Neighborhood Effects, Urban Public Policies and Housing Values. A Spatial Econometric Perspective

Problems of spatial segregation are often associated with segmented housing submarkets where the social status and social capital of a neighborhood appear to be the main driving forces behind housing price formation. Urban regeneration policies aim to raise housing values in poor areas through the construction of new buildings with higher levels of housing services. Structural attributes, neighborhood characteristics, and accessibility variables are the determinants of housing values considered in hedonic models. In this paper, it is assumed that spatial effects in terms of spatial autocorrelation, spatial heterogeneity, and spatial externalities are additional variables worth considering for at least two reasons: (i) in econometrics, OLS estimation problems arise from the occurrence of spatial dependencies among housing values; (ii) in urban policy studies, spatial effects engender a global diffusion process extending beyond housing submarkets. The impact of social housing policies and urban regeneration policies may permeate outside the areas where they are implemented. Our case study is of the urban area of Dijon (France), where two types of urban policy have been implemented in the last three decades: social housing projects in some suburban areas, and a regeneration plan for the old inner-city. Spatial effects are introduced in the hedonic model and a spatial error model is estimated, revealing a positive and significant global spatial effect combined with the usual influence of accessibility to the CBD. We also show the negative influence of location in social housing districts and the spatial diffusion effect they exert on neighboring districts.