Publication Date
11-11-2024
Award Type
Honorable Mention
Dissertation Advisor
John Friedman
Abstract
In this dissertation, I study how public policies and economic shocks impact workers, employers, and the structure of the labor market, aiming to inform optimal policy design and reduce existing inequalities. Chapter 1 studies the optimal design of unemployment insurance financing policies, comparing two approaches to assign unemployment tax rates to employers: assigning individualized rates in proportion to the benefit spending resulting from employers’ layoffs (experience rating) and assigning the same tax rate to all the employers (coinsurance). I derive a sufficient-statistics formula defining the optimal degree of experience rating through a tradeoff between the marginal benefits and costs of coinsurance. Then, I estimate with quasi-experiments or calibrate these benefits and costs for Colorado and South Carolina to evaluate the optimality of their policies. I find that the marginal cost of coinsurance exceeded the benefit in South Carolina, suggesting that its degree of experience rating was suboptimal. Consistent with its higher degree of experience rating, I find the opposite for Colorado. Chapter 2 studies whether gender quotas governing municipal elections promote the election of female mayors. The quotas mandate minimum female representation in electoral lists and government bodies, but do not directly target mayoral positions. Their impact on mayoral representation hinges on the existence of spillover effects. I leverage the adoption of three gender quota policies in large Italian municipalities and use event-study and regression discontinuity methods to decompose the aggregate effects of the quotas into mechanical effects, due to compliance, and residual spillover, effects. The quotas increased female representation in lower-level government positions beyond legal requirements but had no impact on mayoral positions. Chapter 3 investigates gender disparities in the labor market effects of COVID-19 for skilled Ugandan workers. Leveraging a high-frequency panel, we find that the Ugandan lockdowns disproportionately reduced female employment, generating a novel employment gender gap which persisted for eighteen months. Additionally, the lockdowns shifted women from wage-employment to self-employment and into unskilled sectors, and widened the gender pay gap. Women’s pre-pandemic sorting into severely hit economic sectors and childcare responsibilities induced by schools’ closure explain up to 65% of the employment gap.
Link to dissertation full text