The Effect of Party Networks on Congressional Primaries

• In Event: Political Parties, Primaries, and Candidate Selection

Abstract

Recent scholarship on political parties has conceptualized these institutions as dynamic networks of interest groups, activists, and politicians that work together to support candidates for elected office. Because most general elections are generally safe for one party and the resilience of incumbency, Extended Party Network (EPN) theory places a premium on nominating faithful agents in the primary. Work addressing the influence of party network support on the electoral prospects of candidates has thus far focused on general elections (Desmarais et al. 2015), small samples of races (Masket 2009), or limited the scope to only the central portions of the party network (Hassell 2018).

In this paper, I measure the effect of EPN support on the primary prospects of candidates to open-seat races in the House and Senate between 1980 and 2016 in two fashions. First, Using the political contributions made to federal candidates between 1979 and 2014, I construct donor networks for individual primary candidates. I then analyze the structure of these networks to determine which candidates have the support of durable networks within their respective parties. My measure -- Existing Network Density -- indicates how dense or interconnected a candidate's donor network is across election cycles. Those candidates who score high on this measure are those who have the support of campaign contributors who frequently coordinate in primary elections. Having the support of a dense network of donors can serve as a proxy for the support of durable groups within their party. Second, by constructing networks of candidates who are connected in proportion to the number of shared campaign contributors they had in the primary, I can detect which candidates are embedded in the full party’s network and which ones find themselves on the outside looking in. More specifically, I measure the centrality of individual primary candidates within the network of co-partisan candidates.

Using both matching and regression based methods, I find large and significant electoral benefits for candidates with their party network’s support. To further address the issues of endogeneity and to better isolate the causal mechanism, I supplement these quantitative results with qualitative interviews of political elites from the 2014 primary cycle. Collectively, these results suggest that organized interests within the larger party network are able to systematically assist their preferred candidates in their pursuit of office. These results have important implications for the study of issue representation in Congress.

Author

• Shawn Thomas Patterson