Occupational change, employer change, internal migration, and earnings

In this paper I use microdata from the Panel Study of Income Dynamics to measure the financial returns to intercounty and interstate migration for individuals in a temporal framework accounting for gains that accrue over time. Earnings are estimated for a pooled sample of migrants and non-migrants in a two-stage procedure correcting for a self-selection bias. To account for the indirect effects of migration on earnings, explanatory variables are created by interacting migration status with: (1) occupational change, (2) employer change and (3) changes in both occupation and employer. These interaction terms are then included in the earnings functions. Earnings are estimated for three years subsequent to the migration decision to account for the financial returns to migration accruing over time. Results indicate that, when estimating earnings, the use of a simple migration dummy variable will mask the indirect effects of migration on earnings.