Lei Fang and Jun Nie
Working Paper 2014-2
February 2014

Download the full text of this paper (343 KB) pdf

The high U.S. unemployment rate after the Great Recession is usually considered to be a result of changes in factors influencing either the demand side or the supply side of the labor market. However, no matter what factors have caused the changes in the unemployment rate, these factors should have influenced workers' and firms' decisions. Therefore, it is important to take into account workers' endogenous responses to changes in various factors when seeking to understand how these factors affect the unemployment rate. To address this issue, we estimate a Mortensen-Pissarides style of labor-market matching model with endogenous separation decisions and stochastic changes in workers' human capital. We study how agents' endogenous choices vary with changes in the exogenous shocks and changes in labor-market policy in the context of human capital dynamics. We reach four main findings. First, once workers have accounted for and are able to optimally respond to possible human capital loss, the unemployment rate in an economy with human capital loss during unemployment will not be higher than in an economy with no human capital loss. The reason is that the increase in the unemployment rate led by human capital loss is more than offset by workers' endogenous responses to prevent them from being unemployed. Second, human capital accumulation on the job is more important than human capital loss during unemployment for both the unemployment rate and output. Third, workers' endogenous separation rates will decline when job-finding rates fall. Fourth, taking into account the endogenous responses, unemployment insurance extensions contributed 0.5 percentage point to the increase in the aggregate unemployment rate in the 2008–12 period.

JEL classification: E24, J08, J24, J45

Key words: unemployment, unemployment insurance benefits, matching model, human capital, labor market

The authors are grateful for useful suggestions and comments from Yongsung Chang, Russell Cooper, Greg Kaplan, Ayse Imrohoroglu, Christopher Otrok, Jose-Victor Rios-Rull, Richard Rogerson, Peter Rupert, Thomas J. Sargent, and the seminar and conference participants at the Federal Reserve Bank of Kansas City, the Midwest Macro Meeting, New York University's Economics Alumni Conference, the Hong Kong Institute for Monetary Research, the National University of Singapore, and the Society for Computational Economics. They also thank Jing Yu and Lisa Taylor for great research assistance. The views expressed here are the authors' and not necessarily those of the Federal Reserve Banks of Atlanta or Kansas City or the Federal Reserve System. Any remaining errors are the authors' responsibility.

Please address questions regarding content to Lei Fang, Research Department, Federal Reserve Bank of Atlanta, 1000 Peachtree Street, N.E., Atlanta, GA 30309-4470, 404-498-8057, lei.fang@atl.frb.org

Use the WebScriber Service to receive e-mail e-mail notifications about new papers.