Unresolved Issues after Vavilov II: The Doré Framework

In Doré v. Barreau du Québec, 2012 SCC 12, the Supreme Court of Canada held that alleged infringements of Charter rights by administrative decision-makers should be reviewed on the deferential reasonableness standard. What matters is not whether the decision survives the rigours of the proportionality test set out in R v. Oakes but whether it represents an appropriate balance between Charter values and the decision-maker’s statutory objectives.

I criticized Doré at the time and, although I recognize that Doré provides valuable guidance to administrative decision-makers (especially on the front lines), I continue to think that the Doré approach to judicial review is insufficiently protective of Charter rights. It is true that more recent applications of Doré have hewed quite closely (in substance if not in rhetoric) to the Oakes test, but this simply provides further ammunition for those who would return to the pre-Doré position: the law as stated by the Supreme Court should be in line with the law as applied by the Supreme Court. In the meantime, especially on lower courts, there is a risk that Doré will lead to under-protection of fundamental rights.

All this said, however, I am not at all persuaded by the argument that the decision in Vavilov kicks the conceptual legs from under Doré. Mark Mancini argues that, one, the demise of expertise in Vavilov and, two, Vavilov’s relatively formalist, Diceyan approach to reasonableness review mean that Vavilov and Doré are in serious tension:

On one understanding, Vavilov tends to revert to a Diceyan understanding of administrative law, under which courts reserve to themselves the final say on certain issues. It also shows a focus on justification, as a doctrinal requirement in most cases. However, Doré is rooted in a more functionalist understanding of administrative law, under which expertise is taken as a given and administrators are seen as competent to contribute to the content of the law.

By contrast, I would say that Doré emerges strengthened from Vavilov, not weakened.

First, the excision of expertise from the process of selecting the standard of review means that the presumption of reasonableness review certainly applies to Charter issues. In Vavilov, the majority makes a distinction between judicial review of the “merits” of an administrative decision and issues of “procedural fairness” or “natural justice” (at para. 23). On anything to do with the merits of an administrative decision, the Vavilov framework applies and, in that framework, reasonableness is the presumptive standard (at para. 23). Expertise and other substantive considerations are, simply, unnecessary. Reasonableness review is the “starting point” whether the decision-maker has any relevant expertise or not (at para. 23). Inasmuch as expertise was a conceptual basis for deference in Doré, its removal is irrelevant, as it has simply been replaced by another conceptual basis which is (at least) equally solid.

In fact, the conceptual framework of Vavilov supports the continued application of Doré. Exceptions to the presumption of reasonableness review can only be based on legislative intent or the rule of law (at para. 32). In the absence of federal or provincial legislation requiring correctness review for Charter questions, it is only where the rule of law is engaged that Charter issues will be subject to correctness review under the Vavilov framework. But the rule of law, as defined in Vavilov, is engaged only where a “final and determinate” judicial interpretation is necessary to ensure “consistency” (at para. 53).

This rule-of-law exception to the presumption of reasonableness review is engaged in very limited circumstances, described as constitutional questions, questions of central importance to the legal system and questions of overlapping jurisdiction.

What unites these circumstances, conceptually, is the need for judicially imposed uniformity. Professional privilege is an example of a question of central importance to the legal system: if the scope of privilege were to vary depending on whether it was invoked in professional discipline proceedings or access to information proceedings, professional privilege would be undermined; a uniform approach is necessary. Questions of overlapping jurisdiction, similarly, require judicially imposed ‘right answers’: problems would quickly result were Tribunal A and Tribunal B both to claim, concurrently, jurisdiction over the same subject matter.

As to constitutional questions, the same logic suggests facial challenges to the constitutionality of legislation should be given a uniform answer — for the constitutionality of a statute should not depend on whether the statute is relied upon in front of Tribunal A or Tribunal B — and, accordingly, reviewed on a correctness standard (see Doré at para. 38). It is also arguable (and, I think, consistent with the jurisprudence) for questions relating to the scope of Charter rights to be dealt with on a correctness standard. As I have written, there is nothing novel in treating threshold questions of constitutionality as requiring correctness review: see, on the scope of the duty to consult, Rio Tinto Alcan Inc. v. Carrier Sekani Tribal Council, 2010 SCC 43, [2010] 2 S.C.R. 650, at para. 67, and on the scope of a Charter right, s. 2(a), Ktunaxa Nation v. British Columbia (Forests, Lands and Natural Resource Operations), 2017 SCC 54, [2017] 2 S.C.R. 386, at paras. 68-75. Perhaps it will soon be made clear that these threshold questions fall into one of the correctness categories: they are, after all, situations in which the courts ought to provide a final, definitive answer, as the application of the Constitution or the scope of Charter rights should not vary as between different regulatory regimes.

But the discussion in Doré is oriented towards the question of the proportionality of individualized exercises of discretion which infringe Charter rights (or values). Here, it seems to me, answers can legitimately vary as between different regulatory regimes: for example, what is a proportionate restraint on freedom of expression in the workplace may not be proportionate in a municipal election campaign (see Doré at paras. 54, 56). I can see how professional privilege would be undermined by variations in approach in different regulatory regimes; I can see how incoherence might result from different approaches to jurisdictional overlaps; and I can see how the constitutionality of a statute, or the scope of a Charter right, must be the same across the board. Correctness review in such instances rests solidly on the narrow rule of law basis established in Vavilov. With regret, however, I cannot see why the presence of a Charter right requires uniform answers to be furnished by judges in respect of decisions made in different settings by different decision-makers.

I accept Mancini’s point that the scope/application distinction may not be extremely robust. But the question for present purposes is not the robustness of the distinction but whether Doré and Vavilov are compatible. Given the replacement of expertise as the conceptual basis for deference with an across-the-board presumption of reasonableness review and the narrowness of Vavilov’s rule-of-law exception, I do not think there is any incompatibility between Doré and Vavilov.

Second, I do not think Vavilovian reasonableness review can fairly be described as formalist or Diceyan. As I have suggested, Vavilov is an example of the “culture of justification” in administrative law. There is nothing formalist about the detailed articulation of reasonableness in Part III of Vavilov. Indeed, the repeated references to the “demonstrated expertise” of administrative decision-makers strike an unmistakeably functionalist tone. Expertise might now be irrelevant to selecting the standard of review but it is very much relevant to surviving the standard of review.

It is true, as I noted in my commentary on Vavilov, that there are tensions in the majority’s articulation of reasonableness review. Some components of Vavilovian reasonableness review can fairly be described as formalist or Diceyan: the emphasis on the importance of the governing statutory scheme, for example. But reasonableness review post-Vavilov is to begin with the reasons provided by the administrative decision-maker, even where the reasons touch on jurisdictional issues. There is nothing formalist or Diceyan about this. Read fairly, Vavilovian reasonableness review has both formal and functional, Diceyan and non-Diceyan components.

The discussion of the principles of statutory interpretation is perhaps the best example. On the one hand, administrative decision-makers are to apply the principles as courts would (at para. 118). On the other hand, a “formalistic” statutory interpretation exercise is not required in every case (at para. 119). The majority in Vavilov does not go as far as I would advocate but I find it very doubtful that Dicey would have rejoiced at the idea that judicial review would begin, not with the judge’s view of the best reading of a statute but with the reasons provided by the administrative decision-maker “applying its particular insight into the statutory scheme at issue” (at para. 121).

At the core of reasonableness review in Doré was “balancing Charter values against broader objectives” (at para. 57), with courts obliged to uphold an appropriate balance struck by the decision-maker (at para. 58). This is just as possible post Vavilov as it was before. Administrative decision-makers can continue to contribute to our collective understanding of the Charter in its application to particular regulatory settings.

In sum, I have long thought that Doré was a misstep in the Canadian law of judicial review of administrative action. But a post-Vavilov correction is not at all inevitable. Those who wish to see the back of Doré will have to attack it directly and hope their attacks resonate with a majority of the Supreme Court of Canada.

This content has been updated on May 6, 2020 at 12:56.