| Abstract: | The incidence of local property taxes, specifically the degree of capitalizing of property tax differentials in market prices, remains an unresolved question. The studies by Oates (1969) and McMillan and Carlson (1977) , which come to contradictory conclusions typify the diversity of opinion on the question. Many of the previous studies suffer from research design problems including, the use of census data for market values and property tax levels on real property; failure to control for differences in public services quality among the cities samples; and use of assessed value as a proxy for market price. A theoretical model in developed to explain the capitalization effect of property tax differentials within an urban area. Because the model is exactly identified, a single reduced from equation relating market value to a set of physical characteristics of the property and the level of property taxes can be estimated using ordinary least-squares techniques. The function is estimated using data drawn from 107 residential property sales made during 1976. Estimated parameters for the other equations then are determined indirectly. Actual market price and property tax data for each parcel ware used. It is found that property tax differentials on otherwise comparable properties are associated with significant market value differentials. Because the assessed value ranges from 7 to 20 percent of market value, the property tax procedure in the area studied appears to have randomly generated large capital gains for some and capital losses for others. Because of the limited data base, the statistical results are only indicative; the model, howevfer, should have wide applicability in a variety of questions concerning the incidence of property taxes. |
