- Unemployment Duration, Exit Rates and Re-employment Wages (joint with Edgar Preugschat) (Revised & Resubmitted at the Journal of Monetary Economics)
We document for the US that the exit rates decline severely during the first three months of unemployment, while re-employment wages barely fall. Why do workers face similar wages, yet so different exit rates as unemployment duration increases? To address this question, we build a directed search model with heterogeneous workers and incomplete information about a worker's type. Skilled workers are more productive and more likely to be suitable for the job. Firms observe a private signal, which perfectly identifies unsuitable matches. As longer spells signal lower expected productivity, firms rank applicants by unemployment duration in equilibrium. Exit rates decline with duration because of both sorting and ranking. We calibrate the model to the US economy and find that this composite mechanism explains the entire difference between the fall in exit rates and wages. The ranking by duration accounts for approximately one fourth of the difference.
After controlling for observable characteristics, we find that job-finding rates decline with unemployment duration and are procyclical, whereas re-entry wages fall much less with duration and are also much less procyclical. We study whether these outcomes can be explained on the basis of workers' time-invariant characteristics and composition variation with duration and over the business cycle. We build a search model in which firms lack information about the candidates' skills and design self-selection schemes by renting machines of different qualities. The equilibrium is separating due to output-contingent wages together with capital-skill complementarity. In equilibrium, skilled workers experience higher job-finding rates and re-entry wages. We calibrate the model to the US economy. The composition mechanism explains about 40 percent of the decline of the monthly hazard rate after three months of unemployment, yet shows a parallel wage fall, much larger than the actual one. We also show that the composition components of the job-finding rates and re-entry wages are countercyclical.
(This version: December, 2013)This paper analyzes a frictional economy in which workers are privately informed about their skills, and capital and skills are complementary in production. Firms compete for workers by choosing machine qualities and contractual offers. Unskilled workers trade off a larger return from unskilled jobs and a higher employment rate at the skilled ones. Adverse selection arises if unskilled workers crowd out the skilled ones. We show that both the constrained efficient and the equilibrium allocations are positive assortative. To achieve positive sorting, the planner designs mechanisms that compensate the unskilled for not searching in the skilled location if necessary. In contrast, firms have no means to do so, and they may end up committing to inefficiently high machine qualities to prevent unskilled workers from participating in the skilled submarket in equilibrium. Thus, the competitive search equilibrium may fail to be constrained efficient. If so, a system of transfers is welfare-improving.
Work in Progress
- Effects of Unemployment Duration on Labor Market Outcomes
- Intermediation in the Housing Market
- Effects of Switching Occupations along the Job Search