Prior studies show that taxes matter for the residential locations of high-income earners. But, states raise a significant share of income taxes from nonresidents, especially superstars. Using superstar athletes and variation in state tax rates, we provide causal evidence on the effect of the net-of-participation tax rate on the location of labor supply. The elasticity of working in a state is 0.32, with larger effects at the top of the earnings distribution. Our results suggest a novel margin of labor supply responses for top-earners: the spatial relocation of labor supply by nonresidents.
A tax system approach suggests that the definition of the tax base is as important as the tax rate, yet most of the research in public economics focuses on the tax rate. This paper examines the effects of the taxation of food on retail food sales in border areas. Using border pair identification and the Kilts Nielsen Retail Scanner data set, I estimate the effect of both changing the rate at which food is taxed and the choice to include food in the tax base. Overall, I find that reductions in the food tax rate modestly reduce the sale of food in retailers in border areas with a 1 percentage point increase in the gross tax rate of food reduces sales by .8 percent in my primary specification while traditional panel methods with border areas seem to estimate effects over twice as large. Then using a generalized difference-in-difference specification, I find that the removal of food from the tax base does not appear to have any additional effects on the sale of food besides through the rate reduction. These results suggest that consumers treat the change in the tax base as an equivalent rate reduction. Given the gradual decline in the tax rates before removal and the large awareness of cross border tax differentials these results may not scale to the entire population.