RENNIE J.A.
I. Overview
[1] The Governor in Council (GIC) issued an order under subsection 90(1) of the Canadian Environmental Protection Act, 1999, S.C. 1999, c. 33 (CEPA 1999, the Act) listing “plastic manufactured items”
(PMI) on Schedule 1 of that Act as a toxic substance. The Order was published in the Canada Gazette Part II, Volume 155 Number 10 on May 12, 2021, contemporaneously with a regulatory impact analysis statement (RIAS). Two reports preceded the Order. The “Science Assessment of Plastic Pollution”
(Science Assessment) and “A Proposed Integrated Management Approach to Plastic Products to Prevent Waste and Pollution”
(Discussion Paper) were published October 7, 2020.
[2] Two days prior to the issuance of the Order, the Minister of Environment and Climate Change (MECC, the Minister) refused requests to establish a board of review (BOR) under section 333 of CEPA to further assess the environmental risks associated with PMI.
[3] In the Federal Court, the applicants, Responsible Plastic Use Coalition (RPUC), Dow Chemical Canada ULC, Nova Chemicals Corporation, and Imperial Oil Limited, (collectively the industry respondents), challenged the GIC’s order and the Minister’s decision to not constitute a BOR. RPUC is a not-for-profit corporation of which the three industry respondents are members. RPUC’s mandate is to “pursue all legal remedies available to prevent the regulation of plastic manufactured items under CEPA.”
[4] The Federal Court found that PMI were not a “substance”
under CEPA, that the Order listing PMI was unreasonable, that the Order was ultra vires the federal criminal law power and that the decision not to establish a BOR was unreasonable (Responsible Plastic Use Coalition v. Canada (Environment and Climate Change), 2023 FC 1511. The Attorney General appeals.
[5] I would allow the appeal. The principles set forth in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65 (Vavilov) and Auer v. Auer, 2024 SCC 36 (Auer) guide the analysis that follows and, when applied, lead to the conclusion that the decisions of the GIC and the Minister were reasonable. Those principles warrant repeating.
[6] Vavilov requires courts conducting judicial review to assess the reasonableness of a decision in light of the constraints bearing on the decision maker, the primary of which is the empowering legislation. Further constraints include the facts before the decision maker, the common law and the decision maker’s past practices. Vavilov also instructs that reasonableness review be conducted with a view to understanding the decision and to assess the reasons in context against the measures of transparency, justification and intelligibility (Vavilov, at para. 99; Innovative Medicines Canada v. Canada (Attorney General), 2022 FCA 210, at para. 44).
[7] However, there are circumstances such as this where the decision maker is under no obligation to provide reasons; this is particularly the case where the decision maker is at the apex of the executive (Auer, at paras. 52–53, citing Mancini, Mark P. “
One Rule to Rule Them All: Subordinate Legislation and the Law of Judicial Review”
(2024), 55 Ottawa L. Rev. 245, at 278–279). In these situations, “something akin to justification”
may be found in background and contextual documents which shed light on the rationale and evidence underlying the decision (Vavilov, at para. 137). In the case of decisions of the GIC, the rationale may be found in the text of the legislation and associated instruments, such as the RIAS (Portnov v. Canada (Attorney General), 2021 FCA 171, at para. 34). Resort may also be had to Hansard, Parliamentary committee debates and government policy papers foreshadowing or accompanying the legislation.
[8] Of particular pertinence to this appeal is the admonition that reviewing courts must be careful not to “make [their] own yardstick and then use that yardstick to measure what the administrator did”
(Vavilov, at para. 83, citing Delios v. Canada (Attorney General), 2015 FCA 117, at para. 28). This is a corollary of the proposition just noted, that judicial review methodology requires courts to read the reasons with a view to understanding and not to engage in the oft-quoted “treasure hunt for error”
(Vavilov, at para. 102).
[9] In the ordinary course of judicial review and consistent with a “reasons-first”
approach, a reviewing court does not start with its own independent analysis of the statutory scheme. As Jamal J. cautioned in Mason v. Canada (Citizenship and Immigration), 2023 SCC 21 (Mason) at para. 79 “starting with its own perception of the merits may lead a court to slip into correctness review.”
Here, however, other than what can be gleaned from the context, past practice and documents associated with the legislative process as mentioned, there are no reasons. Consequently, the point of departure for reasonableness review in this case is an understanding of the legislative scheme.
[10] An order under subsection 90(1) (reproduced at Annex A) does not require that the substance, in fact, enter the environment. Nor does listing require a substance be harmful in all its manifestations. To list, the GIC must “be satisfied”
that the criteria of section 64 are met, and section 64 only requires that a substance “may”
enter the environment and “may”
cause harm:
|
64 For the purposes of this Part and Part 6, except where the expression “inherently toxic” appears, a substance is toxic if it is entering or may enter the environment in a quantity or concentration or under conditions that
|
64 Pour l’application de la présente partie et de la partie 6, mais non dans le contexte de l’expression « toxicité intrinsèque », est toxique toute substance qui pénètre ou peut pénétrer dans l’environnement en une quantité ou concentration ou dans des conditions de nature à :
|
|
(a) have or may have an immediate or long-term harmful effect on the environment or its biological diversity;
|
a) avoir, immédiatement ou à long terme, un effet nocif sur l’environnement ou sur la diversité biologique;
|
|
(b) constitute or may constitute a danger to the environment on which life depends; or
|
b) mettre en danger l’environnement essentiel pour la vie;
|
|
(c) constitute or may constitute a danger in Canada to human life or health.
|
c) constituer un danger au Canada pour la vie ou la santé humaines.
|
[11] The decision to list a substance and the decision as to how and to what extent, if any, to regulate it, are discrete steps; first under subsection 90(1) and subsequently, under subsection 93(1) (reproduced at Annex A). That is the scheme established by Parliament and it is, as I will explain, also the scheme as understood by the Supreme Court in R. v. Hydro-Québec, [1997] 3 S.C.R. 213, 1997 CanLII 318 (SCC) (Hydro-Québec).
[12] The Order itself, on its face, reflects the two-stage approach orchestrated by sections 90 and 93.
[13] The Order anticipates the winnowing or narrowing of a broad listing through a triage of risks. The Order enables “[M]i
nisters to propose risk management measures under CEPA on certain
plastic items to manage the potential ecological risks associated with those items
becoming plastic pollution”
(RIAS, emphasis added). This is unsurprising given CEPA’s two-stage process, which demands exquisite particularization, if and when subsection 93(1) regulations are enacted.
[14] The Federal Court’s conclusion that the Order was too broad was predicated on an incorrect premise, one that was inconsistent with the express language of the statute and with the principles governing the interpretation of subordinate legislation. By requiring precision in the description of individual plastic items at the listing stage, the Federal Court effectively collapsed the legislative process into a single step, rendering subsection 93(1) superfluous. This misunderstanding of the scheme lead the Federal Court to reason that because not all plastics enter the environment, not all plastics cause harm; consequently, it required the specific identification of the particular plastics that enter the environment, and that only those particular plastics be listed in the Order.
[15] But this is not the test that Parliament imposed. At the listing stage, with which we are concerned, all that section 64 and subsection 90(1) of CEPA require is the potential to cause harm. On this point, as we shall see, the evidence before the GIC was unequivocal.
[16] The Federal Court also noted that only 1% of PMI enter the environment each year. The respondents latch on to this, contending that this fact is conclusive evidence that the Order is unreasonable. This argument fails as it, in effect, invites this Court to develop its own measurement or yardstick to assess the reasonableness of the Order. Simply put, the respondents set aside the problem of plastic pollution as considered by the GIC and define the problem in a different manner. This reasoning runs head-first into a cardinal principle of judicial review as expressed by the Supreme Court:
It follows that the focus of reasonableness review must be on the decision actually made by the decision maker, including both the decision maker’s reasoning process and the outcome. The role of courts in these circumstances is to review, and they are, at least as a general rule, to refrain from deciding the issue themselves. Accordingly, a court applying the reasonableness standard does not ask what decision it would have made in place of that of the administrative decision maker, attempt to ascertain the “range” of possible conclusions that would have been open to the decision maker, conduct a de novo analysis or seek to determine the “correct” solution to the problem.
[Vavilov, at para. 83 (emphasis in original).]
[17] A court cannot redefine the problem before the decision maker to its own liking and then, on those judicially constructed criteria, find it unreasonable. That only 1% of plastics produced enter the environment was not the problem that the GIC was addressing. The RIAS and the Discussion Paper, which I will turn to shortly, tell us that the problem that the GIC was addressing was that this 1% represents 29 kilotonnes (kt) of PMI annually, cumulatively, entering the environment as plastic pollution. We are required to assess the reasonableness of the Order in light of the problem it was designed to address.
[18] The industry respondents’ focus on the 1% is illustrative of the inherent tension between respect for the choices made by the executive branch and the rule of law that Vavilov, at paragraph 83, speaks to. If 1% (or 29 kt) is not enough to make the listing reasonable, at what percentage does it, in fact, become reasonable—10%, 25%, 60%? Courts do not sit in judgment of the science that underlies the decision; rather, on the facts before the decision maker, a court is to ask whether the conclusion was reasonably open to the decision maker given the statute, common law, and evidence.
[19] The respondents also point to the absence of quantitative testing establishing which particular plastics cause what particular harm in support of its argument that the Order over-reaches.
[20] Again, the respondents impose a test or requirement of their own making, one not required by the statute.
[21] Under section 64 of CEPA, a finding that a substance is toxic arises from evidence of potential or actual harm to the environment. In the context of the harm that the GIC was considering, and to which the Order is directed, namely the harm and potential harm of plastic pollution generally, a requirement for quantitative testing to determine the chemical composition of a particular plastic is irrelevant. To put the matter bluntly, and as the Science Assessment and RIAS make clear, the chemical content of PMI is irrelevant to the sea otter choking on a plastic straw. The problem is the plastic item itself, not its chemistry. I also note, parenthetically, that on the evidence that was before the Federal Court, quantitative testing would be impractical and ethically reprehensible. No court should require that a decision maker engage in unethical testing to meet the threshold of reasonableness.
[22] There was overwhelming scientific evidence before the GIC supporting a finding that PMI were ubiquitous in the environment and, to mirror the language of section 64, may be present in such conditions and concentrations that they were, or could be, harmful to the environment or its biological diversity. Further, section 68 of CEPA (reproduced at Annex A) grants the Minister broad discretion to gather and evaluate data to determine whether it is harmful within the meaning of section 64. In urging a different or higher research standard, the respondents, again, substitute their own norms to assess the reasonableness of the Order.
[23] I conclude this overview by turning to the constitutional issue.
[24] The Federal Court found that the dominant purpose of the Order was to prevent harm to the terrestrial and marine environment. I agree. Apart from vague assertions by the Attorneys General of Saskatchewan and Alberta that given the ubiquitous nature of plastic, the Order is a covert attempt to regulate all aspects of the Canadian economy, no incidental effects are pled and no evidence of such was led. There is no challenge to the relevant provisions of CEPA, which is unquestionably valid federal legislation (see Hydro-Québec, at paras. 110, 130). The Federal Court’s conclusion that the Order was unconstitutional flowed from the conclusion that it was an unreasonable exercise of a delegated authority—not that it was an intrusion into a provincial domain.
[25] There is no constitutional issue here. The criminal law power has not been engaged.
[26] The Order is simply an enabling provision. It imposes neither a prohibition nor a sanction. There are no consequences for anyone. The Order opens the door for Ministers to consider potential regulations, which, if enacted, must themselves meet constitutional and administrative law requirements. As the Supreme Court explained, the Order is the first of a two-stage legislative process, one which winnows “from the vast number of substances”
to only those that are ultimately subject to the criminal law. La Forest J. captures the point well, noting that section 64 is simply “a drafting tool”
(Hydro-Québec, at paras. 147, 142). Further, as I will explain later, Ministers may, in fact, decide not to regulate and elect for non-regulatory responses.
[27] Before turning to a more detailed consideration of these issues, a slight digression on the standard of review is required. On an appeal from a judicial review decision of the Federal Court, such as here, this Court must determine whether the Federal Court correctly selected and applied the standard of review (Northern Regional Health Authority v. Horrocks, 2021 SCC 42, at para. 10 [Horrocks]; Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, at paras. 45–46). The administrative decision is considered afresh and we are essentially engaging in a de novo review (Horrocks, at para. 10; Sun v. Canada (Attorney General), 2024 FCA 152, at para. 4). That said, where the respondents rely heavily on the reasons of the applications judge, as they do here, those reasons serve as a useful foil to understanding and illuminating their arguments (Bank of Montreal v. Canada (Attorney General), 2021 FCA 189, at para. 4).
[28] The standard of review to be applied to the judicial review of subordinate legislation is reasonableness, with the criteria in Katz Group Canada Inc. v. Ontario (Health and Long Term Care), 2013 SCC 64 (Katz) continuing to inform the analysis. A finding that delegated legislation is ultra vires no longer requires that it meet the threshold of being “irrelevant,”
“extraneous,”
or “completely unrelated”
to the enabling legislation’s statutory purpose (Auer, at paras. 4, 32, 114).
[29] However, as de Montigny C.J. observed in Canadian Coalition for Firearm Rights v. Canada (Attorney General), 2025 FCA 82 at paragraph 28, “[t]his is not to say that all the principles enunciated in Katz should be discarded.”
The Chief Justice continued, noting:
Justice Côté explicitly stressed that Katz continues to provide ‘valuable guidance’, and that Auer only marks a ‘narrow departure’ from it. More particularly, a reasonableness review of the vires of subordinate legislation should still be informed by the following principles: 1) subordinate legislation must be consistent both with specific provisions of the enabling statute and with its overriding purpose or object; 2) subordinate legislation benefits from a presumption of validity; 3) the challenged subordinate legislation and the enabling statute should be interpreted using a broad and purposive approach to statutory interpretation; and 4) a vires review does not involve assessing the policy merits of the subordinate legislation to determine whether it is necessary, wise, or effective in practice.
[30] The respondents’ arguments cannot be reconciled with these principles, the third and fourth in particular. They hinge, implicitly and at times expressly, on the insertion of the adjective “all”
before “plastic manufactured items”
. The respondents then point to the lack of evidence to support the statement in the Science Assessment that “all”
PMI have the potential to become plastic pollution. They ask of section 90 questions that are answered by section 93. They are impatient.
[31] Questions as to what, when, where and how are answered by regulations under section 93. While not challenged in these proceedings, the Single-use Plastics Prohibition Regulations, S.O.R./2022-138 are instructive. These regulations derive from the Order and its general proposition that plastics can cause harm but supply the answer to the question of which PMI are targeted in exacting detail. As I will explain later (at paras. 113–122), imposing qualifying language on the Order is also inconsistent with long-standing practice in the application of section 90.
[32] Reasonableness, it is to be recalled, is a single standard that accounts for context. Reviewing courts are to examine administrative decisions “in light of the history and context of the proceedings in which they were rendered”
(Vavilov, at para. 94). The contextual component of reasonableness analysis is broad and includes the statute, evidence, submissions, policies, guidelines and past decisions. History and context may reveal that what is an apparent shortcoming in the reasons or decision is no shortcoming at all. Reasonableness is also informed by the degree of flexibility assigned to the decision maker by the statute and the extent that the statute expects the decision maker to apply the purpose and policy underlying the legislation. Here, the legislation sheds considerable light on how Parliament expects the purpose of CEPA to be reflected in decision-making (Vavilov, at paras. 88–94, 97, 110; Mason, at paras. 61, 67, 70).
[33] By way of a roadmap of what follows, I will begin with the question whether “plastic manufactured items”
are a “substance”
within the meaning of paragraph 3(1)(f) of CEPA (reproduced at Annex A). I will then turn to the statutory scheme and, applying the Auer framework, assess whether there was sufficient evidence before the GIC to allow it to reasonably conclude that PMI caused, or had the potential to cause, harm within the meaning of subsection 64(a) of CEPA. I will then address whether subsection 90(1) requires that the Order list individual plastics that are the source of potential harm and the closely related question of whether there is a requirement for quantitative testing before a conclusion can be reached with respect to harm. At the end of these reasons I will address the questions of constitutionality, the Minister’s refusal to establish a BOR and mootness.
II. The facts before the GIC
[34] By way of summary, the RIAS and Science Assessment describe the ubiquitous and persistent nature of plastics in the environment and the varied, sometimes unknown, sources of plastic pollution. These documents explain that the problem targeted by the GIC in listing PMI as toxic was plastic pollution at large and speak to a near universal scientific consensus that macroplastic pollution (>5 millimeters (mm)) is an environmental hazard. The RIAS provides examples of types of macroplastic pollution that harm terrestrial and marine animals, and the Science Assessment describes how the harm is caused, whether through ingestion, suffocation, strangulation, internal hemorrhaging or disease transmission. Macroplastic pollution was also found to affect the integrity of animal habitats and breeding (RIAS, Appeal Book, at 789–790, 792–793; Science Assessment, Appeal Book, at 1049–1053).
The Science Assessment
[35] On October 7, 2020, following a lengthy period of public comment, the Ministers published the final version of the Science Assessment. The Science Assessment was a review of over 600 peer-reviewed scientific publications and “summarize[d] the current state of the science regarding the potential impacts of plastic pollution [defined as plastic that is discarded, disposed of, or abandoned in the environment outside of a managed waste stream] on the environment and human health”
. Its purpose was to “
guide future research and inform decision-making on plastic pollution in Canada”
(Science Assessment, Appeal Book, at 1007).
[36] The Science Assessment built on the 2018 report of the Canadian Council of Ministers of the Environment (CCME). The report recognized the “
exponentially increasing global environmental problem”
of plastic pollution, including its harms to wildlife, habitats, and fisheries, and concluded that it may worsen significantly without further action.
[37] The Science Assessment documented widespread environmental harm, including death of aquatic life and marine mammals from plastic pollution, and concluded that macroplastic pollution causes or potentially causes physical harm to animals, habitats, ecosystems, and plants. The Science Assessment also addressed the broader ecosystem impacts of macroplastic pollution on species not directly affected and human life. Reflecting the findings of the CCME, the Science Assessment determined that these environmental effects would continue to worsen if no mitigation measures were adopted (Science Assessment, Appeal Book, at 1009–1010, 1049).
[38] The Science Assessment explained that the sources and quantities of plastic pollution were vast and varied, and that the slow degradation of plastic magnified its persistence in the environment. It noted that 29 kt of plastic entered the environment in 2016 and that plastic pollution will continue to accumulate. While plastics may degrade in the environment, the process is slow and affected by multiple factors. It recommended that actions be taken to reduce the quantity of both macro and microplastics that end up in the environment (Science Assessment, Appeal Book, at 1007, 1020, 1032, 1049).
The Discussion Paper
[39] The Discussion Paper was published the same day as the Science Assessment. It indicated that listing PMI on Schedule 1 would allow government to “enact regulations that target sources of plastic pollution and change behaviour at key stages in the lifecycle of plastic products…in order to reduce pollution”
. The prospective regulations were proposed to target “
certain [PMI]”
(Discussion Paper, Appeal Book, at 553 (emphasis added)).
[40] The Discussion Paper proposed the development of regulations to manage single-use plastics. It considered which PMI were prevalent in natural and urban environments, known or suspected to cause environmental harm, value-recovery problematic, or should otherwise be exempted. Certain single-use plastics, such as garbage bags, disposable personal care items, and drink cups and lids were not deemed environmentally problematic. The Discussion Paper recommended banning or restricting six single-use plastic items: checkout bags, stir sticks, six-pack rings, cutlery, straws, and foodservice ware (Discussion Paper, Appeal Book, at 559–562).
The Discussion Paper and Science Assessment are relevant to different phases of the CEPA framework
[41] As is apparent, the Science Assessment and the Discussion Paper speak to different stages of the legislative process under CEPA, with the latter being farther downstream in the regulatory process.
[42] The RIAS explains the respective purposes of these publications: the Science Assessment was meant to “summarize the current state of the science regarding the potential impacts of plastic pollution on the environment and human health, as well as to inform future research and decision-making on plastic pollution in Canada”
, while the Discussion Paper was meant to “outline potential risk management measures on
certain [PMI]”
and guide what should be managed and regulated following the listing of PMI as a toxic substance. That is, PMI having been listed, the Discussion Paper reflected the conversation as to what measures, if any, should be enacted to deal with them. (RIAS, Appeal Book, at 792, 795 (emphasis added)).
[43] The Discussion Paper was one piece of a regulatory puzzle; a piece that served a very specific purpose. Specifically, section 68 of CEPA outlines an extensive, non-exhaustive list of measures that the Ministers may utilize in assessing the toxicity of a substance and, once a substance is listed, whether or how to control it. This is an ongoing investigative process: “research, investigation and evaluation”
is conducted to assess “whether a substance is toxic or is capable of becoming toxic,”
and then “whether to control, or the manner in which to control, a substance”
.
[44] The focus of the Discussion Paper was single-use plastics; it notes that “[s]ome single-use plastics that end up in the environment cause harm to ecosystems and wildlife”
and that “[t]he Government of Canada has committed to banning or restricting certain harmful single-use plastics, where warranted and supported by science”
. The Discussion Paper engages in a triage, assessing the environmental harm of specific single-use plastics against other factors, such as whether they can be recycled or recovered, whether they perform essential functions and whether they have alternatives (Discussion Paper, Appeal Book, at 556, 559–562).
[45] CEPA, therefore, contemplates that listing may be informed by different principles and evidence than those germane to the question of whether a substance should be regulated. The reasonableness analysis must be calibrated accordingly.
[46] The Federal Court relied on the “findings in the Discussion Paper indicating that not all PMI are harmful”
to support its conclusion that the Order is overly broad (Federal Court Decision, at para. 117 (emphasis added)). In so doing, the Court misunderstood the purpose of the Discussion Paper within CEPA’s two-stage triage process. If only those PMI that were ultimately triaged for a regulatory response should be listed, listing under section 90 would serve no purpose.
The RIAS
[47] As noted, the RIAS accompanying the Order stated that the Order’s objective was “to add [PMI] to Schedule 1 to CEPA, which enables the [M]inisters to propose risk management measures under CEPA on
certain [PMI] to manage the potential ecological risks associated with
those items becoming plastic pollution”
(RIAS, Appeal Book, at 794 (emphasis added)).
[48] The RIAS refers to the Science Assessment and summarizes research relevant to the Order, including that “[c]urrent scientific evidence confirms that plastic pollution is ubiquitous in the environment, and that macroplastic pollution poses an ecological hazard, including physical harm, to some animals and their habitat,”
and provides examples of how PMI may enter the environment. The RIAS notes that “
[a]ll [PMI] have the
potential to become plastic pollution”
(RIAS, Appeal Book, at 788 (emphasis added)).
[49] The RIAS notes that, unlike macroplastics, harms associated with microplastics (≤5 mm) are less certain. Further research was required and the jury was still out, so to speak, with respect to the harm of microplastics on human and environmental health. Reflecting the RIAS, the Order does not encompass microplastics.
[50] Importantly, for the purpose of the reasonableness analysis, the Science Assessment and the RIAS emphasize that due to the degradation of plastic in the environment, identification by source is often problematic. This fact or characteristic of the problem also informs the reasonableness review.
Public consultation
[51] A period of public consultation followed publication of the proposed Order and RIAS. Sixty notices of objection were filed, 52 of which included requests that the Minister establish a Board of Review. Seventeen civil society organizations, one territorial government, two local governments, and an organization representing municipalities offered support for the proposed Order; 123 industry associations or companies, two provincial governments, and one foreign government expressed opposition to the proposed Order.
[52] In response to the objections the MECC directed that the notices of objection be reviewed by scientists drawn from multiple government departments. This was followed by a second review by expert external and internal scientists with no prior involvement in the matter. Based on this two-stage review, the Minister concluded that the information provided in the notices of objection did not cast doubt on the core findings of the Science Assessment.
[53] Against this background, I turn to the first issue.
III. Whether PMI are a “substance”
within the definition of section 3
[54] The GIC’s determination that PMI are a “substance”
within the meaning of paragraph 3(1)(f) was reasonable having regard to the text, context and purpose of CEPA.
[55] The Federal Court accepted the argument that, as “plastic manufactured items”
is plural and the English version of paragraph 3(1)(f) references “any manufactured item”
in the singular, PMI do not fit under the definition of “substance”
since the term “substance”
itself is singular. This led the Court to conclude that PMI as a category was broader than the definition of “substance”
provided at paragraph 3(1)(f) of CEPA (Federal Court Decision, at para. 80).
[56] I do not agree.
[57] While “
substance”
is defined in the singular in section 3 of the English version of CEPA to include “any manufactured item,”
subsection 33(2) of the Interpretation Act, R.S.C. 1985, c. I-21 provides that “[w]ords in the singular include the plural, and words in the plural include the singular.”
This basic rule of statutory interpretation stipulates that “substance”
includes “substances”
and “item”
includes “items.”
PMI fall within the definition of “substance”
as the singular necessarily encompasses the plural and the plural the singular.
[58] This conclusion is reinforced by the French version of the statute, which speaks in the plural—“
les articles manufacturés”
at paragraph 3(1)(f). The search for the common meaning between the two official language articulations, when guided by the Interpretation Act, leads to the conclusion that PMI readily fall within the ambit of “substance”
(Schreiber v. Canada (Attorney General), [1998] 1 S.C.R. 841, 1998 CanLII 828; Chrysler Canada Ltd. v. Canada (Competition Tribunal), [1992] 2 S.C.R. 394 at p.432 per McLachlin J., dissenting but not on this point).
[59] While the past practices of a decision maker are not determinative of reasonableness, they can be instructive (Canada v. Honey Fashions Ltd., 2020 FCA 64 at para. 38). A review of other substances listed on Schedule 1 supports the interpretation that the plural description of an item necessarily includes a singular iteration of the same substance. Numerous other items are listed in the plural on Schedule 1 including, for example, “plastic microbeads,”
“inorganic fluorides,”
and “inorganic arsenic compounds.”
[60] While this is sufficient to dispense with the matter, I note that this textual reading also aligns with a contextual and purposive understanding of subsection 3(1).
[61] With respect to context, PMI are also “matter”
within the ambit of paragraph 3(1)(a). While “matter”
is not defined in CEPA, its ordinary definition is wide and can include “the substance of which a physical object is composed,”
“a material substance that occupies space and has mass”
and “a material substance of a particular kind or for a particular purpose”
(see Oxford Encyclopedic Dictionary / Merriam Webster definition of “matter”
).
[62] With respect to purpose, the objective of section 90(1) is to enable a threshold survey of numerous, diverse candidate substances for toxicity and potential regulation (Hydro-Québec, at paras. 146–147). To this end, it provides for the review of substances that pre-exist in Canada or are new, are living or inanimate; indeed, it expressly includes “any distinguishable kind of…matter”
(CEPA, s. 3(1)). This aligns with the objective of enabling the detection of substances that may cause harm to health or the environment before it occurs. Given CEPA’s pollution prevention purpose, PMI fits comfortably within the definition of “substance,”
particularly given that the RIAS’ definition of PMI mirrors paragraph 3(1)(f), except for its plurality:
any items made of plastic formed into a specific physical shape or design during manufacture, and have, for their intended use, a function or functions dependent in whole or in part on their shape or design.
[RIAS, Appeal Book, at 788 (emphasis added).]
[63] To conclude, the GIC’s finding that PMI are a “substance”
falls squarely within the relevant constraints of the Act and aligns with the norms of statutory interpretation (Auer, at paras. 37, 39). The narrow, limited and overly technical reading accepted by the Federal Court does not.
IV. Whether the Order is reasonable within the statutory and common law constraints
Overview of the statutory scheme
[64] The primary purpose of CEPA, 1999 is to prevent pollution. This objective is enshrined in both the Act’s preamble and long title (“An Act respecting pollution prevention and the protection of the environment and human health in order to contribute to sustainable development”
), and is embedded throughout its provisions (see, for example, subsection 90(1.1), which requires that the Ministers prioritize pollution prevention actions).
[65] CEPA, 1999’s pollution prevention purpose effected a legislative shift away from managing pollution after it was created, which was the approach under its predecessor statute, the Canadian Environmental Protection Act, R.S.C. 1985, c. 16 (4th Supp.) (CEPA, 1988). CEPA, 1988, the version of the statute in force at the time of Hydro-Québec, itself was also implemented to address the failure of its precursor, the Environmental Contaminants Act, S.C. 1974-75-76, c. 72, to effectively identify toxic substances before they went into use in Canada (Hydro-Québec, at paras. 136, 145).
[66] The 1999 amendments expanded the range of substances canvassed by CEPA and introduced Part 4 “Pollution Prevention,”
which aimed at fortifying CEPA’s ability to pre-empt environmental harm. The precautionary principle was also established as a binding administrative duty under paragraph 2(1)(a) and the “ecosystem approach”
was embedded in a new paragraph of the preamble. The addition of “biological diversity”
and “environment”
in section 64 introduced the assessment of harm to ecosystems and reflected Canada’s increasing obligations under international environmental commitments (see, for example, the 2018 Ocean Plastics Charter, which commits G7 nations to greater plastics stewardship).
[67] Sections 64 and 68 of CEPA are intimately connected, with the latter providing an extensive, non-exhaustive list of tools the Ministers may use to assess the risk posed by a substance against the criteria outlined under section 64. If a substance causes, or may cause, harm within one or more of the categories described in section 64, the Ministers may recommend to the GIC that an order be issued under subsection 90(1) adding it to Schedule 1.
[68] Section 90 is an enabling authority that, as part of the decision-making machinery of government, opens the door for the possibility of enacting subsection 93(1) regulations. Non-regulatory tools are also available, such as codes of practice (ss. 54(1)(d), 55(1)), guidelines (ss. 54(1), 69), or pollution prevention plans (s. 56(1)). To date, only the Single-use Plastics Prohibition Regulations have been enacted in respect of PMI by order of the GIC under subsection 93(1).
The GIC’s discretion
[69] I turn next to the nature of the GIC’s discretion under subsection 90(1) to list a substance on Schedule 1. This is important because the breadth of the discretion afforded to a decision maker by statute frames the application of the reasonableness standard. No matter how broad the discretion, however, the decision must still reflect the hallmarks of reasonableness; justification, transparency and intelligibility (Vavilov, at paras. 88–90, 99, 103, 109–110, 120–121; Mason, at para. 60).
[70] In reviewing the vires of subordinate legislation, however, the Court does not interpret the governing statute de novo; “[s]tatutory delegates are empowered to interpret the scope of their authority”
and a reviewing court’s role is to ensure that the exercise of authority falls within some reasonable interpretation of the enabling statute (Auer, at paras. 60–65, citing John Mark Keyes, Executive Legislation, 3rd ed. (Toronto: LexisNexis, 2021), at 175 [Keyes]; Vavilov, at paras. 101, 105, 106, 108).
[71] An Order under section 90 may be made where the GIC “is satisfied”
that one or more of the criteria under section 64 are met. The language of “is satisfied”
is subjective: it requires that the GIC itself decide whether a substance is toxic prior to acting. As discussed by John Mark Keyes, subjective language in an enabling provision “
augments the scope
of executive legislative authority”
(Keyes, at 427–428; see also Paul Salembier, Regulatory Law and Practice, 3rd ed. (Toronto: LexisNexis, 2021) at 119–123). Even though the law reads the requirement of reasonableness into the language (i.e. the GIC must be “reasonably satisfied”
), the text reinforces the discretionary nature of cabinet-level decisions.
[72] The words “is satisfied”
are also a direction from Parliament that perfection is not the standard. To be satisfied does not mean to be certain.
[73] CEPA provides other clues as to how the GIC’s discretion is to be exercised.
[74] The precautionary principle is a mandatory consideration in the GIC’s administration of CEPA, 1999 (para. 2(1)(a)), along with the duty to “act expeditiously and diligently to assess [substances]”
(para. 2(1)(k)). Paragraph 2(1)(a) stipulates that full scientific certainty shall not be a reason to postpone cost-effective measures to prevent environmental degradation where there are threats of serious or irreversible damage. The addition of the word “may”
throughout section 64 in the 1999 amendments to CEPA infused the precautionary principle into the scheme’s framework for controlling toxic substances, underscoring CEPA, 1999’s risk-based approach to thwarting environmental harm before it occurs.
[75] The respondents argue that the GIC has discretion only if it has been determined by “objectively reasonable standards”
beforehand that a substance is harmful. The matter could not go to the GIC unless there was a threshold decision by someone, presumably by the Minister, that PMI have the potential to cause harm.
[76] There is no merit in this argument. It contradicts the plain text of subsection 90(1) and introduces redundancies.
[77] If accepted, the word “satisfied”
would be redundant—there is nothing for the GIC to be “satisfied”
about as, under the respondents’ theory, it is the Minister (or someone else not identified) that must be satisfied. The words “satisfied,”
“substance,”
and “toxic”
appear in the same clause in relation to the decision-making authority of the GIC, strongly militating against separating these concepts. The section places the discretion with the GIC, not elsewhere, and it is the exercise of that discretion that the law requires be assessed against the standards of reasonableness. I reject the argument that the discretionary decision-making authority can be cleaved away from the GIC. To do so would be to ignore the crystal-clear language of the statute.
[78] The respondents do not confront the language of subsection 90(1); rather, they shift the debate away from the language of the statute by asking whether the evidence could have been stronger, whether laboratory tests should have been conducted or whether only those items that might ultimately be regulated should have been listed in the Order. These are the wrong questions. The question that should have been asked was whether, considering the evidence, the GIC could have reasonably been satisfied of the straightforward question posed by Parliament under section 64; namely, whether PMI have the potential to cause harm to the environment.
The decision to list and the decision to regulate are discrete steps
[79] The effect of the addition of a substance to Schedule 1 by subsection 90(1) is to enable the Ministers to further consider how to control the substance, including whether to enact regulations or non-regulatory responses. Determining “appropriate preventative or control actions to manage the risk posed by a substance is a separate and distinct function from the assessment of whether a substance is toxic”
(Goodyear Canada Inc. v. Canada (Environment), 2017 FCA 149, at para. 42 [Goodyear]).
[80] As I have explained, the GIC’s decision to list a substance on Schedule 1 is distinct from the decision to regulate it: different considerations underpin each decision, necessarily contextualizing the reasonableness analysis. The decision to list a substance under subsection 90(1) flows from the GIC’s determination that it is toxic within the meaning of section 64, posing a risk of harm to human and/or environmental health. A subsequent decision to regulate a substance under section 93 is grounded in further investigation into how to manage the substance’s identified risks, and involves the balancing of policy-based considerations, such as the existence of cost-effective alternatives. This process is played-out in the Discussion Paper with respect to the Single-use Plastics Prohibition Regulations.
[81] Again, CEPA provides for non-regulatory tools that may be implemented to manage listed substances, such as codes of practice (ss. 54(1)(d), 55(1)), guidelines (ss. 54(1), 69), or pollution prevention plans (s. 56(1)). The Act also contemplates that ongoing research may reveal that a substance that was listed as toxic no longer meets the criteria for listing and should be deleted (s. 90(2)). For example, this provision was used to remove BNST from Schedule 1 by order published, with a RIAS, in the Canada Gazette on October 6, 2020.
[82] The respondents’ core argument is that the Order is an unreasonable exercise of the GIC’s authority because there is insufficient evidence to support the conclusion that “all”
PMI cause harm. This argument fails, and for several reasons.
[83] The argument ignores the express language of section 64, which speaks of substances that “may enter”
or have the potential to enter the environment.
[84] Second, the argument collapses the two-stage decision-making process into a single step. Reduced to its core, the effect of their argument is that only those PMI that, in fact, are subsequently regulated can be listed. Imposing a requirement that all manifestations of PMI cause harm as a prerequisite to listing is nowhere to be found in the statutory language and inconsistent with CEPA’s two-stage process. The respondents, in effect, ask the Court to rewrite the statute.
[85] Third, this argument, if accepted, would also constitute a substantial departure from how Schedule 1 has been conceived and utilized in the past. A court should be cautious about deviating from how highly technical statutes related to public health and safety have been understood and applied in the past.
[86] The argument also conflicts with the Supreme Court’s guidance with respect to how CEPA operates.
[87] Hydro-Québec found that section 11 (now section 64 under CEPA, 1999) outlines the types of risk targeted in subsequent assessments, “weed[ing] out from the vast number of substances potentially harmful to the environment or human life those that pose significant risks of that type of harm”
(Hydro-Québec, at para. 147). Of note, La Forest J., writing for the majority, described this provision as “
simply a drafting tool,”
which is not operative in and of itself, but which, when combined with section 68, creates a gateway to the process by which CEPA progressively whittles away the number of substances that could be “candidates for regulations”
(Hydro-Québec, at paras. 141–142).
[88] When considered in light of section 93, the GIC’s listing under subsection 90(1) without proof that every manifestation of the substance is toxic in all circumstances, falls within a reasonable interpretation of the GIC’s authority under that provision. This makes sense. As I will explain later, there are many examples of substances that are listed as toxic, but which have safe uses or permissible concentration levels. Those limits may be fixed by regulations enacted under subsection 93(1).
[89] The scope of the Order also aligns with the Science Assessment. The evidence before the GIC was that PMI, as a substance, when found in the environment in certain conditions and quantities (the language of section 64) have the potential to cause harm to the environment and biodiversity, and that certain PMI cause more direct harm. The Order, situated and understood in light of the interaction between sections 64, 68, 90 and 93, simply states the uncontroverted—that PMI could be harmful.
[90] In this vein, the RIAS notes that “[PMI]
can enter the environment as plastic pollution through a wide range of activities,”
including littering, environmental emergencies, wear-and-tear of certain items, or accidental release (RIAS, Appeal Book, at 790 (emphasis added)). The Science Assessment reflects these findings, noting that there are multiple sources of plastic pollution. It is no large leap for the GIC to infer from this that all PMI may indeed become plastic pollution—an inference that can be made, as a matter of deference, in conducting a reasonableness review.
[91] The text of the Order aligns with the purpose of CEPA. The RIAS states that the objective of adding PMI to CEPA’s List of Toxic Substances is to allow the Ministers to “propose risk management measures under CEPA on
certain [PMI] to manage the potential ecological risks associated with
those items becoming plastic pollution”
(RIAS, Appeal Book, at 794 (emphasis added)). The Discussion Paper speaks of “using…CEPA to regulate
certain [PMI]”
and to “
target sources of plastic pollution”
(Discussion Paper, Appeal Book, at 553 (emphasis added)).
[92] Courts must respect Parliament’s choice of legislative design; in this case, a broad enabling authority followed by a tailored prohibition. CEPA asks two separate and discrete questions. The first asks whether the substance causes or has the potential to cause harm. If it does, that leads to the second question as to what should be done about it. This unique legislative framework underpins the reasonableness inquiry into the GIC’s answers to those questions.
[93] It bears repeating that the respondents’ arguments compress the two steps of the regulatory process of CEPA into one – namely, the only manifestations of PMI that can be listed in an order are those that will, in fact, be regulated.
[94] The CEPA scheme does not work that way, never has, and for good reason.
[95] First, it would be absurd to require the Order to prejudge the outcome of the very downstream legislative processes that it contemplates and initiates. The mandatory public consultations associated with publication of regulations in Parts I and II of the Canada Gazette would be redundant. As the Discussion Paper illustrates, it is not until the consultation process is over that the GIC can, in fact, decide what will be regulated and how.
[96] Second, the two-stage nature of the scheme aligns with government’s duties under the Act. The precautionary principle provides that a lack of scientific certainty should not postpone the development of measures to prevent environmental degradation, and paragraph 2(1)(k) directs that substances are assessed expeditiously and diligently. A broad listing followed by narrow regulations allows for an early launch of discussions, both within and outside of government, about the nature and extent of the problem and what, if anything, to do about it. The Supreme Court has already told us that “broad wording is unavoidable in environmental protection legislation because of the breadth and complexity of the subject and has to be kept in mind in interpreting the relevant legislation”
(Hydro-Québec, at para.134, citing Ontario v. Canadian Pacific, [1995] 2 S.C.R. 1031, 1995 CanLII 112 (SCC), at para. 43).
Harm and toxicity under section 64
[97] The central theme of the respondents’ attack on the reasonableness of the Order is that PMI, alone, are not toxic: the plastic fork or spoon used in the food court, the steering wheel on the car that we hold each day, the contact lenses that we put in our eyes and the computer screens that we look at, absent further studies, are not toxic. In support, the respondents place considerable emphasis on several paragraphs in Hydro-Québec where La Forest J. speaks of CEPA capturing substances that are “toxic in a real sense”
and “poisonous”
; reprising La Forest J., they state that PMI are not “akin”
to poisonous substances (Hydro-Québec, at paras. 141, 144–145).
[98] Beguiling as the argument may be, I do not agree. It is based on a misreading of the statute and Hydro-Québec.
[99] “Toxicity”
is a consequence of the GIC being “satisfied”
that a substance may cause harm within the meaning of section 64. If such “harm”
exists, the result is a finding that a substance is “toxic.”
The respondents argue the reverse, turning the statute on its head. They say, “prove that the substance is toxic and, if so, it might cause harm.”
[100] This argument directly conflicts with the text of section 64, which provides that a substance is “toxic”
if it causes or may cause “harm.”
Harm and toxicity are two different concepts—harm is a science-based finding, toxic is the label given to a harmful or potentially harmful substance. The ordinary meaning of “toxic”
is contextualized by section 64. While substances that are “toxic”
in their ordinary sense will likely cause harm, substances that are not toxic in their ordinary sense may cause harm under certain conditions.
[101] The respondents also latched onto the Supreme Court’s comment that substances that will be listed are likely akin to those already listed on Schedule 1. The Federal Court agreed that PMI is broader than the existing substances on Schedule 1, but found it insufficient on its own to render the Order unreasonable (Federal Court Decision, at para. 80). I agree.
[102] The plain text of CEPA is sufficient to dispose of this argument, but there are further problems with the respondents’ position: it is predicated on a misreading of La Forest J.’s language and does not consider the amendments to CEPA, 1999, the version of the statute with which we are concerned. Three brief points may be made.
[103] First, in Hydro-Québec, which was decided when CEPA, 1988 was in force, the Court determined that “toxic”
took its meaning from section 11, the contemporary to CEPA, 1999’s section 64 (at paras. 144–145). Given that Schedule 1, at its inception, comprised only nine substances, all of which were chemicals or elements, including chlorobiphenyls which were the subject of the Court’s consideration, La Forest J. suggested “poisonous”
as a dictionary definition for “toxic”
(Hydro-Québec, at para. 141). However, and importantly, he noted that section 11 (now section 64) guides risk assessments because the more nuanced effects of substances that are not toxic in the ordinary sense of the word, are generally not known (Hydro-Québec, at para. 141).
[104] Second, Hydro-Québec did not create a new test for toxicity under CEPA. Nor was the Court reading a limitation into the Act—that harm must arise from “poisoning.”
[105] Section 64 is agnostic as to how harm is caused, whether through poisoning, suffocation, strangulation, habitat destruction, disease transmission or internal hemorrhaging, these being the vectors of harm in respect of biodiversity and the environment in issue. Indeed, it is common knowledge that many of the substances on Schedule 1 are not “poisonous”
in the ordinary sense but may cause harm because of their emergent hazardous properties. La Forest J. chose lead as an example, noting that “[l]ead…is not
per se toxic but it can be so when it enters the environment in the course of its use”
(Hydro-Québec, at para. 141). Carbon dioxide, to choose another, but contemporary, example, poses a threat to the viability of the environment and life on the planet through its contribution to global warming; at the same time it is essential to human existence.
[106] At the time PMI were listed on Schedule 1, it contained 163 substances, including non-chemical entities such as “plastic microbeads,”
“particulate matter,”
“ceramic fibre,”
and “wastewater effluent,”
as well as substances with long-term, emergent dangerous properties, such as the greenhouse gases. None of these substances would have been “akin”
to those already listed if the respondents’ narrow interpretation is accepted.
[107] It can therefore be appreciated how the list of substances on Schedule 1as it stood in 1997 informed La Forest J.’s choice of the word “poisonous.”
Nor should the facts of the case before the Supreme Court be ignored – there was no dispute that PCBs were “poisonous”
in their ordinary sense. But that label is inconsequential to the actual conclusions reached by the Supreme Court; that “toxic”
takes its meaning from the interaction between sections 64 and 68. Importantly, as the scientific community’s understanding of the sources of threats to the environment has evolved, so too has Parliament’s. At the time, CEPA, 1988 did not include the Act’s present ecological approach or the precautionary principle, did not mention “biological diversity”
and referred to “life”
only as it related to “human.”
[108] The removal of the word “human”
qualifying “life”
and the addition of “biological diversity”
to section 64 in CEPA, 1999, support a broader, more ecologically comprehensive understanding of harm. Taken together, these factors reflect a legislative intent favouring a meaning of “harmful or dangerous to life”
for “toxic,”
which aligns with CEPA, 1999’s pollution prevention purpose (see Oxford English Dictionary definition of “toxic”
). This point was made with clarity by McLachlin, C.J.C. in Reference re Assisted Human Reproduction Act, 2010 SCC 61 (AHR) where she said:
[t]he complexity of modern problems often requires a nuanced scheme consisting of a mixture of absolute prohibitions, selective prohibitions based on regulations, and supporting administrative provisions. Such schemes permit flexibility, vital in a field of evolving technologies, and they have repeatedly been upheld as valid criminal law: RJR-MacDonald; R. v. Hydro-Québec, 1997 CanLII 318 (SCC), [1997] 3 S.C.R. 213. To take but one example, the list of toxic substances capable of harming the populace is ever-changing. It is unrealistic to expect Parliament to enact new laws every time a change occurs, and the criminal law power does not require it to do so.
[at para. 36 (emphasis added).]
[109] The Supreme Court recently returned to this point in Telus Communications Inc. v. Federation of Canadian Municipalities, 2025 SCC 15 (Telus). Writing for the majority, Moreau J. noted that, while statutory interpretation is centered on the intent of the legislature at the time of enactment, this does not prevent courts from applying statutes to new or evolving circumstances. To the contrary, Moreau J. observed:
[i]t is uncontroversial that, in the exercise of their legislative authority, enacting legislatures can use broad or open-textured language to cover circumstances that are neither in existence nor in their contemplation… Indeed, they frequently do so to ensure the long-term objects of an enactment can be achieved without constantly reopening the statute. (at paragraph 33)
[110] This guidance resonates loudly in the environmental, public health and safety contexts. The concepts under consideration in this case—those of “substance”
and “harm”
—are “broad, or open-textured language”
or concepts, and indicate a legislative intention that the provision “be interpreted dynamically, in that the provision should be capable of applying to new sociological or technological circumstances as they arise”
(Telus, at para. 34). I, therefore, reject an understanding of those terms as being limited to the types of substances listed in CEPA as it stood in 1988.
The GIC’s past practices
[111] Vavilov tells us that past practices may shed light on a decision maker’s understanding of their enabling statute and inform the reasonableness inquiry (Vavilov at para. 106). This is particularly so where the statute establishes a regulatory scheme that is dependent on specialized expertise. However, there are limits to the extent to which past practices can inform the reasonableness of a decision. A long history of unreasonable decisions will not save an unreasonable decision.
[112] A review of Schedule 1 demonstrates that there are substances listed as toxic, but which are entirely unregulated and managed through other means, such as guidelines or codes of practice, that there are some substances regulated in part, substances which are sanctioned for certain applications only and substances which are banned essentially in their entirety.
[113] The GIC’s past practices in listing toxic substances is consistent with the understanding of the statutory scheme described earlier. Listings can be both broad and narrow, reflecting the risk of harm identified and anticipated management options. As the appellants point out, many substances are listed that are not harmful in all their applications or under all conditions, such as lead, carbon dioxide, asbestos, and selenium. No carve-outs are made directly in these listings; their harmful aspects are dealt with by subsequent regulatory or non-regulatory responses.
[114] At the listing stage there is no legislative onus on the GIC to prove that each individual PMI is harmful, any more than there is an onus on the GIC to prove that all lead, all selenium, all carbon dioxide, or all nitrous oxide is harmful. Selenium is available as a dietary supplement, we breathe carbon dioxide every moment of our lives and its solid form, dry ice, is used in shipping temperature-sensitive products, and nitrous oxide (laughing gas) is a dental anesthetic. But they are all listed outright on Schedule 1.
[115] To take another example, asbestos is listed outright, but regulations make exceptions to its use, such as in museum displays or scientific research (Prohibition of Asbestos and Products Containing Asbestos Regulations, S.O.R./2018-196, ss. 12–13). Mercury is listed outright, but regulations set out the maximum total quantity of mercury allowed in certain products on a product-by-product basis (Schedule to the Products Containing Mercury Regulations, S.O.R./2014-254). “Plastic microbeads that are ≤5 mm in size”
are listed outright on Schedule 1, but only their presence in toiletries is subject to regulation; even in this context exceptions are made for certain manifestations, such as prescription drugs (Microbeads in Toiletries Regulations, S.O.R./2017-111). This narrowing or winnowing, based on further assessments of risk at the latter regulatory stage, is precisely what CEPA contemplates.
[116] The Federal Court dismissed this argument since “
all of these examples [lead and carbon dioxide] are of different forms of the same substance; the breadth does not engage a large group of disparate items like PMI”
(Federal Court Decision, at para. 109).
[117] Several points may be made about this.
[118] First, the reasoning is inconsistent with the evidence outlined in the Science Assessment and the RIAS. Both of these documents identify unifying characteristics, of PMI, characteristics that are relevant to the exercise of discretion under section 90; they are persistent and ubiquitous in the environment and their provenance is often difficult to determine. The fact that plastics come in a multitude of forms and shapes was part of the problem under consideration by the GIC. The GIC was not considering whether individual plastic items should be regulated.
[119] Second, as I have described above, a decision maker’s past practice can inform our understanding of reasonableness, particularly in a highly technical domain. As explained, nothing in the prior use of Schedule 1 requires different forms of the same substance be listed individually.
[120] Third, the Federal Court’s reasoning rails against established guidance on statutory interpretation, and I repeat what McLachlin C.J.C. said in AHR at paragraph 36, and what Moreau J said in Telus at paragraph 33. I cannot think of a better way to thwart Parliament’s intent than to require an order under section 90 in respect of each, singular iteration, type or use of PMI. Where text and context allow, courts should adopt interpretations that support the statutory objective, not defeat it.
[121] The Single-use Plastics Prohibition Regulations illustrate how a subset of PMI may be extracted for tailored regulation. Regulations were put in place targeting six specific categories of single-use plastic items: “single-use plastic foodservice ware”
is defined as PMI with plastic that “contains expanded polystyrene foam, extruded polystyrene foam, polyvinyl chloride…”
. As the Discussion Paper explains, these six items were distilled from the broader class of PMI as those where the identified environmental risks demanded the implementation of control measures.
V. Factual constraints
[122] The principal challenge to the reasonableness of the Order is that there is no evidence that all PMI cause harm. I will not revisit why an order under subsection 90(1) does not, as a matter of law, require that this be so. The reasonableness analysis is framed by its statutory context, and this means that an order under section 90 must be assessed in light of its role as an enabling authority.
[123] There are, however, three other elements of the arguments before us that warrant consideration.
[124] The Federal Court found the RIAS’ statement that “all [PMI] have the potential to become plastic pollution”
to be a “peremptory conclusion [that] will rarely assist a reviewing court”
(at paras. 110–111). The Federal Court determined that “the evidence available to the GIC did not support the finding that all PMI are toxic”
(at para. 118).
[125] The respondents reinforce this argument by pointing to the absence of quantitative testing, laboratory studies or chemical analyses to support the listing of all PMI. They highlight the observation of the Federal Court that there was no evidence of harm associated with a particular chemical composition or structure or of the “doses”
or “concentrations”
of PMI that give rise to harm (Federal Court Decision at paras. 113–115). The respondents also dismiss the Science Assessment on the basis that it was “a literature review,”
and emphasize the statement in the RIAS that the Order was not based on a traditional risk assessment. The respondents point out that only “a small subset of PMI”
have been proven to cause actual harm and that therefore the Order is overbroad.
[126] While the “quantitative testing”
and the “RIAS argument”
are inter-related, I will deal with them separately.
The 1% problem
[127] As previously mentioned, while the problem before the GIC was that of plastic pollution generally, the respondents test the reasonableness of the Order by asking a different question; whether the listing of PMI was reasonable given that only 1% of all plastics enter the environment. This is an error in reasonableness review methodology as it invites us to redefine the problem under consideration by the GIC. It is precisely reasoning of this nature that Vavilov proscribes (at para. 83).
[128] The RIAS and Science Assessment make clear that it was not the 1% versus the 99% that was the problem—the problem was what the 1% represented; over 29 kt of plastic items, annually, cumulatively, entering the marine and terrestrial environments. The RIAS itself states that the Order was motivated by the quantity of plastic with the potential to enter the environment and the conditions under which it may enter the environment; quantity and conditions being the legislative triggers in section 64.
The Science Assessment and the absence of quantitative testing
[129] In their oral and written submissions, the respondents characterized the Science Assessment as a literature review that did not support the Order.
[130] I do not agree that the Science Assessment can be discounted or disparaged for this reason, or by an absence of quantitative testing.
[131] The Science Assessment was clearly more than a literature review; it engaged with the substantive question before the GIC, whether PMI in the form of plastic pollution caused harm. As the MECC explained:
In addition, the draft Science Assessment was peer-reviewed, both internally within the Government of Canada and externally by leading experts in the field. The internal review component involved a sequential process beginning with experts internal to [ECCC] and Health Canada and expanding to other government departments (Natural Resources Canada, National Research Council, and Fisheries and Oceans Canada). The external review component involved six peer reviewers. These reviewers provided expert comments and input into the report. Reviewers were chosen because of their known expertise in plastic pollution. The draft Science Assessment was also subject to a 90-day public comment period.
[Response Letters to Notices of Objection; e.g. to Vinyl Institute of Canada]
[132] I note that even if the Science Assessment was merely a literature review, it does not affect the reasonableness of the decision. If decision makers reasonably find that there is scientific consensus with respect to a problem, they may rely on that consensus. Reasonableness does not require that they reinvent the wheel and conduct further testing to arrive at what is likely the same result.
[133] Fairly read in light of the topic under consideration—that of the effects of macroplastic pollution on the environment—the caution that the Science Assessment was “not intended as a substitute for a chemical risk assessment”
means only that there may be particular plastics that merit, by reason of their own particular composition, further scientific investigation or other unique listings or risk management. Indeed, CEPA contemplates this; some substances may be carved out of section 64 for being “inherently toxic.”
This caveat, to the extent that it is noted in brief in the Science Assessment, does not undermine the reasonableness of the Order.
[134] It makes good sense that where the harm targeted is not a chemical substance, but rather a gaseous or a manufactured substance, it will require different testing methods than those traditionally employed for chemical substances. In fact, the GIC recently relied on a literature review to list plastic microbeads (Federal Court Decision, at para. 92). Determinations of the safe thresholds for consumption and the chemical characteristics of individual plastic products are simply irrelevant to the harm to which the Order was directed.
[135] The respondents enthusiastically contend that assessing toxicity requires a “rigorous scientific assessment,”
highlighting that there was no evidence of “a chemical assessment of toxicity…looking at both hazard and exposure”
(industry respondents’ memorandum, at para. 18). They draw on the language of the Federal Court which concluded that the Order was “devoid”
of consideration of the particular chemical composition or structure, or of the “dose”
or “concentration”
of PMI that leads to harm:
The basic principle of toxicity for chemicals is that all chemical substances have the potential to be toxic; however, for a chemical substance to be toxic it must be administered to an organism or enter the environment at a rate (or dose) that causes a high enough concentration to trigger a harmful effect.
[Federal Court decision, at paras. 113 (emphasis added)]
[136] The legal error in this reasoning is patent; toxicity is not the test; it is the consequence of a finding of harm or prospective harm under section 64. The evidentiary error is also patent; the Science Assessment and RIAS describe the nature of the environmental harm posed by PMI: death by strangulation, suffocation, internal hemorrhaging, disease transmission and habitat destruction.
[137] Additionally, a requirement for the traditional quantitative testing used for assessing chemicals is inconsistent with the evidence of Dr. Chelsea Rochman, a recognized expert on the environmental and health effects of plastic pollution. Dr. Rochman opined that the quantitative risk assessment methods used for chemicals cannot be used for macroplastics; they are neither practical nor ethical and, in any event, are unnecessary to understand the harm occurring.
[138] This explanatory background evidence from Dr. Rochman was not before the GIC and is ultimately unnecessary to reach the conclusion that the Order was reasonable. It was, however, before the Federal Court and further frames the nature of the problem. As stated earlier, no court should require unethical testing in order to find a decision reasonable.
[139] The respondents’ requirement for traditional laboratory testing essentially asks the Court to second-guess and reject the science that underlies the GIC’s decision. Not all science takes place in a laboratory. Much research takes place in the field, particularly with respect to marine and environmental science.
[140] CEPA provides a suite of tools that the Ministers may rely on in assessing a substance for its capacity to cause harm. Section 68 gives the Ministers “broad powers”
to collect and generate data and conduct investigations relating to the features or effects associated with a substance (Hydro-Québec, at para. 143). It provides a list of approaches that may be used to generate and analyze data for the purposes of assessing toxicity and, to the extent that they are described, these methods are diverse and extensive. Section 68 does not limit the Ministers to any particular methodology or technique; “investigations”
are not required to be qualitative or quantitative (see e.g. s. 68(a)); and the provision is explicitly not self-limiting, stating that these options are available “without limiting the generality of the foregoing”
(ss. 68(a)).
[141] Neither sections 64 nor 90 impose a requirement for an analysis of the chemical content of individual plastics for listing. In advancing this argument, the respondents invite this Court to read into CEPA a different test than that which is set out in the statute.
The RIAS is not supported by the Science Assessment
[142] I turn next to the crux of the respondents’ position. The respondents argue that there was no evidentiary foundation for the conclusion that all PMI cause or might cause harm. They point out that there is no express statement in the Science Assessment that “all PMI have the potential to become plastic pollution”
or that “all PMI are toxic”
(industry respondents’ memorandum, at paras. 42, 45–48). They argue that there is a break or rupture in the evidentiary chain between the Science Assessment and the RIAS that renders the Order unsupportable.
[143] I would dismiss this argument for three reasons.
[144] First, as I have explained, this argument proceeds on a false premise. The Order does not state that all plastics may cause harm. It says that PMI do or may cause harm. The Order is an administrative precursor to regulations under section 93; the latter tells us which PMI do, in fact, require regulatory sanction. “Plastic manufactured items”
is a statement, not a list.
[145] Second, I flag an important point about language, and the need for absolute precision in a case such as this. “
PMI”
are not “
all plastics.”
PMI is, itself, a defined term; it comprises PMI that are larger than 5 mm. Plastic microbeads are excluded, based as they are on their own discrete listing and regulations. Given their conclusion that further study was required for microplastics, the authors of the Science Assessment would probably agree that the statement “
all plastics are toxic”
is too broad and ask that the reader be precise and recognize that its conclusions on harm are in respect of macroplastics.
[146] Third, Vavilov instructs that a decision will be unreasonable if the reasoning process is not rational or logical. In particular, “a decision will be unreasonable if the reasons for it, read holistically, fail to reveal a rational chain of analysis or if they reveal that the decision was based on an irrational chain of analysis”
(Vavilov, at para. 103). The test is whether there is a “fundamental gap”
in the GIC’s reasoning such that the Order fails against the criteria of transparency, justification and intelligibility (Vavilov, at para. 99).
[147] A careful reading of the Science Assessment and the RIAS, undertaken with a view to understanding, reveals no fundamental gap. The scope of the Order and the reasons for it are self-evident when assessed in light of the problem at hand, namely plastic pollution comprised of undifferentiated and often untraceable sources of PMI in the environment.
[148] The Science Assessment makes clear that the problem of plastic pollution poses unique challenges; it is ubiquitous in the environment, has multiple and diverse entry points and changes shape over time, rendering identification by source problematic. Read in the context of those findings, the absence of an express statement in the Science Assessment that “
all PMI”
have the potential to become plastic pollution is at best a technical or editorial deficiency insufficient to render the Order unreasonable.
[149] In light of this, finding the Order unreasonable on the basis that there is no express statement that “
all PMI”
have the potential to become plastic pollution in the Science Assessment is a classic example of a court engaging in a “line-by-line treasure hunt for error.”
[150] That said, I agree with the respondents that the GIC could have drafted the Order differently so as to narrow its scope. The Order could, for example, have been limited to “PMI used in commercial fisheries”
or “PMI used in the food service industry”
or “PMI used in agriculture”
. The fact that the GIC chose to do it differently, through a broad generic order with the distillations under section 93 to follow, does not mean that the choice was unreasonable.
[151] To repeat, the reasonableness of an order under subsection 90(1) cannot be unhooked from section 93. This Court has characterized decisions such as these as being “like nesting dolls,”
and a context-driven reasonableness inquiry must take account of what came before and what comes after the decision in issue (Canadian National Railway Company v. Halton (Regional Municipality), 2024 FCA 160, at paras. 70, 99).
[152] The final word as to why the respondents’ argument fails is best left to the Supreme Court itself, which said that “[t]he effect of requiring greater precision would be to frustrate the legislature in its attempt to protect the public against the dangers flowing from pollution”
(Hydro-Québec, para. 134). Limiting the reasonableness of the Order to PMI that are ultimately subject to regulatory control is to impose a legal constraint inconsistent with the legislative scheme.
VI. The constitutional question
[153] I begin by delineating what is not in issue in the constitutional challenge to the Order.
[154] The legislative competence of Parliament to prevent environmental harm under section 64 is not challenged, nor is the constitutionality of any of CEPA’s provisions. Whether certain PMI cause environmental harm and can be regulated under section 93 is not an issue. The only question asked by the Notice of Constitutional Question is whether the Order under subsection 90(1) was too broadly drafted to fall within the scope of the criminal law power. Thus, the issue is not whether the GIC can do what it did, but whether it did it the right way.
[155] The Federal Court found, and I agree, that the dominant purpose of the Order is to list PMI to enable regulations to manage the environmental harm associated with plastic pollution. When situated in the context of the enabling statute, the Order’s purpose is environmental protection. However, the Federal Court also reasoned that the Order “
threatened the balance of federalism”
as it enabled the potential regulation of PMI that may not cause harm to the environment (Federal Court Decision, at paras. 165, 166, 175, 184–185).
[156] As is apparent by now, this reasoning is based on a misunderstanding of the legislative scheme. Symmetry between the Order and what is regulated is not required. A regulated substance must fall within the scope of the Order, but not all substances or all their manifestations must be regulated.
[157] Nor does the Order give the GIC the constitutionally undisciplined authority that the Federal Court assumed. A law is not ultra vires simply because it authorizes a discretion that might be exercised unconstitutionally.
[158] Any regulations made under the authority of the Order must, in and of themselves, conform to administrative law principles, including that they are a reasonable exercise of discretion in accordance with the statutory purposes of the enabling legislation. Any regulations must also conform to constitutional law limits (see References re Greenhouse Gas Pollution Pricing Act, 2021 SCC 11, at para. 87 [GGR]; Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038, 1989 CanLII 92 (SCC), at 1076–1080 [Slaight]).
[159] It bears repeating that the Order imposes neither sanction nor penalty and, as such, is not an exercise of the criminal law power. It is the first of a two-stage legislative process by which harmful substances may be brought under regulatory control (see paragraphs 25, 87 above). It is only at the second stage, when regulations are put in place, that the criminal law power is engaged. The only issue before us is the constitutionality of the Order.
[160] The guidance of the Supreme Court requires that legislation granting discretion be assessed based on what it does, not on what it might do. Legislation, particularly legislation directed to the protection of human health or the environment, should not be pre-emptively struck down because it is possible of unreasonable or unconstitutional application. The GIC must be given the chance to interpret and act upon its regulatory powers.
[161] No assumption can be made that a regulatory power will be exercised unconstitutionally. As Jamal J. noted in Reference re Impact Assessment Act, 2023 SCC 23 (IAA Reference), a cardinal principle of constitutional law requires courts to resist a finding of unconstitutionality “simply because [legislation] could conceivably be misused”
(at paras. 230–231). Continuing, Jamal J. points out that “
treating broad grants of statutory discretion as unconstitutional based on the text’s furthest reaches, without regard for constitutional or administrative law constraints, would render ultra vires
many provincial and federal statutes currently in force”
(IAA Reference, at para. 230). I note, parenthetically, that while Jamal J. was in the minority in the result, the Court did not disagree about the principle, only its application in the circumstances.
[162] This is precisely the error in the Federal Court’s reasoning. The fact that the GIC could in theory at some future date, exercise its discretion to sanction objects beyond the scope of the criminal law power does not invalidate the Order.
[163] The existence of executive discretion, on its own, is not unconstitutional (GGR, at paras. 84–86). This is not a new point, nor is it unique to the division of powers; it has been applied in the context of the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c. 11 [Charter] as well. A statutory provision that grants discretion cannot be interpreted in a manner to infringe the Charter unless such infringement is mandated by Parliament. The comments of Lamer J. (as he then was) in Slaight are apposite (at 1078):
[…] As the Constitution is the supreme law of Canada and any law that is inconsistent with its provisions is, to the extent of the inconsistency, of no force or effect, it is impossible to interpret legislation conferring discretion as conferring a power to infringe the Charter, unless, of course, that power is expressly conferred or necessarily implied. Such an interpretation would require us to declare the legislation to be of no force or effect, unless it could be justified under s. 1. Although this Court must not add anything to legislation or delete anything from it in order to make it consistent with the Charter, there is no doubt in my mind that it should also not interpret legislation that is open to more than one interpretation so as to make it inconsistent with the Charter and hence of no force or effect. Legislation conferring an imprecise discretion must therefore be interpreted as not allowing the Charter rights to be infringed.
[164] I do not agree, therefore, with the argument that the litmus test for constitutionality is that it must be impossible to exercise discretion in an unconstitutional manner. A statutory grant of discretionary power should be read on the basis that it will be exercised in a constitutional way, unless the statutory power itself impliedly or expressly authorizes infringement of the Charter or the Constitution Act, 1867. This principle resonates in the context of matters of health and the environment where broad grants of regulatory powers are essential (Slaight, at 1078; Brown v. Canada (Attorney General), 2022 FCA 104).
[165] The respondents place particular emphasis on the IAA Reference where the Impact Assessment Act, S.C. 2019, c. 28, s. 1 required the administrator to consider factors outside of federal jurisdiction and decisions made under the scheme would have had immediate and tangible effects on matters within provincial jurisdiction (IAA Reference, at paras. 162–169, 190–203).
[166] In stark contrast to the legislative provisions at the heart of the IAA Reference, there are no consequences to anyone from the promulgation of the Order. There is, therefore, an essential assumption in the respondents’ position and the Federal Court’s reasoning that the regulatory power will be brought to bear on matters that cannot conceivably cause harm as it is understood in the constitutional law context. That is not an assumption that the Court can make.
VII. Ancillary points with respect to the division of powers
[167] There is no substantive constitutional question before us as the criminal law power has not been engaged. I propose to dispose of the constitutional question on that basis alone.
[168] Albeit obiter, I wish to briefly address two aspects of the arguments before us concerning the division of powers analysis: dominant purpose and the substantive content of the criminal law power.
Dominant purpose
[169] The Federal Court found that the dominant purpose of the Order was to address environmental harm. The Court rejected arguments that the Order was a colourable attempt to regulate the economy at large or matters of provincial responsibility, and rightly so. In reaching this conclusion the Federal Court correctly applied the governing jurisprudence.
[170] To decide whether a law is valid under the division of powers analysis, courts first characterize the impugned law and then, on that basis, classify it by reference to the heads of power listed in sections 91 to 95 of the Constitution Act, 1867 (Murray-Hall v. Quebec (Attorney General), 2023 SCC 10, at para. 22).
[171] The first or characterization stage aims to ascertain the pith and substance of the law or statutory provisions in issue. To do this, courts analyze both the purpose and effects of the law. Intrinsic evidence, such as purpose clauses and the general structure of the statute, may reveal the purpose of a law. Extrinsic evidence, such as Hansard or other documents tabled as part of the legislative process, such as white papers and Parliamentary committee reports, may also assist in determining a law’s purpose. The effects of a law include the legal effect of the text as well as practical consequences of the application of the statute (Reference re Securities Act, 2011 SCC 66, at para. 64 [Securities Reference]).
[172] The second or classification stage involves determining whether the pith and substance of the law can fall under one of the heads of power of the enacting level of government. As the Supreme Court notes in the Securities Reference, at paragraph 65, “[t]his may require interpretation of the scope of the power.”
Since Citizens’ and The Queen Insurance Cos. v. Parsons, [1880] 4 S.C.R. 215, 1880 CanLII 6 (SCC), it has been generally accepted that the classes of subjects referred to in sections 91 to 95 of the Constitution Act, 1867 must be read together. It follows that very broadly worded heads of power, like subsections 91(2) or 92(13), cannot subsume the more specific heads of power that have been explicitly assigned (Securities Reference, at para. 72).
[173] It is well understood that the same subject or “matter”
may possess both federal and provincial aspects. This means that a federal law may govern a matter from one perspective and a provincial law from another. The federal law pursues an objective that in pith and substance falls within Parliament’s jurisdiction, while the provincial law pursues a different objective that falls within provincial jurisdiction. This concept, known as the double aspect doctrine, allows for the concurrent application of both federal and provincial legislation.
[174] The industry respondents and the Attorneys General of Saskatchewan and Alberta point to the existing involvement of the provinces in the management of plastic waste to argue that the Federal Court erred in not finding that the Order was an intrusion into areas of provincial competence. This argument can be quickly disposed of. Two points need only be made; one legal, one evidentiary.
[175] First, in insisting that a conclusion of overreach necessarily follows from the fact that there might be effects on provincial jurisdiction, the Attorneys General seek to resuscitate the “watertight compartments”
approach to federalism, an approach long rejected as unworkable. In contrast, the Supreme Court has consistently urged a cooperative approach to federalism. Since overlapping powers are practically unavoidable in a modern, complex society, “a court should favour, where possible, the ordinary operation of statutes enacted by
both levels of government”
(Canada Western Bank v. Alberta, 2007 SCC 22, at paras. 37, 42 (emphasis in original)).
[176] The environment is, by its very nature, necessarily an area of shared legislative competence, where overlap is expected (Hydro-Québec, at para. 112; Friends of the Oldman River Society v. Canada (Minister of Transport), [1992] 1 S.C.R. 3, 1992 CanLII 110 (SCC), at paras. 63–67). The Supreme Court in Hydro-Québec specifically noted that CEPA would not preclude the provinces from regulating and controlling pollution and that “there is a wide measure of cooperation between the federal and provincial authorities to effect common or complementary ends”
(at para. 131). Notably, the Court also observed that the fear that CEPA would distort the balance of federalism was “overstated.”
[177] Consistent with the Supreme Court’s guidance, this Court has also applied the cooperative approach doctrine, noting that “protection of the environment…is not a matter or subject of legislation with clearly defined wording likely to be wholly attributed to either of the two levels of government”
(Groupe Maison Candiac Inc. v. Canada (Attorney General), 2020 FCA 88, at para. 48 [Groupe Maison]). CEPA itself anticipates a cooperative approach, recognizing in its preamble that environmental protection involves “all governments in Canada.”
[178] The Attorney General of British Columbia (AGBC) argued, compellingly, that there are practical limitations on the ability of provinces to combat plastic pollution and that the federal government is best placed to regulate many sources of plastic pollution. The AGBC also noted that the achievement of the province’s environmental goals depends, in part, on action by the federal government. In my view, the position of the AGBC aligns with the Supreme Court’s guidance on overlapping jurisdiction and the need for cooperation and coordination between governments in addressing environmental challenges.
[179] Therefore, if necessary, I would decline the invitation by the Attorneys General of Saskatchewan and Alberta to turn the clock of constitutional jurisprudence back a century.
[180] Turning to the evidentiary point, there is nothing in the record that would sustain the argument, let alone a credible argument, that the listing Order is an attempt, colourable or otherwise, direct or indirect, to regulate matters falling within provincial jurisdiction. Apart from a vague reference in the Discussion Paper to the desirability of a “circular economy”
there is no evidence of how the Order intrudes, in any way, let alone in a constitutionally impermissible way, into a provincial domain. It is as simple as that.
[181] I conclude my observation with respect to dominant purpose by noting that, if, through its pith and substance, a law falls under one of the heads of power of the enacting level of government, its effects on the exercise of the other level of government’s jurisdiction are irrelevant (Quebec (Attorney General) v. Canada (Attorney General), 2015 SCC 14, at para. 38; Groupe Maison, at para. 33). Resort need not be had to this doctrine here, as there is no evidence of effects. No colourable purpose having been found, there is, with one exception, no basis on which to hold the Order ultra vires. The exception would be if, in substance, the Order addressed matters that could not be the subject of the criminal law power.
[182] This brings me to the second point.
The substantive content of the criminal law power
[183] The Supreme Court considered the question of the substantive content of the criminal law power in Reference re Genetic Non‑Discrimination Act, 2020 SCC 17 (GNDA).
[184] In GNDA, Karakatsanis J., writing for the majority, wrote “[a]s long as Parliament is addressing a reasoned apprehension of harm to one or more of these public interests [protected by the criminal law, such as peace, order, health, etc.], no degree of seriousness of harm need be proved…”
(at para. 79). For the minority, Kasirer J. wrote that Parliament must be responding to a “threat [that is] ‘real’, in the sense that Parliament had a concrete basis and a reasoned apprehension of harm”
(at para. 234).
[185] I do not accept the respondents’ argument that the requirement for a “concrete basis”
constitutes a departure from the established test. Nor do I accept the argument that this language means the harm must occur before it can be targeted; most criminal law is preventative in nature. Rather, I read Kasirer J. as putting down a salutary marker that the harm cannot be illusory, speculative or a pretext. It must have a grounding, whether in empirical evidence, logic or common sense, but beyond that the courts will not second guess whether resort to the criminal law power is warranted. In this respect, the respondents’ arguments conflate principles of constitutional law with those of administrative law.
[186] That said, it is unnecessary for us to wade into these waters; it would be obiter and, in any event, the Order is grounded in a long history of scientific research and policy consideration by government establishing, in the words of Kasirer J., a “concrete”
or, in the words of Karakatsanis J., “a reasoned apprehension of harm”
(GNDA, at para. 234). I decline the invitation to precisely describe the contours of the criminal law power by picking and choosing which particular plastics are today causing harm, any more than I would crawl through Schedule 1 of CEPA and ask whether a particular substance of the 163 listed substances causes “sufficient harm”
to warrant the attention of the criminal law power. So too with respect to the restricted drugs listed in the Schedule to the Food and Drugs Regulations, C.R.C., 870, (ss. J.01.001, J.01.002, J.01.004) or the myriad of items listed under the Regulations Prescribing Certain Firearm and Other Weapons, Components of Parts of Weapons, Accessories, Cartridge Magazines, Ammunition and Projectiles as Prohibited or Restricted, SOR/98-462, Schedule 3, Parts 1–5, under the Criminal Code, R.S.C. c.C-46.
[187] The Federal Court reached the conclusion that the Order was not directed to a criminal law purpose for the very reason that it found it to be an unreasonable exercise of discretion, concluding as it did that it was not restricted to those items that “truly”
caused harm to the environment (Federal Court Decision, at para. 184). There was no daylight between the administrative law test of reasonableness and the test for constitutionality. In the result, the Court sat in judgement of the “degree of seriousness of harm”
contrary to Supreme Court guidance (GNDA, at para. 79).
[188] The language of “justification”
is the language of administrative law and not criminal law. This point is made by the Supreme Court in GNDA where, after a review of the scope of the criminal law power, Karakatsanis J. restated McLachlin C.J.C.’s words that “the language of justification has no place”
in the division of powers analysis: rather, the test is:
So long as Parliament’s apprehension of harm is reasoned and its legislative action is, in pith and substance, a response to that apprehended harm, it has wide latitude to determine the nature and degree of harm to which it wishes to respond by way of the criminal law power, and the means by which it chooses to respond to that harm: Malmo‑Levine, at para. 213, per Arbour J.; RJR‑MacDonald, at para. 44; Firearms Reference, at para. 39.
[GNDA, at para. 78.]
[189] Constitutional analysis is discrete from administrative law analysis; “justification”
in the sense of Vavilov does not apply. In respect of the criminal law power, “[t]he focus is solely on whether recourse to the criminal law power is
available under the circumstances,”
and the answer to that lies in asking whether there is a reasoned basis for the prohibition (GNDA, at para. 79 (emphasis in original)).
VIII. The BOR decision
[190] The reasonableness of the Minister’s decision not to establish a BOR is assessed in light of the statutory framework empowering the decision and the relevant evidence before the Minister. Both lead to the conclusion that the decision was reasonable.
[191] Beginning with the legal constraints, subsection 333(1) of CEPA (reproduced at Annex A) provides that if a person files a notice of objection with respect to a proposed Order, the MECC “may establish a [BOR] to inquire into the nature and extent of the danger posed by the substance [at issue].”
The statute does not establish any criteria, considerations or other factors to be taken into account. The language is permissive, not mandatory, and speaks to a broad Ministerial discretion over whether or not to establish a BOR.
[192] This understanding of subsection 333(1) is reinforced when contrasted with subsections 333(3) and (4). These provisions, by use of the word “shall,”
make it mandatory for the Minister to establish a BOR in certain circumstances.
[193] The Minister’s task in determining whether a BOR should be established is to decide “whether there is sufficient uncertainty or doubt in the underlying science”
(Goodyear, at paras. 45, 49). The purpose of the BOR is, in essence, to inquire into areas where the Minister is not satisfied as to the underlying science.
[194] The Federal Court did not identify anything in the notices of objection that questioned the reasonableness of the Minister’s conclusion in this respect; rather, it found that the Minister’s decision was not responsive to “the issue of the breadth of the proposed Order,”
which, it said, “was a central argument that challenged the sufficiency of the science”
(Federal Court Decision, at para. 136).
[195] I appreciate the argument that some of the objections before the Minister were somewhat Janus-like. They could be framed as a question of science (i.e. there is no evidence that a certain type of plastic is causing harm) or policy (i.e. what subset of PMI ultimately warrants regulation). The Minister characterized the objections as raising questions that went to the underlying science.
[196] A BOR is not a surrogate for the question of whether to regulate and, if so, which substances or manifestations and how. Those are policy considerations that arise downstream in the legislative process, and which have their own mandatory consultations. It is after a section 90 listing order that the debate with stakeholders about regulatory scope takes place; namely, whether, following a risk assessment, any prohibitions or sanctions ought to be put in place. The Single-use Plastics Prohibition Regulations, derived as they are from the Order listing PMI, are a good teaching point. They arose following a lengthy consultation with the public and industry about the scope of the proposed regulations, generating limitations on their reach, numerous exemptions and tailored timeframes for implementation.
[197] Here, the MECC considered the notices of objection and concluded that the core finding of the Science Assessment—that macroplastics are ubiquitous and may harm the environment—was not challenged. It was reasonable for the Minister to conclude that the thrust of the objections was to debate which PMI should escape a regulatory net.
[198] Not all the notices of objection related to the science; some of the objections “raised legal issues or concerns about risk management approaches”
(Kruidenier Affidavit, Appeal Book, at 16597). Certain objections described the important role of plastics in society and identified concerns about trade and economic harm resulting from decreased investment, reduced consumer demand, or increased costs. These matters are irrelevant to the mandate of the BOR.
[199] Regarding the scientific concerns raised in the notices, a two-step process was followed to determine whether the information submitted merited establishing a BOR. First, scientists reviewed the 27 objections made regarding the Science Assessment’s core findings on the environmental impact of macroplastic pollution to determine whether they raised sufficient uncertainty or established doubt about the evidentiary basis of the proposed Order. The scientists found that they did not. Second, a group of independent scientists with experience in risk assessment reviewed the 60 notices of objection to ensure the first-round review was unbiased, scientifically-sound, and considered all scientific information and arguments submitted. This second, neutral review was to ensure that “the scientific process had been respected”
(Kruidenier Affidavit, Appeal Book, at 16597–16598; Memorandum to Minister, Appeal Book, at 1246).
[200] The criteria relied on by the Minister to assess and consider the notices of objection were legally relevant and consistent with the jurisprudence (Goodyear, at paras. 45–47). The RIAS summarized the issues raised in the notices of objection and explained on an objection-by-objection basis why they did not undermine the scientific approach or conclusions of the Science Assessment (RIAS, Appeal Book, at 799–800).
[201] The Federal Court held that the Minister did not “grapple”
with the arguments raised in the notices of objection. Some of the objectors argued that CEPA was not the appropriate tool to address the problem, some suggested that new legislation should be enacted, some that the Order was too broad and should be narrowed and still others objected that the provinces should deal with the problem.
[202] The MECC addressed these objections, head-on, and gave reasons why they did not justify constituting a BOR. The Ministerial decision letters differentiated between those objections that raised no scientific issue (Appeal Book, CKF Inc. Response Letter, at 2641–2642), those that challenged the data (Appeal Book, Vinyl Institute Response Letter, at 2751–2752), those that raised a scientific reason (Appeal Book, Layfield Group Response Letter, at 2643–2644), and those that raised a specific scientific reason (Appeal Book, Gowling WLG Response Letter, at 2675–2676). The Minister decided that many of the objections were policy questions about the scope and nature of the regulatory response under section 93.
[203] A reasoned disagreement with a submission does not lead to the conclusion that the decision maker failed to “grapple”
with the issue. Grappling does not mean acceding. The only way the MECC could better “grapple”
with the objections would be for the Minister to agree that that the objections undermined the findings and conclusion of the Science Assessment. Here, the Minister noted the objections and explained, in respect of each, why he disagreed. This was a reasoned exercise of discretion considering the purpose and text of section 333.
IX. Whether the appeal is moot
[204] In March 2023, the Strengthening Environmental Protection for a Healthier Canada Act, S.C. 2023, c. 12 received Royal Assent, with the result that Schedule 1 of CEPA was amended and re-introduced in two parts, and the Order in question in these proceedings no longer existed at law. Once Schedule 1 was re-introduced, there was no live controversy between the parties (Borowski v. Canada (Attorney General), [1989] 1 S.C.R. 342, 1989 CanLII 123 (SCC), at pp. 353 [Borowski]).
[205] The Federal Court erred in finding that the judicial review application was not moot and in failing to consider whether, in the exercise of its discretion, it ought to hear the application (Federal Court Decision, at para. 32).
[206] However, this error was of no consequence. This Court should still hear and determine the matter.
[207] In deciding whether to hear a matter that is moot, a court can consider the existence of an adversarial context, judicial economy, and the adjudicative role of the courts (Borowski, at pp. 358–363). Here, there exists an adversarial context between the parties and Schedule 1, containing identical substances, was re-introduced under the new legislation. Further, the Single-use Plastics Prohibition Regulations were enabled by the listing of PMI on CEPA’s Schedule 1. The time and expense invested by the parties, as well as the continued relevance of the legal issues, justify a determination on the merits.
X. Disposition
[208] I would allow the appeal and dismiss the application for judicial review with costs in this Court and below. The constitutional question is answered in the negative.
[209] Should the parties fail to agree on costs, they are to advise the Registry within 30 days of the date of this Judgment, following which the Court will give directions for the determination of costs.
“Donald J. Rennie”
|
“I agree. |
|
George R. Locke J.A.” |
|
“I agree. |
|
Sylvie E. Roussel J.A.” |
Email this Content