Job Market Paper
Motivated or Mobilized? Competitiveness, Campaign Effort, and Turnout in U.S. Elections
Voter turnout is substantially higher in battleground states than in non-battleground states during U.S. presidential elections. It remains unclear whether this is an organic response to closer elections or a manufactured outcome of campaign activity. This paper develops and estimates the first structural model in which turnout, competitiveness, and campaign effort are jointly determined in equilibrium. Candidates allocate resources across states, anticipating how their effort shapes both voter preferences and the overall competitiveness of the election, which in turn affects participation. The model is estimated by maximum likelihood using county-level data from the 2008-2020 elections and validated against observed ad spending and reduced-form estimates from a border discontinuity design. Counterfactual simulations show that roughly two-thirds of the turnout differential is driven by campaign effort, with the remaining third attributable to intrinsic competitiveness.
Other Research
Why People Vote: Comparing Models of Voter Turnout
(with Maxime Cugnon de Sévricourt)
We conduct a formal comparison of five leading theories of voter turnout. Using data from U.S. House of Representatives elections and a novel dataset of state special elections, we structurally estimate pivotal-voter, leader-driven, and group-based models. We then compare the fit of the non-nested models using the Vuong and Clarke tests. The results show that the best-fitting theory depends on electoral context: a flexible calculus-of-voting model performs best in high-salience House elections, while a leader-mobilization model fits equally well in low-salience special elections.
Tempting FAIT: Flexible Average Inflation Targeting and the Post-COVID U.S. Inflation Surge
(with R. Duncan and E. MartÃnez GarcÃa)
In August 2020, the Federal Reserve replaced Flexible Inflation Targeting (FIT) with Flexible Average Inflation Targeting (FAIT), which allows inflation to temporarily exceed the 2% target. Using augmented and spillover-adjusted synthetic control methods, we estimate that FAIT raised headline CPI by about 1 percentage point and core CPI by 0.5 percentage points. Short- to medium-term inflation expectations increased by approximately 0.8 percentage points, while long-term expectations remained anchored. Effects on economic activity were minimal. Results are robust across alternative indices, estimators, control groups, and adjustments for supply chains, fiscal policy, and monetary aggregates. These findings are consistent with a steeper-than-expected post-pandemic Phillips curve in the New Keynesian framework.
Works in Progress
Historical Redistricting and the Incentives of Gerrymanders
(with Laurent Bouton, Garance Genicot, and Juan Felipe Riaño)This project uses historical Census data and redistricting maps from the 19th century to the present to study how sudden institutional changes reshaped the incentives of political actors engaged in district design.
Optimal Employee Allocation in Hierarchical Organizations
(with Juan Felipe Riaño)This project develops a formal model of bureaucratic resource allocation, examining how central authorities assign tasks and personnel across multiple layers to balance efficiency and specialization.