Deference on Questions of Procedural Fairness after Law Society of Saskatchewan v. Abrametz, 2022 SCC 29, The Implications

In my last post, I explained the position we took on the standard of review of procedural fairness issues in Law Society of Saskatchewan v. Abrametz, 2022 SCC 29 (overview here). In this post, I will tease out the implications of the decision.

During my oral submissions (all seven minutes of them!), Justice Rowe asked whether it was not “metaphysical” to switch from appellate standards to judicial review standards: in other words, when hearing an appeal from the Hearing Committee, why should a court apply the appellate standards on issues of substance but not on matters of procedure? I avoided giving the principled answer to this question: our goal at the hearing was to convince the Supreme Court that the Court of Appeal should have, but failed to, defer to the Hearing Committee; we had other points to make, especially about the quality of the Hearing Committee’s analysis of the facts, and did not want to get bogged down in the metaphysics of standard of review.

Justice Rowe’s question was portentous, for he ended up writing the majority judgment. I was not surprised that he simply applied the appellate standards of review:

In Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, the Court held that when the legislature provides for a statutory appeal mechanism from an administrative decision maker to a court, this indicates that appellate standards are to apply: paras. 33 and 36-52. While this proposition was stated in the context of substantive review, the direction that appeals are to be decided according to the appellate standards of review was categorical. Thus, where questions of procedural fairness are dealt with through a statutory appeal mechanism, they are subject to appellate standards of review (at para. 27).

So, what happens next?

Procedural Fairness within the Vavilov Framework

Justice Rowe’s line of analysis sends us back to pre-Vavilov times as far as procedural fairness is concerned, in the following sense.

The years after Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 SCR 190 featured debates about the place of procedural fairness in the Supreme Court’s administrative law framework (see here and the links therein). These debates arose because the Dunsmuir framework purported to be comprehensive and, indeed, the Supreme Court suggested, by referring to a standard of correctness on procedural fairness questions, that procedural fairness had been or should be subsumed into the Dunsmuir framework. As judges and commentators pointed out, some procedural fairness questions, because they involve fact or discretion or interpretation of a decision-maker’s home statute, fell into Dunsmuir’s reasonableness categories; for my part, I argued for unification of the poles of procedure and substance.

Vavilov, on the understanding I shared with Justice Côté, brought that debate to an end by making clear that the framework therein set out would only apply to matters of substance, not to matters of procedure.

But Abrametz brings procedural fairness into the Vavilov framework. It does this quite obviously in respect of procedural fairness issues which arise by way of statutory appeal: here it is straightforwardly the case that correctness applies to questions of law but palpable and overriding error to everything else.

As for procedural fairness issues arising by way of judicial review, the position is less straightforward.

Justice Rowe attempts to limit the effect of his analysis by insisting that he does not “depart” (at para. 28) from the Supreme Court’s prior procedural fairness jurisprudence. But he refers to two cases in which the standard of review of procedural fairness was stated to be “correctness”: Canada (Citizenship and Immigration) v. Khosa, 2009 SCC 12, [2009] 1 S.C.R. 339, and Mission Institution v. Khela, 2014 SCC 24, [2014] 1 S.C.R. 502. “Correctness” is one of the Vavilov standards. As such, we are back where we were a decade ago, wondering about whether and where procedural fairness questions fit in the Supreme Court’s administrative law framework.

Now, it may be that, going forward, the lower courts will uncritically accept that “correctness” applies to procedural fairness issues arising on judicial review. Even leaving aside the bifurcation between appeals (where procedural fairness issues will be subject to the deferential standard of palpable and overriding error in some cases) and judicial review, however, it is difficult to square the application of “correctness” on procedural fairness questions with the limited scope of the “correctness” categories set out in Vavilov. Correctness review is appropriate only where the rule of law requires consistent treatment of legal issues. Can it really be said that the contemporary, context-sensitive and fact-specific law of procedural fairness requires correctness review? Where an administrative decision-maker has provided detailed reasons on a procedural issue, can a reviewing court simply conduct a de novo assessment? A positive answer to these questions seems as implausible to me as it did a decade ago. The emergence of the “margin of deference” concept in Khela, just a few paragraphs after the invocation of correctness review, is testament to this.

Justice Côté suggests that negative answers are appropriate, for she advocates correctness review on all procedural fairness issues regardless of the route by which they reach the courts (with, however, deference on findings of fact: at para. 174). To begin with, she notes that procedural fairness “exists independently of statutorily confined administrative regimes” (at para. 165). This is true, but it is also true of the doctrine of reasonableness review articulated in Vavilov, as well as the common law concepts which, according to Vavilov, administrative decision-makers can fashion to their own ends. All principles of judicial review are independent in this sense; the question is whether it is appropriate to give deference when they have to be applied in a particular setting.

Justice Côté also argues that procedural fairness, as a common law principle, “falls squarely within the expertise of the judiciary” (at para. 168). Again, even if this is true in the abstract, it will not invariably be true when, for example, an administrative decision-maker refuses an adjournment, or finds that a notice of hearing was adequate, or concludes that someone complaining of undue delay did not suffer prejudice. In these heavily fact-specific scenarios, procedural fairness is not some brooding omnipresence in the sky uniquely accessible to judges but a context-sensitive concept which administrative decision-makers are well placed to apply. Just as administrative decision-makers may be granted deference when implementing common law principles or judicial precedent (Vavilov, at paras. 111-113) so too could they be given deference on procedural fairness issues.

In addition, for Justice Côté, deference is the trade-off for fairness: “deference to substantive decisions presupposes that the administrative process is fair and thus subject to review by courts on a correctness standard” (at para. 167). But there is no authority for this proposition, other than the reference in Vavilov to the requirement to give reasons for a decision which, if breached, undermines the rationale for deference. And deference in procedural fairness cases typically pre-supposes detailed reasons from the administrative decision-maker: deference has to be earned in any event. As long as reasons are required for deference to be afforded, procedure and substance are indistinguishable in Vavilovian terms. Indeed, Justice Côté accepts that administrative decision-maker’s are entitled to deference in some instances, where statutory provisions or factual findings inform the relevant decision-making context (at para. 174). As Professor Mullan commented in the mid-1980s, deference on procedural fairness issues is perfectly appropriate when “the nature of the decision-maker, including its capacities for making procedural judgments (particularly in comparison with the courts’ own expertise in such matters) and the seriousness with which it has dealt with the procedural question under review” militate in favour of judicial restraint (“Developments in Administrative Law: the 1983-84 Term” (1985) 7 Supreme Court Law Review (2d) 1, at p. 19)

I would also observe that, in situations where reasons have not been given on a procedural fairness issue, either because it is raised for the first time on judicial review or because the administrative decision-maker did not deign to explain itself, reasonableness review can still be applied, focusing “on the outcome rather than on the decision maker’s reasoning process” (Vavilov, at para. 138). Without reasons, reasonableness review of procedural fairness issues would amount to a conventional application of the factors set out in Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817 and other relevant jurisprudence.

Lastly, and with respect, Justice Côté’s argument that correctness review is analytically well suited to procedural fairness issues is unconvincing. She writes:

Khela mandates only that the question of compliance with the duty of procedural fairness be subject to the correctness standard; all procedural rulings falling within the bounds of fairness are entitled to deference. For example, while an opportunity for a party to make submissions may be required at common law in a given context, the administrative decision maker might opt for a hearing, written submissions, or both. Where all three options satisfy the duty of fairness, the reviewing court should not substitute its view for that of the administrative decision maker concerning the most appropriate procedure (at para. 177).

But as Professor McKee has demonstrated in his excellent article “The Standard of Review for Questions of Procedural Fairness” (2016), 41 Queen’s Law Journal 355, correctness review and procedural fairness are uneasy bedfellows. In fact, when answering questions of law on a standard of correctness, courts do not proceed as Justice Côté’s analysis might suggest. On correctness review, a court has to identify the best possible interpretation of a statute, not merely determine that there are multiple available interpretations. This is not what courts do in procedural fairness cases: they do not identify the best possible procedure but rather focus on whether the procedure actually used was fair. Justice Côté’s explanation of correctness review suggests that courts should determine the fairest process and then measure the process used against that. But we have long since abandoned the notion that judicial procedures are the “gold standard” against which fairness is measured. Accordingly, “correctness” will always be a poor fit for procedural fairness issues, especially where an administrative decision-maker has provided reasons on the issue in question.

Law, Fact and Fairness

For Justice Rowe, the ultimate question of whether there has been an abuse of process is “a question of law” reviewable on the correctness standard (at para. 30). There are two ways to read this assertion. On one view, the ultimate question in any procedural fairness case, including ones about delay (at para. 38) will be whether the procedure was fair, a question to be answered by the courts without any deference. On an alternative view, however, one only gets to the question of abuse of process after finding that the delay was inordinate and that significant prejudice resulted (at para. 43): conclusions on inordinate delay and/or significant prejudice would be subject to deferential review because they involve factual assessments (at para. 29).

I think the fairest reading of Justice Rowe’s reasons on standard of review, read in the context of the decision as a whole, is that the courts retain the last word on the ultimate issue in a procedural fairness case (i.e. was it “correct” to find (un)fairness?) but that an administrative decision-maker’s reasoned analysis is subject to deference. Notice that Justice Côté does not fundamentally disagree with this proposition, at least as applied to a decision-maker’s “findings of fact”. But Justice Rowe is clear that mixed questions of law and fact — which would include an administrative decision-maker’s application of a legal test to facts — is also subject to deferential review in the context of a statutory appeal (at paras. 29, 105). Indeed, Justice Rowe held that the Court of Appeal had erred by failing to defer to the Hearing Committee. On inordinate delay, the Hearing Committee had made findings which the Court of Appeal should have respected:

The Court of Appeal departed from its proper role when it substituted its own findings of fact, notably on the scale and the complexity of the investigation. The Hearing Committee’s conclusions were grounded in the evidence before it, in particular, the affidavit of the auditor of the Law Society, Mr. Allen. The Court of Appeal gave no deference to the Hearing Committee’s findings, and simply reweighed this evidence and substituted its own findings. The Court of Appeal found that there were errors in the Hearing Committee’s conclusions regarding the delay attributable to Mr. Abrametz: para. 193. However, the Hearing Committee’s finding that, at a certain point, Mr. Abrametz stopped cooperating in the investigation was supported by uncontroverted evidence. There was no proper basis for the Court of Appeal to contradict the Hearing Committee’s attribution of certain delays to Mr. Abrametz. (at paras. 114-115).

Again, in respect of prejudice, the Court of Appeal failed to defer to the Hearing Committee’s analysis of the evidence (at paras. 120, 122 and 123): “It was not shown that the Hearing Committee committed any palpable and overriding error when it concluded that Mr. Abrametz failed to demonstrate any significant prejudice. The Court of Appeal erred by interfering with its findings of fact and findings of mixed fact and law” (at para. 124).

Going forward, then, in statutory appeals of procedural fairness issues, whilst the courts might retain the last word on fairness, they are going to have to defer to administrative decision-makers’ findings of fact and applications of law to fact.

This change could be profound. Most areas of procedural fairness are regulated by settled principles which set out, for example, the required content of notice, the circumstances in which legal representation should be permitted and the grounds for refusing an adjournment. Accordingly, when administrative decision-makers apply those settled principles to the facts before them, they can expect deference from appellate judges, who will only be able to intervene where the administrative decision-maker made a palpable and overriding error.

There is an underlying practical difficulty here, inasmuch as it has not heretofore been necessary to distinguish between law and fact in procedural fairness cases. But courts will now be required to do so, and one imagines that counsel for administrative decision-makers will be quite eager to assist in this endeavour. The “classification game” will henceforth be played on another stage.

There is a further endeavour for clever respondent counsel. Consider appeal clauses which are limited to questions of law or jurisdiction. Courts have traditionally treated breaches of procedural fairness as falling within such clauses: Canadian National Railway Company v. Emerson Milling Inc., 2017 FCA 79, [2018] 2 FCR 573. But if the only question of “law” is the ultimate question of compliance with the duty of fairness, because the decision involved the administrative decision-maker’s application of settled legal principles to facts, the limited appeal clause may not be broad enough to accommodate an appeal: see by analogy Bell Canada v. British Columbia Broadband Association, 2020 FCA 140. This conclusion is not irresistible, because lower courts are likely to hesitate before concluding that such a change has been effected by a sidewind in Abrametz, but it is a legitimate point for counsel to take.

The last observation I would make here, drawing together the analysis in this and the preceding subsection, is that it makes little sense to refuse to defer to an administrative decision-maker’s reasons in a judicial review case but to defer to an administrative decision-maker’s reasons in a statutory appeal. Sauce for the statutory goose should be sauce for the judicial review gander. Deference in statutory appeals should be accompanied by deference on judicial review. Maintaining a bifurcated world in which an administrative decision-maker gets more deference on a statutory appeal than on judicial review strikes me as entirely implausible.

Summing Up

The Attorney General of Quebec submitted in its intervention in Abrametz that the approach to standard of review of procedural fairness issues should be as follows:

(1) On appeal, an administrative decision-maker’s application of law to fact and findings of fact should be subject to review for palpable and overriding error: no intervention is permissible absent an error in principle.

(2) In judicial review cases, where the administrative decision-maker has given reasons for decisions on matters of procedural fairness, these reasons will be subject to reasonableness review consistent with the articulation of reasonableness in Vavilov.

(3) in judicial review cases where no reasons have been given, the court will assess whether the procedural issue was addressed consistently with the legal and factual constraints of procedural fairness case law.

It seems to me that the first of the AGQ’s propositions is entirely consistent with the majority reasons. And for the reasons I have given, the second and third propositions are corollaries of the majority reasons. The only question is how long it will take the Supreme Court and lower courts to bow to the logic of the approach in Abrametz and hold that these corollaries are law.

This content has been updated on August 26, 2022 at 00:15.