TESTING MODELS OF DISTRIBUTIVE POLITICSUSING EXIT POLLS TO MEASURE VOTERPREFERENCES AND PARTISANSHIP

This paper tests various hypotheses about distributive politics by studying the distributionof federal spending across U.S. states over the period 1978-2002. We improve onprevious work by using survey data to measure the share of voters in each state that areDemocrats, Republicans, and independents, or liberals, conservatives and moderates. Wefind no evidence for the "swing voter" hypothesis { that is, no significant associationbetween the amount of federal funds a state receives and the fraction of independents ormoderates in the state. We also find no evidence for the "battleground state" hypothesis -no significant association between the amount of federal funds and the degree of partisanbalance in a state. Modest support is found for the \partisan supporters" hypothesis, whichconjectures that politicians will favour areas that contain a large percentage of their coresupporters.