Facts:
Section 34(1)(e) of the Immigration and Refugee Protection Act (“IRPA”) states that permanent residents or foreign nationals are “inadmissible on security grounds” for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada.”
M and D are both foreign nationals. M was previously charged with two counts of attempted murder and two counts of discharging a firearm following an argument in a bar (the charges were stayed due to delay). D allegedly engaged in violent acts against former partners and other persons. Some of the charges against D for these incidents were stayed; for three charges, D pled guilty and got a conditional discharge. It was not alleged that either M or D engaged in acts of violence with a link to national security or the security of Canada.
M and D underwent admissibility hearings before the Immigration Division (“ID”), where the focus was on whether their past conduct fell within the scope of s. 34(1)(e). The ID held that M’s conduct lacked any element that would elevate it to “security grounds” and thus s. 34(1)(e) could not apply. On appeal before the Immigration Appeal Division (“IAD”), this decision was reversed, on the basis that “security” should be understood in a broader sense—namely, as ensuring Canadians are secure from acts of violence that would or might endanger their lives or safety. Thus, M was inadmissible under s. 34(1)(e). Applying this same approach to D’s case, the ID concluded D was also inadmissible.
M and D applied for judicial review. The Federal Court allowed the applications, holding that it was unreasonable to interpret s. 34(1)(e) as applying to acts of violence without a nexus to national security. It certified two “serious questions of general importance” for appeal relating to the interpretation of s. 34(1)(e), pursuant to the certified questions regime in s. 74(d) of the IRPA. The Minister appealed and the Federal Court of Appeal allowed the appeal, holding that it was reasonable to interpret s. 34(1)(e) as not requiring a nexus to national security or the security of Canada. M and D then appealed, with leave, to the Supreme Court of Canada.
Decision:
Appeal allowed; administrative decisions at issue set aside (per Wagner CJ and Karakatsanis, Rowe, Martin, Kasirer, Jamal and O’Bonsawin JJ; Côté J, concurring).
The standard of review is reasonableness. None of the six categories of correctness review apply.1 In particular, although the proper interpretation of s. 34(1)(e) is important for the affected persons and the proper administration of the IRPA, it does not affect the legal system or the administration of justice as a whole, have legal implications for many other statutes, or affect other institutions of government. The fact that the Federal Court certified questions under s. 74(d) of IRPA does not affect the standard of review to be applied.
The Federal Court of Appeal strayed from the reasonableness review methodology set out in Vavilov.2 Rather than starting its analysis with the reasons of the decision-maker, the Court of Appeal grafted on an extra step of conducting a preliminary analysis of the text, context and purpose of the legislation to understand the statutory “lay of the land”. This is inconsistent with Vavilov and risks leading a court to slip into correctness review.
Reviewing courts must conduct reasonableness review mindful of the impact of the decision on the affected individual. The reasons provided by the administrative decision-maker must reflect the stakes. Here, the impact on M and D’s rights and interests is severe, as the decisions impact whether two individuals could be deported from Canada. Yet the IAD’s reasons failed to address critical points of statutory context and the broad consequences of its interpretation of s. 34(1)(e) that were raised by M and were “core planks” of his position (para 97). While not raised by the parties, the IAD also failed to consider that the IRPA must always be interpreted with Canada’s international obligations in mind, including the principle of non-refoulement, which is the cornerstone of the international refugee protection regime.
These significant omissions reflected a failure of responsive justification and, cumulatively, rendered the IAD decision unreasonable. The ID’s decision in D’s case, which simply followed the IAD’s interpretation in M’s case, was unreasonable for the same reasons.
The relevant legal constraints point overwhelmingly to only one reasonable interpretation of s. 34(1)(e): the provision requires a nexus to national security or the security of Canada. The provision can only be invoked to render a person inadmissible when their acts of violence that would or might endanger the lives or safety of persons in Canada have a nexus with national security or the security of Canada.
In her concurring opinion, Justice Côté would have recognized a new category of correctness review for certified questions under s. 74(d) of the IRPA. Questions certified under that provision will, by definition, have implications beyond the immediate parties and raise issues of broad significance within Canada’s immigration and refugee protection scheme. Both Parliamentary intent and the rule of law require a singular, determinate and final answer to a question certified as a serious question of general importance under the IRPA. She held that the majority’s interpretation of s. 34(1)(e) is correct.
Commentary:
The Court’s decision in Mason is notable for a number of reasons.
First, with respect to standard of review, the Court holds the line on limiting the categories of correctness review. Despite a rather compelling set of arguments from Justice Côté on why correctness review would be more consistent with the underlying rule of law and respect for legislative intent rationales set out in Vavilov, the majority is clearly not interested in opening that door when it comes to certified questions, reaffirming the Court’s previous holding that such matters do not require correctness review.3 In the result, the decision effectively prioritizes predictability and stability over any further evolution when it comes to the standard of review analysis, sending a message to litigants that they should not be quick to argue for new categories of correctness review—especially where the matter has been settled (even if pre-Vavilov).
Second, the majority’s instruction that reviewing courts should not adopt the “lay of the land” approach to reasonableness review of statutory interpretation is an important, if somewhat puzzling, directive. It is important because counsel will now have to take extra caution not to frame cases in a way that could invite a “lay of the land”-type analysis, and to adhere closely to a “reasons-first” approach. At the same time, it is puzzling because it is hard to object to reviewing courts simply looking at the full relevant statutory context and purpose to ensure they have a basic understanding of the statutory landscape before their review of an administrative decision that focuses on particular aspects of that scheme. Of course, this analysis cannot go too far: a reviewing court cannot conduct its own statutory interpretation analysis and then use that as a “yardstick to measure what the administrator did” (Vavilov at para 83). But merely gaining familiarity with the statutory lay of the land, in order to ensure the reviewing court has the relevant background for its consideration of the administrative decision, seems not only appropriate but advisable in many cases.
In the end, it is not practically possible to conduct judicial review without getting a “lay of the land” with respect to the statute at issue; indeed, Vavilov’s requirement that administrative decisions must be consistent with statutory text, context and purpose of the relevant provisions requires such an undertaking by reviewing courts. The question is really one of the analytical order in which this step is undertaken. Following Mason, the “lay of the land” step will likely occur—at least on paper—following the initial review of the administrative decision. This respects Mason’s insistence on the need for a “reasons-first” approach, while not sacrificing the obvious need for a reviewing court to properly understand the statutory context and purpose of the legislative regime at issue when assessing reasonableness.
Third, Mason serves as a stark reminder that responsive justification requires the reasons of administrative decision-makers to grapple with the submissions of the parties, at least insofar as they relate to “key issues”, “central arguments” (Vavilov at para 128) or “core planks” (Mason at para 97). Decision-makers should address such issues explicitly, otherwise reviewing courts will not be quick to draw an inference that an argument has been considered or addressed; for example, the majority in Mason parted ways with the Federal Court of Appeal’s conclusion that the decision-makers had implicitly considered some of M’s arguments (see paras 96 and 101). Thus, while reviewing courts must review decisions “with sensitivity to the institutional setting and in light of the record” (Vavilov at para 96), Mason suggests there are limits to how far reviewing courts will go in overlooking an administrative decision-maker’s failure to expressly address a key issue raised by one of the parties.
Fourth, Mason illustrates that there will be other legal constraints that bear on administrative decisions—even if those issues are not raised by the parties. In particular, the majority relies on the fact that Parliament intended the IRPA to be interpreted in conformity with Canada’s international obligations (as made explicit in the text of the statute), which is a point that the administrative decision-makers never considered. Despite the parties never raising this issue, the Court concluded it was unreasonable to adopt an interpretation that allows foreign nationals to be returned to countries where they may face persecution, contrary to Canada’s non-refoulement obligations under international law.
Finally, when it comes to the ongoing debates over whether Vavilov has ushered in an era of heightened reasonableness review, Mason is an important data point suggesting that it has. Despite paying heed to the need for a “reasons-first” approach, the substance of the Court’s reasonableness review is quite robust indeed—in terms of the standard expected for responsive justification given the stakes, in terms of taking the decision-makers to task for failing to (expressly) consider points of statutory context, and in terms of relying on issues not raised by the parties as a limiting legal constraint on what constitutes a reasonable decision.
- Those categories are: (i) where the legislature explicitly prescribes standard of review; (ii) where the legislature provides for a statutory right of appeal to a court; (iii) constitutional questions; (iv) general questions of law of central importance to the legal system as a whole; (v) questions related to jurisdictional boundaries between two or more administrative bodies; and (vi) where courts and administrative bodies have concurrent first instance jurisdiction over a legal issue in a statute. ↩︎
- Canada (Minister of Citizenship and Immigration) v Vavilov, 2019 SCC 65. ↩︎
- Kanthasamy v Canada (Citizenship and Immigration), 2015 SCC 61 at para 44. ↩︎