Optimal Tax Audits using Predictions

• In Event: Tax Evasion and Enforcement: Measurement and Estimation

Abstract

Tax authorities use audits to detect and deter tax evasion. In practice, they commonly rely on quantitative predictions about taxpayers’ noncompliance to inform their decisions about which taxpayers to audit. We study the problem of optimal audit selection in this context. Specifically, we investigate how variation in the distribution of predicted noncompliance across taxpayers, including the uncertainty in quantitative predictions, shapes optimal audit policy. Our results highlight the contribution of these factors through a sufficient statistics characterization of the optimal audit selection rule. We leverage this characterization to quantify the social welfare benefit of varying the information available to the tax authority, for example by expanding third-party information reporting.

[In progress:] We implement our sufficient statistics characterizations of the welfare effects of audits using US administrative data, including machine learning predictions of audit revenues and administrative costs constructed from random audit data, together with data on the private costs of audits.

[Note to organizers:] This presentation will focus much more on the empirical part of the paper using NRP data, which is currently underway with the help of Sree Kancherla, who's recently come on as a coauthor. The current, draft of the paper, which is available on the authors' websites, only contains the theory and does not include Sree as a coauthor. An early version of the theory was presented at NTA 2023.

Authors

• Aviv Caspi, Stanford University

• Jacob Goldin, University of Chicago Law School

• Sreeraahul Kancherla, US Treasury Office of Tax Analysis

• Daniel Reck, University of Maryland

• Daniel Ho