Essays on public finance.

Abstract

The first chapter of this dissertation is concerned with the difficulty with identifying tax effects, due to the correlation between tax rates and other determinants of dependent variables. This chapter investigates the role of the Alternative Minimum Tax (AMT) taxpayers in reducing collinearity among regressors. Because AMT taxpayers usually face different tax rates from non-AMT taxpayers with similar incomes, AMT taxpayers can be included in the regressions to increase independent variation in tax rates. In order to test the potential contribution, the price elasticity estimates of charitable giving using the whole sample and those using the subsample without AMT taxpayers are compared. It is found that the standard errors of the price elasticity estimates substantially decline by including AMT returns in the estimation. The inclusion of AMT returns thus provides another approach to obtaining more precise estimates of tax effects. The second chapter of the dissertation deals with the fact that taxpayers probably feel uncertain about their applicable marginal tax rates when making most of their economic decisions. Therefore, the conventional estimate of tax effects based on marginal tax rates from tax returns may be misleading because of measurement error. This chapter argues that the tax effect of consumption depends on not only the income effect and the price substitution effect but also the risk effect of stochastic marginal tax rate. On the basis of the 1990 tax return data, even though we find that the conventional measure of tax price does not significantly show measurement errors, the randomness of marginal tax rates is found to cause a negative effect on charitable giving. A policy implication is that a more graduated tax system could increase taxpayers uncertainty about marginal tax rates and increase taxpayers' costs of charitable giving. In the final chapter of the dissertation, we examine the implications of the tax elasticity of taxable income for policymakers. Using the 1983--1989 SCF panel data, this chapter finds that the tax elasticity of the reported rates of return on privately held business (PHB) is positive and high. The finding supports the argument that high positive tax elasticity of reported rate of return is most likely caused by the tax avoidance activity involved in PHBs. We use the net-of-tax elasticity of the reported rate of return and the tax elasticity of PHB assets to estimate the effects of tax rate increases in the 1993 legislation on taxable income and revenue. The calculation demonstrates that the share of taxable income from PHBs is critical to estimating the revenue effects since taxable income from PHBs is very responsive to tax change.