In a Class of their Own? Applications for Leave to Appeal to the Supreme Court of Canada in Class Actions

This blog post deals with two areas of law that are near and dear to my heart: class actions and appeals to the Supreme Court of Canada. The question I tackle in this post is whether class actions are more likely than other types of cases to be granted leave to appeal to the Supreme Court of Canada.

This is a hypothesis that I have heard from other lawyers over the years, and it seems plausible. There are a number of reasons to believe that class actions would be granted leave to appeal relatively often compared to other types of cases. Class actions are generally high stakes and can affect thousands, or even millions, of Canadians. Many class actions raise complicated legal issues, and the lawyers on both sides of class actions tend to be sophisticated. As a result, it seems plausible that class actions are more likely to meet the test of public importance to be granted leave to appeal to the Supreme Court of Canada.

We wanted to test these intuitions empirically. As readers of our blog will know, we maintain a database that sets out dozens of data points relating to every Supreme Court of Canada leave decision decided from January 1, 2018 onwards. One of the features we track is whether the case involved a proposed or actual class action. (In our dataset, a proposed class action that has not yet been certified as a class action is coded as a class action. Even before certification, the possibility that the case might impact such a large number of Canadians and raise complex issues warrants similar treatment to a case already certified.) This data allows us to empirically analyze the question as to whether class actions are more likely to get leave to appeal than other types of cases.

Having now analyzed the data, the short answer to the question of whether class actions are more likely to get leave to appeal to the Supreme Court of Canada than other types of cases is a resounding “maybe, it depends on how you look at the data”. A nuanced look at the data provides significant insights.

Class Actions Are No More Likely to Get Leave to Appeal than All Non-Class Action Cases

We started by looking at every leave application decided between the beginning of 2018 and the end of 2022. Our database contained a total of 2,371 decisions during this period, of which:

• 170 leave applications were granted (7.2%);
• 2,189 leave applications were dismissed (92.3%); and
• 12 leave applications were remanded to lower courts (0.5%).

For the balance of our analysis, we exclude remands, because they represent a very small percentage of leave application decisions overall and raise unique consideration.

Leaving aside remands, we then split the leave applications into class actions and other cases. Our dataset contained a total of 121 leave applications in class actions, of which leave to appeal was granted in 12 (9.9%). By contrast, there were 2,238 leave applications in cases that were not class actions, of which leave was granted in 178 (7.1%). While class actions appear to get leave slightly more often than other leave applications, this difference is not statistically significant at conventional levels. Put differently, while the data show a minor difference, the difference is not significantly large relative to the overall size of the dataset such that we can conclude that the difference is the product of anything other than random chance.

Class Actions Are More Likely to Get Leave to Appeal than Other Private Law Cases

Perhaps comparing class actions to all other leave applications is the incorrect lens through which to consider this question. Included in the non-class action cases are many constitutional and criminal law cases which may be more likely to raise issues of public importance than private law cases. By contrast, class actions are usually private law cases. Consequently, perhaps the better question is whether class actions are more likely to get leave to appeal to the Supreme Court of Canada than other private law cases.

The answer to this appears to be yes. Our data contained 1,057 private law cases that were not class actions, of which 58 got leave (5.5%). This is just a little over half the rate at which class actions get leave (9.9%). Moreover, the difference appears to be statistically significant at the conventional 0.05 level. As a result, just comparing class actions cases to other private law cases, it does appear that class actions do get leave at a somewhat higher rate.

Class Actions May Get Leave More Often, but It’s Not Just Because They’re Class Actions

But that is not the end of the story. The next question is to ask why class actions are more likely to get leave to appeal than other private law cases. Are class actions more likely to get leave simply because they are class actions? Or are they more likely to get leave because class actions tend to have other features that are associated with a higher probability of getting leave? Our analysis indicates that the answer is the latter: while class actions may be more likely to get leave than other private law cases, that difference is fully explained by other features of those cases, and not the mere fact that those cases are class actions.

As we have done in some of our past work, we analyze this question by using a logistic regression model. This is a statistical model that allows us to isolate the impact of various factors on the probability of getting leave to appeal to the Supreme Court of Canada. In other words, it allows us to control for other factors that we have previously found correlate with the likelihood of getting leave, such as the length of the decision below and whether there has been some disagreement between the lower courts in the case. We used a number of different model specifications to try to investigate this question. In every logistic regression model we ran, we found that, once other features were considered, the fact that the case was a class action had no statistically significant impact on the likelihood of leave being granted.

So what does that mean? It means that class actions on average tend to have features that are generally associated with a higher likelihood of leave to appeal being granted, but the mere fact that a case is a class action is irrelevant. Put differently, as we surmised above, class actions may raise more difficult, contentious, or important legal issues than the average private law case, such that they are more likely to get leave than the average private law case. But as compared to other private cases of equal difficulty, contentiousness, or importance, class actions are no more likely to get leave.

In terms of how we think about the probability of cases getting leave to the Supreme Court, this means that it is reasonable for class actions practitioners to think that their cases have a slightly higher probability of getting leave than other types of cases. But what practitioners should not do is consider both the fact that a case is a class action as well as other positive factors when determining the likelihood of a case getting leave. For example, while it would be correct to reason that because a case received a lengthy treatment at the court of appeal and there was a dissent, that it has a good chance of getting leave. However, it would be erroneous reasoning to then also conclude that the case has an even higher chance of getting leave because it is a class proceeding.

After-Thought: The Golden Age of Class Action Leave Applications in 2018-2019?

The purpose of the analysis that led to this blog post was to investigate whether class actions are more likely to get leave to appeal than other cases and, if so, to what extent and why. However, as often happens in empirical analysis, the work we did led us to uncover some additional facts of interest that we were not looking for.

The one thing we noted when we did this analysis is that there appears to be a dramatic difference in the rate of class actions that got leave in 2018-2019 as compared to 2020-2022:

• In 2018-2019, there were leave applications in 50 class actions, of which 11 got leave to appeal (22%).
  
  Some of those cases that got leave in that period have gone to become leading cases in a variety of areas of private law, such as TELUS v Wellman (addressing the enforceability of arbitration clauses), Pioneer v Godfrey (deciding various issues that commonly arise in competition class actions), Atlantic Lottery v Babstock (holding that there is no cause of action in waiver of tort), Uber Technologies v Heller (now the leading case on unconscionability), and 1688782 Ontario Inc v Maple Leaf Foods (articulating the principles applicable to negligence claims in pure economic loss cases).
   
• By contrast, in 2020-2022, there were leave applications decided in 71 class actions, of which only 1 got leave to appeal (1.4%).

While this is on its face seems to show a striking difference, we are reluctant to infer much. These are small sample sizes. It may simply be that there was a series of leave applications in class actions in 2018 and 2019 that raised issues of particular public importance that warranted getting leave, while the later class actions in 2020-2022 where leave was sought did not raise such issues. However, we cannot exclude the possibility that there might be other factors at play that might explain this type of clustering of class actions. Further investigation is warranted to see whether this kind of clustering of cases getting leave is a more general phenomenon at the Supreme Court of Canada.