LAUSR

Abstract We apply an approach to building a dense and consistent data set for income inequality that was developed for international comparisons to the case of the fifty states (and the District of Columbia) within the United States. This permits us to measure the change of economic inequality year-to-year for each state going back to 1969, something that was previously difficult to do for years before 2000, owing to the small sample size of the Current Population Survey and the fact that the Census is conducted only once in ten years. Given that US presidential elections are decided on a state-by-state basis through a winner-take-all allocation of votes in the Electoral College, we ask whether and to what degree levels or changes of economic inequality at the level of individual states affect the partisan alignment of those states and therefore the outcome of US presidential elections. There is a strong association, and one that suggests an economic model of current American presidential politics, as well as making a prediction for its future direction.