Research
Unemployment and Forward-Looking Congressmen
(JMP) - Draft
Abstract: Higher unemployment alters ideology in Congress through two mechanisms: a composition effect and a moderation effect. Using survey and electoral data, I find that unemployment is negatively correlated with incumbents' vote share and positively correlated with moderate positions, reflecting each of the mechanisms. I develop and calibrate a dynamic model of electoral competition where two forward-looking candidates compete for a seat in the House of Representatives. Candidates choose their positions in response to ideological drift and changes in district economic conditions, under the assumption that bad economic conditions hinder the incumbency advantage and make challengers more likely to win, creating incentives for moderation. Because candidates understand their political stances persist, they anticipate future economic conditions and ideological drift and optimally balance present and future electoral prospects. I focus on the 112th Congress, elected in 2010 following the Great Recession, and find that adverse economic conditions were the main reason for the surge in Republican challengers, while voters’ preferences explain only a small part of this increase. Had the Great Recession not happened, Democrats would have lost only nine seats, and the 112th Congress would have looked similar to its predecessor. Because of the Great Recession, Republican incumbents’ win probability rose by 17 percentage points on average.
Household Time Allocation Decisions: Evidence from Germany's Introduction of the Minimum Wage
With Javiera Garcia - Draft
Abstract: This paper examines how income, preferences, and productivity influence household time allocation decisions, with a focus on the effects of Germany’s 2015 minimum wage introduction. Using data from the German Socio-Economic Panel (SOEP), we perform a difference-in-differences analysis followed by an event study to assess the reform’s impact on time spent on childcare, housework, and paid labor. We find that the minimum wage reduced household time spent on home production between 30-60 minutes per day, primarily by decreasing the time women allocate to childcare. This shift is sustained over time, suggesting a re-optimization of household time allocation. We then propose a structural model where spouses receive wage offers and jointly allocate time to supplying labor market hours and to producing a public good. We estimate the model and find that men have a higher preference for leisure and women are more productive in providing the public good. Differences in wage offers largely explain disparities in paid working hours, while preferences and productivity account for most variation in home production hours. Finally, our estimated model suggests that a minimum wage of €10 would have increased women’s labor participation by 14%, which is 4 percentage points higher than observed with the actual minimum wage. These results suggest that, in the absence of other frictions, minimum wage policies can help reduce the gender wage gap.
Correlated Innovations and Biased Decision-Makers: A Tale of Policy Take-Up in the States
With Javiera Garcia
Abstract: We present a model of policy adoptions across US states, where the value of those policies is correlated. Our model consists of districts with potentially biased decision makers who care about re-elections and households that vote for these decision makers. We show that introducing correlation makes decision makers behave more frequently as if they were unbiased. This happens because there is more public information about the signal and thus it is more costly for the biased decision maker to pick their favorite action. The model allows for learning among districts while relaxing strict assumptions that the literature has made so far. We then show motivating evidence of this mechanism with US data.