Correctness Review (Lauren J. Wihak)

Lauren J. Wihak (@lauren_wihak) is an Associate at McDougall Gauley LLP in Regina

Dunsmuir v. New Brunswick (“Dunsmuir”) was intended to simplify and clarify the question of the standard of review used to review administrative action, making this issue more predictable to litigants.  I have been invited to comment on Dunsmuir and correctness review or, I suppose, what is left of it.  As I have argued previously[1], and the subsequent jurisprudence shows, there is very little space for correctness review in today’s administrative law landscape.

However, I would argue that the case most relevant to the availability of correctness review is not Dunsmuir at all but, rather, Alberta Teachers’ Association (“ATA”).  In my view, Dunsmuir preserved a meaningful role for correctness review in judicial review, through its reference to the continued applicability of pre-Dunsmuir jurisprudence on the applicable standard of review, and in the application of the standard of review analysis.   ATA, however, mandates that reasonableness presumptively applies whenever a decision-maker – whether a tribunal or a deputy minister or a high school principal – decides a question of law that originates in its enabling legislation or a statute closely related to its function, which will be almost always.  I think a fair reading of the Supreme Court’s post-ATA jurisprudence – and most notably the majority’s decision in Edmonton East – suggests that it will be only in extraordinary circumstances that the presumption will be rebutted, and that correctness review will generally not be available outside of the four categories set out in Dunsmuir.  Moreover, those correctness categories have been narrowly construed.  The concept of a jurisdictional question is on life support, and the Dore framework prevents true correctness review for some Charter-related questions.   Correctness review has been confined to a narrow band of questions, their scope interpreted narrowly, with the ATA reasonableness presumption left for everything else.  Efforts to expand new categories of questions to which correctness review applies, for example when there is a statutory right of appeal, with leave, to the court of appeal, have been rejected.  The result is a one-size fits all categorical approach to reasonableness, which I would argue does not appropriately reflect the diversity of the administrative state.

Certain areas of law are appropriately governed by rules, and others are better governed by principles.  Rules are perfectly appropriate when the scope or range of situations to which they apply is relatively narrow and factually confined, for example, or where the benefits of the certainty that bright line rules provide is considered paramount.  However, administrative law is an area that, as a result of the thousands of different types of decision-makers, enabling statutes, and questions arising under them, cannot, I would argue, be appropriately governed by “rules”.  Rather, we can only hope for general, principled standards, based on factors, reflected in the Dunsmuir standard of review analysis, which we know are important to determining legislative intent, the “polar star” of judicial review.  But the shift away from principles and towards “rules”, “categories” and “presumptions” has, I suggest, led to an approach where decisions that would have been reviewed for correctness under the Dunsmuir standard of review analysis are now being reviewed for reasonableness.  We have abandoned any search for actual legislative intent, viewed through the lens of the particular function of the decision-maker and its relationship to the courts, and the particular nature of the question that has been decided.

As Justice Binnie noted in his concurring reasons in Dunsmuir, there is no uniform or prototypical administrative decision-maker.  On the one hand, the legislature may create a comprehensive scheme to deal with a particular issues, with a comprehensive dispute resolution system outside of the courts to adjudicate disputes in relation to those issues.  Examples include the Competition Tribunal, labour boards, human rights tribunals and workers compensation tribunals.  The adjudicators making up these bodies are often legally trained, have actual expertise in the subject matter of their regime, and oftentimes both.  The tribunal or adjudicative body may have an established body of jurisprudence that all individual decision-makers generally follow.  Their enabling legislation will often include strong privative clauses.  Sometimes, these tribunals are created precisely because the disputes were being unsatisfactorily addressed by inexpert courts, their creation by the legislature aimed specifically at taking the adjudication of the disputes out of the courts.  The message that courts should generally “stay out” makes good conceptual sense when talking about these types of adjudicative decision-making bodies.

But contrast these adjudicative bodies with other types of decision-makers, such as the municipal councils and professional colleges who engage in legislative and policy-making functions, and the governmental and bureaucratic fonctionnaires – or, as Justice Lauwers has suggested, “line decision makers” – who primarily make administrative decisions that may, on occasion, cross paths with general legal or constitutional principles.  The public school principal, for example, who must decide to what to accommodate a student’s religious practices.  The role of the courts in relation to these “front line” decision-makers has always been supervisory.  Courts need not be reminded that these decision-makers decide issues that have never been, in our system of government, theirs to decide.  However, it is equally true that, because these “front-line” decision-makers have no particular legal or interpretive expertise, their decisions on legal questions should generally not be deferred to, at least as a matter of course.

There might also be indications from the legislature, in the creation of a privative clause or a statutory right of appeal, for example, that the relationship between the decision-maker – tribunal or front-line decision-maker – and the court suggests that deference is not warranted; at least not in relation to all questions or issues the decision-maker may decide, including those decided under the enabling statute. It cannot be that the mere act of exercising statutory authority, alone and without more, warrants deference on all questions of law.  Rather, the actual expression of the legislature’s intentions, in not only creating but in outlining the scope of authority of decision-makers and their relationship to the courts on review, is surely relevant to how courts ought to review their decisions.  Indeed, as Justice Cromwell noted in his concurring reasons in ATA, any limits on the types of decisions that a statutory decision-maker can make will only ever be found in its enabling statute.  How can legislatures assign to decision-makers some responsibilities and not others, if reviewing courts are precluded from enforcing the very boundaries that the legislatures have established?

The nature of what has been decided also matters.  As noted, the fact that the decision originates in the enabling statute cannot be enough.  Is the decision-maker applying the facts to a well-established body of legal or policy principles?  Is the decision a discretionary one, with wide latitude to the decision-maker to consider the “public interest”?  Deference to these types of decisions seems the obvious result.  Dunsmuir tells us that a reasonable decision is one that falls within a range of reasonable alternatives.  But what of “questions of law”?  Is the question one of statutory interpretation, or the common law?  Does the determination of the legal question involve human rights or liberty interests, or the scope of constitutional rights?  Does consistency in the meaning of the provision matter to the rights and interests of the public and to individual citizens?  Or does the meaning of the provision take colour from the facts of the dispute?  The answer to these questions, in my view, points towards the need for correctness review.

It may just be that we do not have the right vocabulary for dealing with the varied types of decisions that can be made by a decision-maker under its enabling statute.  Is it really sufficient to say that a decision to resolve a particular dispute that it turns out the Legislature has not authorized the decision-maker to resolve is “unreasonable”?  Isn’t the decision just wrong?  And, is this really the kind of question that can lead to more than one reasonable answer?  Rather, isn’t the reviewing court being called upon to police the parameters of the decision-maker’s authority – what is at the very heart of the courts’ constitutional role on judicial review?  Is it any better to say that these decisions are subject to reasonableness review, only to go on and perform “disguised correctness review”, concluding that there is really only one reasonable outcome?

There may still be a place for correctness review under the current jurisprudence.  The Supreme Court has not gone so far as to hold that correctness review outside the four Dunsmuir categories will never be available, or that the correctness standard should be eliminated.  And, the Supreme Court has taken a somewhat more expansive view of legal questions of general importance and outside the expertise of the decision-maker when the scope of human rights or other general legal principles such as, for example, the state’s obligation of neutrality or solicitor-client privilege are at issue.  Some provincial appellate courts have also started to push back.

That said, what we now have is a “rule” that reasonableness will apply anytime a decision-maker renders a decision that arises out of its enabling legislation, with no guidance as to what, if any, exceptions to that rule there might be.   The Supreme Court has superimposed a categorical, as opposed to contextual, approach to determining the standard of review, with little regard to the actual functions and powers of the decision-maker in any particular case, or the decision under review.  It has substituted a presumption about the legislature’s intent for actual legislative intent, as if legislative intent was uniform with respect to all issues that can arise under any given statute.  I cannot put it any better than did Justice Binnie in CUPE: “Given the immense range of discretionary decision makers and administrative bodies, the test [must be] necessarily flexible, and [proceed] by principled analysis rather than categories, seeking the polar star of legislative intent”.

The interaction between the decision-maker’s function and the scheme that it administers, and the nature of the question that it has decided, may not always lead to the conclusion that the legislature intended deference.  The administrative state is complex.  No two enabling statutes will be identical, conferring identical expertise and identical decision-making powers.  It simply will not always be that for all administrative decision-makers, the presumption of reasonableness, and what it presumes about a decision-maker’s familiarity and expertise with its enabling statute and the type of decision-making that it engages in, works.  When the focus is on actual, as opposed to presumed, legislative intent, it’s not hard to see that there can be a wholly appropriate role for correctness review in administrative law, which does not otherwise jeopardize the core teachings of the Dunsmuir deference doctrine.

[1] Wihak, L.J., “Whither the correctness standard of review? Dunsmuir, six years later” (2014), 27 C.J.A.L.P. 173.

This content has been updated on February 20, 2018 at 19:53.