Tax evasion is a pervasive worldwide phenomenon. It is widely believed that high personal income tax rates are partially responsible for high levels of tax evasion everywhere, especially in emerging markets. High personal income tax rates are also often associated with negative effects on the real side of the economy. Indeed, the high elasticity of taxable income with respect to tax rates typically reported in the public finance literature implies sizable deadweight losses in the presence of high personal income tax rates (e.g., Feldstein, 1995). Given these considerations, it is not surprising that over the last two decades there has been a global trend toward the gradual reduction in the level and progressivity of personal income tax rates. Recently, some countries have even taken more radical steps in the same direction.
In January 2001, Russia introduced a fairly dramatic reform of its personal income tax, becoming the first large economy to adopt a flat tax. The Tax Code of 2001 replaced a conventional progressive rate structure with a flat tax rate of 13 percent. Over the next year after the reform, while the Russian economy grew at almost 5% in real terms, revenues from the personal income tax increased by over 25% in real terms. Besides this revenue yield performance, advocates also have credited the flat tax with beneficial changes in the real side of the economy. The Russian experience would appear to have been so successful that many other countries have followed suit with their own flat rate income tax reforms, and an increasing number of countries around the world are considering the adoption of a flat rate income tax.
What happened in Russia?
While Russia’s flat tax reform has created much excitement, little solid evidence on its impact on tax evasion or the real side of the economy has been provided thus far.1 In recent research (Gorodnichenko et al., 2008), we argue that the flat tax reform was instrumental in decreasing tax evasion in Russia, and that, to a certain extent, greater fiscal revenues in 2001 and several years beyond can be linked to increased voluntary tax compliance and reporting. We also find that the productivity effect on the real side of the economy was positive, although smaller than the tax evasion effect.
As income underreporting is not observable by definition, we use data on reported consumption and income to infer tax evasion. Under the permanent income hypothesis, current consumption should equal a share of permanent income. Assuming that consumption expenditures are fully reported, the discrepancy between consumption and income, which we call the consumption-income gap, indicates that households underreport a portion of their income. We perform a series of checks to verify that the consumption-income gap is an informative indicator of tax evasion and find that the behaviour of the consumption-income gap is consistent with common tax evasion determinants.
Since Russia’s reform decreased marginal tax rates for only some groups of people, we use the variation across time and taxpayers to identify and estimate the effects of the flat rate income tax reform. We find that, ceteris paribus, the consumption-income gap decreased by 9 to 12% more for those households that experienced a reduction in marginal tax rates. That is, the most significant reduction in tax evasion was for taxpayers that experienced a decrease in tax rates upon introduction of the flat tax. We also find that this decline in tax evasion was likely due to changes in voluntary compliance, as opposed to greater enforcement efforts by the tax administration authorities. In contrast, the productivity effect, measured by the relative increase in consumption for households that faced smaller tax rates after the reform, was zero at the threshold and about 4% for those taxpayers that experienced the decrease in tax rates over a four-year period.
This strong evidence of a positive relationship between (lower) tax rates and (lower) tax evasion stands out from previous studies of tax evasion that used cross-sectional data and provided only ambiguous results. Our method of using longitudinal household budget surveys to estimate the effects of significant changes in the tax structure might be used to revisit the topic. Given the data required, it may be possible to infer tax evasion and the effects of tax reform for a wide range of countries.
Our Russian results have several important policy implications. The adoption of a flat rate income tax is not generally expected to lead to significant increases in tax revenues because labour supplies are believed to be fairly inelastic. However, if an economy is plagued by ubiquitous tax evasion, as was the case in Russia, then a flat rate income tax reform may lead to substantial revenue gains via increased voluntary compliance. At the same time, a strong evasion response suggests that the efficiency gains from the Russian tax reform perhaps are smaller than previously thought. Using observable responses of consumption and income to tax changes, we find that the tax-evasion-adjusted deadweight loss from the personal income tax is at least 30% smaller than the loss implied by the standard method based on the response of reported income to tax changes. While a flat tax offers tangible efficiency gains, its reduction of tax evasion may be most important.
Feldstein, Martin, 1995. “The Effect of Marginal Tax Rates on Taxable Income: A Panel Study of the 1986 Tax Reform Act,” Journal of Political Economy 103(3): 551-572.
Gorodnichenko, Yuriy; Jorge Martinez-Vazquez; and Klara Sabirianova Peter, 2008. “Myth and Reality of Flat Tax Reform: Micro Estimates of Tax Evasion Response and Welfare Effects in Russia,” NBER WP 13719.
Ivanova, Anna; Michael Keen; and Alexander Klemm, 2005. “The Russian Flat Tax Reform,” Economic Policy 20(43): 397-444.
Martinez-Vazquez, Jorge; Mark Rider; Riatu Qibthiyyah; and Sally Wallace, 2006. “Who Bears the Burden of Taxes on Labor Income in Russia?” AYSPS International Studies Program Working Paper, No. 06/21.
1 Some recent papers (Ivanova et al., 2005 and Martinez-Vazquez et al., 2006) estimated the effect of the flat tax reform in Russia on earnings growth but they did not distinguish between tax evasion and real productivity effects.