« Back to Results

Urban Economics

Paper Session

Saturday, Jan. 6, 2018 10:15 AM - 12:15 PM

Marriott Philadelphia Downtown, Meeting Room 415
Hosted By: Econometric Society
  • Chair: Minchul Shin, University of Illinois

Local Labor Markets and Human Capital Investments

Russell Weinstein
University of Illinois-Urbana-Champaign


I study whether human capital investments are based on local rather than national demand, using three shocks with differential local effects: the dot-com crash, the 2008 financial crisis, and the shock making Delaware a financial headquarters. Event-study analyses show universities more exposed to sectoral shocks experience greater changes in sector-relevant majors. Using students' home and university locations and nearest-neighbor matching, I develop a test for whether information frictions explain this local elasticity, separately from migration frictions. Information frictions do not appear to explain the result. Findings are consistent with migration frictions, implying encouraging investments based on national demand may increase mismatch.

Residential Segregation and Social Segregation by Race

Kirsten Cornelson
University of Toronto


In this paper, I examine a determinant of social segregation by race in the United States:
physical distance. Because U.S. cities are highly segregated, the time cost of interacting with a
member of another race is typically higher than the cost of interacting with a same-race friend.
The goal of this paper is to quantitatively assess the importance of this channel in explaining
why people typically interact with members of their own race. Based on external estimates of
consumers' costs of time, I simulate the frequency of cross-racial interactions that would occur
if only distance mattered in determining individuals' choice of interaction partners. I compare
the simulation results to a new measure of the actual frequency of inter-racial interactions based
on Flickr photographs. I estimate that 25-30% of social segregation for whites in the U.S. is
attributable to physical distance alone.

Metropolitan Land Values

David Yves Albouy
University of Illinois
Gabriel Ehrlich
University of Michigan
Minchul Shin
University of Illinois


We estimate the first cross-sectional index of transaction-based land values for every U.S. metropolitan area. The index accounts for geographic selection in location, and incorporates novel shrinkage methods using a prior belief based on urban economic theory. Land values at the city center increase with city size, as do land-value gradients; both are highly variable across cities. Urban land values are estimated at 1.5 times GDP in 2006. These estimates are higher and less volatile than estimates from residual (total - structure) methods. Average values in the five most expensive metros are 35 times higher than in the five cheapest.
Bruce Ian Sacerdote
Dartmouth College
Ricardo Perez-Truglia
University of California-Los Angeles
Victor Couture
University of California-Berkeley
JEL Classifications
  • R2 - Household Analysis
  • R3 - Real Estate Markets, Spatial Production Analysis, and Firm Location