The Vavilov Framework II: Reasonableness Review

In their hard-hitting concurring reasons in Vavilov, Abella and Karakatsanis JJ charge the majority with “reviv[ing] the kind of search for errors that dominated the pre-C.U.P.E. era” (at para. 199). Although there are some differences of detail, and some internal tensions in the majority’s articulation of a new methodology for reasonableness review, on balance the majority is right that the dissent’s approach is not “fundamentally dissimilar” (at para. 75).

On the key propositions, all nine judges are ad idem: reasonableness review is robust; reasons are fundamental to the legitimacy of administrative decision-making; unreasonableness must be demonstrated by the applicant; reasonableness review should begin with the reasons given by the administrative decision-maker; reasonableness review is contextual; and reasonableness review should be conducted with a healthy appreciation that “‘[a]dministrative justice’ will not always look like ‘judicial justice’ (at para. 92).

Notice, before delving into the details, that when the reformulation of reasonableness review is placed alongside the rearticulation of the framework for selecting the standard of review, some fascinating antimonies emerge. Although contextual factors have been swept away from the selection of the standard of review, they are central to the approach to reasonableness review articulated by the majority. Whereas the brute fact of a legislative choice to create an administrative decision-maker is sufficient to create an almost irrebuttable presumption of reasonableness review, in its articulation of reasonableness review, the majority emphasizes the need for reasons to be “responsive” (at paras. 127 and 133) and stresses “the principle that the exercise of public power must be justified, intelligible and transparent, not in the abstract, but to the individuals subject to it” (at para. 95) with legitimacy flowing ultimately from the reasoned exercise of authority granted by legislation, not the mere fact of its existence. And expertise is irrelevant to selecting the standard of review but the “demonstrated experience and expertise” of an administrative decision-maker will help to support the conclusion that a given decision was reasonable (at para. 93, emphasis added). Read holistically, the majority reasons setting out the Vavilov framework have something for everyone.

In the reasonableness review part of the Vavilov framework, there is significant consensus but there are also tensions in the majority reasons.

As the majority acknowledges, in its previous decisions the Court has provided “relatively little guidance on how to conduct reasonableness review in practice” (at para. 73). It made up for this in Vavilov. The majority begins by setting out its key propositions (with which Abella and Karakatsanis JJ do not quibble):

• “Reasonableness review is…a robust form of review” (at para. 13. See the concurring reasons at para. 294);
• “where reasons are required, they are the primary mechanism by which administrative decision makers show that their decisions are reasonable — both to the affected parties and to the reviewing courts” (at para. 81. See the concurring reasons at paras. 291 and 296);
• “The burden is on the party challenging the decision to show that it is unreasonable” (at para. 100. See the concurring reasons at para. 312);
• “a court applying the reasonableness standard does not ask what decision it would have made in place of that of the administrative decision maker, attempt to ascertain the ‘range’ of possible conclusions that would have been open to the decision maker, conduct a de novo analysis or seek to determine the “correct” solution to the problem…A principled approach to reasonableness review is one which puts [the decision-maker’s] reasons first” (at paras. 83-84. See the concurring reasons at paras. 306 and 313);
• “what is reasonable in a given situation will always depend on the constraints imposed by the legal and factual context of the particular decision under review” (at para. 90. See the concurring reasons at paras. 292-293); and
• “In conducting reasonableness review, judges should be attentive to the application by decision makers of specialized knowledge, as demonstrated by their reasons. Respectful attention to a decision maker’s demonstrated expertise may reveal to a reviewing court that an outcome that might be puzzling or counterintuitive on its face nevertheless accords with the purposes and practical realities of the relevant administrative regime and represents a reasonable approach given the consequences and the operational impact of the decision” (at para. 93. See the concurring reasons at paras. 297-299).

In addition, the majority insists that where reasons are defective, a reviewing court is not “to fashion its own reasons in order to buttress the administrative decision” (at para. 96). Abella and Karakatsanis JJ do not accept this proposition (at paras. 302-305) but it is surely fair to say that even if it did not quite attract a consensus, it nonetheless reflects the recent direction of travel, which has moved the Supreme Court away from the laissez-faire attitude it had to defective reasons in the early part of this decade. Alberta Teachers, notably, is limited to its particular facts (at para. 98), and there is no repetition of some of the more permissive language of the most recent leading precedent on defective reasons, Delta Air Lines, which is cited instead for the principle that “it is not open to a reviewing court to disregard the flawed basis for a decision and substitute its own justification for the outcome” (at para. 96).

In terms of the methodology of reasonableness review, this is about it. The majority goes on, at some length, “to consider two types of fundamental flaws” but emphasizes that these flaws are simply “a convenient way to discuss the types of issues that may show a decision to be unreasonable” (at para. 101). First, the absence of “reasoning that is both rational and logical” (at para. 102), such as reasons which “fail to reveal a rational chain of analysis”, ones which “read in conjunction with the record do not make it possible to understand the decision maker’s reasoning on a critical point” (at para. 103), or ones which “exhibit clear logical fallacies, such as circular reasoning, false dilemmas, unfounded generalizations or an absurd premise” (at para. 104). Plainly, these are intended as examples which illustrate a general point — the absence of logic and reason — and not as a set of categories into which dubious administrative decisions can be pigeonholed by reviewing courts (though, I am sure, this will be a tempting course of action for lower courts, urged upon them by clever counsel).

Second, a decision must be “justified in relation to the constellation of law and facts that are relevant to the decision” (at para. 105). The majority emphasizes that it is impossible to “catalogue” all the considerations which will be relevant to the constellation of particular individual cases but sets out a set which will “generally be relevant”:

…the governing statutory scheme; other relevant statutory or common law; the principles of statutory interpretation; the evidence before the decision maker and facts of which the decision maker may take notice; the submissions of the parties; the past practices and decisions of the administrative body; and the potential impact of the decision on the individual to whom it applies (at para. 106).

That these “elements” are not intended as “a checklist for conducting reasonableness review” (at para. 106) clearly emerges from the ensuing discussion where the formulation “may be unreasonable” is repeatedly employed.

Despite the scholarly analysis in the majority’s reasons and notwithstanding the large area of common ground between the majority and their concurring colleagues, there are tensions in the majority reasons which will have to be worked out in future cases.

Consider, first, this passage: “Before a decision can be set aside [for unreasonableness], the reviewing court must be satisfied that there are sufficiently serious shortcomings in the decision such that it cannot be said to exhibit the requisite degree of justification, intelligibility and transparency” (at para. 100). What is the operative test here? Are “justification, intelligibility and transparency” to be taken as the (high) bar administrative decision-makers are expected to scale (consistent with the emphasis on the principle of responsive justification)? Or are reviewing courts to look only for the presence of “sufficiently serious shortcomings” rather than the absence of justification, intelligibility and transparency”? The latter answer seems to me more consistent with the tenor of the majority reasons but lower courts and future Supreme Court coalitions could conceivably prefer the former. [UPDATE, added December 22: Indeed, whereas in Dunsmuir one of the issues for a reviewing court was whether a decision “falls” the range of possible acceptable outcomes in view of the facts and the law, in Vavilov the issue is whether the decision is “justified in relation to the relevant factual and legal constraints that bear on the decision” (at para. 99). It seems to me that this is not merely a difference of emphasis: requiring justification of a decision rather than merely falling within a range is a more demanding standard.]

Consider, second, the discussion (at paras. 108-110) of the relevance of the “governing statutory scheme” to the constellation of law and fact in which the justifiability of an administrative decision is to be assessed. Whereas in respect of the other contextual considerations considered by the majority the permissive term “may” is invariably used, it is replaced by the imperative “must” in paras. 108-110. For example, “[a]lthough a decision maker’s interpretation of its statutory grant of authority is generally entitled to deference, the decision maker must nonetheless properly justify that interpretation” (at para. 109, emphasis added). Must, not may.

Furthermore, the two closing sentences of the section are almost irreconcilable: “What matters is whether, in the eyes of the reviewing court, the decision maker has properly justified its interpretation of the statute in light of the surrounding context. It will, of course, be impossible for an administrative decision maker to justify a decision that strays beyond the limits set by the statutory language it is interpreting” (at para. 110, emphasis added). The first sentence posits a test of justification, the second sentence posits clear limits enforceable by a reviewing court. What to make of this? Have jurisdictional questions been killed off as far as selecting the standard of review goes only to re-emerge, vital as ever, in the application of the reasonableness standard? On balance, I think the first sentence must prevail. After all, the majority insists that, in general, a reviewing court “may” rely on the contextual considerations to support a finding of unreasonableness (at para. 107). But courts minded to police what they perceive to be jurisdictional boundaries can certainly fasten on the language in paras. 108-110 to engage in intrusive reasonableness review.

There are other examples of slippage from the permissive to the imperative: compare para. 111 with para. 112; and see paras. 126 and 133. But none seems as portentous as the slippage in paras. 108-110.

Consider, third, the discussion of the principles of statutory interpretation. Whilst it is pellucidly clear (and most welcome) that a reviewing court is not to conduct its own statutory interpretation exercise to establish a benchmark or yardstick against which to measure an administrative decision-maker’s interpretation of law (at para. 124), the finer details are murky. On the one hand, the reader is told: “Administrative decision makers are not required to engage in a formalistic statutory interpretation exercise in every case” (at para. 119). On the other hand, a few paragraphs later, the administrative decision-maker’s task is said to be to “interpret the contested provision in a manner consistent with the text, context and purpose, applying its particular insight into the statutory scheme at issue” (at para. 121). That looks awfully like a “formalistic statutory interpretation exercise”, one which judges suspicious of an administrative decision-maker’s ability to issue interpretations of law might well require. I hope such judges take particular note of the majority’s insistence that sometimes an administrative decision-maker need “touch upon only the most salient aspects of the text, context or purpose” (at para. 122). But there is a risk that many anti-deference judges will fasten upon the emphasis on text, context and purpose to constrain administrative interpretations of law. Here, I would wholeheartedly endorse the approach set out by Abella and Karakatsanis JJ at paras. 306-311.

Fourth, Vavilov was a case where reasons were “available and required” (at para. 78). Not very fulsome reasons, but reasons nonetheless. In such cases, the direction to begin with the administrative decision-maker’s reasons is easy to follow. But in other cases, where reasons are sparse or non-existent, deferential review will be harder to conduct. The majority is quite open about this: “it is perhaps inevitable that without reasons, the analysis will then focus on the outcome rather than on the decision maker’s reasoning process” (at para. 138). In general, most administrative decision-makers now provide reasons as a matter of course. But this will not always be the case. Going forward, the absence of reasons, or presence of sparse reasons, may function as an invitation to reviewing courts to conduct an outcome-focused assessment of the decision, in which substitution of judgment is an ever-present risk. Abella and Karakatsanis JJ admonish that, in such scenarios, “a reviewing court should remain focussed on whether the decision has been shown to be unreasonable” (at para. 312), but the majority seems to have little patience with non-existent and sparse reasons, consistent with its emphasis on the primacy of reasoned decision-making.

These tensions will have to be worked out in future cases. Judicial review hawks might resolve them in different ways to judicial review doves. Again, the consensus may not hold; the Vavilov coalition may fracture, with different members emphasizing different aspects of Vavilov. What lower courts do with these tensions will be fascinating and may soon portend just how effectively the Vavilov framework will achieve the majority’s goals of increased clarity and predictability.

This content has been updated on December 23, 2019 at 02:50.