Become a Patreon!


Excerpted From: Jeremy Bearer-Friend, Colorblind Tax Enforcement, 97 New York University Law Review 1 (April, 2022) (291 Footnotes) (Full Document)


JeremyBearerFriendOn June 30, 2020, IRS Commissioner Charles Rettig testified before the Senate Finance Committee that IRS audit rates do not disproportionately impact Black and brown taxpayers. His outright denial surprised many of the Senators attending the hearing, who were familiar with the widely reported disparate audit rates by race covered by ProPublica. To clarify the IRS position, Senator Sherrod Brown submitted additional questions for the record on potential racial bias in tax enforcement, to which the IRS responded:

The IRS does not have, nor does it collect, any information or data related to the race and ethnicity of taxpayers. For example, the Form 1040 does not ask for the race or ethnicity of the taxpayer, and, therefore, the IRS cannot track any of this information.

If the IRS does not track any information related to the race and ethnicity of taxpayers, how could the IRS Commissioner so confidently assert there was no disparate racial impact of IRS enforcement activity when answering questions from a United States Senator?

The answer is a de facto IRS policy that I dub “Colorblind Tax Enforcement.” Across multiple presidential administrations and in a variety of public and private fora, the IRS has repeatedly taken the position that, because it does not ask about race or ethnicity on its tax forms, it does not discriminate. Restated as a formal “if A, then B” proposition, the position taken in this IRS statement is that, if the IRS does not ask about race or ethnicity, then the IRS does not base its actions on a taxpayer's race or ethnicity. The IRS has persistently maintained this position even though it is unwilling to evaluate the racial impact of enforcement due to its decision not to collect such data.

The accuracy of the IRS claim of colorblindness is an issue of urgent proportions. Over 150 million households file income tax returns each year. Over three trillion dollars are remitted. And while the IRS is limited to only the enforcement activity required by Congress to implement the Internal Revenue Code, much of this work also entails discretion. The IRS settles disputes on tax debts, seizes assets, and both assesses and abates civil penalties. The IRS can refer taxpayers to the DOJ for criminal investigation and prosecution, with some of its specialized staff even carrying firearms. According to the DOJ's Justice Manual, “tax enforcement potentially affects more individuals than any other area of criminal enforcement.” As a poignant point of contrast, the DOJ includes race and ethnicity in its own public reporting of criminal tax enforcement, while the IRS does not.

The lack of internal review and public reporting of IRS enforcement activity by race has taken on a new urgency as the Biden Administration has sought to double the size of the IRS and increase its funding by eighty billion dollars. The Biden Administration hopes to secure an additional $700 billion in tax revenue collections obtained through heightened enforcement. Within this context, the IRS's colorblind tax enforcement approach deserves serious scrutiny.

In this Article, I identify the wide variety of ways that race and ethnicity could determine tax enforcement outcomes even when race and ethnicity are not asked about on IRS tax forms. I begin by describing the nature of the IRS colorblind enforcement position. I then present three conceptual models of racial bias that produce disparate tax enforcement by race even when race and ethnicity are not asked about on IRS forms. These models are racial animus, implicit bias, and transmitted bias from non-tax policies. I then demonstrate how all three models would predict racial bias in a variety of tax enforcement settings even without asking taxpayers to identify their race and ethnicity, affecting such tax enforcement outcomes as settlement amounts, civil penalty assessments, and DOJ referrals. While some areas of tax enforcement present fewer opportunities for racial bias, in no enforcement setting is the current policy sufficient to ensure the absence of racial bias.

Providing a bird's eye view on the potential for racial bias in tax enforcement is a novel addition to tax scholarship on race. While there is a well-established body of work on the intersection of race and tax--a rich field that continues to grow--this scholarship has primarily focused on racial disparities embedded in the Internal Revenue Code itself. Scholars have documented disparate tax treatment of housing, marriage, retirement, estate planning, and employment discrimination settlements. Because it is the responsibility of Congress to write and revise the Internal Revenue Code, the interventions that might address racial inequality in the Code have accordingly focused on legislative solutions. But in the context of tax enforcement, new legislation is not required before the IRS can address racial bias. An additional contribution of this Article, then, is to identify areas where the IRS has sufficient discretion to remedy racial inequality on its own, provided institutional leadership has the willpower to do so.

The current ignorance over racial disparities in tax enforcement stands in stark contrast to what we know about law enforcement outside of tax. Criminal law scholars have long documented racial bias at nearly all stages of the criminal justice system, including profiling of suspects, police use of force, access to counsel, grand juries, trial juries, and sentencing. The research on racial bias in the enforcement of criminal law parallels research on racial bias in the design of criminal law. This cumulative body of criminal law scholarship, developed over multiple generations, raises doubts about the IRS presumption of race neutrality in tax enforcement.

The international experience with tax enforcement also suggests potential for disparate enforcement by race or ethnicity. In the Netherlands, tax personnel were able to use the surnames that appeared on tax documents to infer the ancestry of the filer. Those filers who did not appear ethnically Dutch by surname were then targeted for heightened enforcement. As originally reported by The New York Times, “an administrative mistake like a missing signature was enough for the tax authority to label parents as frauds and fine families as much as tens of thousands of euros.” How tax personnel respond to errors on a tax form is just one example of the many types of discretion that occur in most tax enforcement settings. And while the publicity of the discriminatory tax enforcement in the Netherlands led to substantial political consequences, including the resignation of the Prime Minister, current IRS data policy obstructs such potential accountability.

The demonstrated potential for racially biased tax enforcement that I identify in this Article, presenting three original models of racial bias in tax enforcement and illustrating how these models can operate within seven distinct tax enforcement settings, further erodes the case for omitting race and ethnicity from the collection and analysis of federal tax data. Without adequate data collection and data analysis, any disparate tax enforcement by race will go unacknowledged and, subsequently, unaddressed.

This Article proceeds as follows. Part I lays out the current approach to race and ethnicity by the IRS. Part II provides three distinct models of racial bias in tax enforcement. Part III identifies the ways that such biases can operate within seven tax enforcement settings even though race and ethnicity are not asked about by the IRS. I conclude with a menu of alternative approaches to ensuring racial equity in tax enforcement other than the current IRS policy of purported colorblindness.

[. . .]

This Article asks: Under what conditions is racial bias in tax enforcement possible? After providing three original models of racial bias in tax enforcement, this Article then determines that such conditions can be present across every major enforcement function of the IRS, including appeals, collections, and criminal investigations. This finding is a direct rebuttal to the IRS proposition that because the IRS does not ask taxpayers to identify their race or ethnicity, the IRS cannot discriminate. Although current IRS data practices prevent the public from determining the full scope of racial bias in tax enforcement under current law, the credible risks identified here demonstrate the ongoing threat of racial bias in tax enforcement that will persist absent an immediate change to IRS policy and practice.

Jeremy Bearer-Friend, Associate Professor of Law, George Washington Law School.

Become a Patreon!