But the trilogy is equally important for two principles that bolster the burden on plaintiffs seeking certification in class actions generally. The first is the top court’s ruling that judges must scrutinize proposed methodologies at the certification stage to ensure they offer a “realistic prospect of establishing loss on a class-wide basis.” The second is the requirement that the class be identifiable by potential members, as established by another of the trilogy cases, Sun-Rype Products Ltd. v. Archer Daniels Midland Co.
Because these last two principles are applicable to class actions generally, they’ve arguably had greater potential impact than the ruling affecting indirect purchasers as it applies only to competition class actions.
“Pro-Sys is an attempt by the Supreme Court of Canada to make certification a meaningful screening device,” says Michael Eizenga of Bennett Jones LLP in Toronto.
Indeed, initial speculation that the methodology requirement in Pro-Sys applied only to competition cases has proven to be unfounded. In Andriuk v. Merrill Lynch Canada Inc., for example, the plaintiffs put forward a novel theory of damages that linked the loss to Merrill Lynch’s conduct and not to other market forces. But in May 2014, the Alberta Court of Appeal upheld a decision at first instance denying certification on the basis that the plaintiffs hadn’t established a methodology that could maintain that theory on a class-wide basis. The upshot was that common issues relating to causation or damages couldn’t be certified.
“The court did not distinguish Pro-Sys as an antitrust case,” says Eizenga.
Still, Chris Naudie of Osler Hoskin & Harcourt LLP in Toronto says the courts haven’t gone far enough in dealing with the methodology requirement.
“Some courts simply point to conflicts in the evidence as to the effectiveness of the methodology and then just stop there instead of assessing whether the methodology is plausible,” he says. “Their position is that the certification motion is not the time and place to address conflicts in the evidence, but that in my view should not end the inquiry as to whether there is some basis in fact that the proposed methodology will establish the loss on a class-wide basis.”
By way of example, Naudie points to Superior Court Justice Helen Rady’s ruling on the issue in Crosslink Technology, Inc. v. BASF Canada, a products liability case. He also points to the B.C. Supreme Court’s ruling in a price-fixing class action, Fairhurst v. Anglo American PLC.
“My concern in both cases is that the analysis is not faithful to the [Supreme Court]’s requirement that the expert evidence be assessed to some degree,” says Naudie.
“Justice Rady’s analysis, for example, is limited to three paragraphs.”
Naudie derives some optimism, however, from the January 2015 decision of the B.C. Court of Appeal in Charlton v. Abbott Laboratories Ltd. The court held that the absence of generalized risk data establishing causation on a class-wide basis was an insurmountable evidentiary barrier to certification.
“Charlton shows that there is some traction in appellate courts to breathe meaning into the standard, but it’s not clear that all courts are getting the message,” says Naudie.
Since the release of Sun-Rype, defendants have also taken aim at the requirement for an identifiable class requirement. “This requirement is becoming an important battleground in class actions,” says Eizenga. “But the results have been mixed.”
In March 2014, for example, the Ontario Superior Court denied certification in Keatley Surveying Ltd. v. Teranet Inc. after concluding that the plaintiff had failed to meet the requirement for an identifiable class.
The plaintiffs had alleged Teranet, the manager of Ontario’s electronic land registry system, had infringed copyright by selling copies of surveys found in provincial land registry offices. The defendant opposed certification on the basis that the plaintiff had failed to demonstrate the existence of an identifiable class of two or more people.
Justice Carolyn Horkins agreed with the defendant. The evidence led by the plaintiff failed to show that any of the surveyors affected had a complaint they wished to determine in a class proceeding. This failure barred certification.
But the Divisional Court overturned the decision and certified the case. The court ruled the identifiable-class criterion didn’t require evidence of two or more people who wished to have their claims determined in a class proceeding.
“The court noted that Sun-Rype does not mention that class members must desire to join the litigation,” says Eizenga.
“The requirement is simply that individuals be able to establish that they belong to the class.”
Last week, the Ontario Court of Appeal dismissed Teranet’s appeal. “The Court of Appeal swung the pendulum back again to a very pro-certification, low-bar certification stance,” says Kirk Baert of Koskie Minsky LLP, who with colleague Celeste Poltak represents the class.
Justice Elaine Adair of the B.C. Supreme Court also had occasion to deal with the issue in Ladas v. Apple Inc. The case involved allegations that Apple’s operating system performed synchronization functions in a manner that breached the privacy rights of users.
The only evidence tendered in support of the identifiable class criterion was an affidavit from the plaintiff’s law firm listing 17 people said to be members of the class with retainer agreements attached.
Adair ruled this evidence was insufficient to establish that these individuals were potential class members. While agreeing with the court in Keatley that plaintiffs don’t have to demonstrate that the potential members desired to join the class, they do have to show they may fall within it.
“Ladas is a promising sign that the courts will take the self-identification requirement seriously,” says Naudie.
“Class members should be able to know whether or not they fit into a class because otherwise it’s questionable whether a class action is doing much to advance access to justice.”