Agency, Advocacy Coalitions, and Network Structure: Differentiating Political Brokers and Entrepreneurs

While the Advocacy Coalition Framework developed the concept of political brokerage, the concept of political entrepreneurship is mainly rooted in the Multiple Streams Framework and the Punctuated Equilibrium Theory. In their article Christopoulos and Ingold (2015) show that both types of exceptional actors may coexist in policymaking processes and simultaneously influence policy outcomes. What remains unclear is how to methodologically disentangle political brokerage from entrepreneurship. By going back to the theoretical roots, the present article elaborates on the conceptual differences between both types of strategic network positions and proposes a distinction between brokerage and entrepreneurship by using social network analysis. The added value of the present work is that it does not only consider actors’ centrality measures, but takes into consideration the entire network structure of advocacy coalitions, actors’ embeddedness therein, and actors’ policy beliefs in order to identify brokers and entrepreneurs. The ACF literature describes brokers as mediators with neutral beliefs who form an essential link between conflicting coalitions. By taking this theoretical definition seriously, the present paper operationalizes brokers as actors with exceptionally high betweenness centrality scores, a neutral belief system, and a high density of ties to all advocacy coalitions. By contrast, the literature portrays political entrepreneurs as self-interested actors who are able to promote their preferred policy designs through their connections to powerful others. Consequently, the present paper identifies entrepreneurs by their exceptionally high eigenvector centrality scores, non-neutral policy beliefs, and their embeddedness within a dominant coalition. The paper employs data from a survey conducted by the author between 2011 and 2014 in four European countries including Switzerland, Germany, France, and the Netherlands. The dataset comprises relational profiles and policy beliefs of 149 actors participating in decision-making processes regarding emerging issues of water protection policy. Results reveal that the proposed approach to distinguish brokers and entrepreneurs empirically proves useful in the studied cases. The article not only provides more clarity to researchers who wish to study political agency, it also demonstrates the added value of triangulating diverse theories of the policy process in order to better understand the processes guiding policymaking.