The question of whether two parties are dealing at arm's length is an important one for many provisions of the Income Tax Act. The arm's length determination has implications that range from income valuation purposes, to the applicability of tax liability and anti-avoidance rules. The Act does not define the term "arm's length;" rather section 251(1) provides the basic guidelines regarding when parties will be deemed not to be acting at "arm's length." In order to determine whether two parties are dealing at arm's length, one must answer three main questions: whether the parties are related persons, whether the parties are a taxpayer and a personal trust within the meaning of paragraph 251(1)(b), and whether the parties are in fact dealing with one another at arm's length. The first two questions can be answered by statute, while the third question is answered by an application of the common law test.

A recent Tax Court decision, Damis Properties Inc. v. The Queen, highlights how the determination of whether two entities are dealing at arm's length has a significant impact on the taxpayer's tax liability. Courts can apply three common-law indicia to determine whether the parties were in fact operating at arm's length, which ask: 1) whether there is a common mind that directs bargaining for both parties to the transaction, 2) whether the parties to a transaction were acting in concert and without separate interests, and 3) whether there was de facto control. After conducting an analysis of the relationship between the parties, the Court in Damis found that none of the tests were satisfied and therefore concluded that the parties were dealing at arm's length.

Blue J's data reveals that of the three common-law indicia, courts most frequently examine whether parties are acting in concert. Apart from our data collection process, Blue J has also developed an algorithm that distills the common-law test into a series of factors to reliably generate a prediction on whether a court is likely to find that parties are dealing at arm's length. Based on Blue J's algorithm, the question of whether the parties transacted at fair market value is the most significant factor in determining whether the parties acted in concert.

Overview

Damis involved five taxpayers who realized significant gains from the sale of land that was held indirectly in various subsidiary companies. Each taxpayer entered into a share put agreement with an unrelated third party, which gave the taxpayers an option to sell the shares of the subsidiaries who still owed taxes from the sale for a negotiated price that was higher than the amount of the proceeds that would have remained in the subsidiaries had they paid those taxes. By entering into the transaction with the third party, the taxpayers realized financial gain. The Court found that the taxpayers and the third party had separate interests in the transaction and were therefore dealing at arm's length.

Takeaway 1: Acting-in-concert is the most frequently relied upon indicia

Blue J's data shows that the acting-in-concert test is the most frequently relied upon in the case law. In 62.5% of cases where courts only examined one of the indicia, the acting-in-concert analysis has been used and deemed sufficient for a determination; compared to 25% of cases where courts find that the parties had a common mind, and 12.5% of cases where one of the parties had de facto control. Courts often find that parties are acting in concert where one party acts to control the other parties' operations and transactions. In Damis, the taxpayers had neither board-level, officer-level control, nor legal control of the subsidiaries at the time of the transfer. The Court concluded that there was "no evidence to suggest that the taxpayers continued to direct the actions of the subsidiaries or continued to influence or control the activities of the subsidiaries."

It is important to note that even if the Court found that the parties were working toward a common goal, the parties may still be found to be operating at arm's length if they had separate interests. In this case, Owen J. accepted that the overall objective was economic gain, but noted that there was nothing "unusual or untoward about one party to a transaction accommodating the requirements of the other party to achieve a result that satisfies the separate interests of each of those parties."

Blue J's machine learning model reflects all three factual arm's length tests trained by previous court decisions and considers various factors, including whether one of the persons fills the role of executive-level decision-maker, whether either party can decline to participate in the transaction, and whether the transaction deviated from fair market value.

Takeaway 2: Fair market value is the most significant factor in the acting-in-concert indicia

Blue J's machine learning algorithm has determined that fair market value is the most significant factor in determining whether the parties are acting in concert. Blue J's data shows that 30% of cases involved a failure to carry out a transaction at fair market value, and of those cases, 96% resulted in a non-arm's length determination.

In Damis, although the parties traded at a value that was higher than traditional valuation metrics based on general financial considerations, the Court noted that "the fair market value test is flexible and recognizes that parties acting in their own interests is the best gauge for determining value." The Court accepted the taxpayer's argument that the subsidiaries and the taxpayers were trading at fair market value consideration by definition, because the parties traded at the highest price a willing buyer would pay for the shares in a market not exposed to undue pressures. Conversely, Blue J's algorithm predicts that had the Court examined the same set of facts but instead made a finding that the transaction was not conducted at fair market value, this would likely have been sufficient for the Court to find that the parties were not dealing at arm's length.

Leveraging technology

Tax lawyers and tax professionals can leverage Blue J's Related Persons and Dealing at Arm's Length tool to determine whether two parties are related and whether two parties are operating at arm's length. If parties are related, then they are deemed not to operate at arm's length. In cases where the parties are unrelated, Blue J's predictor analyses all the relevant factors to determine whether the parties are operating at arm's length.

Footnotes

1 Damis Properties Inc. v. The Queen, 2021 TCC 24

2 In ruling in favour of the taxpayers, Owen J. noted that "the application of the GAAR in this case could potentially have serious adverse implications for other arm's length share sale transactions where assets in the target are used in whole or in part to fund the purchase price of the shares." Owen J. concluded that neither section 160 nor the GAAR applied in this case. Para 354

3 Ibid at para 184

4 Ibid at para 304

5 Ibid at para 346

6 Ibid at para 221

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.