Insider Strategies, Outsider Strategies and Influence Reputation in Climate Change Policy Networks

In the policy networks literature, it is assumed that the actors that participate in policymaking processes can influence political outcomes using a variety of strategies. But which are more effective: insider strategies, such as lobbying and taking part in official hearings, or outsider strategies, such as media visibility and mobilization through demonstrations and mass meetings? We test the hypothesis that actors using insider strategies have more influence over the climate change policy process than those using outsider strategies. We use Exponential Random Graph Models (ERGMs) and survey data on climate change policy networks in seven countries: Ireland, Finland, Sweden, Portugal, Germany, Czech Republic and Australia. Our dependent variable measures how influential each actor is perceived as by others in the network. The main independent variables of interest measure the use of different insider and outsider strategies by national policy actors. Our network approach makes a unique contribution to the literature on insider versus outsider strategies because it enables us to simultaneously test several alternative explanations for political influence. We also include in our models a range of other variables that have previously been found to predict influence reputation in policy networks. Following Ingold & Leifeld (2016), we include each country’s national collaboration networks and each actors’ betweenness centrality score in those collaboration networks as structural determinants of influence reputation. To investigate formal institutional determinants of influence reputation, we include two actor type variables: The first of these tests if government departments are perceived as being influential; the second tests actors if political parties are perceived as being influential. Following Fischer & Sciarini (2015), we include variables that test for actor type homophily and belief homophily. Finally, we include the GWESP (geometrically weighted edgewise shared partner) and the GWDSP (geometrically weighted dyadwise shared partner) endogenous terms to capture triangular and hierarchical network sub-structures. Results show that conducting scientific analysis is the only insider activity found to be a significant predictor of perceived influence across multiple countries. Working through the media, an outsider activity, is significant in Finland, Portugal and Sweden, while in Ireland engaging in political mobilization is significant. The structural determinants of influence reputation identified by Ingold & Leifeld (2016) are significant in all countries. Government departments are perceived to be influential actors in nearly all countries, while in Finland, Germany and Sweden political parties are seen by other actors as influential. The homophily measures (Fischer & Sciarini, 2015) are not positive and significant in any country. Finally, in all four countries, the gwdsp terms are negative and significant and gwesp terms are positive and significant, indicating the hierarchal nature of influence reputation.