Abstracts Economics

Migration, job change and wage growth among young men

by Jeffrey J. Yankow

This dissertation explores the relationship between migration, job mobility and wage development during the early stages of the working career. Using data drawn from the National Longitudinal Survey of Youth 1979, the first section provides an extensive description of the job-changing behavior of young male workers. Attention is focused on the between-job wage growth accompanying job transitions, comparing the returns from changing jobs across labor markets relative to local job changes. Between-job wage growth regression analysis reveals that, holding other factors constant, workers receive a positive return to migration above the return to job changing as predicted by the human capital model of migration. Workers migrating between jobs collect an additional 3.2 percent wage boost above the average return to within-location job change. However, when the sample of job transitions are partitioned along education level, significant between-job returns are only found for workers with twelve or less years of completed schooling. Recognizing that economic rewards to migration need not be forthcoming immediately upon a change of locations, the next portion of this research complements much of the preceding work by allowing for the full pecuniary gain to be realized over an extended time horizon. Time-varying returns are measured using an extended panel of data to estimate a more flexible earnings specification than used in previous studies of migration. In particular, this research demonstrates that young migrants receive significant positive returns to geographic mobility. These pecuniary returns generally accumulate over a five-year period following migration, during which time migrants experience superior wage growth vis-à-vis non-migrants. Starting at levels nearly identical to non-migrants in the years just prior to migration, migrant wages increase steadily over the first five years post-migration relative to the non-migrant benchmark. After five years, migrant wages peak nearly 5 percent higher than the non-migrant wage level. In accord with the findings of the previous section, extended returns are especially pronounced for highly educated migrants. The final section estimates a Probit model of migration accounting for the selectivity associated with the decision to change jobs. Because migration is only observed conditional on a change of employers, full model specification necessitates both a migration and job change equation. The model is shown to be particularly useful for disentangling the impact of variables on the decision to migrate from their separate effect on the decision to change jobs. In particular, this section demonstrates that simple migration models failing to account for job change in estimation have computed coefficients for explanatory variables that are, in many cases, the summation of two oppositely signed factors affecting the likelihood of migration and job change.