Why is Administrative Law So Complicated? Complex Concepts

I have uploaded a pre-proof version of the first chapter of A Culture of Justification: Vavilov and the Future of Administrative Law to SSRN: “Why is Administrative Law So Complicated?” Here is an extract:

Three concepts have been of central importance to Canadian administrative law in recent decades. One of them – jurisdiction – is found in every textbook in every common law country, often occupying many pages which are often not (sometimes by the author’s own admission) especially illuminating. The two others – deference and legislative intent – can be found in other countries from time to time but rarely play the important roles they have played and continue to play in Canada. All three concepts are complex and difficult to grasp.

Let us take “jurisdiction” first of all. You will often find lawyers – be they advocates, judges or professors – saying things like “The Tribunal exceeded its jurisdiction”, “The Minister had no jurisdiction to take that decision”, or “The Regulator made a jurisdictional error”. What are they trying to convey? There are two basic ideas. First, the variety of decision-makers covered in the last section get their powers from statutes made by a legislature (there are some nuances here about prerogative powers and entering into contracts, but they can be safely ignored). This is their “jurisdiction”, which they cannot exceed.[1] Second, the statute grants powers – but it also limits them. If a statute provides that a decision-maker can do Y only if X is present, then the presence of X is a pre-condition to the doing of Y.^[2] So, if a tribunal is granted the power to make findings of discrimination in respect of the letting of self-contained dwelling units, then, that a given premises is a self-contained dwelling unit (X) is a pre-condition to making a finding of discrimination (Y).^[3] X has been laid down in a statute by the legislature. If a decision-maker does Y although X is not present, then it is acting in excess of its powers: “any grant of jurisdiction will necessarily include limits to the jurisdiction granted, and any grant of a power remains subject to conditions”.^[4]

The primary problem with the concept of jurisdiction is that no formula has ever been devised for distinguishing X from Y: “[n]o satisfactory test has ever been formulated for distinguishing findings which go to jurisdiction from findings which go to the merits”.^[5] All statutory provisions can be cast in the basic form: if X is present, then the decision-maker shall or may Y. The problem then becomes acute, as the boundaries of “jurisdiction” are “impossible to draw precisely because the two matters [X and Y] were inextricably interwoven”.^[6] Because an X component may be identified in all statutory provisions, a court can invoke the doctrine to justify intervening whenever it so pleases, a risk borne out by the historical record: “[t]here was no predictability as to how a case would be categorised before the court pronounced on the matter. There was also no ex post facto rationality that could be achieved by juxtaposing a series of cases and asking why one case went one way and another was decided differently”.^[7] To say things like “The Tribunal exceeded its jurisdiction”, “The Minister had no jurisdiction to take that decision”, or “The Regulator made a jurisdictional error” announces a conclusion (in impressive-sounding language) without explaining it. What matters are the reasons for coming to the conclusion – and these are often obscured by the language of “jurisdiction”.

Although there is much more to say about the concept, it is unnecessary to wade any further into the morass here. For as we will see subsequently – especially in Chapter 3 – the concept of “jurisdiction” has been marginalized in Canadian administrative law. Nonetheless, it is difficult to understand the process by which it was marginalized without understanding why the concept is complex. And it is hard to appreciate why administrative law is so complicated without spending some time discussing “jurisdiction”. Hence the infliction of the last few paragraphs on the reader.

“Jurisdiction” has been marginalized and, in large part, replaced in Canada by the concept of “deference”. This concept is helpfully discussed by Gary Lawson and Guy Seidman in a recent book, Deference: The Legal Concept and the Legal Practice.[8] They note that the use of “deference” in judicial decisions is in contrast to the conventional use of the term deference, which tends to involve complete obeisance, say, in ‘deferring’ to another’s choice of restaurant. When lawyers used the term, it is typically “to describe a sliding scale of weight rather than the kind of yield likely to be meant in ordinary conversation”.[9] In deciding to uphold an administrative decision (or not), a judge will give weight to the views of an administrative decision-maker explaining its preferred interpretation of a statutory provision or justifying a policy choice. Deference is, therefore, a way “of representing an allocation of decision-making responsibility among multiple actors”.[10]

This allocation can be made in a variety of different ways. Giving weight is one possibility. Making space is another. Consider an everyday example. Patients visit doctors to receive advice about their ailments and possible cures. Ultimately, the patient will decide what to do, but in making the decision, will allocate responsibility as between the patient and the doctor. Some patients might simply give weight to the views of the doctor, to go into the mix with what the patient learned on Google or, less spuriously, what the patient has learned from past experience. Indeed, a patient who is a trained doctor might give much less weight to the views of their doctor (hence the expression that doctors make the worst patients). Other patients might accept the advice of their doctor, subject to the general quality of the doctor’s explanations: as long as these seem reasonable and grounded in the evidence, the patient will accept them; if they are not, the patient might reject them, or go and seek a second opinion. “Deference” in administrative law can function the same way. Stay in the doctor’s office: say the patient complains to a disciplinary body about the way the doctor treated them; say the disciplinary body finds that the doctor mistreated the patient; and say that the doctor challenges this finding in court. The court can give weight to the disciplinary body’s findings on what the professional standard of conduct is but keep the final word for itself; or the court can say something like “a judge can only intervene if the findings were unreasonable or unsupported by evidence”, carving out a space for the disciplinary body to develop its interpretation of the standard of conduct. Unsurprisingly, in this area, courts much more often create space rather than give weight.

The qualifier “in this area” is important. Determining the professional standard of conduct typically involves a value judgement, which is heavily dependent on the context of the patient-doctor relationship and the general approach to such relationships in the community of doctors. It is a question of fact, or perhaps of applying standards to fact (a “mixed” question) – but it is certainly not a question of law. Relatedly, if a Minister determines that it is not in the “national interest” to fund a particular project or allow a particular permanent resident to remain in Canada notwithstanding serious criminal offences, this is best characterized as an exercise of discretion. When questions of fact, “mixed” questions or exercises of discretion come up before the courts in judicial review proceedings, the courts will generally be very respectful of the decision-maker. They will not simply give weight to the decision-maker’s views; rather, they will accept those views as conclusive, unless there is something seriously wrong with the reasons for those views and/or the underlying evidence.

Where matters become trickier, as you might imagine, is on questions of law. When it comes to interpreting statutes, or case law, judges find themselves doing something they were trained to do since their very first days as law students. Why would they give any weight at all to the views of others, let alone carve out a space for non-lawyers? Over the years, the response to this question has occupied many pages of Canadian law journals and judicial decisions. For the moment, let us simply acknowledge that in some areas of regulation answering questions of law might require technical knowledge beyond the ken of judges – where a body of economists has to determine whether a merger would lesson competition “substantially”, for example. Here, space might even be appropriate, notwithstanding that the word “substantially” appears in a statute, and weight surely would be. Of course, expertise is a slippery concept: if you prefer, the rationale for carving out a space could be efficiency, inasmuch as the optimal way for judges to spend their time is not second-guessing economists about the meaning of “substantially” in a statute about competition law.

This leads us to the last complex concept, of “legislative intent”. It provides another potential justification for deference, alongside expertise and efficiency: the legislature might require the courts to either give weight to the views of a decision-maker, or simply to accept them as long as they are reasonable and based in the evidence. Here, great care is needed. To begin with, “legislative intent” does not require us to look into the hearts and minds of legislators. We are concerned with the words they used in their statutes, not the message they intended to communicate. “Legislative intent” does not have a free-standing meaning floating in the ether above the words used in the statute. Those words sometimes include so-called ‘privative clauses’, designed to deprive the courts of the authority to review particular decisions. These clauses can take a variety of forms: they targeted particular prerogative writs (a “no-certiorari clause”, for example) when this was appropriate; or they provided that “no decision of the Minister shall be called into question in a court of law”. Canadian courts (and their counterparts elsewhere in the common law world) have regularly had to grapple with such clauses. Beyond privative clauses, however, one can say that the choice to create a decision-maker and give it significant powers, be it a competition authority, a minister, a disciplinary body, also evidences a “legislative intent” that the decision-maker should not be routinely second-guessed by the courts. Whether this is true and, if it is, whether it requires the giving of weight or the carving out of space, has also occupied much of the time of Canada’s administrative lawyers.

Evidently, a legislature need not intend, through its language, for deference to be given. It might give decision-making powers to the courts, or provide for appeals from decision-makers to the courts. Here, again, a wide variety of provisions can be found in Canada’s law libraries: giving power to courts directly; allowing decision-makers to ask courts to give a binding ruling on a question of law; creating a right of appeal on questions of law only (sometimes only with the court’s permission); providing for an entirely new proceeding before a court; or simply giving individuals the option of an “appeal”. “Legislative intent” and the relationship between clear expressions of intent, such as privative clauses and rights of appeal, and not-so-clear expressions of intent, such as delegating significant decision-making authority, has been a central concept in Canadian administrative law.

The purpose of this discussion has been, on the one hand, to assist in answering this Chapter’s main question – why is administrative law so complicated – but on the other hand to introduce some concepts which are going to be central to the story told in subsequent chapters. Jurisdiction, deference and legislative intent will all rear their heads at various points in the pages to come. There be dragons, but at least the reader knows what to fear from them.

[1] See also Martine Valois, « Si l’histoire de la norme m’était contée : évolution et circonvolutions du principe de déférence au Canada » in Paul Daly & Leonid Sirota, eds, A Decade of Dunsmuir/Les 10 ans de Dunsmuir (Toronto: Carswell, 2018) 19.

[2] X need not be singular. There could be a set of such pre-conditions. I refer to X for ease of exposition.

[3] See Bell v Ontario Human Rights Commission,[1971] SCR 756, 18 DLR (3d) 1. See also R v London Rent Tribunal, ex parte Honig,[1951] 1 KB 641; R(A) v London Borough of Croydon,[2009] UKSC 8.

[4] Union des employés de service, local 298 v Bibeault,[1988] 2 SCR 1048 at 1086, 95 NR 161, Beetz J.

[5] SA de Smith et al, Judicial Review of Administrative Action,5th ed (London, UK: Sweet and Maxwell, 1995) at 255. See similarly Louis Jaffe, “Judicial Review: Constitutional and Jurisdictional Fact” (1957) 70:6 Harv L Rev 953 at 959.

[6] Paul Craig, “Jurisdiction, Judicial Control, and Agency Autonomy” in Ian Loveland, ed, A Special Relationship? American Influences on Public Law in the United Kingdom (Oxford: Clarendon, 1995) 173 at 177.

[7] Paul Craig, Administrative Law,6th ed.(London, UK: Sweet and Maxwell, 2008) at 441. See similarly William Wade, “Constitutional and Administrative Aspects of the Anisminic Case” (1969) 85 LQR 198 at 210-211.

[8] New York: Oxford University Press, 2020. See also Dean Knight, “Locating Dunsmuir’s Meta-structure Within Anglo-Commonwealth Traditions” in Paul Daly & Leonid Sirota, eds, A Decade of Dunsmuir/Les 10 ans de Dunsmuir (Toronto: Carswell, 2018) 191.

[9] Lawson & Seidman, supra note 71 at 75.

[10] Ibid at 77.

This content has been updated on August 9, 2023 at 08:51.