Esteban Rossi-Hansberg, Princeton University.

Department seminar. Esteban Rossi-Hansberg is a Professor in the Princeton Economics Department. He will present a paper entitled "Cognitive Hubs and Spatial Redistribution", co-authored by Pierre-Daniel Sarte and Felipe Schwartzman.

Photo of Esteban Rossi-Hansberg

Esteban Rossi-Hansberg. Photo: Princeton University


We study the allocation of workers with different occupations across U.S. cities. We propose and quantify a spatial equilibrium model with multiple industries that em- ploy cognitive-non-routine (CNR) and alternative (non-CNR) occupations. We allow for city-industry-occupation specific productivity levels that are partly determined by externalities across local workers. We estimate the relevant parameters that determine these externalities within and across occupations using a number of instruments based on past migration patterns and the location of land-grant colleges. The productivity of CNR workers in a city depends significantly on its share of CNR workers and total employment size. Heterogeneous preferences for locations and the estimated external- ities imply that the equilibrium allocation is not efficient. In equilibrium, the social value of CNR workers exceeds their private value by more than 70%. An optimal policy that benefits workers in both occupations equally, incentivizes the formation of cognitive hubs with larger fractions of CNR workers in some of today’s largest cities. These cities should become smaller, so non-CNR workers receive transfers that incen- tivize them to move to small cities that have the appropriate industrial composition. We show that the spatial redistribution of CNR and non-CNR workers implied by the optimal policy reinforces equilibrium trends observed since 1980. These trends were generated by a nation-wide increase in the share of CNR workers combined with the local CNR specific externalities and low real-estate productivity in CNR-intensive cities.


Published Aug. 15, 2019 10:14 AM - Last modified Aug. 17, 2019 10:50 AM