Abstract

This thesis examines three fundamental questions in labor economics. The first chapter constructs a multi-sector search-and-matching model with sector-specific productivity. In the model, workers receive idiosyncratic productivity shocks that make them move across sectors. Within this framework, we explore how unemployment insurance (UI) benefits affect unemployment, labor mobility, and per-worker productivity. The key finding is that UI benefits raise per-worker productivity regardless of whether sector-specific productivity features an absolute or comparative advantage. The quantitative impact of UI benefits on per-worker productivity critically depends on the dispersion and correlation of sector-specific productivity.
In the second chapter, we examine how the skewness of the wage offer distribution affects a worker's reservation wage. Building on the classic job search model, we study how mean-and -variance preserving (MVP) skewness affects the reservation wage. For this purpose, we construct an analytically tractable wage offer distribution that preserves its mean and variance while tuning skewness. Our analysis reveals that the relationship between the MVP skewness and the reservation wage is non-monotonic. The level of UI benefits plays a crucial role in how skewness affects an unemployed worker's decision.
In the third chapter, we construct, estimate, and validate a semi-structural dynamic schooling enrollment model with unobserved heterogeneity. Using data from a randomized experiment in Canada, the Future to Discover (FTD) project, we estimate a model for control-group students aged 16-26. The estimation results suggest that both short-run credit constraints and long-run family environment and ability factors play significant roles in explaining schooling enrollment decisions. To validate the model, we compare its predictions to observed enrollment patterns and show that it closely matches the data, both within- and out-of-sample. Counterfactual policy experiments show that the original design of the Learning Account (LA) program would have led to a larger increase in post-secondary enrollment than the version implemented. Overall, targeted and front-loaded financial aid is the more effective way to increase enrollment in the post secondary education.