For those interested in understanding taxpayer compliance—including what motivates taxpayers to report honestly and how to reduce tax evasion – there is a robust body of empirical and legal literature. A number of economists and lawyers have examined the effect of traditional deterrence mechanisms like audits and penalties, as well as non-economic factors like social norms, guilt, taxpayer attitudes about how their tax money is spent, and other psychological factors. While this growing body of literature provides a rich description of the myriad of factors that influence tax compliance decisions, our understanding of taxpayer behavior is far from complete, and further study continues to be necessary. In expanding our understanding of taxpayer motivations, one revealing but possibly overlooked resource is the IRS Taxpayer Advocate Service (“TAS”), which each year publishes a number of empirical studies and other reports relevant to tax compliance. The authors of TAS studies are uniquely situated in that they have access to IRS tax return data, which should provide the best evidence of how taxpayers make decisions in the real world.
As part of the Taxpayer Advocate’s most recent Annual Report to Congress, the TAS published Audits, Identity Theft Investigations, and Taxpayer Attitudes: Evidence from a National Survey (the “Report”). The Report surveyed 2,729 Schedule C filers, that is, taxpayers reporting income from self-employment. Of this group, roughly half (1,363) had been previously audited and half (1,366) had not. One of the primary goals of the Report was to examine how audits influence taxpayer attitudes and behavior. While several of the findings are predictable and consistent with other research on audits, three of the Report’s findings are quite surprising.
The first and most surprising finding is that nearly 40 percent of the audited taxpayers surveyed did not know they had been audited. You read that right. When audited taxpayers were asked if they had been audited in the past, a significant number of them (38.7 percent) did not appear to realize they had been. Why? The answer is at least partly in the details. Taxpayers who experienced a “correspondence audit” (in the form of letters from the IRS) were far more likely to not realize they were being audited compared to those who experienced an in-person audit. It’s plausible that taxpayers envision audits as involving an IRS agent digging through their files, and that they simply do not perceive an exchange of letters as an audit. But since more than 70 percent of audits are done through correspondence, perhaps audits don’t have as strong of a deterrent effect as we might assume.
The second surprising finding is that audited taxpayers have more favorable views about justice in the tax system than non-audited taxpayers. Specifically, audited taxpayers perceive higher levels of procedural justice, informational justice, interpersonal justice, and distributive justice as compared to unaudited taxpayers, perhaps indicating that their experience with the IRS was “transparent, respectful, and appropriate” according to the Report. On the one hand, this may indicate that the IRS gets a bad rap: we might assume an audit will be an invasive and stressful experience when, in actuality, perhaps it’s not so bad. On the other hand, less than half of taxpayers surveyed owed the IRS money after their audit, and nearly 19 percent were owed a refund. This happy and perhaps unexpected outcome likely influenced the attitudes of some survey respondents.
The third surprising finding is that taxpayers who receive a refund after an audit have less favorable views about the tax system than those who receive no adjustment after an audit. In fact, taxpayers with no change after an audit reported the highest perceived levels of distributive justice in the tax system and most positive attitudes overall as compared to taxpayers who owed money and those who were owed a refund. Interestingly, this indicates that post-audit refunds don’t necessarily promote positive attitudes towards the IRS and the tax system. Perhaps the explanation here is that an audit with no adjustment indicates that everything was working correctly: the taxpayer got it right on her return and IRS inquiry did not change the result. On the other hand, it is somewhat surprising that taxpayers who got their returns right the first time aren’t more resentful that they were audited at all, especially as compared to those who receive a surprise refund.
Other aspects of the Report were not surprising. Self-employed taxpayers tend to consider cheating more than employed taxpayers, which makes sense because the former group has significantly more opportunity to cheat. It is also unsurprising that taxpayers who owed money after an audit report more negative attitudes about the tax system compared to those who weren’t audited, including lower levels trust in the IRS and higher levels of anger.
Overall, there are a number of possible takeaways from the Report. Audits appear to have a deterrent effect, but we may overestimate their power given the number of taxpayers who don’t realize they have been audited. At the very least, we can’t necessarily analyze the impact of audits collectively, as the type of audit (correspondence versus in-person) matters a great deal. Furthermore, the IRS’s bad reputation—at least when it comes to audits—might be undeserved, given the positive attitudes reported by many audited taxpayers. More generally, the Report highlights the fact that taxpayer attitudes about the tax system are heterogeneous and complex, underscoring the importance of continued study. Finally, the Report illustrates the wealth of information on tax compliance that is available thanks to TAS.