Date: 20231130

Docket: A-245-20

Citation: 2023 FCA 234

Heard at Toronto, Ontario, on October 31, 2023.

Judgment delivered at Ottawa, Ontario, on November 30, 2023.

Date: 20231130

Docket: A-245-20

Citation: 2023 FCA 234

REASONS FOR JUDGMENT

MONAGHAN J.A.

[1] In Deans Knight Income Corp. v. Canada, 2023 SCC 16 (Deans Knight), the Supreme Court of Canada was faced with deciding whether a series of transactions undertaken to monetize a corporation’s non-capital losses was subject to the general anti-avoidance rule (GAAR) in section 245 of the Income Tax Act, R.S.C., 1985, c. 1 (5^th Supp.). The Supreme Court concluded that the transactions achieved a result that frustrated the object, spirit and purpose of subsection 111(5) of the Income Tax Act and thus GAAR applied. (In these reasons, references to statutory provisions refer to provisions in the Income Tax Act.)

[2] In this appeal, we face the same question, but with respect to a different series of transactions undertaken by a different corporation. The Tax Court of Canada (2020 TCC 82, per Bocock J.) concluded that GAAR did not apply. The Crown appeals that decision, submitting that the transactions at issue here, like those in Deans Knight, abused subsection 111(5) and so GAAR similarly applies.

[3] The respondent disagrees. It submits that its circumstances are significantly different from those in Deans Knight such that GAAR does not apply.

[4] While there are factual differences between the two cases, when we apply the teachings from Deans Knight, the inescapable conclusion is that GAAR applies. Accordingly, I would allow the appeal.

I. Background

[5] Before the Tax Court, the parties filed an agreed statement of facts and a common book of documents, and called no witnesses. Only a brief summary of the largely undisputed facts is necessary for purposes of the appeal.

[6] Prior to July 2009, the respondent, then named National Convergence Inc., carried on a voice over internet protocol (VOIP) business, which was never profitable. In July 2009, one of the respondent’s creditors put the respondent into receivership and its business assets were sold to an unrelated third party. The proceeds from the sale, together with certain tax refunds, were used to settle the respondent’s senior secured indebtedness to MMV Financial Inc. (MMV Financial Inc., MMV Finance Inc., and wholly-owned subsidiaries of MMV Financial Inc. participated in certain transactions summarized below. However, neither party suggested that anything turned on which corporation undertook a particular transaction. Accordingly, to simplify matters, I will refer to each particular corporation simply as MMV.)

[7] By August 2009, the respondent ceased carrying on business and had no assets of value. It had 78 shareholders holding preferred and common shares and unpaid indebtedness of approximately $2 million, including US$850,000 owing under convertible secured debentures held by two shareholders. Clearly, it was insolvent. However, it had certain tax attributes, notably unused non-capital losses of approximately $27 million.

[8] Starting in November 2010, a number of transactions involving the respondent were undertaken. By the end of March 2011, the respondent had different shareholders, a different name, a different capital structure, different directors, different officers, a different source of financing, and a different business.

[9] In particular, the millions of shares outstanding before November 2010 were reorganized and consolidated, leaving only five of the previous 78 shareholders (the original five shareholders) holding 18 common shares as of December 21, 2010. Shortly thereafter, MMV subscribed for 17 common shares and, in February 2011, subscribed for 100,000 non-voting common shares. As a result, MMV owned more than 99.98 percent of the respondent’s outstanding common share equity.

[10] In December 2010, four individuals who had not previously been directors of the respondent were elected to the respondent’s board. Each was indemnified by MMV. The board appointed individuals who were also MMV officers as the respondent’s only officers. Initially two (and by the middle of March 2011, three) of the respondent’s four directors were also MMV directors.

[11] Using funds advanced to it by MMV, the respondent settled all of its unsecured debt for approximately 4 percent of the amount owing. MMV purchased the convertible debentures from the two holders for approximately 12 percent of the principal amount.

[12] In February 2011, MMV extended a US$75 million demand revolving credit facility to the respondent. In March 2011, the respondent purchased a portfolio of venture capital loans and related assets from MMV for approximately US$67 million, satisfying the purchase price by drawing down on the credit facility and issuing US$23 million of redeemable preferred shares to MMV. The respondent also engaged MMV to provide management services and marketing services.

[13] In May 2011, the respondent purchased additional assets from MMV, drawing down US$43 million on the credit facility to pay for them.

[14] In its 2011 to 2015 taxation years, the respondent deducted the non-capital losses incurred in the VOIP business against its income from the venture lending business.

II. The Minister’s reassessments

[15] Subject to certain limitations, the Income Tax Act permits a corporation to carry forward non-capital losses realized in one taxation year for deduction in a later taxation year. However, non-capital losses incurred by a corporation before control is acquired cannot be carried forward unless, following the acquisition of control, the corporation continues to carry on the business in which it incurred the loss: s. 111(5). Control for this purpose means de jure control, which generally means acquiring sufficient shares to elect a majority of the corporation’s board of directors: Duha Printers (Western) Ltd. v. Canada, [1998] 1 S.C.R. 795, 159 DLR (4th) 457, at 815; Deans Knight at para. 2.

[16] Here, the parties agreed that MMV had not acquired de jure control of the respondent. Thus, subsection 111(5) did not apply. However, the Minister reassessed the respondent and disallowed its deduction of the non-capital losses, asserting an abuse of subsection 111(5) so that GAAR applied.

III. The appeal to the Tax Court

[17] The respondent appealed the reassessments to the Tax Court.

[18] GAAR applies only where a “tax benefit” results from an “avoidance transaction” (or a series of transactions that includes an avoidance transaction) and, but for the application of GAAR, the provisions of the Income Tax Act would be abused: ss. 245(1)-(4). The respondent conceded the existence of a tax benefit and an avoidance transaction. Thus, the only issue before the Tax Court was whether there was an abuse.

[19] Analyzing whether transactions are abusive involves two steps. First, a court must identify the object, spirit and purpose of the relevant provisions. This is a question of law. Once it has done so, it must decide whether the result of the transactions frustrates that object, spirit and purpose. This is a fact-driven exercise.

[20] As it was obliged to do, the Tax Court first sought to determine the object, spirit and purpose of the provision at issue. Although it conducted its own analysis, the Tax Court ultimately adopted the object, spirit and purpose identified by the Tax Court in the decision that the Supreme Court ultimately overturned in Deans Knight. In Deans Knight Income Corporation v. The Queen, 2019 TCC 76 (Deans Knight TCC), the Tax Court said the object, spirit and purpose of subsection 111(5) is “to target the manipulation of losses of a corporation by a new person or group of persons, through effective control over the corporation’s assets”: at para. 134.

[21] Accordingly, the Tax Court here considered whether the facts supported a finding that MMV had acquired effective control of the respondent. It concluded they did not. Consistent with its framing of the object, spirit and purpose of subsection 111(5), the Tax Court’s inquiry focused solely on whether MMV had effective control of the respondent.

[22] The Tax Court first observed that in Deans Knight TCC the Tax Court “found that change in management, business activity, assets and liabilities and name are not markers of a change of effective control of a corporation”: reasons at para. 176. It rejected “the suggestion that failing to exercise voting power by removing directors ‘friendly’ to a minority shareholder is evidence of effective control in the hands of the minority shareholder”, and said “[n]o evidence showed that MMV Financial required de facto or effective control of [the appellant]” to effect the loss utilization: reasons at paras. 179-180. It said it had difficulty “comment[ing] on whether the [respondent’s] financing arrangements…shifted the locus of effective control…to MMV Financial”: reasons at para. 181. (In all cases, my emphasis.)

IV. Object, spirit and purpose does not focus on control

[23] Following Deans Knight, there can be no debate that the object, spirit and purpose of subsection 111(5) is “to prevent corporations from being acquired by unrelated parties in order to deduct their unused losses against income from another business for the benefit of new shareholders”. The Supreme Court tells us this no less than five times in the course of its decision: Deans Knight at paras. 6, 78, 113, 124, 140.

[24] Moreover, the Supreme Court expressly criticized the Tax Court’s focus on “effective control” in Deans Knight TCC. It explained that “[t]o define the object, spirit and purpose of s. 111(5) based on Parliament’s choice of test or substitute it for another test would…result in prioritizing the means (the how) over the rationale (the why)”: Deans Knight at para. 115 (emphasis in original).

[25] As it did in Deans Knight TCC, the Tax Court here erred in identifying the object, spirit and purpose of subsection 111(5). We must accept the Supreme Court’s determination of the object, spirit and purpose. Therefore, the only question before us is whether the transactions at issue here were abusive (i.e., frustrated that object, spirit and purpose).

V. The transactions abused subsection 111(5)

[26] Determining whether a transaction or series of transactions abused the relevant provisions is “necessarily fact-intensive”: Deans Knight at para. 121, citing Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54 at para. 44 and Canada v. Oxford Properties Group Inc., 2018 FCA 30 at para. 39.

[27] In this case, there is little dispute about the facts; the dispute centres on whether those facts support a finding of abuse.

[28] The respondent submits that there are two “major distinctions” between the transactions in Deans Knight and those here, which largely explain why the transactions in Deans Knight abusively circumvented subsection 111(5), and why the transactions at issue here do not.

[29] The first is “the artificial manner in which the transactions in Deans Knight circumvented de jure control, as compared with this case where MMV simply acquired fewer votes than would have conferred de jure control”: respondent’s further supplementary submissions at para. 4. The second is “the complete transformation of the shareholder base in Deans Knight, as compared with this case where the [five original shareholders] continued to hold a majority of the votes before and after the transactions”: respondent’s further supplementary submissions at para. 4.

[30] The Crown submits those distinctions are irrelevant because the Court’s task is not to compare the respondent’s circumstances to those in Deans Knight. Rather, to determine whether subsection 111(5) was abused, the Court must compare the result of the transactions to the provision’s underlying rationale. Once that comparison is made in this case, says the Crown, the abuse is clear.

[31] I agree with the Crown.

[32] Deans Knight sets out the framework to be applied at the second stage of the abuse analysis at paragraph 69 of the reasons. There it tells us that to determine whether the underlying rationale of the provisions has been frustrated, we must compare the result of the transactions to that underlying rationale. The Supreme Court provides three examples from prior decisions that illustrate how the object, spirit and purpose of different types of tax provisions can be frustrated: Deans Knight at para 70.

[33] Citing paragraphs 124-127 of Copthorne Holdings Ltd. v. Canada, 2011 SCC 63, the Supreme Court states that abuse may be found “where a series of transactions ‘achieved a result the section was intended to prevent’ while narrowly avoiding application of the provision”: Deans Knight at para. 70. That example applied in Deans Knight (paras. 6, 122, 140). It applies in this case as well.

[34] The object, spirit and purpose of subsection 111(5) – its rationale – is “to prevent corporations from being acquired by unrelated parties in order to deduct their unused losses against income from another business for the benefit of new shareholders”.

[35] As in Deans Knight, what happened here is exactly what subsection 111(5) seeks to prevent.

[36] MMV, unrelated parties not previously shareholders, acquired well in excess of 99 percent of the respondent’s equity (more than 98 percent of its common share equity for $1,000 and 100 percent of its preferred share equity for US$23 million). MMV also became the respondent’s only secured creditor and the only source of the respondent’s funding. By any measure, MMV acquired the respondent. That the five original shareholders remained with an infinitesimal equity interest, but de jure control, does not change this.

[37] More than 18 months after the respondent ceased to carry on the VOIP business in which it incurred its losses, MMV sold a business to the respondent. That business was of an entirely different nature than the business in which the respondent incurred its losses.

[38] The respondent deducted its losses against the income earned from that new business. MMV, the new shareholder with more than 99.99 percent of the respondent’s equity, is the only shareholder that benefitted from those losses.

[39] How did MMV benefit?

[40] But for MMV selling the assets to the respondent, the income generated by those assets would have been earned by MMV, but without the benefit of the respondent’s losses to reduce taxable income. In 2012 and 2013, the respondent paid dividends to MMV, entirely to the exclusion of the five original shareholders. Those dividends exceeded the respondent’s aggregate income in 2011, 2012 and 2013 by nearly $14 million. This was possible both because the losses meant the respondent paid no taxes on its income and because the respondent used principal repayments received on its loan portfolio to fund the dividends paid to MMV.

[41] Thus, MMV received all of the business income free of corporate tax. Moreover, MMV received significant capital that previously had been deployed in the business it sold to the respondent, albeit as dividends rather than as a reduction of its invested capital in the respondent (i.e., the advances under the credit facility or the preferred shares). The obvious inference is that this enabled MMV to retain its 99.99 percent equity interest and maintain its priority debt claim against the respondent while reducing the respondent’s asset base.

[42] The respondent argues that when examining shareholder continuity, the focus should be on de jure control. Here, it says, the five original shareholders with the majority of the common shares remained constant; in contrast, in Deans Knight, the original shareholder “was completely taken out of the picture” and the company “ended up with a completely new shareholder base”: respondent’s further supplementary submissions at para. 10.

[43] While those are the facts in Deans Knight, nothing in that decision suggests a completely new shareholder base is required to find abuse of subsection 111(5). Rather, the question Deans Knight posed was whether a new shareholder base benefitted from the losses. Here, for reasons explained above, only the new shareholder benefitted.

[44] The respondent also submits that the five original shareholders could have acted to elect a new board that would pay them dividends, and no contract precluded them from doing so. That may be true. But that they did not, when doing so would appear to have been in their economic interests, is not at all surprising because of the practical constraints they faced.

[45] All five original shareholders would have had to agree to a new slate of directors to outvote MMV. Moreover, MMV’s interest in the respondent (more than 99.99 percent of the equity that cost more than US$23 million) was of an entirely different scale and import than that of the five original shareholders (less than 0.01 percent of the equity valued at pennies when MMV became a shareholder). The power imbalance is self-evident. MMV had the right to demand immediate repayment of the credit facility and to demand redemption of the preferred shares within 30 days. Had it done so, the respondent would have been unable to pay dividends and, once those demands had been met, would have had little, if any, assets of value. Given this, and their collective infinitesimal interest in the respondent, the prospect of the five original shareholders electing directors and paying themselves dividends was illusory.

[46] Put another way, while their common shares provided the five original shareholders with de jure control, they had no effective way to use that control to benefit from the respondent’s losses. Their de jure control was meaningful only because it prevented MMV from acquiring de jure control. Doing so would have been fatal to MMV being able to benefit from the respondent’s losses.

[47] The transactions at issue “achieved an outcome that Parliament sought to prevent” and “all while narrowly circumventing the application of s. 111(5)”: Deans Knight at para. 122.

[48] Drawing from Deans Knight (at paragraphs 124-127), “the reorganization transactions resulted in the [respondent’s] near-total transformation: its assets and liabilities were shifted”. Here, the respondent’s assets were sold in 2009 and 18 months later its remaining liabilities had been settled or purchased by MMV for a small fraction of the amount owing. MMV became the respondent’s only creditor. As a result, like in Deans Knight, all that remained of the respondent were its non-capital losses. The respondent was “gutted of any vestiges from its prior corporate ‘life’ and became an empty vessel with Tax Attributes” that “were preserved to benefit another party”, MMV. The respondent “was, in practice, a company with new assets and liabilities, [a new shareholder/owner] and a new business”. Thus, “the transactions resulted in a fundamental change in the [respondent’s] identity”.

[49] Accordingly, the transactions frustrated the object, spirit and purpose of subsection 111(5) and were abusive.

VI. Other ground of appeal – Admission of Canada Revenue Agency communications

[50] In its notice of appeal, the Crown asserted that the Tax Court erred in admitting certain communications between the Canada Revenue Agency and the Tax Policy Branch of the Department of Finance. The respondent says the Tax Court gave no weight to those documents but, for purposes of the appeal, conceded the issue. Accordingly, I need not address this issue and I express no view on whether the Tax Court erred.

VII. Conclusion

[51] I would allow the appeal with costs, set aside the judgment of the Tax Court and, giving the decision the Tax Court should have given, dismiss the respondent’s appeal from the reassessments in issue with costs.

“K.A. Siobhan Monaghan”

FEDERAL COURT OF APPEAL

NAMES OF COUNSEL AND SOLICITORS OF RECORD

APPEARANCES:

SOLICITORS OF RECORD: