In Gladwin Realty Corporation v Canada (2020 FCA 142), the Federal Court of Appeal upheld the Tax Court's finding of abuse relying on a GAAR analysis. However, the FCA reached its conclusion in a slightly different way than the Tax Court. The difference is indicated in two interesting comments that the Court made. One concerns what constitutes abuse of a SAAR under the GAAR.
In Gladwin, a series of transactions was conducted with the effect of inflating the Capital Dividend Account of the taxpayer. The taxpayer was a corporation that was a limited partner, holding a 99% interest, in a partnership. In sum, a distribution from the partnership triggered a deemed gain for the taxpayer under subsection 40(3.1) by causing the adjusted cost base of the taxpayer’s partnership interest to become “negative”. Half of the gain went to the taxpayer’s CDA and therefore increased its CDA balance. The taxpayer then paid out the capital dividend to its shareholders on a tax-free basis. After that, it elected an offsetting capital loss under subsection 40(3.12), a special rule directed at limited partners to avoid eventual double taxation of the same gain, effectively neutralizing the gain on which the CDA distribution depended. Since the parties agreed that the transactions were avoidance transactions that resulted in a tax benefit, the issue before the Court was whether the series of transactions frustrated “the provisions of the Act that were relied upon in order to achieve this result” (para 50).
At the first stage of the abuse analysis, the Court largely adopted Hogan J’s interpretation of the rationale underlying the related provisions. As part of the CDA mechanisms in the context of how and to what extent the Act imposes tax on capital gains, subsections 83(2), 89(1) and 184(2) were enacted to promote integration. As part of the negative ACB rules the effect of which is to trigger capital gains when more than the cost of a property is realized by a taxpayer, which is what subsection 40(3.1) accomplishes, a companion measure subsection 40(3.12) provides corresponding relief from potential double taxation caused by subsection 40(3.1).
The second stage of the abuse analysis is to determine whether that “single taxation” rationale was frustrated by arranging for no gain in the original amount to be taxable. In its analysis, the Tax Court focused on the overall result of the transactions, which was a significant over-integration achieved through deliberate triggering of subsections 40(3.1) and 40(3.12). The Court found the overall result inconsistent with the rational of subsections 40(3.1), 40(3.12) and the CDA mechanism.
However, FCA commented that in abuse analysis, the courts must “refrain from assessing the abuse on the basis of the overall result achieved” (para 70). The reason seems to be to avoid “economic reality” assessment of the transactions (para 86). Instead of relying on the overall result, the Court linked over-integration with the permanent CDA deficit (roughly $12 millions) created in the process. The CDA deficit was deliberately created to coincide in time with the cessation of taxpayer’s operations and therefore will never need to be offset by a taxpayer’s further gain. According to the Court, this permanent CDA deficit broke the integrity of the CDA regime. It also constituted a plain misuse of subsections 40(3.1) and 40(3.12) whose underlying rationale envisioned a CDA neutral application.
More interestingly, the FCA made a further comment on what constitutes abuse of SAARs. It was in response to the Crown’s assertion that the use of a SAAR to obtain a tax benefit is abusive under the GAAR. According to the Court, this assertion was too broad. Relying on Lipson (2009 SCC 1), the Court stated that “the abuse lies in using an anti-avoidance measure in order to obtain the result that it is intended to circumvent” [emphasis added] (para 68). In the view of the Court, “[a]s is the case for any other provision, what must be shown at the abuse stage of the analysis is that the anti-avoidance provision was used in a manner that defeats its underlying rationale” [emphasis added] (para 69).
In my view, this is a rather narrow approach because it imposes two restrictions on what constitutes abuse of a SAAR. First, a SAAR must be used before it can be abused. Accordingly, since merely circumventing the application of a SAAR does not involve using it, how is the primary requirement of use triggered? I am not sure this requirement of use can be reconciled with the Supreme Court’s decision in Copthorne (2011 SCC 63), where the transaction was found abusive because it circumvented subsection 87(3), an anti-avoidance rule, to achieve what it was intended to prevent (para 127). Secondly, under this approach, the use of a SAAR cannot be for any kind of tax benefit; it has to be for the tax benefit that the SAAR was intended to prevent. I wonder whether this restriction is necessary. In my view, as long as a SAAR was used to achieve a tax benefit it was not intended to provide, the courts may find that its rationale frustrated, and the provision misused or abused.
Despite this narrow approach, the Court still found a plain misuse of subsections 40(3.1) and 40(3.12). The two provisions, according to the Court, envisioned a CDA neutral application but were used to defeat this rationale. The Court provided two bases for evaluating this rationale. First, the deemed loss under subsection 40(3.12) was intended to neutralize the deemed gain under subsection 40(3.1). Over time, the CDA reduction caused by the deemed loss would be expected to neutralize the CDA increase caused by the deemed gain. Secondly, the subsequent amendment of subsection 89(1) in 2013 excluded a deemed gain under subsection 40(3.1) and a deemed loss under subsection 40(3.12) from CDA computation. In the view of the Court, this amendment was to ensure that the negative ACB rules would “continue to have a CDA-neutral application”, which was “an observable policy that was already in place” [emphasis added]. However, it remains controversial how much light the hindsight of a subsequent amendment necessarily should shed on the purpose of the existing tax provision. For example, in its textual, contextual and purposive interpretation of the CDA mechanisms, the Tax Court found that the subsequent amendment to subsection 89(1) was not instructive. Since the amendment had a much broader effect than if the GAAR applied, in the Court's view it modified the existing law more than clarified it.