People occasionally ask me which areas of tax administration worry me the most. There is certainly no shortage of candidates, but if I had to narrow it down, the IRS’s math and clerical error authority under IRC § 6213(b) and (g) would be right up there at the top of the list, along with the IRS correspondence examination program. In fact, I expressed my concerns about the IRS’s administration of its math error authority since my first Annual Report to Congress in 2001, when I was fresh from representing low income taxpayers as Executive Director of a Low Income Taxpayer Clinic. I’ve gone on to write, conduct research studies, and make legislative recommendations about math error authority in my Reports to Congress, as well as congressional testimony in 2011 and 2015. My past concerns take on new meaning, what with the current Administration’s budget proposal echoing the former Obama Administration’s call for the IRS to be granted “correctible error” authority.

While I have offered many proposals to minimize improper payments, I believe Congress should not address the problem by delegating to the Treasury Department the authority to expand the IRS’s power to summarily assess additional tax liabilities, at least not without sufficient limits and oversight. The IRS is currently authorized to assess tax to correct math and clerical errors – arithmetic mistakes and the like – under summary assessment procedures that bypass procedural taxpayer rights protections. The Administration has proposed legislation that would delegate authority for the Treasury Department to expand the IRS’s summary assessment (or “math error”) authority to other “correctible” errors (by regulation) where:

• The information provided by the taxpayer does not match the information in government databases;
• The taxpayer has exceeded the lifetime limit for claiming a deduction or credit; or
• The taxpayer has failed to include with his or her return documentation that is required by statute.

In my opinion, summary assessment authority is appropriate in only one of the instances described above; namely, where there can be no doubt that the taxpayer has claimed amounts in excess of a lifetime limitation, income cap, or age requirement. In fact, Congress’ original legislation back in 1926 was intended to limit the IRS’s authority to summarily assess math errors to situations involving such unambiguous errors. (See H.R. Rep. No. 69-1, at 10-11 (1926); S. Rep. No. 94-938(I), at 375 (1976); H.R. Rep. No. 94-658, at 289 (1976)).

For example, in cases where it is clear on the face of the return that a taxpayer has claimed a credit in excess of a statutory limit, such as overclaiming the American Opportunity Tax Credit (AOTC), then the summary assessment process may be appropriate. The AOTC is a partially-refundable credit for qualified post-secondary education expenditures that is available only for the first four years of a student’s post-secondary education (see IRC § 25A(i).) Because the number of years claimed for each student is apparent on the face of current and past income tax returns, allowing the IRS to use math error procedures to stop the improper payment of capped claims may be appropriate and cost effective, although probably not as cost effective as alerting the taxpayer to the problem before or at filing, through software checks and warnings, and e-filing rejection of the return in real time so it can be corrected and resubmitted.

Without adequate safeguards and congressional oversight, however, significant expansion of the IRS’s math error authority could permit the IRS to take property without adequate due process, as described below. It may also violate taxpayer rights, discourage eligible taxpayers from claiming the Earned Income Tax Credit (EITC) and other credits, and waste resources by requiring taxpayers to contact the IRS to correct the IRS’s errors and inaccurate inferences. In the face of such risks, in my opinion, Congress should not grant the IRS broad discretion to use its summary assessment authority.

The Right to Judicial Review Before Paying an Audit Assessment is the Cornerstone of Due Process in the U.S. Tax System.

Under current law, if the IRS during an audit proposes a deficiency, the IRS must issue a Statutory Notice of Deficiency (SNOD), also known as a “90-day letter.” This letter explains the basis for the proposed deficiency and gives the taxpayer 90 days to file a petition with the Tax Court to contest the proposed deficiency (see IRC § 6213). A taxpayer who misses this deadline for filing a Tax Court petition can only seek judicial review by paying the assessment and filing a claim for refund. If the claim is denied or if no action is taken on the claim within six months, the taxpayer may file a refund suit in the federal district court or the Court of Federal Claims within the limitations period (see IRC §§ 6511, 6532, 7422). Low income taxpayers are less likely to be able to afford to pay the assessment before disputing it or navigate these more complicated procedures.

Empowering taxpayers to seek judicial review in a prepayment forum (i.e., before they pay) protects them from arbitrary administrative actions by the IRS, which might otherwise unjustly deprive them of property without due process. Taxpayers who cannot understand the IRS’s position, determine if they agree or disagree, and respond appropriately within the 30- and 90-day periods may be deprived of this key right.  Therefore, even under normal deficiency procedures, confusing IRS correspondence, illiteracy, language barriers, and unequal access to competent tax professionals can cause taxpayers – particularly low income taxpayers – to miss these deadlines and lose access to judicial review in a prepayment forum.

Because prepayment judicial review is a cornerstone of due process in our tax system, any limitations on affording taxpayers access to prepayment judicial forums should be made in only the most compelling circumstances and when no other reasonable alternative solutions to a compliance problem exist.

Math Error Assessments Place the Burden on Taxpayers to Ask for the Right to Petition the Tax Court, Rather than Automatically Receiving That Right Under Normal IRS Procedures.

IRC §§ 6213(b) and (g) authorize the IRS to use its math error authority to summarily assess and immediately collect tax without first providing the taxpayer the right access to the Tax Court. If the taxpayer wants to preserve her right to petition the Tax Court, she must request an abatement of the assessment within 60 days.  Initially Congress limited this summary assessment authority to situations involving mathematical errors (e.g., 2+2=5), (see Revenue Act of 1926, enacting IRC § 274(f); H.R. Rep. No. 69-1, at 10-11 (1926)). Congress later expanded math error authority to address “clerical errors” (e.g., inconsistent entries on the face of the return), and other circumstances where a return is clearly incorrect (e.g., omits a required Taxpayer Identification Number, uses a Social Security Number that does not match the one in the Social Security Administration’s Numident database, or claims tax credits in excess of statutory maximums).

Math Error Adjustments Are Intended to Allow Correction of Unambiguous Errors That Are Easy to Explain.

As I noted in my 2014 report, Congress was concerned about removing more situations from the deficiency procedures and placing them under the summary assessment procedures, particularly in the case of complicated errors. If taxpayers do not understand the supposed error, they may have difficulty deciding whether to request an abatement (assuming they understand that requesting an abatement is an option), and they are less likely to request an abatement within the shorter 60-day period applicable to summary assessments. Accordingly, Congress enacted IRC § 6213(b)(1), requiring that “[e]ach notice under this paragraph shall set forth the error alleged and an explanation thereof.”

In legislative history, Congress provided an example of how simple it expected math error notices to be, which we have paraphrased below:

Example from Legislative History: You entered six dependents on line x but listed a total of seven dependents on line y. We are using six. If there is one more, please provide corrected information.

Although the IRS has been working to simplify these notices for nearly 40 years, even its current notice on this very issue (i.e., inconsistent number of dependents on the return) does not identify the discrepancy as clearly as Congress envisioned. The notice states:

Current Math Error Notice (Document 6209): “We changed your total exemption amount on page 2 of your tax return because there was an error in the:

• number of exemptions provided on lines 6a – 6d and/or
• computation of your total exemption amount.”

Other math error notices are inscrutable. The IRS’s problem with math error notice clarity is a serious, longstanding, and well-documented problem that disproportionately affects low income taxpayers – the very taxpayers that Congress intends to claim the EITC and similar credits. Moreover, unclear math error notices jeopardize the taxpayer’s rights to be informed, to challenge the IRS’s position and be heard, and to appeal an IRS decision in an independent forum.

The IRS Should Attempt to Resolve Minor Inconsistencies With Third-Party Data Before Burdening Taxpayers and Issuing Math Error Notices.

Not every return that contains a typo or similar error contains an understatement. For example, the IRS should not automatically conclude that a taxpayer does not have a qualifying child just because the Taxpayer Identification Number (TIN) of the child listed on the return does not match a TIN in the IRS’s database.  Such mismatches can be typos.

As reported in a research study published in my 2011 Annual Report to Congress, TAS studied a statistically valid sample of tax year 2009 accounts in which the IRS reversed all or part of its dependent TIN math error corrections. The IRS ended up abating all or part of the math error in 55 percent of the returns in which it originally assessed additional tax.  Further, the study found that the IRS could have resolved 56 percent of these errors using information already in its possession (e.g., the correct TIN listed on a prior year return), rather than charging a math error and asking the taxpayer to explain the apparent discrepancy. In other words, the IRS imposed a burden on taxpayers in a large percentage of math error cases, generating phone calls and letters it could not timely handle, rather than investing a few minutes of research at the front end.

Based on this study, in 2011 I recommended that even if it finds a mismatch between the return and a reliable database, the IRS should not use summary assessment procedures before taking additional steps to reconcile the mismatch. The same holds true today.  Until the IRS decides to (1) develop clear, specific and informative math error notices as Congress has directed at least since 1976; and (2) uses the data it has in its own possession to identify easily resolvable typographic errors without resorting to summary assessment authority, Congress should be very wary about granting the IRS additional authority.

In my next blog, I will discuss my concerns about specific aspects of the correctible error proposal.