Subsection 245(4)

A Calgary shopping mall was sold by Harvard Properties and the other co-owners to a third party (“Abacus”) in a share sale transaction but at a price representative the mall’s asset value and, thus, at a premium to its share-sale value. This was accomplished by transferring their co-ownership interests on a s. 85(1) rollover basis to respective Newcos (“HP Newco”, in the case of Harvard Properties) in consideration inter alia for voting and non-voting shares, followed by a sale of those voting shares to an Abacus subsidiary (NH Properties) for promissory notes for under half of the sale price. The Newcos then sold the shopping centre to a third party (Bentall), and the co-owners then sold their Newco non-voting shares to NH Properties for the balance of the purchase price (receiving, by direction, the Bentall sales proceeds), at no gain due to an ACB step-up pursuant to a stated capital increase coming out of the newly-created capital dividend accounts of the Newcos. Real estate counsel for the vendors negotiated for these transactions to all occur in one integrated interdependent closing.

After finding that s. 160 applied to these transactions in the cash amounts referenced by him, subject to a determination in still-pending proceedings as to the quantum of any unpaid tax liability of NH Properties, on the basis inter alia that Harvard Properties was not dealing at arm’s length with NH Properties, Boyle J found, in the alternative, that if the series of transactions had avoided the application of s. 160, there was an abusive avoidance of s. 160 through the structuring of a supposed arm’s-length through a sale of the voting shares of the Harvard Properties’ Newco (HP Newco) before the asset sale, capital dividend and sale of the non-voting Newco shares. He J stated (at para. 206):

[I]f these avoidance transaction steps in the series succeeded in changing Harvard Properties’ relationship to its controlled subsidiary, HP Newco, in the midst of the single closing into an arm’s length relationship and instead of when the closing was complete, and that in turn led to Harvard Properties being arm’s length with NH Properties, the results of these impugned avoidance transactions are a section 160 hat trick or trifecta of abuse. They accomplished what section 160 sought to prevent – its application, they defeated its underlying rationale, and they circumvented its application.

Furthermore, Boyle J considered that the reasonable way in the circumstances to deny the tax benefit was to treat s. 160 as applying to the extent of the shortfall in consideration given on the transfer (para. 208).

A week before realizing a $119 million capital gain, the taxpayer, which until then was wholly-owned by a Canadian-resident individual, issued voting preference shares to his non-resident children (who thereby acquired de jure control), but not to a resident son. CRA had decided not to take the position that the taxpayer had remained a CCPC by virtue of continued de facto control by the resident individual, but that GAAR should instead be applied. CRA assessed the taxable capital gain on sale, so as to deny the general rate reduction under s. 123.4 and impose refundable tax under s. 123.3, on this basis.

Based on the Crown’s submissions at the hearing of this motion, its position appeared to be that, although Lark was no longer a CCPC because of the de jure control of the non-resident children, there was a GAAR abuse because de facto control was maintained in Canada. However, this position was not reflected in the Reply, which instead contained vague references to the integration system and abuse of ss. 123.3 and 123.4, and did not refer to de facto control.

St-Hilaire J found (at para. 60) that the relevant part of the Reply “may prejudice the fair hearing of the appeal and is an abuse of process” and should be struck – but with leave to the Crown to amend its pleadings.

After noting (at para. 43) that the taxpayer “appears to complain that there was no GAAR analysis conducted by the GAAR Committee specifically for Lark’s case” and that the GAAR Committee minutes provided to the taxpayer instead referred to a continuance of the corporation as a non-Canadian corporation, St-Hilaire J stated (at para. 45) that she adopted the view in Aikman “that there is no requirement that the Minister consult the GAAR Committee.”

The taxpayer (“KQI”), a Canadian operating subsidiary in the Kone multinational group, used group funds advanced to it by a group company (Kone Canada), in part as an interest-bearing loan, to purchase, for a cash purchase price of $394 million, cumulative preferred shares of a US affiliate (Kone USA) from the non-resident affiliated company (Kone BV) to which such shares had recently been issued as a stock dividend. At the same time, KQI agreed to resell such preferred shares at pre-agreed higher prices, to Kone BV in three and five years’ time, which in fact occurred. The gain arising under this resale was deemed under s. 93 to be dividends coming out of exempt surplus of Kone USA.

The ARQ sought to impute interest income to KQI under TA s. 127.6, the Quebec equivalent of ITA s. 17(1), on the basis that the above “repo” transaction was a sham that should instead be characterized as an interest-free loan by KQI to Kone BV or, alternatively, that the repo transaction represented an abusive avoidance of such s. 17 equivalent for Quebec GAAR purposes.

After rejecting the sham argument, the Court went on to find that the Quebec GAAR did not apply, stating (at paras. 34-36):

One cannot ignore the fact that financing transactions that are not loans will not generate interest but may provide for other forms of return. … A repo with a reasonable return in the form of dividends does not defeat the OSP [object, spirit and purpose] of Section 127.6.

… KQI is taking advantage of … a mismatch between the tax treatment of its income (the dividends from Kone US are not taxable because they are paid out of its exempt surplus) and its expense (the interest in pays to Kone Canada is deductible). …

However, the mismatch arises from the Taxation Act and the policies underlying it … . The exempt surplus in Kone US is made up of its earnings from an active business on which it has already paid income taxes.

With a view to its imminent disposition of the shares of a subsidiary which it had acquired on a rollover basis, the taxpayer continued to the British Virgin Islands, with the result that it ceased to be a Canadian-controlled private corporation (CCPC) and became a private corporation that was not a CCPC (its central management and control remained in Canada). CRA assessed on the basis that the resulting non-application of s. 123.3 (imposing refundable tax) and s. 123.4 (denying the general rate reduction on aggregate investment income) was an abuse of those provisions and of s. 250(5.1) (which deemed the taxpayer on its continuance to have been incorporated outside Canada).

In finding no abuse of those provisions, D’Arcy J first turned to s. 123.3 and stated (at paras. 212, 226-227):

Parliament has chosen, for policy reasons, to have different sets of rules for different corporations. …

The object, spirit and purpose of the definition of CCPC is to provide a dividing line between those corporations that are taxed under the specific regime for CCPCs and those corporations that are not taxed under this regime. Prior to being continued in the British Virgin Islands, the Appellant was on one side of the dividing line and, after it was continued, it was on the other side of the dividing line. This is exactly what Parliament intended. …

[T]he Appellant’s choice to be taxed as a non-CCPC did not abuse section 123.3 since Parliament only intended it to apply to a corporation’s investment income that is taxed under the regime for CCPCs.

After having indicated that the absence of a rate reduction under s. 123.4 for aggregate investment income reflected that it was subject to a low rate of tax once distributed as dividends, D’Arcy J noted that there was no abuse regarding the s. 123.4 rate reduction, stating (at para. 228):

Since the income was not eligible for any other special tax preferences, section 123.4, as intended by Parliament, lowered the tax rate to 15%, which Parliament intended to be the highest non-refundable tax rate levied under the Act.

Finally, the application of s. 250(5.1) accorded with its rationale, which was “to equate the place of continuance of a corporation with its place of incorporation” (para. 229).

In early 2008, the sale of petroleum and natural gas (PNG) rights by an arm’s length vendor (Transglobe) to the parent (Direct Energy) of the appellant (DEML) was structured on the basis that Transglobe transferred 99% and 1% of the PNG rights, at a nominal elected amount pursuant to a s. 85(1) rollover, to two wholly-owned Newcos (137 and 138, respectively), which then transferred the PNG rights on an s. 97(2) rollover basis to a newly-formed partnership (DERP2). Direct Enery then acquired the shares of 137 and 138 for $51 million and $0.5 million, respectively.

A year later, Direct Energy transferred the shares of 137 to DEML on an s. 85(1) rollover basis, with 137 then distributing its partnership interest in DERP2 to DEML on its winding up, with the ACB of that partnership interest being bumped under s. 88(1)(d), and with associated CCOGPE pools flowing up to DEML.

DERP2 then distributed its resource properties to DEML as a return of capital, thereby increasing the COGPE balance of DEML and reducing the ACB of DEML’s partnership interest by the FMV of the rights (higher than the value a year earlier) – but with these items effectively being approximately reversed at the partnership year end as a result of DERP2’s proceeds of the distribution of the PNG rights being allocated to its partners.

After then seeding DERP2 with a small resource property that was of interest to a third-party purchaser, DEML sold its partnership interest to that purchaser, thereby realizing a capital loss, which it then carried back to offset a capital gain it had previously realized.

In confirming CRA’s GAAR assessment to deny the capital loss, Russell J stated (at paras. 47, 56):

Here the substantial Capital Loss was claimed where there was no economic loss or impoverishment, thus per Triad Gestco breaching the OSP [object, spirit and purpose] of the Act’s capital loss provisions, including paragraph 39(1)(b). …

As the purpose of the capital loss provisions is to recognize real losses, there is clear abuse where artificial losses are deducted. That is even more so when those losses are based on non-capital CRP [Canadian resource property], that will also be deducted through CCOGPE pools at a 100% inclusion rate thus creating a double deduction.

Similarly, regarding the use of the s. 88(1)(d) bump to effectively create most of the quantum of the loss, he stated (at para.70):

[I]t seems incomprehensible that an artificial loss would signal misuse of capital loss provisions of the Act without equally indicating misuse of the very “bump” provisions of the Act used to achieve the artificial loss through the “bumping” of an ACB.

In September, 2007, most of the equity of an insolvent public corporation (“Xillix”) was acquired by the company (“Nexia”) of two individuals involved in acquiring and selling loss companies. As a result, Nexia held all of non-voting common shares of Xillix (representing 80% of its equity), 45% of its voting common shares and a non-interest-bearing demand loan owing by Xillix (now called “Biomerge”). The individuals then identified an income fund (“Total”) which was becoming subject to tax under the “SIFT” rules and was interested in accessing the Biomerge losses. In May 2009, a plan of arrangement was implemented under which the Total units were exchanged for new common shares of Biomerge, the existing voting common shares of Biomerge were largely cashed-out, and Total was wound-up into Biomerge (now, “New Total”) pursuant to s. 88.1(2). The former Total unitholders held 99.8% of the equity of New Total.

In following Deans Knight (as well as MMV) in finding that these transactions were an abuse of s. 111(5), Pizzitelli J stated (at para. 112):

[T]he reality of what happened here is that a willing seller in the business of selling tax attributes of failed companies takes the reins of such a company and markets and sells them to a willing unrelated buyer for use against their income. If these are not the type of transactions Parliament sought to stop by the enactment of the loss streaming rules in s.111(5) and parallel provisions, I don’t know what are.

In rejecting the New Total position (quoted at para. 114) that while “there were detailed rules in subsection 111(5) that dealt with the streaming of losses for corporations there were no such rules or evident policy related to the streaming of losses of trusts,” he noted inter alia that the transaction entailed the acquisition of a corporate lossco that effectively was merged with the corporate operating subsidiary of Total and that the subsequent introduction of s. 256(7)(c.1) “serves to clarify the policy as well as provide automatic denial of such losses rather than resorting to the GAAR” (para. 126).

The appellant (“MPP”) was an insolvent, publicly traded, mining company with accumulated net capital losses of $72.7 million. In order for two companies (“Madison” and “Vanac,” which dealt with each other and MPP at arm’s length) to access those losses and shelter gains from portfolios of rental properties, transactions were implemented, which first entailed the existing MPP mining business being transferred to a subsidiary, whose shares were effectively spun-off to the existing shareholders. Now that MPP was an empty shell, Madison and Vanac transferred respective portfolios of rental properties to MPP in consideration for the assumption of liabilities and for the issuance of a mixture of Class B voting shares and Class C non-voting shares (with the same attributes other than being generally non-voting) so that Madison and Vanac collectively held (and in equal proportions, after giving effect to some catch-up transactions to equalize those holdings) 46.6% of the voting rights and 92.8% of the equity of MPP. This transaction deliberately overvalued the shares that were so issued by MPP to Madison and Vanac so as to effectively transfer around $2.8 million of equity value to the existing public Class B common shareholders of MPP, thereby paying them for the losses.

After finding that MPP had received a tax benefit, and regarding the abuse test under s. 245(4), Graham J concluded (at para. 185):

Subsection 111(4) is supposed to prevent a corporation from being acquired by unrelated parties in order to deduct its unused net capital losses against new capital gains for the benefit of its new shareholders. The series of transactions completely frustrated that purpose.

In this regard, after having noted (at para. 170) that “the majority in Deans Knight highlighted that, while there had been no acquisition of control, there had been ‘the functional equivalent of such an acquisition of control’ by the company who effected the series of transactions (Matco)”, he indicated that:

• “[T]the Madison-Vanac Group fundamentally transformed the Appellant” (para. 184)
• “They structured the series of transactions in a way that ensured they would receive substantially all of the benefit from the application of those losses to a completely new business” (para. 184).
• “[T]hey selected the share compensation that they received in a way that ensured that, absent very unlikely circumstances, they could control the Appellant as if they had de jure control without actually taking that control” (par. 184). (For instance, taking into account the Class B shares held by friendly parties, such as directors, they effectively had more than half the voting rights and, conversely, a significant portion of the public shareholders did not exercise their voting rights.)

Before a Canadian public corporation (“Husky”) paid a dividend on its shares, two significant shareholders of Husky resident in Barbados (the “Barbcos”) transferred their shares under securities lending agreements to companies resident in Luxembourg with which they did not deal at arm’s length (the “Luxcos”). On payment to the Luxcos of the dividends on those shares, Husky withheld at the Luxembourg treaty-reduced rate of 5% (based on the Luxcos being the beneficial owners of the dividends and controlling at least 10% of the voting power in Husky).

After finding that Husky was liable under s. 215(6) for not having withheld at the non-Treaty rate of 25%, Owen J went on to consider the GAAR assessments of the successors to the Barbcos for the difference between the 15% withholding tax they would have borne without the securities loans, and the claimed rate of withholding at the 5% rate.

As to whether there were avoidance transactions, he rejected submissions that the transactions were carried out primarily to avoid the risk of Barbados tax on the dividends, and found that the purpose of the arrangements was primarily to reduce Part XIII tax.

The transactions were not an abuse, because they did not reduce Part XIII tax, and instead increased the rate from 15% to 25%. However, if for completeness, one assumed that the conditions for the 5% rate under the Barbados Treaty had been satisfied, then under this hypothesis there would appear to be no abuse. He stated (at paras. 376, 416):

Given the absence of any rule in Article 10 or elsewhere in the Luxembourg Treaty to supplement the residence requirement, the beneficial owner requirement, and the voting requirement, it is reasonable to conclude that Canada and Luxembourg were satisfied with the protection against “conduits” and flow-through arrangements afforded by the inclusion in Article 10(2) of those requirements. In other words, the true intentions of Canada and Luxembourg are fully reflected in the scope of the concepts of residence, beneficial owner and voting power adopted in Article 10(2). …

Consistent with the theory of economic allegiance described by the majority in Alta Energy, which recognizes that a recipient of passive income need not have any allegiance to the paying country, the focus of the rationale of Article 10(2) is not how the common shares of Husky came to be owned by the Luxcos, but whether the Luxcos satisfy the residence requirement, the beneficial owner requirement and the voting power requirement.

In December 2005, the appellant (“Québecor”) held 54.72% of the shares of another holding company (“Québecor Media”) which, in turn, held all the common shares of a third holding company (“366”) directly and preferred shares of 366 through a fourth holding company (“9101”). The common shares of 366 had an adjusted cost base (“ACB”) of $400 million and a nil fair market value (“FMV”) and the preferred shares of 366 had an FMV, ACB and paid-up capital (“PUC”) of $195.6 million, $166.5 million and $166.5 million, respectively. 366 held shares of an operating subsidiary with a net accrued capital loss of approximately $200 million.

In December 2005:

1. The PUC of the preferred shares held by 9101 in 366 was reduced to a nominal amount.
2. Québecor transferred its common shares of Abitibi Consolidated Inc. (“Abitibi”), having an FMV of $191.8 million and a nominal ACB, to 366 on a s. 85(1) rollover basis in exchange for preferred shares of 366.
3. 366 then redeemed such preferred shares in consideration for issuing a note to Québecor, and repaid the note by transferring the Abitibi shares back to Québecor.
4. On a winding-up of 366, all its assets were transferred to 9101. Given that s. 69(5) applied to this winding up, it thereby realized a capital loss on its shares of its subsidiary, which more than offset the capital gain realized by it on its transfer of Abitibi shares in 2 above, and was deemed by s. 84(2) to pay a dividend to 9101. Furthermore, Québecor Media thereby realized the accrued capital loss on its common shares of 366. S. 112(3) did not apply to deny this capital loss given that only 9101, not Québecor Media, received a deemed dividend on the winding-up.

In finding that the transfer of the Abitibi shares on a rollover basis to 366 did not constitute an abuse of the object and spirit of s. 85(1), Ouimet J stated (at para. 280, TaxInterpretations translation) that “this provision allows two related corporations ... to transfer a capital gain to be realized on a property from one corporation that does not have a capital loss to another that does have a capital loss, so that the latter can deduct it from the capital gain to be realized on the transferred property.” Although he accepted (e.g., at para. 281) that the object of s. 85(1) was to provide for the deferral rather than the elimination of tax, here the step-up of the ACB of the Abitibi shares in the hands of Québecor entailed the receipt by it of a deemed dividend that was included in its income (albeit, eligible for the s. 112(1) deduction), and the capital gain realized by 366 on disposing of the Abitibi shares was included in computing its income (para. 293).

In finding that it was not an abuse of ss. 88 and 69(5) for 366 to have been wound-up on a non-rollover basis so as to realize a capital loss, Ouimet J noted (at para. 274) that the scheme of s. 88 was to provide for both wind-ups on a rollover basis under s. 88(1) and for wind-ups on a taxable basis under ss. 88(2) and 69(5) and stated (at para. 273) that the specific exclusions of stop loss rules and other anti-avoidance rules in ss. 69(5)(c) and (d) “indicates the intention to allow, in the wound-up company's final year, the recognition of losses that would otherwise not have been realized under the specific anti-avoidance rules.”

Regarding the realization of capital losses at two levels (by 366 and by Québecor Media), Ouimet J stated (at para. 242) that “there is no provision that prevents a capital loss at two levels to the extent that the ACB of the shares exceeds the proceeds of disposition of the shares at both levels.”

The appeal was allowed.

The taxpayer (“KQI”), which was a resident Canadian corporation engaged in an elevator and escalator installation and repair business and which was an indirect subsidiary of a Finnish corporation (“Kone Corporation”) assisted in the financing of the acquisition by the Kone group in 2001 of two targets with complementary businesses by using money lent to it at interest (and advanced to it as share subscription proceeds) by another group corporation (whose Canadian parent had borrowed such funds in the European mid-term note market) to purchase, for a cash purchase price of $394 million, cumulative preferred shares of “Kone USA” (a group company with an active business) from the non-resident affiliated company (“Kone BV”) to which such shares had recently been issued as a stock dividend. At the same time, KQI agreed to resell such preferred shares in two tranches, and at pre-agreed higher prices, to Kone BV in three and five years’ time, which in fact occurred. The gain arising under this resale was deemed to be a dividend under the Quebec equivalent of ITA s. 93 which, along with the cumulative dividends received in the interim, came out of exempt surplus of Kone USA. The funds so received by Kone BV were used indirectly to fund the target purchases.

The ARQ sought to impute interest income to KQI under TA s. 127.6, the Quebec equivalent of ITA s. 17(1), on the basis that the above “repo” transaction was a sham that should instead be characterized as an interest-free loan by KQI to Kone BV or, alternatively, that the repo transaction represented an abusive avoidance of such s. 17 equivalent for Quebec GAAR purposes.

After rejecting the sham argument, Fournier JCQ went on to reject the application of the Quebec GAAR. He first found that the above repo transaction gave rise to a tax benefit and a tax avoidance transaction (noting, in the latter regard, at para. 181, TaxInterpretations translation) that “KQI served only as an instrument utilized by Kone Corporation to thus permit the acquisition of the Targets by way of a financing which was required to be the most advantageous possible from a tax viewpoint.”

However, in finding that the required element of abuse of the s. 17-equivalent rule had not been established, he noted (at para. 196) that such rule “contemplated blocking the exporting of income and preventing Canadian corporations from using their capital outside Canada by means of loans or advance not bearing a reasonable rate of interest and which remains unpaid for more than one year,” whereas here, no such loan or advance had occurred and that KQI had “instead acquired from Kone BV the shares of Kone USA, which it had agreed to hold for a certain passage of time and to then resell them” (para. 198). Accordingly, the repo transaction “accorded with the object and spirit of TA section 127.6” (para. 202).

The taxpayer, agreed to sell his shares of a Canadian-controlled private corporation (“Créaform”), having a nominal ACB, to third parties. On the day of the closing of the sale, about two months later, the following transactions occurred:

• He donated ½ of his shares, having an FMV of approximately $1M, to his wife, with s. 73(1) applying for this to occur on a rollover basis.
• He assigned the other ½ of his shares to her in consideration for a promissory note of approximately $1M, and elected out of the application of s. 73(1), so that he realized a capital gain of approximately $1M and her ACB increased by this amount.
• She completed the sale of all the shares to the purchaser and used ½ the proceeds to repay the note.

As a result of the ACB-averaging under s. 47(1), she realized a taxable capital gain (of around $250,000) on the sale of the ½ of the shares which she had purchased at full cost, so that the exception in s. 74.5(1) for FMV purchases allowed her to retain that half of the taxable capital gain and claim the capital gains deduction under s. 110.6(2.1), rather than such gain being attributed to him. The other half of the taxable capital gain realized by her (from her sale of the donated shares) was attributed to him pursuant to s. 74.2(1).

After finding that the sale of ½ of the taxpayer’s shares to his wife for a note clearly was a tax avoidance transaction, Favreau J went on to find that there was an abuse under s. 245(4), noting in this regard (at para. 56) that the facts were “very similar” to those in Gervais, where Noël CJ had stated (at para. 51) that the result of the spouse retaining half of the capital gain realized by her due to ACB averaging was “contrary to the object, spirit and purpose of subsections 73(1) and 74.2(1), the purpose of which is to ensure that a gain (or loss) deferred by reason of a rollover between spouses or common-law partners be attributed back to the transferor.”

Accordingly, the s. 245(2) assessment of the taxpayer to include all (rather than ½) of the taxable capital gains in his hands was confirmed.

In January 2002, a New Zealand “Buyco” (DHNZ) in the Danone group had acquired a NZ target company. In March 2003, in connection with the (planned-since-February 2002) refinancing of the 2002 acquisition, Deutsche Bank advanced $204 million (the maximum amount permitted under the NA thin capitalization rules) to DHNZ in exchange for a convertible note redeemable at maturity in five years’ time at Deutsche Bank’s election by the issuance of a specified number of non-voting shares in DHNZ. Interest on the advance was payable semi-annually in arrears at 6.5% per annum (amounting to $66 million over the term).

Contemporaneously with the advance, the Singapore immediate parent of DHNZ (DAP) paid Deutsche Bank $149 million pursuant to a forward purchase agreement to acquire the shares to be received on maturity of the note. It was exempt under Singapore on its resulting gain (the $55 million excess of the value of the shares received over the advance purchase price.

S. BG1(1) of the Income Tax Act 2004 (NZ) provided that a tax avoidance arrangement (defined to include an arrangement that has “tax avoidance as its purpose or effect … [or] as 1 of its purposes or effects … if the purpose or effect is not merely incidental”) was void as against the Commissioner. However, Ben Nevis had found (at para. 108, quoted at para. 53):

The ultimate question is whether the impugned arrangement, viewed in a commercially and economically realistic way, makes use of the specific provision in a manner that is consistent with Parliament’s purpose. If that is so, the arrangement will not, by reason of that use, be a tax avoidance arrangement. If the use of the specific provision is beyond parliamentary contemplation, its use in that way will result in the arrangement being a tax avoidance arrangement.

William Young J indicated that the economic substance of the arrangements was that Deutsche Bank advanced only $55 million to DHNZ (being the difference between the $204 million advance and the $149 million paid by DAP under the forward purchase agreement) and that the $66 million paid by DHNZ to Deutsche Bank amounted to repayment of that $55 million and interest on an amortizing basis. In particular, he stated (at para. 76):

In economic substance, Deutsche Bank advanced $55 million to DHNZ which was fully repaid on an amortising principal and interest basis over the term of the note. There was never any alteration of substance in relation to the ownership by DAP of DHNZ. DAP started off owning 100 per cent. And as was always intended, it wound up owning 100 per cent. The net effect of everything that happened was that the advance of $55 million was repaid on a basis that ostensibly permitted DHNZ to deduct all payments made, including repayments of principal.

In finding that there was thus a tax avoidance transaction (so that the Commissioner appropriately treated $55 million of the “interest” payments as being non-deductible), he further stated (at para. 86):

[T]he effect of the arrangement was that DHNZ sought to obtain deductions in relation to $55 million in principal repayments. These are provided for in the Act to meet financing expenses and not repayments of principal. DHNZ was thus claiming deductions for expenses which, in economic substance, it had not incurred. This use of the relevant deduction provisions of the Act lay outside of parliamentary contemplation as to the use of those provisions.

The taxpayer, which had been assessed under s. 245(2) to deny the carryforward of losses and credits, sought the discovery of proposals made by CRA to two unrelated taxpayers that set out its understanding of the facts and its legal analysis thereof. The CRA auditor had considered such documents (which he had placed in the file) but had not relied on them in auditing the taxpayer.

In finding that they were discoverable, Owen J stated (at paras. 39, 60):

[I]n GAAR cases, the legal analysis of the Minister in support of the policy relied upon is subject to discovery. …

[R]eliance is not the test for relevance. … [C]onsideration of the documents in the context of the audit of the Appellant is sufficient to make them relevant for the purposes of discovery.

CRA had relied on the GAAR Committee’s analysis of a similar case in deciding to assess the taxpayer under GAAR, so that a GAAR Committee review of the taxpayer’s transactions was considered unnecessary.

After noting (at para. 85) that “[i]f the GAAR Committee had considered the Appellant’s case, there is no doubt that the Appellant would be entitled to discovery of all non‑privileged documents considered by the GAAR Committee in deciding to assess the Appellant under the GAAR,” Owen J stated (at para. 86):

[T]he Appellant is equally entitled to all non-privileged documents considered by the GAAR Committee in deciding to assess under the GAAR the unrelated taxpayer described in the Similar Case … because that decision directly resulted in the subsequent decision to assess the Appellant under the GAAR.

Accordingly, such documents were discoverable, subject to redaction of all information identifying third parties, and subject to any claims of solicitor-client privilege.

Following preliminary transactions, on June 28, 2004, a Canadian holding company (“Micsau”) held all the shares, having a fair market value (“FMV”) and adjusted cost base (“ACB”) of $101.8M and $48.1M respectively, and a nominal paid-up capital (“PUC”), of a holding company (“3295940”), which held a minority shareholding (having an FMV of $88.5M and an ACB of $4M, reflecting the crystallization of the safe income on hand of such shares, and a nominal PUC) in another holding company (“Holdings” – which was majority-owned by an non-resident arm’s length investor), which carried on a Canadian generic-pharmaceuticals business. (The reasons for judgment do not disclose the reason for the discrepancy between the $101.8M and $88.5M FMVs.) Micsau wished to take advantage of the high ACB of its shares of 3295940 by selling those shares to a third-party purchaser (Novartis), but Novartis insisted that the sale be of the shares of Holdings. In order to indirectly take advantage of the high ACB of Micsau’s shares of 3295940, the following transactions were implemented:

1. 3295940 effected a “dirty 85” exchange of its shares of 3295940 for Class D preferred shares of 3295940 having an ACB and FMV of $31.4M and for common shares having an ACB of $16.6M and higher FMV.
2. Micsau transferred its Class D preferred shares of 3295940 to a Newco formed by it (“4244”) in exchange for Class D preferred shares of 4244 having the same full FMV/ACB.
3. 3295940 transferred its shares of Holdings to 4244 on a partial s. 85(1) rollover basis in consideration for $57M of Class D preferred shares of 4244 (with $57M being the agreed amount, so that it thereby realized a capital gain of $53M) and for common shares having a nominal cost pursuant to s. 85(1)(h). As a result, the capital dividend account (“CDA”) of 3295940 increased by $26.5M to $32.3M.
4. 3295940 redeemed the Class D preferred shares in its capital held by 4244 for a $31.5M note, and elected for the resulting $31.5M deemed dividend to be a capital dividend paid to 4244.
5. 4244 redeemed 31.5M common shares (along with a small number of Class D preferred shares) held by 3295940 for a $31.5M note, and elected for the resulting $31.5M deemed dividend to be a capital dividend paid to 3295940.
6. The two notes referred to above were set off.
7. On August 12, 2004, Micsau transferred its shares of 4244 to 3295940 in consideration for 31.5M preferred shares of 3295940, with no gain being realized.
8. As a result of these transactions, 3295940 held 1 common share of 4244 with a nominal ACB and FMV and 88.4M Class D preferred shares of 4244 with an FMV and ACB of $88.4M. These shares were then sold to Novartis at no gain.

In confirming the addition of $31.5M to the capital gains realized by 3295940, Favreau J stated (at paras. 120-122, TaxInterpretations translation):

Through its elections, the appellant was able to avoid the application of subsection 55(2) on the cross redemption of its shares and the shares of 4244. Knowing that subsection 55(2) would have transformed the taxable dividend paid by 4244 on the redemption of the appellant's shares, the only way to avoid such a disadvantageous result was to circulate the CDA through a capital dividend, which was ultimately restored to its original position. The capital dividend was used in a way that is not consistent with its purpose: instead of allowing tax-free amounts to flow upwards in the corporate group, the amount circulated back to its starting point, in the hands of 3295940.

This recycling of the capital dividend prevented subsection 55(2) from applying and converting the deemed dividend paid by 4244 to 3295940 into a capital gain. The application of subsection 55(2) would have allowed the entire appreciation in value of the Holdings shares to be taxed, thereby preserving the integrity of the capital gains tax regime.

The highlight of this case is that the circulation of the CDA allowed the appellant to establish an interest in its newly incorporated company at a value equal to its fair market value, resulting in no tax being payable on the sale of the 4244 shares to Novartis. The appellant suffered only a portion of the disadvantages associated with the nominal tax basis of its interest in Holdings

Regarding the taxpayer’s submission that it was appropriate for it to effectively access the high ACB of Micsau in the shares of 3295940, Favreau J noted that, in light of s. 88(1)(d.2), Micsau could not have used s. 88(1)(d) to bump the low ACB of the shares of Holdings to 3295940, and stated (at para. 143):

[T]he appellant used the capital dividend mechanism in order to reduce the value of its shares and thus benefit from a reduced capital gain. Through the series of transactions, the appellant reached the same result as if there had been a bump of the shares held by it, but without having to satisfy the required conditions.

Regarding the submission of the taxpayer that (but for commercial roadblocks) it could have achieved a similar result by more directly using the high ACB of Micsau in 3295940 under three alternative transactions (e.g., amalgamated 3295940 with Holdings before a sale of the Amalco, or engaging in a “tuck under” transaction), Favreau J stated (at para. 159):

[I]t was demonstrated that the sale of Holdings' shares was paramount. That being the case, the alternative transactions involving the sale of 3295940 shares cannot be submitted for comparison, since the result is not necessarily the same, at least not in terms of economic reality. In fact, the three proposed alternative scenarios involved, in one way or another, the disposition of the 3295940 shares and not a sale of the shares of Holdings by 3295940.

The appellants engaged in transactions which were intended to result in the realization by them of substantial capital gains (resulting in additions to their capital dividend accounts), that were immediately distributed by them), followed by the realization of largely offsetting capital losses later that day. These transactions depended in part on the appellants being considered to have acquired units of an income fund ("FMO") from another income fund ("TOM") (which in turn had purportedly acquired the FMO units from an RRSP in consideration for issuing TOM units to it) at a cost equaling the units’ FMV, followed immediately by a distribution to them of capital gains that had been realized by FMO – with that distribution not reducing the ACB of their units by virtue of s. 53(2)(h)(i.1)(A) and (B)((I). Accordingly, such repurchases were intended to result in the realization of largely offsetting capital losses.

Smith J found that, in light, inter alia, of the very transitory nature of the appellants’ holding of the FMO units, that the appellants had not acquired their beneficial ownership, so that their purported disposition on their repurchase did not generate the capital losses. In going on to find, in the alternative, that GAAR would apply to deny the CDA additions, he stated (at para. 256):

The intent of Parliament was to ensure that the positive components of the CDA would be ‘added’ and that the negative amounts would be ‘subtracted’. The payment of a capital dividend by a corporation that has realized capital gains and capital losses in an equal amount, more or less, on the same day and before the capital dividends are declared, would defeat the “underlying rational of the provisions (…) in a manner that frustrates or defeats the object, spirit or purpose of those provisions” ... . It can be said that the “object, spirit and purpose” of the CDA regime is to ensure that it mirrors the tax treatment of capital gains for an individual and that the Minister can only seek to tax gains that give rise to ‘real’ economic gains. By the same token, only one half of the ‘real’ economic gains realized by a corporation can be added to the CDA.

In order that the taxpayer’s RRSP could indirectly invest in operating businesses in which he and/or two business colleagues had a management role, he instigated the formation of various unit trusts (the “Income Funds”) which were intended to be mutual fund trusts on the basis that 171 individuals (the “Investors”) - immediate and extended family members, friends, employees, business associates and others - each subscribed $750 for units of each Income Fund. The taxpayer’s RRSP (along, in two instances, with those of his two colleagues) then invested a large sum (e.g., over $150 million for one of the Income Funds) in subscribing for additional units. The Income Funds invested in underlying LPs carrying on businesses, usually through an intermediate sub trust and master limited partnership.

Smith J found that there had been a failure to satisfy the requirement in Reg. 4801(a)(i)(A) that there had been a “lawful distribution” of the units to the Investors in accordance with the offering memorandum exemption from a prospectus-filing requirement, given that a substantial number of the Investors were adults who did not pay for their own units, or minors. As the Income Fund units thus did not qualify as MFT units, the RRSP was subject to tax under s. 146(10.1) and penalty tax (under former s. 207.1(1)) on the basis of not holding qualified investments.

In the alternative, Smith J considered whether GAAR should apply even if the units in the Income Funds were qualified investments. After finding that there was a tax benefit and avoidance transactions (i.e., the establishment of the Income Funds through which the RRSP could invest in the businesses), Smith J also found that there was an abuse under s. 245(4), stating (at paras. 592-593, 600, 602):

[T]he object, spirit and purpose of subsection 146(4) is to prevent an annuitant from making tax deductible contributions (at great cost to the public treasury, at least in the short term) and then using those funds for business purposes and thus take advantage of the tax-exempt status of the plan.

… [I]ncome earned from qualified investments, being investments that are not “non-qualified investments”, will accrue on a tax-exempt basis but not so if the annuitant has somehow managed to use the contributions or accumulated assets in the RRSP to operate a business that is not at arm’s length. …

[T]he acquisition by the RRSP Trust of 99% of the units of the Income Funds defeated the object, spirit and purpose of the provision and was contrary to the Parliament intention that a mutual fund trust was to be widely held. It was certainly not within the contemplation of Parliament that a mutual fund trust that was a qualified investment for RRSP purposes would effectively become one investor’s alter ego. …

[T]he Appellant sought to abuse the RRSP regime and the provisions of the Act by establishing the Income Funds … .

After finding that the Minister’s assessment of the taxpayer respecting the RRSP income pursuant to s. 56(2) was unsupported by the wording of that provision, Smith J went on to indicate that he would have upheld the reassessments of the taxpayer in those amounts on the basis of the GAAR, but for this resulting “in a duplication of the tax which the Minister has also sought to impose on the RRSP Trust pursuant to subsection 146(10.1)” – which could not “be considered ‘reasonable in the circumstances’ as contemplated in subsection 245(5)” (para. 623). Accordingly, only the assessments of the RRSP under s. 146(10.1), and not of the taxpayer under s. 56(2), were sustained.

A venture capital corporation (“MMV Finance”) was issued, early in 2011, voting common shares of a corporation (“MMV”) (that was in interim receivership pursuant to the Bankruptcy and Insolvency Act and that had significant accumulated losses and investment tax credits) so as to acquire and hold approximately 49% of the outstanding voting common share and, under a proposal that was accepted by the creditors, lent $50,000 to MMV Finance in order for MMV Finance to settle in full its unsecured debts. In addition, MMV’s outstanding secured debts of US$850,000 were acquired by MMV Financial for a total of $100,000. MMV Finance then subscribed $1,000 for non-voting shares representing over 99.9% of the total outstanding common shares. In March and June of 2011, MMV purchased loan portfolios from MMV Finance for an aggregate purchase price of U.S.$86 million, which was paid with cash that had been lent to MMV by MMV Finance, and through the issuance of preferred shares. The parent of MMV Finance managed this portfolio for MMV. CRA ultimately accepted that MMV Finance had not acquired de jure control of MMV as a technical matter, but applied s. 245(2) to deny the use by MMV of its losses against the income generated by these portfolios.

In finding that such use of the losses was not an abuse, Bocock J stated (at paras. 170, 180, 184)):

Duha determined that Parliament’s aim in choosing a de jure standard was to further its purpose of achieving certainty and predictability when restricting loss utilization. … Parliament … deliberately kept the reference to de jure control in 111(5) instead of adopting a de facto standard. …

Evidence was not presented to show that the board did not have the actual authority to make material decisions on behalf of MMV … . No evidence showed that MMV Financial required de facto or effective control of MMV in order to make the utilization of the losses work. …

The presence of the longstanding, bright-line test of de jure control bears further witness to the rejection of applying the GAAR in the circumstances of this appeal as regards subsection 111(5).

To simplify somewhat, a group of private corporations owned for the benefit of the Rogers family structured their affairs such that one corporation (“CGESR”) was the beneficiary of policies on the life of Ted Rogers whereas a company (“ESRL”) holding shares of CGESR through another CCPC (“ESRIL 98”) was the policyholder and had been paying the premiums, so that its adjusted cost basis (“ACB”) was $42M. On the death of Mr. Rogers in 2008, the full proceeds of the policies ($102M) were added to the capital dividend account (“CDA”) of CGESR, which used such proceeds to pay actual and deemed capital dividends in that amount.

CRA considered that it was abusive for the group to benefit from the full $102M addition to the CGESR CDA rather than an amount net of the $42M ACB to the policyholder (ESRL). Sommerfeldt J.found that there was no “tax benefit” given that inter alia not all of these proceeds were fully distributed up the chain, so that, on the trial date, ESRIL 98 continued to have a CDA of approximately $42M.

However, he went on to find that there was no abuse under s. 245(4). In 1977, there had been a legislative change to reflect a policy that the addition to a corporate beneficiary’s CDA would now only be reduced by the policy’s ACB to it rather than by the ACB of the policy to any person. Accordingly, taking advantage of the fact that CGESR itself had a nil ACB for the policies, so that the bump to its CDA was $102M rather than $60M ($102M - $42M), accorded with the object and spirit of the provisions. Sommerfeldt J. stated (at paras. 131-132):

[T]his is one of those, perhaps rare, situations where the underlying rationale of the Reduction [for ACB] Provision in 2008 and 2009 was no broader than the text itself. …

The Crown has not persuaded me that in 2008 and 2009 the purpose of the Reduction Provision was to do anything other than to reduce the addition of life insurance proceeds to a corporation’s capital dividend account by the amount of the adjusted cost basis of the life insurance policy to that corporation.

Notwithstanding some “self-serving” language of Finance to the contrary, a 2016 amendment “clearly changed” the object and spirit of the provisions so that, from thence onwards, there was to be a reduction in the bump to the corporate CDA for the policyholder’s ACB (para. 105).

The “Tax Attributes” of a Lossco (the taxpayer) were effectively sold to arm’s length investors pursuant to transactions under which:

• The existing shareholders of Lossco exchanged their Lossco shares for “Newco” shares under a Plan of Arrangement
• A private company “facilitator” (Matco) acquired a debenture of the Lossco that was convertible into shares representing 79% of its equity shares but only 35% of its voting shares (this occurred before the introduction of s. 256.1). Matco also obligated itself to find a purchaser for the remaining shares of Newco for a pre-agreed price (failing which Matco itself would be required to pay that amount to Newco, without a right ot acquire those shares).
• Lossco transferred it assets to Newco for a note.
• Matco then identified a mutual fund management company who wanted to IPO Lossco and use the IPO proceeds (of $100M) for a new bond trading business to be carried on in Lossco.
• The IPO subscription price for the newly-issued common shares caused the securities of Lossco held by Newco and Matco to appreciate which, in the case of Matco, effectively was its fee.
• Newco sold its remaining shares of Lossco to Matco for the pre-agreed price notwithstanding that this was at a 20% discount to the shares’ current market value.

In finding that the utilization of the Tax Attributes did not constitute an abuse of the loss-streaming rules (and, in particular, of ss. 256(8) and 251(5)(b)), Paris J noted (at para. 132) that “Parliament’s aim in choosing the de jure control test was to achieve certainty and predictability,” and the stated (at paras. 134, 147):

It has been suggested that the acquisition of control test in subsection 111(5) is a reasonable marker between situations where the corporation is a free actor in a transaction and when it is only a passive participant whose actions can be manipulated by a new person or group of persons in order to utilize the losses or Tax Attributes of the corporation for their own benefit. I agree … and find that the object, spirit and purpose of subsection 111(5) is to target manipulation of losses of a corporation by a new person or group of persons, through effective control over the corporation’s actions.

A change of management, business activity, assets and liabilities and name are not markers of a change of effective control of a corporation under the common law and the change of shareholders that came about as a result of the IPO did not result in a change of effective control because presumably there was no common link between the post-IPO shareholders.

In allowing the appeal, he concluded (at para 166):

… [T]he circumstances referred to by the Respondent do not, in my view, indicate that Matco had effective control over the majority of the voting shares of the Appellant prior to the IPO and I find that the Avoidance Transactions do not amount to abuse of subsection 256(8) and paragraph 251(5)(b) of the Act.

The taxpayer, which carried on a commercial real estate business and had September 30 year ends, transferred on a s. 97(2) rollover basis a commercial real estate property to a newly-formed LP (the “Partnership”) on April 10, 2007. The Partnership closed the sale of the property to a third party at a gain on August 8, 2007 (before the first fiscal period end of the LP on October 1, 2007) and lent $24.5M to the taxpayer’s shareholder (“Shabholdings”). On September 26, 2007, the taxpayer was discontinued under the CBCA and continued under the BVI Business Companies Act, 2004 in order to cease to be a Canadian-controlled private corporation and to not be subject to additional refundable taxes under s. 123.3.

On September 28, 2007, the Partnership distributed $24.6M to the taxpayer, so that a negative ACB capital gain to the taxpayer of $24.3M resulted on October 1, 2007 pursuant to ss. 53(2)(c) and 40(3.1) and (before CRA’s application of s. 245(2)) produced an increase of $12.2M to the taxpayer’s capital dividend account (“CDA”). (The CDA definition was subsequently amended to eliminate such an addition.)

Immediately after the Partnership’s first fiscal year end, the taxpayer’s share of the capital gain from the sale increased the ACB of its Partnership interest by $24.3M, and its CDA by ½ thereof, or $12.2M (or, in fact, $11.7M taking Partnership operating losses into account). The taxpayer then paid a capital dividend to Shabholdings of $23.8M through the distribution of a portion of the Shabholdings note. In its tax return for its 2008 taxation year, the taxpayer elected under s. 40(3.12) to realize a capital loss equal to the amount of the s. 40(3.1) capital gain, so that it avoided paying income tax on that gain.

Hogan J concluded (at para. 85), in confirming the Minister’s application of s. 245(2) to reduce the taxpayer’s CDA by ½ the amount of the s. 40(3.1) capital gain (thereby generating Part III tax subject to a s. 184(3) election being made):

[T]he GAAR applies because the Avoidance Transactions were specifically designed to achieve a result that was, in the case of subsections 40(3.1) and 40(3.12), inconsistent with the rationale underlying each of those provisions and equally inconsistent with the rationale of the provisions that form part of the CDA Mechanism … .

Respecting the second inconsistency, he had stated (at para. 42) that “consistent with the principle of integration, the CDA Mechanism was adopted to ensure that only one-half of a capital gain would be subject to income tax if the gain was realized indirectly by a private corporation,” whereas (para. 86):

Here, the Appellant achieved a result which led to significant over-integration and, but for the application of the GAAR, would have allowed the Appellant to pay a capital dividend equal to the entire capital gain realized from the sale of the Property.

Respecting the first inconsistency, “the purpose and effect of subsection 40(3.1) are to dissuade taxpayers from extracting from a partnership on a tax-free basis funds in excess of their investment in the partnership” (para. 58) and “subsection 40(3.1) was not intended to encourage taxpayers to deliberately trigger its application for the purpose of creating a capital gain which through intentional planning, would soon be offset by an offsetting capital loss by means of an election under subsection 40(3.12)" (para. 59). Thus s. 40(3.1) “and the alleviating rule in subsection 40(3.12) were not enacted to encourage taxpayers to deliberately create offsetting gains and losses for the purpose of inflating their CDA” (para. 67).

Three affiliated Quebec corporations (Développements, Location and Estrie) would have realized gains totaling $728.9 million (plus recapture of depreciation of $67.1 million) on their sale of various Quebec real estate properties. They transferred those properties on a rollover basis to limited partnerships controlled by them, and then transferred their units of those LPs to the two appellant numbered companies, also on a rollover basis. The two numbered companies, selected February 28, 2006 as their taxation year-end for federal (and Ontario) tax purposes, but March 19, 2006 for Quebec taxation purposes. On March 1, the two numbered companies then acquired units in two Ontario LPs with business income, which had the effect of most of their income being allocated to Ontario for Quebec purposes in accordance with the formula in Rule 771R3, so that virtually no Quebec tax was payable on the above gains, which were realized in March between the two year ends. For the year ending February 28, 2006, the two numbered companies declared no capital gains or recapture of depreciation.

The Court of Quebec found that establishing different year ends for provincial and federal purposes was contrary to the purpose of the Taxation Act, s. 7, which defined a company’s taxation year as well as Rule 771R3, whose purpose was to ensure an equitable distribution of the tax burden on companies carrying on business in more than one province, and also found that the rollover transactions abused the legislative intent of the rollover provisions, which was to defer and not to eliminate tax. Accordingly, the avoidance of the Quebec tax was reversed through the application of the Quebec general anti-avoidance provision to treat the year end for Quebec purposes (February 28) the same as it was federally and in Ontario.

All three findings were confirmed by Schrager JA. In confirming the finding on Rule 771R3, he stated (at para. 48, and in disagreeing with what he characterized as obiter on a similar point in Veracity):

The rationale [of Rule 771R3] is to allocate income amongst the provinces where a corporation carries on business. This is done in order to achieve equitable taxation, meaning that not more than 100% of total income should be taxed. The same rational dictates that no less than 100% of the income should be taxed.

Respecting s. 7, he stated (at para. 52, 55-56):

The definition of fiscal period essentially copied from the federal legislation some years ago cannot in context be characterized as tax policy and certainly not a policy to treat capital gains differently from other provinces or from the federal government. The [definition] … by no means speaks to specifically allowing different financial year-ends for Quebec purposes.

The BCCA [in Veracity] … conclude[s] that because there is no uniform system of provincial taxation in Canada, then something may well “fall through the cracks. Such reasoning may well apply where provinces ascribe different tax treatment to the same category of revenue … .

[However, here] the income, or capital gain is treated the same in Quebec, Ontario and by the federal government for tax purposes so that upon application of the allocation formula, 100% of the income must result in taxation somewhere as the Quebec Court judge correctly decided.

Respecting the abuse of the rollover provisions, he stated (at paras. 61-62):

[T]he Appellants transferred their business to Ontario knowing that because of the creation of two fiscal periods, tax would not be paid there …[and] knew that no tax would be payable in Quebec because (theoretically) it was payable in Ontario upon application of the allocation formula, but because of the different fiscal periods, no tax would ultimately be paid in Ontario.

Thus, the rollover provisions have been used, as in OGT Holdings, to avoid the payment of tax and not simply defer its payment. ln this manner, the Appellants have acted contrary to the object and spirit of [the rollover provisions].

The taxpayer, an indirect wholly-owned subsidiary of Loblaws (a Canadian public company), wholly-owned a Barbados subsidiary (GBL), that was licensed in Barbados as an international bank and that used funds mostly derived from equity injections by the taxpayer to invest in U.S.-dollar short-term debt obligations, loans to several thousand independent U.S. distributors of Weston baked goods and intercorporate loans – and entered into cross-currency and interest rate swaps with an arm’s length bank to effectively convert much of its income stream into fixed rated Canadian-dollar interest. CRA assessed the taxpayer on the basis that GBL had realized $473 million of foreign accrual property income (FAPI) between 2001 and 2010.

C Miller J confirmed the assessments on the basis that the business was conducted principally with the Loblaw group (i.e, it was not conducted principally with arm’s length persons, as required under para. (c) of the exclusion from an investment business as defined in s. 95(1).)

It thus was unnecessary to consider “misuse or abuse” under s. 245(4). However, he indicated (if application of s. 245 had been relevant and if, contrary to his finding, there had been an avoidance transaction) that there would have been such misuse, stating (at paras. 322-323):

The policy, or underlying rationale, of the exemption … is to promote competition of affiliates operating in international markets. …

[I]t follows that Loblaw Financial was misusing this exemption as it was not competing in any manner in any international market. It basically managed an investment portfolio for Loblaw.

In further rejecting a submission that the specific anti-avoidance rule in s. 95(2)(l) “makes clear that Parliament made an explicit policy choice to exclude foreign subsidiaries of Canadian regulated financial institutions” (para. 324), he noted that the s. 95(2)(l) exception was there to permit non-resident subsidiaries of Canadian banks and stock brokers to compete with foreign financial institutions (“which emphasizes the requirement for third party business” (para. 325), whereas GBL was merely “trading for proprietary purposes.”

When the taxpayers’ corporation (“Opco”), a small business corporation, was in financial difficulty, a Quebec regional development fund agreed to inject equity capital in Opco on terms dictated by the fund – which entailed the fund investing in the common shares of Opco, so that the paid-up capital of the taxpayers’ shares was stepped-up from a nominal amount to $1.45 million (as a result of the aggregate class PUC being averaged on a per share basis). About five years later, the taxpayers engaged in capital gains crystallization transactions in which they transferred most of their common shares of Opco to personal holding companies, realizing capital gains of $1 million, and took back preferred shares with a correlative adjusted cost base and also a paid-up capital that reflected the earlier step-up in the transferred shares’ PUC. Following the repurchase of all of the fund’s common shares, the taxpayers had their Holdcos distribute most of the PUC of their preferred shares in cash.

The ARQ considered there to have been abusive surplus-stripping, and applied the Quebec general anti-avoidance rule to treat most of the paid-up capital distributions as taxable dividends. After finding that there was no avoidance transaction, Dortélus JCQ went on to find that there was no abusive tax avoidance. He noted that the onus was on the ARQ to demonstrate an abuse (para. 83), and accepted the taxpayers’ submission that Pomerleau and 1245989 were distinguishable on the basis that in those two decisions “the surpluses were stripped as part of 'internal' transactions between individuals and corporations not dealing at arm’s length, which is not the case here, as there is an arm’s length relationship between the plaintiffs and [the fund]” (para. 86, TaxInterpretations translation).

A Blackstone LP and a U.S. shale company held their investment in a Canadian subsidiary (Alta Canada), that was to develop a shale formation in northern B.C., through a Luxembourg s.à r.l. About two years after the acquisition by Altas Canada of the exploration licences, it was sold to Chevron Canada at a significant gain. Hogan J found that the gain of the s.à r.l.was exempted from Canadian capital gains tax by virtue of the exclusion in Art. 13(4) of the Canada-Luxembourg Treaty which provided that the Alta Canada shares were not deemed immovable property (and thus not subject to Canadian capital gains tax) if the Alta Canada licences qualified as property of Alta Canada “in which the business of the company … was carried on.”

In rejecting the Crown’s GAAR argument that accessing the Treaty exemption was abusive in that “the rationale and purpose of the Treaty is to prevent or reduce double taxation on activities or transactions that potentially may be subject to tax in both Contracting States at the same time’’ (para. 76) whereas here, there was tax in neither state, Hogan J stated (at para. 85):

[I]f Canada wished to curtail the benefits of the Treaty to potential situations of double taxation, Canada could have insisted that the exemption provided for under Article 13(5) be made available only in the circumstances where the capital gain was otherwise taxable in Luxembourg. Canada and Luxembourg did not choose this option. It is certainly not the role of the Court to disturb their bargain … .

He also stated (at paras. 91, 100):

There is nothing in the Treaty that suggests that a single purpose holding corporation, resident in Luxembourg, cannot avail itself of the benefits of the Treaty. There is also nothing in the Treaty that suggest that a holding corporation, resident in Luxembourg, should be denied the benefit of the Treaty because its shareholders are not themselves residents of Luxembourg. …

The rationale underlying the carve-out is to exempt residents of Luxembourg from Canadian taxation where there is an investment in immovable property used in a business. The significant investments of the Appellant to de-risk the Duvernay shale constitute an investment in immovable property used in a business.

In order to reduce Quebec capital tax, the taxpayer lent $350M on a non-interest-bearing basis to its parent on January 27, 2005, and claimed this amount as an investment allowance in computing its capital tax liability at its taxation year end of January 31, 2005. On February 15, 2005, the loan was repaid so that the parent did not account for this loan in its capital when it, in turn, calculated its capital tax liability. In confirming the application of the Quebec general anti-avoidance rule to deny the benefit to the taxpayer of the investment allowance, Vézina JCA stated (at paras. 29, 35, TaxInterpretations translation) that the reduction under the investment allowance:

is intended to “avoid double taxation” and is not intended to exclude all taxation.

…The reduction did not serve to rectify a prejudice, but created a benefit, which is contrary to the object and spirit of this disposition of simply addressing double taxation and not creating a tax exemption.

A newly-launched public corporation ("Birchcliff") sought to access the losses of a lossco ("Veracel") in order to shelter the profits from producing oil and gas properties which it was acquiring. Rather than financing the properties directly, private placement investors were told that they could subscribe for subscription receipts of Veracel instead, which was not a problem to them because their subscription receipts would be converted into Veracel Class B common shares as a transitory step under a Plan of Arrangement in which Veracel was then amalgamated with Birchcliff. As the investors received a majority voting equity interest in Amalco, the loss streaming rules otherwise engaged by ss. 256(7)(b)(iii)(B) and 111(5)(a) were avoided.

In finding that there was an abuse under s. 245(4), so that there was an acquisition of control of Veracel, Jorré J stated (at paras. 135-136):

The statutory provision itself shows that the policy underlying the provision is that the “minority” predecessor, i.e. the predecessor whose shareholders, were they acting in concert, would not have control in the amalgamated corporation, will lose its tax attributes, its losses, on an amalgamation with another corporation. It does not include what might be described as contingent shareholders like the Class B shareholders here.

The artificial insertion of Class B shareholders of Veracel, persons whose shares’ only purpose was to be converted into common shares, and whose shares had such a short existence that they had to be deemed by the plan of arrangement to be created before the amalgamation, is a manipulation of the shareholdings of a predecessor contrary to the object of the rules in subsection 256(7).

An individual did not sell his shares of an operating company (Groupe AST) directly to a third-party purchaser. Instead he rolled his shares into a holding company (276), following which some internal transactions occurred in which the adjusted cost base of the Groupe AST shares was stepped up to fair market value - including a non-rollover drop-down of those shares to a subsidiary (9144) in exchange for high-basis common shares - with 276 realizing corresponding capital gains. The Groupe AST shares were then sold to the purchaser at no additional gain.

276 then engaged in “value shift” transactions of the same general type as were struck down in Triad Gestco and 1207192, i.e., a stock dividend of high-low preferred shares was paid on the high-ACB common shares that 276 held in 9144, thereby rendering those common shares almost worthless, and then the capital loss was realized by selling those common shares for $1 to a corporation owned by the son of 276’s shareholder.

Although the realization of the capital gain – to be offset by the value-shift loss – was realized in internal transactions, unlike Triad Gestco and 1207192, this was not a relevant difference.

After quoting with approval from Triad Gestco, Paris J stated (at para. 56, TaxInterpretations translation):

For the same reasons, there is an abuse with respect to the transactions in this case having regard to the ITA.

$6.3 million in funds was extracted from a private operating company (“Gennium”) by paying those dividends to a holding company (“9134”), which contributed those funds to a trust to which s. 75(2) applied (namely, the taxpayer, or “Satoma Trust”), which then contributed those funds to another holding company (“9163”), which then paid those funds back to Satoma Trust as dividends. Although the funds thus landed in Satoma Trust, which it used to make new investments, those dividends were attributed under s. 75(2) to 9134, which used the dividends-received deduction in s. 112(1) to avoid any inclusion in its taxable income. Lamarre ACJ noted (at para. 24) that the result of the above transactions was that “9134 was stripped of its assets in favour of the appellant without any payment of tax.” Although this description of the steps may be sufficient, they are described in somewhat more detailed (but still summary) fashion below.

1. Gennium paid dividends to a family trust (Louis Pilon Family Trust) to which s. 75(2) did not apply.
2. That trust distributed the dividends to a corporate beneficiary (“9134”), which excluded the dividend amounts from its taxable income under s. 112(1).
3. 9134 then made a $100 gift to the taxpayer (Satoma Trust), which used those funds to acquire Class F shares of another corporation (9163) that did not have an active business. The gift was intended to engage the application of s. 75(2), so that dividends on the shares would be attributed to 9134.
4. 9134 made capital contributions of the dividends amounts received by it in 2 to 9163.
5. 9163 then paid those funds to Satoma Trust as dividends. Those dividends were not required to be included in the income of Satoma Trust as s. 75(2) applied to include them in the income of 9134, which claimed the s. 112(1) deduction to exclude them from its taxable income.

CRA assessed under s. 245(2) to include the dividends paid in 5 in the income of Satoma Trust.

Consistently with Lipson, which established that GAAR can apply where a specific avoidance rule is misused to achieve a result that is contrary to its object and spirit (para 84), Lamarre ACJ found that the series of transactions represented abusive tax avoidance for purposes of s. 245(4), stating (at paras. 115, 117-119, TaxInterpretations translation):

It was only by taking advantage of the introduction of the reversionary trust, established by a beneficiary corporation, that Gennium's profits could end up in the hands of the Satoma Trust without any tax impact, through the combined operation of the attribution rule in subsection 75(2) and the dividend deduction accorded to corporations under subsection 112(1)… .

…The interaction of these two provisions created a tax advantage, in that Gennium was stripped of its surplus in favour of a trust at zero tax cost, which is contrary to the object and spirit of these two provisions.

Subsection 75(2) seeks to prevent income splitting by a transferor who transfers property to a trust while enjoying the right to recover the property. Subsection 112(1) provides for the non-taxation of inter-company dividends.

The object and spirit of these two provisions is not to allow the transfer of funds from a corporation to a trust by taking advantage of a total tax reduction. The avoidance transactions at issue run counter to the purpose of these provisions.

Predecessors of the taxpayers had been acquired for their losses in transactions where less than 50% of their voting shares, but more than 90% of their non-voting participating shares, had been acquired. The Minister had reassessed to deny the acquired losses primarily on the basis that there had been an acquisition of control, but secondarily through applying the general anti-avoidance rule.

In discussing the scope of the obligations of CRA and Finance to provide correspondence relating to loss utilization schemes that had been requested by the taxpayers, V.A. Miller J stated:

… It is my view that in a GAAR appeal, draft documents prepared in the context of a taxpayer’s audit or considered by officials involved in or consulted during the audit and assessment of the taxpayer should be disclosed. They inform the Minister’s mental process leading up to an assessment. They may also inform the Minister’s understanding of the policy at issue.

V.A. Miller J found that most of the redacted portions did not have to be produced, including on the basis that there was no evidence that the particular document was considered by CRA during the audit.

An individual (Mr. Wild) stepped up the adjusted cost base of his investment in a small business corporation (PWR) by transferring his PWR common shares to two new Holdcos for him and his wife in exchange for preferred shares of the Holdcos, and electing under s. 85 at the right deemed proceeds amount to use up his capital gains exemption. However, the paid-up capital of those preferred shares was ground down to essentially nil under s. 84.1.

The solution adopted was for PWR to then transfer high basis assets to the Holdcos in consideration for preferred shares of the same class, so that the PUC of the preferred shares held by Mr. Wild personally could be bumped due to the class-averaging rule in s. 89(1). It was of no import that there was a corresponding shifting of PUC away from the preferred shares taken back by PWR, because those shares were then redeemed for a note in reliance on the s. 55(3)(a) exception (and their relatively high basis anyway).

Lyons J, in finding that there was an abuse under s. 245(4), so that the MInister's assessment reversing the step-up to the preferred shares held directly by Mr. Wild was confirmed, stated:

[T]he…transactions…achieved a result (extraction of corporate surplus indirectly and use of his [capital gains] exemption) that section 84.1 was intended to prevent and defeats its underlying rationale and did so by misusing the PUC computation in subsection 89(1) to trigger the share averaging thus artificially inflated the PUC in The Shares held by Mr. Wild without any new capital contribution made by him.

The following transactions occurred under a KPMG-advised “Q-Yes” plan for the avoidance of 90% of the B.C. income tax that otherwise would have been payable on the taxable capital gain realized on the July 8, 2002 closing of the arm’s-length sale of the shares of a corporation (“ALI”):

1. On June 14, 2002, shareholders of ALI transferred their shares on a rollover basis under ITA s. 85(1) to a newly-incorporated B.C. corporation (“Veracity,” which was the taxpayer), which had selected a June 30 fiscal year-end for federal and B.C. income tax purposes.
2. Shortly after the closing, Veracity paid $2,000 in directors’ fees to two B.C. directors and lent the net sales proceeds of approximately $23.5 million on a non-interest-bearing basis to one of its shareholders (a corporation).
3. In late July 2002, Veracity purchased 15,000 units (with a cost $266,337) of a publicly-traded limited partnership (“Gaz Metro”) carrying on regulated natural gas utility business in Quebec, having a permanent establishment there and having a September 30 fiscal year end (so that a proportionate part of its gross revenues and payroll would be allocable to the purchased units at that year end on the basis inter alia that Veracity as a partner also had a Quebec establishment).
4. Veracity selected an August 31, fiscal year end for Quebec income tax purposes.
5. In September 2002, Veracity acquired a further 30,000 Gaz Metro units.

The payment of the directors’ fees before August 31, 2002 (together with allocations on the Gaz Metro units not occurring until after August 31, 2002) ensured that when the taxable capital gain was reported for the August 31, 2002 Quebec taxation year, the Quebec allocation formula allocated 100% of that income to B.C. On the other hand, in the federal/B.C. return for the year ended June 30, 2003, the allocation of gross revenues and payroll on the Gaz Metro units in that year resulted in 90% of the income (mostly the taxable capital gain from the sale) being allocated to Quebec for B.C. allocation purposes.

The Minister assessed Veracity under the B.C. GAAR provision (s. 68.1 of the Income Tax Act (B.C.)) on the basis that 100% of the taxable capital gain was taxable in B.C.

After noting that s. 68.1(1)(d) (the B.C. equivalent of ITA s. 245(4)) referenced a misuse or abuse of the provisions of the B.C. Act, MacKenzie JA rejected the Crown’s submission that the transactions constituted an abuse of the ITA s. 85 rollover rule, stating (at paras. 68-69):

[S]. 68.1(1)(d)… does not apply to provisions in the federal Act that are not expressly incorporated into the BC Act. …

Had the Legislature intended the BC GAAR to cover misuse or abuse of provisions beyond the BC Act and its regulations, such as the federal Act, it could easily have adopted the explicit language in subsection (c) [the equivalent of ITA s. 245(3), which referred to the reduction, avoidance or deferral of tax under the B.C. Act or “under any other federal or provincial Act.”]

(Similarly, Veracity exploited the ability to choose a different year end under the Quebec Act “but the provisions of the Quebec Act are clearly not within the scope of s. 68.1(1)(d)…” (para. 123).)

She went on to note that, in any event, although “s. 85’s purpose encompasses both deferral and preservation of a taxable capital gain such that upon the transferee’s ultimate disposition, the taxable capital gain, if any, will be realized and included in the transferee’s taxable income” (para. 85), its purpose was not “ensuring that the gain will give rise to an actual tax liability” (para. 86, emphasis in original), so that here “s. 85…operated as intended” even though “100% tax did not result” (para. 90). To reach a finding of a general abuse of the capital gains provisions of which s. 85(1) was a part would require an impermissibly “broad reading of the GAAR whereby any transaction that leads to lower taxes payable would trigger an abuse of general provisions such as s. 3” (para. 92).

In contrast to ITA s. 85(1), the inter-provincial “Allocation Rules” in Part IV of the ITA Regulations were specifically referenced in s. 13.3 of the B.C. Act. However (paras. 102-103, emphasis in original):

The purpose of the Allocation Rules is to allocate the income to the provinces. How the provinces tax that income, if at all, is beyond the purpose, object and spirit of the Allocation Rules. … I note the… comment in Copthorne…that:

… in some cases the underlying rationale of a provision would be no broader than the text itself. …

[T]he underlying rationale of providing a mechanism of allocation is fully explained by the formula it provides, choosing, as it did, to use gross revenues and salaries and wages as the factors.

S. 68.1(1)(d) should not have been applied.

The taxpayer was a holding company and Canadian-controlled private corporation which wholly-owned a “Partnerco” holding a 24.975% limited partnership interest in a strata development partnership (“HLP”) which, by May 25, 2006, had realized income of $13 million from the sale of most of the strata units, and was projected to realize another $863,546 of profit from the sale of the remaining six units. The three siblings of the shareholder of the taxpayer (“Rossano”) held their interests in HLP under the same Holdco-Partnerco structure, and the general partner, holding a 0.1% interest, was held by Rossano.

On May 25, 2006, HLP lent $8.5 million ($2.1 million each) to the four Partnercos, and then each Partnerco immediately declared a stock dividend of preferred shares with a paid-up capital and redemption amount of $2.1 million to each Holdco, with each Partnerco using the proceeds of such loan to redeem the preferred shares. Each Partnerco then paid a further stock dividend of preferred shares in the amount of the anticipated gain on the unsold strata units (for which it had been granted a put option by the developer).

On May 29, 2006, each Holdco sold its shares of its Partnerco to an arm’s length public corporation (“Nuinsco”) with substantial resource pools (and Nuinsco acquired the shares of the general partner for $1.) HLP then lent its cash of $4.4 million to Nuinsco.

On May 30, 2006, each Partnerco was wound up into Nuinsco and Nuinsco was admitted as sole limited partner of HLP. On May 31, 2006, HLP allocated its income of $12.1 million to its partners (mostly, Nuinsco). HLP then sold its remaining strata units, and HLP was dissolved.

In finding that the transactions did not entail an abuse of s. 111(5), Rossiter CJ stated (at paras. 87-89):

Subsection 111(5) was never contemplated to apply to a situation where it is the entity with accrued losses (Nuinsco) that acquires another entity. To put it another way, I am unconvinced that subsection 111(5) applies to profit trading with a loss corporation acquirer such that it can even be said that the parties relied on subsection 111(5) in transacting as they did. …

[T]here is no indication in subsection 111(5) that it was meant to prevent the acquiring party from using its own losses against post-acquisition income allocable to that acquiring corporation indirectly due to its acquisition of the target corporation. …

In finding that there also was no abuse of the provisions of the Act for allocation of partnership income (ss. 103 and 96), he stated (at paras. 99, 101, 109):

In this case, the allocation scheme in the HLP partnership agreement had not changed since the creation of the partnership. … There is no indication that this scheme was chosen for tax purposes or that it was unreasonable. …

It was open to the Respondent to argue that the combination of paragraph 96(1.01)(a) and section 103 demonstrate a general policy against profit trading between new and former members; having failed to do so, it is not for me to expand the lis of the parties. …

[N]othing in the partnership regime prevents a partnership agreement from basing its allocation of income on the membership at its fiscal year-end. …

He acknowledged that the transactions entailed an indirect transfer of property from the LP to the Holdcos from a s. 160 perspective, but did not consider that there had been any receipt of property on other than FMV terms. However, he considered that if he were wrong on this mechanical valuation point, the transactions would have entailed an abuse of s. 160, stating (at para. 167):

I find that the Respondent has sufficiently demonstrated the existence of an underlying policy against allowing those who incur liability for income tax from reducing, through transfers to non-arm’s length parties at a time where the transferor is so liable or reasonably likely to anticipate incurring such liability, the pool of assets with which they may satisfy that tax debt.

An individual taxpayer engaged in a surplus-stripping transaction similar to transactions in a ruling which CRA had resiled from following Descarries. He held shares of a holding company whose adjusted cost base reflected the step-up of predecessor shares’ ACB in capital gains crystallization transactions by him and other family members. Such additional “soft” ACB would have been ignored under s. 84.1 if those shares had been transferred to a personal holding company in exchange for the issuance of shares with a purported paid-up capital equal to the transferred shares’ soft ACB.

Instead, he retracted his own soft ACB shares (as well as soft ACB shares that had been gifted to him by family members), which resulted in a capital loss under s. 40(3.6) that was added to the ACB of his common shares of the corporation under s. 53(1)(f.2). This s. 53(1)(f.2) bump was not caught by s. 84.1(2), so that he could transfer the bumped common shares to a personal holding company, taking back high PUC shares which he promptly redeemed.

Favreau J agreed with CRA that this conversion of soft ACB into hard ACB, in order to receive a tax-free return of capital, contravened GAAR, stating:

This series of transactions permitted the appellant, on the redemption of the Class G shares of [his new holding company], to extract as a tax-free return of capital, $994,628 derived from the surplus of his corporation by virtue of utilizing his capital gains deduction and that of his mother and sister. …

A corporation (“BPC”), which was mostly owned by a Canadian pension fund (“OMERS”), obtained the agreement of a predecessor of the taxpayer (“OPGI Amalco”) that, prior to BPC’s acquisition of OPGI Amalco, it or an OPGI Amalco subsidiary (“MRC Amalco”) would transfer various of its rental real estate properties (including the “Three Real Estate Properties”) on a s. 97(2) rollover basis to newly formed LPs (the “First Level LPs”). Following the acquisition on October 16, 2001 of OPGI Amalco by a subsidiary of BPC, the cost of the interests in the First Level LPs was bumped under s. 88(1)(d) on an amalgamation which formed the taxpayer. In 2004, the First Level LPs transferred the Three Real Estate Properties to respective newly-formed LPs (the “Second Level LPs”) on a s. 97(2) rollover basis and the First Level LPs were wound–up, thereby permitting the high ACB in their units to be pushed down onto the cost of the interests in the Second Level LPs under s. 98(3)(c). The taxpayer then sold its interests in the Second Level LPs, somewhat after the three-year period referenced in s. 69(11), to entities that were exempt from Part I tax.

In finding that the use of the s. 97(2) rollovers in these transactions was not abusive, D’Arcy J stated (at para. 193):

Parliament…made the positive decision to limit the application of subsection 69(11) to transfers to tax-exempt entities that occur within the three-year period [in s. 69(11).] In my view, it is reasonable to conclude that Parliament was of the view that transfers after this three-year period did not abuse subsection 97(2).

In rejecting a Crown position that an “indirect” bumping of the cost of depreciable property frustrated s. 88(1)(d) (and similarly respecting s. 98(3)(c)), D'Arcy J noted that CRA had not reassessed transactions in which tax exempts had been sold bumped LP interests in LPs which had been formed before planning for the OMERS transactions had commenced, as well as not reassessing respecting sales of bumped LP interests (in LPs that had been formed as part of the series) to non-tax exempts, so that the Crown effectively was "arguing that it is only when all three elements are present, i.e., the prepackaging, the bump and the sale to the tax-exempt entity that abuse occurs" (para. 203), and stated (at para. 204) that this approach was contrary to Copthorne as it sought a “finding of abuse on a broad policy statement…without attaching the policy to specific provisions of the Act.” He further stated that “section 88…did not require the Appellant to look at the nature of the assets of the First Level LPs to determine the amount by which it could bump its interest in the limited partnerships” (para. 206) and that when subsequently amending the bump rule “Parliament decided to substantially narrow the amount by which a partnership interest may be bumped” (para. 212).

Finally, in arguing an abuse of s. 100(1), the Crown was in effect asking him “to find that one of the purposes of subsection 100(1) is to base the capital gain not on the gain otherwise calculated under the Act, but rather on accrued gains (including recapture) of property held by the partnership,” which was not reflective of the non-look through drafting of s. 100(1).

Accordingly, GAAR did not apply.

After a third party ("BW") had made an offer to purchase the shares of a Quebec manufacturing company ("Vulcain"), Mr. Gervais sold 1.04M Vulcain preferred shares (having a modest ACB) to his wife ("Mrs. Gendron") in consideration for a promissory note (payable in five annual instalments) and elected that s. 73 not apply to produce rollover treatment. Mr. Gervais then gifted a further 1.04M preferred shares to Mrs. Gendron on a rollover basis. Two weeks later and at the same time as the Vulcain common shares were sold by the other shareholders, Mrs. Gendron sold her 2.08M preferred shares to BW for $2.08M. She treated half of her $1M capital gain as being attributed to her husband (as that was the portion which she had acquired on a rollover basis), and treated the other half of the capital gain as being hers (for which she claimed the capital gains exemption). After Mrs. Gendron had succeeded before the Federal Court of Appeal on other grounds, the case was remitted to Jorré J to consider the correctness of the Minister’s application of GAAR to Mr. Gervais to increase his capital gain from $0.5M to $1M.

In finding that there was abuse under s. 245(4), Jorré J stated (at paras. 135, 141-142, TaxInterpretations translation):

Here, one must consider all the circumstances surrounding the series: the recourse to transfers of two blocs of Class E shares, one bloc by sale and one bloc by gift, the election to not transfer the shares sold to Mrs. Gendron on a rollover basis, and the election to have the rollover apply to the shares donated to Mrs. Gendron with the result that subsection 47(1) of the Act applied at the time that Mrs. Gendron sold those shares. It is evident that this produces a result that subsection 74.2(1) seeks to prevent and which thwarts the purpose of subsection 74.2(1) and the scheme of the Act by avoiding the attribution of part of the taxable capital gain to Mr. Gervais which would normally have occurred at the time Mrs. Gendron sold the shares. …

The legislator, in enacting subsection 69(11) and 110.6(7) of the Act, had no intention of neutralizing the effect of subsection 74.2(1) in circumstances such as these. …

In the context of the attribution rules, the purpose of the [s. 73] election is to permit a taxpayer to defer or not the realization of a gain, and not to permit a taxpayer to avoid attribution. Here, the election was made in order to circumvent attribution. This is an abuse.

A family corporation (“Ontario”) used proceeds of a daylight loan to redeem shares of Holdco, which used those proceeds to purchase a life insurance policy (or, to be more precise, to purchase an annuity to fund the premiums on the acquired policy) from an accommodating offshore insurance company, with those funds making their way back, through a series of equally accommodating intermediaries, to the sole individual shareholder of Holdco (“Paul Sr.”) as a loan. This “Metropac” loan was guaranteed by Holdco, with the guarantee secured by Holdco’s life insurance policy. The loan terms limited the lender’s recourse thereunder to realization of such security. Paul Sr. used the loan proceeds to subscribe for preferred shares of Ontario, which repaid the daylight loan.

After finding that there was a shareholder benefit arising from Holdco permitting its policy to be used to pay off the Metropac loan, C Miller J went on, in the alternative, to find (at para. 139) that if it had been necessary to rely on GAAR, he would have found “there is an abuse of the underlying policy of subsection 84(1) of the Act and the Minister’s assessment of a deemed dividend is correct,” after having noted that the policy of s. 84(1) was “a limitation on returns to shareholders on a tax‑free basis to only the shareholder’s tax paid investment in a corporation, where such investment creates an equivalent increase in the company’s assets or decrease in its liabilities.”

An acquisition of the shares of a Netherlands public company indirectly holding the shares of a valuable Canadian sub (Univar Canada) with nominal paid-up capital was structured to effectively step-up the PUC of the shares of Univar Canada to fair market value by using (or, according to V. Miller J, abusing) the pre-2016 version of s. 212.1(4) rule. This was accomplished by setting up a sandwich structure immediately after the acquisition under which a new Canadian ULC, with high outside PUC, held the shares of a U.S. corporation holding the low-PUC shares of Univar Canada – so that such U.S. corporation could distribute the shares of Univar Canada to its controlling Canadian purchaser (the ULC) without technically being affected by the s. 212.1(1) deemed dividend rule.

Her analysis was informed by her starting point (at para. 83):

Subsection 212.1(4) is placed as an exception within an anti-avoidance section… . [I]t is reasonable to infer that subsection 212.1(4) cannot be used so that it would defeat the very application of section 212.1. …[S]ubsection 212.1(4) is aimed at a narrow circumstance where the purchaser corporation, which is resident in Canada, actually controls the non-resident corporation without manipulating the corporate structure to achieve that control.

She went on to say (at para. 97):

The exception should not apply in the situation where a non-resident owns shares of the Canadian resident purchaser corporation.

and referred (at paras. 96-8) to the 2016 proposed amendments to s. 212.1(4) as evidencing this policy of the old s. 212.1(4) rule.

The following transactions occurred under a KPMG-advised plan in order to avoid 90% of the B.C. income tax that otherwise would have been payable on the taxable capital gain on a sale of the shares of a corporation (“ALI”) which closed on July 8, 2002:

1. Shareholders of ALI transferred their shares on a rollover basis under ITA s. 85(1) to a newly-incorporated B.C. corporation (the taxpayer, and “Veracity”), which had selected a June 30 fiscal year-end for federal and B.C. income tax purposes.
2. Shortly after the closing, Veracity paid $2,000 in directors’ fees to two B.C. directors and lent the net sales proceeds of approximately $23.5 million on a non-interest-bearing basis to one of its shareholders (a corporation).
3. In late July 2002, Veracity purchased 15,000 units (with a cost $266,337) of a publicly-traded limited partnership (“Gaz Metro”) carrying on regulated natural gas utility business in Quebec, having a permanent establishment there and having a September 30 fiscal year end (so that a proportionate part of its gross revenues and payroll would be allocable to the purchased units at that year end on the basis inter alia that Veracity as a partner also had a Quebec establishment).
4. Veracity selected an August 31, fiscal year end for Quebec income tax purposes.
5. In September 2002, Veracity acquired a further 30,000 Gaz Metro units.

The payment of the directors’ fees before August 31, 2002 (together with allocations on the Gaz Metro units not occurring until after August 31, 2002) ensured that when the taxable capital gain was reported for the August 31, 2002 Quebec taxation year, the Quebec allocation formula allocated 100% of that income to B.C. On the other hand, in the federal/B.C. return for the year ended June 30, 2003, the allocation of gross revenues and payroll on the Gaz Metro units in that year resulted in 90% of the income (mostly the taxable capital gain from the sale) being allocated to Quebec for B.C. allocation purposes.

Before affirming CRA’s assessment to apply the B.C. GAAR to allocate 100% of the taxable capital gain to B.C., Macintosh J found (at paras. 23, 27) that although the LP units purchased were “a reasonable, stand-alone investment,” the targeted tax savings were five-times the investment return on the LP units and the units were purchased primarily for the purposes of these “Quebec Year-end shuffle” transactions.

Before finding that the transactions were “abusive of the inter-provincial income allocation provisions” (para. 56), Macintosh J stated (at para. 50) that the

legislative scheme intends that capital gains earned in Canada by corporations be taxable as income by the provinces based on an allocation formula designed to prevent both the over-taxation and the under-taxation of income earned by a corporation which is active in more than one province.

Other abuses also were identified, including the use of s. 85(1) as “using a tax deferral provision in that way, resulting in the tax being avoided completely” (para. 54).

Shortly after it became a public company, a predecessor ("Birchcliff") of the taxpayer entered into an agreement for a major acquisition of producing oil and gas properties and also negotiated a plan to merge with a corporation ("Veracel"), which had discontinued its medical equipment business, in order to access Veracel's non-capital losses and credits. Investors subscribed for subscription receipts of Veracel and received voting common shares of Veracel therefor under a Plan of Arrangement, Veracel and Birchcliff amalgamated immediately thereafter under the Plan, and the proceeds were used in the purchase of the oil and gas properties. The previous Veracel shareholders could elect to receive retractable preferred shares on the amalgamation, which most did.

As the voting common shares received by the investors on the amalgamation represented "a majority of the voting shares of the amalgamated entity (the "Majority Voting Interest Test")" (para. 13), no acquisition of control of Veracel occurred under s. 256(7)(b)(iii)(B), so that the loss-streaming rules under s. 111(5) were avoided. In upholding the Minister's finding that this avoidance was an abuse under s. 245(4), Hogan J stated (at paras. 105, 106):

[T[he Majority Voting Interest Test indicates that Parliament did not want amalgamations and reverse takeovers being used as techniques to avoid an acquisition of control in situations where the original Lossco shareholders do not collectively receive shares representing a Majority Voting Interest in the combined enterprise.

…Parliament adopted the Majority Voting Interest Test to prevent Lossco from being subsumed by Profitco without an acquisition of control of Lossco.

A scheme, which exploited a UK VAT rule which was intended to avoid double-taxation, so as to avoid tax, was found to be abusive. See summary under ETA, s. 274(4).

The Superior Plus Income Fund (the "Fund") effectively converted (in accordance with the distribution method contemplated under s. 107(3.1)) into a public corporation using an existing corporation (the taxpayer, a.k.a. "Old Ballard") with non-capital losses and other tax attributes as the new corporate vehicle - rather than using a new corporation. The transactions were designed to ensure that there was no acquisition of control of the taxpayer (which would have resulted in a streaming of its losses). In particular, although the unitholders of the fund became shareholders of the taxpayer, this was considered not to entail an acquisition of control of the taxpayer by a group of persons. Subsequent to the conversion, s. 256(7)(c.1) was introduced, which would have deemed there to be an acquisition of control of the taxpayer, if it had had retroactive effect (which it did not). The Minister disallowed the use of the tax attributes, on the basis that there in fact had been an acquisition of control of the taxpayer or, alternatively, under s. 245(2) (i.e., GAAR).

At the discovery stage, the taxpayer moved to compel the Minister to answer various questions, or to produce documents, or previously redacted portions of documents previously requested under the Access to Information Act, which CRA had resisted principally on the grounds of irrelevance. The questions included whether the Department of Finance considered making the 2010 SIFT amendments retroactive, why it had changed its explanatory notes to say that s. 256(7)(c.1) "clarified" rather than "extended" the change-of-control rules and whether the Attorney General agreed that initially the policy choice of the SIFT conversion rules was to allow the use of existing corporations. Requested documents included GAAR Committee minutes including comments of individual members (whereas CRA had provided only the final Recommendation of the Committee), and correspondence between CRA and Finance (resulting in the drafting of s. 256(7)(c.1))and between the GAAR Committee and Aggressive Tax Planning, with the questions seeking particulars on the questions posed above and policy considerations brought to bear on this file, and respecting what initially may have been diffidence on the part of Finance as to how to proceed, if at all.

Following the reasons in Birchcliff, Hogan J granted the taxpayer's motion in the main (including all the above-mentioned questions and documents). The scope of "relevance" in discovery is extremely broad. Hogan J stated (at para. 33):

As correctly pointed out by the Appellant's counsel, discovery serves a much broader purpose than eliciting evidence that is admissible at trial. For example, the discovery process allows the parties to gauge the weakness of their opponent's case.

Certain documents were irrelevant, or contained specific portions that should be redacted because of privilege, and certain questions improperly asked the Minister to draw conclusions of law.

The taxpayer, who had realized a capital gain of $30 million earlier in the year, subscribed for Class A common shares of a corporation newly incorporated by him ("Immeubles Molibec") in consideration for a demand promissory note of $22.5 million. The next day, Immeubles Molibec paid a stock dividend to the taxpayer consisting of preferred shares having a nominal paid-up capital and a total redemption value of $22.5 million, thereby effecting a "value shift" away from the common shares. On the same day, the taxpayer sold all his common shares to his two sons for their nominal value, and realized a capital loss of approximately $22.5 million. He engaged in similar transactions in the two subsequent years to realize capital losses of approximately $35 million and $7.5 million.

The taxpayer sought to distinguish 1207192 and Triad Gestco on the basis that he was an individual rather than a corporation (so that offsetting gain would be realized on the preferred shares no later than his death) and had sold to his sons rather than family trusts. Paris J considered that "the fact that the appellant is an individual and not a corporation does not change the academic ["théorique"] nature or quality of the claimed losses" (para. 19) as "during the years under appeal, the appellant simply had not sustained a financial loss from the sale of the shares to his sons" (para. 20), and furthermore, "as in Triad Gestco, the taxpayer has not presented any evidence that he had sold the shares received as a stock dividend and has not requested an adjustment to the tax attributes arising under the application of the GAAR to take into account an eventual sale of the shares" (para. 21). Respecting the second point, the "decisions… in the two cases would have been the same irrespective of the identity of the share purchaser… with whom the taxpayer did not deal at arm's length" (para. 23).

His appeal from reassessments denying the losses under GAAR was dismissed.

The six taxpayers, who were siblings (or a step-daughter of their deceased father), held all the shares, having an aggregate fair market value (FMV), adjusted cost base (ACB) and paid-up capital (PUC) of $617,466, $361,658 and $25,100 respectively, of a Canadian real estate company (Oka). They sought through the transactions described below to reduce the deemed dividend of $592,362, that otherwise would have arisen on the redemption of their shares, to approximately $265,505 (see 4 and 7 below [minor discrepancy in figures]). CRA assessed on the basis that a deemed dividend of the larger amount had been realized, as there had been an indirect distribution from Oka described in s. 84(2).

1. Oka lent $544,354 to a newly-incorporated company (9149) in December 2004.
2. In March 2005, the taxpayers exchanged their shares of Oka under s. 85(1) for Class B and C preferred shares of Oka, so that they realized a capital gain of $361,658 (which was offset by the current year's capital loss realized in 4 below and a carryback of the capital loss in 7 below) and with the Class B and C preferred shares having ACBs of $269,618 and $347,848 and low PUC.
3. They exchanged their shares of Oka for Class A common shares of 9149 having an ACB and PUC of $347,848 (with such PUC "step-up" complying with s. 84.1) and Class B preferred shares of 9149 having an ACB and PUC of $269,618 and nil.
4. Also in March 2005, 9149 redeemed the Class A common shares for their PUC and ACB of $347,848 (so that no deemed divided or capital gain resulted) and redeemed approximately ¾ of the Class B preferred shares, giving rise to a deemed dividend and capital loss of $196,506 to the taxpayers.
5. Oka sold its real estate in December 2005, but with title issues not resolved until December 2006.
6. Oka was wound-up into 9149 in December 2006, with the loan in 1 being extinguished.
7. At the end of 2008, 9149 redeemed the (¼) balance of the Class B preferred shares for $69,000, giving rise to a deemed dividend and capital loss to the taxpayers of $69,000 and $73,112.

In finding that the transactions abused the object of s. 84.1, Hogan J stated (at paras. 56, 59):

The tax specialist was aware of the fact that section 84.1 of the Act would cause the Class B shares issued by 9149 [step 3] to have an adjusted cost base that is higher than their paid-up capital, which would prevent the additional value accumulated before 1971 from being used to strip Oka of its surpluses. However, applying this rule ensures that the redemption of these shares will generate a capital loss that is sufficient to erase the capital gain realized in the preceding step, namely, the internal rollover of Oka's common shares [step 2].

...

The result of all three transactions ... is that the tax-exempt margin made it possible for part of Oka's surplus to be distributed to the appellants tax-free in a manner contrary to the object, spirit or purpose of section 84.1 of the Act.

Having regard to $66,940 of capital gain having been realized on the death of the taxpayers' father, the Minister was directed to assess on the basis that the taxpayers received a deemed dividend of $525,422.

A Canadian subsidiary ("Greenleaf") of Ford U.S. paid down to $9,750,000 (including accrued interest) a debt of $24,369,439 (plus accrued interest) owing by it to Ford U.S. through the application of share subscription proceeds of $14,843,596 received by it from Ford U.S.A. and the application of a small amount of cash on hand of $100,706. Eighteen days later, a predecessor of the taxpayer ("392"), which was at arm's length with Ford U.S., purchased all the shares of Greenleaf from Ford U.S. for consideration of $1, and purchased the debt at only a slight discount to the amount owing, so that the debt-parking rules in ss. 80.01(6) to (8) did not apply. The Minister applied s. 245(2) on the basis that Greenleaf had sustained a debt forgiveness in the amount of the debt-paydown.

In dismissing the taxpayer's appeal, Bédard J found (at para. 126-127) that the transactions were abusive having regard to s. 80(2)(g) (TaxInterpretations translation):

The spirit and object of paragraph 80(2)(g) are to ensure that, when a debt is settled in exchange for shares, the debt forgiveness rules apply by taking into account the actual value of the shares which are issued. ... In adopting this paragraph, the legislator sought to prevent a taxpayer from transforming a debt into shares with low value, thereby avoiding the debt forgiveness rules. ... In proceeding in two stages rather than effecting a direct conversion of debt to shares, the appellant circumvented the application of paragraph 80(2)(g) and, thus, a gain on debt settlement.

The taxpayers ("Brianne and Steven") were minors and beneficiaries of a family trust (the "Trust"). The Trust held all the common shares (as well as Class C preferred shares) of a management corporation ("FHDM") for a dental practice of the beneficiaries' father (Dr. Gwartz). The following transactions occurred:

• 1a. In 2003 and again in 2005, FHDM declared a stock dividend consisting of 150,000 Class D preferred shares, redeemable and retractable for $1 each, and having a total paid up capital of $1 (300,000 shares, $2 total PUC).
• 1b. In each of 2003 to 2005, the Trust disposed of 75,000 Class D shares to Dr. Gwartz for a $75,000 promissory note.
• 2. In 2005, Dr. Gwartz sold the 225,000 Class D shares to a numbered corporation ("206"), wholly owned by Dr. Gwartz's spouse, for $225,000 in promissory notes.
• 3. FHDM redeemed 206's Class D shares for $225,000.
• 4. Both sets of notes were repaid out of such redemption proceeds, effecting a transfer of $225,000 from 206 to the Trust.

The Trust reported a capital gain of $74,999.50 from step 1b for each of 2003 to 2005, which it allocated to the taxpayers for inclusion (as to the taxable capital gains portions) in their returns. The Minister reassessed the taxpayers under the general anti-avoidance rule on the basis that, by effecting the realization of capital gains and their distribution to the beneficiaries, rather than the distribution of taxable dividends, the series of transactions had abusively circumvented s. 120.4 of the Act. (The Minister conceded that reassessment of Steven's 2005 taxation year should be vacated as he was 18 in that year.)

Hogan J. allowed the taxpayers' appeals. Although s. 120.4 was circumvented, there was no abuse. He stated (at para. 65):

The fact that specific anti-avoidance provisions were enacted long before the introduction of section 120.4 leads me to infer that Parliament was well aware of the fact that taxpayers could arrange to distribute corporate surpluses in the form of taxable dividends or of capital gains subject to the application of those specific anti‑avoidance provisions. The fact that those provisions were not amended and that a specific rule was not included in section 120.4 to curtail well-known techniques leads me to infer that Parliament preferred simplicity over complexity when it enacted section 120.4.

The present case was distinguishable from Triad Gestco, which involved the creation of an artificial capital loss, rather than the shifting of an already-accrued capital gain on common shares of a taxpayer to other shares (namely, the Class D shares) of the taxpayer.

Paris J. noted obiter (at para. 75) that Overs v. The Queen, 2006 TCC 26, "has been implicitly overruled by the Lipson decision."

The taxpayer planned to move to the U.S.. Although the capital gain on the shares of his wholly owned corporation ("PC") that would have arisen (under s. 128.1(4)(b)) on his emigration would have been sheltered for Canadian purposes by unutilized capital losses, there would have been no corresponding step up in the tax basis of his shares of PC for US tax purposes. In light of this issue, he sold his shares of PC (which held liquid assets) to his brother-in-law ("J.C.") in exchange for a promissory note, which was satisfied (within the following serveral months) with funds that J.C. extracted from PC (by first transferring his high-basis shares of PC to a "Newco" in consideration inter alia for a promissory note, which was repaid by Newco with funds which it, in turn, extracted from PC).

After finding that s. 84(2) did not apply to deem any portion of the sale proceeds to be a dividend, Hershfield J. found that s. 245(2) did not apply to produce the same result. The alleged abuse was that the capital gains, which enabled the use of the taxpayer's losses, arose from payments that, being funds from the taxpayer's own corporation, would normally be treated as dividend income. Hershfield J. stated (at para. 116):

Indeed, triggering capital gains to utilize capital losses is not discouraged by the Act in any way. Transfers to a corporation without a section 85 election can be used to realize capital gains as can transfers between spouses. There is nothing abusive about realizing capital gains for no other purpose than to utilize available net capital losses.

Furthermore, allowing capital gains treatment of the sale produced a "fair result" (para. 139), whereas "to deny a tax benefit to which he was entitled by an express provision of the Act [i.e., s. 128.1(4)(b)] because he achieved it by a different legally effectively means is, frankly, bizarre" (para. 121). Although the transactions entailed surplus stripping, "it is doubtful whether in an integrated corporate/shareholder tax system, a surplus strip per se can be said to abuse the spirit and object of the Act" (para. 101).

A plan for the taxpayer to avoid capital gains on his sale of shares of a corporation ("PM") would have entailed him gifting the shares to a Barbadian trust of which his wife was the sole beneficiary (so that the trust was intended to be a spousal trust), with the trust then selling the shares to his wife for a promissory note equal to the fair market value of the shares, and the trust distributing the promissory note to her on a wind-up of the trust and her selling the shares of PM to the purchaser. In finding that the transaction, if effective, would have given rise to an abuse for GAAR purposes, Campbell Miller, J. stated (at para. 119):

"Sleight of hand to inject a non-resident trust (not a legal entity but deemed an individual only for tax purposes) in the middle of a Canadian resident couple to take advantage of the tax treatment of a non-resident trust's own jurisdiction is intended to defeat Canada's policy of taxing residents on their capital gains."

Woods, J. rejected a submission of the Crown that it would have represented an abuse of the Canada-Barbados Income Tax Convention to use the gains exemption in Article XIV(4) to avoid an anti-avoidance rule such as s. 94 of the Act (after noting that in the OECD Commentary, the OECD indicated that contracting states may need to explicitly provide in income tax conventions for the preservation of the right to apply domestic anti-avoidance provisions).

She also rejected a submission that it represented an abuse of that Convention to rely on such treaty exemption where the trust in question had very little connection with Barbados. She noted (at para. 381) that such an approach would result in a selective application of the Treaty to residents of Barbados, on the basis of criteria other than residence, and that the object and spirit of the Treaty was that residents of Barbados qualified for exemption.

The taxpayers engaged in a "Québec shuffle" transaction in which they transferred shares of subsidiaries to a related Ontario purchaser on a rollover basis for Québec purposes but on a non-rollover basis for federal purposes (so that for Ontario corporate purposes the Ontario purchaser had full basis in the acquired subsidiaries) with the Ontario purchaser then selling the shares without any Ontario corporate tax being payable. In finding that these transactions represented abusive tax avoidance under s. 1079.10 of the Québec Taxation Act, the Court noted that the purpose of the Québec rollover provision (s.518) was to defer the realization of a capital gain, and not to be used in a scheme to avoid the payment of any Québec income tax on the gain.

A partnership of which the taxpayer was a member ("Roseland II") and another partnership ("Roseland I") owning a similar and adjacent condominium development, sold all their assets to a newly-formed partnership ("RPM") with the purchase price being paid for by way of set-off against the subscription price for the partnership interests in RPM with such partnership interest in RPM having been distributed to the partners of Roseland I and II.

After finding that these transactions were undertaken primarily to realize a tax benefit (the realization of terminal losses), Paris J. went on to find that the transactions did not result in an abuse or misuse and rejected a submission of the Crown (at para. 122) that "there is a general or overall policy in the Act prohibiting losses on any transfer between related parties, or parties described by counsel as forming an economic unit". Paris J. stated (at para. 120):

"In my view, the particularity with which Parliament has specified the relationship that must exist between the transferor and transferee for the purpose of each stop-loss rule referred to by the Respondent is more indicative that these rules are exceptions to a general policy of allowing losses on all dispositions."

He went on to state (at para. 123):

"I would also point out that Parliament has chosen to define the circumstances in which the terminal loss would be denied on transfers of depreciable property between partnerships in subsection 85(5.1) (now subsection 13(21.2)) and in doing so would appear to have chosen to allow taxpayers who are not within the circumstances set out in that provision to claim their terminal losses."

The taxpayer and an unrelated individual ("DeBruyn") accomplished a split-up of the business of a corporation ("DEL") of which they were equal common shareholders by transactions under which (i) DeBruyn converted his (Class A) common shares into Class B common shares, (ii) the taxpayer sold his Class A common shares of DEL to a newly-incorporated holding company for DeBruyn's wife ("114") for a purchase price of $150,000, (iii) DEL issued a promissory note to 114 in satisfaction of a $150,000 dividend declared by it on the Class A shares, (iv) 114 as signed the promissory note to the taxpayer in satisfaction of the purchase price for the Class A shares, (v) the taxpayer transferred the promissory note owing to him by DEL to a holding company ("HHCI"), and HHCI purchased assets of the Kingston branch of the business of DEL in consideration for satisfaction of the promissory note.

After finding that none of the transactions was an avoidance transaction, Lamarre J. went on to indicate (at p. 298) that the Crown had not established "that the policy of the Act read as a whole is designed so as to necessarily tax corporate distributions as dividends in the hands of shareholders".

The taxpayers accomplished an indirect sale of their indirect investment in an operating company ("Canstar") to an arm's length purchaser ("Nike") by transferring their shares of holding companies for Canstar to a newly-incorporated Ontario corporation on a rollover basis for Québec purposes but not for federal purposes, followed by a sale of their shares of the Ontario corporation to Nike.

In determining that this transaction was abusive for purposes of the Québec anti-avoidance rule, De Michele J.C.Q. found that the purpose of the Québec rollover provision was to allow the deferral of capital gains tax, not its complete avoidance.

The taxpayer transferred assets of a business to a partnership in what was intended to be an s. 97(2) rollover transactions in consideration for cash, promissory notes and assumption of debt ("boot") totalling an amount less than the cost amount of the transferred assets, and a Class "F" partnership interest stipulated to have a value equal to the balance of the purchase price. The partnership used cash (derived in part from a third party who had subscribed for ¾ of the equity in the partnership) to subscribe for preferred shares of a sister company of the taxpayer ("Holdings"), with Holdings in turn using the proceeds to subscribe for preferred shares of holding companies ("Holdcos") for the various indirect individual shareholders of the taxpayer. A "back-flow preventor" clause in the Partnership Agreement provided that in the event that the partnership received any payments in respect of preferred securities held by the partnership, the partnership would make distributions to the holders of Class F units equalling such payments received.

After noting (at p. 3015) that "in the real world" the Class F partnership unit "represented nothing more than an undertaking to pay $18.7 million for Holdings preference shares, which were of no practical value to the Partnership by reason of the Partnership Agreement", Bonner J. went on to find that the transactions were abusive in that in substance it could be said that the supposed "tax deferred proceeds" for the sale (represented by the Class F units) had reached the Holdcos. The Minister had been substantially correct in reassessing the taxpayer on the basis that it had received additional boot of $18.6 million.

The taxpayer's wife ("Jordanna") borrowed $562,500 from the Bank of Montreal to fund the purchase of shares of a family company from the taxpayer for $562,500. A day later, the taxpayer and Jordanna borrowed, on a joint and several basis, $562,000 from the Bank secured by a mortgage on a new personal residence that they had just purchased, with the proceeds of that loan being used to pay off the loan the Bank had made to Jordanna. The taxpayer used the share sale proceeds to pay the vendor of the residence. The taxpayer filed his return on the basis that the inter-spousal rollover applied to the share sale and that s. 74.1(1) attributed to him the loss sustained by Jordanna resulting from the deduction of the interest expense on the mortgage loan from the dividend income she received on her purchased shares.

Bowman C.J. indicated (at p. 2692) that "paragraph 20.(1)(c) was intended to permit interest on money borrowed for commercial purposes to be deducted" and "that interest on money borrowed for personal use (such as buying a residence) is not deducible" and (at p. 2691) that "subsection 20(3) allows a deduction for interest on money borrowed to repay money previously borrowed for commercial purposes." Here, subsection 20(3) was being abused through "the purported attachment to the subsequent mortgage loan [of] the tax incidents of Jordanna's original and fleeting use of the proceeds of the first loan" (p. 2692).

The taxpayer, who held 14.28% of the common shares of a Canadian private corporation ("Consercom") transferred a 9.76% block to a wholly-owned holding company ("6311") in consideration for preferred shares of 6311 with a high paid-up capital (thereby giving rise to a capital gain eligible for the capital gains exemption). The taxpayer also transferred shares of a Canadian private corporation ("Gestion") that he owned together with his brother to 6311 in consideration for shares of 6311. The redemption of the taxpayer's preferred shares of 6311 was financed through dividends received by 6311 from Gestion.

After finding that s. 84.1 was intended to prevent the stripping of surpluses of an operating company, that although Parliament had assumed that a shareholder with less than a 10% block of shares would not be able to strip the surpluses of that company, such influence could be exercised when two related shareholders held all the shares of a company (Gestion), and that there would not have been an abusive transaction if the taxpayer had transferred to 6311 only the Consercom shares, Archambault J. found that there was an abuse here where 6311 used the surpluses from Gestion to redeem the preferred shares that had been issued in consideration for the Consercom shares.

The tax consequences to the taxpayer were to be determined on the basis that the sums received by him on such preferred shares in excess of their paid-up capital were a dividend to him.

A corporation ("117679") owned by the taxpayer issued a stock dividend of non-voting shares to the taxpayer that were redeemable and retractable for an aggregate of $487,000. The next day, the taxpayer sold these shares to a partnership of which his wife was a general partner and three of his children held a 99% limited partnership interest in consideration for a $487,000 promissory note of the partnership bearing interest at the rate prescribed for purposes of s. 74.5. The taxpayer utilized the enhanced capital gains deduction in respect of his gain on the sale. Thereafter, dividends and proceeds of redemption of the redeemable shares of 117679 were paid by way of set-off against payments of principal and interest on the promissory note. The Minister recharacterized under s. 245 everything that the taxpayer received from the partnership as dividends.

In reversing the s. 245 reassessment of the taxpayer, Bowman C.J. indicated that he could not find that there was "some overarching principle of Canadian tax law that requires that corporate distributions to shareholders must be taxed as dividends, and where they are not the Minister is permitted to ignore half a dozen specific sections of the Act" and also noted that the transactions did not lack economic substance in that there was "a genuine change in legal and economic relations that took place as a result of the transactions".

A partnership owned by the taxpayers admitted a third party ("Woodward") as a member of the partnership with a view to selling an apartment building of the partnership and allocating 80% of the resulting gain for tax purposes to Woodward, with Woodward also receiving 80% of the net proceeds of sale of the property (which were substantially reduced due to the leveraging of the property) and then ceasing to be a partner.

The allocation of 80% of the income to Woodward was found to be unreasonable given that "Woodwards' contribution was both ephemeral and for all practical purposes risk free". The reasonable treatment of the arrangement under s. 103 would be to treat Woodward's share of the income as the amount of income it actually received.

However, Bowman C.J. went on to find that if he had concluded that s. 103(1) did not apply, on balance s. 245 would be applied to reach the same conclusion, but with the possibility that perhaps all of the profit, rather than most of the profit, should have been allocated to the taxpayers.

In order to avoid the application of the pre-1994 version of section 80, debt owing by the taxpayer was sold by a bank (for consideration equal to 2.2% of the total amount owing) to a corporation ("W720") that had similar ownership to the taxpayer.

In finding that the avoidance of the application of section 80 by this transaction did not represent a "misuse". Rothstein J.A. stated (at p. 5028) that:

"If a provision of the Income Tax Act is not used cannot be misused",

and also found that there was no "abuse" stating (at pp. 5028-5029) that:

"We are not satisfied from the references given to us that there is a clear and unambiguous policy that debts that are not legally extinguished are to be treated as if they were."

In order that the sale of a real estate property could be accomplished in a manner that utilized the losses of a loss company ("Lobro Stables") the taxpayers transferred the property to a partnership utilizing the rollover in subsection 97(2), transferred their partnership interests to Lobro Stables utilizing the rollover provision of subsection 85(1), with Lobro Stables then selling the property at a gain.

In finding that these transactions did not result in an abuse or misuse for purposes of s. 245(4), Campbell T.C.J. stated that the principles in OSFC Holdings with respect to loss trading should not be extended to profit trading, and the transactions simply utilized the provisions of the Act for the very purpose for which they were designed.

Under a non-recourse loan owing by the taxpayer and other tenants of an office building to the non-resident landlord, 90% of the cash flow was applied first to the payment of interest and then designated and paid as rent. If the interest expense (which had been reduced to a 10% rate in the taxation years in question) exceeded the cash flow, the taxpayer could request in writing that the landlord advance the excess to him as an addition to principal, with such excess interest also being added to the principal if no such request was made. By the taxation years in question, the principal had accumulated to well over twice the value of the property.

After finding that the excess interest was deductible in full notwithstanding that a portion of it accrued on amounts that had been reinvested by the landlord as stipulated in the loan, Miller T.C.J. declined to find that this result represented in an abuse or misuse in the absence of being referred to any material that would assist him in understanding why the government permitted the deduction of simple interest on a payable basis and only permitted the deduction of compound interest on a paid basis. He stated (p. 1763):

"What is the policy? It is not my role to speculate; it is the Respondent's role to explain to me the clear and unambiguous policy. He has not done so."

A partnership ("SDF") of which the taxpayer and two trusts for his children were the partners owned substantially all the units in a second partnership ("SA") which, in turn, owned rental properties. SDF and SA had February 28 and January 31 fiscal periods with the result that net rental income of SA was not included in the income of the taxpayer for approximately two years. Archambault T.C.J. found that although the Act permitted income to be carried over for one year, the provisions of the Act, read as a whole, were contravened if a second, third or fourth partnership was interposed to defer the taxation of income for two, three or four years. Parliament's intent in enacting ss.96(1)(b), 248(1) and 249(2) was not to enable a taxpayer to defer the taxation of its income indefinitely. Accordingly, the fiscal period of SDF was to be treated as if it ended on February 28.

Given that s. 245(4) did not reference the Regulations, it was inappropriate for the Minister to assess on the basis that there was an abuse of the Reg. 1100(11) rental property restriction rules for the taxpayer to finance his investment at a personal level and deduct interest personally, so that the partnership's capital cost allowance claims were not restricted. Archambault, J.T.C.C. stated (at para. 72):

It is clear in administrative law that an act and regulations, although both enactments, are very different in nature. An act is passed by Parliament or a provincial legislative assembly, whereas regulations are most often adopted by a government (the executive) under the authority of an act. In my view, since subsection 245(4) of the Act does not say "the Act and Regulations read as a whole", one must not take into account the rules adopted by the government in the Regulations. If Parliament had wanted them to be considered, it would have clearly so stated in subsection 245(4) of the Act, as it has done in a number of other provisions of the Act.

Furthermore, even if Reg. 1100(11) could be considered in determining whether there was an abuse, there was none, given that it is quite common for a shareholder (or partner) to borrow in order to provide capital, and given that s. 245 cannot be used by the Minister as a tool to force taxpayers to structure a transaction in a manner most favourable to the tax authorities.

The taxpayer, who owned a portion of the shares of the holding company ("GH") which, in turn, owned an operating company ("GBC") utilized the enhanced capital gains exemption in connection with the sale of a cement plant operated by GBC through the following transactions: the taxpayer and the other shareholders of GH transferred a portion of their shares of GH to a newly-incorporated company ("Newco") in consideration for shares of Newco having a value of $500,000; GBC paid a dividend-in-kind of most of the cement plant assets (having a value of $1 million) to GH; GH redeemed the common shares held in its capital by Newco by transferring to Newco the assets it had received from GBC; and the shareholders of Newco's sold their interests in Newco to the purchaser (who also purchased the remaining cement-plant assets directly from GBC).

Bowman T.C.J. found (at p. 250) that even if the transactions had been avoidance transactions, there would have been no abuse or misuse for purposes of s. 245(4):

"Simply put, using the specific provisions of the Income Tax Act in the course of a commercial transaction, and applying them in accordance with their terms is not a misuse or an abuse. The Income Tax Act is a statute that is remarkable for its specificity and replete with anti-avoidance provisions designed to counteract specific perceived abuses. Where a taxpayer applies those provisions and manages to avoid the pitfalls, the Minister cannot say 'Because you have avoided the shoals and traps of the Act and have not carried out your commercial transaction in a manner that maximizes your tax, I will use GAAR to fill in any gaps not covered by the multitude of specific anti-avoidance provisions'."

After it was determined as a result of settlement of litigation between the taxpayer and a co-venturer that it would sell its 50% interest in 13 properties to the co-venturer, the taxpayer gifted the 13 properties to a private foundation that had been funded and was controlled by its controlling shareholder, with the foundation then selling the properties to the co-venturer. Although the taxpayer avoided capital gains under s. 110.1(3), it realized recapture of depreciation and as part of the same transactions, received a loan from the private foundation sufficient to more than cover the tax realized by it.

In concluding that there was no abuse or misuse, Bowman TCJ. found that the taxpayer had used s. 110.1(3) "for the very purpose for which it was designed", namely, allowing "the tax consequences of a charitable gift to be mitigated", and further noted that section 245 was "an extreme sanction" which "should not be used routinely every time the Minister gets upset just because a taxpayer structures a transaction in a tax-effective way".

The taxpayer acquired units in limited partnerships that had incurred Canadian exploration expense and then, within 24 hours and after the occurrence of the partnership year ends, resold their units for nominal consideration pursuant to "put" agreements.

Before going on to find that former s. 245(1) of the Act denied the deduction by the taxpayers of the Canadian exploration expense allocated to them by the limited partnerships, Linden J.A. found that such deductions by the taxpayers would have fallen outside the intended object and spirit of the resource provisions of the Act because the goals of the resource provision (the encouragement of exploration) were not furthered as a result of the taxpayer's fleeting investment in the partnerships at a time when no exploration or money was being carried out or contemplated.

The taxpayer which owned one-half of the shares of a company ("Wascana") having non-capital losses that were about to expire, transferred depreciable property to Wascana on a rollover basis and then, a few days later, purchased the depreciable property back at the same price but on a non-rollover basis. The other unrelated shareholder of Wascana engaged in similar transactions.

In finding that the deduction by the taxpayer of capital cost allowance on the stepped-up tax basis of the property did not give rise to an undue or artificial reduction of income for purposes of former s. 245(1), Teskey TCJ. stated that he did not believe that these transactions offended the object and spirit of the Act.

A non-resident corporation ("EC") approached a business associate who, along with two other individuals, formed a Canadian corporation ("RMM") to buy the shares of a Canadian subsidiary ("EL") of EC for a cash purchase price approximating the cash and near cash on hand of EL and a Canadian subsidiary of EL ("ECL"). Immediately following the purchase, EL was wound-up into RMM and ECL was amalgamated with RMM; and three or four days later, RMM used the cash received by it from EL and ECL to pay off a loan that had financed the acquisition.

In finding that if this transaction had not already been caught by s. 84(2), s. 245 would deem the appropriate portion of the sale proceeds to be a dividend, Bowman TCJ. stated:

"The Income Tax Act, read as a whole, envisages that a distribution of corporate surplus to shareholders is to be taxed as a payment of dividends. A form of transaction that is otherwise devoid of any commercial objective, and that has as its real purpose the extraction of corporate surplus and the avoidance of the ordinary consequences of such a distribution is an abuse of the Act as a whole."

He also found that the Canada-U.S. Income Tax Convention could not prevent Canada from applying GAAR to recharacterize the transaction as one to which section 84 applied.

The taxpayers sold their shares of a corporation ("Bec"), whose assets (following a sale of real estate) consisted largely of cash, to a corporate purchaser for a cash purchase price that reflected, in part, the savings that would accrue to the taxpayers from effectively receiving that cash as an exempt capital gain rather than as a liquidation dividend from Bec. Following the acquisition, the purchaser amalgamated with Bec and used the cash received on the amalgamation to pay off a bank loan that had funded the acquisition.

In finding that the taxpayers were correctly reassessed under s. 245(5) to treat the purchase price received by them as a taxable dividend from Bec, Bonner TCJ. stated (at p. 121):

"The transaction in issue which was designed to effect, in everything but form, a distribution of Bec's surplus results in a misuse of sections 38 and 110.6 and an abuse of the provisions of the Act, read as a whole, which contemplate that distributions of corporate property to shareholders are to be treated as income in the hands of the shareholders. It is evident ... that the section [245] is intended inter alia to counteract transactions which do violence to the Act by taking advantage of a divergence between the effect of the transaction, viewed realistically, and what, having regard only to the legal form appears to be the effect."