Henry Overman is an outstanding young scholar, who develops innovative ways to investigate key spatial issues and, in the process, often changes our perspective on them. Henry isolates difficulties with existing statistical approaches and works on reformulations, which enhance our ability to draw inferences from data describing spatial relationships in economics and geography.
One example is his work on Zipf’s Law. Together with collaborator Yannis Ioannides, Overman developed non-parametric methods for examining Gibrat’s Law which underlies the regularities summarized by Zipf’s law, and he has developed more formal tests of Zipf’s Law itself. In an extension of this work, the authors applied non-parametric methods to examine the evolution of the U.S. urban system during the last century, comparing on the roles of neighboring cities as competitors and their roles as sources of demand.
A second example is Overman’s work with Puga and Turner on urban sprawl. This massive undertaking used high altitude photo and satellite imagery to compare land use on the urban fringes in the US over a fifteen-year year period. In their first paper, published in the Quarterly Journal of Economics, the authors explored how underlying geographic, historical, and political-economic factors have affected the extent and distribution of urban sprawl across metropolitan areas. Subsequent research using these data examined whether suburbanization contributes to poor health through obesity. With careful attention to issues of measurement and identification, the authors cast doubt on the widely held view that people gain weight from inactivity when they move to the suburbs. They establish, on the contrary, that heavier people are instead more likely move to the suburbs.
Another example of careful attention to identification is Overman’s paper with Duranton on evaluating the effect of tax policy on employment and industrial location, applying a “border” discontinuity research design to detailed data from the UK.
Henry’s development of new statistical techniques to analyze spatial patterns may be best reflected in his work with Duranton on localization in U.K, taking as a benchmark Ellison and Glaeser’s “dartboard” approach to analyzing non-randomness in location patterns. The Duranton-Overman innovations are twofold. The first is a detailed and creative geo-coding of data, to remove the usual discreteness in the analysis of space. The second is the description of location patterns by distributions of pairwise distances between plants (in continuous space). With these measurements, the authors are able to use distributions of the pairwise distances between plants in different industries to compare the extent of within-industry agglomeration; they also develop techniques to make inferences about when estimated distributions differ significantly from those which would arise under purely random location choices. The work, published in the Review of Economic Studies is sophisticated and innovative.
Finally we note that Henry has taken seriously the role being a “good citizen” by contributing to public debate and to academic knowledge on an international scale. He has been a powerful force in the development of the urban economics stream of papers in the NARSC meetings and has given generously of his time to making these meetings a success.
For these reasons, we are pleased to recommend that the North American Regional Science Council present the 2009 Geoffrey J. D. Hewings award to Henry Overman.