Explaining Interest Groups' Information Networks in the European Union

Interest groups are key actors in the design, formulation and implementation of EU policies. They constitute a valuable source of information, expertise and legitimacy for supranational decision-makers and play a key role in the process of European integration. Although networks are a distinctive feature of EU lobbying and policymaking, we lack a systematic analysis of interest groups’ information networks. We address this gap in research and examine the formation and structure of information networks based on social media data. We conceptualize interest groups as information and reputation maximizers, and develop a theoretical argument outlining how key actor-level characteristics shape and affect the likelihood of contact-making and tie formation on Twitter. Our analyses account for the effects of network interdependencies on actors’ propensity to establish an information tie. We test our argument on a new dataset containing information about the network ties between 7,388 interest groups registered in the EU Transparency Register and having a Twitter account. Our empirical analysis employs exponential random graph models (ERGM) to study networks across 40 EU policy areas. We show that, across EU policy areas, homophily is key in understanding the information networks: interest groups are more likely to establish an information tie with actors who share their interest type and country of origin. Also, interest groups are more likely to establish a tie with an organization that is a policy insider and has frequent access to EU decision-makers, in line with the argument that perceived influence of an actors is key in determining its network ties and positioning. In an additional, exploratory analysis, we examine policy and interest group community characteristics that drive the variation in the effects of interest type and country homophily across policy areas. While variation in country homophily is driven by multiple factors, clearer patterns emerge in relation to the effect of interest type homophily. It is lowest in regulatory policies, medium high in (re-)distributive policy domains and highest in foreign and interior policies. The observed differences between regulatory and (re-) distributive policies are explained by lower levels of diversity in the interest group community in regulatory policy areas and the predominant presence of business groups.