Workplace-anchored migration in US counties

Residential amenities and access to employment are major factors in migration decisions. Yet traditional migration models are unable to reveal the causes (and effects) of migration because current data capture only movers who change their place of residence; depending on how far they move, these migrants may or may not also change their jobs. Migration flows thus could be categorized into two groups depending on whether the migrants also change their workplace. In this paper, we identify the number of movers who do and do not change their workplace by using overlapping county-to-county migration and commuting data. We refer to this as workplace-anchored and unanchored migration. Then we compare the local factors that affect both types of migration. Our analysis reveals that the most important local factors that separate workplace-anchored and unanchored migration are the poverty rate, commuting time, and age demographics of the origin and destination counties.