A Longitudinal Analysis of Interest Group Influence in Retirement Policy
In this paper, I explore some of the political outcomes of networked relationships within a policy domain. Specifically, the outcomes for this paper are an organization's propensity to testify before a congressional committee, the number of times an organization is mentioned in a major U.S. newspaper or wire service story, and the selection of proposed legislation for lobbying. These outcomes I have collectively labeled as ‘influence'. The broad research question of this paper is whether relationships matter for influence in interest group politics. This paper looks at influence from a network perspective using two different applications. In the first, influence consists of being selected to provide information, which is both a tool of influence and an indicator of influential status, and I use congressional testimony and being quoted in the news media as case studies. Does working with others or being highly connected to others lead to being selected? The second application involves influencing the agenda of other lobbying organizations. Do the choices of one lobbying organization affect the choices of another organization? While in both cases I use longitudinal data and network analysis methods, I rely on traditional statistical regression for the first approach but use a relatively new dynamic model of network change for the second application.The discussion in this paper will begin with a review of prior work on influence, which work shows the difficulty of defining and operationalizing influence. I then discuss the main argument of this paper as well as hypotheses that follow from the broad argument regarding networked relations and influence. I next discuss how I operationalized the idea of influence and review the applicable methodology. This paper uses longitudinal data from a single policy domain, which is federal retirement policy. The policy domain is reviewed in terms of the characteristics of the participating organizations, and I then present and discuss the main empirical results. The paper concludes with some thoughts about the nature of influence in the context of the prior paper's empirical findings