52nd Congress of the European Regional Science Association: "Regions in Motion - Breaking the Path", 21-25 August 2012, Bratislava, Slovakia
One of the classic predictions of the urban economic theory is that high-income and low-income households choose different residential locations and therefore, conditional on workplace location, have different commuting patterns. Empirical tests of this theory are not standard, due to unobserved heterogeneity of workers, the presence of residential moving costs and a reverse causational relationship of income and distance. This study seeks to examine the long-run causal effect of household income on the workers' commute. Using German panel data, our empirical approach entails analysing whether changes in household income lead households to locate farther from their workplace - or closer. In our study, we avoid the need of using instrumental variables to deal with endogenity of household income by observing that commuting may affect wages only when workers change job. Only selecting observations of individuals who keep the same employer and job over the observation period, ensures that an observed change in distance must be due to (exogenous) residence relocation. By formulating the model in terms of first differences, we deal with unobserved heterogeneity. In addition, we focus on changes which occur during a period of at least five years for workers who during this period moved residence at least once. In this way, we identify long-term effects, and avoid that due to the presence of residential moving costs, few workers immediately change commuting distance by moving residence. Our empirical study shows that the long-run income elasticity of commuting distance is positive and around 0.18. The estimated elasticities are higher for single wage-earners than for dual wage-earners, and higher for female workers than for male workers. At least for Germany, these results suggest, as workers tend to commute to workplace locations where land prices are higher, that the income elasticity for residential space exceeds the income elasticity for commuting costs. One of the main consequences of this study is then that in the absence of historic city centres amenities or other amenities close to workplace centres, rich households tend to move farther away from the city centre, as is observed in most US cities. This is in contrast to the results of the seminal paper by Wheaton (1977, AER) which states that the standard urban economics model accidentally does not explain variation in household income over space.