Rising income inequality and tight government budgets have spurred discussions in many developing nations regarding the taxation of high-income earners. We study taxpayer responses to an increase in the top marginal tax rate in South Africa drawing on exceptionally rich tax administrative data and a transparent empirical identification design. We establish that treated taxpayers strongly reduce their reported taxable income in response to the tax reform. Our preferred estimates suggest an elasticity of taxable income of around 1.2. Responses are driven by both reductions in broad income and increases in tax deductions. While regular, third-party reported earnings remain unaffected, we find significant decreases in fringe benefits, allowances, annual incentive and bonus payments. Consistent with the latter response reflecting less effort provision by leading workers in South African firms, we find that businesses which employ workers treated by the tax increase experience a significant decline in business output after the reform.

This study estimates the elasticity of income with regard to the net of tax rate using panel data from Japanese tax returns from 2014 to 2020. The following conclusions were drawn: First, in Japan, the elasticity of income with respect to the net of tax rate ranges from 0.25 to 0.52. These estimates are greater than or close to the ETIs for most OECD countries but are much greater than those provided by previous research for Japan. Second, income elasticities for business-income earners are larger than those for salary income earners. Third, the ETI estimates are much higher than the income elasticity estimates. This result probably reflects the fact that there was no large-scale reform in income tax rates during the sample period, thus making it difficult to precisely identify taxpayers’ behavioral responses to tax rates.

This paper estimates the elasticity of broad income (EBI) with respect to the marginal net-of-tax rate for high-income taxpayers. We study the introduction of a new top US income tax bracket in 2013 using a large panel of high-income taxpayers drawn from US administrative tax records. Taxpayers in the top tax bracket experience tremendous income volatility – more than one in four have year-on-year changes in broad income of more than 30 percent. This volatility shifts some across time and drives potential bias in all possible estimators. We propose a method for estimating this expected bias and use this information to select our estimator and bound any remaining bias. Our results have important policy implications for the optimal top marginal tax rate in the US.