Class actions are often complex: thousands if not millions of putative class members may unite, seeking millions if not billions of dollars.  ‎With so much at stake, there is often a fierce initial battle at the class certification stage over whether the diverse individual claims are sufficiently common to proceed together.

On September 20, 2019, the Supreme Court of Canada (“SCC”) released Pioneer Corp. v. Godfrey1‎ an important and remarkable decision. By an 8:1 majority, the Court outlined a test for commonality of ‎loss at certification that diverges from what the plaintiff must show to establish liability at trial. While the ‎decision has direct implications for actions involving allegations of anticompetitive conduct, the Court’s ‎approach to certification will undoubtedly impact the approach to commonality in other instances. ‎


The case is similar to the many price-fixing class actions that have proliferated over the past decade in ‎Canada. The plaintiff alleged that between 2004 and 2010, the defendants - manufacturers or suppliers of ‎optical disc drives (“ODD”) - had unlawfully conspired to raise the prices of ODDs (and indirectly, prices ‎of products containing ODDs like computers) contrary to Section 45 of the Competition Act. The plaintiff relied on the statutory right of action in the Competition Act (Section 36), the tort of civil conspiracy, unlawful means tort, unjust enrichment and waiver of tort‎.

The plaintiff sued not only on behalf of purchasers who directly or indirectly purchased ODDs made by the defendants, but also on behalf of “umbrella purchasers” - persons who directly or indirectly bought ODDs made by others manufacturers not involved in the cartel who allegedly increased their prices in response to the defendants’ pricing. While the ability of umbrella purchasers to sue had been litigated in the U.S., this case was one of the first where the defendants invited a Canadian court to decide the issue.  

‎Mr. Justice Masuhara of the B.C. Supreme Court certified the action as a class proceeding, subject to certain exceptions and conditions.‎2   The B.C. Court of Appeal upheld the decision.3

Issues determined by the SCC

The SCC last addressed certification in competition class actions in its 2013 “Trilogy”.4‎ ‎ Since then, every competition case brought before the courts has been certified. Nevertheless, these cases raised a number of hotly debated issues, five of which were addressed in Godfrey:

  1. ‎At certification, sufficient if methodology can show loss reached ‎purchaser “level” ‎
  2. The majority agreed with the two lower courts that “loss-related” issues would satisfy the ‎commonality criterion under the Class Proceedings Act if the plaintiff established that there is a ‎plausible methodology capable of showing “one or more” purchasers at the requisite purchaser ‎‎“level” suffered loss as a result of the defendants’ conduct.‎ In dissent, Justice ‎Côté would have ‎required that the methodology be capable of sorting out at trial which purchasers did (or did not) ‎suffer loss.‎ ‎  ‎

  3. At trial, each class member must show loss
  4. The SCC parted ways with the lower courts and held that at trial, only class members who ‎suffered loss would be able to establish liability and recover damages. Merely showing that loss ‎reached a purchaser level would not establish liability to any given class member.‎ ‎ 

  5. ‎ ‎Umbrella purchasers can sue ‎‎
  6. The majority held that umbrella purchasers could sue under Section 36(1) of the Competition Act. ‎They disagreed with Justice Côté, who held that such claims should be barred because they are ‎too remote and expose defendants to indeterminate liability.‎ The majority reasoned that ‎umbrella effects are not ‎indeterminate but are pre-determined by the alleged intention of the ‎conspirators to raise prices across the market.‎  Nevertheless, the majority recognized that ‎umbrella purchasers would have a significant burden at trial: demonstrate that they had suffered ‎a loss that was causally linked to the alleged conspiratorial conduct.‎  

  7. ‎No complete code ‎‎
  8.  ‎All nine judges agreed that Section 36 of the Competition Act is not an exclusive remedy: ‎plaintiffs are not precluded from also suing for breaches of the Act based under other legal ‎grounds such as common law torts and equitable doctrines like unjust enrichment.‎ ‎ ‎

  9. ‎Discoverability and fraudulent concealment apply to the limitation ‎period
  10. Section 36(4) of the Competition Act prohibits actions under Section 36(1) based on a violation ‎of Section 45 from starting more than two years from the later of “a day on which the conduct ‎was engaged” or the final disposal of related criminal proceedings. Certain defendants ‎contended the claim was out of time, being commenced more than two years after the end of the ‎alleged cartel. The majority concluded that the limitations “clock” should only start when the ‎action was “discoverable” - when the plaintiff knew or ought to have known with reasonable ‎diligence the material facts supporting the claim.‎ ‎ Justice Côté disagreed, holding that the clear ‎language of Section 36(4) did not permit one to read in discoverability.‎ ‎ However, she agreed ‎with the majority (for slightly different reasons) that it was not plain and obvious that the ‎fraudulent concealment doctrine could not ‎apply to delay the running of the limitation period, ‎leaving the issue to be determined by the trial judge.‎ ‎  ‎    ‎

The balance of this article explores the Court’s controversial approach to commonality. 

Discussion ‎

 In Microsoft, the SCC reaffirmed the importance of certification as a “meaningful screening device”, being ‎neither a determination of the merits nor mere superficial, symbolic scrutiny.‎5‎‎ ‎ The Court explained that ‎there must be a sufficient factual  basis to allow the matter to proceed “without foundering at the merits ‎stage” because the statutory criteria for certification had not been met‎.6

The Microsoft Court also reaffirmed that commonality should be approached purposively and is the ‎‎“central notion of a class proceeding:” individuals resolve common concerns in one proceeding rather ‎than through inefficient, repetitive proceedings‎‎. 7

All judges in Godfrey accepted that in order to prove liability and recover damages, an individual class ‎member must show that they actually suffered loss. Taking a purposive approach, the logical starting ‎point for certification would be to ask whether the ‎plaintiff has a plausible plan to establish such losses ‎at the common issues trial. Otherwise, the case would founder, devolving into individual inquiries.   ‎

However, the majority posed a different question  to determine certification of ‘loss-related’ questions: ‎whether the plaintiff’s method is capable of showing at least one purchaser at the “level” of the ‎distribution chain suffered loss. They reasoned that‎:‎‎8‎‎

‎[S]howing that loss reached the indirect purchaser level satisfies the criteria for certifying ‎a common issue, since it will significantly advance the litigation, is a prerequisite to ‎imposing liability … and will result in “common success” as explained in Vivendi, given ‎that success for one class member will not result in failure for another. Showing loss ‎reached the requisite purchaser level will advance the claims of all the purchasers at that ‎level.‎

 In a spirited dissent, Justice Côté observed “a determination at a common issues trial of whether loss reached the indirect purchaser level in the distribution chain is of no assistance in resolving the question of whether the defendants are actually liable to any or all of the indirect purchasers".9‎ Since the level approach did not dispose of any element of liability for the class members, it did not “advance the litigation in any meaningful way”‎‎‎‎10‎‎, and did not satisfy the commonality requirement:‎11‎‎  ‎  ‎ ‎

‎[T]he fact that losses might have occurred somewhere at the indirect purchaser level in ‎the distribution chain does not assist us in determining which specific indirect purchasers ‎suffered losses in order to identify the class members to whom the Defendants might be ‎liable. If the common issues trial judge finds that overcharges were passed on to at least ‎one unidentifiable indirect purchaser, there would still be a need for individual trials; ‎therefore, duplication of fact-finding would not be eliminated (Dutton, at para. 39). And if ‎such individual trials are indeed required, then proof that loss occurred somewhere at the ‎indirect purchaser level is not truly “necessary to the resolution of each class member’s ‎claim”, is not a “substantial common ingredient” of their causes of action, and cannot in ‎fact result in “success” for any of those indirect purchasers…. ‎

The majority identified only one extreme scenario where the “level” approach might advance the litigation: ‎if the plaintiff failed to show any purchasers in the level suffered harm, then the action would be ‎dismissed.‎ ‎  Justice Côté responded that “it is unclear why any representative plaintiff would seek the ‎certification of a question that can meaningfully ‘advance the litigation’ only if it results in failure for all ‎indirect purchasers” and held that ‎“it would be a gross waste of private and public resources to litigate if ‎the only prospective ‎‎‘benefit’ was to show that there was no point bringing the case in the first place.”‎ ‎‎12‎‎

Loss-related common issues and preferable procedure

When a class is certified, the court approves a list of common issues to be answered by the trial judge.  The wording of such questions is important since they govern the scope of the trial.

‎In Godfrey, the  Court referred to “loss-related” issues. The actual questions certified were ambiguous, i.e. ‎‎“Did the Class Members suffer economic loss?” The majority observed:‎13

These questions were stated broadly enough that they could be taken as asking whether all class members suffered economic loss or whether any class members suffered economic loss. And, because they could be taken in two different ways they might, following the common issues trial, be answered in different ways.  

A methodology capable of establishing whether loss reached one or more purchasers at a level would be sufficient to answer the “any” question. In effect, the majority’s decision approved the “any” question as a proper common issue.

The formulation of the common question matters because the certification court must also be satisfied that a class action is a “preferable procedure.” Normally this is not a hurdle if positive answers to the common questions are capable of establishing liability across the class. But if such questions are not capable of establishing liability (i.e. the question only asks whether “anyone” suffered loss) then the court must evaluate if the individual issues (i.e. proving which individuals suffered loss) overwhelm the common issues, and the court may well decide that the case should be not certified.‎14‎‎‎ ‎

This is of particular concern in competition cases where proving loss is typically the most contentious issue, and where it is difficult to imagine individual trials for proving loss that would not overwhelm the process. Indeed, while assessing preferrability, the B.C. Court of Appeal recently remarked:‎15‎‎ 

The concept of comparing the import of common issues in relation to individual issues has particular resonance in indirect purchaser cases. In these cases, there are normally some common issues relating to the cause of action, and some individual issues relating to the individual circumstances of the class members. Whether common issues predominate over individual issues will often depend on whether loss on a class-wide basis can be considered a common issue, which would support certification, or whether loss will have to be established individually for the class members, which will likely make a class proceeding unmanageable.

The preferable procedure issue was not before the SCC in Godfrey. Going forward, one would expect defendants and courts to pay more close attention to the preferrability analysis ‎and the precise formulation of loss-related issues.


Respectfully, the Canadian “level” standard for certification of loss-related issues approved in Godfrey is ‎not a meaningful screening device. The standard stands in sharp contrast to antitrust class actions in the ‎United States, where at certification the plaintiffs must provide a methodology showing that all or nearly ‎all class members were injured.‎ ‎ The implications of the SCC’s generous approach to commonality are ‎not limited to competition actions: we expect to see ripple effects in applications to certify other types ‎of cases. ‎

Still, Godfrey is positive for defendants in that it confirms the plaintiff’s trial burden to prove which class members suffered loss. This should allow defendants to explore more aggressive post-certification strategies and inform their evaluation of liability in settlement negotiations. 


1 ‎2019 SCC 42. DLA Piper (Canada) LLP acted for a defendant which was released from the action before ‎the case reached the SCC.

2‎ ‎ 2016 BCSC 844. ‎

3‎‎  ‎2017 BCCA 302. 

4 The trilogy of cases released by the Supreme Court of Canada in 2013 consists of "Pro Sys" Consultants ‎Ltd. v. Microsoft Corporation, 2013 SCC 57; Sun Rype Products Ltd. v. Archer Daniels Midland Company, ‎‎2013 SCC 58; and Infineon Technologies AG v. Option consommateurs, 2013 SCC 59.‎

‎At para. 103‎‎‎

At para. 104.‎‎‎‎

At para. 106. ‎‎ ‎‎‎

At para. 108. ‎‎‎

9 At para. 222, emphasis in original.‎‎‎

10 At para. 217. ‎‎‎‎

11 At para. 224, emphasis in original‎.‎‎‎

12 At para. 225. Justice Côté quoted from our article analyzing the “level” standard adopted by lower courts:    

13  At para. 91, emphasis in original.‎‎‎

14 This happened to the umbrella purchaser claimants in Ewert v. Nippon Yusen Kabushiki Kaisha, 2019 ‎BCCA 187 at para. 137 [Ewert]. The authors are counsel for certain defendants in that case. ‎‎‎

15Ewert at para. 117. ‎‎‎ 

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.