Measuring and Mitigating Racial Disparities in Tax Audits.

This article examines racial disparities in Internal Revenue Service (IRS) audit rates between Black and non-Black taxpayers, focusing particularly on claimants of the Earned Income Tax Credit (EITC), the largest cash-based safety net program in the United States. Using a novel partial identification method to impute taxpayer race from names and geography, the study finds that Black taxpayers are audited at 2.9 to 4.7 times the rate of non-Black taxpayers, with the disparity primarily driven by higher audit rates among Black EITC claimants. The authors show that this disparity arises not from disparate treatment or differences in total underreported taxes but largely from the IRS’s audit selection algorithm prioritizing detection of overclaimed refundable credits (such as the EITC) over other forms of noncompliance, which disproportionately affects Black taxpayers due to differences in types of errors, such as dependent eligibility. The study further highlights that shifting audit selection to maximize detection of total underreporting would reduce racial disparities but increase audit complexity and costs, underscoring trade-offs in enforcement priorities and resource allocation.