Resorting to GAAR?

By James J. Tobin, Esq.  

Ernst & Young LLP, NY


I had the occasion recently to read through the recommendations of the so-called GAAR Study Group commissioned by the U.K. tax authority (HMRC) with respect to the question of whether a General Anti-Abuse Rule (GAAR) regime was needed in the United Kingdom and, if so, what it should look like. The report recommended that a GAAR be introduced, and since then the HM Treasury has held an informal public consultation, announced in its Autumn Statement that indeed it does intend to introduce a GAAR, and published draft legislation and guidance. GAAR provisions are quite common now around the world and seem to be increasingly relied on by tax authorities as a means to challenge tax results that they view to be inconsistent with the "spirit" or intent of the tax law. While the underlying rationale and motivation behind enacting a GAAR is understandable, I find the global trend of increased resort to GAAR provisions to be concerning. The inconsistency of the standards for applying GAAR country to country creates huge uncertainty for multinational corporations and the subjectivity inherent in those standards can encourage tax authorities to over-rely on GAAR as their tool of choice in their administration of the tax system.

So my first reaction to the potential for the United Kingdom to join the GAAR club was not positive. As a non-U.K. practitioner, I always had the impression that U.K. jurisprudence was rich with substance-over-form case law, like Ramsay, Furniss v. Dawson, Craven v. White, etc., which I perceived as an adequate means to address "tax schemes" that did not conform to economic reality. However, after reading the report and recommendations of the Study Group, which was chaired by eminent QC Graham Aaronson, I'm more inclined to see the usefulness of a U.K. GAAR. My epiphany is in large part driven by the balanced recommendations of the Study Group, which not only recommended a GAAR as an effective tool for tax authorities with respect to artificial tax transactions but also recommended building in safeguards for taxpayers to protect against the GAAR being asserted with respect to reasonable tax planning measures a taxpayer would implement in the normal course of business. Hopefully, these essential recommendations of the Study Group will be adopted by HM Treasury in the United Kingdom and also will be carefully studied by other countries that have an existing GAAR or are considering the adoption of one.

The Study Group's rationale for concluding that a GAAR was needed in the United Kingdom was based on a thoughtful analysis of the case law precedents, leading to the conclusion that the courts did not always have the power to reach a tax result consistent with the economic substance of a transaction. A case heavily relied upon to reach this conclusion was the so-called "SHIP2 case," HMRC v. Mayes ((2011) EWCA Civ 407). In that case, an individual generated a significant allowable loss in excess of his economic loss due to anomalies in the calculation of gains and losses with respect to a life insurance contract. The fact pattern was somewhat similar to the so-called rent-strip transactions seen in the United States.  Large gains were realized outside the U.K. tax net, followed by a taxable transaction in which the disproportionate remaining tax basis exceeded eventual proceeds of sale, producing a large onshore loss.  The court in Mayes felt unable to override the formulaic statutory approach to computing gain or loss, which in this case resulted in a large pre-arranged tax loss. The Study Group found this result unreasonable and saw it as a good example of why a broader GAAR was needed.

In its recommendations for a GAAR, the Study Group was emphatic about the need for adequate taxpayer protections and thus the overall design parameters were that any GAAR provision should have limited focus, contain taxpayer protection against overreaching by the tax authorities, have clear procedures for its application, put the burden of proof on the government in applying it, and provide published precedents as guidance on when it had and had not been applied.  Seems like a very sensible and civilized approach to me - unfortunately, one not necessarily consistent with the GAAR practice in some countries.  But, of course, at this point these are only Study Group recommendations, still to be formally adopted and applied in practice, so wait and see.

So what are the Study Group's recommended core terms for the GAAR? The GAAR would apply to abnormal arrangements that achieve an abusive tax result and are designed to achieve such a result. However, a tax result would not be considered abusive if it is regarded as the result of reasonable tax planning.

If the GAAR applies, the "counteraction" available to the tax authority would be either to ignore the effect of the abnormal arrangement or to substitute a more normal arrangement and determine the tax on that basis.

The Study Group goes on to elaborate on these core elements. In defining an abnormal arrangement, it focuses on a transaction or step within a transaction that can reasonably be considered to have as its sole purpose or one of its main purposes the achievement of an abusive tax result. Note this is an area of some inconsistency among those GAARs around the world that have some standards regarding "abuse." In the United States, we are used to the "principal purpose" test of §269 (not quite a full-blown GAAR but a more focused anti-avoidance provision).  In Canada, a taxpayer must demonstrate a bona fide non-tax purpose.  But in New Zealand, the GAAR can apply if the tax result sought was more than merely an incidental purpose. And case law interpretation of these standards seems to vary even more widely. Hence, my concern about the uncertainty multinationals face when dealing with varying standards across multiple jurisdictions.

An exception for reasonable tax planning is a very important taxpayer safeguard. The Study Group recommends a so-called double reasonableness standard: "An arrangement does not achieve an abusive tax result if it can reasonably be regarded as a reasonable exercise of choices of conduct afforded by the provisions of the Acts." The use of the term "reasonable" twice - is it reasonable to view the tax planning as reasonable? - should allow most normal tax-driven choices, such as decisions regarding debt or equity, holding company locations, treaty choices, etc., in the typical multinational commercial transaction. Ideally, this should operate with effect much like the famous opinion of Lord Tomlin in the Duke of Westminster case: "Every man is entitled if he can to order his affairs so as that the tax attaching under the appropriate Acts is less than it otherwise would be…." Anything less than this type of safeguard would seem unreasonable!

The second key safeguard area is with respect to the procedures recommended for the use of the GAAR. First of all, the Study Group clearly recommends that the burden of proof of asserting the GAAR be on HMRC. It recommends that there be central control within HMRC by a designated officer who would approve all GAAR assertions.  And it recommends that an independent GAAR panel be formed to review all proposed GAAR assertions and make public recommendations on whether such assertions are appropriate. The proposed panel would consist of three members. Its findings would not be binding but certainly would be expected to be influential in HMRC's decision to go forward with a GAAR assertion.

All this sounds very comforting and likely to avoid "I don't like the tax result"-motivated assertions of GAAR by tax authorities in the field, which in other countries often occur or often feel like an unspoken threat at the exam level.  Of course, the GAAR is intended as an additional tool for HMRC so, if enacted as proposed, there will no doubt be situations where it will be successfully asserted and all U.K. taxpayers will need to assess their planning with a view of what would be considered reasonable.  That is the type of question that if you have to think about it for too long, you probably have a problem and should consider alternative transactions or steps.

So the prospect of a U.K. GAAR as recommended to the Study Group doesn't seem worrying and perhaps makes good sense from a tax policy standpoint. And as I was coming to this view on the U.K. GAAR proposal, I came across the recently released recommendations from the European Union (EU) regarding "an Action Plan to strengthen the fight against tax fraud and tax evasion." There are a number of categories of recommendations for the Member States included in the Action Plan. Most relevant to this commentary are those under the category of recommendations on Aggressive Tax Planning. And most relevant from that category is, lo and behold, a recommendation that "Member States are encouraged to introduce a GAAR in their national legislation." And the EU Commission was also kind enough to provide suggested draft language.

The core language recommended is as follows: "An artificial arrangement or series thereof which has been put into place for the essential purpose of avoiding taxation and leads to a tax benefit shall be ignored. National authorities shall treat these arrangements for tax purposes by reference to their economic substance." The Recommendation's elaboration on some of the terms includes some troubling elements and some seemingly reasonable elements. The purpose of an arrangement consists in avoiding taxation where "regardless of any subjective intentions of the taxpayer, it defeats the object, spirit, and purpose of the tax provisions that would otherwise apply." Not sure if I would know that when I see it, given the spirit of many technical tax provisions may be difficult to discern. In defining "essential," it seems as long as a non-tax purpose is more than "negligible," the tax purpose would not be essential. But given that purpose is an objective not subjective test, it is hard to see how this would be evaluated.

It is my understanding that nearly all EU countries except the United Kingdom already have something akin to a GAAR, I am not sure why this is recommended - perhaps primarily to have a consistent approach throughout the EU. In principle, I think that is a good idea, given my earlier comments about inconsistent standards.  However, it seems to me it will take a major effort to come up with both a consistent rule and one that can be drafted to insure consistent and "reasonable" application in practice across all EU countries.

Overall, I am obviously not a big fan of GAARs, which I think can lead to uncertainty and risk of tax authority misuse.  Rather, I prefer specific and focused anti-avoidance rules that are more clearly targeted and thus easier to apply and administer. However, I also understand the attraction of a GAAR from a tax administration perspective. And the reality in today's world is that they are a tax system feature in more and more countries globally. The thoughtful safeguard features in the U.K. Study Group recommendations are critically important while in no way undermining the intended effectiveness of the GAAR provision. I am sure the debate about how GAARs should operate will continue and I hope these features will be a significant focus of the debate. A GAAR rule with appropriate safeguards that is adopted consistently across the EU, replacing the varying standards that exist today, could be a good thing. But it would be a big step and no doubt a very difficult one to achieve.

This commentary also will appear in the February 2013 issue of the  Tax Management International Journal. For more information, in the Tax Management Portfolios, see Nias, Ross, Khvat and Morison, 989 T.M., Business Operations in the United Kingdom.

  * The views expressed herein are those of the author and do not necessarily reflect those of Ernst & Young LLP.