Urban spatial structure and the value of residential housing

This dissertation examines a number of recent developments in the quantitative modeling of land use and land values. The work particularly addresses a set of methods that explicitly model the economic process leading to an urban spatial structure. The study includes three separate essays that address (i) the determination of urban spatial structure, (ii) the effects of spatial fragmentation on residential real estate prices, and (iii) the construction of aggregate real estate price indexes through spatial econometric techniques. The methods employed in this study borrow from several fields ranging from actuarial science to ecology, and the studies show that each method, when properly applied, provides relevant economic information and interpretation. Each application draws from a rich dataset of land use and single family residential house prices in the City of Milwaukee, Wisconsin. Collectively, this dissertation shows how the unique needs of various economic agents leads to observed patterns of land use activities in an urban environment, and the spatial arrangement of these land use activities directly affects property values. The results suggest the urban spatial structure of the City of Milwaukee is consistent with the classical theories of urban form. The model supports the concentric zone model in that commercial activities appear clustered near the city center, followed by manufacturing and residential uses. The results also support the sector model in that direction, as well as distance, is shown to play a role in the determination of land use activities. Direction is shown to have an impact particularly for commercial, manufacturing, and low class residential land use activities. Urban spatial structure is also shown to have an impact on residential property values. Larger residential land use patches are associated with significantly higher property values which suggests that single-use zoning protects residential property values. Similarly, mixed land use practices may lead to reduced residential property values. The study also demonstrates the benefits of quantitative methods that explicitly consider the spatial aspects of land use and land value data