Evidence of Tax-Induced Individual Behavioral Responses

2 " The Clinton revenue estimates are based on the fallacy that taxpayers will not change their behavior in response to a 37% jump in their marginal tax rates... " Martin Feldstein (1993) Estimating the effect of income taxes on individual behaviors such as labor supply, compensation, saving, and taxable income have been focal points of economic research for several decades. Because income taxes account for nearly 80 percent of all federal revenues collected, their effects on individual behaviors and the attendant tax collections figure prominently in tax policy debates. For example, in 1993 the Clinton administration proposed increasing marginal tax rates by about one-third for the highest income Americans. In estimating how much revenue the tax increase would produce, they assumed hours worked by American taxpayers would change little and that tax collections would then rise by an amount directly proportional to the rate hike. Feldstein (1993) countered that the tax increase would substantially lower work effort by encouraging primary workers to reduce their hours of work and encouraging secondary workers in high-income households to leave the labor force, and that the tax increase would induce high-income workers to alter the form and timing of their compensation. Feldstein concluded that the combined negative behavioral responses of work effort and compensation would raise tax revenues much less than predicted, and that revenues might even fall. Whether because of or in spite of the Clinton tax policies, actual tax collections went well above both the Administration's and Feldstein's forecasts in the late 1990s, re-emphasizing our limited understanding of the behavioral consequences of income taxes. We describe here what is known, knowable, and likely unknowable about the effects of income taxes on individual and household decisions. Knowledge of the empirical consequences of income taxation on labor supply, consumption, and saving is of first-order importance not only to inform policy about the magnitudes of the possible behavioral and distributional consequences of fundamental tax 3 reform, but also to inform policy on the direction of the response. Consider first the case of labor supply. Supporters of replacing current graduated marginal tax rates with a flat tax typically cite positive labor supply consequences of a flat tax. Economic theory does not provide a definitive answer on the effect of a flat tax on labor supply over much of the income distribution. The ambiguity is from possible offsetting substitution and income effects when the after-tax wage …