Population And Employment Density Functions Revisited: A Spatial Interaction Approach

ABSTRACT The gravity‐based Garin‐Lowry model, which generates urban population and service employment distributions for a given pattern of basic employment, has often been used in case studies. However, few analytical results or general functional forms have been derived from it. To remedy this shortcoming, this article proposes a generalized urban spatial structure and transportation network, and adapts the Garin‐Lowry model to simulate both population and service employment densities in this hypothetical, yet realistic, city. The model is solved numerically while varying exogenous factors such as the distance friction coefficients and the spatial distribution of basic employment. The computed density functions often display regularities and features commonly observed in empirical density analyses. The results are generalized by estimating, via regression analysis, density functions over a large sample of simulated density patterns, pointing to the critical importance of transportation costs and basic employment distribution, and providing a basis for further empirical studies.