Research

"Leveraging an October Surprise to Estimate Coattail Effects" | pdf

Abstract: Does a presidential candidate's popularity impact her party's down-ballot success? So-called presidential "coattail effects" can be challenging to identify: it is difficult to distinguish between voters' opinion of a prominent candidate and her political party. In this paper, I exploit a shock to candidate popularity created by late-election information in the 2016 presidential election to estimate coattail effects. Using a difference-in-differences design and variation in the availability of early voting, I find that counties only able to cast their ballot after the release of FBI Director James Comey's letter to Congress on October 28, 2016, saw an increase of 3.2 percentage points in Republican presidential vote share. Using an instrumented difference-in-differences model, I find that a one percentage point increase in Republican presidential vote share led to a 0.86 percentage point increase in down-ballot Republican vote share. This suggests that late-election information not only affects the election of the candidate in question, but can lead to spillover effects for others in her party as well.

Abstract:  Convenience voting (any form of voting that does not take place on Election Day at one's precinct) offers voters a "low-cost" method of voting, but at a price: those choosing to cast a ballot early forfeit their ability to incorporate late-election information into their vote. Given this information can matter for election outcomes, normative questions arise: from an ex-ante point of view, would society benefit from a wider availability of early voting, or is early voting already too available? To answer these questions, I develop a model in which voters choose whether to vote early, late, or not at all, and where both information and the realized cost of Election Day voting affect whether a particular voter votes or not. I show that early voting is beneficial to society when 1) Election Day voting costs are correlated with ideological preferences and 2) late-election information is not too "big."

"Abortion Ballot Measures Have Spillover Effects on Election Outcomes" with Graham Gardner and Melissa K. Spencer | pdf

Abstract: In the 2022 U.S. midterm elections, Democratic candidates lost fewer than predicted seats and stymied an expected red wave. News coverage and polling data represent this surprise Democratic success as a result of voters’ response to the Supreme Court overturning Roe v. Wade. Using county-level vote data, we find that the decrease in Republican vote margin in 2022 can be explained by demographic and economic factors. However, relative to the national average, the Republican vote margin decreased by 4.8 percentage points more in states with abortion-related ballot measures. Our results indicate that abortion ballot measures have spillover effects on election outcomes of a magnitude large enough to determine competitive races.

"Do Newer Methods Deliver? Re-evaluating the Impact of Universal Vote by Mail" | pdf

 Abstract: Universal vote by mail (VBM) offers a theoretically large decrease in voting "costs" to voters: ballots are mailed to all registered voters in advance of Election Day -- no request needed -- who mail back their completed ballot to cast their vote. VBM's reduction in voting costs is thus two-fold: first, costs associated with in-person voting (travel time, lines, etc.) are eliminated, and second, automatic ballot receipt provides a "nudge" to all registered voters, reminding them to turn out. However, previous studies find VBM leads to relatively modest increases in turnout, contrasting the intuition from costly voting models. In this paper, I use recently-developed estimation techniques appropriate for settings with variation in treatment timing and find that VBM leads to a mild increase in turnout. Having shown that these results are not simply an artifact of the chosen empirical specification, I turn to heterogeneity analyses to understand where VBM is most effective in boosting turnout. I find that VBM has its largest turnout effects in counties with higher baseline voter registration rates and higher baseline Democrat vote shares. Additionally, I find that turnout effects are largest in rural counties and in counties with older voters.