Job Search and Intermediation under Discrimination: Evidence from Terrorist Attacks in France

Using detailed, high frequency data on potential job matches made through the French Public Employment Service (PES), I present evidence showing that search intensity both by and for minority jobseekers is highly sensitive to a shock that increases bias against their type. In the 10 weeks following the January 2015 “Charlie Hebdo” attacks, employers significantly reduce their search for minorities – jobseekers defined as having a first name of Arabic origin – to fill their vacancies as compared to majority jobseekers – those with classically French sounding first names. Minorities also drastically reduce their job search intensity after the shock. These drops are oset by a substantial increase in matching eort made by job counselors for their minority jobseekers after the shock. This counselor “compensatory eect” is driven by counselors who are themselves minorities and for majority counselors who specialize in getting the most marginalized jobseekers back to work. In addition, these eects are strongest in areas of low latent discrimination, proxied for by the local extreme-right vote share. Overall, I find no significant employment eects, but this belies strong heterogeneity: Significant negative employment eects on minorities are observed in micromarkets outside of the job counselors’ purview. This suggests that labor market intermediaries can play an important role in mitigating adverse shocks that reduce the eciency of the labor market matching technology.