Firm Relocation and Spatial Mismatch: Evidence from Natural Disasters

The spatial mismatch hypothesis is that large differences in black/white employment patterns could be produced by race-neutral employers, were there discrimination in the housing market combined with the suburbanization of work. Debate over the hypothesis hinges on employer intentions: do they use relocation as a tool to alter the racial makeup of their workforces? Prior work has foundered on the endogeneity of employer decisions. We exploit natural disasters to study relocations where employer discretion is curtailed. We develop models of employer behaviors consistent with different theories of employer intentions and hypothesize how these behaviors would differ in the wake of disasters. We test these hypotheses on longitudinal workforce-composition data from every large establishment in the United States between 1971 and 2014. Our preliminary results are more consistent with “race-based” decisions by employers than with “space-based” decisions. We discuss the implications of these findings for work on reducing employment segregation.