Many urban revitalization programs focus policy resources on spatially defined target neighborhoods. The impacts of these programs can include direct effects in the neighborhood chosen for intervention and spillover effects in other neighborhoods. These unintended and sometimes unpredicted effects may be positive or negative. This dissertation argues that without analyzing the urban spatial structure as a set of inter-related housing submarkets, planners will not be able to adequately predict and evaluate the effects of revitalization policy. In doing so, the research investigates the theorized nature of neighborhoods, discussing this socio-cultural geography in light of the theorized nature of the housing market in space. The empirical investigation, a case study of Philadelphia, defines spatial housing submarkets as distinct market segments based on housing quality. These submarkets are compared with first the housing market implied by the theories of Alonso-Mills-Muth and Tiebout, and second with the defined policy neighborhoods. The city and metropolitan area of Philadelphia are found to be highly heterogeneous in housing types, with no uniform pattern predicted by theory. Additionally, policy target neighborhoods are often comprised of different submarkets, leading to confusion in policy targeting. This research suggests that urban planners should consider the geography of housing submarkets in developing revitalization policies in order to choose appropriate geographic targets and to predict the spatial extent of market responses.