Secular Labor Reallocation and Business Cycles

We revisit an old question: does industry labor reallocation affect the business cycle? Our empirical methodology exploits variation in a local area's exposure to industry reallocation based on the area's initial industry composition and national industry employment trends for identification. Applied to confidential employment data over 1980-2014, we find sharp evidence of reallocation contributing to higher local area unemployment if it occurs during a national recession, but little difference in outcomes during an expansion. A multi-area, multi-sector search and matching model with imperfect mobility across industries and downward nominal wage rigidity can reproduce these cross-sectional patterns. The same model implies that labor reallocation substantially amplifies the increase in aggregate unemployment during recessions, but has little impact on aggregate employment in expansions.