Three Essays in Labor and Education Economics

Abstract

This dissertation constitutes of three papers on labor economics. In the first chapter I study how the returns to experience change across educational groups. In contrast to past work, in the construction of the experience variable I distinguish working and non-working periods after an individual leaves school. The results using this accurate measure of work history are remarkably different from those in the existing literature. I consistently find that more educated workers have a higher wage increase with actual experience but suffer a greater wage loss after unemployment periods. In addition, I propose an economic model that can rationalize these novel findings. In the model, high ability workers have greater returns to human capital. Furthermore, employers have imperfect information about workers and use past unemployment periods to predict their unobservable quality. Under these assumptions, the model predicts the empirical findings of the paper.

The second chapter is a joint work with Paola Bordón, in which we study the impact of graduating from an elite university on earnings. Using a regression discontinuity design, we estimate a substantial wage premium for individuals that graduate from a prestigious university in Chile. However, this wage premium declines as workers progress in their career. We interpret this result as employers using college prestige as a signal of workers' unobservable quality when they leave school but rewarding them based on their true productivity as they progress in their careers.

The third chapter is a joint work with John Bound, Joe Golden, and Gaurav Khanna. We build and calibrate a dynamic model to investigate the impact of immigrants on the market for computer scientists in the US. We find that wages and employment of American computer scientists would be higher if firms had restrictions on the number of foreigners they could hire.