Vavilov: Alive at Five

Today marks the fifth anniversary of the Supreme Court of Canada’s landmark decision in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 SCR 653.

2019 is really a world away. On the afternoon of the release of the decision, I held a dial-in session where I delivered my thoughts into a telephone from an office at the University of Ottawa. The session was a bit of a disaster! The dial-in technology used by the university only allowed 100 callers at a time — and many more than that wanted to dial in. Nonetheless, whilst fielding emails from those irritated they could not join the call, I managed to get my thoughts across, until someone dialed in from what sounded like a wind tunnel and did not mute the line, making it impossible for me to think and for anyone to hear me. We have come a long way since then. Nowadays, I would send out a Zoom link and people could join from around the world, with a minimum of fuss. How times have changed.

Vavilov itself is in rude health. I opened that call by riffing on Benjamin Franklin, describing Vavilov as a compromise, if you can keep it. Kept it has been.

Compare and contrast with its predecessor at this point in its life span. Five years after the decision in Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 SCR 190,, Professor Mullan published an article entitled “Unresolved Issues on Standard of Review in Canadian Judicial Review of Administrative Action- The Top Fifteen!” (2013), 42 Adv. Q. 1, tipping the scales at 85 pages. Questions about the scope and nature of judicial review abounded. Consider the numerous issues identified by Stratas JA in Toussaint v. Canada (Attorney General), 2011 FCA 213, [2013] 1 FCR 374, at para. 19, about how much deference should be paid to a ministerial delegate interpreting an order in council:

(a) We are dealing with a ministerial delegate, not a “tribunal” in any formal sense. In Dunsmuir the Supreme Court used the word “tribunal” on this point. In my view, although it is not perfectly clear, in Dunsmuir the Supreme Court did not intend to restrict this position of deference to interpretations by formal tribunals. Throughout its discussion of the standard of review, the Supreme Court used the terms “tribunal”, “decision maker”, “exercises of public authority”, “administrative bodies”, “adjudicative tribunal”, “adjudicative bodies”, “administrative tribunal”, and “administrative actors”: Dunsmuir, above, at paragraphs 28–29, 31, 33, 41, 47–50, 52, 54–56, and 59. It seems to have used the terms interchangeably and, collectively, they are wide enough to embrace a ministerial delegate such as the director.

(b) In a relatively recent decision, albeit before Dunsmuir, the Supreme Court did not defer to the interpretation of a ministerial delegate who was interpreting a statute closely related to his function: Hilewitz v. Canada (Minister of Citizenship and Immigration); De Jong v. Canada (Minister of Citizenship and Immigration), 2005 SCC 57, [2005] 2 S.C.R. 706 (a visa officer making an assessment under subparagraph 19(1)(a)(ii) of the Immigration Act, R.S.C., 1985, c. I‑2); see also Patel v. Canada (Citizenship and Immigration), 2011 FCA 187, [2013] 1 F.C.R. 340 and cases cited at paragraph 27 of Patel. This is certainly consistent with how we today approach decisions involving some other ministerial delegates. For example, in the income tax context, income tax assessors—ministerial delegates—are very familiar with the Income Tax Act [R.S.C., 1985 (5th Supp.), c. 1]. One might think that the normal administrative law standard of review analysis would apply to appeals of these administrators, with deference to their legal interpretations being the result: see, e.g., Canada (Director of Investigation and Research) v. Southam Inc., 1997 CanLII 385 (SCC), [1997] 1 S.C.R. 748 and Dunsmuir, above, at paragraph 54. But it does not. The Tax Court of Canada, sitting in appeal on income tax assessments, and this Court do not defer at all to the statutory interpretations of the Minister’s delegate.

(c) The Supreme Court spoke in Dunsmuir of deference to interpretations of certain types of “statutes”. Did it mean to restrict this principle to “statutes”? There would appear to be no principled basis to do so. Deference probably also applies to interpretations of other types of laws, such as the Order in Council in this case.

(d) The director’s title seems to suggest that he administers programs such as this, and so he could be considered to be interpreting what Dunsmuir [at paragraph 54] described as a law “closely connected to [his] function”, warranting our deference. But there is no evidence in the record on this one way or the other, nor would one expect there to be such evidence given the narrow nature of a record on judicial review.

(e) The position of deference for administrative interpretations of statutes is said in Dunsmuir to apply only “usually” or “normally”. Does this qualification refer to the situations mentioned in Dunsmuir where the correctness standard applies? Perhaps not, as these situations largely do not involve issues of statutory interpretation. Does this qualification refer to some as yet unidentified situations? We simply do not know.

(f) In this particular case, as we shall see, the director did not engage in any actual interpretation of the Order in Council. Rather, he simply interpreted and applied an administrative policy made under that Order in Council. Does this mean that the director’s decision is subject to correctness review? I am not so sure. There are statements in Dunsmuir that suggest that the director’s failure to interpret the Order in Council may not matter. In two places in Dunsmuir, the Supreme Court suggests that in assessing the substance of decision making under the reasonableness standard we are to examine the outcome reached by the decision maker and not necessarily the plausibility of the reasons actually given. At paragraph 47, we are directed to ask ourselves “whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law” and at paragraph 48 we are told that an administrative decision can be supported on the basis of reasons that “could [have] be[en] offered” (emphasis added).

(g) I am not alone in my doubts on this issue. Recently, this Court discussed Dunsmuir and the standard of review that should apply to the Governor in Council’s interpretation of a statute. It found the law in this area to be unclear: Public Mobile Inc. v. Canada (Attorney General), 2011 FCA 194, [2011] 3 F.C.R. 344, at paragraph 35.

Things did not get any better, with splits on the Supreme Court and dissatisfaction on the lower courts around the country creating an unenviable and untenable position. So muchso that before the Vavilov revolution, I was sceptical that the Court would be able to cobble together a workable majority for a set of workable propositions.

How wrong I was!

In a remarkable act of leadership, Chief Justice Wagner managed to put together a strong majority for a set of propositions that have proved eminently workable over the past five years.

Vavilov did not resolve all questions about judicial review of administrative action. But it resolved most of them and left only a handful which, in turn, the Court has sought to resolve in the post-Vavilov era. Here, it is notable that many of the decisions have been unanimous (Law Society of Saskatchewan v. Abrametz, 2022 SCC 29 on procedural fairness; Commission scolaire francophone des Territoires du Nord-Ouest v. Northwest Territories (Education, Culture and Employment), 2023 SCC 31 on Charter values; Yatar v. TD Insurance Meloche Monnex, 2024 SCC 8 on the constitutional foundations of judicial review; Auer v. Auer, 2024 SCC 36 on judicial review of regulations).

Where novel questions arise, Vavilov provides a framework to address them (e.g. the problem of ‘boilerplate’ reasons for decision). Where there are divergences between courts of appeal (as James Casey notes is the case in respect of professional regulation in his contribution to Professor Mancini’s invaluable ‘Vavilov at 5′ symposium), Vavilov provides the tools to resolve them. Where the Court arguably gets it wrong in its application of the reasonableness standard, a critique can be made on the basis of Vavilov rather than by an appeal to first principles.

I concur with Justices Stratas and Rowe that this is a significant achievement. Credit to the Chief Justice, under whose tenure the number of concurrences and dissents has risen, for producing and maintaining consensus in an area of Canadian law long plagued by disagreement.

I remember the ‘Dunsmuir Decade’. I spent more of it than I would have liked chronicling the difficulties caused by a framework for judicial review of administrative action that neither resolved all existing questions nor provided adequate resources to resolve novel questions or provide meaningful critiques. My abiding memory of these years is of throwing my hands up in frustration. Vavilov has been a huge improvement, not least for lawyers and judges who have to argue and decide cases: anecdotally, everyone is happier than they were under the ancien regime.

True, part of Vavilov’s success is due to its ‘all things to all people’ quality — how can both applicant and respondent counsel be so positive about the decision? — and its vagueness on key issues, like the appropriate approach to review of the reasonableness of interpretations of law. Again, though, I tend to think that where Vavilov does not provide a first-order answer to a question, it at least provides second-order resources to construct arguments about the answer to a question.

All in all, the last five years have been very good. Five more please!

This content has been updated on December 19, 2024 at 11:52.