Understanding the tax gap.

INTRODUCTION

The term "tax gap" has been used repeatedly by pundits and policy makers. These individuals exhibit a wide range of understanding of the concepts underpinning the tax gap estimates. Our aim with this article is to provide public finance professionals with a basic understanding of the various aspects of the tax gap so these professionals can help shape and guide future debate on this subject. (2)

DEFINING THE TAX GAP

The federal "tax gap" is defined for a given tax year as the difference between the amount of tax owed by taxpayers under the Tax Code and the amount that is actually paid to the federal Treasury on time. This definition seems straightforward, but in many cases there are some difficulties applying it in practice. For instance, consider the case of a complex multinational enterprise, with far-flung operations and complicated financial undertakings. The amount of tax imposed by the Tax Code for this entity may not be viewed as certain by the taxpayer and its advisers and investors. Often, many interpretations of tax law are required to match the entity's fact patterns, and these interpretations may differ between the taxpayer and the tax administrator. If there is some uncertainty about the true tax liability imposed by the Tax Code, then this uncertainty will flow right through to estimates of the tax gap.

In principle, the tax gap covers all taxpayers and all taxes (individual and corporate income tax, payroll taxes, estate and gift taxes, and excise taxes). In practice, however, the Internal Revenue Service (IRS) does not yet have estimates of all components of the tax gap for a few types of taxes.

The Internal Revenue Service limits its estimates of the tax gap to tax due on legal-sector activity only. That is, taxes that may be due on income from activities such as illegal gambling, prostitution, and drug dealing are not included in the estimates of the tax gap. The primary rationale for this exclusion is somewhat philosophical, in that the overriding government interest in these areas is not in eliminating the tax gap that might be associated with illegal activities, but rather to eliminate the activity itself. A secondary (though more practical) rationale is that it would be extraordinarily difficult to estimate the amount of taxable income from illegal activities, net of deductible expenses associated with earning that income.

The tax gap is not synonymous with the so-called "underground economy," though there is some overlap. The underground economy is made up of activities that are not very visible to tax and other government authorities. Some of these are legal-sector activities and some, illegal-sector. As noted above, illegal-sector income is not included in estimates of the tax gap. While some of the tax gap arises from legal-sector income generated by underground economy participants, some of it arises from noncompliance that is completely unrelated to the underground economy--such as claiming the wrong filing status or overstating exemptions or tax credits. So, while there is substantial overlap between the tax gap and the underground economy, it is best to maintain the distinction between these two concepts.

COMPONENTS OF THE TAX GAP

The Internal Revenue Service classifies every dollar of the tax gap into one of the following three mutually exclusive categories.

Nonfiling gap--The tax not paid on

time associated with returns that are

filed after the due date (with extension,

if any) or not filed at all. Any withholding

(or other timely payments) paid on

behalf of the taxpayer is netted out in

computing estimates of the nonfiling

gap.

Underreporting gap--The additional

tax due on timely filed returns arising

from the misreporting of tax liability

on those returns (compared to the true

liability owed under the Tax Code).

Underpayment gap--The tax that is

reported on timely filed returns, but

that is not paid on time.

For each of these components, the gross tax gap is defined as underpayments net of overpayments, so taxpayer errors in favor of the federal government are appropriately accounted for in these estimates.

These three components are depicted in the chart displayed below, called the Tax Gap Map. From the Map it is apparent that for Tax Year 2001 (the most recent year for which estimates are available) underreporting accounts for over 80 percent of the $345 billion gross tax gap, while nonfiling and underpayment each account for less than ten percent of the total gross tax gap.

The Map also indicates that the largest portion of the tax gap is attributable to the individual income tax (which is primarily because it is the federal government's largest source of revenue). The biggest contributor to the individual income tax gap is underreported business income (from sole proprietors or business owners, farmers filing Schedule F, or recipients of partnership, S-Corporation, rental, or royalty income). This is due largely to the fact that much of this income is much less "visible" to the IRS through third-party information reporting or withholding than are other types of income.

[FIGURE 1 OMITTED]

The Tax Gap Map organizes estimates of the various components of the tax gap (where these estimates exist). It also shows that an estimated $55 billion of Tax Year 2001 liabilities will eventually be collected through enforcement actions taken by the Internal Revenue Service, or via other late payments, leaving a net tax gap of $290 billion for Tax Year 2001. Finally, the Tax Gap Map provides an indication of the strength of the estimates. White boxes (no shading) indicate that the figures are generated from internal accounting data and can be considered very solid. This characterization applies only to the underpayment data, which are derived from Internal Revenue Service accounting data. Light shading indicates that the estimates were generated from relatively recent studies and, thus, can be considered quite reasonable. This covers the individual income tax and self-employment tax estimates, which were generated as a result of a Tax Year 2001 study of the reporting compliance of individual taxpayers. The darkest shading indicates that the estimates are based on older data, with results projected to Tax Year 2001 assuming constant compliance behavior. These estimates are generally extrapolated from studies of taxpayer compliance covering periods prior to 1988 and are clearly weaker than the others.

ESTIMATING THE TAX GAP

The elements of the Tax Gap are estimated separately, with the various components combined on the Tax Gap Map. As noted above, the underpayment gap is computed directly from IRS accounting data. IRS knows both how much tax is reported on timely filed returns and how much of that is paid on time. The underpayment gap is merely the difference between the two. Since these figures are derived from internal accounting data, these calculations can, in principle, be done on a fairly timely basis.

Nonfiling gap estimates have been derived for the individual income tax and the estate tax. In both cases, the estimates result from comparing estimates of overall tax liability among the population required to the a return developed from a non-tax data set (such as the Current Population Survey) with corresponding figures derived from tax data pertaining to timely filers. The estimate of overall tax liability associated with late filers and nonfilers is then reduced by the amount of tax withheld or otherwise paid on time by nonfilers and late filers to get an estimate of the nonfiling gap. Since these estimates require comparisons of multiple data sets, generally there is a noticeable time lag between when returns are (or should be) filed and when these estimates are generated.

Underreporting gap estimates typically are derived either from thorough analyses (including audits--sometimes called "research audits") of randomly selected returns, or from the regular operational audits that the IRS conducts. For Tax Year 2001, the IRS undertook a reporting compliance study of 46,000 randomly selected individual income tax returns under the new National Research Program (NRP). (3) Examinations started in October 2002 and continued through September 2004. Comprehensive tax gap estimates were produced from these data in February 2006, illustrating the substantial time required to develop these sorts of estimates.