Canada's Supreme Court sets the standards for permissive tax avoidance.

Introduction

On October 19, 2005, the Supreme Court of Canada released its first judgments on the general anti-avoidance rule (GAAR) under section 245 of the Income Tax Act (Canada) (Act). The judgments, which emanate from a unanimous court, were rendered in the companion cases of Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54 (Trustco), and Mathew, et al. v. Canada, 2005 SCC 55 (formerly known as Kaulius, et al. v. The Queen) (Mathew). These judgments were much anticipated by tax executives and their advisers since more than 20 decisions from lower courts have been rendered since GAAR was introduced in 1988.

By way of background, the GAAR was introduced after the predecessor of the Canada Revenue Agency (CRA) lost an important battle before the Supreme Court in the tax avoidance area 21 years ago. In Stubart Investments Ltd. v. The Queen, [1984] 1 S.C.R. 536 (Stubart), the Supreme Court held that the Act did not contain an inherent or implicit "business purpose test." With Stubart, the so-called Duke of Westminster principle was rejuvenated and taxpayers continued to be "entitled to order their affairs to pay the least amount of tax allowed by law." Generally speaking, this meant that the taxpayer could minimize or avoid tax by bringing itself within (or keeping itself outside) the specific wording of the Act. This generally led Canadian courts, including the Supreme Court, to favor a "literal interpretation or application" of the Act.

In 1988, the GAAR became part of the Act law and "superimposed a prohibition on abusive tax avoidance, with the effect that the literal application of provisions of the Act may be seen as abusive in light of their context and purpose." (1) In the Supreme Court's view, its task in the Trustco and Mathew appeals was "to unite these two approaches in a framework that reflects the intention of Parliament in enacting the GAAR and achieves consistent, predictable and fair results." (2)

Trustco and Mathew are good reads. A first review, however, will leave the tax professional wanting more. While the judgments contain certain unequivocal and helpful diktats, ultimately they fail to define precisely what constitutes permissible tax avoidance. It is only after several readings and discussions that the tax executive will have a good appreciation of the effect of the Trustco and Matthew decisions. This article is intended to facilitate such an appreciation.

Statutory Context

The provisions of the GAAR, as they applied to the Trustco and Matthew decisions, (3) read in part, as follows [paragraphing modified]:

245. (1) [Definitions] In this section, "tax benefit" means a reduction, avoidance or deferral of tax or other amount payable under this Act or an increase in a refund of tax or other amount under this Act; [and] "transaction" includes an arrangement or event.

(2) [General Anti-Avoidance Provision.] Where a transaction is an avoidance transaction, the tax consequences to a person shall be determined as is reasonable in the circumstances in order to deny a tax benefit that, but for this section, would result, directly or indirectly, from that transaction or from a series of transactions that includes that transaction.

(3) [Avoidance Transaction.] An avoidance transaction means any transaction (a) that, but for this section, would result, directly or indirectly, in a tax benefit, unless the transaction may reasonably be considered to have been undertaken or arranged primarily for bona fide purposes other than to obtain the tax benefit; or (b) that is part of a series of transactions, which series, but for this section, would result, directly or indirectly, in a tax benefit, unless the transaction may reasonably be considered to have been undertaken or arranged primarily for bona fide purposes other than to obtain the tax benefit.

(4) [Where subsection (2) does not apply.] For greater certainty, subsection (2) does not apply to a transaction where it may reasonably be considered that the transaction would not result directly or indirectly in a misuse of the provisions of this Act or an abuse having regard to the provisions of this Act, other than this section, read as a whole.

(5) [Determination of Tax Consequences.] Without restricting the generality of subsection (2), (a) any deduction in computing income, taxable income, taxable income earned in Canada or tax payable or any part thereof may be allowed or disallowed in whole or in part, (b) any such deduction, any income, loss or other amount or part thereof may be allocated to any person, (c) the nature of any payment or other amount may be recharacterized, and (d) the tax effects that would otherwise result from the application of other provisions of this Act may be ignored, in determining the tax consequences to a person as is reasonable in the circumstances in order to deny a tax benefit that would, but for this section, result, directly or indirectly, from an avoidance transaction.

248. (10) [Series of transactions] For the purposes of this Act, where there is a reference to a series of transactions or events, the series shall be deemed to include any related transactions or events completed in contemplation of the series.

The Transactions under Scrutiny

In Trustco, the transaction resembled a sale-leaseback transaction. Canada Trustco was a mortgage lender and lessor, and as part of its business, obtained revenues from leased assets. In 1996, it paid $120 million to acquire a fleet of trailers from a U.S.-based company, TLI. To make the purchase, Canada Trustco used $20 million of its own money and borrowed the rest from Royal Bank. Immediately after purchasing the trailers, Canada Trustco leased them to a U.K. company, MAIL, which in turn sub-leased them back to TLI, the original owner. TLI then pre-paid the $120 million due to MAIL under the sub-lease, and this amount was placed on deposit for purposes of making MAIL's lease payments to Canada Trustco. Canada Trustco then assigned to Royal Bank the rent payments owed to it by MAIL to retire the loan and it was agreed that Royal Bank would have no further recourse against Canada Trustco. Canada Trustco went on to claim capital cost allowance (CCA) on the trailers totaling $98 million, which exceeded its lease income on the deal and could therefore shelter other income.

In Mathew, the facts were simpler. In 1991, the Standard Trust Company (STC) was insolvent and a liquidator was appointed. The liquidator transferred a portfolio of STC's non-performing mortgages into a general partnership called STIL II. The mortgages had significant accrued but unrealized losses at the time of the transfer. Because STC did not deal at arm's length with STIL II at the time of the transfer (and for the 30 days following the transfer), former subsection 18(13) denied recognition of the loss on the transfer and preserved it in the hands of STIL II. The liquidator then sought purchasers for the partnership and, after 8 months, found them in the taxpayers. The taxpayers acquired an interest in a second general partnership, SRMP, that had been established to hold 76 percent of STIL II. Thereafter, in 1993, STIL II realized on the mortgages and recorded a business loss of $52 million, $40 million of which was allocated its majority partner (SRMP). The taxpayers, as partners in SRMP, then claimed their respective shares of SRMP's losses, which amounted to $10 million, and sought to apply it against their own incomes.

Procedural History

The Minister of National Revenue challenged the tax result in each case under the GAAR (and some other arguments) and reassessed accordingly. Practically speaking, this meant incremental tax and (non-deductible) interest in both cases. In Trustco, the court of first instance, the Tax Court of Canada, found in favor of the taxpayer on the basis that there was no "misuse or abuse" under subsection 245(4), and the Federal Court of Appeal dismissed the Minister's appeal. In Mathew, however, both the Tax Court and Federal Court of Appeal upheld the GAAR assessment on the grounds that the subject transaction constituted a "misuse or abuse" for purposes of the same provision.

Before the Supreme Court, the Minister's argued in Trustco that the transactions in issue were completely circular, involved no economic risk for the taxpayer, and were designed solely to achieve a manufactured CCA deduction. As such, they offended the general policy that only money actually spent to acquire depreciable property should be respected for purposes of the CCA rules. In Mathew, the Minister's primary argument was that arm's-length taxpayers ought not be able to acquire tax losses of others on the ground this was contrary to the general policy against loss trading in the Act. Also, the argument before the Supreme Court involved whether the preferred approach developed to date by the lower courts was the correct one. In particular, in OSFC Holdings Ltd. v. Canada, [2002] 2 F.C. 288, 2001 FCA 260, a case involving essentially the same transactions as in Mathew, two of the three justices of the Federal Court of Appeal interpreted the phrase "misuse or abuse" in subsection 245(4) disjointedly and determined that it mandated a two-step enquiry to determine whether it applied. According to that approach, the first step was to determine whether there was a "misuse" of the particular provisions of the Act that were relied upon to achieve the tax benefit. The second enquiry was whether there was an "abuse" of any policy of the Act when read as a whole.

The Supreme Court Lays Out Three-Step Analytical Framework for GAAR

According to the Supreme Court, the application of the GAAR involves three requirements, all of which must be present before the GAAR can apply.

Step 1: Is there a tax benefit?

The expression "tax benefit" is widely defined in subsection 245(1) and its existence will depend on the particular facts of the case. In some cases, determining whether a tax benefit exists will require the comparison between alternative tax treatments. Practically speaking, this first step analysis will rarely be contentious.

Step 2: Is there an avoidance transaction?

Given the wide definition of "avoidance transaction" in subsection 245(3) and "series of transactions" at subsection 248(10), and the Supreme Court's adoption of the U.K. jurisprudence on the meaning of series, the preliminary requirement of having a "transaction" will often be easily met. Regarding the avoidance requirement, the subject transaction will not fall under the GAAR under subsection 245(3) if the transaction "may reasonably be considered to have been undertaken or arranged primarily for bona fide purposes other than to obtain the tax benefit." Interestingly, for the Supreme Court, the majority of tax benefits claimed by taxpayers will be immune as a result of subsection 245(3) since "the GAAR was enacted as a provision of last resort to address abusive tax avoidance, ... [and] was not intended to introduce uncertainty in tax planning." (5)

According to the Supreme Court, the determination whether there is an avoidance transaction requires an "objective assessment of the relative importance of the driving forces of the transaction." (6) Further, this enquiry requires the examination of "the relationships between the parties and the actual transactions that were executed between them." (7) In other words, whether there is an avoidance transaction is not to be determined by reference to a normative transaction (i.e., an alternate transaction with a higher tax cost). Instead, the purpose of the subject transaction will be determined based on the facts (and evidence) surrounding the transaction. If this factual enquiry leads to concluding that there are both tax and non-tax purposes, then it will become necessary to determine which is the primary one. Lastly, the Supreme Court emphasized that the expression "non-tax purpose" had a broader scope than "business purpose" (e.g., it can include family or investment purposes).

The issue whether there is an avoidance transaction should not be overlooked by tax executives. Based on the Trustco and Mathew decisions, this enquiry is aimed at the facts surrounding the transaction, and tax executives (and their counsel) are in the best position to assemble those facts and collect the requisite evidence contemporaneously with the transaction. The reward can be gratifying since the absence of an avoidance transition will be a complete answer to the GAAR.

Step 3: Is there abusive tax avoidance?

This third step has been the most contentious one and is perhaps the most difficult to apply, though the Supreme Court laid out an analytical framework to answer this difficult question. First, the Supreme Court clarified that the phrase "misuse or abuse" under subsection 245(4) ultimately calls for a single test (i.e., whether there is "abusive tax avoidance"). In doing so, the Supreme Court rejected the two-step approach under subsection 245(4) that was originated in the OSFC case. In the Supreme Court's view, whether there is abusive tax avoidance cannot be determined by searching for an overriding policy on the Act; rather the determination must be "anchored in a textual, contextual and purposive interpretation of the specific provisions that are relied upon." (8) In other words, what the Supreme Court is saying is the interpretation must go through "concentric circles" (i.e., text [right arrow] context [right arrow] purpose) or "from the bottom to the top" and not the reverse. To justify its approach, the Court referred to the "overall policy of Parliament that tax law be certain, predictable and fair, so that taxpayers can intelligently order their affairs." (9) In the Supreme Court's view, adopting any other approach would frustrate this paramount policy. In this regard, consider the following passages from the Court's decision:

44 The heart of the analysis under s. 245(4) lies in a contextual and purposive interpretation of the provisions of the Act that are relied on by the taxpayer, and the application of the properly interpreted provisions to the facts of a given case. The first task is to interpret the provisions giving rise to the tax benefit to determine their object, spirit and purpose. The next task is to determine whether the transaction falls within or frustrates that purpose. The overall inquiry thus involves a mixed question of fact and law. The textual, contextual and purposive interpretation of specific provisions of the Income Tax Act is essentially a question of law but the application of these provisions to the facts of a case is necessarily fact-intensive.

45 This analysis will lead to a finding of abusive tax avoidance when a taxpayer relies on specific provisions of the Income Tax Act in order to achieve an outcome that those provisions seek to prevent. As well, abusive tax avoidance will occur when a transaction defeats the underlying rationale of the provisions that are relied upon. An abuse may also result from an arrangement that circumvents the application of certain provisions, such as specific anti-avoidance rules, in a manner that frustrates or defeats the object, spirit or purpose of those provisions. By contrast, abuse is not established where it is reasonable to conclude that an avoidance transaction under s. 245(3) was within the object, spirit or purpose of the provisions that confer the tax benefit....

55 In summary, s. 245(4) imposes a two-part inquiry. The first step is to determine the object, spirit or purpose of the provisions of the Income Tax Act that are relied on for the tax benefit, having regard to the scheme of the Act, the relevant provisions and permissible extrinsic aids. The second step is to examine the factual context of a case in order to determine whether the avoidance transaction defeated or frustrated the object, spirit or purpose of the provisions in issue.

What the Supreme Court has done is to require one to look beyond the mere text or context of the subject provisions when there is a GAAR enquiry. This means that it is now permissible when considering the GAAR to search for the purpose and, in doing so, to rely on extrinsic materials, especially the Explanatory Notes of the Department of Finance. Traditionally, the reliance on such materials was not authorized to look beyond the words of the tax statutes in the absence of ambiguity. (10)

Taxpayers can nonetheless find solace in the fact that the Court stated unambiguously that "GAAR can only be applied to deny a tax benefit when the abusive nature of the transaction is clear." (11) In the same vein, the Court ruled that the burden of showing that an avoidance transaction is an abusive one will fall on the Minister while the burden on the other issues test will be on the taxpayer, as is the case generally in tax cases.

How Did the Supreme Court Apply its Own Principles?

Step 1: Is there a tax benefit? In both the Mathew and Trustco cases, the Supreme Court found that the first requirement was easily met and effectively confirmed the extremely low threshold established by lower courts for the existence of a "tax benefit." It noted that the magnitude of the "tax benefit" was irrelevant and that the sole question was "whether the taxpayer reduced, avoided or deferred tax payable under the Act."

Step 2: Is there an avoidance transaction? As with the first requirement, the Supreme Court in both cases had little difficulty concluding the second requirement was met on the facts. Interestingly, while the taxpayer in Trustco had specifically argued that this requirement was not met, the Supreme Court did not even address this point in its reasons, concluding simply that "the only issue" was whether the third requirement, discussed below, was met.

Step 3: Is there abusive tax avoidance? Applying these tests to the two cases, the Supreme Court upheld the lower courts' findings that the transactions in Trustco were not abusive, while those in the Mathew case were. In particular, the Supreme Court held that the facts in Trustco did not demonstrate abusive tax avoidance because the scheme of the CCA rules in the Act was premised upon "legal cost" (rather than "real cost" or "economic cost") and that is precisely what was used to determine the taxpayer's CCA entitlement in this case. Thus, applying its "textual, contextual and purposive approach" to statutory interpretation, the Court held:

   Textually, the CCA provisions use "cost" in the
   well-established sense of the amount paid to acquire
   the assets. Contextually, other provisions of
   the Act support this interpretation. Finally, the
   purpose of the CCA provisions of the Act, as applied
   to sale-leaseback transactions, was, as found by the
   Tax Court judge, to permit deduction of CCA based
   on the cost of the assets acquired. This purpose
   emerges clearly from the scheme of the CCA provisions
   within the Act as a whole. (12)

Accordingly, the Supreme Court upheld the lower court judgment and refused to apply the GAAR to disallow any portion of the full CCA claimed. By its decision in Trustco, the Supreme Court rejected (once again) the notion of economic substance not grounded in the relevant provisions of the Act, as a basis to find abusive tax avoidance.

In contrast, the Court held in Mathew that the clear purpose of former subsection 18(13) was to prevent a taxpayer who is in the business of lending money from claiming a loss upon a non-arm's-length transfer of a mortgage, and that this loss was preserved in the hands of the transferee only "because of its special relationship with the transferee partnership." Accordingly, to allow a new, arm's-length partner to buy into the transferee partnership and benefit from these losses "would violate the fundamental premise underlying subsection 18(13) that the loss is preserved because it essentially remains in the transferor's control." Therefore, to allow the new partners to benefit from the loss "would contradict the main purpose of s. 18(13) and the premise on which it operates." As a result, the Supreme Court upheld the lower court's determination that GAAR applied to deny the loss deduction in the hands of the affected taxpayers.

Understanding the Ramifications of the Two Cases

The judgments reveal more of their meaning when one analyzes what the Supreme Court did and how it arrived at its conclusions. Similarly, reviewing concrete examples or actual transactions is also helpful in understanding the true import of the two decisions.

First, consider the 2002 decision in Imperial Oil Limited v. The Queen, 2004 DTC 6044 (F.C.A.), confirming 2002 DTC 1954 (T.C.C.) (Imperial Oil). The facts of this case were quite straightforward. The taxpayer was a Canadian corporation and evidently had some excess cash toward the end of its 1993 tax year, which ended on December 31, 1993. Before its year-end, the taxpayer loaned $500M to two entities (having different year-ends) which were foreign subsidiaries of two Canadian banks. The loans were fully guaranteed by the banks in question. The loans were made at what were admitted to be "commercial rates of interest." The term of the loans were for 33 to 35 days, so the $500M, plus interest, was repaid to the taxpayer in early 1994.

The case involved the application of large corporations tax (LCT) under Part 1.3 of the Act. Roughly described, this tax applies to the capital of the corporation (i.e., the right-hand side of the corporation's balance sheet) while providing taxpayers with an "investment allowance." The taxpayer included the amounts of the loans to the two foreign entities in its "investment allowance" under paragraph 181.2(4)(b) of the Act. This provision includes in a corporation's investment allowance "a loan or advance to another corporation (other than a financial institution)." Since the entities to which the taxpayer loaned money were not themselves "financial institutions," the loans were qualifying assets for the purpose of the investment allowance. The result was that the taxpayer's investment allowance was increased by $500M at the end of its 1993 tax year, and so its taxable capital for that tax year was effectively reduced by the same amount.

The CRA reassessed the taxpayer. The Minister characterized the transactions with the offshore companies as avoidance transactions for the purpose of section 245 of the Act. In particular, the Minister alleged that the purpose of the transactions was to inflate the year-end carrying values of the assets that would qualify for the investment allowance under Part I.3, and convert what would otherwise be non-qualifying assets into qualifying assets. Basically, the Minister alleged that the transactions were nothing more than a means of avoiding payment of the LCT, resulted in a tax benefit to the taxpayer, had no bona fide non-tax purpose and constituted a "misuse or abuse" of the provisions of the Act.

The first question (i.e., is there a tax benefit?) is easily answered because the taxpayer obtained a deduction in the form of an increased investment allowance. The second question (i.e., is there an avoidance transaction?) is also easily answered because the evidence was such that there were internal memoranda and favorable tax opinions leading to the conclusion that the transaction was undertaken primarily to obtain a tax benefit.

The third question (i.e., is there abusive tax avoidance?) is not as easily answered. The investment allowance provision and its context do not prevent or discourage short-term loans to banks. Under the LCT scheme, the loans to financial institutions are simply excluded from the investment allowance base because such loans are excluded from the tax base of the financial institutions. Financial institutions and non-financial institutions compute their large corporations tax under different rules and there is no rule, deeming provisions or specific anti-avoidance rule in the Act that make the subsidiary of a financial institution a financial institution for purposes of the LCT. Along the same lines, the fact that the borrowers had a different year-end does not provide more ammunition to the GAAR, as the Act did not contain prohibition in this regard nor a requirement that the loan be for a longer term to qualify under the investment allowance. (13) Based on that, it would be difficult to find a purpose that would be "frustrated" when the provisions are literally applied.

When the Supreme Court's methodology is used to review the Imperial Oil situation, the result is a conclusion that there was no abusive tax avoidance. In Imperial Oil, the same conclusion was reached by the lower courts, though they had used (understandably) a different analytical approach (i.e., the approach in OSFC).

The second situation is the one that took place in Antosko, et al. v. The Minister of National Revenue, 94 DTC 6314 (S.C.C.) (Antosko). At the outset, it should be noted that the case was not decided on GAAR but simply on the applicability of subsection 20(14) which provided [paragraphing modified]:

20(14) Accrued bond interest--Where, by virtue of an assignment or other transfer of a debt obligation, other than an income bond, an income debenture, a small business development bond or a small business bond, the transferee has become entitled to an amount of interest that accrued on the debt obligation for a period commencing before the time of transfer and ending at that time that is not payable until after that time, that amount (a) shall be included as interest in computing the transferor's income for the transferor's taxation year in which the transfer occurred, except to the extent that it was otherwise included in computing the transferor's income for the year or a preceding taxation year; and (b) may be deducted in computing the transferee's income for a taxation year to the extent that the amount was included as interest in computing the transferee's income for the year.

The basic facts in Antosko were, as follows. A provincial board held stock and debt in a corporation which it had financed. The corporation failed and, in the end, owed the provincial board $5 million. The taxpayers intervened in an attempt to turn around the corporation. They acquired the corporation's shares for nominal consideration. The board also agreed to postpone the interest obligation on the indebtedness provided the taxpayers ran the corporation in a businesslike manner for two years. After the two years, the provincial board agreed to sell the indebtedness to the taxpayers for $10. The taxpayers then received from the corporation interest that had accrued before the transfer. The taxpayers included such interest in their respective incomes but at the same time claimed a deduction for the same amount under paragraph 20(14)(b) of the Act. The Minister denied the deduction on the ground that the provision's purpose is to provide for a deduction only when the income is otherwise included into income of the transferor. Given that the transferor of the indebtedness was not subject to tax, the Minister objected to the deduction (but not under the GAAR).

Consider now how the case would be analyzed under the Supreme Court's methodology under GAAR. Under the first step, clearly there would be a tax benefit inasmuch as a deduction was claimed. Under the second step, the evidence discloses that the purpose of the transaction was to allow the taxpayers to step in and turn around the corporation. Therefore, GAAR should not apply. For discussion purposes, however, consider whether one would conclude that the (hypothetical) avoidance transaction is an abusive one.

At first blush, the situation in Antosko appears somewhat similar to that in Mathew. As the Minister contended, the taxpayers effectively obtained a windfall not intended by Parliament. Examining subsection 20(14) more closely following the "textual, contextual and purposive approach," a more nuanced answer results. First, the text of the provision clearly allows for the deduction. Second, the context of the provision is such that it provides a tie-breaker rule between the transferor and transferee, nothing more. In the absence of some other rule in the statute or materials, it becomes difficult to imply a purpose justifying the imposition of a requirement that the transferor be a taxpaying entity, had not suffered losses, or even filed tax returns.

In the end, although it is not free from doubt, Antosko did not "defeat the underlying rationale" of subsection 20(14), and the case appears closer to Imperial Oil in that there was not a clear indication by Parliament of a purpose which can override the textual and contextual application of the provision.

Did the Supreme Court also Change the Rules of Statutory Construction of Tax Statutes?

Another important non-GAAR aspect has suffused the Supreme Court's judgments. It relates to the Court's "textual, contextual and purposive" approach while generally recognizing that its approach on how to construe tax statutes in past cases has been based on the text alone. In particular, in Trustco and Mathew, the Court stated: "When the words of a provision are precise and unequivocal, the ordinary meaning of the words play a dominant role in the interpretative process." (14) (Emphasis added.) By contrast, in Shell Canada Limited v. The Queen, et al., 99 DTC 5669 (S.C.C.) (Shell), one of the very same justices who wrote for the Supreme Court the judgment in Trustco and Mathew stated: "Where the provision at issue is clear and unambiguous, its terms must be simply applied." (15) This stark contrast between the two statements and in particular the addition of the adjective "dominant" warrants commentary.

Before 1984, the "strict interpretation" or "literal approach" guided the interpretation of taxation statutes. (16) In 1984, with its decision in Stubart, the Supreme Court recognized that taxation statutes had gone beyond a simple mechanism by which to raise finances and had evolved into tools of economic and social policy used by governments to influence society, (17) and that a new doctrine of statutory interpretation was required. Writing for the majority, Justice Estey adopted the reasoning of an eminent scholar, Elmer Dreidger, that:

   [T]here is only one principle or approach, namely,
   the words of an Act are to be read in their entire
   context and in their grammatical and ordinary
   sense harmoniously with the scheme of the Act,
  the object of the act, and the intention of Parliament. (18)

This approach, which is sometimes referred to as the "modern rule of statutory construction" or "the words in total context" approach (the so-called Modern Rule) is the preferred approach, according to the Supreme Court. (19) While superficially the Modern Rule appears similar to the purposive approach, it is not precisely the same. The purposive approach requires the interpreter first to ascertain the purpose of the legislation and then to select the meaning of the word or expression interpreted that suits that purpose. By contrast, the Modern Rule requires the interpreter "to paint the subject words with the colour of their surroundings" (i.e., the context). (20) The Supreme Court's preference for the Modern Rule over the purposive approach explains why it has consciously refrained from adding words to a provision or embellishing it. (21) The rationale for this approach is best explained by the following statement of Justice Iacobucci writing for the majority of the Supreme Court in 65302 British Columbia Ltd. v. H.M.Q.:

   [T]his Court has also often been cautious in utilizing
   tools of statutory interpretation in order
   to stray from clear and unambiguous statutory
   language. In Canada v. Antosko, [1994] 2 S.C.R.
   312, at p. 326-27, this Court held: "While it is true
   that the courts must view discrete sections of the
   Income Tax Act in light of the other provisions of
   the Act and of the purpose of the legislation, and
   that they must analyze a given transaction in
   the context of economic and commercial reality,
   such techniques cannot alter the result where the
   words of the statute are clear and plain and where
   the legal and practical effect of the transaction is
   undisputed.

   "In discussing this case, P.W. Hogg and J.E. Magee,
   while correctly acknowledging that the context
   and purpose of a statutory provision must always
   be considered, comment that 'lilt would introduce
   intolerable uncertainty
   into the Income Tax Act
   if clear language in a
   detailed provision of the
   Act were to be qualified by
   unexpressed exceptions
   derived from a court's
   view of the object and
   purpose of the provision':
   Principles of Canadian
   Income Tax Law 2nd ed.,
   1997 at pp. 475-76. This
   is not an endorsement of
   a literalist approach to
   statutory interpretation, but a recognition that in
   applying the principles of interpretation to the Act,
   attention must be paid to the fact that the Act is one
   of the most detailed, complex, and comprehensive
   statutes in our legislative inventory and courts
   should be reluctant to embrace unexpressed notions
   of policy of principle in the guise of statutory
   interpretation." (22)

Even though the Supreme Court declared in Stubart that tax statutes were not to be construed under the "literal or literalist" approach, the reality is that the Court has, in its subsequent decisions, paid close attention to the words of the tax statutes. This appears to be confirmed, as the state of the law perhaps, by the following passage in Trustco:

11 As a result of the Duke of Westminster principle (Commissioners of Inland Revenue v. Duke of Westminster, [1936] A.C. 1 (H.L.)) that taxpayers are entitled to arrange their affairs to minimize the amount of tax payable, Canadian tax legislation received a strict interpretation in an era of more literal statutory interpretation than the present. There is no doubt today that all statutes, including the Act, must be interpreted in a textual, contextual and purposive way. However, the particularity and detail of many tax provisions have often led to an emphasis on textual interpretation, where Parliament has specified precisely what conditions must be satisfied to achieve a particular result, it is reasonable to assume that Parliament intended that taxpayers would rely on such provisions to achieve the result they prescribe. [Emphasis added.]

The remaining question, given the apparent conflict between Trustco/Mathew and Shell, and certain commentaries made in the past, (23) appears to be: When is there clear enough language to allow the interpreter to limit itself to the "plain meaning" of the words? In other words, using the language in Trustco, when does the ordinary meaning "dominate" the interpretation process? The Supreme Court, in Bell Express-Vu, a non-tax case, answered that question, as follows:

   It is necessary, in every case, for the court charged
   with interpreting a provision to undertake the
   contextual and purposive approach set out by
   Driedger, and hereafter to determine if the "words
   are ambiguous enough to induce two people to
   spend good money in backing two opposing views
   as to their meaning" (Willis, supra, at pp. 4-5). (24)
   [Emphasis added.]

Optimally, the use of the word "dominant" to qualify the "ordinary meaning rule" was simply an attempt to state what the rule is supposed to be as opposed to opening the door to some other interpretative approach.

Conclusions

Reading the Trustco and Mathew decisions only reinforces the old saying that one can only tell what the Supreme Court meant after reading, in a subsequent judgment, what it says it meant. In the meantime, some important conclusions and limitations on the GAAR are contained in the recent judgments:

1. Economic Substance. Economic substance was at the heart of the Minister's argument in Trustco. While the Supreme Court did acknowledge that economic substance "may be relevant" in analyzing a transaction, it clearly stated that the concept "has little meaning in isolation from the proper interpretation of specific provisions in the Act." As such, the Supreme Court clearly rejected any notion that the GAAR itself imposed an economic substance test that could be used to override specific provisions of the Act that did not.

2. Certainty, Predictability, and the Role of the Courts. Consonant with past pronouncements, the Supreme Court reiterated (a) the importance of certainty and predictability in tax laws so that taxpayers are able to order their affairs intelligently, and (b) that the courts are not the proper place for formulating fiscal policy. While some commentators have argued that the first goal is merely one of many (such as fairness, neutrality, and raising revenue) and the second concern is overblown, the Supreme Court's continuing strong support for these principles in the context of the GAAR is a welcome development.

3. Minister's Burden to Establish Clear Misuse and Abuse. The Supreme Court specifically endorsed the twin propositions that GAAR can only be applied "when the abusive nature of the transaction is clear" and that the burden is on the Minister to establish that the transaction frustrates the "object, spirit, or purpose" of a particular provision of the Act. This reaffirms the extraordinary nature of section 245 and the view that, in cases of doubt, the benefit of that doubt should go to the taxpayer.

In addition, even if the recent judgments do not precisely tell what constitutes abusive tax avoidance, they nonetheless provide guidance on how a transaction and the relevant provisions of the Act are to be scrutinized and interpreted under the GAAR. Although the decisions were made to deal with two precise (and different) transactions, they appear to be more of a guide to deal with GAAR issues rather than a code. The Supreme Court reminded the lower courts that GAAR is in many respects a factual enquiry 25 with the consequence that "where the Tax Court judge has proceeded on a proper construction of the provisions of the Income Tax Act and on findings supported by the evidence, appellate tribunals should not interfere, absent a palpable and overriding error." (26)

On a more practical note, the recent judgments do reveal more usefulness when real transactions are examined. In Trustco (again), the court would not find an explicit or implicit prohibition against the subject transaction in the legislation, hence no abusive tax avoidance. In other words, there was nothing to frustrate Parliament's intent. In Mathew, the approach was different as the Court determined that it would frustrate Parliament's intent to give the taxpayer a benefit that was never contemplated in the first place. In the same vein, one could conclude that a court will likely apply GAAR if a transaction runs contrary to the Department of Finance explanatory notes but is saved by the actual (inadequate) wording of the legislation.

In the end, there is no doubt that the judgments in Trustco and Mathew will resonate many years to come in the Canadian tax jurisprudence. Ultimately, however, the judgments should be interpreted as laying the foundation for a GAAR analysis and nothing more.

(1) Trustco, para. 1.

(2) Trustco, para. 1.

(3) Subsection 245(4) was amended on May 13, 2005, effective with respect to transactions entered into after September 2, 1988. The stated purpose of this amendment was to extend the application of the GAAR to the misuse or abuse of the Regulations, Income Tax Application Rules, tax treaties and other legislation. The amendment, however, also altered the language of subsection 245(4), by removing the double negative form in the paragraph. While the Supreme Court in Trustco took notice of the amendment, it refused to apply the new language to the appeals before it on the basis that such retroactive changes "cannot apply at this stage of appellate review"(Trustco, para 7). The Court did expressly note that "even if this amendment were to apply, it would not warrant a different approach to the issues on appeal" (Trustco, para. 7) without however expanding further on this specific point. Contrast with Trustco, para. 62.

(4) The Supreme Court endorsed "the test as adopted by the House of Lords that a series of transactions involves a number of transactions that are 'pre-ordained in order to produce a given result' with no practical likelihood that the pre-planned events would not take place in the order ordained Craven v. White, [1989] A.C. 398 at p. 514, per Lord Oliver; see also W.T. Ramsay Ltd. Inland Revenue Commissioners, [1981] ALL E.R. 865" (Trustco, para. 25). Also of note is the Supreme Court's interpretation of the phrase "related transactions on events in contemplation of the series in subsection 248(1) in the broader sense to mean "related events or transactions in relation to the series" to capture events occurring before or after the basic subject transaction.

(5) Trustco, para. 21.

(6) Trustco, para. 28.

(7) Trustco, para. 30 (emphasis added).

(8) Trustco, para. 42.

(9) Trustco, para. 42.

(10) See, e.g., Rooke v. The Queen, 2002 DTC 7442 (F.C.A.).

(11) Trustco, para. 50.

(12) Trustco, para. 74.

(13) By contrast, several provincial capital tax statutes raising taxes similar to the LCT contain specific anti-avoidance rules to address such type of transaction.

(14) Trustco, para. 10.

(15) Shell, para. 40.

(16) For a good review of the evolution of the principles guiding the interpretation of tax statutes, see P.W. Hogg, J.E. Magee & T. Cook, Principles of Canadian Income Tax Law 8-9 and 556-65 (4th ed. 2002).

(17) Id. at 6321-6322.

(18) Id. note 3 at 6323.

(19) See Ludco Enterprises Ltd., et al. v. The Queen, 2001 DTC 5505 (S.C.C.), at para. 37. See also The Queen v. Markevich, et al., 2003 DTC 5185 (S.C.C.), at para. 12 for the Court's most recent pronouncement on point.

(20) This is paraphrased from a quotation ("words like people, take their colour from their surroundings") from the seminal article by Professor John Willis entitled Statute Interpretation in a Nutshell (1930), 18 C.B.R. 6, that was recently adopted by the Supreme Court in Bell Express-Vu Limited Partnership v. Rex et al. [2002] 2 S.C.R. 559, at para. 27.

(21) See The Queen v. Markevich, et al., 2003 DTC 5185 (S.C.C.), at para. 15, where the Court unanimously approved Friesen v. The Queen, 95 DTC 5551 (S.C.C.), at para. 27. See also The Queen v. Walls and Buvyer, 2002 DTC 6960 (S.C.C.), at para. 22.

(22) 65302 British Columbia Ltd. v. The Queen, 99 DTC 5799 (S.C.C.), at para. 51 (emphasis added).

(23) See Brian J. Arnold, Reflections on the Relationship between Statutory and Tax Avoidance, CTJ (2001), Vol. 49, No. 1. See the dissenting reasons of Justice Lebel. (concurred by Justice Bastarache) in The Queen v. Singleton, 2001 DTC 5533 (S.C.C.), at paras. 59-68.

(24) Bell Express-Vu Limited Partnership v. Rex et al. [2002] 2 S.C.R. 559, para. 30.

(25) According to the Supreme Court, step 1 in the analysis (i.e., is there a tax benefit?) is (surprisingly) a question of fact. Even in a simple case, it contains a preliminary requirement to read the Act and, in some instances, to compare alternate tax treatments. With respect, this question is akin to asking "is there a contract?" or "is there a tort?," which are more properly viewed as questions of fact and of law. According to the Court, step II (i.e., is there an avoidance transaction) is also a factual enquiry. Step III (i.e., is the transaction abusive?) is partly a legal analysis and partly the application of legal principles to the facts.

JACQUES BERNIER leads the Tax Litigation and Dispute Resolution Practice Groups of Bennett Jones LLP. A graduate of the University of Ottawa, he is resident in the firm's Toronto office. He has significant experience in representations to federal and provincial tax authorities to resolve disputes in the areas of income, capital, commodity and mining taxation. The author thanks Martin Sorensen of Bennett Jones for his contribution to this article.