This decision continues the acceptable practice of appellate courts hearing certification matters de novo.
In Good, Justice Ian Nordheimer, writing for a unanimous Divisional Court, was critical of the approach but nonetheless noted it as an acceptable practice:
It is entirely unclear why this peculiar approach is permitted in class actions. Normally, if a party has failed to plead a material element of their claim, the action is dismissed and the party is required to start over. That is not the approach taken in class proceedings, however, as I have already set out. One can point to many certification decisions in this Province where this phenomenon is demonstrated.
Nordheimer went further to state: “The fact that the representative plaintiff can significantly alter almost every facet of the proposed class proceeding on appeal makes it somewhat difficult to adhere to a normal standard of review by an appellate court” and “it does seem somewhat unfair, if not artificial, to decide, for example, that the motion judge made an error in principle in reaching her conclusion when the nature of the claim, and the foundation for it, has been entirely rewritten subsequent to that decision.”
The judge even indicated if the court had been confined to a review of the record from the original certification motion, it might well have deferred to the decision of the motion judge and dismissed the appeal.
Nevertheless, in Good, Nordheimer relied on the Ontario Court of Appeal’s decision in Pearson v. Inco Ltd. to declare that this practice is “well established and . . . implicitly (if not expressly) countenanced by the Court of Appeal.”
In Inco, the claim had been significantly narrowed at the certification appeal in the Divisional Court and the Ontario Court of Appeal considered it an error that the Divisional Court did not take this change into account in its reasons, which left it open to the Court of Appeal to determine whether the elements for certification had been made out.
As a result of the Inco precedent, Nordheimer determined that “it is not open to this court to refuse to consider the case as it is presented before us as opposed to considering solely the case that was presented to the motion judge especially where, as here, the respondent does not suggest that it has suffered incurable prejudice from the revisions.”
As he noted, this practice was also subject of recent comment in Keatley Surveying Ltd. v. Teranet Inc. In that case, Justice Harriet Sachs was critical of the de novo approach, while still certifying the action on appeal on revised common issues and a revised class definition.
In both Teranet and Good, despite making a clear effort to call the practice of de novo appellate certification determinations into question, the courts still certified the actions at issue as class actions.
Leave to appeal to the Ontario Court of Appeal in Good is being sought, and recently leave to appeal to the appeal court was granted in Teranet. This means the issue of revisions and narrowing of certification motions on appeal will likely be addressed in the near future.
Despite what this recent trend of criticism seems to suggest, a re-casting of a proposed class action on a certification appeal is not simply a practice that has gone on unchecked. Further, contrary to what can be discerned from the reasons in Teranet and Good, lower courts should not view this issue as one where the only reason they consider the re-casted case as presented on appeal is because they are begrudgingly bound to do so.
Rather, there are valid justifications for the support and continuation of this established practice. Such justifications include the following, the list of which is by no means exhaustive:
1. The core principles of class actions — access to justice, judicial economy, and behaviour modification — in fact support the practice of appellate courts making certification determinations de novo, as evidenced by Sachs in Teranet, where despite criticizing the practice, she still stated:
While the class action motion judge certainly has specialized expertise, in my opinion, access to justice and judicial economy militate in favour of us making a determination. . . . If it is sent back to the motion judge, it is likely to be appealed again, as almost all certification decisions at first instance are appealed. Once all certification appeals are exhausted, there are still two additional phases of proceedings remaining: the common issues trial and the damages phase/individual assessments. Remitting the matter to the motion judge will essentially restart the process and could add years to the litigation. This, in turn, will hamper the goals of judicial economy and access to justice for which the class action vehicle is designed.
2. Further, as stated by the Ontario Court of Appeal in Brown v. Canada (Attorney General), “[t]here is no question that class proceedings evolve as they work their way through the certification and case management process.”
This sort of procedural flexibility is unique to class actions and is necessary to realize the purpose of class action legislation; to hold plaintiffs strictly at the certification stage to their pleadings and arguments as they were initially formulated would be contrary to this aim.
Indeed, the B.C. Court of Appeal took this position in Halvorson v. British Columbia (Medical Services Commission) to hold that “there is nothing wrong with plaintiffs reformulating their [certification] approach on appeal.” Sachs echoed a similar point in Teranet.
3. In any event, recall that post-certification, a party may still seek to amend the certified common issues.
4. Lastly, as evidenced, this “phenomenon” is not unique to Ontario. Altering this practice would amount to a dramatic shift in both law and practice nation-wide.
Such justifications must not be going unnoticed, as the Ontario Court of Appeal, for example, in Inco, Kumar v. Mutual Life Assurance Co. of Canada, and Markson v. MBNA Canada Bank, has recognized that absent prejudice to the defendant cannot be cured by costs, appellate courts are entitled to decide on certification in a de novo fashion if the case as presented calls for it. And the Supreme Court of Canada has also approved of doing so: see Rumley v. British Columbia.
In conclusion, despite the recent criticism of the practice, the Divisional Court’s recent decision in Good should be taken in essence as another high-profile stamp of approval on the practice of appellate courts hearing certification matters de novo. This practice is well-established in Canadian class actions case law and practice, and as demonstrated, is amply justified and should be continued unabated.