Henrik J. Kleven (London School of Economics, Department of Economics), Martin B. Knudsen (Danish Inland Revenue, Denmark), Claus T. Kreiner (University of Copenhagen, Department of Economics), Søren Pedersen (Danish Inland Revenue, Denmark) & Emmanuel Saez (UC-Berkeley, Department of Economics) have posted Unwilling or Unable to Cheat? Evidence from a Randomized Tax Audit Experiment in Denmark on NBER. Here is the abstract:
This paper analyzes a randomized tax enforcement experiment in Denmark. In the base year, a stratified and representative sample of over 40,000 individual income tax filers was selected for the experiment. Half of the tax filers were randomly selected to be thoroughly audited, while the rest were deliberately not audited. The following year, "threat-of-audit" letters were randomly assigned and sent to tax filers in both groups. Using comprehensive administrative tax data, we present four main findings. First, we find that the tax evasion rate is very small (0.3%) for income subject to third-party reporting, but substantial (37%) for self-reported income. Since 95% of all income is third-party reported, the overall evasion rate is very modest. Second, using bunching evidence around large and salient kink points of the nonlinear income tax schedule, we find that marginal tax rates have a positive impact on tax evasion, but that this effect is small in comparison to avoidance responses. Third, we find that prior audits substantially increase self-reported income, implying that individuals update their beliefs about detection probability based on experiencing an audit. Fourth, threat-of-audit letters also have a significant effect on self-reported income, and the size of this effect depends positively on the audit probability expressed in the letter. All these empirical results can be explained by extending the standard model of (rational) tax evasion to allow for the key distinction between self-reported and third-party reported incomes.
The article is featured in the June 2010 NBER Digest in Evidence from a Tax Audit Experiment in Denmark:
[Announcing high probabilities of future audits generated] substantial future tax revenue through behavioral responses to a higher perceived probability of detection.
[The authors] conclude that the low overall rate of tax evasion enjoyed by advanced economies has more to do with an information environment in which third-party reporting makes it difficult to cheat than it does with any moral reluctance to cheat on the part of taxpayers.
Observing the behavior of a random sample of 42,784 Danish taxpayers in 2007 and 2008, they find that income categories in which both taxpayers and third parties report payments have evasion rates that fall “between 0.2% and 0.9%” of each type of income. Income categories that are self-reported and not subject to third-party reporting have evasion rates of roughly 37%. Although self-reported income constitutes only about 5% of total income among the sample in this study, it is responsible for 87% of the detected tax evasion.
The authors also report on the results of an experiment in which taxpayers were randomly divided into two groups, one group in which all were audited and the other in which none were audited. The people in the 100%-audit group had comprehensive unannounced tax audits of their 2007 tax returns. None of the returns of the other (0%-audit) group were audited. The following year, people from each group were randomly selected into three sub-groups with different audit probabilities (0%, 50%, and 100%, respectively) and informed in advance about the probability that their 2008 returns would be audited.
The combination of pre- and post-audit data makes it possible for the authors to examine taxpayer behavioral responses to actual audits and to the threat of audits, and to study the effect of high marginal tax rates on tax evasion decisions. The results show that audits had a strong and significant effect on subsequent reporting of self-reported income and it is concluded that audits generate “substantial future tax revenue through behavioral responses to a higher perceived probability of detection.” Similarly, audit threats led to upward adjustments in self-reported income. Many Danish taxpayers bunch at the kink points of the income tax schedule where marginal tax rates jump from 49% to 62%, which is evidence of behavioral responses to marginal tax rates. However, the audits show that the vast majority of those bunching taxpayers do not evade taxes. This implies that behavioral responses to marginal tax rates are due mostly to labor supply or tax avoidance rather than tax evasion. Overall, this evidence shows that broadening information reporting requirements can have a substantial impact on tax compliance.
