Tax Evasion and Tax Rates: Properties of a Penalty Structure

Abstract

Allingham and Sandmo demonstrated that under decreasing absolute risk aversion, when the penalty for tax evasion fell on evaded income, an increase in the tax rate might increase, decrease, or leave unaltered the level of evaded income. Yitzhaki resolved the ambiguity by considering the impact of increased tax rates when the penalty for tax evasion fell on evaded taxes. However, his result (that an increase in tax rates leads to a decrease, rather than an increase, in evaded income) has been heavily criticized as counterintuitive, and this criticism has spawned a literature aimed at exploring different assumptions that would generate an "intuitive" result. Preoccupation with this concern has deflected analysis from appraisal of the properties of the penalty structure as described by Yitzhaki. In this article, the authors challenge the view that Yitzhaki's analysis is so counterintuitive and, by offering a broader assessment, the intention is to refocus the discussion. Yitzhaki's analysis is better interpreted in the context of an evaluation of different penalty structures. Building on Yitzhaki's analysis, it can be shown that penalties are generally more efficient when punishment is designed to fit the crime.