I would like to thank Chris Flinn for his guidance and constant encouragement. I would also like to thank for highly valuable supervision Raquel Fernandez and Ricardo Lagos and for helpful comments Jim Albrecht, Alberto Bisin, Daniele Checchi, Xiaohong Chen, Daniela Del Boca, Paolo Dudine, Hanming Fang, Anna Fruttero, Chris Ferrall, Andrea Ichino, Boyan Jovanovic, Mike Keane, Donghoon Lee, Alessandro Lizzeri, James Mabli, Francesc Ortega, Vincenzo Quadrini, Shannon Seitz, Giorgio Topa, Frank Vella, Gianluca Violante, Susan Vroman, and participants at seminars in the following Universities: NYU, Georgetown, Yale, Wisconsin-Madison, Penn State, UC-Riverside, Rutgers, Milan, and Northwestern; and at the following conferences: 2005 Duke Conference on Structural Models, 2005 EEA, 2004 SED, 2003 ESPE, 2003 NASM of the Econometric Society. An editor and two referees have been extremely helpful in improving the article. All errors are my own. Please address correspondence to: Luca Flabbi, Georgetown University, Department of Economics, 37th & O Streets, NW, Washington, DC 20057, U.S.A. Phone: 202-687-6213. Fax: 202-687-6102. E-mail: email@example.com.
GENDER DISCRIMINATION ESTIMATION IN A SEARCH MODEL WITH MATCHING AND BARGAINING*
Version of Record online: 4 AUG 2010
© (2010) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association
International Economic Review
Volume 51, Issue 3, pages 745–783, August 2010
How to Cite
Flabbi, L. (2010), GENDER DISCRIMINATION ESTIMATION IN A SEARCH MODEL WITH MATCHING AND BARGAINING. International Economic Review, 51: 745–783. doi: 10.1111/j.1468-2354.2010.00600.x
Manuscript received October 2006; revised December 2008.
- Issue online: 4 AUG 2010
- Version of Record online: 4 AUG 2010
This article develops a search model of the labor market with matching, bargaining, and employers' taste discrimination in which—under necessary but standard distributional assumption—it is possible to separately identify gender discrimination and unobserved productivity differences. The equilibrium shows that both prejudiced and unprejudiced employers wage discriminate. Maximum likelihood estimates on CPS data indicate that half of the employers are prejudiced, average female productivity is 6.5% lower, and two-third of the gender earning differential may be explained by prejudice. An affirmative action policy is implemented resulting in a redistribution of welfare from men to women at no cost for employers' welfare.