15 mai 2009, 14h00–15h30
Toulouse
Salle MF 323
Department Seminar
Résumé
This paper estimates the impact of the reason for layoff (economic or personal) on the length of the subsequent unemployment spell, exploiting the French Labour Force Survey. In order to control for an eventual endogeneity of the reason for layoff, a joint model of the type of layoff and unemployment duration is proposed. The length of the unemployment spell is estimated to be significantly shorter after a layoff for economic reasons than after a layoff for personal reasons, especially for nonsupervisory workers with less than a high school degree. We interpret these findings as a sign that the type of layoff is used by prospective employers as a signal of unobserved worker’s productivity.
Mots-clés
layoffs; unemployment; duration; signals;
Codes JEL
- C34: Truncated and Censored Models • Switching Regression Models
- C35: Discrete Regression and Qualitative Choice Models • Discrete Regressors • Proportions
- J63: Turnover • Vacancies • Layoffs
- J64: Unemployment: Models, Duration, Incidence, and Job Search