For many decades, the relationship between urban and rural places was well understood. Beginning in the early 20th century, however, this distinct dichotomy broke down as large numbers of businesses and people migrated out of the central city. As a result of this expansion of the urban center and the growth in suburban and exurban development, many urban fringe areas in the U.S. have become characterized by low-density and fragmented development. As a result of these changing land use patterns and the potential for both positive and negative outcomes, researchers and policymakers have become interested in understanding both the demand-side and supply-side incentive mechanisms that have led to this type of fragmented development. The objective of this research is to fill several gaps in the empirical literature on residential land conversion and land use policy by using unique micro-level data on historical subdivision development, land conversion, the platting and subdivision approval process, house prices and policy changes.
In our first essay, we build on the growing literature that looks at the effect of regulation on housing supply decisions and focus specifically on the question of whether the expected time to completion affects both the decision to develop as well as the quantity of lots chosen by the individual landowners. Using a unique micro dataset on the timing of subdivision approvals by a local planning agency and a sample selection Poisson model, we test the effects of implicit costs that arise from uncertain subdivision approval on the timing, quantity and pattern of residential subdivision development. Consistent with theory, we find that these regulation-induced implicit costs reduce the probability and size of subdivision development on any given parcel. Our results contribute to the growing supply-side literature on housing and land use, and provide a new explanation of scattered residential development as the outcome of heterogeneous regulatory costs and optimal land development.
In our second essay, we compare two competing hypotheses of exurban land development. Previous research has shown that local land use spillovers can impact land use outcomes and patterns. In this paper, we simultaneously test this hypothesis and our regulation hypothesis by combining both hypotheses into the same sample selection Poisson model to determine the power of each factor in explaining the timing and intensity decisions of subdivision developers. Our results show that, while many of the land use variables are significant, the impact of regulatory uncertainty is larger and that it provides a better explanation of the timing and intensity decisions of developers.
In our last essay, we develop a nonparametric estimation technique for spatial panel data. Using a Monte Carlo experiment, we show how extending current geographically weighted regression (GWR) models to account for temporal heterogeneity provides a better fit to the data when coefficient heterogeneity exists in both the spatial and temporal dimensions. We also show how the technique can be used in modeling real-world land use change by applying the proposed technique to a panel dataset of historical subdivision development.