Issue #114: November 19, 2023
Plastics, best interests of the child, a new general question, expertise
Welcome back after the week-long hiatus. This issue covers cases released over the last two weeks. I may include other cases released during this period in future newsletters.
If there is a case that you think others might be interested in reading, please do not hesitate to send it my way.
Sharp v Autorité des marchés financiers, 2023 SCC 29 (November 17, 2023)
Category: Selection of the standard of review (general questions of central importance)
Context and Analysis: This is a case involving private international law. Four BC residents challenged the Financial Markets Administrative Tribunal’s [FMAT] adjudicative jurisdiction over them as out-of-province defendants. The Court (per Wagner CJC and Jamal J) holds that the standard of correctness applies to the FMAT’s decision regarding its jurisdiction, since it raises a “constitutional issue regarding the territorial reach of provincial legislation” [38].
However, the appellants also argued that the agency’s jurisdiction over out of province parties is to be resolved by applying the Civil Code of Quebec [CCQ]. Taking this argument on its face, the Court says that this question is a “general question of law of central importance to the legal system as a whole” inviting correctness review [39]. While this is the first time—to my knowledge—that this specific question has been recognized as a general question attracting correctness review, it is entirely consistent with the theory underlying this category. The CCQ is the “general law” of Quebec, and relating that general law to “special statutes” that may derogate from the CCQ transcends the particular parties before the Court. Given the CCQ’s status as a general norm, the question of how it relates to specific statutes should be reviewed under the correctness standard. Though reliance on this category is rare, this case is likely one of the situations where correctness is the appropriate standard.
For more on this case, see Paul Daly.
Responsible Plastic Use Coalition v Canada (Environment and Climate Change), 2023 FC 1511 (November 16, 2023).
Category: Application of the reasonableness standard (regulations).
Context: The applicants advance two challenges to a Governor in Council regulatory order listing “plastic manufactured items” as toxic substances under the Canadian Environmental Protection Act [CEPA]. The first, on administrative law grounds, argues that the GIC unreasonably listed plastics as toxic substances without complying with the CEPA; the second argues that the listing is beyond the powers of the federal government, particularly its criminal law power.
Issue: Is the listing unreasonable/unconstitutional?
Holding: “[Plastics were] too broad to be listed on the the List of Toxic Substances in Schedule 1 and this breadth renders the Order both unreasonable and unconstitutional” [10].
Analysis: A significant case that implicates the federal government’s public policy agenda involving the environment and climate change, particularly its approach to plastics. I will focus only on the reasonableness of the regulatory instrument; there is also an issue about the reasonableness of the decision of the Minister of Environment and Climate Change to deny requests for a “Board of Review” to assess alleged risks [4]. Additionally, there is also the constitutional issue that I do not address.
On administrative law grounds, the Court—following the line of cases in the Federal Courts—applies Vavilov to the review of the regulation rather than the more deferential, orthodox approach outlined in Katz (see Issue #68, Issue #70…too many more to count, at this point). The parties differed on the characterization of the regulation. The government—as is common—argued that the GIC order was a policy decision owed extensive deference, and that finding the order unreasonable “limits the options open to government” [53-54]; the applicants suggest that nothing in the order justifies any higher level of deference [55].
As the Court holds, this debate is somewhat besides the point. The question is whether the order is consistent with the statutory authority in the CEPA, which will prescribe the margin of movement the GIC has in listing substances as “toxic”. Here, relying on Vavilovian reasonableness review, the Court conducts a close reading of the statutory authority. The relevant authority, s.90(1) of the CEPA (at the time of the application) provided that “…the Governor in Council may, if satisfied that a substance is toxic, on the recommendations of the Ministers, make an order adding the substance to the List of Toxic Substances in Schedule 1” [62]. Once listed, the CEPA permits the GIC—on recommendation of the Ministers, to make regulations respecting the substance. The question, then, is a legal one prescribed by the statute: are plastics as a broad category a “substance” for the purposes of the CEPA? [68].
The Court finds that the Governor in Council failed to demonstrate that its interpretation of the statute—as permitting the listing of plastics as a toxic “substance”— was justified. Begin with the text. The government suggested, relying on the Interpretation Act, that the singular “substance” in the statute should also be read as plural [73]. This would permit a broad category of “plastics” to be listed. But the Court rightly concludes that this simply isn’t plausible. The use of singular and plural in the CEPA indicates, for the Court, an intentional legislative choice that should not be glossed over by a general gesture towards the Interpretation Act [74, 76]. Doing so would subvert the plan of the statute [76].
The Court is bolstered in this reading—rightly so, in my view—by a purposive interpretation of the text. The regulatory impact statement accompanying the regulatory order contained the supposition that “all plastic manufactured items have the potential to become plastic pollution” [110]. This statement supported, for the govenrment, a broad reading of the statute that permits it to use “robust and efficient tools to prevent pollution” [78]. For the government, a reading of the statute that permitted these tools would run against a requirement for each substance to be listed on its own merits and according to its own assessment. This would, for the government, cause significant delay in the achievement of its environmental goals [78].
But as the Court notes, this reading is based on a “peremptory conclusion” that all plastic items have the potential to become plastic pollution, as noted in the regulatory impact statement [110]. Under Vavilov, not only are these sorts of statements suspect, but taken literally, it would abridge the careful delineation in the statute that requires the listing of individual items (or, in some cases “classes of substances”). No matter how noble the goal of forbidding plastic pollution, the government’s argument expands its power beyond the statute, without an adequate evidentiary basis.
Two observations about the administrative law grounds in this case. First, the case is an excellent example of a rejection of overly-liberal “purposive” argumentation. The government’s arguments were based on a broad reading of the statute that would maximize its ability and flexibility to list broad categories of items as a singular “substance.” But, if accepted, this argument would upset the plan of the statute, which requires a preliminary assessment, to the satisfaction of the GIC, that a substance is, in reality, toxic. Here, the Court finds there is no evidence of this, and the GIC’s extrapolatory statement that all plastic could become pollution was too overbroad. It would mean that the statute’s contemplation of an individualized assessment for a toxic substance would be undermined.
Second, as the Alberta Court of Appeal suggested in Auer v Auer, 2022 ABCA 375, one might argue that the Court’s holding here waded too deep into the policy merits of the regulatory action—the Court, for example, looked carefully at whether there was evidence to support a conclusion that plastics as a broad category are inherently toxic, or would become so [110]. However, I do not see things this way. The evidentiary link is itself required by the statute, which mandates that the GIC be “satisfied” that a substance is toxic before listing it. This is a condition precedent that invites a qualitative analysis on the part of the GIC. For a court to assess this analysis under a reasonableness standard, some evidence must be present. The Court does not itself weigh this evidence—this would be the court arrogating to itself a policy-making function—but rather, it is only an assessment of the evidence as it pertains to the legal standard.
Farhat v Canada (Citizenship and Immigration), 2023 FC 1427 (October 26, 2023)
Category: Application of the reasonableness standard.
Context: This is a decision of a Senior Immigration Officer refusing Farhat’s application for permanent residence on the basis of humanitarian and compassionate factors [H&C]. An aspect of Farhat’s H&C application pertained to her status as a significant caregiver for her grandchild, in addition to her establishment in Canada and the potential hardships she would encounter on returning to Lebanon. The Officer decided that, in relation to the grandchild, there was “insufficient evidence to show that the Applicant’s return to Lebanon would compromise her granddaughter’s best interests” [13].
Issue: Is the decision reasonable?
Holding: No.
Analysis: The Court finds multiple flaws in the decision—a decision which repays careful reading. Among other things, the Court finds that the Officer erected a requirement of “exceptional” establishment, and the Officer “failed to explain ‘what more’ could have been reasonably expected of the Applicant, in her particular circumstances, for her establishment to be deemed worthy of consideration” [31].
Importantly, Regimbald J explores an interesting issue in the best-interests-of-the-child analysis. At the hearing, the government conceded that “cultural, linguistic, and religious grounds are also relevant in determining the best interests of the child, in relation to the child’s development and education” [21]. Here, Regimbald J finds that the Officer failed entirely to consider the best interests of the child “in her emotional, cultural, linguistic and religious development” [47]. It was not enough for the Officer to say that childcare arrangements could be made for the child in Canada in Farhat’s absence, because “[r]egular childcare arrangements in Canada do not necessarily cater to specific cultural, linguistic and religious considerations…” [47].
Regimbald J’s decision in this case is one of the clearest statements out there on the best interests of the child analysis, particularly as it pertains to the linguistic and cultural needs of children. The analysis will need to be individualized to the child’s global needs.
International Union of Elevator Constructors, Local 102 v Kone Inc, 2023 SKKB 234 (November 1, 2023)
Category: Application of the reasonableness standard.
Context and Analysis: Local 102 applied for judicial review of a labour arbitrator’s decision on a question of interpretation involving the collective bargaining agreement. In applying the reasonableness standard, Kone argued that a “high level of deference” was owed to a labour arbitrator, specifically [41]. The Court rejects the proposition that additional deference “based on the decision-maker’s credentials” is an appropriate application of the reasonableness standard [43]. The reasonableness standard is a single standard, and “it is not appropriate to further subdivide the degree of deference applied…based on the identity of the decision-maker” [43].
I agree wholeheartedly with this comment. A single reasonableness standard, coupled with Vavilov’s rejection of an always-applicable assumption of administrative expertise, means we should not contextualize the reasonableness standard based solely on the identity of the decision-maker. This is not a full rejection of a contextual approach to the reasonableness standard—indeed, Vavilov’s constraints will change the margin of appreciation owed to a decision-maker. And it might be true that the line between the statement that reasonableness is a single standard and that Vavilov is contextual is thin. Nonetheless, we can at least say that the reasonableness standard does not apply more or less intensely based solely on the identity of the decision-maker.
On the other hand, the Court’s comment that “[i]t may be appropriate to consider the context including the involvement of an experienced labour arbitrator when the issue is whether the standard of review is reasonableness or correctness” [43]. This is an oddity. As I’ve noted, presumptive expertise is no longer the name of the game, and Vavilov’s rubric for selecting the standard of review is rather categorical rather than contextual. The status of the decision-maker as an “experienced labour arbitrator” is not relevant to the selection of the standard of review.
This aside, the SKKB’s rejection of a multi-layered reasonableness standard is welcome. It stands in stark contrast to the approach in the Ontario courts, where arguably a subdivided reasonableness standard has emerged in labour board decisions: see Turkiewicz, 2022 ONCA 780. Though it is true that the application of expertise is an appropriate consideration in conducting reasonableness review—and perhaps the Ontario approach can be read in this light—Turkiewicz seems to come close to granting “more” deference on the basis of the assumed expertise of labour boards as a class of decision-maker. To the extent this is true, or that Turkiewicz can be taken in this direction, the SKKB approach is to be preferred.