The cycle of fragmentation and sprawl: a conceptual framework and empirical model

The political and spatial dimensions of US metropolitan areas are eminently interconnected through a recurring cycle of fragmentation and sprawl. In this paper we demonstrate the cycle at work by refining a land-use model developed in a previous paper and applying it to a national dataset of metropolitan counties. The recursive simultaneous-equations model is structured around five-year intervals, and enables us to observe how the political landscape (urban development patterns) at time t-5 affects spatial outcomes (municipal fragmentation) at time t. The results suggest that regulatory failure may bear as much fault for urban sprawl as the more commonly cited market failures, and that it may therefore be worthwhile to shift the focus of the sprawl/antisprawl debate from its physical to its political dimensions. Future research should focus on identifying the systematic nature of sprawl and, as an extension, the various policy levers that may be used to mitigate its negative consequences. The paper improves on previous research in four key ways: by describing the underlying theory in greater detail; by rounding out the recursive relationship; by enriching the set of interdependent variables; and by expanding the geographic scope of the model. Its primary contribution is to deepen the pool of empirical evidence linking political structure to patterns of growth and change in US metropolitan regions.