Essays on local labor markets

Abstract

This thesis studies empirically several issues regarding the functioning of local labor markets. In Chapter 1, I follow the methodology developed by Autor, Dorn, and Hanson (2013) to estimate the impact of Chinese imports competition onto French local labor markets, with an emphasis on the spill-overs e ects beyond the manufacturing sector on the structure of employment and wages. Local employment and total labor income in both manufacturing and non-manufacturing are negatively a ected by rising exposure to imports. Imports competition from China polarized the local structure of employment in the manufacturing sector. Hourly wages distribution is negatively a ected but overall wage dispersion is not increased. The non-traded sector even experiences a decrease in lower-tail inequality. Exploiting geographical variation in the bite of the minimum wage, I nd evidence suggesting that the minimum wage explains this e ect. In Chapter 2, I use a re nement of empirical strategy in Chapter 1 to look at whether communities suddenly a ected by rising economic integration with low-wage countries tended to vote more for the far-right parties over the last four French presidential elections. I nd evidence of a small but signi cantly positive impact of imports competition exposure on votes for the far-right: a one standard-deviation increase in imports-per-worker causes the change in the far-right share to increase by 7 percent of a standard deviation. Further results suggest that this e ect has been increasing over the time period considered. We conduct a simple sensitivity test supporting the notion that (i) omitting local share of immigrants is likely to bias our estimate downward, and that (ii) this bias is likely to negligible. In Chapter 3, co-authored with Camille H emet, we study the impact of local diversity on labour market outcomes, at two di erent level of aggregation: local labor market and i immediate neighborhood. We nd that employment correlates positively with local labor market diversity, but negatively with neighborhood diversity. Using an instrumental variable approach to deal with local labor market diversity drives the positive correlation to zero, con rming the suspicion of self-selection. Regarding neighborhood diversity, we adopt the strategy of Bayer et al. (2008), taking advantage of the very precise localization of the data: the negative e ect of diversity is reinforced. We also show that nationality-based diversity matters more than parents' origin-based diversity, giving insights on the underlying mechanisms. In Chapter 4, co-authored with Camille H emet, we exploit some speci cities of the French Labor Force Survey, in order to detect the presence of referral networks among neighbors. We show the presence of referral networks, provide extensive robustness checks and investigate two rather understudied issues in the literature: (i) what kind of job transition are local referrals associated with (job-to-job or unemployment-to-job), (ii) how has the strength of local referral e ects evolved overtime?