The Vavilov Framework I: Selecting the Standard of Review

As I mentioned in my previous post, the Supreme Court of Canada addressed two long-standing issues in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65. In this post, I will focus on the Court’s “revised framework for selecting” the standard of review (adopted by a majority of 7-2) (at para. 10).

In brief, reasonableness review is the starting point in all situations. But statutory appeals will now attract correctness review (at least on extricable questions of law). And in three other non-exhaustive scenarios correctness review is required by the rule of law: the resolution of constitutional questions, questions of central importance to the legal system as a whole and issues of overlapping jurisdiction. Jurisdictional questions, even in their purest form, will no longer attract correctness review.

The goal of the broad coalition of judicial review hawks and doves assembled in the majority is to “bring greater coherence and predictability to this area of law” (at para. 10). As yesterday’s post foreshadows, whether this is successful will ultimately depend on what happens next, for there are many details still to be worked out.

On first glance, all trace of nuance and context has been swept from this area. As the majority forthrightly states: “this decision conclusively closes the door on the application of a contextual analysis to determine the applicable standard, and in doing so streamlines and simplifies the standard of review framework” (at para. 47). Where once the contextual pragmatic and functional approach, with its four interlocking factors, reigned supreme until dethroned by a presumption of reasonableness review rebuttable by reference to contextual factors, context will no longer play any role in the selection of the standard of review. Rather, the standard applied “must reflect the legislature’s intent with respect to the role of the reviewing court, except wheregiving effect to that intent is precluded by the rule of law” (at para. 23).

The starting point is a presumption of reasonableness review. Significantly, this presumption is based on the brute fact of a legislative choice to delegate decision-making authority to an administrative decision-maker. Other justifications, such as the expertise of the decision-maker in question, are irrelevant to the selection of the standard of review: “it is the very fact that the legislature has chosen to delegate authority which justifies a default position of reasonableness review” (at para. 30, emphasis original). This is dressed up (in garb provided by the amici curiae) as “institutional design choice” (at para. 30) but in reality reflects a judicial choice to do away with the complexities of a contextual approach.

The corollary is that where the legislature has provided for an appeal, correctness review applies in all circumstances: “Where a legislature has provided that parties may appeal from an administrative decision to a court, either as of right or with leave, it has subjected the administrative regime to appellate oversight and indicated that it expects the court to scrutinize such administrative decisions on an appellate basis” (at para. 36. Legislated standard of review frameworks must also be respected: paras. 34-35).

This is a significant change. As Abella and Karakatsanis JJ note, “the majority’s reasons strip away deference from hundreds of administrative actors subject to statutory rights of appeal” (at para. 199). The majority’s rejoinder is essentially that there is “no principled rationale for ignoring statutory appeal mechanisms” (at para. 45). With respect, however, this rejoinder is aimed at a straw man. The balance of the academic and judicial criticism cited by the majority (at paras. 38-46) was not that statutory appeal provisions should never be taken into account. It has been more common to propose a nuanced approach which would take account of the difference between, say, an appeal on a point of law with leave of a court and a full de novo appeal right. Such nuances are, on the majority’s view, irrelevant: “While the existence of a leave requirement will affect whether a court will hear an appeal from a particular decision, it does not affect the standard to be applied if leave is given and the appeal is heard” (at para. 50. See also the comments at para. 52). In the Vavilov framework, appeal rights equal correctness review, context and nuance be damned.

Nuance cannot be avoided forever, though. Consider, for instance, the certified question regime in immigration law. Immigration decisions are subject to judicial review in the Federal Court. Under Vavilov, reasonableness will presumptively be the standard of review. From the Federal Court, however, there is an appeal on a certified question of law. What role should the Federal Court of Appeal play in immigration law post-Vavilov? Should it apply the appellate standard of review (presumably leaving questions of fact off the table as outside the “scope” of the appeal: para. 37)? Or should it also consider factual matters addressed by the Federal Court, as has been the case under Supreme Court authority such as Baker and Kanthasamy? This is one example but there will be many other nuances to work through (see especially para. 52).

And one wonders whether, despite the injunction to perform correctness review in such cases, whether courts hearing appeals from specialized administrative decision-makers will nevertheless give significant or perhaps even dispositive weight to the decision-makers’ views on matters within their expertise.

Any other departures from the starting point of the presumption of reasonableness review are only justifiable by reference to the rule of law. These are the categories set out in Dunsmuir, minus true questions of jurisdiction: “respect for the rule of law requires courts to apply the standard of correctness for certain types of legal questions: constitutional questions, general questions of law of central importance to the legal system as a whole and questions regarding the jurisdictional boundaries between two or more administrative bodies” (at para. 53). The majority’s conception of the oft-controversial concept of the rule of law is quite thin: it is engaged in situations where “consistency” and thus “a final, determinate answer” to a legal question is necessary (at para. 53).

Three points are notable here: the expansion of the “central questions” category, the elimination of jurisdictional questions and the letting of the door ever-so-slightly ajar to the possibility that a new rule-of-law category will be recognized in the future.

With expertise now out of the way, it no longer performs a limiting function in the definition of “questions of central importance to the legal system”. As the concurring judges warn, this “inevitably” widens the scope for intrusive judicial oversight of expert bodies: “Issues of discrimination, labour rights, and economic regulation of the securities markets (among many others) theoretically raise questions of vital importance for Canada and its legal system [and could require correctness review]” (at para. 244). The majority insists that “questions of central importance are now transformed into a broad catch-all category for correctness review” simply “because expertise no longer plays a role in the selection of the standard of review” (at para. 61). Only time will tell, however, whether the broad coalition in the majority will fracture as future cases present enticing arguments that some question or other is one of vital national importance. Lower courts too will, undoubtedly, face such arguments in the near future.

Jurisdictional questions seem to have been finally consigned to the dustbin of history. It is not “necessary to maintain this category of correctness review” as the concerns about decision-makers overstepping the boundaries of their authority can be addressed by suitably robust reasonableness review (at para. 67). Herein lies the rub. Jurisdictional questions have such a stubborn hold on the legal imagination that any measure short of a stake through the heart cannot be assured of success. In future cases, some members of the Vavilov majority coalition might use the insistence that reasonableness review “does not give [administrative decision-makers] licence to enlarge their powers beyond what the legislature intended” and that the “governing statutory scheme will always operate as a constraint on administrative decision makers and as a limit on their authority” (at para. 68, emphasis added) as justification for non-deferential reasonableness review. This would represent, as I will suggest in my next post, an erroneous reading of Vavilov but it is not difficult to envisage circumstances in which such an error might be made by the Supreme Court or a lower court.

Portentously, the “rule-of-law” door is left slightly ajar. The majority “would not definitively foreclose the possibility that another category could be recognized as requiring a derogation from the presumption of reasonableness review in a future case” (at para. 70). In language reminiscent of Alberta Teachers and CHRC 2018, the reader is warned that any new category would be “exceptional” (at para. 70). If history is any guide, however, such equivocation will be treated by lawyers as a wedge with which to open another door to correctness review (which, by virtue of the place of appeals in the Vavilov framework, will be more familiar to the courts than it has been in the recent past). Again, whether the coalition can hold and whether lower courts resist the temptation to take a peek behind the rule-of-law door is impossible to tell at this point.

In conclusion, the removal of the “vexing” contextual factors (at para. 200) will doubtless bring good cheer to many Canadian lawyers. I suspect the majority consensus also reflects a consensus in the country’s legal community. Time will tell, however, whether this consensus holds or fractures as context and nuance exert the pressure they will inevitably exert.

This content has been updated on December 20, 2019 at 18:50.