This paper uses the neoclassical growth model to examine the extent to which a tax cut pays for itself through higher economic growth. The model yields simple expressions for the steady-state feedback effect of a tax cut. The feedback is surprisingly large: for standard parameter values, half of a capital tax cut is self- nancing. The paper considers various generalizations of the basic model, including elastic labor supply, general production technologies, departures from in nite horizons, and non-neoclassical production settings. It also examines how the steady-state results are modi ed when one considers the transition path to the steady state.