Vavilov on Standard of Review in Canadian Administrative Law

By: Shaun Fluker

PDF Version: Vavilov on Standard of Review in Canadian Administrative Law

Case Commented On: Canada (Minister of Citizenship and Immigration) v Vavilov, 2019 SCC 65 (CanLII)

In the summer of 2018, I wrote about disagreement within the Supreme Court of Canada over the role of contextual factors in the selection of a standard of review in Canadian administrative law (see The Great Divide on Standard of Review in Canadian Administrative Law). At that time, the Court had arrived at yet another fork in the road on standard of review and stated it would address the matter head-on in a hearing scheduled for late 2018. Canada (Minister of Citizenship and Immigration) v Vavilov, 2019 SCC 65 (CanLII) (Vavilov) is the result of that hearing, and Vavilov has, once again, amended the law applicable to selecting and applying the standard of review. This post is my analysis of Vavilov, and is organized as follows: (1) an overview on the law regarding standard of review up to Vavilov; (2) the law as per Vavilov on selecting the standard of review; (3) the law as per Vavilov on applying the standard of reasonableness; and (4) a short conclusion. I am admittedly somewhat late to this party. My colleague Nigel Bankes has previously written on an aspect of Vavilov here (which I reference below) and I also encourage readers interested in this topic to have a look at Paul Daly’s analysis of Vavilov here.

Overview on the Law regarding Standard of Review up to Vavilov

The substance of a statutory decision varies extensively from the exercise of broad ministerial discretion, to recommendations made by a board of inquiry, to a determination of legal rights by a statutory tribunal. Despite this diversity in the type of decision-maker or character of decision, the common thread in these decisions is that they are made pursuant to the exercise of power granted in legislation. The review (or appeal heard) by a superior court on the substance of a statutory decision invokes the standard of review analysis.

The first step in a review or appeal on the substance of a statutory decision is for the reviewing court to select the standard of review, and the second step is for the court to apply the chosen standard to the alleged errors contained in the decision. The essential function performed by the standard of review is to determine the measure of deference owed by the court to the findings and conclusions made by the statutory decision-maker.  This has been an ongoing source of difficulty for Canadian courts ever since the era of judicial deference was ushered in by CUPE v NB Liquor Corporation, [1979] 2 SCR 227, 1979 CanLII 23 (SCC) (CUPE 1979), and the problems are most apparent when superior courts are reviewing legal determinations made by statutory decision-makers (as opposed to factual or policy-based decisions). By my count, Vavilov represents the fourth significant overhaul since CUPE 1979, and the Court continues to spin its wheels trying to find a principled resolution to what seems to be an irreconcilable tension: on the one hand, to respect the intention of the legislative branch to empower the executive and its delegates with the authority to decide legal rights and interests; while on the other hand, to supervise the exercise of that authority to ensure it is lawful.

There are many difficulties associated with establishing a principled approach to deference in these cases, and here is just a sampling of them: (1) what might be appropriate deference for the review of legal determinations made by a specialized agency will not be appropriate for an ad-hoc tribunal or possibly even a minister; (2) statutory decisions cover the full range of outcomes from the exercise of full discretion with application to case-specific facts, to the adjudication of legal rights with implications for the law more generally; (3) statutory decision-makers are often charged with implementing policy direction and, by definition, these decision-makers cannot exercise legal power independent of political influence; and (4) statutory decisions-makers may have little experience with legal reasoning and it can be problematic to expect them to be well-versed in the nuances of statutory interpretation or the application of legal doctrine.

Over the past decade, the Supreme Court has been trying to develop an approach to deference that selects a standard of review (correctness or reasonableness) without getting stuck in the weeds of ‘context’. The contextual approach initially arose after CUPE 1979 in a series of decisions that focused on trying to balance the supervisory role of superior courts with a need to defer to the relative expertise of a statutory decision-maker. These decisions are nicely summarized by Justices Abella and Karakatsanis (Vavilov at paras 202-219), who concur in the result of Vavilov but dissent on the standard of review framework set by the majority (I refer hereinafter to their concurring (but dissenting) reasons as the minority in Vavilov).

The contextual approach to selecting a standard of review was explicitly articulated by the Court as the ‘pragmatic and functional approach’ in Pushpanathan v Canada (Minister of Citizenship and Immigration), [1998] 1 SCR 982, 1998 CanLII 778 (SCC). Pushpanathan distilled the selection process into a consideration of four contextual factors related to the statutory decision: (1) the presence or absence of a privative clause or statutory appeal in the governing legislation; (2) the purpose of the statutory regime and the specific powers in question; (3) the nature of the question at issue; and (4) the expertise of the statutory decision-maker relative to the court in relation to the question at issue. None of these factors were determinative of a standard on their own, however a reviewing court was to consider the factors collectively and determine the standard of review for a particular decision. The standard of review analysis was not informed by precedent, and thus a reviewing court would consider these factors each time it was tasked with the determining the standard of review. Even decisions made by the same statutory decision-maker could attract a different standard (and thus a different measure of deference from the reviewing court).

In Dunsmuir v New Brunswick, [2008] 1 SCR 190, 2008 SCC 9 (CanLII), the Court simplified the standard of review selection process by eliminating the ‘patent unreasonable’ standard and by making a series of declarations on the standard of review applicable to certain categories of question decided by a statutory decision-maker (Dunsmuir at paras 51-61). The ‘pragmatic and functional’ factors were renamed the ‘standard of review’ factors, and Dunsmuir ruled that it would no longer be necessary to consider each factor in every case. Much of Dunsmuir seemed like a refinement of Pushpanathan, rather than an overhaul.

The overhaul came later, when the Dunsmuir approach evolved into a presumption that the standard of review is reasonableness (deference owed) for a statutory decision that involves the interpretation by a decision-maker of their ‘home’ legislation. After a series of decisions following Dunsmuir, the leading authority for the presumption became Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd., [2016] 2 SCR 293, 2016 SCC 47 (CanLII). In East Capilano, the Court ruled that the presumption of reasonableness would apply to the interpretation by a decision-maker of their ‘home’ legislation unless the question: (1) relates to the constitutional division of powers; (2) is a true question of jurisdiction; (3) engages in competing jurisdiction between two or more tribunals; or (4) is of central importance to the legal system and outside the expertise of the statutory decision-maker (East Capilano at paras 22-24).

In Supreme Court decisions post-East Capilano, majority judgements upheld the presumption of reasonableness, with strong dissents warning that the presumption had become irrefutable because an assessment of expertise in the statutory decision-maker had evolved away from actual or demonstrated expertise and into a formalistic one based solely on institutional setting or statutory design. These disagreements were usually articulated alongside spirited debates which fractured over two of the mechanisms for rebutting the presumption, namely: (1) true questions of jurisdiction; and (2) questions of general importance to the legal system. In the face of this presumption of reasonableness, the role of  contextual factors in selecting the standard of review was left somewhat uncertain by East Capilano and the decisions which followed it.

The Law as per Vavilov on Selecting the Standard of Review

The majority in Vavilov describes its work as a ‘recalibration’ of the governing approach to selecting the standard of review (at para 143). Vavilov expands the presumption of reasonableness as applicable to any substantive decision made by a statutory decision-maker (in other words, the presumption is no longer limited to decisions that involve the interpretation by a decision-maker of their ‘home’ legislation). This blanket presumption applies to the determination of questions of law (including jurisdiction), fact, or mixed law and fact (Vavilov at para 25). Expertise is eviscerated as a relevant consideration in selecting the standard of review – a superior court is not to consider the expertise of a statutory decision-maker in its selection of the applicable standard of review (Vavilov at paras 30, 31).

Vavilov sets out five exceptions to this presumption, where the standard of review will be a standard other than reasonableness, and bundles these exceptions under two categories: (1) a legislature has indicated that a different standard of review should apply to a statutory decision; and (2) rule of law principles require that a different standard of review should apply to a statutory decision. These two categories are set out below, with my commentary on each of the exceptions included therein.

(1) The Legislature Has Indicated that a Different Standard Should Apply to a Statutory Decision

There are two exceptions to the presumption of reasonableness under this category.

A legislature may explicitly prescribe the standard of review that a superior court must apply in its review of a statutory decision, and where a legislature does so that is the standard which must be applied by a reviewing court (Vavilov at para 35).

Where a legislature provides for a statutory right of appeal to the courts in relation to a statutory decision, and the legislature has not prescribed the applicable standard of review, the standard of review is determined in accordance with the principles of appellate review currently established in Housen v Nikolaisen, [2002] 2 SCR 235, 2002 SCC 33 (CanLII) (Housen) (Vavilov at para 37). In accordance with the principles set out by Housen, the standard of review on a pure extricable question of law emanating from an administrative decision is correctness (Housen at paras 8, 36). The standard of review on a finding of fact is ‘palpable and overriding error’ (Housen at para 10). The standard of review is also ‘palpable and overriding error’ on a question of mixed fact and law where the legal question is not readily extricable (Housen at para 36). The Housen principles apply regardless of whether the statutory appeal has or does not have a leave requirement (Vavilov at para 50). Vavilov makes a point of confirming that a statutory right of appeal is distinct from judicial review, and thus the Housen principles do not apply in selecting the standard of review in a judicial review proceeding (Vavilov at paras 51, 52).

This category most clearly reflects the renewed emphasis on legislative intent (and a departure from expertise) in selecting the standard of review: “Where a legislature has not explicitly prescribed that a court is to have a role in reviewing the decisions of that decision maker, it can safely be assumed that the legislature intended the administrative decision maker to function with a minimum of judicial interference.” (Vavilov at para 24)

As Nigel Bankes explains here, there are several problems with how the majority in Vavilov has formulated this exception to the presumption of reasonableness (the minority in Vavilov also discusses some of these problems at paras 245-253). For example, the presence of a statutory right of appeal in governing legislation means that under Vavilov no deference is now owed to certain specialized administrative agencies which have previously enjoyed significant deference from superior courts on determinations of law within their regime. As well, Vavilov invites the creation of new standards of review by legislatures, or perhaps even the resurrection of a ‘patent unreasonable’ standard by legislatures keen to protect their delegates from judicial scrutiny. I find it curious that Vavilov only references how a legislature may prescribe the standard of correctness (at para 35), yet it references, as an example, legislation in British Columbia which also prescribes ‘patent unreasonableness’. Perhaps there are constitutional limits to what standards may be prescribed by a legislature, but it is disappointing to think Vavilov may have sparked years of litigation to sort that out.

These problems, of course, emanate from the fact that the majority in Vavilov hitched its wagon to legislative text. To be blunt, statutory drafting and design is hardly a guiding light for rational or principled action. This work is often rushed, poorly thought out, and driven largely by partisan interests. I am not aware of any evidence to support a rational or principled approach followed by legislatures in deciding when to place a statutory right of appeal into governing legislation, and I don’t think it’s a stretch to suggest legislatures will respond to Vavilov by prescribing standards of review based largely on policy grounds. This would be a larger concern if not for how the majority has ‘clarified’ the application of reasonableness (as discussed below).

The complete removal of expertise as a fundamental basis for judicial deference is also a significant reversal in the jurisprudence, as the minority in Vavilov forcefully asserts. The majority provides a very thin justification for this reversal (at paras 27, 28). In addition to the critique given by the minority (at paras 230-278), I would add that the justification given by the majority for this significant change in course does not even appear to live up to the very standard of reasonableness it dictates for decision-makers who must justify a departure from prior decisions!

(2) Rule of Law Principles Require that a Different Standard Should Apply to an Administrative Decision

There are three exceptions to the presumption of reasonableness under this category.

The standard of review is correctness in relation to the following questions: (1) constitutional questions regarding the federal-provincial division of powers, the relationship between the legislature and other branches of the state, the scope of Aboriginal and treaty rights under section 35 of the Constitution Act, 1982, and other constitutional matters such as when the issue on review is whether a provision of the decision-maker’s enabling legislation contravenes the Charter (Vavilov at paras 55-57); (2) questions of law which are of central importance to the legal system as a whole because their resolution is of broad applicability and has implications for legislative regimes or the law more generally beyond the legislative framework governing the particular administrative decision-maker under review, and thus requires a single determinative answer (Vavilov at paras 58-62); (3) questions regarding the jurisdictional boundaries between two or more administrative bodies (Vavilov at paras 63-64). These exceptions are more or less adopted from Dunsmuir; given that they have been used sparingly since that decision, one would expect these exceptions to the presumption of reasonableness will continue to make rare appearances in the case law.

Vavilov oddly makes only a passing reference to Doré v Barreau du Québec, [2012] 1 SCR 395, 2012 SCC 12 (CanLII), and thereby conveniently avoids having to decide the more thorny question of whether reasonableness still applies to the judicial review of statutory decisions which allegedly contravene the Charter. The ‘questions of central importance’ exception is one which I believe has been under-utilized by the courts. Unfortunately, the majority provides almost no guidance on how to recognize such questions of law, other than providing a list of examples from earlier cases. Hopefully, future decisions will seize on opportunities to embark upon setting out an intelligible and transparent methodology for identifying this type of question.

Jurisdictional questions have been erased from existence as a category for correctness review (Vavilov at para 65-68). Instead, the application of a more ‘robust’ reasonableness review will enable superior courts to ensure statutory decision-makers do not exceed their powers.

The Law as Per Vavilov on Applying the Standard of Reasonableness

Before getting into the details of this new ‘robust’ version of reasonableness, it is worth noting that Vavilov does not change the law on how a reviewing court should apply the standard of correctness. As summarized in Dunsmuir at para 50: “When applying the correctness standard, a reviewing court will not show deference to the decision-maker’s reasoning process; it will rather undertake its own analysis of the question. The analysis will bring the court to decide whether it agrees with the determination of the decision-maker; if not, the court will substitute its own view and provide the correct answer.”

The majority describes its work in Vavilov as a ‘clarification’ of the proper application of the reasonableness standard, and accordingly notes that pre-Vavilov decisions that address how to apply the reasonableness standard should be read cautiously and considered in light of the guidance set out in Vavilov (Vavilov at para 143). Vavilov provides extensive guidance on how a reviewing court should apply the standard of reasonableness, under the overall direction given by Dunsmuir that a reasonable decision is one which exhibits a requisite degree of justification, intelligibility, and transparency (Vavilov at para 99). Vavilov confirms that the onus is on the party challenging the statutory decision to establish that the decision is unreasonable. Alleged flaws in the decision must be more than merely superficial, minor, or peripheral to the substance of the outcome. A party challenging the decision must establish that the flaws in the reasoning or outcome are sufficiently central or significant to render the decision unreasonable (Vavilov at para 100).

While much of the guidance provided by the majority in Vavilov is a compilation of how the standard of reasonableness has been applied in earlier decisions, this guidance does seem to intensify the scrutiny on the reasons provided by a statutory decision-maker. In cases where a statutory decision-maker is required (by statute or the common law) to provide reasons for its decision, Vavilov confirms that the reasons given by a decision-maker are the primary focus of a reviewing court under a reasonableness review (Vavilov at paras 82-87). This focus on reasons, however, applies awkwardly in cases where the law does not require reasons to be provided for a decision (for example, the enactment of subordinate legislation).Here the majority suggests a reviewing court examine the record, other ‘relevant constraints’, or perhaps simply the outcome of the decision (Vavilov at paras 136-138). Similar to how it avoids any reconsideration of Doré in relation to reasonableness and the Charter, the majority oddly fails to acknowledge the difficulties in applying reasonableness to review the legality of subordinate legislation enacted by an administrative agency (a primary example of statutory decision-making without reasons). This difficulty is illustrated by the recent decision in West Fraser Mills Ltd. v British Columbia (Workers’ Compensation Appeal Tribunal), [2018] 1 SCR 635, 2018 SCC 22 (CanLII), which contains four sets of reasons (the majority judgement and three dissents) on the proper approach for selecting the standard of review to assess the vires of a regulation enacted by the provincial workers compensation board.

Turning to cases where a statutory decision-maker is required by law to give reasons for its decision, we see that the contextual approach is alive and well when it comes to determining whether a statutory decision survives judicial review as a reasonable decision. What constitutes a reasonable decision will vary based on the legal and factual context for the particular decision under review and a statutory decision cannot be divorced from its institutional context. As the majority puts it: ‘Administrative justice’ will not always look like ‘judicial justice’ (Vavilov at paras 90-92). Vavilov sets out a long list of relevant contextual factors for consideration in the review of an administrative decision for reasonableness, and here is my summary of them:

• • The specialized expertise of the decision-maker: A reasonableness review requires attention to the application of specialized knowledge by the statutory decision-maker. A consideration of the relative expertise of a statutory decision-maker, and respect for that expertise in relation to the institutional context of the statutory decision-maker, may provide justification for what would otherwise be seen as shortcomings in reasons given or the outcome of a decision. For example, the application of specialized knowledge may reveal why, particular attention in the reasons is given to some evidence over other evidence, or why the reasons focus analysis on certain issues over others (Vavilov at para 93);
  • The record of proceedings and prior decisions: Consideration of the institutional context for the decision may provide some additional relevant explanation for a statutory decision that is not apparent in the reasons themselves. However, institutional context cannot serve to remedy non-transparency or a gap of logic in the reasoning process; a reviewing court cannot disregard flawed reasoning by supplementing the decision with reasons which could be offered or relying solely on the outcome of the decision as the basis for a reasonableness review (Vavilov at paras 94-97). A statutory decision-maker must demonstrate in its reasons how the evidentiary record was considered and the decision itself must show how the outcome is justified or supported by the evidence tendered before the decision-maker (Vavilov at paras 125, 126). Reasons provided must demonstrate that the decision-maker was responsive to, and grappled with, the submissions made by the parties and that the decision-maker addressed the key or central issues raised before it; however, reasons do not have to address each argument made by the parties (Vavilov at paras 127, 128);
  • Internal coherence: Reasonableness review is not to be conducted as a “line-by-line treasure hunt” (Vavilov at para 102); however, the reasons must demonstrate a connection, or a path of analysis, between the evidence and the decision made by the statutory decision-maker. Reasons that merely set out the submissions made by the parties and then immediately arrive at conclusions will rarely be sufficient to demonstrate the path of analysis undertaken by the decision-maker, and similarly, reasons which contain circular analysis, state unfounded generalizations or rely on absurd premises will also be suspect of unreasonableness (Vavilov at paras 102-104);
  • Statutory scheme: The statutory context for a decision is likely to be the most salient factor to be considered in a reasonableness analysis (Vavilov at para 108). A statutory decision must comply with any prescribed limitations on the scope of the outcome. Conversely, it is unreasonable for a statutory decision-maker with broad discretion to fetter that discretion in its decision-making. The level of assessment on whether a decision is sufficiently justified will vary based on the legislative context for that decision. This context varies from statutory provisions which strictly construe decision-making power to others which provide more open-ended power to decide matters ‘in the public interest’ (Vavilov at para 110). Generally speaking, a reasonable decision is one which is consistent with the statutory grant of power given to the decision-maker;
  • Prior statutory decisions: While statutory decision-makers are not bound by stare decisis, the extent to which a particular decision is consistent with prior decisions made by that decision-maker is a relevant consideration in whether a decision is reasonable. Prior decisions relevant to the issue(s) constrain what will constitute a reasonable decision made by the statutory decision-maker. The failure to explain or justify a departure from a precedent or established internal authority concerning the same or similar issue(s) may constitute an unreasonable decision (Vavilov at paras 112, 129-132);
  • Non-statutory law: Reasonableness does not necessarily require a statutory decision-maker to apply common law or international law principles in the same manner as the superior courts. Whether or not it is unreasonable to depart from, or to be inconsistent with the common law relevant to the decision in question will depend on statutory and other context, and the explanation or justification provided by the decision-maker for the departure or inconsistency (Vavilov at paras 112-114);
  • Statutory interpretation: Vavilov directs special attention to questions of statutory interpretation. A decision-maker who interprets a statutory provision as part of its decision must consider the text, context, and purpose of the provision, particularly where that provision is directly in issue. It is unreasonable for a decision-maker to “adopt an interpretation it knows to be inferior – albeit plausible – merely because the interpretation in question appears to be available and is expedient” (Vavilov at para 121). A failure by the decision-maker to consider a key element of a disputed statutory provision demonstrates unreasonableness if the omission is of such significance that it leads a reviewing court to question the outcome of the decision (Vavilov at paras 115-124);
  • Impact on the affected person: In an interesting reference to Baker v Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817, 1999 CanLII 699 (SCC), the threshold of what constitutes a reasonable justification in a decision is heightened in cases where the impact of the decision on the rights or interests of a person is severe or particularly harsh (Vavilov at paras 133-135).

Pity the statutory decision-maker who is not trained in law, has limited or no access to legal counsel, and must navigate through these signposts in setting out their reasons for decision; the complicated and messy contextual approach is alive and well, however Vavilov has simply shifted it away from selecting a standard of review and into the application of reasonableness going forward. There is a healthy dose of judicial scrutiny embedded in this ‘clarified’ version of reasonableness provided by the majority in Vavilov. This is also demonstrated by how the majority seems to encourage reviewing courts to exercise remedial discretion in judicial review to not only quash but also reverse an unreasonable statutory decision in limited cases (Vavilov at paras 139-142). And this is what the Court does here, deciding that it was unreasonable for the Canadian Registrar of Citizenship to revoke Vavilov’s citizenship and declaring Vavilov to be a Canadian citizen, rather than remit the matter back to the Registrar.

Conclusion

In spite of disagreement and uncertainty over the fundamentals on standard of review post-East Capilano, I am fairly sure that East Capilano did simplify the analysis for selecting the standard of review set out in written briefs and oral argument for judicial review cases. Even parties seeking to have a statutory decision quashed would be more likely to concede reasonableness is the standard of review under East Capilano. Accordingly, I think the majority in Vavilov overstates the extent of uncertainty and non-coherence in selecting the standard of review post-East Capilano (Vavilov at paras 4-9). As I wrote in The Great Divide on Standard of Review in Canadian Administrative Law, the actual problem post-East Capilano was uncertainty on applying the standard of reasonableness. And as is hopefully evident from my discussion above, I think Vavilov does little to resolve this. As a concluding point, consider how the majority describes the basis for reasonableness under Vavilov – reasonableness finds its starting point in the principle of judicial restraint but remains a robust form of judicial review (Vavilov at para 13). A vigorous yet controlled form of judicial supervision: it is difficult for me to weave these together. The minority in Vavilov accuses the majority of charting a version of reasonableness that is far more ‘robust’ than ‘restrained’ (Vavilov at paras 284, 285), and I agree with the minority on this observation. Whether this is a positive or negative development in Canadian administrative law is in the eye of the beholder, but nonetheless a reign of confusion is sure to emerge as the courts try to make sense of reasonableness going forward.

This post may be cited as: Shaun Fluker, “Vavilov on Standard of Review in Canadian Administrative Law” (February 6, 2020), online: ABlawg, http://ablawg.ca/wp-content/uploads/2018/07/Blog_SCF_Vavilov.pdf

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg