MEASURING PERMANENT RESPONSES TO CAPITAL-GAINS TAX CHANGES IN PANEL-DATA

We use panel data and information about differences in state tax rates to separate the effects of transitory and permanent tax rate changes on capital-gains realizations behavior. The estimated effect of perman ent change is substantially smaller than the effect of transitory chan ge. The difference is even larger than differences between estimates f rom past micro data studies, which have primarily measured the transit ory effect, and time-series studies, which have primarily measured the permanent effect. Our results resolve a long-standing conflict betwee n micro data and time-series studies of how marginal tax rates affect capital-gains realizations behavior.