Standard of Review of Regulations: Auer v. Auer, 2024 SCC 36

The Supreme Court of Canada handed down its much-anticipated decision on standard of review of regulations in Auer v. Auer, 2024 SCC 36 this morning. I was co-counsel for the appellant, Roland Auer. After the hearing back in April, two things seemed quite clear to me: the Supreme Court would apply the Vavilov framework to judicial review of regulations; but was also likely to find that the regulations at issue here (the federal Child Support Guidelines made by the Governor in Council) were valid. And so it transpired.

The major issue of administrative law principle was the standard to be applied in cases where an applicant seeks to challenge a regulation. In Katz Group Canada Inc. v. Ontario (Health and Long‑Term Care), 2013 SCC 64, [2013] 3 SCR 810, at para. 28 the Supreme Court held that judicial intervention would only be appropriate “on the basis of inconsistency with statutory purpose” where regulations were demonstrated to be “irrelevant”, “extraneous” or “completely unrelated”. This standard was criticized as being “hyperdeferential” (Portnov v. Canada (Attorney General), 2021 FCA 171, [2021] 4 FCR 501, at para. 19) and an inappropriate carve-out from the general Vavilov framework (British Columbia (Attorney General) v. Le, 2023 BCCA 200, at para. 93). But in the Alberta Court of Appeal in Auer, the majority held that regulations have a special status recognized by the unique Katz standard (2022 ABCA 375, at paras. 47-63).

The Supreme Court unanimously, in reasons written by Côté J, resolved that dispute in favour of Vavilov. The Vavilov framework is “comprehensive” (at para. 21) and represents a “sea change” (at para. 32) in Canadian administrative law, setting out the starting point for any future consideration of issues related to standard of review (at para. 19). Indeed, Vavilov specifically folded in cases involving challenges to the lawfulness of regulations (at para. 22). Furthermore, the majority in Vavilov contemplated that “robust reasonableness review” would suffice to ensure that administrative decision-makers (including those who make regulations) remain within the boundaries of their authority, even where no formal reasons have been provided for the decision (at para. 26).

However, some aspects of the decision in Katz survived the Vavilovian sea change. There were five aspects to the Supreme Court’s prior decision in Katz:

[1]         “A successful challenge to the vires of regulations requires that they be shown to be inconsistent with the objective of the enabling statute or the scope of the statutory mandate” (para. 24);

[2]         “Regulations benefit from a presumption of validity . . . . This presumption has two aspects: it places the burden on challengers to demonstrate the invalidity of regulations . . . and it favours an interpretive approach that reconciles the regulation with its enabling statute so that, where possible, the regulation is construed in a manner which renders it intra vires” (para. 25 (emphasis deleted));

[3]         “Both the challenged regulation and the enabling statute should be interpreted using a ‘broad and purposive approach . . . consistent with the Court’s approach to statutory interpretation generally’” (para. 26, quoting United Taxi Drivers’ Fellowship of Southern Alberta v. Calgary (City), 2004 SCC 19, [2004] 1 S.C.R. 485, at para. 8);

[4]         “This inquiry does not involve assessing the policy merits of the regulations to determine whether they are ‘necessary, wise, or effective in practice’” (para. 27, quoting Jafari v. Canada (Minister of Employment and Immigration), 1995 CanLII 3592 (FCA), [1995] 2 F.C. 595 (C.A.), at p. 604). “It is not an inquiry into the underlying ‘political, economic, social or partisan considerations’” or an assessment of whether the regulations “will actually succeed at achieving the statutory objectives” (para. 28, quoting Thorne’s Hardware Ltd. v. The Queen, 1983 CanLII 20 (SCC), [1983] 1 S.C.R. 106, at pp. 112‑13);

[5]         The regulations “must be ‘irrelevant’, ‘extraneous’ or ‘completely unrelated’ to the statutory purpose to be found ultra vires on the basis of inconsistency with statutory purpose” (para. 28).

Several of these propositions are entirely unobjectionable, as Stratas JA observed in Portnov, at para. 20. In Auer, only the last of them was held to be inconsistent with Vavilov, as it created a carve-out based on the status of the decision-maker and nature of decision at issue:

In my view, all of the above‑mentioned principles in Katz Group,except for the “irrelevant”, “extraneous” or “completely unrelated” threshold, remain good law and continue to inform the review of the vires of subordinate legislation. As I will explain, the significant sea change brought about by Vavilov in favour of a presumption of reasonableness as a basis for review erodes the rationale for the “irrelevant”, “extraneous” or “completely unrelated” threshold, and maintaining this threshold would perpetuate uncertainty in the law. Accordingly, there is sound basis for a narrow departure from Katz Group (see Canada (Attorney General) v. Power,2024 SCC 26, at paras. 98 and 209; R. v. Kirkpatrick, 2022 SCC 33, at para. 202, per Côté, Brown and Rowe JJ., concurring). Otherwise, Katz Group continues to “provide valuable guidance on the application of the reasonableness standard” (Daly (2023), at p. 148). To the extent that the principles in Katz Group do not conflict with Vavilov, they “are to form part of the application of the reasonableness standard” (p. 149) (at para. 32).

…

To summarize, unless the legislature has indicated otherwise or if a matter invokes an issue pertaining to the rule of law which would require a review on the basis of correctness, the vires of subordinate legislation are to be reviewed on the reasonableness standard regardless of the delegate who enacted it, their proximity to the legislative branch or the process by which the subordinate legislation was enacted. Introducing these distinctions into the standard of review framework would be “contrary to the Vavilovian purposes of simplification and clarity” (P. Daly, Resisting which Siren’s Call? Auer v Auer, 2022 ABCA 375 and TransAlta Generation Partnership v Alberta (Minister of Municipal Affairs), 2022 ABCA 381, November 24 2022 (online); Daly (2023), at p. 147) (at para. 44).

I would also note that the first principle was not retained in its entirety in Auer. The word “requires” used in Katz suggested that this was an exclusive basis on which regulations could be challenged. However, in Auer, this principle is reframed: inconsistency with statutory objectives or scope is a basis on which to attack regulations, as they must be “must be consistent both with specific provisions of the enabling statute and with its overriding purpose or object” (at para. 33, citing References re Greenhouse Gas Pollution Pricing Act, 2021 SCC 11, [2021] 1 S.C.R. 175, at para. 87) but this is not said to be an exclusive basis for challenge. Similarly, some of Vavilov’s legal constraints — the statutory scheme, common law principles and the rules of statutory interpretation — will be “particularly relevant” in reviewing regulations (at para. 60) but clearly, in appropriate cases other constraints might be important and, indeed, the regulations might not follow a coherent chain of reasoning (at para. 51; see especially at paras. 52-54).

As to the second principle, Côté J noted the criticism that had been levelled at it but nonetheless saw it as consistent with Vavilov. For one thing, Vavilov also puts the burden on the applicant to demonstrate substantive unreasonableness (at para. 38, citing Vavilov, at para. 100). For another, the corollary of the presumption of validity that regulations should be interpreted to conform to the statutory scheme authorizing them does not impose any heavier a burden than Vavilov, because “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” (at para. 39).

Here, I confess to having some sympathy for the contrary view advanced by the Federal Court of Appeal. As Stratas JA explained in Innovative Medicines, at para. 30, “under Vavilov, the challenger does not have to overcome a presumption the decision is reasonable”. The applicant has a burden and so, in a de facto sense, a decision is reasonable until proven otherwise. But there is no de jure presumption of validity. The more modest (and I suggest better) way to think about this issue is that, on judicial review, a court must always characterize the decision at issue. Vavilov suggests that the characterization should, generally, be in favour of the decision-maker (Vavilov, at paras. 91-94, explaining the need to read administrative decisions “with sensitivity to the institutional setting and in light of the record” (Vavilov, at para. 96).

Retaining the presumption from Katz is a step too far, in my view, if it means that a judge must, as a preliminary matter, undertake an interpretive exercise designed to bring a regulation into conformity with the governing statutory scheme. Giving a regulation a fair characterization for the purposes of reasonableness review (just as is done for other administrative decisions) is one thing and would accord with the third Katz principle on broad and purposive interpretation but  requiring preliminary interpretive gymnastics would be quite another. Of course, it is possible that I am making too much of the de facto/de jure distinction, especially given Côté J’s insistence on the primacy of the Vavilov framework, her rejection of a special carve-out for regulations, her finding that the “very high degree of deference” required by the fifth Katz principle is inconsistent with Vavilovian reasonableness review (at para. 46) and her comments on the fourth principle from Katz, which I turn to now.

As far as the fourth Katz principle is concerned, it is clear that reasonableness review of regulations, no more than reasonableness review of any other type of administrative decision, does not permit a court to second-guess the merits of a determination made by an administrative decision-maker. There is, therefore, no violation of the separation of powers by applying the reasonableness standard to regulations (at paras. 55-56). Quoting Professor Mancini, Côté J emphasized that the court must be “mindful” that it plays only a reviewing role when assessing the reasonableness of regulations:

Importantly courts must organize these various sources properly to preserve the focus on the limiting statutory language. Again, the reasonableness review should not focus on the content of the inputs into the process or the policy merits of those inputs. Rather, courts must key these sources to the analysis of whether the subordinate instrument is consistent with the enabling statute’s text, context, and purpose. For example, Regulatory Impact Analysis Statements can inform a court as to the link between an enabling statute’s purpose and a regulatory aim, much like Hansard evidence. These analyses can help show how the effects of a regulation which, at first blush appear unreasonable, are enabled by the primary legislation (at para. 57).

This is closer to what I have in mind as far as the presumption of validity goes: treat it as de facto rather than de jure and characterize the administrative action at issue fairly.

With Katz put back in the bag, Côté J turned to the salient constraints here. Ultimately, the appellant’s case foundered on Côté J’s interpretation of the authority granted by s. 26.1 of the Divorce Act. The general regulation-making power (s. 26.1(1)) granted “extremely broad authority” to the Governor in Council (at para. 75). This general power is limited by s.26.1(2), imposing the “principle that spouses have a joint financial obligation to maintain the children of the marriage in accordance with their relative abilities to contribute to the performance of that obligation” but this too, Côté J held, was “expressed in broad terms” (at para. 79. Accordingly, on the various issues identified by the appellant with the substantive reasonableness of the Guidelines relating to lines drawn (or not drawn) by the Governor in Council, there was no basis for judicial intervention:

The GIC was entitled to choose an approach to calculating child support that (1) does not take into account the recipient parent’s income; (2) assumes that parents spend the same linear percentage of income on their children regardless of the parents’ levels of income and the children’s ages; (3) does not take into account government child benefits paid to recipient parents; (4) does not take into account direct spending on the child by the payer parent when that parent exercises less than 40 percent of annual parenting time; and (5) risks double counting certain special or extraordinary expenses. Each of these decisions fell squarely within the scope of the authority delegated to the GIC under the Divorce Act (at para.116).

With the framework now clarified, it will be interested to observe how future challenges to the reasonableness of regulations will be framed. The elimination of the special carve-out for regulations is certainly welcome (and, to my eye at least) inevitable in light of Vavilov. As ever, though, there are aspects of the Supreme Court’s analysis that will require close examination in future cases to determine the extent to which challengers to regulations are likely to prevail.

This content has been updated on November 8, 2024 at 18:00.