STATE AND LOCAL GOVERNMENT POLICIES AND THE LOCATION CHOICE OF HOUSEHOLDS: AN ANALYSIS OF POLYTOMOUS CHOICE (MIGRATION, AMENITIES)

The effect of state and local government policies on the location choice of households who move between areas is investigated in this research. The primary focus is placed on the effect of education policy and property taxation on location choice. The hypothesis is that education policy has a positive impact and property taxation a negative impact on the location choice of households. It is further hypothesized that these variables will have a differential impact on households with differing characteristics.;Education spending was not found to be a significant determinant of location choice for any of the subgroups examined, including households with children. Property taxation, on the other hand, was found to be a significant determinant of location choice for households with and without children and renters. Although property taxation was not found to have a significant effect on the location choice of owners, differential results were observed when this group was further dissaggregated: owners without children were found to be significantly affected by property taxation; owners with children were not found to be significantly affected by property taxation. Lastly, local income taxation was found to have a significant effect on the location choice of each subgroup examined.;In this research the household is viewed as choosing among J mutually exclusive discrete alternatives. Thus, the problem of location choice is analyzed empirically in a discrete choice (logit) model. The effect of place attributes on location choice is analyzed separately for those groups of migrants hypothesized to be differentially affected by specific policy variables, reducing the possibility of aggregation bias. Measures of the effective property tax rate, local imcome tax rate and state income tax rate are included in the estimating equation to capture the tax cost of locating in an area. The other variables included in the analysis that measure place attributes include: education spending, per capita income, the unemployment rate and a climatic variable. The income and education variables are deflated by an inter-city cost-of-living index. The model is estimated using micro data on the location choice of households extracted from the 1976 and 1977 Annual Housing Survey.