Some months ago, I mentioned that I will be running a symposium on the newsletter in celebration of Vavilov’s 5th year. The “Vavilov at 5” symposium entries will be sent to newsletter subscribers on Fridays. The regular newsletter will continue on Sundays.
I am excited to say that the symposium will begin this upcoming Friday, with an entry from Justice David Stratas of the Federal Court of Appeal. We have a great lineup of entries planned for the coming weeks. If you want to get a bird’s eye view of Vavilov at 5, I hope the symposium will interest you.
**dicta recording to follow**
Auer v Auer, 2024 SCC 36; TransAlta Generation Partnership v Alberta, 2024 SCC 37 (November 8, 2024)
Category: Standard of review for subordinate legislative instruments.
Context and Summary: As I explored in Issue #68, these significant appeals jointly considered the question of the standard of review for subordinate legislation. Here is what you need to know from these decisions (the products of a unanimous court, per Côté J):
Generally speaking, Vavilov and not Katz Group Canada Inc v Ontario, (Health and Long-Term Care), 2013 SCC 64, is the “comprehensive” starting point for the review of subordinate legislation (Auer, at para 21). Put differently, Vavilov was a “significant sea change” (Auer, at para 32). No longer will courts need to ask whether a subordinate instrument is “completely unrelated” to the purpose of an enabling statute, as Katz instructs. Rather, the application of reasonableness review is conditioned by the principles of statutory interpretation and the legal constraints on the regulator.
In applying reasonableness review, some of Katz’s uncontroversial elements remain—and so Auer and TransAlta together constitute a “narrow departure from Katz Group” (Auer, at para 32). For example, regulations continue to benefit from a “presumption of validity,” putting the onus on the challenger; and contra the ABCA, reasonableness review does not entitle the court to assess the consequences or wisdom of the regulatory policy writ large (Auer, at paras 29 and 58).
Practically speaking, and in applying reasonableness review to regulations, it is “fundamentally an exercise of statutory interpretation to ensure that the delegate has acted within the scope of their lawful authority under the enabling statute” (Auer, at para 59). The most relevant constraints will be the “governing statutory scheme, other applicable statutory or common law and the principles of statutory interpretation…” (Auer, at para 60).
Much was made in the ABCA about whether and how reasonableness review can be conducted in absence of formal reasons. It is true that “[m]ost of the time formal reasons are not provided for the enactment of subordinate legislation” (Auer, at para 52). However, as Vavilov affirms, even in the absence of formal reasons, “[t]he reasoning process can often be deduced from various sources (see Auer, at paras 52-54). This does not entitle a court to wade through regulatory materials with the aim of determining whether there is a “rational basis,” as a matter of policy, for the regulation. Since reasonableness review in this context is really a species of statutory interpretation, “[a] court must be mindful of its proper role when reviewing the vires of subordinate legislation, especially when it relies on the record, other sources or the context to ascertain the delegate’s reasoning process” (Auer, at para 57). Indeed, “[t]he potential or actual consequences of the subordinate legislation are relevant only insofar as a reviewing court must determine whether the statutory delegate was reasonably authorized to enact subordinate legislation that would have such consequences” (Auer, at para 58).
The Court upheld the validity of the Federal Child Support Guidelines in Auer, and the regulation at issue in TransAlta. TransAlta also contained an interesting issue of administrative discrimination.
Analysis: As readers will know, the doctrinal method deployed in Auer/TransAlta is the one I prefer. In general, Côté J’s reasons are excellent. Because they preserve Vavilov, and places a laser-focus on statutory intepretation, I think Auer/TransAlta are welcome cases in the administrative law pantheon. I highlight four substantive points from the decisions, and what they may portend for administrative law going forward:
The method for determining the appropriate standard of review in these circumstances is principled and simple—start with Vavilov. This is so not only for Cabinet regulations (the regulations at issue in Auer), but all types of subordinate legislation, as Côté J does not draw distinctions between types of subordinate legislation and the legal reasoning would capture all types. This method, of course, was championed by Stratas JA in a series of cases in the Federal Court of Appeal (Portnov v Canada (Attorney General), 2021 FCA 171; Innovative Medicines Canada v Canada (Attorney General), 2022 FCA 210) and these cases undoubtedly animated the Court’s decisions in Auer/TransAlta (an important signal about how central lower courts are to the development of our law—these decisions did not come from nowhere). I take these decisions to affirm a now-central truth about our administrative law. Vavilov is not only a set of operational rules about how to review administrative instruments of all types, but it is a simple and comprehensive set of principles, keyed to those operational rules, that can be deployed across the law of judicial review. For this reason, it is good news that Côté J rejects the ABCA’s approach, which would have trifurcated the standard of review based on the identity of the promulgator: (1) “true” Cabinet regulations; (2) rules and regulations adopted by municipalities or tribunals, reviewed on a modified standard of reasonableness; (3) adjudicative decisions, reviewed on Vavilov’s standards. This trifurcation would have unravelled the simplicity and comprehensiveness of Vavilov.
As these cases demonstrate, the Court appears committed to ensuring this does not happen. This is reassuring, especially after some rocky cases in which aspects of Vavilov appeared, at least from some perspectives, as open to exception (Mason v Canada (Citizenship and Immigration), 2023 SCC 21 (Issue #108); Ontario (Attorney General) v Ontario (Information and Privacy Commissioner), 2024 SCC 4 (Issue #123). Whether the stability and predictability of Vavilov remains depends on cases like Auer preserving it.
The application of Vavilov in this context is significant, because Côté J affirms that the principles of statutory interpretation tell almost the whole story. In the ABCA decision, the Court seemed to suggest that Vavilov reasonableness could not properly consider questions of vires; that every application of reasonableness review would descend into the policy merits, and that some of Vavilov’s constraints would simply not make sense in the context of subordinate legislation. But as I mentioned in Issue #68 in response to the ABCA decision, “Not all of the constraints have to apply in a given case for Vavilov to be relevant.” Litigants and their counsel should become even more familiar with the principles and rules of statutory interpretation, because those principles and rules will take on centre stage—as they should—in the review of subordinate legal instruments, which is fundamentally a question of legality.
On that note, the dispatch of Katz meaningfully changes how the principles of interpretation apply in this context. Recall that Katz’s standard of deference was predicated on a litigant showing that a regulation was “completely unrelated” to the purpose of the enabling statute. That standard was problematic. It permitted courts to gloss over the enabling statutory text, inviting a value-laden analysis of the connection of the subordinate legislation to an overarching purpose. To the extent Auer/TransAlta cut back on this tendency by reinforcing the normal principles of interpretation, it is a welcome adjustment. Now, the purpose of the enabling provision is still relevant. But Auer reinforces that it is not the centerpiece of the analysis. Instead, a regulation must fit with both the specific enabling provisions—the text—read in light of the enabling statute’s purpose (see Auer, at para 33).This is not a small change, in either practice or principle. Accordingly, many will be tempted to ask whether this decision makes it “easier” or “harder” to challenge government action. In an absolute sense, the narrow overruling of Katz means that the artificial deference it implemented no longer exists. The problem with that deference was its lack of correspondence to the statutory grant of authority—it compelled wide deference (the regulation must be “completely unrelated” to the enabling statute, a high bar) even if the enabling statute was circumscribed and prescriptive (see Innovative Medicines, at para 40). In this sense, this artificial burden—unprincipled, disconnected from the scope of statutory authority—is no more.
But I generally think that it is the wrong question to ask whether Auer/TransAlta will lead to “stronger” review. Rather, the only question is much more straightforward: what is the breadth of the statutory grant of authority enabling the regulation-making power? If it is broad under the text and purpose of the legislation, then the deference afforded to the regulator will be broader. If it is narrower, and the text tells the regulator to consider specific factors, then we should expect less room to move. In other words, deference is a function of the statutory text, no more or less. So it was in Auer itself: the Child Support Guidelines were intra vires because the enabling legislation grants the Governor-in-Council “extremely broad authority to establish guidelines respecting child support” (Auer, at para 5). The authority is not limitless (Auer, at para 77). Even so, the Child Support Guidelines respected the one mandatory, relevant legal constraint—that “the guidelines be based on the principle that spouses have a joint financial obligation to maintain the children of the marriage in accordance with their relative abilities to contribute” (Auer, at para 5).
There are still some unanswered questions going forward. As Paul Daly writes, one of these refers to the scope of the “presumption of validity.” As he does, I take this presumption to simply mean that the onus is on the challenger to show that a regulation is unreasonable because it does not “fit” with the enabling statute. It is true that Côté J notes that this presumption also includes that proposition that “where possible, subordinate legislation should be construed in a manner that renders it intra vires” (Auer, at para 39, Côté J’s emphasis). This could mean that there is a substantive presumption of interpretation that a regulation that better conforms with the enabling statute should be preferred. The Federal Court of Appeal, rightly in my view, did not assign substantive value to the presumption—beyond the statement of fact that the challenger bears the onus (see Portnov, at paras 20-21). And, reading between the lines, I do not think Côté J does so either. All the presumption of validity means is that “[i]f the reasonableness standard applies, to overcome the presumption of validity, challengers must demonstrate that the subordinate legislation does not fall within a reasonable interpretation of the delegate’s statutory authority” (Auer, at para 39). As Côté J says, “[t]his aspect does not heighten the burden that challengers would otherwise face pursuant to Vavilov” (Auer, at para 39). It may be tempting for government counsel and some courts to read these statements on the presumption of validity to erect a higher barrier when the object of the challenge is a regulation. For my part, I think this would be a mistake.
There will be more work to do in determining how reasonableness review works in the context of subordinate legislation, though Auer and TransAlta go some distance to clarifying any confusion. I suspect more ink to be spilled about the concept of “consequences.” This much is clear: the ABCA was mistaken in concluding that reasonableness review amounts to an inevitable reweighing of the policy merits/consequences of the regulatory action. In normal times—the application of reasonableness review to an adjudicative decision—this is not so. Nor is it so in this context.
However, consequences are relevant to the extent they link to the enabling statute. Again, the entire business is a question of statutory interpretation. For that reason, the consequences of the regulation are only relevant if there is an argument that the legislation does not authorize those consequences; “[w]hether those consequences are in themselves necessary, desirable or wise is not the appropriate inquiry” (Auer, at para 58). This is a fine line, but the line exists. The arguments made about consequences must be rooted in textual evidence that the statute authorizes those consequences. It is appropriate to look to consequences of a proposed interpretation not to question whether they are sound from a moral or political perspective. Instead, one looks to consequences to measure them up against evidence derived from the statutory scheme itself (the same is true when we are evaluating whether certain consequences are more or less consistent with the purpose of a provision: see Williams v Canada (Public Safety and Emergency Preparedness), 2017 FCA 252 at para 52).
Pelletier v Canada (Attorney General), 2024 FC 1669 (October 24, 2024)**Thank you to a SEAR reader for sending this case**
Category: Remedy.
Context: The applicant, a veteran, applied for a critical injury benefit under the Veterans Well-Being Act. After an initial decision, a first appeal panel of the Veterans Review Appeal Board [VRAB] denied his appeal. On judicial review, the Court found the decision unreasonable. A second appeal panel denied the benefit for different reasons—on a ground on which the first panel had previously accepted the applicant’s argument. Pelletier now seeks judicial review of this decision. In this case, the Court (per Ngo J) again finds the decision unreasonable. For our purposes, the question is the appropriate remedy.
Analysis: Ngo J declines to remit, bucking the usual rule in Vavilov, because “the circumstances of this case meet the threshold of exceptionality” [51]. This is a classic case of the endless merry-go-round that Vavilov warns against. As Ngo J says, at para:
The Court’s concerns about the undue delay and additional delay would also not be addressed with a further redetermination. The matter has already reached eight years, from the date of the Injury and initial application for the benefit in 2016. A further delay of two more years if the matter is remitted to a third appeal panel and if any party seeks judicial review of that decision would total ten years. Furthermore, the parties described the Critical Injury Benefit to the Court as a “one time payment” to a veteran who has been injured. I also do not believe that Parliament could have intended for the final disposition of a Critical Benefit Injury application for a “one time payment” to take so long (D'Errico at para. 19). This is a significant delay that threatens to bring the administration of justice into disrepute.
Note two things about this case. First, it is a rare outcome. Ngo J declines to remit and directs that Veterans Affairs grant the benefit. Second, note how Ngo J couches the discussion of remedy—in terms of the legislative intention.
Disclaimer: Nothing in the SEAR is legal advice. The SEAR is not designed to be comprehensive. It is designed to collect some cases that I believe are interesting to practitioners and students, and to provide my quick thoughts. Not all of the cases presented are fully analyzed ie) I’ve selected the most relevant issues in some of the cases. Please complement the SEAR with your own research and study. Any mistakes are my own