Category Archives: Inter-Corporate Dividends

Anti-Avoidance Rule in Paragraph 95(6)(b) Clarified – Lehigh Cement

Download PDF

Anti-Avoidance Rule in Paragraph 95(6)(b) Clarified

Canada v Lehigh Cement Limited, 2014 FCA 103

An appeal from the TCC decision in 2013 TCC 176, granting the taxpayer’s appeal of the MNR’s reassessment.

At issue was the proper interpretation of paragraph 95(6)(b) of the Income Tax Act, specifically how wide the inquiry into the “principal purpose” of a share transaction could be.


Lehigh and CBR acquired shares of a non-resident corporation as part of a complex and wide restructuring involving many steps (Details at paras 9-10 of the FCA decision).  They received dividends from this non-resident company and claimed a deduction under paragraph 113(1)(a).  The MNR disallowed the deductions and applied paragraph 95(6)(b) on the basis that the taxpayers had acquired the shares of the non-resident with the  principal purpose of avoiding the payment of tax that would otherwise be payable under the ITA.

The FCA agreed with the TCC outcome (taxpayer succeeding) but for different reasons. The TCC had determined that the principal purpose of the acquisition or disposition was a question of fact to be determined in light of all relevant circumstances and that one salient consideration is whether the share purchase or sale was part of a series of dispositions or acquisitions conducted with a view to avoid tax.  The TCC then considered the effect of the phrase “tax…that would otherwise be payable”, and stated that this required looking at the alternate scenario where the share transaction has not occurred.  In this case, the TCC determined that there was no change in tax consequences before or after the share transaction and therefore 95(6)(b) did not apply.

The MNR’s position at the FCA was that in applying 95(6)(b), was that it was proper to look at a series of transactions to see if there was a tax avoidance purpose. The Taxpayer’s position was that only the share transaction itself must be looked at.


The FCA began by providing a thorough overview of the legislative regime which paragraph 95(6)(b), in subdivision I of Division B of the ITA is part of.  How a Canadian resident is taxed on income derived from a non-resident corporation depends on two factors:

  1. the type of income; and
  2. the ownership status of the non-resident corporation.

Subsection 90(1) requires that dividends received by a Canadian taxpayer from  non-resident corporation be included in income when received.  But, 113(1)(a) provides an offsetting deduction for dividends paid by “foreign affiliate” corporations paid out of exempt surplus.  Very simply stated, 95(1) defines “foreign affiliate” to be a non-resident corporation in which a Canadian taxpayer holds at least 1% interest of any class of shares AND that taxpayer’s holdings combined with the holdings of any related person total 10% or more of the class.

The court noted that the ownership status of the non-resident corporation can easily be manipulated so as to realise tax savings. The court quoted from Vern Krishna, The Fundamentals of Canadian Income Tax (9th ed., 2006) at page 1327, where it is said that ““taxpayers jockey to get on the right side of the distinctions to take advantage of the rules”.  Paragraph 95(6)(b) was enacted to address taxpayers’ ability to manipulate the ownership status of non-resident corporations.  The court stated (para 20):

 Broadly speaking, paragraph 95(6)(b) provides that where a person acquires or disposes of shares of a corporation and it can reasonably be considered that the principal purpose of the acquisition or disposition is to permit a person to avoid, reduce or defer the payment of tax, the acquisition or disposition is deemed not to have occurred.

The FCA referred to the decision in Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54, where the rule for statuary interpretation was laid out. The FCA then set out the rules in relation to Income Tax provisions:

[39]           The provisions in taxation statutes are often detailed and particular. The Income Tax Act is “an instrument dominated by explicit provisions dictating specific consequences,” and this invites “a largely textual interpretation”: Canada Trustco, at paragraph 13.

[40]           As a result, “[w]here 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”: Canada Trustco, supra at paragraph 11. Where the provision at issue is “clear and unambiguous,” its words “must simply be applied”: Shell Canada Ltd. v. Canada, [1999] 3 S.C.R. 622 at paragraph 40. In such circumstances, a supposed purpose “cannot be used to create an unexpressed exception to clear language” or “supplant” clear language: Placer Dome Canada Ltd. v. Ontario (Minister of Finance), 2006 SCC 20, [2006] 1 S.C.R. 715 at paragraph 23, citing P. W. Hogg, J. E. Magee and J. Li, Principles of Canadian Income Tax Law (5th ed. 2005), at page 569.

[41]           When interpreting provisions in taxation statutes, we must keep front of mind their real life context: many taxpayers study closely the text of the Act to manage and plan their affairs intelligently. Accordingly, we must interpret “clear and unambiguous” text in the Act in a way that promotes “consistency, predictability and fairness,” with due weight placed upon the particular wording of the provision: Canada Trustco, at paragraph 12, citing Shell Canada Ltd., supra at paragraph 45.

[42]           We must not supplant or qualify the words of paragraph 95(6)(b) by creating “unexpressed exceptions derived from [our] view of the object and purpose of the provision,” or by resorting to tendentious reasoning. Otherwise, we would inject “intolerable uncertainty” into the Act, undermining “consistency, predictability and fairness”: 65302 British Columbia Ltd. v. Canada, [1999] 3 S.C.R. 804, at paragraph 51, citing P. W. Hogg and J. E. Magee, Principles of Canadian Income Tax Law (2nd ed. 1997) at pp. 475-76; see also Canada Trustco, at paragraph 12.

[43]           In the course of applying these principles, legislative history and explanatory documents such as technical notes, budget papers and committee minutes can offer assistance.

[44]           Overall, though, our task is to discern the meaning of the provision’s text using all of the objective clues available to us.

In applying this test, the court accepted the taxpayer’s interpretation.  Paragraph 95(6)(b) focuses precisely and unequivocally on the principal purpose of the acquisition or disposition of the shares and NOT on that of the series of transactions on which the particular acquisition or disposition is part of.  The FCA held that 95(6)(b) is aimed at a particular species of tax avoidance (manipulation of foreign affiliate status), and does not use the words “series of transactions or event”.  The FCA stated (para 50):

Rather, the words of paragraph 95(6)(b) require that the tax benefit must flow from the share acquisition or disposition itself and obtaining the tax benefit must be the principal purpose of the share acquisition or disposition.

The court also considered other contexts, including amendments made to other anti-avoidance provisions and the architecture of the ITA (provision in a specific area not a general area).  The provision is only aimed at dealing with manipulations of foreign affiliate status (para 55).

The FCA also dealt with the MNR’s argument based on the purpose of the provision. It did so by considering the potential effects of the alternative interpretations, and held that the MNR’s interpretation would result in unacceptable (board and ill-defined) discretion leading to uncertainty (para 64):

[64]           Unacceptability is in the eye of the beholder. It can shift depending on one’s subjective judgment and mood at the time. Using it, as the Crown suggests, to restrain the indiscriminate use of paragraph 95(6)(b) creates the spectre of similarly-situated taxpayers being treated differently for no objective reason. This would violate the principle that, absent clear legislative wording, the same legal principles should apply to all taxpayers: Bronfman Trust v. The Queen, [1987] 1 S.C.R. 32 at page 46.

The court concluded:

[68]           […]  paragraph 95(6)(b) is targeted at those whose principal purpose for acquiring or disposing of shares in a non-resident corporation is to meet or fail the relevant tests for foreign affiliate, controlled foreign affiliate or related-corporation status in subdivision i of Division B of Part I of the Act with a view to avoiding, reducing or deferring Canadian tax.

[69]           The principal purpose of the acquisition or disposition of shares in the non-resident corporation is a question of fact to be determined on the basis of all relevant circumstances. An entire series of transactions may form part of the circumstances relevant to discerning the principal purpose of the acquisition or disposition of shares in the non-resident corporation. But it is not open to the Minister to look at an entire series of transactions to discern a tax avoidance purpose that is not the specific target of paragraph 95(6)(b).

[70]           Manipulating the shareholdings in the non-resident corporation to change its status in subdivision i of Division B of Part I of the Act in order to avoid, reduce or defer Canadian tax by itself does not necessarily trigger paragraph 95(6)(b) of the Act. The purpose must be the principal – i.e. dominant or main purpose – not just one of many different purposes.

In this case, the principal purpose of the share transaction was to avoid US tax and not Canadian Tax, and 95(6)(b) had no application.

Sas Ansari, BSc BEd PC JD LLM PhD (exp) CPA In-Depth Tax 1, 2 &3

If you like this website, please share it with others.

Back To Top OR Home

Spruce Credit Union v The Queen, 2012 TCC 357

Download PDF

GAAR analysis framework – Purpose of ITA s 112 – meaning of “in proportion to”

Spruce Credit Union v The Queen, 2012 TCC 357

There were two issues in this case:

  • Whether a dividend received by the credit union (“CU”) from a BC deposit insurance corporation (a taxable Canadian corporation) is deductible under s 112 inter-corporate dividend deduction in the ITA?
  • If so, does GAAR apply to the receipt of the dividend?

The TCC provided a detailed guide to how a court ought to approach GAAR analysis.  It first determined that the dividend paid met the legal definition of a dividend and was thus, absent GAAR, deductible pursuant to s 112.  The Court then considered the GAAR question and noted that in light of the overall non-tax purpose of the transaction, and the fact that no step was inserted or undertaken primarily for obtaining a tax result, there was no avoidance transaction to which GAAR could apply.

Importantly, the Court noted that even if tax considerations play a primary role in a taxpayer’d choices this doesn’t necessarily make the transaction primarily tax motivated. The Court was very clear to differentiate between a “choice” and a “transaction”, and stated that the manner of distributing funds was a choice, and that the “act of choosing or deciding between or among alternative available transactions or structures to accomplish a non-tax purpose, based in whole or part upon the differing tax results of each, is not a transaction [as] making a decision can not be an avoidance transaction” (para 93). Taxpayer’s are free to choose among alternatives purely on the basis of tax outcomes.


The TCC did not deal with the question of whether there was a “tax benefit”, but this may be due to the arguments raised by the taxpayer (ie that the lack of tax benefit was not argued, and therefore the burden on the taxpayer on this issue was not met).

The TCC also noted, but only in passing without analysis, that the decision here was that of the deposit insurance corporation, and not that of the taxpayer and not one that could have been influenced by the taxpayer (lack of control).

The decision does, however, highlight the freedom of taxpayer’s to elect the least tax onerous manner of structuring their affairs – thus freedom of choice among alternatives purely on the basis of tax consequences- so long as no steps taken to achieve the non-tax purpose are not primarily for a bona fide non-tax purpose – ie a necessary connection between steps to acheive the non-tax purpose.


The CU was involved with two deposit insurance corporations (CUDIC and STAB) as per BC regulations of CUs.

CUDIC is a table Canadian corporation controlled and operated by the FI Commission, which is an arm of the BC government.   It is funded by assessments paid by CUs, which are deductible by the CUs pursuant to section 9 and subsection 137.1(11) of the ITA.  CUDIC was responsible for maintaining the deposit insurance fund.

STAB is a taxable Canadian corporation, and is a central credit union under the relevant BC statute. Each CU in BC was require to be a member and shareholder of STAB. STAB was funded by assessments, which were deductible to the payor CU pursuant to paragraph 137.1(11) of the ITA.  STAB would rebalance its members’ shareholdings to reflect the current size of the members. STAB was responsible for maintaining the stability of CUs.

Due to some changes in the FI Commission’s views on the proper allocation of funds between CUDIC and STAB, there was a need to transfer funds between the two organizations.  Because of the legal (lack of) relationship between these two corporations, and as result of discussions, alternatives were considered.  STAB could make distributions to its members either by way of refund of premiums or by dividend.  There was a need to distribute funds from STAB to CUs to pay for the imminent CUDIC assessment required by BC law.  CUDIC assessed and independently STAB paid dividends to its members, but split it between two dividends due to the RUlind Directorate’s position that there is no GAAR concern with respect to the aggregate cumulative investment income amount (vs assessment income).


The crown argues that despite s 112, the dividend received by the Credit union is not deducible because of s 137.1, because ” (i) the dividend amounts were paid to the credit unions as allocations in proportion to assessments received by the deposit insurance corporation from the credit unions, and required to be included in the credit unions’ incomes under paragraph 137.1(10)(a); and (ii) section 137.1 is a complete code with respect to such amounts and does not permit them to then be deducted” (para 2).

Alternatively the crown argues that GAAR applies to the extent that the amount of the dividend is also an amount described in paragraph 137.2(10)(a)  as being a refund of premiums that must be included in Income by subsection 137.1(10) of the ITA, so that its not a dividend that is thereafter deductible (para 3).


Section 112

The TCC noted that there is no reason on the basis of s 112 of the ITA to treat the two dividend payments differently. The amounts paid were in law dividends, and they are taxable dividends required to be included in income by 82(1).

The court stated that the purpose of s 112 ” inter-corporate dividends received deduction is to avoid double taxation of the after-tax profits of a corporation as they are paid by way of dividends to shareholder corporations. This forms part of the Act’s approach to achieving a degree of integration of corporate and personal taxation of income earned through a corporation.” (para 43).

The TCC did not believe that the Crown’s position was supported by the evidence – ie that the amount was paid by STAB as an allocation in proportion to assessments received from the CUs as contemplated by 137.1(10)(a).  This is because the section uses the word “as” and not words such as “in respect of” or “as, on account, or in lieu of”, or more expansive language of that kind (para 46).  ” It does not speak of amounts that could reasonably be considered to relate to, or directly or indirectly be funded by, assessments previously received. The form or nature of the amount of the payment is specifically described given the use of the word “as” in English and the words “à titre de” in French.” (para 46).

Definition of “in proportion to”

The amounts were paid by STAB as dividends in proportion to each shareholder’s shareholdings.  The TCC said that the amounts could both be dividends and fall under 137.1(10)(a).  The Court went on to say at paragraph 49:

The meaning of the term “in proportion to” is neither unclear nor ambiguous. A proportion is a comparative ratio that is a part considered in comparative relation to a whole. For two things to be in proportion to one another there must be an equality of ratios. For an amount to be paid to persons in proportion to their assessments, it is a requirement that the person receive that portion of the aggregate amount paid that assessments received from them is of the total of all assessments received. That is, there must be an equality of ratios. That the amount paid to them was arguably funded by the payer in whole or in part directly or indirectly, with assessments received, or income earned on such assessments, is clearly not sufficient. That position would require that the meaning of the words “in proportion to” be ignored. Similarly, the fact that a member received a portion of the pool paid out does not lead in any way to the conclusion it was paid proportionate to their assessments. For the Respondent’s position to be correct on the facts of this case, I would have to read “shareholdings” for the word “assessments” used in the legislation.

Since the court held that the amounts were not “paid in proportion to”, and therefore didn’t need to decide whether section 137.1 is a complete code.

GAAR Issue:

[NOTE that the court provides a very detailed review of how to approach GAAR – not all of which is reproduces here]

The TCC referred to the decisions of the SCC in Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54, [2005] 2 S.C.R. 601, 2005 DTC 5523 and in Lipson v. Canada, 2009 SCC 1, [2009] 1 S.C.R. 3, 2009 DTC 5015, and most recently reaffirmed in Copthorne Holdings Ltd. v. Canada, 2011 SCC 63, 2012 DTC 5006, as setting out the test in GAAR.

The Court noted that absent GAAR, “taxpayers are entitled to select courses of action or to enter into transactions that will minimize their tax liability relying upon the Duke of Westminster principle [2]. Taxpayers are entitled to know with a degree of certainty that the provisions of the Act apply to transactions with real economic substance [3]” (para 56).  The Court went on to state at paragraph 57:

The GAAR is an exceptional provision of last resort that may be invoked by the Minister of National Revenue (the “Minister”) if he believes that the taxpayer’s chosen transactions, notwithstanding that they comply with the literal requirements of the provisions in question, are not in accord with the object, spirit, rationale or purpose of the provisions and indeed frustrate and abuse them[4]. The GAAR creates an unavoidable degree of uncertainty for taxpayers and, for this reason, a court must undertake its analysis cautiously[5]. It is the obligation of the Minister to demonstrate clearly the abuse he alleges[6]. Any residual doubt is resolved in favour of the taxpayer[7].

This question of whether there is a tax benefit” requires the court to identify and isolate the tax benefit from the non-tax purpose of the taxpayer’s chosen transactions in one of several ways:

[61]        If a deduction against taxable income is claimed in the impugned transaction or series of transactions, the existence of a tax benefit is clear since a deduction results in a reduction of tax[8].


[62]        Alternatively, the existence of a tax benefit can be established by comparing the taxpayer’s chosen transactions with an alternative transaction that might reasonably have been carried out but for the existence of the tax benefit[9].


[63]        The burden is on the taxpayer to refute the Minister’s assumption of the existence of a tax benefit[10].


[64]        Whether or not there is a tax benefit is a question of fact, subject to review on the basis of palpable and overriding error[11].

The Question of whether there is an “avoidance transaction”, looks to whether the step that led to the tax benefit was undertaken primarily for a bona fide non-tax purpose, and ” If there is a series of transactions that results directly or indirectly in a tax benefit, any transaction or step in the series will be an avoidance transaction if that step is not undertaken primarily for a bona fide non‑tax purpose” (para 66).  The TCC stated at paragraph 68:

[68]        The courts must at this stage examine the relationship between the parties and the actual transactions that were executed between them. A transaction cannot be an avoidance transaction because some alternative transaction that might have achieved an equivalent result would have resulted in more tax. That will not suffice to establish an avoidance transaction[16], though, as summarized above, it may suffice to establish a tax benefit.

The limitation of this approach was highlighted by the TCC at paragraph 69:

 [69]        Consistent with its decision in Trustco, the Supreme Court of Canada in Copthorne does not suggest that it is appropriate at the avoidance transaction stage of the analysis to compare the taxpayer’s chosen transaction or series to other available structures to see if the taxpayer chose among the alternatives primarily based on tax considerations or consequences. This makes sense. If it were otherwise, taxpayers would be obliged to choose a more taxable alternative and the Duke of Westminster principle would be completely for naught. It appears to be at least to this extent that the Supreme Court of Canada repeatedly sets out that the Duke of Westminster principle co-exists with the GAAR.”

So long as each step of a transaction is primarily for a non-tax purpose, incidental tax benefits will not render the transaction an avoidance transaction. (para 70).  The TCC noted that ” tax considerations may play a primary role in a taxpayer’s choice of available structuring options to implement a transaction or series of transactions without necessarily making the transaction itself primarily tax motivated” (para 71).

The Court note that the “overall non-tax objective” of the transaction was admitted by the Crown, and this was not a case were a step was inserted or undertaken primarily for being able to obtain a desired tax result (paras 91-92).  The Court stated at paragraph 93:

[93]        The act of choosing or deciding between or among alternative available transactions or structures to accomplish a non-tax purpose, based in whole or in part upon the differing tax results of each, is not a transaction. Making a decision can not be an avoidance transaction.

In this case STAB made a decision in line with the Duke of Westminster Principle, and such a decision “can not be considered a transaction for GAAR purposes” (para 94).  The Court concluded beginning at paragraph 95:

[95]        This is an example of a case where a taxpayer:

(i)               decides to do something for entirely business or other non-tax purposes – (that is, put money in its shareholders’ hands to allow them to pay their business obligations and to recalibrate the level of the deposit insurance and stabilisation fund maintained for their benefit to the level now needed);

(ii)            considers the alternatives available to them to accomplish what is needed to be done, including a consideration of the tax consequences and costs of each; and

(iii)         chooses an available option that is not the one with the greatest tax cost and may be the one with the least tax cost or no tax cost at all.

[96]        Provided no steps or transactions were inserted into the commercial transactions implementing the chosen structure primarily to obtain the tax benefit, neither the taxpayer’s choice nor its implementation can meet the statutory definition of “avoidance transaction” as interpreted by the Supreme Court of Canada.

The Court also noted as an aside that the decision here was that of STAB and not that of the taxpayer (para 97), and went on to hold that there is no fault for choosing a more tax efficient route, ” Provided the structure of the transactions used to implement their choice does not include any step the primary purpose of which is to position themselves to obtain the desired tax benefit, or is otherwise primarily tax driven, their tax benefit can not result from an avoidance transaction and the GAAR by its terms can not apply. See, for example, former Chief Justice Bowman’s decision in Geransky v. Canada, 2001 DTC 243.” (para 97).

The Court said that there was no need to move on to consider the “abuse or misuse” test as there was no “avoidance transaction” allowing GAAR to apply.