Larger pay, longer drives? Location specific wage effects on commuting distances

Over the past decades, commuting has become more and more important as a means for labor market adjustment. This is reflected in increasing commuter shares and distances. Nevertheless, identifying the determinants of commuting distances is by no means straightforward due to complex job and housing decisions. We argue that commuting results from a two-stage process where sorting leads to a temporary optimal housing/work combination that is adjusted in a second step. Due to differences in employment density, optimization results in diverse wage effects depending on the housing location. We focus on commuters who change their job and keep their place of residence. Using a mixed-effects design and matched data from the German ALWA-ADIAB survey, combined with precise wage information from administrative data, we are able to single out two aspects. Firstly, we identify the effects of wage level and wage change on commuting distance, and secondly, we analyze how these vary with the housing location. We find support for our assumptions that wage effects differ substantially as employees with higher wages generally commute longer distances but adjust their distance to a wage change predominantly in urban areas. Residents in rural areas keep their travel distances constant.