Interacting agents, spatial externalities and the evolution of residential land use patterns

We develop a model of land use conversion that incorporateslocal spillover effects among spatially distributed agents.The model is used to test the hypothesis that fragmented patternsof development in rural-urban fringe areas could be due to negativeexternalities that create a ‘repelling’ effect amongresidential land parcels. Identification of the hypothesizedinteraction effect is complicated by unobserved, spatially correlatedheterogeneity. Using an identification strategy that boundsthe interaction effect from above, we find empirical evidencethat is consistent with a theory of negative interactions amongrecently developed residential subdivisions in exurban Maryland.The result offers an alternative explanation for low densitysprawl to that which is frequently posited in the economicsliterature and one with potentially quite different efficiencyimplications.