Neighborhood effect and Labor Market Integration∗

The 2008 economic crisis has resulted in a dramatic increase of youth unemployment in most developed countries. A 26.5 per cent increase in their number and a rate remaining close to the crisis peak is still recorded in 2012. These striking levels are not yet a new feature for some labour market like in France and recent studies reveal major disparities according to the residential location, even at a very local level. One hypothesis is that youth may be influenced by the behavior of their neighborhood peers in getting a job. The identification of this endogenous social effect from the sorting process requires implementing specific identification strategies. Two complementary approaches are developed in this paper using representative samples of youth leaving the French educational system (Génération 1998 and Génération 2004 panel surveys from the Céreq). We first implement an instrumental variable approach using employment conditions in different nearby areas as instruments of the neighborhood employment level. Then, we assume a random assignment within the neighborhood, taking the very local variation of employment level as exogenous. In both estimation strategies, the positive impact of local employment conditions on job access remains significant suggesting that the employment situation of local peers matters to successfully enter the job market.