Review of Standards: Please SCC, Provide “Reasonable” Guidance!

Jodi Martin’s post this week on Lévis (City) v Fraternité des policiers de Lévis Inc, 2007 SCC 14, highlights the current uncertainty within administrative law surrounding the application of the “pragmatic and functional” approach in determining the appropriate standard of review for administrative tribunal decisions. If the Supreme Court Justices, whose role it is to provide guidance on the laws of this country, cannot agree on the application of a test that they developed (in Dr Q v College of Physicians and Surgeons of British Columbia, [2003] 1 SCR 226), where does that leave the rest of us (lower court judges, lawyers, and unfortunate students of admin law)?

While the use of different standards did not affect the outcome in Lévis, it did have a direct impact on the two contrasting results in Council of Canadians with Disabilities v VIA Rail Canada Inc, 2007 SCC 15, a judgment released one day later. A summary of the substantive issue raised in the appeal can be found in Corey Wall’s post from yesterday.

Justice Abella, whose reputation as an advocate of equality rights took a hit after her uncharacteristic turn in McGill University Health Centre (Montreal General Hospital) v Syndicat des employés de l’Hôpital général de Montréal, 2007 SCC 4, may have taken a step towards restoring that reputation in this case. She wrote the majority judgment which upheld the Canadian Transportation Agency’s decision requiring VIA Rail to retrofit some of its newly acquired cars to accommodate wheelchair access.

Since the majority characterized the issues of the case as falling “squarely and inextricably within [the Agency’s] expertise and mandate,” they reviewed the decision of the Agency with the highest degree of deference, the “patent unreasonableness” standard. Only if the Agency’s decision was patently unreasonable would the SCC step in to overturn it. Having set the bar so high, it is not surprising that the majority reached the result it did. However, it is interesting to note that the Federal Court of Appeal had also applied this same standard in overturning the Agency’s decision.

[Personally, I have never understood how any decision can be considered patently unreasonable as long as reasonable people may hold a different opinion. By coming to the conclusion in the way it did, wasn’t the Court of Appeal really saying that the Agency was made up of a group of unreasonable knuckleheads?]

Perhaps it was unnecessary for the majority to adopt such a deferential standard to reach its result. From Justice Abella’s comments, it appears the majority would have upheld the Agency’s decision just as easily on a correctness standard:

In my view, … far from being unreasonable for the Agency to adopt a frame of reference premised on achieving personal wheelchair-based accessibility in 13 economy coach cars and 17 service cars out of the 139 cars VIA purchased, it may well have been found to be patently unreasonable for the Agency not to do so.

This is in stark contrast to the approach of the dissenting justices, who settled upon a standard of “correctness” in reviewing the Agency’s decision. Thus, little or no deference was given to that decision and the dissent was free to simply substitute its own opinion for that of the Agency. Perhaps the dissent realized the difficulty it would face in overturning the Agency’s decision had it used a standard of patent unreasonableness or even reasonableness simpliciter and selected the standard they felt necessary to reach the desired result.

Writing for the dissent, Justices Deschamps and Rothstein held that both the jurisdictional issue (whether the Agency had the authority to review a complaint that does not arise from the experiences of an actual passenger) and the determination of applicable human rights principles in the case were matters of law and in these areas, the Agency did not have greater relative expertise than a court:

The Agency is required to resort to human rights principles which are not comprehensively set out in its home statute and in respect of which the Agency, whose prime function is economic regulation of transportation in a largely deregulated environment, does not have specific expertise. This factor points to a standard of review that will be less deferential.

Another difference between the two judgments is that the majority felt it unnecessary to separate the jurisdictional issue from the substantive part of the Agency’s decision and conduct a functional and pragmatic assessment of each component, possibly leading to different standards for the different issues. Speaking for the majority, Justice Abella noted the dangers of the liberal use of this segmentation of issues:

If every provision of a tribunal’s enabling legislation were treated as if it had jurisdictional consequences that permitted a court to substitute its own view of the correct interpretation, a tribunal’s role would be effectively reduced to fact-finding … Just as courts “should not be alert to brand as jurisdictional, and therefore subject to broader curial review, that which may be doubtfully so,” so should they also refrain from overlooking the expertise a tribunal may bring to the exercise of interpreting its enabling legislation and defining the scope of its statutory authority.

Interestingly, these comments echo similar comments made by Justice Abella in her concurring judgment in Lévis, which was probably written around the same time:

Similarly, legal issues ought not to be declared readily extricable when they are legitimately and necessarily intertwined with the adjudicator’s mandate and expertise. In such circumstances, the decision ought to be reviewed as a whole, not as a segmented compilation subject to an increased degree of scrutiny and intervention.

This was in opposition to the majority judgment in that case, which held that segmentation leading to “[m]ultiple standards of review should be adopted when there are clearly defined questions that engage different concerns under the pragmatic and functional approach.”

Thus, while Justice Abella stood alone in advocating a unified standard of review in Lévis, she enjoyed the support of the majority in taking this unified approach on a somewhat similar basis in VIA Rail. From the Lévis decision to the VIA Rail decision, which were released one day apart, Chief Justice McLachlin, Justice Bastarache and Justice Charron had all switched from the view that the standard of review for a tribunal’s statutory interpretation of enabling legislation should be segmented from its analysis of substantive issues to a view that the same standard should be applied to the whole. Justice LeBel, who had not taken part in the Lévis decision, joined them in supporting Justice Abella’s unified approach in VIA Rail.

To be fair, there were some differences between the jurisdictional issue in VIA Rail and that in Lévis. In VIA Rail, the majority gave deference to the Agency’s interpretation of the specific statute from which it derived its authority while in Lévis, the majority chose not to defer to an arbitrator’s interpretation of two apparently conflicting laws, external to the arbitrator’s specific authority, regarding requirements for employment of police and municipal workers. Still, the close timing of the release of the two judgments and the shifting positions of the justices within this debate are quite striking.

While the dissent in VIA Rail applied the correctness standard both to the jurisdictional issue and the substantive issues of the case, they did not concede that segmentation was unwarranted in this case:

Subjecting all aspects of a decision to a single standard of review does not account for the diversity of questions under review and either insulates the decision from a more exacting review where the pragmatic and functional considerations call for greater intensity in the review of specific legal questions, or subjects questions of fact to a standard that is too exacting. A tribunal’s decision must therefore be subject to segmentation to enable a reviewing court to apply the appropriate degree of scrutiny to the various aspects of the decision which call for greater or lesser deference.

Thus, the dissent applied the pragmatic and functional approach to each of the two questions separately in coming to its conclusion that the correctness standard was appropriate for both.

Since this discrepancy on the segmentation issue did not result in two different standards in either judgment, it is unclear whether these statements by Justice Abella or by Justices Deschamps and Rothstein are truly representative of the opinions of the other justices who signed on to their respective judgments. It is possible that, having reached a conclusion on the ultimate outcome of the case, these other justices simply endorsed the judgment which matched this result without writing a concurring judgment on the segmentation issue, the discussion of which was not necessary to the outcome.

One further comment, made by Justice Abella in obiter, offers some hope for a simplification of the standard of review analysis in the future:

I appreciate that it is a conceptual challenge to delineate the difference in degrees of deference between what is patently unreasonable and what is unreasonable. Both, it seems to me, speak to whether a tribunal’s decision is demonstrably unreasonable, that is, such a marked departure from what is rational, as to be unsustainable.

The dissent picked up on this point and appeared to agree with its substance, although they were understandably perturbed by Justice Abella’s introduction of new terminology (“demonstrably unreasonable”) to the current catalogue of adjectives for “unreasonableness” available to the courts:

We must respectfully express reservations about introducing another term to an already complex area of the law which can only lead to ambiguity. We agree with the majority that it is difficult to determine the degrees of differences as between what is unreasonable and what is patently unreasonable. In an appropriate case, of which this is not one, the Court may engage in a review of the standards of unreasonableness and patent unreasonableness. Until that occurs, we do not see the need to add to the lexicon of standard of review terminology.

These comments suggest that the SCC may be entertaining the idea of a shift from the current standard of review model with three different standards to one which only contains two: a single unreasonableness standard and a correctness standard. I certainly hope so. And I await with bated breath a future case that can achieve such clarity within admin law. While they’re at it, perhaps the SCC could also offer more consistent and coherent guidence on the application of the pragmatic and functional approach. It just might improve my mark in administrative law.