Hiding in Plain Sight? Changes to “Transfer Pricing” and the GAAR



Drawing on comparative references to transfer pricing legislation in the United Kingdom and Australia and on the legislative “General Anti-Avoidance Rule” in the United Kingdom, these comments probe to identify and suggest responses to the underlying “problem” that pervades transfer pricing and Canadian GAAR jurisprudence but is, it seems, either thought or thought to have proven to be inadequately addressed by Canadian tax legislation.  Even if there is not a “problem” with our law, the perception exists that there is, and perception being reality and this being such an important issue, it is desirable to identify the core “problem” before considering “solutions”, rather than to hypothesize “solutions” to which then an interpretation of the “problem” must be shaped to fit.  In sum, these comments ask whether it would be desirable, and entirely responsive to the perceived shortcomings of Canadian transfer pricing and anti-avoidance law, to incorporate the kind of Australian and UK infusion into their statutory law of the OECD Transfer Pricing Guidelines on a qualified basis, but without a separate anti-avoidance notion built in to section 247 of the Act, and to incorporate the UK notion of “abusive” transactions in its GAAR into ours with general application to the entire Act including section 247. These comments suggest that the “complaint” underlying most, if not all, “problematic” transfer pricing and GAAR cases is an inability on the part of the Crown to be able to critically question the fiscal significance of various legal constructions and to overlook them for purposes of the Act while still operating within the principled boundaries of what loftily is described as the “rule of law” – an inability fueled by the absence of specific statutory authority and indeed instruction to apply the law this way.  A parallel and like disability for same reasons is, perhaps, insufficient statutory instruction guiding taxpayers and their advisers toward an obligation to adopt such an outlook and prudent awareness with fiscal discipline on advice giving and taking. The UK and Australian legislative approaches, grounded comparatively as they are in familiar legal traditions generally and in taxation specifically, are illuminating about how, if necessary, we might consider how to solve a “problem” that we have and how to avoid advocating and adopting solutions that may miss the mark.


“Calling Them as I See Them”

No matter how well an umpire or other kind of rules official understands and can speak at length about the rules of the game, at some point balls and strikes, and penalties, have to be called in the moment - simply, in a straightforward way, in a way that leaves no doubt about what call was made and why. It may be the “wrong” call - But, a call happens.  And it happens in and in response to the immediacy of the actual circumstances, without becoming an inquiry into the substantive meaning or integrity of the rules.  Too, most of the time, by some mixture of intuition, common sense, and self-awareness (perhaps “guilt”) the player understands the call is right, no matter how animated the theatrics of their occasional “what me?” denials and gesticulations.  Tax is not a game.  But the metaphor about calling the play – the actual play, versus studied introspection about the rules, strikes me as apt.


A Cautionary Preface

This commentary proceeds on the basis of several assumptions. It assumes that dissatisfaction on the parts of the Canadian tax and finance authorities with judicial reactions to alleged unreasonable tax avoidance including in the transfer pricing setting, and changes in the direction and tone of international reactions to tax avoidance arising from the Organisation for Economic Co-operation and Development (“OECD”) - led BEPS (“Base Erosion and Profit Shifting”) and “Pillars One and Two” projects, makes inevitable some refurbishment of the Canadian Income Tax Act’s sections 245 (the General Anti-Avoidance Rule, “GAAR”) and 247 (the Canadian legislative transfer pricing rule inspired by the long-standing “arm’s length standard” enshrined in the OECD Transfer Pricing Guidelines which build out Article 9 of the OECD Model Tax Convention).  It also assumes, however, that a large-scale redesign of the Act to ferret out all manner of specific tax avoidance / minimization opportunities and to neutralize them is neither in the cards nor the scope of ongoing consultations about the adequacy of sections 245 and 247.

It would be an error, however, to assume that this is necessarily correct, and that the Act somehow is inadequate in specific or general ways to address even the kinds of tax avoidance that have not attracted judicial sanction.  It may be, as these comments implicitly reflect, that sections 245 and 247 when properly applied are up to serving well enough the objectives for which they exist so that changes would not be required.  But, sensing that that is not likely to be a satisfactory conclusion, any refurbishment ought to respect not only the goals sought to be achieved by these provisions, but the underlying currents of Canadian tax law, and a careful awareness of which would suggest that any changes thought to be required can be accomplished modestly. Changes that would do the least damage to what we have, that would reinforce the underlying tenets of what we have, and that reflect fundamentally what legislation directed as sections 245 and 247 is intended to accomplish. Modesty is a worthy goal, to avoid any more unnecessary opacity in the legal rules and doctrine associated with transfer pricing and anti-avoidance measures, notably the GAAR. And Canada is not alone in facing legislative challenges in these areas. There are useful reference points elsewhere in the world – described in these comments as hiding in plain sight – that usefully might supply more concrete indicators of more precisely what constitutes untoward tax avoidance without stressing the Act with virtually endless philosophical introspection about the roles of “economics” and “substance” and “reality” from one perspective or another and the opportunities for further unchecked avoidance fueled by the absence of clarity about what fairly the tax “system” can expect of measures to mitigate undue tax avoidance.

This, though, starts with being able actually to clearly identify and say what the core “problem” is that shelters under the unspecific notion of tax avoidance, before trying to address it.  It is not a question of drafting imperfection or the absence of suitable lists of “good” and “bad” planning.  It is something more fundamental.

This is the point of these comments, which, it is acknowledged are deliberately not cast as a typical scholarly paper, in order to try to bring the issues addressed into the starkest possible relief.  It is the luxury and indulgence of a blog post that allows this result, even for comments that have grown well beyond the usual length of even a long post.  If a consultation proceeds, possibly it would engage with some of these ideas and, as these comments also reflect, engage with the experiences of other countries familiar to us where what I suggest here is the core “problem” has attracted direct legislative responses in kind. I acknowledge conversations about these comments with some close friends and colleagues whose insights and judgment I respect, and which improved the clarity of my thinking.


Reviewing the Bidding

In the 2020 Fall Economic Statement and the April 2021 Federal Budget, it was announced that the Canada Revenue Agency and the Department of Finance are consulting on changes to section 247 and the GAAR.

No doubt this consultation has been inspired by all levels of Canadian courts’ notable disinclination to sustain tax assessments grounded in one way or another by pleas – that is, pleadings in the courts and not exhortations, although maybe those too – to overlook, overturn, recharacterize, ignore taxpayers’ arrangements on the basis of their “substance” as advocated by the Crown.  These arguments bear various labels and reflect legal tacking in various directions, most commonly framed by notions of “sham”, “economic substance”, “reality” and the like. The courts’ reactions often have been unsatisfactory to the Crown and indeed more generally as systemic, focused guidance.

More significant for the tax system as a “system”, is the manner in which the cases have unfolded, with the Crown sometimes seemingly trying via argument to shoehorn cases into proscriptions not easily or even readily found in the Act and how it has been interpreted and applied customarily, has left the law in an unpredictable state.

The problem? Shoehorns are not shoes. The courts have noticed.

It may be that positions advanced by the Crown are not fundamentally unsound in terms of what the Act envisages. Rather, even with the benefit of a constructive and purposive approach to interpreting and applying the Act according to the oft chanted mantra “text, context, and purpose” (which, by the way, did not originate in the Canada Trustco decision and is a long-standing approach evident for example even in Driedger’s celebrated guide to statutory interpretation), suitable grounding in the Act for sought outcomes cannot seemingly be found – as a legal matter.  And, as the Supreme Court admonished in the Canada Trustco case and Justice Strayer said in no uncertain terms in a recent non-tax case, Attorney General of Canada v. Dr. David Kattenburg and Psagot Winery Ltd. (2020 FCA 164) (a case that had escaped my attention until I noticed a reference without citation to what Justice Stratas was reported to have said in the Financial Post and I went searching for it), courts are not legislators or political policy makers.

What courts are asked to do and say has to follow a pathway originating in the law, not aspirations or assertions by any litigants about what the law ought to be when it is not that.  In many situations, indeed, there might be general agreement about what is “impugnable” or problematic about adjudicated situations, by taxpayers and their prudent advisers as well as parties on the governments’ side.  But we do operate in a legal realm and we cannot simply shunt it to the side when it yields inconvenient outcomes, no matter how concerning they might be from the vantage of economics and alternate reality.  Absent guidance in the law to deter what the law considers unacceptable outcomes; it is not surprising that outcomes grounded in the law – the private law (non-tax law referred to in these comments is described interchangeably as “private” or “general” law) as well as the tax law – have often prevailed.  There are notable recent examples of this in cases adjudicated before or presented to all levels of court that hear tax cases.

This state of affairs begs a number of questions, if the governments’ inclination about how the tax system should function more effectively as a “system” can prevail.  They can all, however, be encapsulated in these three questions:


  1. If opaque references to purpose and incompatibility with the Act because of contextual abuse (with exclusive reference to legislative context according to the mantra of “text, context, and purpose”, casting tax avoidance cases as more akin to a refence on the meaning of legislation) are insufficient to communicate unreasonable or unacceptable tax outcomes, why?
  2. If these kinds of references - the spine of both sections 245 and 247 - fail to offer expansive but law-grounded direction and authority to courts to achieve outcomes explicitly or implicitly justified by the Crown according to perceptions of what is economic or real, what would the Act need to say to establish the missing authority that would require courts to examine more critically legal constructions adopted by taxpayers, which even if implemented soundly in legal terms have doubtful fiscal significance in the circumstances before the courts?
  3. How could changes to the Act in this regard be achieved which respect the legal environment in which events must occur while critically evaluating the fiscal significance of legal constructions adopted by taxpayers, and over-looking it without otherwise undermining the law, where warranted?

These are the key questions.  Often they are obscured by the chatter of commentary about the GAAR and the search by courts for sensible outcomes and a way of communicating some measure of direction to favour predictable, principled outcomes.  But these are the key questions and from them we can anticipate answers offering more straightforward instruction than the cacophony of inward looking GAAR commentaries, replete with linguistic inquiries that rival the parsing of religious tracts, would suggest is possible.

Are the answers to these questions hiding in plain sight?


A Different Tack – Some Initial Comments:  What is the “Problem”?

Before ever launching into suggested solutions for a problem, it is best to distill and crystalize the essence of the problem. Too much of the discussion in the transfer pricing and GAAR arenas is cast in terms of the perceived legal incapacity or uncertainty of the Act, as if each time we were deciding legal reference cases, when in fact the “problem” to which the consultations is directed is much simpler.  I suggest, too, that this problem – like most transfer pricing and GAAR cases – is not really about “the law” or at least “the tax law” – but instead reflects a dissatisfaction with the fundamental and necessary deference the tax law in Canada (and most everywhere else) pays (because at basis level it has to) to legal constructions.  The legal constructions I have in mind are organizational forms and transactional arrangements – the nature of “property”, “contracts”, entities that are the progeny of statutory or other general law that the tax law recognizes such as “corporations” and “trusts” which respectively exist as the legal progeny of statutory law and as the manifestation of triangular relations between an owner of property dedicating it to the interests of others which in the meantime is in the custody of an intermediary who administers and supervises its existence and use according to the original owner’s instructions.

A question seldom if ever asked seriously in a tax law setting is:  Why do organizational forms and transactional relationships have fiscal significance?  Who ever said that they must?  They may serve a variety of important non-tax purposes but even when the parties to them respect and abide by them, why are they or why must they be the determinants of fiscal and tax policy, of taxation as such?

Even in the Canadian tax environment we rarely if ever ask or answer this question. When concerned for example with preferred shares being used as debt equivalents or to transmit value and tax benefit of losses to unintended third parties, we do not as some countries do recharacterize debt as equity. No, instead we respect that legal construction for what it is, but we change the intensity with which dividends are taxed by denying the inter-corporate dividend deduction in situations that bother us with fundamental features of our “system” of taxation in mind.  When, infamously, public trusts and partnerships – “specified investment flow throughs” or “SIFTS” were employed to fragment corporate income so as to imperil – “base erode and profit shift”, or “BEPS” to use a now prominent notion – the corporate tax base, we did not deny the existence of partnerships and trusts and recast them as corporations. Instead, we took what we found, but for tax purposes reconditioned the application of the Act to them – where it mattered to us – by approximating their relevant circumstances, and in particular “entity” and stakeholder income tax responsibilities, to what would happen if the affected partnerships and trusts had been corporations with shareholders.

The “problem” that is at the heart of both the transfer pricing and GAAR consultations, and indeed can be seen in virtually all judicial challenges in those realms, has been doubt about the necessary fiscal significance of arrangements, particularly the introduction of intermediate legal constructions as entities and transactional arrangements. These, it is contended, effectively are unnecessary to achieve a taxpayer’s objections as they are ordinarily conducted. They are “abnormal” in that sense and often “contrived”, taking on contorted shapes that seem even to challenge the typical application of general law. Their outcomes, and objectively their objectives, are to secure access to “tax benefits” that would not “naturally” arise from the “normal” and otherwise undisturbed arrangements of affected taxpayers. They are “abusive” when examined with the Act’s nature as a system of taxation in mind.  We would like to discard legal determinism in these moments – not absolutely or forever – but only when the otherwise evident underpinning of the tax system demonstrably would be undermined by ceding control of the tax outcomes to the private law that exists for other reasons and which the tax system – any tax system – generally takes simply “as found.”

If we think about recent transfer pricing cases, using various arguments considered later, the fundamental tack is to want to disregard otherwise legally effective organizational and transactional arrangements in favor of an alternate “characterization” – a “recharacterization” – that the OECD now would describe to be the “commercially rational” “accurate delineation” of what a taxpayer has done, even though it’s entirely clear as a legal matter, within the confines of a legal system that is otherwise respected, what the taxpayer did. I do not have to name the recent cases for this to be clear enough.  The same can be said for GAAR cases, and even recent cases that are not, or not cast as, either transfer pricing or GAAR cases but in which, implicitly the concern is unreasonable tax avoidance stoked by organizational and transactional artistry.

That is the problem we are solving for.  We are not looking for a “new” transfer pricing regime or a new GAAR rule.  We are looking for statutory guidance and authority to indicate when it is not only acceptable but necessary to overlook or ignore the effects of the fiscal significance of legal arrangements that are benignly purposeful for other purposes. Correspondingly, we want that statutory direction to supply the means for legally distinguishing “acceptable” from impugnable tax avoidance.  And the courts are communicating, quite clearly, that without that guidance in the Act, they are disinclined to disregard that to which the law gives respectable life, simply because the tax authorities would prefer to assess on the basis of the “economics” of situations – cheered on in this regard by the commonly perceived tendency of OECD-grounded transfer pricing analysis, or they contend in either the transfer pricing or GAAR realms because economically equivalent outcomes may be shaped entirely differently organizationally and transactionally with more or less different tax outcomes, with choosing the least tax outcome perceived as somehow unacceptable, again reflecting the sentiment in the narrow transfer pricing context of the OECD – led base erosion and profit shifting work.

The creeping - and I would say unreasonable - deference accorded to “economics” in both transfer pricing (it might seem odd to say) and GAAR settings has other implications besides an implicit and rather brutish justification for displacing the law.  It is not necessary to stand on a “rule of law” pedestal to be skeptical about the desirability of an indiscriminate or haphazard or episodic denial of the legal outcomes of events actually undertaken by taxpayers to appreciate these implications. Tax law and the law that underlies it, are law. Economically equivalent outcomes can and routinely are implemented relying on different legal constructs. How transactions are undertaken, according to the private law, affects how the tax law receives them. Tax practitioners, and in particular those who are legally trained in at least the interstices of the law underlying the tax law and the general law, are or should be well placed to understand and evaluate theses distinctions which do have differences.  The interpretation and application of the law is not a social science exercise; it is not an exercise in deciding among alternative theoretically plausible alternatives when what happened as an evidentiary matter is clear.  Economics no doubt has a role to play as an empirical resource in a typical legal analysis that requires compiling and understanding evidence about taxpayers’ circumstances, but as Canadian statutory and jurisprudential tax law has long made clear, it is not the basis on which assessments should be raised. Yet, the contemporary inertia of economics is increasingly hard to resist or even subject to critical evaluation, and in various ways it lies at the heart of the “problem” faced by consultations to recalibrate transfer pricing and the GAAR.

So, this is the problem: when, whether in a transfer pricing or GAAR context, is it desirable to detach taxation from organizational forms and transactional arrangements adopted by taxpayers, and where and how should that authority be found in the Act?

Now, some guardrails. What else do we know apart from the necessary dependence on the legal system to avoid interpersonal and commercial chaos, about the environment in which these questions arise that would affect how we conceive the problem as it is now to be addressed by the consultations, which would affect the simplicity and elegance and economy of ensuing proposals?


  1. The situations involving disputed transfer pricing and the GAAR which seem most galling to the tax authorities involve events, circumstances, transactions, and intermediaries, that the usual rhythm of taxpayers lives seemingly do not require, and that are perceived to be “abnormal” or “contrived” in this regard, even if not necessarily by their intrinsic nature, or being considered “bad,” or “revealing” “bad behavior”. The tax authorities would like to disregard these.
  2. Canadian tax practice is infused generally, in practice, by an essentially unreserved application and influence of the OECD Transfer Pricing Guidelines. Assiduously, their incorporation by reference into Canadian tax practice has been as interpretative and administrative guidance which, we have been reminded by the highest Canadian court, is not law and does not displace the law as it has been enacted if determined to be inconsistent with the Guidelines.  The point, though, is that for all intents and purposes the Guidelines are, and for as long as they have existed have been, influential and generally have been the framework within which transfer pricing occurs.  Indeed, some - I am among them - would regard the Guidelines as additional Commentary to Article 9 of the OECD Tax Convention with - as venerable Canadian jurisprudence has confirmed - serious probative significance, in a legal analysis ultimately tethered as transfer pricing analysis to tax treaties, subject to temporal limitations applicable to evolved and changed tax treaty commentary.  However, even in a statutory encapsulation of the sort discussed below, Canada’s even more close embrace of the Guidelines “as law” still would not be, would not need to be, and from a general law perspective should not be unreserved or unqualified, even if only by stipulating that their ambulatory application must be subject to some kind of official legal ratification or confirmation to which Canadian law makers advert to gauge their continuing suitability. Another way, as noted later, would be to adopt the tack of other countries taking this route, by legislating the expectation that the transfer pricing rules as otherwise legislated would be interpreted and applied consistently with the OECD Transfer Pricing Guidelines.  A conscious awareness needs to be, and stay, firmly in view that the Guidelines are no more than the collective statement by tax administrators and finance officials of their views about applicable law and practice (though they may be as useful in practice that they are), but that it is the obligation of law makers to make balanced, purposeful, instructive, balanced, and fairly conceived law.
  3. We have considerable statutory and judicial experience with anti-avoidance legislation and doctrine. Without meaning to be glib or clever, most experienced practitioners have a pretty good idea when planning may cross the line and become “abusive” in a systemic sense by introducing elements of artifice in relation to what taxpayers’ circumstances would ordinarily require.  Stripped of more analysis and discussion than its essence really requires, purposive tax avoidance generally involves in one way or another “unnatural actions” measured with reference to a taxpayer’s normal circumstances or controlled reliance on choices among the “legal fictions” underlying taxpayers’ behavior which are only possible when there is no genuine adversity of interest between parties to transactions because their parties are commonly controlled.  Hence, in the most topical example of this, transfer pricing, the tax system essentially becomes elective in ways far beyond explicit statutory elections. David Rosenbloom, the celebrated US tax practitioner and thinker has referred to this many times, for example in: Banes of an Income Tax:  Legal Fictions, Elections, Hypothetical Determinations, Related Party Debt, Vol. 26 2004 Sydney Law Review 17.

Essentially, because the law is a web of legal constructions that exists because the law says they do, informed judgment is necessary to determine in any particular context whether they ought to be respected as they are, uncritically, or, if not, why and with what effect.  Law and its instructions are necessary as the framework in which “we” in all manner of condition interact without chaos and resulting mutual detriment. That is the case regardless of what theory of law one embraces.  But that necessary adherence to law and its expressions in, for example, organizational forms and transactional expressions, is really the bur under the Canadian transfer pricing and GAAR saddles, and we would be well advised to focus our attention this way, of course on a principled, evidence-grounded, and consequently statutorily coherent basis

What might we do, with these considerations in mind, to find what may be hiding in plain sight?

How can we confine transfer pricing to the so-called comparative or other income allocation “bread and butter” “pricing aspect” in paragraphs 247(2)(a) and (c), which in any event is the essence of the Tax Court’s and Federal Court of Appeal’s decisions in the Cameco Corporation case (2018 TCC 195 (TCC); 2020 FCA 112 (FCA); leave to appeal to the Supreme Court of Canada denied) about the significance of paragraphs 247(2)(b) and (d) as an analytical heuristic for applying paragraphs 247(2)(a) and (c)?  How can we do this by excising from section 247 its GAAR aspect associated with the purposive tax avoidance that ostensibly is the target of paragraphs 247(2)(b) and (d), while still in the Act’s “system” retaining in a more effective, transparent, and principled way that line of contention in circumstances where it is most appropriate?

How do we reinforce the GAAR with a principled articulation of what roughly one hundred years of contending statutorily and doctrinally with tax avoidance tells us is “abusive” and why?

The suggestions, developed in what follows are these:


  1. Remove what amounts to the specific, general anti-avoidance rule from section 247 by making clear that paragraphs 247(2)(b) and (d) serve only the role assigned to them by the Tax Court and Federal Court of Appeal in the Cameco Corporation case, with an overarching statutory statement, if one be needed, reminding that the application of section 247 like the rest of the Act is subject to section 245.
  2. Make statutory what in any event is the reality of Canadian transfer pricing by giving the OECD Transfer Pricing Guidelines the same kind of legal significance that they have in the United Kingdom and Australia, and effectively what they have in the United States in so far as the Guidelines fundamentally are a “global” adoption of US law (and got that way on purpose). Perhaps, as noted earlier, this would not be an unreserved or unqualified “delegation” of legislative effect to the OECD and its members – for public law and constitutional law reasons this conceivably would be important - but would qualify, for example, this authority with the obligation to evaluate the Guidelines’ “consistency” with relevant Canadian tax law and practice, from time to time.
  3. Introduce a definition of “abusive” in section 245 along the lines of the United Kingdom’s GAAR that directs those who interpret and apply the tax law in all capacities to determine whether arrangements with tax outcomes are “abnormal” or “contrived” in relation to a taxpayer’s normal condition and/or are expedient and opportunistic to secure tax outcomes that in circumstances that objectively are not aligned with the normal expectations of the taxpayer’s condition, and supplies the requisite legal obligation to do this.

Now, to explain.


Transfer Pricing

Why a Review?

No doubt, Canadian revenue and finance authorities are disappointed with judicial reactions to cases in either or both of those legislative environments.  Common to both is the inclination of the authorities to want to deny the significance for tax purposes of taxpayers’ organizational and transactional arrangements, even when taxpayers behave consistently with the intrinsic legal expectations of those arrangements establishing those legal arrangements to be not merely “form” but indicative of the kind and quality of behavior enabling justifiable reliance on them. In some cases, the government asserts that arrangements that, in its view, should be disregarded in applying the Act as “shams.” In other cases, notably in transfer pricing cases, the tax avoidance branch of section 247, in paragraphs 247(2)(b) and (d), have been asserted as justification for “recharacterizing” otherwise legally enforceable arrangements, the implicit requirements of which taxpayers have implemented and followed.

On both counts all levels of court have been unsympathetic. These cases have reinforced the requirement that “sham” requires active deceit by taxpayers, not merely an outcome the economic equivalence of which could have been replicated by other more taxable legal configurations.  As far as transfer pricing is concerned, in reconciling the two branches of subsection 247(2) of the Act, the “a-c” axis addressed to the comparative pricing of otherwise commercial transactions and the “b-d” branch supplying an analytical heuristic for transactions that lack an arm’s length analogue and are undertaken for tax avoidance purposes, the courts at all levels have said clearly that the “b-d” branch is not a license to disregard taxpayer’s transactions, that is to “recharacterize” them.  Rather, it supplies a comparative device, if the provision applies – that is, if for anyone and not merely a Canadian taxpayer there is no transactional analogue – it offers on the convenience of finding a kind of transaction that would yield a comparable economic/financial outcome in relation to which a taxpayer’s circumstances would then be tested in a typical way according to the “a-c” axis using that convenient reference formulation.

This is a particularly important judicial conclusion in light of how the BEPS Actions 8-10 wind has blown and how the 2017 OECD Transfer Pricing Guidelines have accordingly been recast.  It is a considerable set-back for the tax authorities – particularly in light of the inertia of evolved international transfer pricing guidance by the OECD embraced by Canada that prominently relies on notions of “accurate delineation” and “commercial rationality” of and “value created” by transactions to enable taxpayers’ transactions to be disregarded from the get-go in superimposing other transactional configurations compatible with economically equivalent outcomes.


What Do Tax Authorities Want?  What Have They Been Denied?  What Are the Questions?

It is helpful to begin an exercise in legislative review by paying attention to the questions as they have been answered to date by those with the authority to answer them. Approaching this with the kind of misplaced philosophical inquiry that transfer pricing and the GAAR seem to inspire, is not very useful.

The question: What do the tax authorities want and is that a justifiable, reasonable outcome, other things considered? If and only if it is, how does one get there? It is about these two questions that all of the transfer pricing jurisprudence is concerned.  And more times than not when the Crown fails, it is because the laws, or the Act more specifically, is not what they want it to be, assume it to be, and think that it ought to be.  Without judging the Crown’s position, when the right questions are asked, the answers may be a lot clearer than the muck of transfer pricing analysis and commentary would otherwise suggest.  They may indeed, be hiding in plain sight.

Essentially, the tax authorities want to recharacterize in light of the permissive direction, in particular after 2010 of the OECD Transfer Pricing Guidelines.  Without judging, they have wanted to substitute for transactions actually undertaken, other formulations they consider to be more compatible with how, in their view, transacting parties otherwise disinterested in each other would have behaved.

But what is lacking?  Legal authority - an anchor in the Act. Long have the OECD Transfer Pricing Guidelines, particularly their empirical and methodological devices, been administrative guidance followed more or less assiduously in line with other countries’ tax authorities in applying the legislative constructions of the arm’s length standard. However, as the Supreme Court of Canada and other lower courts have observed, those Guidelines are not in the same class as law – because they are not law.  It is not clear how Canadian courts would have reacted had Canadian legislative statements been more comprehensive or clearer on the point. Drawing some insight into statutory interpretation from now Lord Burrows, now of the United Kingdom Supreme Court, in the first of his three 2017 Hamlyn Lectures (Thinking About Statutes Interpretation, Interaction, Improvement, The Hamlyn Lectures 2017, Cambridge University Press). Finding a direct statement of the law’s meaning and the means to activate it in the law without resorting to what he described quite directly as an illusory and even fallacious search for parliamentary intention among other ways, so he spoke in another paper, “standing outside the legal system and applying to it some grandiose theory or feeding ideas and information and statistics in to those who may be in a position to influence policy choices” (Challenges for Private Law in the 21st Century, originally presented in 2015 at a conference in Australia), may well be not only salutary but necessary.  A hook, so to speak, for judges to be able to say: there is explicit direction in the law about how we should understand and apply these provisions because the statutory law has supplied us with the tools and the obligation to use them!

It may be that the search now undertaken by Canadian revenue and finance authorities has in mind the “Holy Grail” of prescriptive legislative certainty.  Though maybe a worthy aspiration, it is unlikely to be achievable, or, from the tax and revenue authorities’ perspective, even desirable.  Prescriptive rules are limiting and navigable.  That is the problem.  Transfer pricing, and for that matter its sibling GAAR’s, are highly factual and transaction or event oriented.  It is likely not possible to chronicle all the ways in which it is thought the law should apply in specific legislative instructions.  And even if it were, it would likely take us further into the swamp of transfer pricing rather than in the direction of dry land.

But it is clear from the cases that what the tax and finance authorities yearn for is the ambulatory application of the inspirational OECD Transfer Pricing Guidelines, as if they were law.  They foresee transactional redefinition and critical examinations of the fiscal significance of legal constructions and, subject to legal authority, supply the wherewithal to superimpose on legally effective transactions a veneer with a different tax polish, despite the significance for other purposes of those legal arrangements.  And, in the reasons of all levels of relevant court, it can be detected that with suitable legislative direction, those courts might well entertain that line of inquiry with needing to embark on hoary and often fruitless forays in to sham, or the intricacy, or not, of the “a-c” and “b-d” axes of subsection 247, or for that matter “lifting the corporate veil” by stealth that in a nutshell is often a detectable mindset in transfer pricing and GAAR cases. In effect, the outcomes from such permitted analysis would serve the perceptions of the tax authorities evident in how the decided cases were advocated by the Crown, without venturing into in hospitable legal territory.


Legislative Recalibration – Directionally, Without Legislative Clutter

If that is a fair assessment of the situation, then the refurbishment of section 247 seems pretty clear. Both the United Kingdom and Australia have legislated the OECD Transfer Pricing Guidelines as they exist from time to time, as overarching legal direction on how otherwise the transfer pricing sensitive provisions of their tax acts should be interpreted and applied.  For example, taken from the UK law, section 164 of the Taxation (International and Other Provisions) Act 2010, as amended:


“(1) This Part is to be read in such manner as best secures consistency between

(a)the effect given to sections 147(1)(a), (b) and (d) and (2) to (6), 148 and 151(2), and

(b)the effect which, in accordance with the transfer pricing guidelines, is to be given, in cases where double taxation arrangements incorporate the whole or any part of the OECD model, to so much of the arrangements as does so.

(3)  In this section “the OECD model” means—

(a)the rules which, at the passing of ICTA (which occurred on 9 February 1988), were contained in Article 9 of the Model Tax Convention on Income and on Capital published by the Organisation for Economic Co-operation and Development, or

(b)any rules in the same or equivalent terms.

 (4) In this section “the transfer pricing guidelines” means—

(a) the version of the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations approved by the Organisation for Economic Co-operation and Development (OECD) on 22 July 2010 as revised by the report, Aligning Transfer Pricing Outcomes with Value Creation, Actions 8-10 - 2015 Final Reports, published by the OECD on 5 October 2015, or

(b)such other document approved and published by the OECD in place of that (or a later) version or in place of those Guidelines as is designated for the time being by order made by the Treasury, including, in either case, material which is published by the OECD as part of (or by way of update or supplement to) the version or other document concerned and which is designated for the time being by order made by the Treasury.” [Emphasis added throughout]


To paraphrase, the OECD Transfer Pricing Guidelines are statutory law, in the condition they exist from time to time, and the rest of the application of the tax law, and I would say the private law to which it is accessory, are to be interpreted and applied accordingly and consistently with the Guidelines, whatever else the law other than the tax law might recommend for other reasons.

The Australian rule (subsection 815-135(2) of the Income Tax Assessment Act 1997, as amended to refer to the 2017 OECD Transfer Pricing Guidelines) which accomplishes the same outcome is very similar, with the emphasis being clearly establishing legislative authority for the Guidelines part of the tax law and the obligation to apply the material other provisions of the tax law in ways that reflect the influence and effects of the Guidelines.

There is no list.  There is no wholesale revision of the law.  There is no doubt.  Words to this effect readily and easily could be added to section 247.

If there is doubt about whether there is sufficient authority to recast otherwise legally effective transactions, then the notion of tax abuse already found in subsection 247(2) could be embellished as it is in the United Kingdom GAAR, about which more follows below.

There may be one catch, but seemingly even if noticed it has not deterred the UK or Australian legislatures or, for that matter, seemingly drawn any proverbial fire. This approach could be seen as amounting to a delegation of ambulatory legislative authority to a supranational body of countries’ tax and finance officials – which is what the OECD is (not the cathedral of tax policy that sometimes it seems to be, a body no less that lacks legal authority to do anything and is responsible to no democratic or other manifestation of governance in the exercise of its functions).  That poses an interesting public and constitutional law question.  However, the UK and Australian law seems to have avoided this, or in any event it may be avoidable by requiring changes to the Guidelines extant when first introduced into the law, to be adopted as law in a typical way, for example, as we do by regulation – requiring some evaluation of those changes in light of their context. Another relevant brake is the requirement that the transfer pricing law as it otherwise stands be applied consistently with the Guidelines – in effect, a statutory recognition of their significance considerably short of a surrender of supervening legislative authority.  This is the sort of thing, perhaps, that would have made determinations about the legality and significance of the Guidelines made courts in the well-known GlaxoSmithKline Inc. (2012 SCC 52; [2012] 3 SCR 3) and Cameco Corporation cases about the supremacy of that Act’s legislative statement of the “arm’s length standard” regardless of the Guidelines, more difficult to make.

There is one other jurisdiction, notable for Canada in particular, but for all countries in the circumstances.  It is the United States of America.  The US has an uneasy association with the OECD, even though the US “participates” or offers influential commentary on all international tax matters seized by the OECD.  And, in no way does the US tax law incorporate by reference OECD guidance of any kind; indeed, the compatibility of US law with that guidance might be acknowledged but not as something on which US law is build.  And, for good reasons.  As I have discussed elsewhere, the OECD Transfer Pricing Guidelines are US transfer pricing tax law.  The original late 1960s regulations are the core of the comparative methodological approach found in the original OECD 1979 Transfer Pricing Guidelines.  The two principal subsequent modifications of the US transfer pricing law, captured in detailed and transaction-oriented regulations, in the mid-1980s and the early 1990s, enshrined intangibles-oriented residual profit attribution via the “commensurate with income standard” and against that backdrop the significance of profit splits and the comparable profit method that is, except by name, the OECD’s “transactional net margin method.”

So, too, the US has effectively legislated the OECD transfer pricing guidance, by in fact making and continuing to influence it.

One might ask:  With three significant OECD countries with close ties to Canada having legislated the OECD Transfer Pricing Guidelines on an ambulatory basis, what is holding Canada back?

Maybe we have reached the point, with this awareness and recent judicial experiences, where the answer is no. No, we do not have to reinvent or recast or renovate or reform or any other “re”, section 247. We just have to listen to the legislative muse of other countries and signals from our courts about what to do, or what, adopting Lord Burrows’ outlook, courts could and indeed may want to do with suitable legislative authority. And, with suitable acknowledgment to the inspiration, we could copy-paste. As soon as tomorrow!

There is a caveat, alluded to in earlier comments.  It is that the anti-abuse aspect of section 247 would be conditioned to serve only the purpose of assisting normative transfer pricing analysis in circumstances where the actual transaction, motivated by tax avoidance or not, lacks a convenient market-based commercial analogue. This is essentially the role for paragraphs 247(2)(b) and (d) assigned by the Tax Court and Federal Court of Appeal in the Cameco Corporation case.  Should we be concerned that in so doing we will unleash the rampages of tax avoidance?  No. First, and explicitly if insecurity requires it, it would correspondingly be made clear that section 245, recalibrated to “abnormal” and “contrived” circumstances, still has over-arching application; that, after all, is the implication of taxpayers’ transactions that lies at the heart of “b-d” axis assessments in any event. In fact, not infrequently, the confusion or change in how transfer pricing cases rely or do not rely on the GAAR as well as section 247 and even whether assessing irregularities could be considered to have arisen if both are not pleaded.

The suggested approach would cede nothing in terms of the merits or ability to undertake a purposive tax avoidance-oriented assessment in a transfer pricing setting, but it would avoid the confusion of seeming to apply two different notions of tax avoidance - one evidently distilled, but only partly, from the other.  And it would relieve any additional insecurity that might have attached a particular distinction to the “almost GAAR” in paragraphs 247(2)(b) and (d) that excludes, there, the qualification on the application of the GAAR found in subsection 245(4), namely an abuse of the Act. It might be, or have been, thought that presumptively mis-reported income in a system that depends on self-enforced voluntary declarations of the correct income is implicitly “abusive” so as not, then, to require a further determination of abuse that subsection 245 requires of “avoidance transactions.”  That is plausible, but it also may miss the point that even in situations where income may be properly reported according to legal constructions adopted by taxpayers, questions about what is “abnormal” or “contrived” still may arise. The general application of section 245 according to standards that apply consistently for the Act generally, sacrifices nothing in the transfer pricing world and in fact would be sound systemically.

Hiding in plain sight? Validated, no less? Not a legislative adventure. In good company. The legislative significance and analytical direction, the statutory objectives and context, fully in view.  Rules do not have to be a litany of “rules” to be effective rules. A good idea to investigate? Yes.  A good idea to learn about the experiences and approaches of other countries facing the same issues as confront our transfer pricing regulation?  It is hard to see why not.




Why a Review?

The judicial experience with the GAAR reflects, certainly from the Crown’s perspective, similar considerations as with transfer pricing.


What Do Tax Authorities Want?  What Have They Been Denied?  What Are the Questions?

The questions asked at the outset of these comments, and it may be said the answers, are common to both the most controversial aspects of transfer pricing and the efficacy in practice of the GAAR.  This is not surprising given that GAAR cases are primarily factual despite often getting lost in legal inquiries befitting a “reference case.”  As with transfer pricing, so with GAAR that the principal complaint is that taxpayers have engaged in “unnatural acts” – “abnormal” or “contrived” transactional or organizational activity relative to their typical and ongoing circumstances, to secure tax advantages that viewed in this light are gossamer, ephemeral, “insubstantial”. And like its transfer pricing sibling, GAAR analysis decants all too quickly to assertions about “substance”, “economic substance”, the triumph of “form”, abandoning both rigorous legal analysis and any awareness that fundamentally cases are GAAR cases because of the facts, not the law.


The GAAR is Less (Much Less) Than Its Billing

In fact, despite over thirty years of introspection about the GAAR, it is probably fair to say the GAAR is not nay more than what Justice Estey spelled out as the test for evaluating the substantial compatibility of taxpayers’ planning and the proper application of tax law, in the Stubart case (Stubart Investments Limited v. The Queen, [1984] 1 S.C.R. 536, paying particular attention to the reasons of Justice Estey at pages 579 and 580).

Parsing the five-pronged “interpretative guidelines” (three, but two with subdivisions advanced by Justice Estey near the end of his reasons, one might be hard pressed to distinguish the GAAR as legislated as other than a distinction without a difference). And both the circumstances of Stubart, which inform Justice Estey’s meaning, and the rest of his analysis, essentially supply most of what has been worth saying about legislative or rule based anti-tax avoidance measures despite mountains of GAAR commentary and many GAAR judicial decisions.

The Crown’s complaint in Stubart was that the taxpayer had engaged in unnatural acts, pointless from a commercial point of view and essentially abnormal or contrived when measured with reference to the situation of third parties, to secure a tax outcome, namely de facto consolidation of the two affiliated corporations so as to use tax accounts that, in the event, the Act had already decided existed and but for the structural mitosis within the corporate group would have been used had there been a single corporate cell. In the complaint were allegations that the taxpayer had not fully effected its adopted forms, and so should fail for that reason, too. As the case progressed judicially, all manner of anti-avoidance doctrine was raised in the Crown’s assertion of a business purpose test, among them:  shame, incomplete transaction, and the denial of deductions considered to “unduly or artificially reduce the income” found in the “old” section 245 and even “older” section 137 of the Act.  In effect, the Crown thought that the taxpayers’ actions amounted to contrivance; they were abnormal; they were unnecessary – not in so many words, but that is the clear sentiment and superstructure on which the assessment was build.  The courts, and notably the Supreme Court, directed their judicial minds to the application of the Act evidently, reading between the lines without too much strain, according to these ways of expressing unreality.

The Supreme Court found no unnatural acts, and in so doing both disavowed a determinative “business purpose” test and recognized that in the “real world” tax is a cost that knowledgeable and responsible business parties seek to moderate if possible. Indeed, the Court referred to jurisprudence and particularly influential commentary by David Ward and Maurice Cullity, to establish that rational taxpayers might well, using my (not its) words, justify the impugned planning as quite necessary to reduce the cost of tax to a corporate group in a rational way – a way with many variations, it must be said, that became, and has been ruled on, by the tax authorities for decades. The GAAR takes a circuitous path also to say, that while there is no business purpose test in the law, business considerations need to be relevant to sustain otherwise possibly colourable events that yield tax outcomes.

What Justice Estey said in his six “interpretative guidelines”, after having canvassed the waterfront of anti-avoidance doctrine that generally had had difficulty upsetting legally effective arrangements undertaken by taxpayers in their own interest, is worth repeating.  Paraphrasing, his general anti-avoidance rule had the following contours:


  1. Where a transaction reveals no business purpose, tax attributes still may be denied if there is an “undue” or “artificial” reduction of income, or in other words, events out of step with a taxpayer’s usual circumstances that only achieve tax reduction belied by the Act as seen as a system, instead of a menu of choices to build the fiscal equivalent of a subway sandwich.
  2. Legally ineffective transactions cannot have the tax results they seek.
  3. “Sham” transactions, in the classical deceit oriented sense of “sham”, cannot have the tax results they seek.
  4. Otherwise legally effective transactions cannot have the tax results they seek if those results were manufactured or transmitted outside or beyond the pre-existing circumstances of a taxpayer according to the taxpayer’s condition before those transactions, which is the condition with reference to which the tax results should be evaluated, if the tax results are not consonant with restrictions evident by “the setting in the Act” of the particular provisions relied on to deliver them.
  5. Otherwise legally effective transactions cannot have the tax results they seek if the affected provisions in the Act “necessarily relate to an identified business function” of those provisions that a taxpayer has failed to show existed.
  6. Otherwise legally effective transactions cannot have the tax results they seek if they are unnatural with reference to the usual condition or conduct of the taxpayer, if, to deliberately use language from the United Kingdom GAAR discussed a bit later they are “abnormal” or “contrived” to achieve the tax results. Here, Justice Estey’s own words speak forcefully with the sorts of considerations animating my comments evident at least as undercurrents: the “‘object and spirit’ of the allowance or benefit provision is defeated by the procedures blatantly adopted by the taxpayer to synthesize a loss, delay or other tax saving device, although these actions may not attain the heights of ‘artificiality’ in s. 137 [artificial reduction of income].  This may be illustrated where the taxpayer, in order to qualify for an ‘allowance’ or ‘benefit’, takes steps which the terms of the allowance provisions of the Act may, when taken in isolation and read narrowly, be stretched to support.  However, when the allowance provision is read in the context of the whole statute, and with the ‘object and spirit’ and purpose of the allowance provision in mind, the accounting result produced by the taxpayer’s actions would not, by itself, avail him of the benefit of the allowance.” [Emphasis added]

Leaving aside linguistic considerations – which, to an unreasonable extent, have dominated GAAR jurisprudence and commentary ever since the GAAR’s introduction into the Act, one might be hard pressed to see any meaningful difference between Justice Estey’s “interpretative guidelines” and the GAAR as we “know” it, though how well we know it is debatable. Equally, we see little, if any, difference from the statements of the GAAR found in the Supreme Court’s Canada Trustco and Copthorne decisions, allowing for the inevitability that speaking as a court or a commentator, most find the need to “interpret” the GAAR in their own images.  Again, though, it is useful, as an interpretative heuristic, to ignore linguistic considerations and try to identify the fiscal proscription however it might be expressed in words.

In short, Justice Estey’s “guidelines” and the GAAR, and everything in between, with the occasional blip, say or convey the sense that “abnormal” or “contrived” events in relation to a taxpayer’s usual condition, which have tax results out of step with the taxpayer’s condition and would not have arisen but for a change by the taxpayer in that condition to obtain them, justifiably may be denied.  It is not more complicated than this. What may be lacking, however, is explicit statutory direction that infuses opaque references to purpose and abuse with legislative direction to courts, and in fact to everybody else, that “abnormality” and “contrivance” – with the connotations those words have – are compelling legal grounds for denying the tax effects of otherwise effective transactions. Without resort to “sham”. Without incanting TCP. Without asserting loose notions of economic or business reality, even as evidence about a taxpayer’s conduct loosely fitting the economic or business molds still will be relevant in a fact-based evidentiary analysis – a legal analysis(!).  Without uttering the word “substance” of invoking the tiresome “form over substance” debate favoured by many – many one suspects who have never read or fully comprehended the vaulted Duke of Westminster case, considering the speeches (now reasons) of all the Law Lords and not the single sentence of Lord Tomlin that anchors the common veneration of legal formality.

How might this be done?  Is this answer, too, hiding in plain sight?


Legislative Recalibration – Directionally, Without Legislative Clutter

The United Kingdom’s GAAR, much more recent than Canada’s, offers useful legislative as well as directional insight.  To appreciate its elegance, and also its undercarriage informed by much House of Lords jurisprudence that also has influenced Canada’s conception of impugnable tax avoidance, three sources – cogent and well-grounded in pertinent comparative legal analysis -- beyond any others are worth consulting before we, in Canada do anything.  They are: The Aaronson Committee Report (GAAR STUDY, A study to consider whether a general anti-avoidance rule should be introduced into the UK tax system REPORT BY GRAHAM AARONSON QC, 11 NOVEMBER 2011) which considers the desirability of introducing a statutory GAAR in UK tax law despite the rich judicial experience with tax avoidance, and the incisive and for that Report no doubt leading influence by one of the Committee’s members, Professor Judith Freedman in her profoundly important paper:  Defining Taxpayer Responsibility:  In Support of a General Anti-Avoidance Principle (University of Oxford Faculty of Law Legal Studies Research Paper Series Working Paper No 14/2006 April 2006; [2004] British Tax Review 332–357).  These, together with pertinent Her Majesty’s Revenue and Customs guidance on the UK GAAR and a House of Commons Briefing Paper by Antony Seely, Number 6265, 7 April 2021 Tax avoidance: a General Anti-Abuse Rule, are essential not merely helpful, and possibly the best contemporary guidance in the design of any statutory general anti-avoidance rule and, I would say, the recalibration of ours.


Key Elements of the United Kingdom GAAR

The essence of our long doctrinal and statutory history of addressing untoward tax avoidance is to distinguish between achieving economically equivalent outcomes via a selection among different organizational and transactional choices depending on a taxpayer’s own circumstances with correspondingly different tax outcomes, on the one hand, and events that, as a close friend and colleague reminds me with reference to a phrase long embedded in Canadian law, merely reflect a “patina of compliance” with the tax law taking account of how those events are configured.  That is, the goal is to detect when in relation to a taxpayer’s own condition legal arrangements to achieve tax outcomes are mere fiscal theatre. Importantly, this tests the applicability of those arrangements themselves according to the significance of the intrinsic legal elements of and found within an understanding of the organizational forms and transactional expressions on which taxpayers rely for the tax outcomes they defend.

The linchpin of the UK GAAR is a statutory definition of “abusive” that is descriptive, not prescriptive, but that not only authorizes but directs those applying the UK tax law in any particular context to precisely the reservations about tax outcomes that implicitly animate not only the Canadian GAAR but its judicial and statutory antecedents reaching far back to the earliest history of modern income taxation in Canada – in fact back to the earliest stages of the Income War Tax Act 1917 (as Colin Campbell and Robert Raizenne chronicle in a soon to be released paper that they presented at a Cambridge University tax history colloquium in the summer of 2020). The definition of “abusive” specifically targets “abnormal” and “contrived” circumstances, situations out of the normal rhythm of a taxpayer’s condition and circumstances, which fuel tax outcomes that are not consistent with that condition or circumstances, and among other things exploit perceived gaps in the statutory law which should not be seen as detracting from the law’s evident substantive significance, taking explicit account of pertinent economic factors.

Here is the definition of “abusive” in subsection 207(2) of the Finance Act 2013; (the emphasis is mine – and it is meant to convey an important directional message found in the provision):


“(2) Tax arrangements [defined in subsection 207(1) in a manner consistent with how an “avoidance transaction is conceived in the Canadian GAAR] are "abusive" if they are arrangements the entering into or carrying out of which cannot reasonably be regarded as a reasonable course of action in relation to the relevant tax provisions, having regard to all the circumstances including—

(a) whether the substantive results of the arrangements are consistent with any principles on which those provisions are based (whether express or implied) and the policy objectives of those provisions,

(b) whether the means of achieving those results involves one or more contrived or abnormal steps, and

(c) whether the arrangements are intended to exploit any shortcomings in those provisions.

(3) Where the tax arrangements form part of any other arrangements regard must also be had to those other arrangements.

(4) Each of the following is an example of something which might indicate that tax arrangements are abusive—

(a) the arrangements result in an amount of income, profits or gains for tax purposes that is significantly less than the amount for economic purposes,

(b) the arrangements result in deductions or losses of an amount for tax purposes that is significantly greater than the amount for economic purposes, and

(c) the arrangements result in a claim for the repayment or crediting of tax (including foreign tax) that has not been, and is unlikely to be, paid,

but in each case only if it is reasonable to assume that such a result was not the anticipated result when the relevant tax provisions were enacted.”


In other words, whatever else may be said about the transactions and related events, they will not have their intended tax effects if they are abnormal or contrived, or if they are in other words unnatural or unnecessary with reference to a taxpayer’s ordinary circumstances evaluated according to a reasonableness, i.e., objective, standard, if they produce tax outcomes that are not consistent with their economic effects or implications, if they seek to thread holes in statutory needles without reference to the whole cloth stitched by those needles.

A tax law – induced restatement of a taxpayer’s circumstances including otherwise legally effective arrangements, whether via the GAAR or the “b-d axis” of section 247, is and is meant to be an extreme event.  Even in the existing GAAR jurisprudence, there is little doubt about this.  It is an event that is so devoid of meaning according to the rhythm of a taxpayer’s usual situation in relation to the fiscal objectives served by targeted tax provisions the event unlocks, that regarding the event as effective to keep the benefits of those tax provisions is problematic.  In effect, it is an event that is not meaningful in a taxpayer’s actual universe, and even if short of “window dressing” demonstrably according to an evidence based, i.e., a legal, analysis is not meant to be meaningful in the sense of being relied on or meant to be relied on as an event meaningfully transformative, in fact, of a taxpayer’s circumstances – and it can be shown by relevant evidence developed in typical law-based ways that this is the case.  Nothing really changes but a tax bill. It has the connotation of being legally effective only in the narrowest of ways with the fiscal objectives and the Act’s exposition of them in mind, to secure a tax outcome that the normal rhythm of the taxpayer’s life would not offer but without actually changing that rhythm.  Indeed, it would not be hard to see many if not most of the GAAR cases this way, and that sometimes tortured analysis has been adopted to decide them on grounds that belie the kind and degree of theoretical introspection they often engage, for which, though, that introspection is (merely) the articulation.

The hard question is how to communicate that in the Act in a way that enlivens it with legal authority, as a necessary instruction to adjudicators to consider and as a clear (enough) guardrail for taxpayers and their advisers.

“Abnormal” and “contrived” are powerful words. They are used in the UK context clearly with reference to the objective circumstances of a particular taxpayer in relation to an evaluation of those circumstances in relation to its ordinary condition.  They are not theoretical abstractions – legal or linguistic; the full definition of “abusive” accompanied by a statutory conception of “transaction” not unlike our own in meaning, makes that evident, at least evident enough for purpose.  In fact, they, as embedded notions, actually infuse our pre- and post-GAAR jurisprudence as floating spirits though not in a consistent or systemically coherent way. They convey a sense of context – the taxpayer’s condition before embarking on events that allow it to benefit from salutary tax outcomes.  They convey gravity and judgment that is focused on events themselves and the manner in which they involve departures from how a taxpayer would or would need to conduct itself if harvesting tax benefits were not top of mind.  They direct the pertinent lines of an evidentiary inquiry in order to make the necessary judgments about a taxpayer’s entitlement to tax benefits.  They have the tone of a scolding parent or school master.  They avoid conflating fraudulent conduct, which is not what the GAAR is about, and the civil object of GAAR’s attention, and further do not depend on events fitting any of the hoary anti-avoidance doctrine comprising, for example, incomplete transactions; they apply as reference points in a GAAR analysis for transactions fully effective in law not merely as formal depictions on a whiteboard but according to intrinsic legal requirements of the chosen organizational forms and transactional configurations. They focus on the fiscal significance of those forms and configurations because that is what concerns the tax law.  Even when taxpayers “get it right” for other purposes, it still may not be right for tax purposes if to get the tax prize, contortions are required, contortions that not surprisingly may be relieved somehow once the tax prize has been claimed. They do not invoke untethered notions of “substance” or “reality”, relative as those terms are uninformed by any normative systemic meaning or definition and not oriented to a particular taxpayer’s own situation in its moment.  They accordingly avoid the kinds of implicit judgments and theoretical introspection that pepper typical anti-avoidance - oriented transfer pricing and GAAR discussions in search, among other things, of a fiscally cosmic instruction about what is “an abuse having regard to … [the Act and connected or relevant legal instruments] read as a whole.”  In short, they offer real statutory guidance about what the GAAR as a tax system’s extreme remedy is really “all about,” guidance that that notably consonant with how most GAAR cases and transfer pricing cases with tax avoidance overtones beyond mere income measurement, i.e., cases invoking the “b-d” axis in subsection 247(2) have actually proceeded, if not directly, certainly bubbling in the shallow subsurface.  They speak as law telling courts what to do and taxpayers and their advisers what to consider in their conduct, and in their giving and taking of advice.

Some might observe that this conception of abuse is far from prescriptive and is not much more than what, commonly, the Crown asserts underlies our GAAR. That may well be right.  But it ignores the significance of an explicit statutory statement of abuse or fiscal incongruity, both as a legal matter and in practice.

First, it is unlikely that any attempt to formulate a catalogue or menu of specific impugnable tax-directed events including but not limited to typical transactions, can ever be satisfactory or other than yet another opportunity to navigate the provisions of the Act including that catalogue for defensible outcomes. A satisfactory, effective reconstruction of the GAAR has to targe its essence, the vista of the forest not the random array of trees into which one is likely to bump by starting in the middle of the forest.

Second, a list is not the point.  What is the point, is providing direction in the Act, with the authority to follow it that the enactment of that direction insists, about what specifically makes events (noting that the definition of “transaction” in both sections 245 and 247 is much more expansive than mere contractual transactions and organizational forms) incompatible systemically with the Act.  The use of words like “abnormal”, and “contrivance” convey the tone and implications of judgments to be made, and the permitted means to make them, that defy clever or formalistic threading of legal needles. Enabling them by reference to an objective standard that also anticipates the particular condition and circumstances of particular taxpayers in the moment, and bears witness to the significance of economic compatibility, convey volumes and supply what too many might think is missing statutory authority in the present GAAR formulation.

This is no insignificant point.  A UK colleague and friend, a notable tax expert who is well informed about the UK GAAR offered insight into challenges faced by and responses framed in general anti-avoidance rules in relation to the UK GAAR, and spoke to students in one of my classes last winter (2021) about the UK GAAR.  Without meaning to put words our guest’s mouth, I have the strong impression from that conversation that an explicit statement in the tax legislation of what courts are obliged to consider in interpreting and applying the Act is what empowers fact-based purpose oriented legal construction, shielding it from both the criticism and possibility disability of being cast as insupportable judicial activism beyond the limits of the law.  I think it might be said, accordingly, that in decisions that may seem to be perverse from a government’s perspective, what has been lacking is not judicial will, but judicial tools.  There is a sense of this in Lord (as he now is, from June 2020) Burrows’ Hamlyn Lecture to which I referred earlier. Lord Burrows essentially said that it is permissible, even expected, to interpret statutory law in situations that might not have been directly foreseen when the law was enacted, without needing to resort to the illusory quest for self-speaking statutory intention or the intention of Parliament, even to incorporate “changes in scientific thinking or societal attitudes” – provided that the direction to do so can be found in the law without engaging in what judges are not empowered to do, namely to “update” rather than interpret legislation.  In interpreting legislation however Lord Burrows notes that “the judges may apply the contemporary meaning of the statutory words”.  In fact, judges must do so, where the law so obliges them, to avoid being tethered to the perversion of the guiding hand of statutory intention: “…  [T]he right question is what is the best interpretation now of the Act.  [J]udges are interpreting a legal rule laid down in the public interest, which is not the same as interpreting interpersonal communications in everyday life. The judges must look for the best interpretation now of the legal rule laid down by … statutory words, with reference to their context and purpose … [as may be assisted by recourse to various sources]. …A serious objection to any reference to legislative intention is that it is advocating an approach that favours the law’s ossification by inappropriately freeing the law in the past.  We would not accept such an approach for the common law and there is no good reason why we should regard it as acceptable when interpreting.”

The point, as I was impressed to think by listening to our class guest, is that it is not enough that willing judges may be amenable to coming to reasonable conclusions; they need to be enabled and instructed to do it, by the law that is before them. That makes the UK approach, and in particular the aspects of the UK definition of “abusive” I have highlighted, very important when we think of alleged shortcomings or surprises in our GAAR jurisprudence. As Lord Burrows intimated above, and said quite explicitly in his paper Challenges for Private Law in the 21st Century, it is the law that determines outcomes, not the social science about law that occupies commentators’ advocacy about what in theory the law ought to be.

A similar sentiment can be found in Justice Stratas’ reasons in the Kattenburg judicial review appeal. That case was a judicial review of ultimately unsuccessful prospective intervenors’ motions to intervene in a food labelling case largely for extrinsic political or like reason, as perceived by the Court want to advocate beyond the limits of the applicable law.  Justice Stratas said:


“[44]  Some courts admit into an appeal just about anyone who wants to offer any views, even political or ideological ones oblivious to the legal doctrine that binds the Court: see observations in Teksavvy Solutions at para. 11; Ishaq at paras. 25-27; Atlas Tube at paras. 4-12. And sometimes upwards of twenty or more special interest or political advocacy groups are allowed to pile in, giving appeals the appearance of a sprawling Parliamentary committee hearing or an open-line radio show, and often a one-sided one at that: Gitxaala Nation at paras. 21-24; Zaric at para. 12; Teksavvy Solutions at para. 11; Atlas Tube at para. 12. So much of their loose policy talk, untethered to proven facts and settled doctrine, can seep into reasons for judgment, leading to inaccuracies with real-life consequences: see examples provided in Brown v. Canada (Citizenship and Immigration), 2020 FCA 130 at paras. 156-159, citing Teksavvy Solutions at para. 22, both referring to R. v. Bird, 2019 SCC 7, [2019] 1 S.C.R. 409 and Canada (Public Safety and Emergency Preparedness) v. Chhina, 2019 SCC 29, 433 D.L.R. (4th) 381.


[45]  As for judges, some give the impression that they decide cases based on their own personal preferences, politics and ideologies, whether they be liberal, conservative or whatever. Increasingly, they wander into the public square and give virtue signalling and populism a go. They write op-eds, deliver speeches and give interviews, extolling constitutional rights as absolutes that can never be outweighed by pressing public interest concerns and embracing people, groups and causes that line up with their personal view of what is “just”, “right” and “fair”. They do these things even though cases are under reserve and other cases are coming to them.


[46]  They should not act in this way. They should stay in their proper place. Their place is not in the public square amongst the partisans and the politicians, participating in the fray. Instead, their place is inside their courthouses, hearing each side, weighing and assessing the admissible evidence and discerning and applying the relevant legal doctrine, all in a rational, open-minded and neutral way, both in appearance and actual fact.”


It is important then that the Act speak not only philosophically about what it means to accomplish, but explicitly and specifically about how to accomplish it, in a way that conveys not only legislative authority, but legislative perspicacity, awareness, contextual “reality”, and attitude. This is what the UK law does, untested though it may be in light of its relative newness.  But the reasons for being cast as it is, with the benefit of having considered Canadian, Australian and US approaches to deterring tax avoidance, including by statutory means, are evident in how, in a suitable evergreen and ambulatory way, the UK provision is expressed as it is.

It is worth repeating.  Hiding in plain sight.  Validated, no less.  Not a legislative adventure. In good company.  The legislative significance and analytical direction, the statutory objectives and context, fully in view.  Rules do not have to be a litany of “rules” to be effective rules.


Recalibrating Sections 245 and 247 – If at All, Carefully and With Focus

What is contemplated by the consultations on transfer pricing and the GAAR is the same.   The catalyst for the review is the same.  The underlying relevant experience with the two “rules” is the same.  The Crown’s formulations of anti-avoidance criticism of taxpayers’ arrangements are the same. And there be any, the legislative shortcomings – the absence of a specific enough legislative instruction and tools in the Act to enable the statutory relevance of notions that creep into transfer pricing and GAAR cases as aspirational political economy, are the same.

No exotic retrofit of section 245 or section 247 required. Economical prefabricated changes are readily at hand.  They are modest, but directionally consistent with Canadian tax law and its anti-avoidance undercurrents; they are sensitive to GAAR’s origins and history, to the objectives and concerns of transfer pricing analysis as reinvigorated by the OECD – led reactions to base erosion, to the thematic arguments undeterred by doctrine of the Crown in GAAR and transfer pricing cases, and to sensible and reasonable perceptions of out of the ordinary conduct by taxpayers abetted by their advisers.

Are solutions indeed hiding in plain sight?  At least, are the experiences of others and their notable statutory economy, communicating as they do, exactly what is the problem and precisely what are the law-based instructions and authority to address it, illuminating and instructive?

Hiding in plain sight?  Perhaps.  A good idea to investigate?  Yes.  A good idea to learn about the experiences and approaches of other countries facing the same issues as confront our GAAR?  It is hard to see why not.


 - Professor Scott Wilkie (Distinguished Professor of Practice, Osgoode Hall Law School)

Comment on “Hiding in Plain Sight? Changes to “Transfer Pricing” and the GAAR

  1. Excellent read and post, Professor Wilkie!

    I agree with your conclusion, following our Winter 2021 class with our class guest (an expert in British tax law), that the Court is wanting, or rather needing, tools and explicit instructions to work with, versus their decisions indicating some sort of an unwillingness to deal with (in a way satisfactory to the perspective of the CRA) transactions where a taxpayer may be seen to engage in abnormal or unintended leveraging of otherwise completely lawful and statutory legal constructs.

    Thus, it seems beneficial to have such explicitly spelled out (like the s. 207 abuse definition re UK). Much the same as my thoughts on "preventing" "Treaty Shopping"… spell it out. Define "Treaty Shopping", then make explicit reference to its not being allowed (we now have step 2 but no real step 1, in my opinion, with BEPS Action 6 RE the MLI).

    A Court can act on law - a Court can’t act on that which is not law (such as commentary at the OECD, as much as it may be helpful in colouring those decisions which touch on that which actually is law). So legislators must make it explicit in the law!... Kind of goes back to basic civics, doesn't it?