Working Paper: CEPR ID: DP10030
Authors: Pierre M. Picard; Yves Zenou
Abstract: We develop a model where workers both choose their residential location (geographical space) and social interactions (social space). In equilibrium, we show under which condition the majority group resides close to the job center while the minority group lives far away from it. Even though the two populations are ex ante totally identical, we find that the majority group experiences a lower unemployment rate than the minority group and tends to socially interact more with other workers of its own group. Within each group, we demonstrate that workers residing farther away from the job center tend to search less for a job and are less likely to be employed. This model is thus able to explain why ethnic minorities are segregated in the urban and social space and why this leads to adverse labor-market outcomes in the absence of any discrimination against the minority group.
Keywords: Labor market; Segregation; Social interactions; Spatial mismatch
JEL Codes: A14; J15; R14; Z13
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| proximity to the job center (R23) | social interactions (Z13) |
| social interactions (Z13) | job search efforts (J68) |
| job search efforts (J68) | employment probabilities (J68) |
| proximity to the job center (R23) | employment probabilities (J68) |
| distance from the job center (J60) | job search behavior (J68) |
| distance from the job center (J60) | employment outcomes (J68) |
| social interactions among majority group (Z13) | job information and opportunities (M39) |
| social interactions (Z13) | employment rates for majority group (J79) |
| residential location (R30) | job search behavior (J68) |
| residential location (R30) | employment rates (J68) |