The demand for residence site characteristics

Abstract

The analysis of residential site choice is the primary concern in urban economics. Research on this problem is deficient in that rigorous theoretical models concentrate almost exclusively on access to the Central Business District (CBD) in this decision process while empirical models have been specified in a manner that is inconsistent with standard demand theory. Although these models possess the power of simplicity, they are unable to explain the non-monotonic location patterns which characterize urban areas without resorting to ad hoc theorizing. This article develops and tests a model which is consistent with demand theory and is also capable of explaining non-monotonic location patterns. The hedonic rent function is seen as an intermediate step in the analysis of location which allows one to construct the complete data matrix necessary for the estimation of a set of demand functions for location-specific traits. The analytical importance of the functional form of this hedonic function is briefly noted. Finally, the empirical analysis notes the importance of children in determining residential preferences, with special attention being directed to the endogeneity of children with respect to the process of residential choice. The traditional location model is generally tested by regressing some transformation of the annual rental payments (or alternatively, property value), R, on access (or lack of access) to the city center, ACC, a vector of other locational traits, Z, income, I, and a race variable, MC,