Ensuring Consistency in Administrative Adjudication: Canadian Association of Refugee Lawyers v Canada (Immigration, Refugee and Citizenship), 2020 FCA 196
The Immigration and Refugee Board is Canada’s largest administrative tribunal. It makes tens of thousands of determinations each year in refugee and immigration matters where the stakes are as high as they can be in administrative adjudication. Ensuring consistent decision-making in a high-volume tribunal, with members spread out across the country, is a significant challenge. In particular, refugee claimants may present similar fact patterns to different decision-makers located all over Canada: claimants may come from the same country, from similar regions or have similar fears of persecution.
In Canadian Association of Refugee Lawyers v Canada (Immigration, Refugee and Citizenship), 2020 FCA 196, the Federal Court of Appeal considered the validity of “jurisprudential guides” issued by the Chair of the IRB to refugee decision-makers. These guides are, in essence, decisions of IRB members which are particularly useful in determining issues which may arise in refugee proceedings; indeed, they provide analytical frameworks or starting points for dealing with common issues. At first instance, the Chief Justice of the Federal Court held that these guides were unlawful fetters on the adjudicative independence of IRB members (2019 FC 1126, noted here). In a persuasive set of reasons by de Montigny JA, the Federal Court of Appeal disagreed.
There was a preliminary issue relating to whether the Chair has the authority to issue jurisprudential guides in the first place. The Immigration and Refugee Protection Act, s. 159(1)(h) provides that the Chair “may issue guidelines in writing to members of the Board and identify decisions of the Board as jurisprudential guides…to assist members in carrying out their duties…” CARL argued that the provision could not be interpreted to permit the issuing of guidelines relating to particular factual situations which interfere with the adjudicative independence of members. Assessing s. 159(1)(h) in its context, however, de Montigny JA was clear that it “conferred in the broadest terms” the power to issue guidelines (at para. 46). The crux of the matter, as de Montigny JA recognized (at para. 50), was whether the jurisprudential guides substantively interfered with members’ adjudicative independence, a separate question from whether the Chair was authorized to issue them in the first place.
It is notable that the Chair did not provide reasons explaining his interpretation of s. 159(1)(h). Given that the jurisprudential guides were issued prior to Vavilov, this is unsurprising, but retrofitting Vavilov’s emphasis on reasons to pre-Vavilov decisions is a tricky exercise. Here, de Montigny JA relied on the Chair’s “implicit interpretation” to support the reasonableness of issuing the guides (at para. 44), but the letter and spirit of Vavilov counsel against reliance on implicit interpretations in conducting reasonableness review. The better course here, I think, was to say (as de Montigny did) that CARL’s argument conflated the existence of statutory authority with its exercise. Moreover, as de Montigny JA noted (at para. 45), a decision-maker does not need express statutory authority to issue guidelines in the first place, such that any dispute about the scope of s. 159(1)(h) was moot.
The central issue, therefore, was whether the jurisprudential guides interfered with members’ adjudicative independence. Centrally important here are two venerable Supreme Court precedents: Iwa v. Consolidated-Bathurst Packaging Ltd.,  1 SCR 282 and Ellis-Don Ltd. v. Ontario (Labour Relations Board), 2001 SCC 4,  1 SCR 221. CARL argued that these cases establish that outside interference in factual determinations made by IRB members is unlawful: any policy guidance or internal discussion of decision-making can only focus on issues of law and policy and not on questions of fact. Yet, here, the jurisprudential guides provided starting points on factual issues, from which any divergence by an IRB member would have to be justified. This, in CARL’s submission, was unlawful.
de Montigny JA disagreed. To begin with, consistency is a “legitimate goal” (at para. 66) for administrative tribunals. In the IRB’s case, consistency is particularly important:
[T]he IRB is a high volume tribunal that annually receives and decides thousands of claims and appeals. There are hundreds of decision-makers across various regions of the country. Moreover, judicial review is subject to leave from the Federal Court. In that context, the need for consistency is even more obvious, and as this Court recognized in [Thamotharem v. Canada (Minister of Citizenship and Immigration), 2007 FCA 198,  1 FCR 385], the use of guidelines and other soft law techniques to achieve an acceptable level of consistency is particularly important for large tribunals exercising discretion such as the IRB (at para. 67).
Moreover, in Vavilov, “the Court found that administrative bodies may resort to guidelines and other soft law techniques to address” concerns about consistency and, indeed, “went so far as stating that a departure from longstanding practices or established internal authority without any explanation for so doing may be a badge of unreasonableness” (at para. 68). (Notably, the Chief Justice of the Federal Court made his decision prior to Vavilov.)
Three particular features of the jurisprudential guides supported their lawfulness. First, the factual issues referred to in the guides are not just any type of factual issues:
While these can be characterized as factual findings, they are of a special nature to the extent that they go beyond the evidence specific to any particular claimant. The accuracy of the review of a specific country condition with respect to IFAs, state protection or objective fear is not dependent upon a claimant’s specific circumstances, and is not meant to be. After all, the objective of refugee determination is not so much to determine what has happened, but what will happen if a person is returned to his or her country of origin (at para. 74).
Second, the jurisprudential guides are issued in a transparent manner:
The JGs are meant to apply to all claimants originating from the same country to which they are directed and whose situation broadly raise the same issues. They are also clearly identified and posted on the IRB website, and are readily available to all claimants and their counsel. They are therefore much less susceptible [than the closed-door proceedings in Consolidated Bathurst and Ellis Don] to give rise to an apprehension of coercion or to a perception of interference by superiors.
Third (and I would say, critically), the guides do not actually dictate conclusions to IRB members. Fundamentally, “it is understood that the departure from a JG will be reasonable if justified, that is, when the decision-maker has provided proper reasons for doing so” (at para. 84). Indeed, IRB members “should be expected to know the well established legal principle that soft law tools such as guidelines are non-binding” (at para. 86). de Montigny JA was not persuaded that the evidence in the record proved that IRB members were under so much undue pressure from hierarchical superiors to follow the jurisprudential guides that their adjudicative independence was compromised (at paras. 79-85 and 87).
de Montigny JA did note “that JGs on findings of fact are fraught with risks and difficulties” (at para. 89). Country information is constantly changing, which requires the “utmost vigilance” on the part of counsel and claimants (at para. 90). This was not enough to condemn the lawfulness of the jurisprudential guides, only enough to counsel caution in issuing them (at para. 91). With respect, however, de Montigny JA seems to have been overly concerned about valid claims being dismissed because of reliance on the jurisprudential guides. This is one type of false negative, but a system with no jurisprudential guides could generate other false negatives; indeed, a non-jurisprudential guide system could perhaps more readily generate false positives, wrongly granting claims in the absence of common starting points for analysis or analytical frameworks. That quibble aside, one cannot disagree with de Montigny JA’s ultimate conclusion: “Adjudicative decision-makers…must always use their discretionary powers wisely, and strive to avoid sacrificing fairness to consistency and expediency” (at para. 91). The calculus here is difficult, but I would not be minded to lightly interfere with the IRB’s analysis of the costs and benefits of its strategies for ensuring consistency.
Lastly, de Montigny JA made important observations about administrative tribunal decision-writing in batting away the argument that the jurisprudential guide was tainted by a reasonable apprehension of bias because the member’s hierarchical superiors might have participated in the formulation of the underlying decision. Many administrative tribunal members have a tendency to think that their adjudicative independence means that any discussion of draft reasons is absolutely forbidden. This is absolutely incorrect:
There is also no indication in the record tending to show that the Board member improperly consulted other members while she was deliberating, or that the circulation of draft reasons to other members for their comments was not voluntary. At the deliberation stage, members are not precluded from consulting other members with more experience to ensure consistency in decision-making. The Federal Court appropriately found that the review process was entirely consistent with the principles set out in Consolidated-Bathurst and Ellis-Don. While the Acting Deputy Chairperson reviewed drafts of the member’s decision, under the IRPA, the Chairperson and Deputy Chairperson are also members of the RAD and paragraph 159(1)(h) does not prohibit them from suggesting changes to a draft at a deliberative stage (at para. 98).
Indeed, the IRB has a reasons review policy. It is published on the IRB’s website. There is, as de Montigny JA made clear, nothing improper about consulting on a draft decision. Indeed, it may enhance the quality of the final decision and the overall consistency of a tribunal’s decision-making.
It is possible that the Supreme Court will grant leave to appeal in this case, having not addressed adjudicative independence for many years. Regardless, this is a very significant decision, with de Montigny JA setting out the applicable principles with admirable clarity. As a consequence, the IRB will no doubt be encouraged to further develop its consistency-advancing practices. Other high-volume tribunals should also take note.
This content has been updated on December 9, 2020 at 15:20.