Economic Crime: Tax Offenses

Tax compliance typically relies on voluntary self-assessment, which requires taxpayers to calculate their own tax liability and voluntarily to pay the amount due. For income taxes, there are three kinds of noncompliance: failure to file tax returns; underreporting of taxable income (either through underreporting income or overstating deductions); and failure to pay established liabilities. Tax scholars often distinguish between tax minimization, tax avoidance, and tax evasion, but no precise lines divide this continuum of conduct.

Tax noncompliance is socially harmful. Most obviously, tax noncompliance reduces tax revenues, which is a bad thing if one believes that legitimately elected governments should be able to carry out their policies as they choose. Tax noncompliance also can distort labor markets, as when people select jobs to dodge taxes. Efforts to avoid taxes (like efforts to increase compliance) are deadweight losses to society. Tax evasion can create unfairness and can fuel perceptions of rampant cheating that undermine respect for government. Left unchecked over time, these perceptions would tend to snowball as more people conclude that cheating is common, normal, and inviting.

The tax noncompliance problem in the United States is large and growing. The tax gap (the difference between what taxpayers owe and pay) was an estimated $195 billion/year in 1998, or about $1600 per year for every tax return filed by compliant taxpayers (Rossotti). Estimates are that American taxpayers voluntarily pay 83 percent of the taxes they owe. The compliance rate in other countries is often much lower (IRS, Sept. 1997; Graetz and Wilde). The dollar cost of noncompliance has risen sharply since about 1980 even as the rate of tax noncompliance has remained fairly stable (IRS, Sept. 1997). Most academic focus has been on evasion of income taxes, but noncompliance problems also exist for other taxes (e.g., employment taxes, excise taxes, retail sales taxes, estate and gift taxes).

Who is not paying? Three-quarters of the problem has been the fault of individuals rather than corporations, according to data that Slemrod and Bakija report. The bulk of noncompliance seems to come from the underreporting of personal income, which accounted for 47.5 percent of the tax gap in 1992. Corporations, however, may be closing the gap. In the late twentieth century there has been a dramatic increase in the number of corporate tax shelters that corporate taxpayers use to reduce tax liability by entering into transactions that purportedly lack economic substance apart from tax benefits. Bankman estimates that the lost corporate tax revenues may amount to billions of dollars. It is difficult to generalize about the types of transactions deemed corporate tax shelters. They do share some common elements: (i) they seek to obtain a tax benefit not clearly contemplated by the applicable tax provision; (ii) they lack economic substance, in that the reasonable expected pre-tax economic profit is insignificant relative to the reasonably expected net tax benefits; (iii) they result in inconsistent financial accounting and tax treatment; and (iv) they use tax-indifferent parties (generally foreign and tax-exempt entities) to absorb or deflect taxable income (Department of Treasury 1999, 2000).

Compliance also varies greatly by types of income. For example, the compliance rates for wages and salaries (99.5%), pensions and annuities (98.4%), interest and dividends (94.6%), and income from capital gains (88.3%) are relatively high, while the compliance rates for partnerships and S corporations (42.1%) and self-employment income (41.4%) are quite low (in all cases estimated for reported net income as a percentage of Figure 1 true net income for filers only) (Slemrod and Bakija).

To spur tax compliance, two techniques are coercive incentives and structural components. Coercive incentives seek to induce tax compliance, primarily through a program of audits, civil penalties, and, rarely, criminal prosecutions. Structural components aim to get the owed money (or the information that leads to the money) before taxpayers can hide it, mainly by means of increased withholding and information reporting.

The IRS has changed its enforcement approach by relying more on structural components and less on coercive incentives. This shift has changed the types of noncompliance detected and prosecuted, and may have reduced the deterrent effect that tax cheats face. The shift also has reduced reliance on the ultimate deterrent: criminal prosecution of traditional tax crimes.

The coercive incentive approach begins with auditing, and the decline in IRS auditing has been marked. In the mid-1960s, the audit rate was about 6 percent for individuals (Dubin, Graetz, and Wilde), but this rate fell to less than one half of one percent by the century's end. Audits of individual taxpayers thus were twelve times less common in the late 1990s than in the mid 1960s.

The rate of corporate auditing likewise fell at the end of the twentieth century. For the largest corporations (those with assets over $250 million) the audit probability fell from 54 percent to 37 percent from 1992 to 1998. Over the same time period, the audit rates for small corporations (those with less than $250,000 of assets) fell from 1.18 percent to .75 percent. In contrast, there has been a dramatic increase in the use of structural components in the last thirty-five years. The IRS has been quite successful in requiring information reporting on certain transaction and matching that information to income tax returns. In 1965, the IRS received about 340 million information documents; by the 1990s the annual number had increased to over 1 billion documents. The IRS estimates that over 75 percent of all income that should be reported on income tax returns is subject to information reporting requirements (Andreoni, Erard, and Feinstein).