Anticipating the Supreme Court of Canada’s Administrative Law Trilogy II: The Issues

Why is the Vavilov/Bell Canada/National Football League trilogy so important? In granting leave to appeal in the Bell Canada and Vavilov appeals, the Supreme Court of Canada expressed itself:

…of the view that these appeals provide an opportunity to consider the nature and scope of judicial review of administrative action, as addressed in Dunsmuir v. New Brunswick, [2008] 1 S.C.R. 190, 2008 SCC 9, and subsequent cases. To that end, the appellants and respondent are invited to devote a substantial part of their written and oral submissions on the appeal to the question of standard of review, and shall be allowed to file and serve a factum on appeal of at most 45 pages.

The formulation in the National Football League appeal was slightly different, with no reference to devoting “a substantial part” of the submissions to standard of review, because the underlying issues arise out of the same controversy as the Bell Canada appeal.

“Standard of review” and “deference” rarely hit the headlines but this week I have been fielding calls from reporters from far and wide. Even the Washington Post — the paper that brought down Richard Nixon — is on the case. I am happy that standard of review is having its fifteen minutes of fame. It is central to how courts decide whether to uphold or quash decisions taken by the country’s vast array of administrative decision-makers — many of which, like social welfare, immigration and workers’ compensation, are fundamental to the lives of Canadians. 

This area of law has been riven by uncertainty for many years, especially in the last decade, and some lower court judges have been in open revolt against the Supreme Court. The uncertainty is partly the fault of the Court. On a number of occasions over the last decade the Court has said one thing in one case and then soon after said the exact opposite in another. And repeatedly the Court has instructed lower courts to defer to administrative decision-makers but has failed to follow its own advice in its own decisions. Understandably, lower courts, lawyers, litigants and ordinary Canadians have found themselves perplexed.

Moreover, the idea that judges should “defer” to non-lawyers (doctors, economists and patronage appointees to administrative tribunals) on questions of law is controversial and difficult for many lawyers to accept — indeed, the judges of the Supreme Court themselves are deeply divided on this key issue.

Canadian administrative law has been aptly described by Stratas JA, writing extra-judicially, as “a never-ending construction site”. Although the broad contours of Canadian administrative law are clear, on many important issues there is little concrete guidance from the Supreme Court and, as a result, much damaging confusion.

There are two main areas the Supreme Court will hope to bring clarity to in these appeals: selecting the standard of review and applying the reasonableness standard.

Selecting the Standard of Review

If you lived in France and you wanted to challenge a decision the government had taken about your immigration status or your social security benefits, you would go to the administrative courts. The judges who sit in these courts are generally experienced administrators themselves: they have previously taken exactly the sorts of decisions citizens want to challenge before the administrative courts.

For historical reasons, common law countries never developed special administrative courts. Instead, when a citizen wishes to challenge a government decision, he or she has to go to the same courts which adjudicate contract law, civil liability and family law disputes. The resulting problem is that the courts know little about the details of immigration law, telecommunications law, workers’ compensation law, social security law and so on, because the judges who sit in these courts might have spent their professional careers doing completely different work (corporate litigation, insolvency, mergers and acquisitions, etc). Even where the judges in their past lives did a lot of administrative law, they probably specialized in one area (environmental law, labour relations law or municipal law, for example) and will not have a detailed grasp of other areas. The question then arises: how much respect or “deference” should courts give to the experts (if, indeed, they are really experts)?

In Canada, for the last 40 years, judges have either applied the correctness standard (no deference) or some form of reasonableness standard (defer to the administrative decision-maker unless the decision was demonstrably unreasonable). From 1988 to 2008, the Supreme Court advocated a contextual approach, where the selection of correctness or reasonableness depended on the interplay of a variety of factors. From 2008 onwards, the Supreme Court preferred to rely on a series of categories and then an outright presumption that in most situations the reasonableness standard would apply. But the relationship between categories, context and presumptions was never clarified. The circumstances in which contextual factors would push a decision out of a category or rebut a presumption remain somewhat nebulous. And the position in relation to regulations or by-laws — general norms adopted under statutory authority — is hopelessly confused.

There are some who argue that reasonableness should always be the standard of review: we should do away with correctness review because the presence of categories — “jurisdictional” questions and “general questions of law” — is a distraction from the important task of applying the reasonableness standard to the facts of a given case.

There are others, however, who argue that the constitutional role of the judiciary is to give authoritative answers to questions of law. What do the doctors, economists and patronage appointees to administrative tribunals know which judges do not? Most members of this group will accept that, sometimes, there will be questions of law on which economists, say, might have something useful to say. But, for them, this is the exception, not the norm.

Somewhere between the reasonableness and correctness camps are those who appreciate the arguments for deference but would like correctness review to be retained on some issues of fundamental importance to the legal system or the field of operation of a particular decision-maker (this is usually dressed up as “jurisdiction“).

One particular flashpoint in recent years has been situations in which a legislature has provided for a right of appeal from an administrative decision-maker to a court. Ordinarily, one would assume that, in such circumstances, there would be no “deference” due from the courts to the decision-maker. But, as the Supreme Court pointed out in the late 1980s and mid 1990s, an expert decision-maker is still an expert even if the court is hearing an appeal rather than a judicial review. More recently, however, this sensible observation has been pushed to something of an extreme. Now, the presumption of reasonableness review cannot even be rebutted by a statutory provision which gives a right of appeal on questions of law or jurisdiction.

Reasonableness Review

Despite all the fuss about selecting the standard of review, empirical studies have revealed that most of the time, Canadian courts apply the reasonableness standard of review. But what does “reasonableness” mean? In the Supreme Court’s decisions, it seems to mean, first, that the reasons for the decision were comprehensible and, second, that the decision fell within a range of permissible, acceptable outcomes.

In terms of comprehensibility, the Supreme Court has struggled to address situations where the administrative decision-maker gave sparse or no reasons. On the one hand, it is difficult to give any deference to a nominally expert decision-maker where there is no evidence that the decision-maker applied its expertise to the matter in front of it (as the Federal Court of Appeal noted in Vavilov, at para. 39). On the other hand, the decisions of decision-makers operating in areas like labour relations law, where the participants are repeat players, can be fully comprehensible to the participants but not easily decipherable by an outsider and it is thus inappropriate to treat judicial review as a “line-by-line treasure hunt for error“.

The Supreme Court’s response to this tension has been, variously, to allow decision-makers to bootstrap their defective contemporaneous reasons with material provided later on, to imaginatively supplement the reasons given with reasons that ‘could have been offered’ or to simply proceed to give the decision it thought should have been given (sometimes termed “disguised correctness review”). Bootstrapping and supplementation appear to be on the wane (though one can never be sure) but disguised correctness review is a regular feature in the Supreme Court Reports.

The alternative possibility, sometimes championed by the Federal Court of Appeal (and mentioned in Vavilov, at para. 39), is to send defective decisions back to the decision-maker to do a better job. Perhaps understandably, the Supreme Court is reluctant to take this course after parties have already spent years and many thousands of dollars in litigation, though if it were to do so once or twice the message (“Shape up, or we’ll send it back”) would probably get through pretty quickly.

In terms of the range of reasonable outcomes, the Supreme Court has said little other than that ‘it depends on the context’. Other courts, particularly the Federal Court of Appeal, have tried to flesh out what this might mean, perhaps broader or narrow ranges depending on the interplay of contextual factors (not unlike the 1988-2008 approach to selecting the standard of review) and a requirement to identify “badges” of unreasonableness tainting the decision complained of. But the Supreme Court has refused to engage in great detail.

In the application of reasonableness review to questions of statutory interpretation, the Supreme Court has said next to nothing. The result is that some lower courts (like the British Columbia Court of Appeal) insist that the principles of statutory interpretation must be applied rigorously to set a benchmark against which to measure the administrative decision-maker’s interpretation of law but others (like the Federal Court of Appeal) preach a more restrained approach, in which statutory interpretation principles play a secondary role in determining whether the decision at issue was indeed demonstrably unreasonable. There is support for both of these wildly divergent approaches in the Supreme Court’s recent administrative law jurisprudence.

Running through all issues relating to reasonableness review is another question: should the same standard apply to different types of decision-maker? The administrative state is a vast galaxy, populated by creatures as diverse as front-line officials on borders, in classrooms and behind service counters and highly-skilled experts sitting on environmental, labour relations and workers’ compensation tribunals. What is expected of these creatures by those Canadians whom they serve varies greatly. Perhaps what administrative law expects should vary too. At the very least, the issue deserves careful consideration.

Time will tell whether Thursday’s trilogy of cases is the vehicle which will allow the Supreme Court to drive forward into a new era of Canadian administrative law, one in which lower courts, litigants and ordinary citizens operate in a clear, coherent and predictable framework. I have expressed doubt on this point in the past — in this post I have not really even touched on the deeper philosophical and theoretical questions which lie just beneath the surface of the issues I have discussed, or whether the adversarial process is suited to resolve them.

The task, as you can tell, is great. I can only wish the Supreme Court good luck.

This content has been updated on December 18, 2019 at 15:41.