Compliance effects of risk-based tax audits

Abstract

Tax administrations use machine learning to predict risk scores as a basis for selectingindividual taxpayers for audit. Audits detect noncompliance immediately, butmay also alter future filing behavior. This analysis is the first to estimate complianceeffects of audits among high-risk wage earners. We exploit a sharp audit assignmentdiscontinuity in Norway based on individual tax payers risk score. Additional datafrom a random audit allow us to estimate how the audit effect vary across the risk scoredistribution. We show that the current risk score audit threshold is set far above theone that maximizes net public revenue.