Another in the “Replayed” series.
For the introduction to the series, see “Unjust by Design Replayed”, Post #1 below.
In the passages quoted from Unjust by Design, footnotes are omitted and context-clarifying notes sometimes [added]
THE PROBLEM OF “UNLIKE” TRIBUNAL DECISIONS
Post #6 in the “Unjust by Design Replayed” series addressed one of the most important of the fundamental principles of any democratic society – the right of all members of that society to equality under the law; the principle that, in a democratic society’s adjudication of disputes over members’ legal rights, everyone must be able to count on their rights being validated, or not, on the same procedural and substantive footing as everyone else’s.
Post #6 recalled UBD’s reference to Professor Hart’s observation that for the equality principle to be, and be seen to be, implemented, we must “take seriously” the requirement that those who adjudicate our rights are demonstrably independent and impartial – and UBD had added “and competent”.
That post also reiterated UBD’s constant theme that our administrative justice system – the part of our justice system that is defined, established and controlled by the executive branch and to which we now entrust the adjudication of the majority of our legal rights disputes – has deliberately flown in the face of that principle and has been allowed to do so by our courts. Thus, in the majority of disputes about their legal rights, members of our society cannot in fact expect equality under the law – cannot reasonably count on their disputes being treated the same as others of a like nature.
(UBD called for structural reforms that would make the principle of equality under the law operational in the administrative justice system by ensuring that our adjudicative tribunals were demonstrably independent and impartial and optimally competent. See Chapter 5.)
And because of the unique value of expert adjudication by specialized tribunals, UBD was also clear that tribunal decisions must be shown deference by the courts.
But one of the anomalies inherent in institutional adjudication by tribunals is that in the normal operation of a tribunal it can and does happen that, in terms of outcomes, like cases may receive unlike treatment. This can happen even if the tribunal, qua institution, and the adjudicator members of the tribunal, are all seen to be independent, impartial and competent.
This occurs typically where there is a disagreement among the tribunal adjudicators as to how a particular provision in an applicable statute should be interpreted in the context of particular facts where those facts are not distinguishable.
Of course, out-of-step decisions that are unreasonable outliers can be corrected by judicial review against the deferential standard of reasonableness in the ordinary course. But if the statute is truly ambiguous when considered in the context of particular facts then it may well be that neither of the conflicting decisions can be said to be unreasonable. In that case, the courts are not entitled to quash either of them, and the Supreme Court has so held. See Domtar Inc. v. Quebec … ,  2 S.C.R. 756 at para. 94.
This may lead to claimants before a tribunal being destined to get one result if they happen to draw one particular adjudicator member, and a different result if by bad luck they draw another. This, of course, is, as the Supreme Court noted in Consolidated Bathurst, the very antithesis of the like-treatment principle.
UBD identified that problem on pages 192/93:
Administrative judicial tribunals also operate in a legal environment that differs significantly from that in which the civil courts operate. In the first place, the doctrine of stare decisis does not typically apply. So, unlike judges, judicial tribunal adjudicators are not required, as far as the law is concerned, to follow previous decisions of their tribunal even if the previous decisions cannot be distinguished.
This was recognized in the Supreme Court of Canada’s 1993 decision in Domtar Inc. […] which held that the fact that two contemporaneous decisions from the same tribunal were in direct conflict with one another did not in itself provide grounds for a court to quash either decision – did not make either of them “unreasonable.”11
UBD faced up to that question and answered it in the referenced footnote – footnote 11, published on pages 320/21:
Here you may be thinking, how can one square this tolerance of conflicting tribunal decisions with the application of the ‘like-cases’ principle? On judicial review, should the courts not have the duty to fix this?”
I answer as follows.
First, an effective adherence to the like-cases principle does not guarantee the exact results even in cases that are identical on the facts. The principle calls for “like” results in “like” cases and these are assured when the law is applied, as Hart says, without “prejudice, interest, or caprice,” and applied, I would add, by a competent adjudicator.
[But] The practical fact that the “like” cases will inevitably be adjudicated by different judges means that there will be differences in the interpretation of the law and in the findings of fact arising from the good-faith idiosyncrasies of the individual judges. The principle will be confirmed as having been applied authentically so long as both results fall within the range of results that are reasonable, relative to the law and the evidence.
Moreover, when neither of the conflicting decisions can be said to be “unreasonable,” it would not be appropriate for the courts to intervene. Faced with two conflicting but reasonable decisions, a court, to intervene, would have to decide that one of the reasonable decisions was not in its view “correct,” thus abrogating the standard of review appropriate for that tribunal and effectively usurping the tribunal’s specialized and expert role in the interpretation of its statute.
THE UNJUST BY DESIGN SOLUTION
Footnote 11, continued:
The reconciling of inconsistent tribunal decisions must be left to the tribunal itself, otherwise its essential mandate as the judicial arm of a statutory rights enterprise is emasculated. The tools this book proposes for accomplishing this include the usual tribunal strategies and structures for “institutionalizing” its decisions – educational and collegial consultation strategies and the like – plus the typical power to reconsider, as well as some additional provisions of a like nature in the reform proposals to follow.
THE FEDERAL COURT OF APPEAL’S ELEGANT SOLUTION
In a judgment written by Stratas J.A. and concurred in by Webb J.A and Near J.A. and dated January 22, 2015 (Wilson v. Atomic Energy of Canada, 2015 FCA 17), the Federal Court of Appeal found itself faced with the very question – what should a Court do when faced with competing, and conflicting, adjudicative interpretations of an adjudicator’s home statute when neither interpretation can be said to be unreasonable.
Taken at face value, the Domtar authority would say that the Court should leave them alone – should leave it to the tribunals themselves to sort it out in future cases.
But the Court’s conclusion in Wilson was that if it can be shown that the current state of adjudication on this particular interpretation issue is one of “persistent discord” then the Court’s role as the protector of the rule of law requires it to assume the role of “tiebreaker” and resolve the discord on the basis of deciding which of the competing interpretations is correct.
The decision manages to be respectful of a tribunal’s autonomy in dealing with early conflicts in the interpretation of its home statute and resolving them itself through its internal dynamics while providing a remedy when the conflict becomes intractable – that is when there is a “persistent discord”.
The issue in Wilson did not in fact arise in the context of conflicting decisions issued by one tribunal, but, rather, in the context of competing decisions of labour arbitrators in which over a long period of time two diametrically opposed lines of arbitration decisions had become established and there was no prospect of the conflict being resolved.
Each line of authority had its strong adherents amongst the arbitrators in the labour field and it had become the classic situation where one’s chances of winning or losing depended entirely on which arbitrator one happened to be landed with.
The relevant passages of Stratas J.A.’s judgment read as follows;
 Normally, a labour adjudicator’s interpretation of a provision in a labour statute would be subject to reasonableness review: Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association, 2011 SCC 61 (CanLII),  3 S.C.R. 654. This, however, is an unusual case. For a long time, adjudicators acting under the Code have disagreed on whether Part III of the Canada Labour Code permits dismissals on a without cause basis.
 Some agree with the adjudicator and the appellant in the case at bar and have concluded that the Code does not permit dismissals on a without cause basis: see …
 Others disagree and have concluded that the Code does permit dismissals on a without cause basis: see …
 In circumstances such as these, what is the standard of review?
 Dunsmuir, supra provides the answer in two ways: one by way of concept, another by way of presumptive rule.
 At the conceptual level, the Supreme Court in Dunsmuir identified two principles that underlie our law of judicial review, principles that are in tension with each other (at paragraphs 27-31). First, there is the constitutional principle of Parliamentary supremacy. Absent constitutional objection, courts are bound by the laws of Parliament, including those that vest exclusive power in an administrative decision-maker over a certain type of decision. Second, there is the constitutional principle of the rule of law. In some circumstances, courts must intervene even in the face of Parliamentary language forbidding intervention: Crevier v. A.G. (Québec) et al., 1981 CanLII 30 (SCC),  2 S.C.R. 220, 127 D.L.R. (3d) 1.
 In this case, it is true that Parliament has vested jurisdiction in adjudicators under the Code to decide questions of statutory interpretation, such as the question before us. However, on the statutory interpretation issue before us, the current state of adjudicators’ jurisprudence is one of persistent discord. Adjudicators on one side do not consider themselves bound by the holdings on the other side. As a result, for some time now, the answer to the question whether the Code permits dismissals on a without cause basis has depended on the identity of the adjudicator. Draw one adjudicator and one interpretation will be applied; draw another and the opposite interpretation will be applied. Under the rule of law, the meaning of a law should not differ according to the identity of the decision-maker: Taub v. Investment Dealers Association of Canada, 2009 ONCA 628 (CanLII), 98 O.R. (3d) 169 at paragraph 67.
 In the case of some tribunals that sit in panels, one panel may legitimately disagree with another on an issue of statutory interpretation. Over time, it may be expected that differing panels will sort out the disagreement through the development of tribunal jurisprudence or through the type of institutional discussions approved in Consolidated‑Bathurst … . It may be that at least in the initial stages of discord, without other considerations bearing upon the matter, the rule of law concerns do not predominate and so reviewing courts should lay off and give the tribunal the opportunity to work out its jurisprudence, as Parliament has authorized it to do.
 However, here, we are not dealing with initial discord on a point of statutory interpretation at the administrative level. Instead, we are dealing with persistent discord that has existed for many years. Further, because no one adjudicator binds another and because adjudicators operate independently and not within an institutional umbrella such as a tribunal, there is no prospect that the discord will be eliminated. There is every expectation that adjudicators, acting individually, will continue to disagree on this point, perhaps forever.
 As a result, at a conceptual level, the rule of law concern predominates in this case and warrants this Court intervening to end the discord and determine the legal point once and for all. We have to act as a tie-breaker.