We explore urban earnings premiums gained by native individuals born 1979 to 1981 in rural Norway who, after completing their education and starting fulltime work, moved from a rural to an urban area. We employ for the first time in this field the adjusted interrupted time-series (AITS) model, which offers a superior means of controlling for selection than conventional fixed effects and difference-in-differences estimators because it controls for pre-move level and trajectories of earnings. We find that estimates produced by conventional specifications understate the static premium and overstate the dynamic premium because they fail to account for the upward trajectory of earnings that urban migrants (regardless of gender and education) evince before they move. Parameters estimated from AITS models jointly stratified by gender and education also reveal more heterogeneity in urban earnings premiums than has typically been revealed. Better-educated workers get larger static earnings premiums when they move to the city, men gain more from such a move than women with the same education, and the gender gap for those with less education is much larger. If our results may be generalized, they would suggest that cities primarily generate wage premiums through agglomeration-based efficiencies that are heterogeneously distributed across urban employers according to their capital and knowledge intensivity.
Discussion(0)
No comments yet. Be the first to comment.