Home > City differences in the returns to schooling: Amenity or productivity effect?

City differences in the returns to schooling: Amenity or productivity effect?


Resumen Paper






Victor Iturra1  y Mauricio Sarrias2.

1. Universidad Católica del Norte, viturra@ucn.cl.

2. Universidad Católica del Norte, sarrias@ucn.cl.



Given different preferences to amenities, there might be some workers that have a higher marginal willing to pay to live in more amenable places to enjoy non-pecuniary benefits, while there are others who display a weak preference to location-specific attributes (Beeson 1991; Black et al. 2009). Consequently, considering the empirical data, we should find that not only wage but the return to worker characteristics, i.e. the return to schooling, should be heterogeneous across cities. Nonetheless, spatial differences on the return to schooling might be observationally equivalent to two city-specific shifts: firms’ and workers’ amenities. While more attractive cities for workers should display a lower return to schooling, amenable places for firms would shift the return upward. Accordingly, researchers face an empirical puzzle, because the net effect of more attractive cities (for firms and workers) on the return to schooling, might be positive due to a larger offsetting productivity effect due to firm amenities (Graves, 2014).

The aim of this paper is to disentangle this puzzle. In doing so, its main novelties are two. First, we propose an extension of the model developed by Black et al. (2009), by including explicitly the effect of firm amenities on the city returns to schooling. Second, we apply a novel empirical approach to the study of financial return to schooling called random coefficients, which allows us to model individual unobserved heterogeneity along with observed individual and city characteristics.

Our theoretical and preliminary empirical analysis reveal several interesting results. First, the extended model shows that under a large offsetting effects of production amenities, attractive cities for both workers and firms would display a higher return to schooling. This theoretical prediction con- trasts the model with that of Black et al. (2009), which suggests that more productivity cities would display lower returns to schooling. Second, the key theoretical predictions of the extended model are confirmed by the empirical analysis, that is production amenities are positively associated with the return to schooling, and cities endowed with locally-specific advantages to firms display a higher return to schooling, offsetting any effects derived from consumption amenities. Additionally, our results suggest that the returns to schooling vary across individual and cities, more able individuals seem to get larger benefits from larger cities.


Participación en sesión: Dinámicas regionales del mercado laboral.