It is no secret that tax legislation is extraordinary complex. Part of the reason is the subject matter itself. Economic reality and legal doctrines do not necessarily coincide, and when they do not then taxpayers frequently can exploit the mismatch to achieve beneficial tax results. One of the swords that administrators wield to combat this phenomenon is the general anti-avoidance rule (GAAR). The question of the limits to which taxpayers may go to lower their tax liability was originally – at least in common law countries – a product of judicial doctrine. Today many countries have codified the rule or at least certain key elements of it (the closest the United States has to a statutory GAAR is IRC §7701(o), which clarifies the judicial economic substance doctrine). However, whether codified or not, GAARs by their nature are problematic. They call upon the courts to ignore the express words of the statute to prevent tax avoidance. However, one would have to be extraordinarily naïve to believe that taxpayers do not routinely structure their affairs in response to tax rules. Thus the question of when it is legitimate to invoke a GAAR is not a simple one.
Professor David Duff begins by describing the rationale for GAARs, and he considers both consequentialist and non-consequentialist arguments. Consequentialist arguments concern the revenue loss, the direct economic cost, the resultant allocative inefficiency, vertical inequity, and the affect upon tax compliance generally. A piecemeal approach to fixing the gaps and inconsistencies in the tax code is often impractical due to policy considerations such as reducing complexity, accommodating liquidity concerns, and promoting social and economic policy objectives. Professor Duff adds that when tax avoidance opportunities are the result of poor policy choices, political restraints often make it difficult to change those choices.
Non-consequentialist arguments are concerned with the “inner morality of law,” in Lon Fuller’s phrase. The focus is more on realizing legislative intent and less on realizing the substantive aims that are incorporated into the law. However, as Professor Duff points out, this type of argument cuts two ways. One could rely upon it to delegitimize the entire GAAR enterprise: GAARs are vague, uncertain, and retroactively applied, and so statutes should be implemented as enacted. On the other hand – and this is the view that Professor Duff indicates that he prefers – GAARs actually support the ideal of legality by respecting the objectives and purposes of tax legislation.
The next part of the article discusses the design of a GAAR. As an initial matter, Professor Duff supports the codification of GAARs, particularly given the “quasi-constitutional” right that taxpayers have to arrange their affairs in order to minimize tax. Regarding the design itself, he points to both the subjective and the objective elements in a typical GAAR. The subjective element involves the purpose (a term that he distinguished from “motive” in criminal law) of the transaction. Here, too, there is a debate with regard to the legitimacy of examining the taxpayer’s purpose. On the one hand, it is difficult to talk about avoidance if there is no intent to avoid. On the other hand, it does not seem fair that identical transactions should be taxed differently simply because of the subjective purpose of the taxpayer in undertaking the transaction. Furthermore, invoking a GAAR because of one person’s purpose could affect the tax liability of others had no such purpose. However, the subjective test may be less subjective than it might appear at first glance. In some cases, the rules effectively ask for a reasonable person standard, as in the Canadian GAAR (“may reasonably be considered to have been undertaken or arranged…for bona fide purposes other than to obtain a tax benefit”). Even when the test is nominally subjective, courts generally rely upon objective manifestations of taxpayer purpose.
Professor Duff concludes by discussing the “misuse or abuse” test.
Tax avoidance is not a new phenomenon. It is as old as taxation itself (interesting parlor game for tax geeks: try to come up with the earliest example of tax avoidance) and will doubtlessly continue for as long as taxes exist. The debate regarding the permissible limits of tax avoidance is likely to display equally longevity. This article is an important contribution to the ongoing discussion.