Tax Evasion and the Low Penalty, Low Audit Rate Phenomenon Robert Halperin and Joseph Tzur In the United States, penalties for income tax evasion and the audit rate of taxpayers’ returns are both remarkably low. This phenomenon exists despite the existence of extensive literature which suggests that high audit rates and high penalties can reduce tax evasion. We construct an analytical model and present three numerical examples to explain this phenomenon. Our results suggest that low penalties and low audit rates exist because of the political weight that evaders have with policy makers. We also show that a large number of evaders, by itself, is not sufficient to explain the phenomenon. Ever since the publication of a seminal article by Allingham and Sandmo (1972), a considerable amount of research effort has been directed at the issue of income tax evasion and compliance. Given the fact that noncompliance is estimated to cost the U.S. Treasury at least $100 billion annually, this line of research is of great practical importance (Graetz and Wilde 1985, p. 357). Furthermore, Graetz and Wilde (1985, p. 355) suggest that voluntary compli- ance is declining. Research in the area (Allingham and Sandmo 1972; Srinivasan 1973; Yitzhaki 1974) shows that an effective way of reducing tax evasion is to, first, increase penalties for tax evasion when the penalties are a percentage of the tax evaded and second, to increase the costly audit rate. In the United States, criminal penalties for tax evasion can be as much as five years in prison plus a $lOO,ooO fine. Arguably, these penalties are rather high. However, in 1981, criminal investigations comprised less than 0.1% of all Internal Revenue Service (IRS) investigations (Skinner and Slemrod 1985, p. 348, citing Mansfield 1983). Perhaps, this is due to the high administrative cost to the IRS of gathering evidence and prosecuting these cases in order to show that the tax understatement is willful. Proving that a tax understatement is characterized by the requisite knowledge and deliberate behavior to be called Address zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA rqwint re+ests zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA to zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK Pmfeasor Robert Halperin, GraduateSchool of Business Administration D+t’tmnt of Accounting, Fordbm University, 113 West 60th Street, New York, NY 10023. Jaamal of Acamting and Public Policy, 9, 179- 1% (1990) zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQ 179 0 1990 Elswier Science Publishing Co.. Inc. 02184254/9Q/M3.50