BCSC clarifies the evidentiary requirements for establishing commonality of harm

Canadian courts have consistently cautioned against applying an exacting standard of scrutiny at the certification stage. But the B.C. Supreme Court’s recent decision in Ewert v. Nippon Yusen Kabushiki Kaisha issues an important reminder that the standard of assessing evidence at the certification requires more than just symbolic scrutiny.

In Ewert, the Court denied certification due to the plaintiff’s failure to demonstrate a credible methodology to show harm to indirect purchasers. In doing so, the Court reminded prospective class action plaintiffs that certification is not a “file, smile, and certify” exercise.[1]


The plaintiff commenced an action alleging price-fixing by several major operators of roll-on/roll-off (“RoRo”) vessels for shipping vehicles and heavy equipment to Canada.

The plaintiff alleged that the price-fixing – to which several defendants had pleaded guilty in the U.S. and in Japan – had resulted in manufacturers paying higher prices for shipping, and that all or part of these increased costs were passed down the distribution chain. The proposed class included those dealers and end consumers who purchased vehicles downstream (the “indirect purchasers”).

Relying on several causes of action – breach of the Competition Act, civil conspiracy, unlawful means, unjust enrichment, and waiver of tort – the plaintiffs sought certification of the action as a class proceeding.

Certification Application

The fate of the plaintiff’s certification application hinged on the strength of the proposed methodology for establishing harm to the indirect purchasers. The plaintiff put forward an econometric model developed by an expert economist.

The central issue on the certification application was whether the proposed model offered a “credible and plausible economic methodology to prove harm on a class-wide basis”.[2] The defendants argued that it did not because, among other shortcomings, the plaintiff had failed to show that the essential data required to perform the proposed analysis were actually available.

The plaintiff’s expert expected that the information needed to implement the proposed methodology – including sales information and data regarding supply and demand – would come from the defendants themselves and from other sources, some public. The expert referred to data that were “typically and likely available” from statistical agencies and industry or analysts’ reports,[3] referred to court documents that “might be available and used”,[4] and noted several general categories of data sources that “would include” essential data.[5]

The Court held that the plaintiff’s expert must provide “some identification of the other sources and not merely an expectation as to what they might be”, and the expert must have taken “at least a cursory look at the data to ensure its potential applicability.”[6] In short, merely asserting that the necessary data “possibly” existed and “might” be applicable would be insufficient.

Applying these principles, the Court found that the plaintiff’s expert had made no effort to investigate whether that data actually existed. The Court reasoned that “[i]n the absence of providing some evidence of the data [the plaintiff’s expert] says is necessary his opinion is ultimately purely theoretical.”[7] It continued:

… certification is meant to be more than symbolic scrutiny.  To put the matter more colloquially … it is not a “file, smile and certify” exercise.  Defendants––even major corporations––should not have to fly into the onerous discovery process in a complex class action on a “wing and prayer” that harm may be shown on a class-wide basis when the proper steps have not been taken to meet the low threshold required.[8]

In the result, the Court declined certification.


The message from Ewert rings loud and clear: certification is no rubber stamp. Ewert illustrates that even where the plaintiff’s proposed methodology for establishing commonality of harm in indirect purchaser cases appears, in theory, to be sound, that methodology will nonetheless be found wanting if the plaintiff does not demonstrate that the data necessary to implement the methodology are available. Where the plaintiff’s expert does little more than speculate about “possible” sources of data, the proposed methodology will not likely satisfy the baseline threshold of offering a “credible and plausible” method of establishing class-wide harm.

In light of Ewert, we expect to see courts and class action defendants putting greater pressure on plaintiffs to identify with greater precision the sources of data they intend to rely on in applying their proposed econometric models.

[1] 2017 BCSC 2357 [Ewert] at para. 12.

[2] Ibid at para. 14.

[3] Ibid at para. 40.

[4] Ibid at para. 43.

[5] Ibid at para. 41.

[6] Ibid at para. 48.

[7] Ibid at para. 53.

[8] Ibid at para. 54.



Stay Connected

Get the latest posts from this blog

Please enter a valid email address