A SPATIAL MODEL OF URBAN RESIDENTIAL LOCATION WITH PUBLIC SECTOR EXTERNALITIE

The traditional spatial models of urban areas discuss the interaction of the private business sector, the housing sector, and the residential sector. However, the local public sector and its related externalities also have important effects on the structure of urban areas. Economies of agglomeration in the provision of public services may draw households to urban areas. Diseconomies in public service provision related to factors such as congestion, labor power, and population composition may limit the growth of cities.;Very few urban models have incorporated a local public sector. The inadequate theoretical development of the local public sector's role in an urban economy is reflected in recent empirical work regarding the local public sector's impact on household location. Many of these studies may have serious biases as a result of the failure to consider the determinants of local public expenditures.;The purpose of this work is to formulate a model of residential location that fuses a well-developed local public sector with the other sectors of the urban economy. The theoretical model explicitly considers the supply of public services, the demand for public services, income redistribution and the financing of public expenditures. It demonstrates that public sector labor union power, population composition, and scale and congestion effects in public service provision affect public sector costs and outputs. These have subsequent effects on residential location decisions.;The major implications of the theoretical model are tested using a six-equation empirical model and data for 67 large SMSAs. Two principal findings result from the tests. First, population composition and public sector labor power have significant effects on public sector costs; the hypothesis of insignificant scale and congestion effects cannot be rejected. The importance of population composition and labor power suggests a bias in studies of urban residential location that have used local public expenditures as a measure of benefits. Second, quality of educational services affects the locational distribution of nonpoor households; the hypothesis of an insignificant tax effect could not be rejected.