Abstract
In June of 1969 a tax substitution referendum was submitted to the voters in Oregon. The referendum specified that a general state retail sales tax be adopted. The revenues would be used to provide property tax relief. A rational choice model was utilized to predict voting behavior. The behavioral assumption was that the voter would vote for the referendum if his expected sales tax payment was less than his expected reduction in property tax payments. Using 1960 Census data, a joint distnbution of income and property values was developed. The expected sales tax and expected property tax reduction by county, by family was then computed to predict the referendum outcome. The predicted statewide vote for and against the referendum was very close to the actual totals. The results were consistent with the assumption that the voter maximizes his self interest narrowly conceived in his voting behavior.

This publication has 0 references indexed in Scilit: