When Work Moves: Job Suburbanization and Black Employment

Abstract

I provide novel evidence that job suburbanization has been an important cause of the decline in black employment since 1970, exploiting variation across US local labor markets. I rst present four stylized facts suggesting that any relationship between the two is driven by spatial frictions and not changes in labor demand or supply. I show that: (1) job suburbanization is not associated with signicant changes in the types of work performed; (2) suburbanization is driven by the location decisions of entrant rms rather than the spatial distribution of exiting rms; (3) these entrants locate in areas where blacks are ex-ante less likely to work than whites; (4) despite these shifts in the location of rms, black workers remain equally less likely to work in these suburban areas. I then show that job suburbanization is associated with substantial declines in black (relative) employment rates. Instrumenting for suburbanization using central city intersections with the interstate highway system yields similar estimates, corroborating a causal interpretation. Job suburbanization can explain the majority of the relative decline in black male employment over this period.