Taxpayers can claim the Earned Income Tax Credit (EITC) in more than one tax year, so using the audit as an opportunity to educate them about the requirements for claiming EITC is of particular benefit to them and to the IRS. If a taxpayer claims the credit in error but understands why there was an error, he or she can not only become compliant for the year of any audit, but remain compliant going forward. However, audits are expensive for both the IRS and taxpayers, and are intrusive and intimidating for the taxpayer. There are many EITC returns the IRS does not audit but identifies as containing an error. Thus, while the IRS may not have the resources to audit these taxpayers, through other cost-effective approaches, it can educate them about why they appear to have erroneously claimed EITC, and avert future noncompliance.

To investigate this possibility, in 2016 the Taxpayer Advocate Service (TAS) conducted a study to see whether providing taxpayers with more tailored information about their claims for EITC, which appeared to be erroneous, would help them avoid making errors in the future. (Our methodology is described below.) We learned that this approach does avert noncompliance, especially when the taxpayer’s apparent mistake was not meeting the relationship requirement for claiming EITC. Actually, we projected that sending an educational letter to all taxpayers whose 2014 returns appeared to be erroneous because the relationship test was not met would have averted about $47 million of erroneous EITC claims on these taxpayers’ 2015 returns.

Here is how we carried out the study. We selected a representative sample of taxpayers the IRS had identified as having erred in claiming EITC on their 2014 returns via its Dependent Database (DDb) rules, but whose 2014 returns were not audited. There were over 1.9 million taxpayers who tripped only the DDb rules we studied; only about 6,500 of these taxpayers were actually audited. We sent over 7,000 taxpayers one of three versions of an educational letter. The letter was sent out under my signature and was mailed during the first few weeks of January, before the start of the 2016 filing season. In fact, to enhance the saliency of the communication, TAS sent the letter in envelopes printed with “Important Tax Information Inside,” in red, so taxpayers expecting Forms W-2 might actually open the envelope and read the letter.

Depending on the DDb rule broken, the letter identified the error the taxpayer appeared to have made on the 2014 return as: the relationship test was not met, the residency test was not met, or another taxpayer claimed the credit with respect to the same qualifying child or children. We then described, in plain English, the basic eligibility requirements related to the relevant error. In addition, we reminded each taxpayer that if he or she received Temporary Assistance to Needy Families (TANF), food stamps, or other public benefits for a given child, that did not necessarily mean the taxpayer qualified for EITC with respect to that child. I have not seen any other IRS publication or communication alerting taxpayers to this basic fact. We also told taxpayers that this was not an audit – we were simply contacting them to prevent future problems and help educate them.

Our control group consisted of a representative sample of over 14,000 taxpayers whose 2014 returns were also not audited and had similar characteristics as the returns of taxpayers who received the TAS letter, but who were not sent the TAS letter. The study compared the level of compliance shown on 2015 returns filed by taxpayers who were sent the TAS letter to compliance shown on the 2015 returns filed by taxpayers in the control group, as well as to the 2015 returns of those taxpayers who were actually audited for breaking the same DDb rule on their 2014 returns. Our findings for the population studied are statistically valid at the 95 percent confidence level.

The study showed that the taxpayer’s improved compliance behavior depended on the type of DDb rule that was broken. For example, when the error on the 2014 return appeared to be that the relationship test was not met, taxpayers who were sent the TAS letter were less likely to repeat that error on their 2015 returns than taxpayers in the control group. Specifically, those in the control group repeated their error 77.3 percent of the time, compared to 74.7 percent for the TAS group, an improvement of 2.6 percent, which is statistically significant. Taking into account the number of 2014 returns that appeared to repeat this error (and only this error) in 2015, the TAS letter averted about 20,000 erroneous EITC claims in 2015. The average amount of EITC paid to 2014 claimants was more than $2,400, so we projected that sending the TAS letter to all taxpayers who did not appear to meet the relationship test would have averted about $47 million of erroneous EITC claims. We did not quantify the cost of sending letters to the nearly 1.2 million taxpayers who appeared to have made this error, but even if the cost would be $2 per letter, for a total cost of $2.4 million, the cost of sending the letter would be far outweighed by the increased compliance.

When the error was that there were duplicate claims (i.e., another person claimed the same qualifying child or children), the TAS letter prevented taxpayers from filing returns on which they claimed EITC, compared to the control group. Audited taxpayers were no more likely to file EITC returns compared to taxpayers who received the TAS letter, and were less likely to file EITC returns compared to the control group. However, audited taxpayers were more likely to trip an EITC DDb rule on those returns than taxpayers who received the TAS letter or taxpayers in the control group.

When the error on the 2014 return appeared to be that the residency test was not met, taxpayers who received the TAS letter were slightly less likely to repeat the same error on their 2015 returns than taxpayers in the control group, but this result was not statistically significant.

We conducted a similar study in 2017 prior to the filing season. The educational letter we are using is the same as in the 2016 study, except that this year’s letter also reminds taxpayers that if they cannot claim a child for the EITC, they may still be able to receive the “childless worker” EITC. In addition, in a letter to a separate group of taxpayers who appeared to not meet the residency test we offered an additional resource: a toll-free phone number the taxpayer can call to talk with a TAS employee about their eligibility for the EITC.

Because it appears that taxpayers failing the residency test are not as responsive to the educational letter as other taxpayer groups, it may make sense to follow up the letter experiment with focus groups of these taxpayers. We could then learn how they perceived and understood the information the letter provided. Moreover, this population might be sensitive to repetitive multi-year touches, as residency tends to change within years and between years. Nevertheless, the 2016 TAS study shows that for a minimal expense of sending an informative, direct, personal, tailored letter, there can be a significant compliance impact in at least some situations.