Experimental Evidence on the Relation between Tax Rates and Compliance: The Effect of Earned vs. Endowed Income

Abstract

In this paper, we report the results of an experiment designed to determine whether the manner in which income is obtained (earned vs. endowed) affects the relation between tax rates and taxpayer compliance. Our experiment consisted of an income phase and a tax-reporting phase. In the income phase, participants were either endowed with $20 or were required to earn $20 by performing a one-hour multiplication exercise. In the tax-reporting phase, participants decided how much of their $20 in income to report on their tax returns. Consistent with prior experimental evidence, we find that when income is endowed, participants respond to a tax rate increase by reporting less taxable income. In contrast, but consistent with economic theory and some archival-empirical evidence, we find that when income is earned, participants respond to a tax rate increase by reporting more taxable income. Collectively, the results suggest that income is not a fungible commodity and that tax-payer responses to changes in policy variables such as the tax rate may depend critically on the amount of time and effort required to generate income. Additionally, our results may help explain differences between the results of taxpayer compliance experiments (which typically endow individuals with income) and archival-empirical studies (which use data that typically include earned income) regarding how changes in the tax rate (and other factors) affect taxpayer compliance decisions.