Labor Markets With Endogenous Job Referral Networks PDF Download

This dissertation consists of three chapters. The first chapter empirically studies the effects of job referrals in the labor market, and the second chapter provides a theoretical explanation on the empirical findings. The third chapter explores the aggregate implications of job referrals quantitatively. In the U.S labor market, a majority of workers use referrals when searching for a job, but relatively little is known about how the quality of information transmitted through referrals is determined. In the first chapter, using recent survey data, I document that the effects of referrals are different between the currently employed searchers and the unemployed. Job referrals increase wages and job-finding probability for the employed searchers, but have statistically insignificant effects for the unemployed. In the second chapter of my dissertation, I build an equilibrium model of on-the-job search and referrals to capture these patterns and use it to understand how referrals affect the aggregate labor market. A novel feature of the model is that the extent of information transmission through referrals is endogenously determined by a strategic game. In equilibrium, referrals are endogenously more informative when the match surplus for firms is low. Consequently, the employed searchers benefit more from referrals as they search for high wage jobs, and the quality of information transmitted through referrals improves during recessions. In the third chapter, I calibrate the model and quantify the effects of job referrals on the aggregate labor market. The results show that referrals increase steady-state welfare substantially about 3-5% through better match quality. Referrals increase inequality, but mostly by benefiting workers who can use social connections.