Research

Job Market Paper

“Interest Group Strategy and State Legislative Polarization.”

• 2023 Best APSA Paper Award, Political Organizations and Parties Section

• Abstract: The growth in partisan polarization in US national politics over the past several decades has been mirrored at the state level, a development whose causes are still not fully understood. Although organized groups are theorized to pull parties to the ideological extremes, previous analyses of the relationship between interest group activity and polarization among state legislators have yielded mixed results. I explain these conflicting findings by drawing on the interest group literature to distinguish between two potential strategies groups can pursue—ideological (focused on electing legislators who share particular policy views) and access-oriented (prioritizing obtaining favors from incumbents)—and argue that the former approach, but not the latter, will lead to polarization. To determine the prevalence of these strategies, I draw on campaign contribution records from all fifty states between 1997 and 2016 to implement a supervised machine learning model that classifies organizations as ideological or access-oriented based on their donations to candidates for state legislature. Using this model, I show that existing methods substantially underestimate the proportion of contributions that come from ideological sources. I further demonstrate that increased activity by ideological groups in a state is associated with that state's legislature becoming more polarized. In addition, more extreme legislators rely on ideological groups for a greater share of their funds, consistent with the idea that these groups promote polarization by supporting the election of polarized candidates. These results advance our understanding of interest group activity and the origins of partisan polarization in contemporary politics.

Publications

“Who Runs for Congress? A Study of State Legislators and Congressional Polarization” (with James M. Snyder, Jr. and Andrew B. Hall). 2024. Quarterly Journal of Political Science 19 (1): 1–25.

• Abstract: Why have the people running for Congress become more ideologically extreme over time, causing Congress to polarize? In this paper, we study how the pool of state legislators shapes the ideological distribution of congressional candidates. We geographically match state legislators to the House districts that they could plausibly seek to represent. Although the national pool of state legislators has polarized over time, we find no evidence that they have polarized more where opportunities to win House seats are higher, or that more-extreme state legislators have become more likely to run for Congress relative to other state legislators in the same congressional districts. We conclude that the nationwide polarization of state legislators is particularly important in explaining the polarization of congressional candidates. The results emphasize the need to understand ideological selection into running for lower-level political offices in order to understand congressional polarization.

Working Papers

“Which Issues Are Driving Polarization in US State Legislatures?”

• Abstract: Partisan polarization on the state level has received increased attention recently from both academics and the media, but much remains unknown regarding its causes and features, especially the extent to which polarization varies across different policy issues. In this paper, I employ a dataset of approximately 750,000 evaluations of individual state legislators from almost 1000 interest groups representing a wide range of policy domains to track state legislative polarization in these areas between 1997 and 2016. Specifically, I estimate the level and rate of polarization and its uniformity across states for each issue by investigating how between-party differences in and the dispersion of the relevant interest group grades have evolved. My findings indicate that polarization during these years was primarily associated with “culture war” debates including those concerning LGBTQ+, gun, and especially reproductive rights. These results expand our understanding of the roots of current partisan polarization in the states and indicate directions for building on this work to ask what role interest groups themselves play in this process.

“The Political Consequences of Federated Elections: Evidence from California.”

• Abstract: A distinctive characteristic of American politics is the practice of holding frequent, separate elections for offices at many different administrative levels. Although recent work has highlighted how this multitiered electoral structure impacts local government, less attention has been paid to the effects of this system on state and national politics. In this paper, I address this question by leveraging a law passed in California in 2015 that required consolidation of some municipal elections with federal primary and general elections. Using a difference-in-differences framework, I find that holding on-cycle municipal elections appears to increase turnout on the national Election Day by around 3 percent and provide suggestive evidence that the presence of a municipal election on the general election ballot is what drives this effect. These results support the idea that the US election calendar is an important contributor to the country’s low rates of aggregate turnout in international context.

“Estimating Controlled Direct Effects with Panel Data: An Application to Reducing Support for Discriminatory Policies” (with Matthew Blackwell, Adam Glynn, and Hanno Hilbig). Invitation to Revise and Resubmit. 

• Abstract: Recent experimental studies in the social sciences have demonstrated that short, perspective-taking conversations are effective at reducing prejudicial attitudes and support for discriminatory public policies, but it is unclear if such interventions can directly affect policy views without changing prejudice. Unfortunately, the identification and estimation of the controlled direct effect—the natural causal quantity of interest for this question—has required strong selection-on-observables assumptions for any mediator. We leverage a recent experimental study with multiple survey waves of follow-up to identify and estimate the controlled direct effect using the changes in the outcome and mediator over time. This design allows us to weaken the identification assumptions to allow for linear and time-constant unmeasured confounding between the mediator and the outcome. Furthermore, we develop a semiparametrically efficient and doubly robust estimator for these quantities. We find that there is a robust controlled direct effect of perspective-taking conversations when subjective feelings are neutral but not positive or negative.