Political Connections and Legislative Effectiveness

• In Event: Political Networks, Influence, and Communication in American Politics

Abstract

Understanding what makes legislators effective is a central problem in the congressional literature. Previous work has shown that well-connected legislators are more effective than the average legislator. However, estimating the causal effects of
connections on legislative effectiveness is challenging, as unobserved factors likely influence both the effectiveness and the network of legislators. I propose a novel empirical strategy that estimates how the loss of a connection impacts legislators’ effectiveness in the U.S. House of Representatives. First, I identify the network disruptions induced by the deaths of legislators while in office. Second, I overcome the sample selection problem created by the endogeneity of connections by comparing
legislators who represent the same states as deceased legislators with all other legislators. Legislators who represent the same states as deceased legislators collaborated with them more closely than legislators from other states, but the two groups are otherwise comparable. I find that legislators suffer a 25% decrease in their effectiveness following the death of a legislator from the same state. Substantive and significant bills sponsored by these lawmakers are less likely to receive action in committees, and fewer of them become law. Placebo tests indicate that the estimated effects are not attributable to unobserved differences between legislators who were affected and unaffected by deaths of legislators from the same state. The results provide evidence of the influence of connections on legislative effectiveness.

Author

• Haritz Garro, Stanford University