Decentralized Deterrence, with an Application to Labor Tax Auditing

TLDR

The authors develop a strategic auditing game in which auditors maximize audit success and calibrate its parameters using audit data, aiming to assess decentralized enforcement versus a commitment policy. They calibrate the model to Italian audit data, estimate tax evasion from non‑random audits, and run counterfactual simulations to quantify the costs and benefits of alternative auditing policies. The calibrated model estimates tax evasion levels and shows that decentralized enforcement incurs a measurable loss compared to a commitment policy, while also quantifying the associated costs and benefits. JEL codes: H25, H26, M42.

Abstract

This paper studies a new strategic auditing game in which atomistic auditors maximize the success rate of audits, and provides a method to calibrate its parameters based on audit data. Calibrating the model to Italian auditing data, we provide an estimate of tax evasion based on (non-random) audit data alone. Counterfactual simulation of the model quantifies the costs and benefits of alternative auditing policies. We compare decentralized enforcement with a counterfactual “commitment policy,” and compute the loss from the former. (JEL H25, H26, M42)