The Economic Recovery Tax Act of 1981 lowered the top marginal federal income tax rate from 70% to 50%. Hendren and Sprung-Keyser (2020) use estimates of the response to taxation from Saez (2003) to compute the MVPF of this policy change. Saez (2003) studies the impact of variation in the marginal tax rate induced by high rates of inflation combined with the fact that rates are set in nominal units. He estimates an elasticity of taxable income with respect to the after-tax keep rate of 0.31.
Hendren and Sprung-Keyser (2020) translate this into an implied MVPF of the 1981 top tax rate reform. They arrive at an MVPF that is infinite, but note the statistical imprecision of the estimates lead them to be unable to reject a confidence interval that includes an MVPF of 1 and an MVPF near infinity. Although the elasticity of 0.31 is consistent with findings in more recent time periods, the high tax rates in 1981 lead to large fiscal externalities for a given elasticity, which amplifies both the point estimate and the sampling uncertainty in the MVPF for the reform. Hendren and Sprung-Keyser (2020) suggest this implies a value to future work that provides more precise estimates of this elasticity.
_________
To compute the MVPF, Hendren and Sprung-Keyser (2020) use the equation
where
where
is the Pareto parameter of the income distribution and
is the elasticity of taxable income, y, with respect to one minus the tax rate, t.
Hendren and Sprung-Keyser (2020) measure t as the sum of the federal income tax rate and a 5% state income tax rate assumption. In practice, the reforms are discrete changes in t. To account for this, Hendren and Sprung-Keyser (2020) compute the fiscal externality FE separately for the pre- and post-reform tax rates, and then take an average of the two FEs. Appendix F of Hendren and Sprung-Keyser (2020) provides further details.
The key additional parameter beyond the elasticity of taxable income is the Pareto parameter of the income distribution. Atkinson et al. (2011) find a value of 2.3 for 1981. Combining with the elasticity of taxable income of 0.31 implies an infinite MVPF, with a wide confidence interval of [0.945, \infty]. This means Hendren and Sprung-Keyser (2020) cannot reject a null hypothesis of an MVPF of 1 or an infinite MPVF.
Pays for Itself
Atkinson, Anthony B., Thomas Piketty, and Emmanuel Saez (2011). “Top Incomes in the Long Run of History.” Journal of Economic Literature, 49(1): 3-71. DOI: https://doi.org/10.1257/jel.49.1.3
Hendren, Nathaniel and Ben Sprung-Keyser (2020). “A Unified Welfare Analysis of Government Policies.” The Quarterly Journal of Economics, 135(3): 1209–1318. DOI: https://doi.org/10.1093/qje/qjaa006
Saez, Emmanuel (2003). “The Effect of Marginal Tax Rates on Income: A Panel Study of ‘Bracket Creep.’” Journal of Public Economics, 87(5-6), 1231-1258. DOI: https://doi.org/10.1016/S0047-2727(01)00178-5