Can This Be Correct? Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61

Another week, another underwhelming standard-of-review decision from the Supreme Court of Canada — this time in Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61.

At issue here was the scope of the Minister for Citizenship and Immigration’s authority to waive statutory conditions for “humanitarian or compassionate” reasons, an authority expressly conferred by s. 25(1) of the Immigration and Refugee Protection Act. K fled from Sri Lanka and applied for refugee status in Canada. When his application was refused, he applied to the Minister seeking permission to apply for permanent residence. At the time of this application, he was 17 years old and had been in Canada for almost two years.

Guidelines adopted by the Minister put some flesh on the bones of the bare reference to “humanitarian and compassionate considerations”. Immigration officers are advised that applicants should demonstrate unusual or undeserved hardship — something not anticipated by the regulatory scheme that is beyond the applicant’s control — or disproportionate hardship — an unreasonable impact on the applicant due to their personal circumstances. Relevant factors to consider are also set out in the Guidelines:

•               establishment in Canada;

•               ties to Canada;

•               the best interests of any children affected by their application;

•               factors in their country of origin (this includes but is not limited to: Medical inadequacies, discrimination that does not amount to persecution, harassment or other hardships that are not described in [ss. 96 and 97]);

•               health considerations;

•               family violence considerations;

•               consequences of the separation of relatives;

•               inability to leave Canada has led to establishment; and/or

•               any other relevant factor they wish to have considered not related to [ss. 96 and 97]. (ss. 96 and 97 address asylum claims by refugees)

In her “analysis” section, Abella J., for the majority, described the effect of this as follows: “what officers should not do, is look at s. 25(1) through the lens of the three adjectives as discrete and high thresholds, and use the language of ‘unusual and undeserved or disproportionate hardship’ in a way that limits their ability to consider and give weight to all relevant humanitarian and compassionate considerations in a particular case” (at para. 33).

Under s. 25(1) reference is also required to the “best interests” of any child affected by the application, interests which require “a singularly significant focus and perspective” from the decision-maker (at para. 40). Here, the applicant himself was a child, which “trigger[ed] not only the requirement that the ‘best interests’ be treated as a significant factor in the analysis, [but] should also influence the manner in which the child’s other circumstances are evaluated” (at para. 41).

After taking the time to explain the proper interpretation of s. 25(1) to the world at large (paras. 11-41), Abella J. turned (a little late…) to the question of the standard of review. In the Federal Court of Appeal, Stratas J.A. had argued that because a question of general law had been certified by the first-instance judge for resolution by the Federal Court of Appeal, it should provide an authoritative answer to the interpretation of s. 25(1), the functional equivalent of correctness review. Abella J. rejected this approach:

The Federal Court of Appeal refers to one case from this Court to support this point: Hilewitz v. Canada (Minister of Citizenship and Immigration), 2005 SCC 57 (CanLII), [2005] 2 S.C.R. 706. This case is not particularly helpful. It was decided before Dunsmuir v. New Brunswick, 2008 SCC 9 (CanLII), [2008] 1 S.C.R. 190, there was no discussion of the impact of a certified question on the issue of standard of review, and the parties asked that correctness be applied: para. 71.  In any event, the case law from this Court confirms that certified questions are not decisive of the standard of review: Baker, at para. 58; Chieu v. Canada (Minister of Citizenship and Immigration), 2002 SCC 3 (CanLII), [2002] 1 S.C.R. 84, at para. 23.  As the Court said in Baker, at para. 12, the certification of a question of general importance may be the “trigger” by which an appeal is permitted. The subject of the appeal is still the judgment itself, not merely the certified question. The fact that the reviewing judge in this case considered the question to be of general importance is relevant, but not determinative. Despite the presence of a certified question, the appropriate standard of review is reasonableness: Baker, at para. 62 (at para. 44).

There is a lot to unpack here.

First, the Federal Court of Appeal is taken to task for referring to a pre-Dunsmuir case in support of correctness review. It is said to be “not particularly helpful” because it “was decided before” Dunsmuir. But Abella J. herself refers to a pre-Dunsmuir case (Baker) to support her proposition that the standard of review is reasonableness and to another one to support her proposition that the presence of a certified question is not determinative of the standard of review.

Second, there is no mention of Pushpanathan, [1998] 1 SCR 982, where Bastarache J., for a unanimous Court, stated that the certified question procedure “would be incoherent if the standard of review were anything other than correctness” (at para. 43). He explained:

The general importance of the question, that is, its applicability to numerous future cases, warrants the review by a court of justice.  Would that review serve any purpose if the Court of Appeal were obliged to defer to incorrect decisions of the Board?  Is it possible that the legislator would have provided for an exceptional appeal  to the Court of Appeal  on questions of  “general importance”, but then required that despite the “general importance” of the question, the court accept decisions of the Board which are wrong in law, even clearly wrong in law, but not patently unreasonable? (at para. 43)

Abella J. treats one pre-Dunsmuir  decision as establishing reasonableness as the standard of review, but ignores another pre-Dunsmuir decision, decided just months earlier.

Actually, it is wrong to say that Abella J. ignores Pushpanathan. In fact, at one point in her reasons, she paraphrases it. The closing sentences of her para. 44 paraphrase para. 25 of Pushpanathan. Let’s put them side by side:

Pushpanathan, para. 25: The certification of  a “question of general importance” is the trigger by which an appeal is justified.  The object of the appeal is still the judgment itself, not merely the certified question.

Kanthasamy, para. 44:  As the Court said in Baker, at para. 12, the certification of a question of general importance may be the “trigger” by which an appeal is permitted. The subject of the appeal is still the judgment itself, not merely the certified question.

Why would Abella J. paraphrase a decision, but not cite to it?

Third, there is nonetheless something to Abella J.’s observation that “[t]he fact that the reviewing judge in this case considered the question to be of general importance is relevant, but not determinative” (at para. 44). At the very least, the certification of a question should not be determinative of the standard of review, because a first-instance judge might (for one reason or another) certify a question with significant discretionary elements. And the discussion in Baker, at paras. 57-62, makes a persuasive case for reasonableness as the standard of review of s. 25(1) decisions. However, as Abella J.’s own analysis (at paras. 11-41) demonstrates (as does, indeed, the Federal Court of Appeal decision), it is possible to address the interpretation of s. 25(1) separately from the application of s. 25(1) to the facts of the case. A segmentation of the decision, to be sure, but one that is, as Stratas J.A. put it “an artefact of having a certified question put to us” (at para. 36).

Fourth, it might be said that the certified question procedure is, despite what was said in Pushpanathan and by the Federal Court of Appeal in Kanthasamy, merely a technical provision designed as a filter to ensure the Federal Court of Appeal is not swamped by immigration judicial reviews appealed to it from the Federal Court. That may be so. But in Tervita, another technical-looking provision — that decisions of the Competition Tribunal should be treated as if they were judgments of the Federal Court — was held to require correctness review. How can oblique language in the Competition Tribunal Act justify correctness review but the certified question procedure not justify it? There is no mention of Tervita in Abella J.’s reasons.

Fifthly, the Federal Court of Appeal has repeatedly maintained — and others have recently joined in — that it is possible by reference to contextual factors to rebut the presumption of deference when an administrative decision-maker is interpreting its home statute.  I watched the oral hearing of Kanthasamy and there was lengthy argument on the lack of legal expertise of front-line immigration officers. But there is no discussion of this in Abella J.’s reasons. Recent Supreme Court of Canada cases simply do not engage with this issue. When the presumption of deference is rebutted, it is for highly technical reasons that are divorced from the actual context of the decision-making process at issue. Abella J. merely tells us that the certified question procedure is “relevant” but not “determinative”. What would be?

The end result is that a blanket presumption that courts must defer on questions of law to front-line decision-makers who are not lawyers and whose independence may be compromised. Perhaps this hard-and-fast rule is a better idea than the contextual approach that people like me preach. But if the Supreme Court does not defend it, other courts are likely to undermine it, and we are in for — hurrah — even more lengthy disquisitions on the standard of review.

This content has been updated on December 11, 2015 at 20:14.