Much like the average American voter, legislators are frequently faced with complex political decisions they are ill informed to make. To overcome this problem, he or she can seek the requisite information from a variety of sources at her disposal: public opinion, the executive, interest groups, or the behavior of his colleagues. Given the internal incentives, level of expertise, and reliability of intra-legislative cues relative to other information sources, legislators tend to rely most heavily on the behavior of those around them to inform their own decisions. Decades of Political Science research attests to the persistent value of cue-taking, or influence, to legislative actors across time and political context as a means of providing consistent cues that serve a variety of political ends in an otherwise low-information setting, as well as the informal hierarchy that this process implies. However, while scholars of political science have long known about the value and pervasiveness of influence in American legislatures, the consequences of this process are unknown.

My dissertation explores the concepts of influence and informal hierarchy in U.S. state legislatures, a consequence of representatives’ pervasive use of cue-taking as a low-cost source of information. Using temporal cosponsorship data from eight US states, this project comprehensively defines and models this latent hierarchy, aiming to answer three broad questions: 1) what qualities makes a legislator influential in this process, and how does that differ across political context?, 2) how does variation in state-level institutions impact the utility and accuracy of cue-taking as a behavioral heuristic, and to what extent does the widespread indulgence of this process result in policy outcomes at odds with public opinion?, and 3) given that legislators the relative influence of their colleagues as a metric to condition their decisions, to what extent do voters use it their evaluation of political candidates? In other words, do voters prefer candidates who are powerful – even if they incur other representational costs – to those who are not?

Capitalizing on unique, temporally sequenced cosponsorship data in eight US state legislatures, this project explores the latent hierarchal network through which influence is exercised by examining its role and scope in legislative decisions as well as its value to constituent representation. This project contributes to an important and underdeveloped topic in American politics by advancing three primary research questions: (1) How do behaviors diffuse across legislators embedded in political networks?, (2) To what extent are these influence networks responsive to institutional arrangements, and how do they affect legislators’ policy responsiveness?, (3) How valuable is a legislator’s institutional status to constituents?