Corporate Tax Shelters and Narrowing the Book/Tax “GAAP”

Monday, January 1st, 2001 at 12:00 am by Mitchell L. Engler
Mitchell L. Engler, Corporate Tax Shelters and Narrowing the Book/Tax “GAAP”, 2001 Colum. Bus. L. Rev. 539

America’s largest publicly-held corporations routinely report vastly different pictures of their operating results depending on their audience. Net income reported to investors on public financial statements (book income) dramatically exceeds taxable income reported to the IRS. Furthermore, a recent private study reports that the difference between reported book and taxable income continues to grow. This study characterizes the book/tax difference as corporate tax avoidance, which should be eliminated.

Independently, government officials recently identified corporate tax shelters as the ‘most serious compliance problem facing our tax system. ‘ It has been suggested that corporate tax shelters cost the federal government approximately ten billion dollars per year. Corporate tax shelters lack a precise definition, although they generally share certain characteristics. Typically, corporate tax shelters provide corporate managers with the opportunity to lower their taxable income without either (i) reducing the income reported to investors or (ii) incurring a real loss. Tax shelters therefore generate a loss for tax purposes even though the corporation has not sustained an economic loss. Importantly, no loss is reported for book purposes (because there is no economic loss).

An additional shared characteristic of corporate tax shelters highlights how tax sheltering has a narrower scope than tax avoidance. Tax shelters involve an aggressive interpretation of existing law to apply in unintended ways. As discussed more fully below, however, tax avoidance includes both tax sheltering and the use of intended tax preferences. Tax preferences are instances where Congress deliberately attempts to stimulate certain activities by allowing taxpayers to report less than the true economic income from such activity (e.g., the exclusion of interest income on state and local bonds).

Despite their different reach, tax sheltering and tax avoidance concerns share a common thread: the mismatch of tax and book reporting. This confluence might suggest a comprehensive linked approach as a potential panacea. Under current law, corporations determine book and taxable income under separate rules: (i) book income is determined in accordance with generally accepted accounting principles (‘GAAP‘), while (ii) taxable income is determined under the Internal Revenue Code (the ‘Code‘). Under a comprehensive linked regime, however, there would be one set of rules for public corporations: GAAP. Corporations therefore would pay tax equal to the statutory tax rate times the book income reported on the financial statements. By eliminating the book/tax gap, the comprehensive approach appears promising as a way to eradicate both tax sheltering and broader tax avoidance.

The comprehensive approach, however, has already been discredited adequately by a rich body of existing commentary (largely related to an earlier period of significant book/tax gaps in the mid 1980s). Most significantly, the comprehensive approach selectively denies intended tax preferences to a limited group of taxpayers (i.e., corporations subject to the linked regime). As discussed below, this selective denial has serious flaws from a tax policy standpoint.

Previous commentators have maintained that the comprehensive approach’s shortcomings discredit any linkage between the book and tax systems. This Article argues, however, that a more refined book/tax linkage can curb tax avoidance while simultaneously evading the comprehensive approach’s decisive weaknesses. Under such refined linkage, specified tax deductions would be disallowed in the absence of matching book deductions. This matching rule would target those items more indicative of unintended tax avoidance than the intended use of designed tax preferences. There would not be comprehensive linkage, as other (i.e., unspecified) items would remain unaffected by their book treatment. While the entire book/tax gap would not be eliminated, the limited approach suggested here holds promise as a tool in narrowing the portion of the book/tax gap attributable to corporate tax shelters.

The limited approach, by design, therefore does not tackle the broader tax avoidance issue raised by the recent private study. This Article leaves the broader reach of corporate tax avoidance (i.e., the use of intended tax preferences) to a second well-developed line of prior commentary. That is, independent of the public company book/tax differential, the merits of increasing taxable income through the elimination of tax preferences have been long debated. While elimination of all tax preferences would have the ancillary effect of narrowing the book/tax gap, this Article looks beyond the tax preference debate for two reasons. First, tax preferences will remain part of the current landscape. Second, even assuming unrealistically the removal of all tax preferences, items contributing to the book/tax gap likely would remain. Thus, this Article focuses on what the book/tax gap tells us independent of the tax preference issue.

This Article proceeds as follows. Although previously discredited, the comprehensive approach and its ample commentary retain great utility in assessing the appeal of the more refined approach. Section II.A first describes the initial appeal of using the book income of public corporations as their tax base under the comprehensive approach. Section II.B then briefly describes two different possibilities for comprehensive linkage which have been considered previously: (i) pure conformity which would equate tax and book income, and (ii) a floor version under which taxable income would equal or exceed book income. Section II.C finishes the foundational work by describing the most critical shortcoming of the comprehensive approach, principally related to the tax preference issue raised above.

Section III then considers the more refined, or non-comprehensive, approach suggested above. Section III.A analyzes the extent to which this approach preserves the initial appeal of book/tax linkage without running afoul of the comprehensive approach’s most disabling defect. Section III.B then subjects the non-comprehensive approach to a more thorough evaluation demonstrating how its special attributes respond to some of the major criticisms previously raised with respect to the comprehensive approach. Alternatively, the Section also discusses how the lack of comprehensiveness leaves some gaps in coverage. To provide perspective, Section III.C contrasts the non-comprehensive approach with another recently considered alternative, which subjects transactions generating reporting discrepancies to additional penalties and/or registration requirements. A comparison of these two limited options presents a final way to balance the relative strengths and weaknesses of the non-comprehensive approach suggested.

Finally, Section IV considers two potential deeper tax policy objections arising from the selective application of this special regime to public corporations. Included therein is a discussion of the supporting rationales for subjecting only public corporations to the linkage requirement.

Author Information

Associate Professor of Law, Benjamin N. Cardozo School of Law, Yeshiva University