From Today's City to Tomorrow's City : An Empirical Investigation of Urban Land Assembly

Because urban areas are fundamentally constrained by the boundaries of land ownership, economic and technological shocks cannot modify the urban landscape without also changing the delineation of land. In this paper, we ask if urban land markets are capable of producing such changes in land delineation. Specifically, we test if there is enough land assembly – the legal joining together of two or more parcels of land – to put land to its highest value use. Failure to assemble land may reduce agglomerative benefits, produce urban blight, push growth to the city edge and away from the core; ultimately it may cause cities to forfeit economic growth. We develop a simple theoretical framework which provides a testable hypothesis: in the absence of market frictions, the price of land sold for assembly should not exceed the price of land sold for other uses. This hypothesis does not hold when frictions, such as holdouts and land use regulation, drive the market for land assembly. We test this hypothesis using a novel data set that follows each of the 2.3 million parcels in Los Angeles County over a twelve year period and allows us to observe all instances of assembly. Using panel data methods, we find that to-be-assembled land trades at a 40 percent premium. Thus, we find that urban land markets are subject to significant frictions that prevent assemblies and produce sclerotic urban redevelopment. Additional empirical results suggest that private market frictions, such as holdouts, play an important role in blocking assembly