The urban wage growth premium: sorting or learning?

This paper is concerned with the urban wage premium and addresses two central issues about which the field has not yet reached a consensus: first, the extent to which sorting of high ability individuals into urban areas explains the urban wage premium and second, whether workers receive this wage premium immediately, or through faster wage growth over time. Using a large panel of worker-level data from Britain, we first demonstrate the existence of an urban premium for wage levels, which increases in city size. We next provide evidence of a city size premium on wage growth, but show that this effect is driven purely by the increase in wage that occurs in the first year that a worker moves to a larger location. Controlling for sorting on the basis of unobservables we find no evidence of an urban wage growth premium. Experience in cities does have some impact on wage growth, however. Specifically, we show that workers who have at some point worked in a city experience faster wage growth than those who have never worked in a city.