difficult to be meaningfully offered to groups who come together for a certain broad political cause. In contrast, they are relatively easy to be offered to groups who act collectively for a given joint, and relative concentrated interest (Ibid.).

Figure 1: Relationship between selective incentives and the two phases of collective action

As a last point, the authors ignore organizations with a dominant focus on solidary selective benefits because they assume these organizations to fully avoid direct engagement in public policy making. However, I argue that such organizations value the benefits of coming together as presented above. This is turn, do not directly imply the avoidance of the engagement in public policy. As such, I include this type of selective incentives in my framework.

In sum, the relation between the selective incentives and the first stage of collective action concerns the political nature of the mobilization and have “important consequences for the political engagement of groups relying on these distinct mobilization and maintenance strategies” (Ibid. p. 10). For diffuse interests, “the rationale of mobilization is political” (Ibid.) which leads to a potentially difficult first phase of collective action. This is better explained by acknowledging that political interest aggregation can be a slow and challenging process. Once mobilized on the basis of selective expressive benefits, groups representing diffuse interests will find easier to translate their shared values into explicit political positions. Indeed, “expressive benefits already relate to some sort of political goal” (Ibid.). In contrast, during the

Selective incentives 2nd phase of collective action 1st phase of collective action

(21)

first phase of collective action specific interests face a non-political mobilization. They usually manage to mobilize on the basis of material incentives which are not necessarily policy related, and “the commonality of the interests arises from shared economic or social interests rather than political ones” (Ibid.). Therefore, when positioning themselves in public policy, specific interest groups will have to manage a second phase of collective action. Indeed, Interest groups might have opposing views on the design of a certain policy issue, but they might share similar levels of attention to that issue. Hence, they might agree that something has to be done about a certain issue without having necessarily the same policy objective (Klüver, 2013). Accordingly, the cost to the members is the time, human capacity, information and investment in the processes needed to reach collective options on issues. One benefit of a coalition is greater strength when voicing shared positions, with enhanced informal access to information through trusted relations.

Following this, the aforementioned authors advance that when selective incentives are dominantly expressive, interest group leaders will have relatively clear policy objectives and positions. Conversely, for organization leaders relying on material incentives framing clear policy positions will necessitate some coordination with their constituents (Ibid.).

To complete the picture and in relation with the second phase of collective action, a last aspect deserves to be discussed before moving to the fifth and last subsection. According to De Bruycker et al. (2016), interest organizations “vary in the extent to which leaders can operate autonomously from their members (Ibid. p. 13). That is, when positioning themselves groups have different modes of internal decision-making.

“While interest groups can more swiftly position themselves without consulting members, it also involves the risk of losing support, especially when the adopted position does not enjoy broad support among constituents” (Ibid).

Put differently, it is suggested that the consultation of the members regarding the organizational position of interest groups impact the effectiveness of issue framing and consensus building. Thus, members can play a significant role in the second phase of collective action and therefore it is necessary to include their involvement in the positioning of groups.

Empirically, I will assess whether the incentives offered to members are of expressive, solidary or material nature. This shall indicate how the second phase of collective action occurs for coalitions, and how constraints are managed by them. Moreover, I intend to measure the role and involvement of interest groups during the processes of consensus building and issue

(22)

framing. Subsequently, the second sub-question raised is: What roles have interest groups

regarding consensus building and issue framing within advocacy coalitions?

2.5 Balancing individual and collective action

In the last two subsections, I have comprehensively presented the concepts explaining potential variations during the different phase of collective action. From now on, I draw on another concept to provide theoretical support to the third sub-question which considers the balance between members’ individual and collective actions within a coalition. That is, I assess the reasons for members to choose to act collectively on some issues whereas individually on others. At first sight, the balance between individual and collective actions by coalition members seems puzzling. The current literature does not tell much about this aspect of my research.

To explain how coalition members balance between their individual and collective actions, I mainly rely on the framework developed by Hanegraaff et al. (2016). It encompasses the political science literature that deals with lobbying strategies and explains how transnational lobbyists balance inside and outside strategies. According to the authors, organized interests seek to deploy strategies that follow their organizational maintenance needs (Ibid., p. 571). Following this, an important aspect of maintenance strategies is linked to the level of credibility perceived the group members. For instance, by using inside lobbying strategies interest groups increase their access to policy makers, which signal to their supportive constituencies efficient resource management and their active engagement in the pursuit of the common interest (Ibid.). Thus, this can be perceived by members as a sign of “good health” and credibility, which in turn influence the decision to remain and participate in the organization. In contrast, outside lobbying strategies explicitly indicate to the members that their group representatives are effectively acting, since such tactics directly generate public exposure and visibility (Ibid.). Moreover, some scholars believe that NGOs and citizen groups face significant obstacles to attract resources, increasing the need to actively involve their members (Dür & Mateo, 2013, p. 663). From this perspective, groups are expected to balance their lobbying strategies to pursue their organizational maintenance objectives.

Even though the scope of their research is different, I adjust this approach to the context of advocacy coalitions, to be consistent with my focus of study. Indeed, they note that organizations balance their strategies according to the group type and resources, with the final objective of influencing policies. Similarly, coalition members pursue the goal of achieving the

(23)

common interests either by exploiting the collective action offered by the coalition or by acting individually. According to the Hanegraff et al. (2016), “group type” refers to the distinction already presented between diffuse or specific interests. They consider it more as a proxy for resources than a central explanatory factor (see Beyers, 2004; Hanegraff et al., 2016). On the other hand, “Resources” relate to financial means, policy expertise or staff resources (Ibid.). To verify whether this framework can be used to answer my question, I partly rely on the two previous subsections which already provide some answers, and complete my argumentation with the findings from my interviews and organizations’ documents.

In relation to the second phase of collective action, I argue that depending on the perceived selective incentives and on the role and involvement of interest groups in the framing of issues and building of consensus, they will have different responses to the coalitions regarding their balance of individual and collective action. The above-mentioned concept can be applied to coalitions in the sense that member organizations can have different comportments during the second phase of collective action. Furthermore, when framing issues and building consensus, coalitions prioritize the action conducted collectively, and accordingly members can balance their involvement in the collective action by focusing on their individual priorities. Hence, using this conceptual framework, I intend to answer to the third and last sub-question: How do interest groups balance between their individual and collective actions?

In the next section, I develop and justify the methodology used in order to tackle my research question and its three subquestions. Additionally, I explain the process of case selection and data collection, before explaining the method with which I processed the data to highlight the different concepts.

(24)

3. Methodology

The present section introduces the research design and methods adopted to conduct the empirical research of my thesis. My aim here is to justify my methodological choices, and detail how the process unfolded to answer the following research question:

“How interest organizations join, frame issues and build consensus as part of advocacy coalitions? How do they balance their individual and collective actions?”

Actually, I have chosen a research design adapted to test the interactions between two actors: interest organizations and coalitions. My goal is to see how coalition members perceive the selective benefits of coalescing; what role they have when it comes to issue framing and consensus building; and consequently, explain the balance between organizations’ individual and collective actions. In turn, I intend to prove that the theories of collective action developed to explain the mobilization and maintenance of interest groups also apply for coalitions.

First, I expose the methodological approach I use and explain the selection of my case. In a second step, I discuss the methods employed by giving a particular emphasis on the fieldwork conducted in Brussels. Finally, I describe how I processed the data gathered, before presenting the sub-questions.

3.1 Methodological Approach

The main objective of my study is to apply the collective action theories in the context of coalitions. To do so, I rely on qualitative methods of research and analysis during the whole process. To be able to consider the perception and role of interest organizations, I first needed to collect qualitative data. One can wonder why I used this sort of research design. I did so because I intend to test whether coalitions follow the same patterns as interest organizations with regards to their mobilization—coalition building—and interest aggregation—issue framing and consensus building. Subsequently, I aim to assess the impact of these elements on how interest organizations balance their individual and collective actions. Therefore, in the last part of my research, I expect to contribute to the literature by providing insights on this process. To do so, I have decided to use a single case study as it constitutes “an intensive study of a single unit for the purpose of understanding a larger class of (similar) units” (Gerring, 2004, p. 341), and represents a flexible research design compatible with several methods. According to Gerring (2004), units are “spatially bounded phenomenon—e.g., a nation-state, revolution, political party, election, or person—observed at a single point in time or over some

(25)

delimited period of time” (Ibid.). Basically, this design can help me to reach a deeper understanding of the case-specific dynamics. With regards to the type of case study, I have chosen to conduct a study with a within-case spatial variation but no temporal variation. That is, the members of the coalition taken into account are differentiated by their spatial locations.

In order to gather the relevant data, this methodology focusing on one case has allowed me to use: semi-structured interviews with representatives of interest organizations, a participant observation and content analysis. Also, this research design is beneficial because it provides the researcher a more in-depth knowledge of the selected case with the objective of learning about the general population of cases. Also, this approach is appropriate to understand causal mechanisms as it is the case in my research. Indeed, I want to prove that the provision of selective benefits impacts coalition building which in turn, affects the framing of issues and the formation of consensus. All combined, these elements explain how coalition members balance their actions within coalitions.

Yet, the case study method is often criticized due to the lack of representativeness in the larger population of cases and its limits in terms of generalization—also called external validity. I claim that since my scope of research concerns a coalition acting at the EU level, some generalization can be drawn. In effect, advocating EU institutions involve a large range of actors, including various coalitions working on different issues. Therefore, it can be expected that they follow similar schemes regarding their internal dynamics.

On the other hand, the use of qualitative content analysis in case study research helped me to find out that my initial focus—assessing the gain of leverage for interest organizations when lobbying in coalitions—was not easy to measure by using such methods. In consequence, I had changed my scope of research to be theoretically and empirically consistent.

When it comes to the methods employed, interviews have highlighted the relatively informal and secretive aspects of interactions and internal dynamics of coalitions. For instance, understanding how member organizations perceive the selective benefits received from coalitions is crucial to my research, as this information is generally absent in documents. Overall, interviewing NGOs representatives remains an excellent way to collect data in order to strengthen my arguments. On the other hand, participant observation is a method that allowed to gain an insider’s point of view of the processes at study. Therefore, I acknowledge that observation was important as a secondary method in order to complete a better view of the interactions between the coalition and its members. In effect, while interviews provide individual and objective opinions, a participant observation is beneficial because of its neutral and reflective nature. Lastly, organizations’ documents were necessary to get additional

(26)

background information. Moreover, it allowed me to acquire official information coming directly from the different actors involved.

Practically, I have first divided my research question into three sub-questions. Each of these is expected to provide complementary insights. Consequently, the answer to the main research question synthesize the different elements considered to form a meaningful and credible whole. Indeed, by combining multiple methods and empirical materials, I aim to overcome the weakness or intrinsic biases and the problems that come from single method studies.

3.2 Case Selection

“Often researchers begin their inquiry with a theory in search of a test case or a case in search of a theory for which it is a good test” (George & Bennett, 2005, p. 83). Concerning my research, I started by selecting my case and then searched for a theory that I could test on it. Empirically, I study the case of CAN Europe. CAN Europe is a case of coalitions acting at the EU level, and is Europe’s largest coalition working and advocating on climate and energy issues. It is composed of over 130 member organizations in more than 30 European countries— representing over 44 million citizens. Several reasons motivate this choice:

Firstly, I have chosen CAN Europe because I believe that it is the “most likely” case to assess whether the theories of collective action apply for coalitions. Indeed, as my research is theory-based, the range of possibilities for my case selection strategy consists of choosing a case that is likely or not. Moreover, it has relevance for scholars and practitioners that wish to analyze the internal dynamics of coalitions acting at the EU level. Other reasons concern (1) the increasing importance of the topic of climate and energy at the European level; (2) the concentration of diverse interest organizations within this coalition; (3) and the presence within CAN of interest organizations that have different levels of resources and capacities. Last but not least, focusing on an EU level coalition is interesting because the EU constitutes an appropriate focus since there are multiple interest groups advocating for policy changes.

Another decisive point regarding the selection of CAN Europe concerns the nature of the interests it represents. In effect, CAN Europe focuses on climate and energy issues which according to Beyers (2002) correspond to diffuse interests. As we have seen in the theoretical framework, diffuse interests defend “broad and general segments of society” (Ibid. p. 589) such as environmental issues. Therefore, by considering this typical coalition defending diffuse

(27)

interests at the EU level, it implies that CAN Europe is most likely to offer selective expressive and solidary benefits to its members (De Bruycker et al., 2016).

As regards the within-units of my case, CAN members are mainly NGOs. Practically, I have used a purposive sampling in order to ensure that the NGOs I interviewed were relevant to my research and second, that there was a diversity in the sample. This led me to the following selection criteria: (1) NGOs must be members of CAN Europe; (2) NGOs can have an office in Brussels or not; (3) being either active at the national level, EU level or both; (4) interviewees have sufficient information regarding my topic; (5) representatives shall speak either French or English. Based on these criteria, I listed several organizations and contacted all of them by email. During this process, I presented myself and my research, precisely my focus of study and the interview guide. Likewise, I contacted the secretariat of CAN Europe.

3.3 Data Collection

Using a case study let me the choice between a wide range of methods. I selected these three for their relevance in accordance with my particular topic of research. As a matter of fact, I could not have used document analysis as a primary method, nor quantitative methods. In effect, I was looking for qualitative answers and could not predict the type of responses I received. However, I could have selected additional methods such as focus groups, in order to confront different viewpoints on my questions. As previously mentioned, I used three complementary qualitative methods to collect my data: semi-structured expert4 interviews, participant observation, and documents. Hereafter, I provide a detail of the process of data collection for each method.

Interviews

First of all, I have collected data through semi-structured interviews with experts representing member organizations of CAN Europe. I decided to conduct these interviews based on a predetermined set of open questions in order to be free to explore additional themes and react according to the responses. I adapted my questionnaire throughout the research and ended up with a more structured design as the research topic became more precise. I also had to adapt my interrogations according to the position of the interviewee.

(28)

In practice, I sent emails to the different organizations matching with my selection criteria, and I received six responses: from CAN Europe and five of its members. The interviews have been held with5: (1) the network coordinator of Carbon Market Watch; (2) a policy officer of ClientEarth; (3) a policy officer of WWF European Policy Office; (4) the director of CAN Europe; (5) a policy officer of Oxfam International; (6) the director of Change Partnership. Then, after my participant observation, I conducted two more interviews with: (7) a spokesperson of Plant-for-the-Planet and (8) a policy officer of BOTH Ends. In total, I had eight interviews that lasted about an hour each, and I had the opportunity to ask few more questions to members during my participant observation. I do not really use these interviews since they were not encompassing all the elements of my interview guide. Out of these eight interviews, I conducted five in person (mostly at the organizations’ offices), and three through Skype.

Regarding the interview guide and the semi-structured form I wanted to have, I have designed a template to follow during the interviews. It consists of seven main parts with context-oriented questions. It generally served more as a guideline than a proper questionnaire principally because I intended to have “conversations” with the interviewees. This allowed them to be more at ease and discuss about aspects that I had not predicted. Consequently, I had the occasion to react and ask questions that were not planned.

With respect to the topics discussed, I covered: (1) the organizations’ missions and structures; (2) the reasons of joining CAN Europe; (3) the benefits and pitfalls of being part of CAN Europe; (4) the role of the organizations in framing the collective issues at stake and during consensus building; (5) the actions taken individually by organizations; (6) the actions taken collectively as part of CAN Europe; (7) the reasons explaining the different balance between individual and collective actions.

On the other hand, I first thought that I reached a theoretical saturation point after the sixth interview because of the similarities of the responses provided by the interviewees. However, thanks to the last two interviews with representatives of national members of CAN Europe, I was able to gain additional knowledge that is valuable for my research. This also means that my sample was not fully representative of the population of organizations composing CAN Europe. I recognize that I could have interviewed more national members of the coalition in order to distinguish between two types of within-units: Brussels-based and national members.

(29)

In my analysis, I will primarily rely on the responses provided during these interviews to formulate my arguments. In effect, this is the main method and also the more adapted in my study. When necessary, I will complete sentences by elements gathered through the other methods I used.

Participant Observation

During my fieldwork in Brussels, I had the opportunity to attend the general assembly (GA) of CAN Europe which happens only twice a year. I had access to that GA after having conducted an interview with the director of CAN Europe. Indeed, at the end of the interview, I asked him the permission to attend the next GA that was planned beginning of May 2017. During this event that lasted two days, I had access to almost everything except the formal session. Indeed, as I was granted permission by the secretariat of CAN Europe but not by all its members, I could not attend this session. Moreover, several members from different countries and working on various issues were represented during this GA. In total, at least sixty organizations from seventeen EU countries were present over the two days.

Regarding the agenda of this GA, it was divided between plenary sessions and workshops. During the first plenary session, several speakers from the secretariat of CAN Europe and member organizations presented updates of the EU legislative files. Among these presentations, they talked about: (1) the lessons from the “Emission Trading Scheme” (ETS) work to influence the European Parliament; (2) the incoming Estonian presidency; (3) how to advocate the member states. In addition, four workshops were organized to discuss and exchange knowledge about the EU legislative files. I attended a workshop called “Securing enough effort in the Effort Sharing Regulation (ESD)”, which aimed to lay out the collective strategies on how to influence the EU institutions. At the end of the first day, the formal session which I could not attend was settled to approve CAN Europe budget, review membership applications and define CAN Europe priorities for 2018. On the second day, the plenary session and workshops focused on non-legislative files. For instance, presentations were about the “Brexit”, and the implementation of the Paris Climate Agreement.

All in all, I had the precious occasion to observe what members and the secretariat were saying, how they were talking about what they do. Also, I discussed with members and staff from the secretariat about their views of the coalition. However, I was requested to stay away from intervening in plenary conversations. Yet, I was granted a more active engagement in break-out sessions with the perspective to contribute to their discussions. In brief, I only had

(30)

the possibility to interact informally to get answers useful for my research. Last but not least, I had to sign a confidentiality statement and to present myself as a researcher volunteering with CAN Europe. Despite these conditions, I collected interesting data that I use as a complement of the interviews.

Documents

To add depth to my findings I have used many documents from member organizations. Reading newspapers and NGO websites have also helped me to understand the background of member organizations prior to the interviews. For instance, before each interview, I was considering organizations’ annual reports in order to specify and orientate the follow-up questions. Furthermore, I have read several position papers published by CAN Europe and its members in order to gain a better view of the collective actions. In the same trend, I browsed into the websites of the organizations in order to find information about their individual and joint campaigns. I did so to gain a better picture of the balance between members’ individual and collective actions.

3.4 Data Processing

Herein, I present how I processed the data, and how I plan to use it in my analysis. With regards to the interviews, I have managed to voice record them all, so I first had to transcribe them. Then, I have proceeded to manual coding. Each of the aspect of my research was attributed to a particular code, highlighting relevant aspects. Then, I created tables containing all the similar quotations, and providing insights on the different sub-questions. In particular, selective incentives had to be divided between material, expressive, and solidary. To do so, I have separated them into three distinct codes.

In total, I have distinguished eleven codes, encompassing the different types of information gathered. These range from the perception of selective incentives received (3); the perception of being consulted during the framing of issues and building of consensus (4); the interaction between members and the coalition (2); the balance between individual and collective action (2).

Likewise, I included the notes taken during the participant observation to these tables, by following the same approach. Important to mention is the reflective aspect of the notes.

(31)

Indeed, during my participant observation, I had taken notes before adding my personal reflections to these descriptive elements.

Regarding the organizations’ document used, I have isolated relevant information found in the different documents gathered during my fieldwork. As this information does not directly contribute to answering my research question, I use it as supportive complementary data.

For the sake of my analysis I divide my research question into the three following sub-questions:

(1) What do interest organizations perceive as selective benefits of acting collectively in a

coalition?

(2) What roles have interest organizations regarding issue framing and consensus building

within advocacy coalitions?

(3) How do interest organizations balance between their individual and collective actions? Hereafter, I present the content of my empirical study, by confronting each of these subquestions to their theoretical arguments.

Assessing the interactions of interest organizations and coalitions : mobilization, participation and balance : a case study of Climate Action Network Europe