Correctness, Reasonableness and the Scope and Limits of Judicial Review: Administrative Law in 2024 PDF Free Download
1 / 45/45
100%
Correctness, Reasonableness and the Scope and Limits
of Judicial Review: Administrative Law in 2024
Paul Daly
University Research Chair in Administrative Law & Governance, University
of Ottawa
We are coming up on the fifth anniversary of the Supreme Court of Canada’s landmark
decision in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65,
[2019] 4 SCR 653. At the five-year mark after the release of the predecessor to Vavilov
(Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 SCR 190), Professor Mullan wrote
an article about the issues left unresolved – the top 15!
No such article could be written today. The issues left unresolved by Vavilov are
comparatively few. I identified five subsequent to the decision: internal standard of
review, arbitration appeals, procedural fairness, Charter review and the extent to which
reasonableness review is constitutionally entrenched. I later added the standard of review
for regulations to the list.
Of these six, the Supreme Court has squarely dealt with procedural fairness (Law Society
of Saskatchewan v. Abrametz, 2022 SCC 29); will imminently deal with the standard of
review for regulations; and has decided cases addressing Charter review and the
constitutional foundations of judicial review. It has only avoided resolving one of the
outstanding issues – arbitration appeals – and has not offered any guidance on internal
standard of review (though here it must be pointed out that the Supreme Court has never
addressed this issue).
Accordingly, at the five-year mark, the question is more how faithful the Supreme Court
has been to the Vavilov framework than how it needs to clarify or tweak the framework.
In the last 12 months, the Court has made a number of significant decisions applying the
Vavilov framework and addressing unresolved issues. Most of the focus of this ‘year in
review’ paper will be on those.
In Part I, I will address reasonableness review, focusing on the decision in the Mandate
Letters case and also addressing the most recent entry in the ‘Charter values’ ledger. In
Part II, I will focus on correctness review, with two Charter case from the labour and
employment field forming the core of my analysis. In Part III, I address some recent case
law on justiciability, a pair of recent cases on the public-private divide in Canadian
administrative law and a pair of others on what constitutes a ‘decision’ subject to judicial
review. Lastly, in Part IV, I turn my attention to constitutional issues, on the one hand the
scope of the core constitutional minimum of judicial review post-Vavilov, on the other
the scope of application of the Charter.
I. Reasonableness Review
A. The Mandate Letters Case
The Supreme Court of Canada’s most notable recent entry in its standard of review
catalogue is Ontario (Attorney General) v. Ontario (Information and Privacy
Commissioner), 2024 SCC 4. This decision follows Mason v. Canada (Citizenship and
Immigration), 2023 SCC 21 in concluding that an administrative decision was
unreasonable and, moreover, that there was only one reasonable interpretation of the
provision at issue. I was relatively sanguine about the decision in Mason; I am less
enamoured of the Mandate Letters case.
Here, the Commissioner had ordered that mandate letters issued by the Premier of
Ontario to his cabinet ministers should be released under the Freedom of Information
and Protection of Privacy Act, RSO 1990, c F.31. The government of the province had
resisted disclosure on the basis of s. 12(1) of the Act, which creates an exemption “where
the disclosure would reveal the substance of deliberations” of the cabinet for a range of
documents “including” (but not limited to) agendas, minutes or lists of policy options
presented to cabinet. The basic premise of the Commissioner’s detailed reasons for
decision was that mandate letters, which memorialize decisions that have already been
taken, would not reveal the substance of deliberations.
Applying reasonableness review, Karakatsanis J concluded that the Commissioner’s
decision was unreasonable. The key error was that the Commissioner had failed to have
adequate regard to the constitutional context: “Because s. 12(1) was designed to preserve
the secrecy of Cabinet’s deliberative process, the constitutional dimension of Cabinet
secrecy was crucial context in interpreting s. 12(1)” (at para. 27). This caused a loss of
confidence in the outcome (at para. 23).
The Commissioner erred in two ways. First, he gave too narrow a scope to s. 12(1). The
Commissioner focused only on two rationales for cabinet confidentiality — promoting
ministerial candour and preserving collective solidarity — to the exclusion of a third,
efficient government. Failing to take this rationale into account caused him to ascribe too
narrow a purpose to s. 12(1) and to fail to respond to one of the government’s submissions:
First, had the IPC recognized that the fundamental focus of deliberative secrecy
is effective government, the Commissioner could not have framed the purpose to
focus only on “free and frank discussion among Cabinet members”. Rather, as
Lauwers J.A. noted, a contextual interpretation of s. 12(1) instructs that the
provision more broadly aims to establish the confidentiality necessary for the
executive to function effectively (paras. 187 and 208).
Second, had the IPC framed the purpose of s. 12(1) more broadly, he may not have
rejected a central argument from Cabinet Office going to the scope of s. 12(1).
Cabinet Office argued that, along with ensuring ministerial candour and solidarity,
Cabinet secrecy also helps to ensure the deliberative process runs efficiently by
preserving the confidentiality of deliberations until a final decision has been made
and announced (IPC reasons, at paras. 30-32; A.R., vol. III, at pp. 90, 101-2, 228
and 232) (at paras. 34-35).
As a result, the Commissioner also “did not acknowledge Cabinet Office’s submission that
determining “when and how” to communicate policy priorities to the public and
opposition parties is itself an important part of Cabinet’s deliberative process” (at para.
37, emphasis original). In particular, the Commissioner “concluded that “outcomes” of
the deliberative process are not encompassed by the opening words of s. 12(1), full stop,
without acknowledging that an important part of Cabinet confidentiality is government’s
prerogative to decide how and when to announce policy priorities (see para. 104)” (at
para. 39).
Second, the Commissioner failed to have regard to constitutional conventions and
traditions relating to the nature of cabinet decision-making and the premier’s role in the
process:
The Letters on their face contain communications between the Premier and
Cabinet ministers about policy priorities, many if not most of which would require
decisions from Cabinet, both as to their substance and as to how they should be
communicated to the public. Cabinet “formulates and carries out all executive
policies,” and all major government policy matters are forwarded to Cabinet for
decision (Hogg and Wright, at § 9:5; M. Schacter and P. Haid, Cabinet Decision-
Making in Canada: Lessons and Practices (1999), at p. 1; see also Brooks, at
p. 236). There is no basis in convention or past precedent to separate the Premier’s
role in this process from the rest of Cabinet. Disclosure of the Premier’s initial
priorities, when compared against later announcements of government policy and
what government actually accomplished, would reveal the substance of what
happened during Cabinet’s deliberative process. The IPC’s characterization of the
Letters as “the end point of the Premier’s formulation of the policies and goals to
be achieved by each Ministry”, or “the product of his deliberations” was thus beside
the point, and an unreasonable basis upon which to deny protection under s. 12(1)
(paras. 132 and 134 (emphasis added); see also para. 79).
Relatedly, to the extent the IPC required evidence linking the Letters to “actual
Cabinet deliberations at a specific Cabinet meeting”, that approach was
unreasonable. Such a requirement is far too narrow and does not account for the
realities of the deliberative process, including the Premier’s priority-setting and
supervisory functions, which are not necessarily performed at a specific Cabinet
meeting and may occur throughout the continuum of Cabinet’s deliberative
process. Accordingly, it would be unreasonable for the Commissioner to establish
a heightened test for exemption from disclosure that would require evidence
linking the record to “actual Cabinet deliberations at a specific Cabinet meeting”…
[The Commissioner’s] focus on actual deliberations at a specific Cabinet meeting
underscored his finding that the fact that some policy priorities “may never return
to Cabinet at all or . . . may be altered or amended in significant . . . ways” was a
“deficiency” in Cabinet Office’s continuum argument and meant that the Letters
could not be exempted in their entirety (para. 121). This determination was
unreasonable because it did not account for the fact that disclosure of early policy
priorities not acted on, or changed in significant ways before implementation,
would be revealing of the substance of Cabinet deliberations — whether the
decision to abandon or alter the priority was the decision of Cabinet, its
committees, or the Premier (at paras. 53-55).
These considerations led inexorably to the conclusion that the mandate letters were
covered by s. 12(1). Karakatsanis J refused to remit the matter to the Commissioner.
The methodology of reasonableness review is worthy of comment. Karakatsanis J refused
to grapple with whether the appropriate standard of review was correctness or
reasonableness, on the basis that the decision was unreasonable and thus could not
survive under either standard (at para. 16). In substance rather than in form, however,
the analysis looks very much like correctness review in the guise of reasonableness review.
The first difficulty here is that Karakatsanis J’s analysis of the Commissioner’s reasons
focuses on two aspects of context even though the Commissioner conducted a wide-
ranging analysis. As Côté J accurately observes:
The Commissioner relied on, among other things, the stated purposes of the
legislation (see paras. 106-8); the principle that “exemptions from the right of
access should be limited and specific” (s. 1(a)(ii) of the Act); our Court’s decisions
in Babcock and Dagg v. Canada (Minister of Finance), 1997 CanLII 358 (SCC),
[1997] 2 S.C.R. 403; appellate jurisprudence across the country, notably O’Connor;
and a significant body of past administrative decisions. All of these factors lend
support to his interpretation (at para. 81).
However, Karakatsanis J does not acknowledge the Commissioner’s reliance on these
indicia of reasonableness, nor does she explain why the Commissioner’s decision is
unreasonable because it fell short in two areas even though it could be defended on other
grounds. This was an extensive decision where a loss in confidence in the outcome in one
respect could, in theory at least, be offset by the strengths of the reasons in other respects.
Moreover, second, there is significant force to Côté J’s charge that in order to identify the
shortfalls in the Commissioner’s decision, Karakatsanis J “conducts her own
interpretation” of s. 12 and “uses her conclusions as a yardstick” against which to measure
the Commissioner’s interpretation (at para. 76):
For example, my colleague refers to three rationales for the convention of Cabinet
confidentiality: “. . . candour, solidarity, and efficiency . . .” (para. 30). She finds
that the Commissioner considered the first two of these rationales but that he “did
not engage with a core purpose of Cabinet secrecy to promote the efficiency of the
collective decision making” or with the ultimate goal of effective government
(para. 32). However, this third rationale of “efficiency”, while an important tenet
of Cabinet privilege, has not been articulated by our Court as such. As a result, I do
not agree that it was unreasonable for the Commissioner to not address a concept
that is fully expressed only in scholarly authority (see Karakatsanis J.’s reasons, at
paras. 30 and 36, citing Y. Campagnolo, “The Political Legitimacy of Cabinet
Secrecy” (2017), 51 R.J.T.U.M. 51, at p. 68, and Y. Campagnolo, Behind Closed
Doors: The Law and Politics of Cabinet Secrecy (2021), at p. 26) (at para. 76).
It is worth repeating how Karakatsanis J herself framed the issue: “had the IPC framed
the purpose of s. 12(1) more broadly, he may not have rejected a central argument from
Cabinet Office going to the scope of s. 12(1)” (at para. 35, my emphasis). This gets things
backwards: the submission could only have been a “central argument” if the Commission
had agreed that the “purpose” of s. 12(1) should have been “framed…more broadly”. I have
some difficulty appreciating how this can be described as “reasons-first” reasonableness
review as Vavilov requires. There is an important difference between this case and the
Supreme Court’s recent decision in Mason (noted here), where the tribunal had simply
failed to address a central argument for an unarticulated reason, not because of the
purpose it ascribed to the provision.
Furthermore, and fundamentally, the choice of standard of review is of critical
importance here. On reasonableness review, the Commissioner is in the interpretive
driving seat. If deference is to mean anything it all, it must be up to the Commissioner to
determine whether the law should be changed to incorporate an additional rationale as
background context for s. 12(1). As it was, the Commissioner gave multiple reasons in
support of his interpretation and, ordinarily, on a deferential standard his refusal to
change, extend or expand the law would be respected. On correctness review, by contrast,
the courts have the last word (and, indeed, Côté J applying correctness review agreed that
the law should be updated to incorporate this third rationale). This should have been an
example — a pretty good one, I would have thought — of the standard of review making a
difference. In my view, reasonableness was the appropriate standard, for reasons I
developed here and from which Côté J’s neatly-done argument at para. 55-61 does not
dissuade me, and an appropriately deferential approach would have led to the decision
being upheld as reasonable.
The discussion of legal and factual constraints is also worthy of comment. Again, there is
an important difference from Mason: there, the legally binding constraint of international
law identified by the Supreme Court at least had the merit of being set out expressly in
statute; but here, the conventions and traditions relied upon by the Supreme Court are
entirely unwritten. This is different even from CSFTNO (discussed below), where the
relevant Charter values had been expressed repeatedly in binding Supreme Court of
Canada jurisprudence on the objectives of s. 23 of the Charter.
Here, the conventions and traditions have not and arguably cannot be reduced to precise
textual statements (and, of course, by their very nature are not binding law at all). Indeed,
as political scientist Professor Emmett Macfarlane points out in a critical comment on the
decision, these conventions and traditions shift over time, meaning that any
administrative decision-maker required to consider them will be aiming at a moving
target:
[T]he problem in this case is that convention is wholly silent on the place or
relationship of mandate letters to cabinet deliberations. The Court rests its
decision quite heavily on a discussion of convention that is largely irrelevant to the
central issue.
In fact, the emergent practice of releasing mandate letters to the public (not only
at the federal level but in Ontario under former Premier Kathleen Wynne) is
precisely what led to this controversy in the first place! Stunningly, the Court pays
no heed to this practice – a practice that was unlikely to emerge if those
governments felt it would somehow constrain or impair cabinet confidentiality or
effective decision-making. In this fundamental sense, recent political practice
directly contradicts some of the Court’s conclusions about the effects releasing
mandate letters might have on cabinet confidentiality.
Professor Macfarlane also notes that the Court gets quite far into the “weeds” of the
operation of government; traditionally, of course, making findings of fact and drawing
inferences based on a detailed record is a matter for the administrative decision-maker.
For these reasons, I am somewhat sceptical that the Court arrived at the correct
conclusion in this case. Whereas in Mason and CSFTNO the legal and factual constraints
identified by the Court stood on solid ground (and the decisions under review were silent
on key aspects of the arguments and evidence), in Information Commissioner there is
significant force to Côté J’s charge that the majority engaged in correctness review in
substance if not in form. It will be necessary to read the next entries in the standard-of-
review catalogue very carefully to see if the Court is sending a signal about the level of
intensity of reasonableness review under the Vavilov framework.
In that regard, the recent decision of the Quebec Court of Appeal in Piché c. Entreprises
Y. Bouchard & Fils inc., 2024 QCCA 1374 is very interesting. It suggests that the
reasonableness standard set out in Canada (Minister of Citizenship and Immigration) v.
Vavilov, 2019 SCC 65, [2019] 4 SCR 653 may indeed have become more robust.
Nonetheless, whilst quashing the decision at issue for unreasonableness, the Court of
Appeal refrained from imposing a solution even in circumstances where it had been asked
to find that there is only one possible, acceptable interpretation.
The underlying issue is very interesting. There are two streams of authority in Quebec’s
Tribunal administratif du travail (which has jurisdiction over workplace health and
safety) about compensating workers who withdrew from the workplace during the
COVID-19 pandemic. Both the appellant and the Commission des normes, de l’équité, de
la santé et de la sécurité urged the Court of Appeal to resolve two interpretive questions
on which the Tribunal is divided.
Here, the applicant was working as a paramedic. His doctors recommended that he
temporarily withdraw from the workplace for his own safety as he was taking medication
that made him immunodeficient. In the end, he went back to work about nine months
after the outbreak. For a three-month period, he received no pay. Under Quebec
legislation, an employee who withdraws from work for preventative health reasons can
claim an indemnity. But there are several conditions (at para. 14), one of which is that the
source of the danger to health must come from a “contaminant”. In this case (though not
in others!), the Tribunal held that COVID-19 was not a “contaminant” within the meaning
of the legislation. The Tribunal also held, based on its analysis of the evidence of his
immunodeficiency, that the second condition – danger to the worker – was not met either.
The Tribunal did not go on to consider the third condition – alteration of the worker’s
health.
The Court of Appeal (Moore, Cournoyer and Bachand JJA) held that the decision was
unreasonable. Following Vavilov, the judges did not apply the correctness standard to
resolve the split on the Tribunal but noted that a deficient statutory interpretation
analysis by an administrative decision-maker would justify intervention on judicial
review for unreasonableness (at para. 35). Indeed, they cited the Supreme Court’s
decisions in Mason v. Canada (Citizenship and Immigration), 2023 SCC 21 and Ontario
(Attorney General) v. Ontario (Information and Privacy Commissioner), 2024 SCC 4 as
further support for this proposition:
C’est ainsi que, dans Mason, si la majorité de la Cour suprême a reconnu que le
décideur avait appliqué plusieurs techniques d’interprétation législative
reconnues, elle conclut tout de même au caractère déraisonnable de la décision au
motif que le décideur n’a pas tenu compte de deux arguments du contexte législatif,
des conséquences potentiellement importantes de la décision et des contraintes
imposées par le droit international.
C’est ainsi encore que dans l’arrêt Ontario (Procureur général) c. Ontario
(Commissaire à l’information et à la protection de la vie privée), bien que la
majorité de la Cour suprême souligne que le décideur a pris en compte le texte,
l’objectif général des lois sur l’accès à l’information et deux des objectifs de la
confidentialité du Cabinet, elle conclut tout de même au caractère déraisonnable
de la décision au motif que le décideur n’a pas traité d’un troisième objectif de la
confidentialité du Cabinet, soit l’efficacité du processus décisionnel collectif ainsi
que de certaines conventions constitutionnelles (at paras. 36-37).
Here, the decision was unreasonable because, in the first place, the Tribunal fixated on
an amendment to the statutory scheme made in 2015 and gave it a much broader scope
than the legislature intended (at para. 40), without regard for other textual and contextual
indicators about the meaning of “contaminant” (at paras. 42-44). In addition, the
Tribunal failed to have regard to the purpose of the legislature (at paras. 45-47). In the
result, the Court of Appeal held — bolstered by Mason and Information Commissioner —
that the statutory interpretation exercise undertaken by the Tribunal in the instant case
neglected key elements of text, context and purpose and was, thus, unreasonable (at para.
48).
Interestingly, the Court of Appeal did not conclude, however (unlike the Supreme Court
in Mason and Information Commissioner) that there was only one reasonable
interpretation. Rather, it left the matter to the Tribunal, albeit with a stern warning about
the desirability of legal certainty (at para. 51) and a reminder of the mechanisms available
to achieve that certainty:
En l’espèce, il ressort qu’au moins trois avenues sont possibles : 1) la Covid-19 n’est
pas un contaminant; 2) la Covid-19 est un contaminant; et 3) la Covid-19 est parfois
un contaminant selon les activités de l’employeur. Quant à la controverse
jurisprudentielle au sein du TAT, elle n’est pas ici suffisante pour justifier que la
Cour tranche à la place de celui-ci. Comme la Cour suprême le rappelait, il convient
de laisser d’abord au décideur administratif le soin de résoudre ce désaccord à
l’aide de ses mécanismes internes, que ce soit, en ce qui concerne le TAT, la mise
sur pied d’une formation de trois membres ou la participation des membres à
l’élaboration d’orientations générales (at para. 50).
It said the same about the jurisprudential split on the third condition – the extent to which
the “contaminant” has to affect the health of the worker (at paras. 66-70).
On the question of danger (the second condition), there is an interesting discussion of the
role of the Tribunal. Even though the Tribunal sits de novo and potentially has access to
a wide range of materials, its role in assessing danger is held to be limited. In short, its
role is to assess whether there was danger at the moment the worker withdrew based on
the evidence available at the time. In other words, the Tribunal cannot retrospectively
apply evidence that became available after the worker’s withdrawal to conclude, with the
benefit of hindsight, that there was no danger (at para. 55). The Tribunal may only take a
forward-looking view of the matter, with the right to withdraw dissolving only
prospectively, from the point the Tribunal has new evidence at its disposal (at para. 56).
On the evidence here, there was no doubt in the Court of Appeal’s eyes that there was
ample evidence of danger at the moment the worker withdrew (at para. 57). Either,
therefore, the Tribunal disregarded a legal constraint by focusing on danger
retrospectively rather than prospectively or disregarded factual constraints by
fundamentally misapprehending the evidence: unreasonableness was the inevitable
conclusion (at para. 64). Again, the matter was returned to the Tribunal to determine the
point at which the worker would have been able to safely reintegrate the workplace (at
para. 65).
Overall, this decision indicates that Vavilov certainly does furnish the tools for a robust
reasonableness review of administrative decisions. The Supreme Court’s recent decisions
applying Vavilov are cited as further support for this proposition. Even if the Court of
Appeal did send back the interpretive question about the first issue – “contaminant” – its
analysis strongly favoured one of the streams of jurisprudence. Nonetheless, the Court of
Appeal here did not go as far as the Supreme Court went in Mason and Information
Commissioner, insisting instead that the matter should be returned to the Tribunal for
the identification of a definitive answer albeit with a fairly stern warning that the Tribunal
should find a way to resolve its internal jurisprudential conflict in short order.
B. Charter Values
As is well known, in Doré v. Barreau du Québec, 2012 SCC 12, [2012] 1 SCR 395, the
Supreme Court applied a deferential standard of review to an alleged interference by an
administrative decision with the Charter right to freedom of expression. Casting its
analysis in terms of Charter values rather than Charter rights, the Court held that
administrative decision-makers should focus on ensuring a proportionate balance
between relevant Charter values and statutory objectives. In subsequent decisions, the
Court reaffirmed Doré and, in Law Society of British Columbia v. Trinity Western
University, 2018 SCC 32, [2018] 2 SCR 293, clarified that the Doré analysis requires (1)
a demonstration that Charter values were considered in the decision-making process and
(2) a demonstration that the relevant values and statutory objectives were proportionately
balanced.
However, in Vavilov, the Supreme Court left Doré for another day. So criticized has the
decision been over the years that many expected Doré to be overruled at the first available
opportunity. But reports of Doré’s demise have been greatly exaggerated. In Commission
scolaire francophone des Territoires du Nord-Ouest v. Northwest Territories
(Education, Culture and Employment), 2023 SCC 31, the Supreme Court of Canada
resoundingly restated Doré, now superpowered by the commitment to responsive
reasons in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65,
[2019] 4 SCR 653.
This case involved the review of ministerial discretion. Section 23 of the Canadian
Charter of Rights and Freedoms creates categories of ‘rights holders’ who, because of
their personal circumstances as speakers or learners of a minority language, “have the
right to have their children receive primary and secondary school instruction in that
language in that province”.
In the Northwest Territories (as in most of Canada), French speakers are the linguistic
minority. Years of Supreme Court of Canada jurisprudence on minority language rights
have established that, under certain conditions, provinces must make education available
in English (in Quebec) or French (everywhere else) to cater for their minority-language
community.
In fact, this case involved several exercises of ministerial discretion to deny entry into
French-language schools. The Minister had adopted a policy that expanded the s. 23
categories and made a wider group of children eligible to attend school in French.
However, the children at issue in this case did not fall within the scope of the policy. The
children were either French-language speakers, otherwise embedded in the French-
language community in the Northwest Territories or would contribute to the vitality of
the community by attending school in French. In each case, the Commission (the
provincial Francophone school board) recommended that they be permitted to attend a
French-language school. But the children did not fall within the scope of the policy.
Essentially on that basis the Minister refused to permit them to attend a French-language
school.
This case was, therefore, a pure case of Charter values. No Charter rights were at stake.
Instead, the children (and, in the Supreme Court, the Commission) invoked
the Charter values underpinning s. 23: “Section 23 has a remedial purpose related to
promoting the development of official language minority communities and changing the
status quo” (Conseil scolaire francophone de la Colombie-Britannique v. British
Columbia, 2020 SCC 13, at para. 3). The provision is “preventive, remedial and unifying
in nature”, as it is “intended not only to prevent the erosion of official language
communities, but also to redress past injustices and promote the development of those
communities” (ibid., at para. 15).
Unanimously, the Court found that the Charter values underpinning s. 23 were engaged
and that the Minister had failed to grapple meaningfully with them in his decision.
Notably, the Court’s reasons were written by Côté J, who has previously expressed
scepticism about Doré: Law Society of British Columbia v. Trinity Western University,
2018 SCC 32, especially at paras. 306-309. Here, however, the Court reaffirmed Doré in
no uncertain terms:
Indeed, it has consistently been held that the Doré framework applies not only
where an administrative decision directly infringes Charter rights but also in
cases where it simply engages a value underlying one or more Charterrights,
without limiting these rights (Doré, at paras. 35 et seq.; Loyola, at para. 4; Trinity
Western University, at para. 57). This is the case because administrative decision
makers have an obligation to consider the values relevant to the exercise of their
discretion, in addition to respecting Charterrights. There can be no doubt about
this, because “[t]he Constitution — both written and unwritten — dictates the
limits of all state action” (Vavilov, at para. 56). As L’Heureux-Dubé J. clearly
stated in Baker v. Canada (Minister of Citizenship and Immigration), 1999
CanLII 699 (SCC), [1999] 2 S.C.R. 817, a discretionary decision, to be reasonable,
must be made in accordance with the “fundamental values of Canadian society” as
reflected in the Charter(para. 56). Relying on this statement, Abella J. held
in Doré that discretionary decisions must “always” take Charter values into
consideration (para. 35 (emphasis in original)) (at paras. 64-65).
The methodology of the Doré framework was summarized in the following terms:
a reviewing court must first determine whether the discretionary decision
limits Charter protections. If this is the case, the reviewing court must then
examine the decision maker’s reasoning process to assess whether, given the
relevant factual and legal constraints, the decision reflects a proportionate
balancing of Charter rights or the values underlying them. If not, the decision is
unreasonable (at para. 73).
To begin with, if a Charter right or value is engaged, “the court must, under the approach
laid down in Doré, determine whether the decision is reasonable through an analysis of
its proportionality” (at para. 67). This is underscored by the emphasis in Vavilov on
responsive reasons. First, a reasonable decision “must reflect the fact that the decision
maker considered the Charter values that were relevant to the exercise of its discretion”
(at para. 68). Second, indeed, the reasons “show that the decision maker “meaningfully”
(Vavilov, at para. 128) addressed the Charter protections to “reflect” the impact that its
decision may have on the concerned group or individual (para. 133)” (at para. 68). Third,
the reviewing court’s role, accordingly, must be robust: it “must take into account the role
of the courts as guardians of the Constitution and must reflect the particular importance
of justification in decisions that engage Charter protections (Vavilov, at para. 133)” (at
para. 70). Fourth, unlike in ordinary judicial review cases, where it is forbidden to reweigh
the factors considered by the decision-maker, “the Doré approach requires reviewing
courts to inquire into the weight accorded by the decision maker to the relevant
considerations in order to assess whether a proportionate balancing was conducted by the
decision maker” (at para. 72).
This is a striking commitment to robust judicial review, building on Vavilov, of
infringements of both Charter rights and Charter values. The only limitation (apart from
the Charter being engaged) is that the Charter value must be relevant: “it will often be
evident that a value must be considered, whether because of the nature of the governing
statutory scheme (at para. 108), because the parties raised the value before the
administrative decision maker (at paras. 127-28), or because of the link between the value
and the matter under consideration” (at para. 66). This, incidentally, answers the
objection that the Doré framework is less rights-protective because the onus does not
shift to the state to justify an infringement of the Charter (Trinity Western, at para. 312):
in fact, once the relevance of the Charter has been established, the requirement of
responsive justification kicks in and necessitates a demonstration of proportionality by
the decision-maker.
There is no obvious difference between Charter rights and Charter values in this
framework, although there is a distinction made (at para. 65) between
“consider[ing]…values” and “respecting Charter rights”. It is also said that the “context
constrains” the decision-maker (at para. 68) and one can posit that the context of an
infringement of a Charter right may differ (and require more by way of justification) than
the infringement of a Charter value. Indeed, it is notable in this case that the obligation
on the Minister was to “truly take into account the constitutional values of preservation
and development of official language minority communities, in other words, that she
meaningfully address the considerations arising therefrom” (at para. 92): it was an
obligation to take into account (and meaningfully grapple with) the relevant values rather
than to comply with them, as would be the case with a Charter right. Indeed, Côté J was
careful to observe that she was not elevating the values underpinning s. 23 to the right
itself, as she made clear that there was no “obligation on decision makers in [the
Minister’s] position to admit all children of non-rights holder parents”. More generally,
the legal and factual constraints (at para. 73) may play out differently in rights and values
cases, not least because the jurisprudence on the content of a Charter right will be one of
those legal constraints and limit a decision-maker’s freedom of action more extensively
than a value.
On the facts, the underlying values of s. 23 were engaged, as there was a “clear link”
between s. 23 and the exercises of discretion “because the decisions were likely to have an
impact on a minority language educational environment” (at para. 78). Preserving and
developing the minority-language community (which admission of the children would
have contributed to) are s. 23 values (at paras. 80-82):
Thus, the decisions rendered by provincial and territorial governments regarding
the admission of children of non-rights holder parents to minority language
schools, even when they do not directly infringe the right guaranteed by s. 23, can
nevertheless have a significant impact on the preservation and development of
minority language communities. It follows that these values are always relevant
when the government exercises such a discretion and that they must therefore be
taken into account. For the purposes of this appeal, this means that the Minister
was required to consider the values of preservation and development of minority
language communities in exercising her discretion to decide whether to admit
children of non-rights holder parents to the schools of the Francophone minority
in the Northwest Territories (at para. 83).
Here, the evidence demonstrated that “there was a link between the admission of children
of non-rights holder parents to French-language schools in the Northwest Territories and
the preservation and development of the Francophone community there” (at para. 86).
Indeed, the evidence showed here that “the admission of the children of the appellant
parents would have had considerable benefits for the preservation and development of
the language and culture of the minority language community” (at para. 101).
The Minister failed to justify her decisions given the evidence of this link: she failed to
consider the Commission’s support for the applications (at para. 98); she did not consider
the “individual characteristics of the various applications in relation to the benefits that
could result from a decision to grant them” (at para. 99); she attached “too much
importance to her duty to make consistent decisions” (at para. 102); and “gave
disproportionate weight to the cost of the contemplated services in the exercise of her
discretion” (at para. 102). This meant that her reasons did not demonstrate that she
“meaningfully addressed the values of preservation and development of the Francophone
community of the Northwest Territories so as to reflect the significant impact that the
decisions might have on it” (at para. 102).
Perhaps this decision could have been clearer on the different roles played
by Charter rights and Charter values, and said more on the methodology for
identifying Charter values (or, maybe, Charter purposes, at paras. 75, 78). I have
addressed these issues in forthcoming scholarship: “The Charter in Administrative
Decision-Making: Defending the Duty to Take Charter Values (or Purposes) Into
Account”, where I suggest minor modifications to the Doré duty. CSFTNO represents,
nonetheless, a resounding restatement of a much-maligned framework. Doré will have to
die another day. Vavilov has given it a new lease on life.
II. Correctness Review
A. Charter Rights and Charter Values
The Charter was front and centre in another important administrative law decision in
2024: York Region District School Board v. Elementary Teachers’ Federation of Ontario,
2024 SCC 22. This is an important decision about the application of the Charter of Rights
and Freedoms to public school boards and standard of review of administrative decisions
that infringe Charter rights. I will address the former point below. Here, I will address
the latter point, which attracted a majority and minority view.
In my view, the decision in York Teachers confirms a distinction between the scope and
application of Charter rights for standard of review purposes and continues to develop
the distinction between Charter rights and Charter values as limitations on the freedom
of action of administrative decision-makers. The discussion of the latter point in the
competing judgments invites reflection on the relationship between correctness review
and the “constraints” that bear upon an administrative decision-maker subject to
reasonableness review.
The underlying facts arose in a tense school workplace in Toronto. Two teachers created
and edited a private log recording their concerns about the toxic environment in the
school. The existence of this log became notorious within the school, so much so that the
principal attempted unsuccessfully to access it through an IT search.
Then, one day the principal found himself alone in the classroom of one of the teachers.
The teacher’s computer had been left open and, so, the principal was able to access the
log. He took screenshots which, ultimately, formed the basis of reprimands to the two
teachers. Subsequently, the Board seized this laptop and the other teacher’s laptop and
searched them for the log, albeit fruitlessly. The teachers’ union grieved the reprimands
under the collective agreement on the basis that the Board had violated their reasonable
expectation of privacy in the log.
The arbitrator found in favour of the Board. She did not conduct a Charter analysis.
Section 8 of the Charter, which protects individuals against unreasonable searches and
seizures, and which has the ‘reasonable expectation of privacy’ as its lynchpin, was not
raised before the arbitrator, but she nonetheless applied principles and jurisprudence
drawn from the s. 8 context. She found that the teachers had a reasonable expectation of
privacy but that it was diminished in the circumstances. Further, she analyzed three
potential breaches of s. 8 — the initial IT search, the access to the log on the open
computer and the ‘forensic’ search of the laptops — and found that they did not constitute
unreasonable searches.
The courts below were divided on the applicable analytical framework. At first instance,
the majority and dissenting judge at the Divisional Court applied reasonableness review.
But at the Court of Appeal, Huscroft JA held that s. 8 was engaged and that the arbitrator’s
application of the legal principles of s. 8 was subject to correctness review (although
deference would be given to the arbitrator on her findings of fact).
The Supreme Court also split on standard of review on the s. 8 issue (they all agreed that
the application of the Charter to school boards had to be determined correctly: see paras.
62, 108). The majority applied correctness, the minority reasonableness. As if to
demonstrate the closeness of the question, both majority and minority cited my academic
work in support of their respective positions.
(i) Correctness Review: Scope and Application of s. 8
Rowe J wrote for the majority. He applied correctness review because “the arbitrator
erred in failing to appreciate that a Charter right arose from the facts before her” (at para.
63). This failure brought the case within Vavilov’s constitutional question category, as
questions about “whether a Charter right arises, the scope of its protection, and the
appropriate framework of analysis” must receive a uniform answer from the courts (at
para. 63). It is a matter of the “delimitation of the scope of constitutional guarantees that
Canadians enjoy”, which cannot be permitted to vary from case to case (at para. 64). This
was a narrow approach, applicable in this case because the “scope” of a constitutional
right was at issue (at para. 66) given that the arbitrator had not explicitly addressed s. 8:
Ontario public school teachers are protected from unreasonable search and seizure
in their place of employment under s. 8 of the Charter. Despite their apparent
functional resemblance, a right to a reasonable expectation of privacy that is
entrenched in the Constitution is distinct in source and nature from an arbitral
right to privacy. For one, state actors cannot disavow their constitutional
obligations no matter the terms of the collective agreement. At its core, the
arbitrator’s reasons disclosed a fundamental error because she had the wrong right
in mind. The arbitrator ought to have applied the Charter, but failed to do so. Once
she failed to appreciate the constitutional dimension of the searches conducted by
the principal, there was no intelligible way for her to continue the analysis while
fully engaging with the gravity of the alleged violations of the Charter right at
issue. Courts cannot dilute the sacrosanct nature of Charter rights by accepting a
different substitute. Nor can courts supplant the reasons proffered by the decision-
maker and read the reasons as if it applied a Charter right when in fact it applied
a different right (at para. 68).
The narrowness of this approach can be appreciated by contrasting Rowe J’s approach
with that of the Ontario Court of Appeal.
There, as noted above, Huscroft JA found that “whether the grievors had a reasonable
expectation of privacy is a question of law that is subject to review for correctness” whilst
affording deference to the arbitrator’s findings of fact (2022 ONCA 476, at para. 37). In
defence of this broader approach, one might note that in recent cases involving fact- and
context-sensitive determinations of challenges to the constitutionality of legislation —
mixed questions of fact and law — the Supreme Court has applied exactly the standard
described by Huscroft JA (see especially Société des casinos du Québec inc. v. Association
des cadres de la Société des casinos du Québec, 2024 SCC 13, at paras. 94-97).
But Rowe J declined to endorse this approach (at para. 67). And it should be noted that
the Casinos case involved a challenge to legislation, unlike York Teachers, which involved
a challenge to administrative action. Correctness review encompasses mixed questions of
law and fact in cases like Casinos but it does not stretch so far in cases like York Teachers.
In sum, Rowe J does not seem to have required correctness review for all s. 8 cases, just
those cases in which the “scope” of s. 8 is in issue. His conclusion was that the arbitrator’s
failure to apply the proper legal framework was a “fatal” error justifying the quashing of
her decision (at paras. 69, 95). The majority did not, therefore, need to take a position on
whether reasonableness review applies to the application of a Charter right. Accordingly,
I do not think York Teachers can be read as changing the standard of review of
individualized decisions applying the Charter in specific statutory settings (settled
since Doré v. Barreau du Québec, 2012 SCC 12).
His concurring colleagues Karakatsanis and Martin JJ applied the reasonableness
standard to the arbitrator’s decision, essentially because the issue before the arbitrator
involved “an application and assessment [of privacy rights] which heavily depended on
the specific factual and statutory context” (at para. 112). As they noted, s. 8 is heavily
context-sensitive and applies differently in different settings (at para. 115). The fact that
the arbitrator did not specifically reference s. 8 could not justify quashing her decision (at
para. 118), as she had meaningfully “engaged with the concepts of reasonable expectation
of privacy, diminished reasonable expectation of privacy, plain view, and biographical
core” all central to the s. 8 jurisprudence to which, moreover, she had been referred by
the parties (at para. 117). This meant that this was a case of application of s. 8 rather than
determination of its scope (at para. 122) and, accordingly, it was not necessary (and
would, indeed, be inappropriate) for the courts to provide a final and determinate answer
applying the correctness standard:
[I]ndividualized decisions involving the application of the Charter that are
intrinsically linked to a specific factual and statutory context will generally not
engage the same rule of law concern about potential inconsistency as that which
motivated the correctness exception for “constitutional questions” in Vavilov. As
Professor Daly explains, “variations between individualized decisions about the
appropriate application of the Charter in a particular regulatory setting do not
compromise the integrity of the legal system: different balances may perfectly
legitimately be struck in different areas of regulation between individual rights and
the public interest” (p. 347). This explanation reflects the conceptual
underpinnings of reasonableness review, with its overall policy of deference and
the recognition that courts do not possess a monopoly over the adjudication
of Charter-related issues in the administrative context. Administrative decision-
makers are empowered and required to consider the Charter in exercising their
statutory functions (Conway, at para. 78).
Determining the engagement and scope of Charter rights will sometimes entail a
highly context specific exercise, which this case exemplifies. Not only will a search
in an employment setting differ from a search executed by police officers in the
course of a criminal investigation, the questions the arbitrator had to answer were
heavily interconnected and dependent on the particular factual and statutory
context (see I.F., Attorney General of Canada, at para. 20). For instance, did the
principal interfere with the Grievors’ reasonable expectation of privacy such that a
search within the meaning of s. 8 occurred? Was the search reasonable in that it
was authorized by law and conducted reasonably? In answering these questions,
the arbitrator had to consider workplace realities in an educational setting, the
interpretation of the Education Act, R.S.O. 1990, c. E.2, and the authority it
confers to conduct workplace searches in schools, and the reasonableness of the
principal’s exercise of authority at a particular place and time. Deference should
be afforded to the arbitrator’s understanding of this critical, case-specific context,
one in which there is no pressing need for “judicially imposed uniformity” (see P.
Daly, “Unresolved Issues after Vavilov” (2022), 85 Sask. L. Rev. 89, at pp. 106-7)
(at paras. 122-123).
Here, “the arbitrator’s decision and interpretive exercise were highly fact-specific,
depended on a particular statutory context, and concerned the application of legal
principles to the particular grievance presented” (at para. 127).
The standard of review debate in this case therefore involved a very fine distinction
between scope and application. I have observed in previous scholarship (the “Unresolved
Issues” piece cited above) that the Supreme Court has regularly applied the correctness
standard to determinations of the scope of Charter rights. I readily accept that the
distinction is liable to be somewhat unstable: the difference of opinion between majority
and minority is evidence of that; the different views turned primarily on how much can
be ‘read in’ to the arbitrator’s reasons based on background context (compare paras. 68,
94 with para. 117) and on competing characterizations of whether the arbitrator’s position
depended on her view of the scope or application of s. 8. Instability is particularly evident
in respect of s. 8, because the scope of the right is determined in large part by the context-
sensitive concept of a ‘reasonable expectation of privacy’, meaning
that scope and application bleed one into the other. But as mentioned above, I am not
persuaded that the majority’s approach here effects or portends any change in the
approach to judicial review of administrative decisions that allegedly violate the Charter.
(ii) Reasonableness Review, Legal Constraints and s. 8
This leads me to the majority and minority approaches to the implications of s. 8 for
administrative decision-making. It is useful to begin with the minority’s proposition that
the arbitrator’s decision “is not reasonable in light of the constraints bearing on it” (at
para. 132). The particular flaw, for Karakatsanis and Martin JJ, was that the arbitrator
focused too much attention on the content of the log accessed by the principal (at paras.
137-141): the arbitrator neglected the “principle of content neutrality” (at para. 138); the
arbitrator failed to appreciate how the “s. 8 jurisprudence makes clear…[that] what
mattered was the potential for the search to reveal information touching on the Grievors’
biographical core” (at para. 139); and the arbitrator’s analysis was “not reasonably
reflective of privacy’s normative aspirations” as reflected in the jurisprudence (at para.
140). As such, the minority’s position was that the arbitrator’s decision was untenable in
view of the applicable legal constraints, principally Supreme Court jurisprudence on s. 8.
Rowe J’s approach for the majority was strikingly similar. Just as his concurring
colleagues do, Rowe J used the language of “constraint”:
Where a Charter right applies, an administrative decision-maker should perform
an analysis that is consistent with the relevant Charter provision. Administrative
tribunals are empowered — and, for the effective administration of justice, called
upon — to conduct an analysis consistent with the Charter where a claimant’s
constitutional rights apply … the Charter and relevant s. 8 jurisprudence
were legal constraints that applied to the arbitrator’s decision (Vavilov, at
para. 101). In other words, the arbitrator was required to decide the grievance
consistent with the requirements of s. 8 (at para. 91, my emphasis; see also, at para.
4).
As noted above, Rowe J did not go on to perform a full-fledged s. 8 analysis. Had he done
so, however, it is quite likely that he would have ended up in the same place as his
concurring colleagues, concluding that the arbitrator’s decision was unlawful for failure
to comply with relevant legal constraints, namely s. 8 and associated jurisprudence. (To
coin a phrase, this might have been “disguised reasonableness review”.)
(iii) Charter Rights and Charter Values
There is one final point to make here about the distinction between Charter rights
and Charter values. Pace critics such as Dr Sirota and Professor Mancini, York
Teachers does not contradict the recent decision in CSFTNO. In understanding these two
decisions, it is important to appreciate the rights/values distinction. I put the point as
follows in my paper defending CSFNTO:
Charter values and Charter rights have different functions, with values operating
essentially as relevant considerations to be grappled with by an administrative
decision-maker and rights imposing harder legal constraints on decision-
makers…Constitutional text (rights) binds by imposing hard constraints on
decision-makers. Constitutional purposes (values) simply require certain matters
to be taken into account by administrative decision-makers (and it is open in
principle to the administrative decision-makers to formulate their own views on
these matters).
Rowe J’s approach here, in my view, is perfectly consistent with the rights/values
distinction and, indeed, strengthens it by framing rights as constraints that
administrative decision-makers must comply with (see above and at para. 88). Similarly,
the minority judges treat s. 8 as a hard constraint on the arbitrator. Simply put, on both
the majority and minority approach, it was not enough for the arbitrator to conduct an s.
8-lite analysis, rather she had to demonstrate compliance with the constraints of s. 8.
What is the relevance (if any) of Charter values in cases where Charter rights are
engaged? Charter rights impose additional justificatory burdens on decision-makers.
Where a Charter right is engaged, as in York Teachers, there is no need for a decision-
maker to reason in terms of Charter values because writing reasons that satisfy the
exigencies of Charter rights will also by definition discharge the administrative decision-
maker’s procedural duty to take Charter values into account: compliance
with Charter values is an inherent part of compliance with Charter rights (see Lauzon v.
Ontario (Justices of the Peace Review Council), 2023 ONCA 425); decision-makers can,
of course, reason in terms of values rather than in terms of rights and, as long as their
decision is consistent with the infringed Charter right there will be no reason for judicial
intervention. This may play out slightly differently in cases where the decision-maker
does not give reasons (as with a by-law): in such instances, the decision must comply with
a Charter right and the record must reflect consideration of Charter values. But where
reasons are given and a Charter right is engaged, the decision-maker’s focus must be on
demonstrating through its reasons the compliance of the decision with the Charter.
Where no Charter right is engaged, as in CSFTNO, the duty to take account
of Charter values applies with full force but there is — evidently — no duty to comply with
the Charter right.
In summary, then, this is a very interesting decision, most important for its holding on
the applicability of the Charter to public school boards (to which I will return) but also
significant in developing the Charter rights/values distinction. As for the standard of
review, the case is a slightly unusual one, as the decision-maker did not overtly consider
the relevant Charter right, and I am not persuaded that it will have a lasting impact
outside circumstances where a decision-maker simply has not considered a right at all in
circumstances where it ought to have.
(iv) Counsel in Regulatory Investigations
Under the broad heading of the Charter, it is useful to consider the decision in
Neustaedter v Alberta (Labour Relations Board), 2024 ABCA 238, where the issue was
the ability of a regulated entity to insist on the presence of counsel during a regulatory
investigation. Generally speaking, regulatory statutes provide for wide investigative
powers in respect of regulated activities (here, occupational health and safety), including
the ability to enter on premises and interview staff.
Here, the regulatee objected to interviews with staff being conducted in the absence of
counsel. This objection was rejected by the occupational health and safety officer assigned
to the matter and by the Board. The Court of Appeal was of the same view:
[T]he appellants argue, on a proper interpretation, OHS officers did not have
authority to compel interviews. The OHS officer came to a different conclusion. He
noted section 51(j) of the OHSA expressly gave officers the authority to interview
and obtain statements for the purposes of the Act (“For the purposes of this Act,
an officer may… interview and obtain statements…”), section 53(2) mandated that
witnesses comply with an OHS officer’s request for information (“shall, on the
request of an officer, provide to the officer any information respecting the injury
or incident that the officer requests”), and section 54 required witnesses to
cooperate (“No person shall interfere with or in any manner hinder an
occupational health and safety officer… who is exercising powers or performing
duties or functions under this Act.”). He also noted, with reference to Ebsworth,
that OHS had the authority to determine its own procedure as necessary to carry
out its legislated function. He concluded from the foregoing that the OHSA gave
“an OHS Officer the power to compel a witness to attend an interview for the
purpose of requesting information pursuant to section 53(2)”. The ALRB
characterized the officer’s reasoning in this regard as “coherent, rational and
justified”: ALRB Decision at para 54. We agree.
Before the chambers judge, the appellant advanced a variation of this argument by
submitting OHS officers have no authority to compel a person to provide information. At
paragraph 105 of the Substantive Decision, the chambers judge held the appellant’s
interpretation “would render the OHSAessentially toothless. If a request for information
regarding a workplace incident can be ignored with impunity, OHS’s mandate to protect
worker safety would be rendered nugatory. This cannot have been the Legislature’s
intention.” We agree and extend this reasoning to the argument before us (at paras. 16-
17).
The appellant had raised the Charter in support of the argument that interviews could
not be conducted in the absence of counsel but this was to no avail. The Charter simply
does not apply with significant force in respect of regulated activities (see e.g. here).
B. Mixed Questions of Law and Fact
In Société des casinos du Québec inc. v. Association des cadres de la Société des casinos
du Québec, 2024 SCC 13, the Supreme Court of Canada dealt with a number of important
issues that are significant for the law of judicial review of administrative action and for
regulation more broadly. Most significantly, the Court gave guidance on the resolution of
mixed questions of law and fact arising in the correctness categories established by
Vavilov. A clear marker has been laid down: on constitutional questions and other issues
requiring correctness review, there is no deference to the decision-maker’s application of
law to fact. Beyond this, Société des casinos provides guidance about positive and negative
rights, and the role of tribunals in resolving constitutional issues.
The case involved a claim for certification by casino managers based in Quebec.
Certification would allow them to bargain as a group with their employer. But managers
are excluded from the provincial collective bargaining legislation. So the managers
invoked s. 2(d) of the Charter of Rights and Freedoms, which protects freedom of
association. Applying the context-sensitive Charter test to the facts, the expert tribunal
concluded that the exclusion of the managers represented a “substantial interference”
with their freedom of association. The remedy was that the legislative exclusion was
inoperable as applied to the managers who, accordingly, were entitled to go ahead with
their claim for certification. This conclusion was, ultimately, upheld by the Court of
Appeal albeit that the effects of the decision were suspended for 12 months to allow for
legislative intervention.
However, the Supreme Court allowed the appeal, an outcome on which the seven judges
who heard the case were agreed, concluding unanimously that there was no “substantial
interference” with associative freedom. In doing so, they addressed: (1) the standard of
review for mixed questions of fact and law in the ‘constitutional questions’ category of
correctness review; (2) whether there are distinct tests or standards applicable to Charter
claims where a positive or negative right is being asserted; and (3) the jurisdiction of
administrative tribunals to grant remedies relating to the inoperability of legislation. The
third question was addressed only by Côté J in her concurring reasons but the others were
treated by all seven (Jamal J for the majority and Rowe J concurring separately).
(i) Confirming Correctness on Mixed Questions
The Court of was unanimous on standard of review. Côté J addressed the point in detail
and the others agreed with her analysis (at paras. 45, 199). That analysis is brief and to
the point. Here is what she said:
In this case, determining whether the exclusion from the L.C. regime constitutes
substantial interference with the freedom of association of the Association’s
members is not a simple question of fact. Such an inquiry involves weighing “the
constitutional significance” of the findings of fact made on the basis of the
members’ situation by reference to freedom of association (Westcoast Energy, at
para. 39). To some extent, this amounts to defining the constitutional standard of
“substantial interference”.
The definition of this standard requires a determinate and final answer (Vavilov,
at paras. 53 and 55). In Westcoast Energy, cited with approval in Vavilov, at
para. 55, our Court noted that no deference is owed in respect of questions of mixed
fact and law that arise in connection with a constitutional question because it is
important that constitutional questions be answered correctly (paras. 39-40).
It follows that the Superior Court did not owe deference to the ALT’s findings of
law and findings of mixed fact and law, but only to the findings of fact made by that
tribunal.
A reviewing court must show deference to findings of pure fact that can be isolated
from the constitutional analysis (Consolidated Fastfrate Inc. v. Western Canada
Council of Teamsters, 2009 SCC 53, [2009] 3 S.C.R. 407, at para. 26). Such
deference to findings of this kind is based on considerations related to “judicial
efficiency, the importance of preserving certainty and public confidence, and the
relatively advantageous position of the first instance decision maker” (Vavilov, at
para. 125). The rule of law does not require that there be a determinate and final
answer to questions of pure fact, as they will vary from case to case (at paras. 94-
97).
The parties agreed (see para. 93) that this case fell into the ‘constitutional question’
category set out in Vavilov: the question was the Charter compliance of the legislative
exclusion of the managers and so, indeed, the correctness standard properly applied, as
the answer to this question should be given definitively by the courts. Evidently, however,
the parties did not agree on what exactly the correctness standard should apply to.
The question was how much deference, if any, is due to the tribunal in this instance. The
standard of review of constitutional questions, such as the consistency of the statutory
exclusion with the Charter, is correctness. But what about the factual determinations
underpinning the tribunal’s analysis? And, in any event, can the tribunal’s interpretation
of the statutory exclusion and the Charter be separated from its factual analysis?
The Supreme Court’s prior jurisprudence had not been especially clear on this point. On
the ‘no deference’ side of the ledger is Westcoast Energy Inc. v. Canada (National Energy
Board), [1998] 1 SCR 322, a case about whether a pipeline and related facilities
constituted a federal undertaking. Iacobucci and Major JJ were sceptical about whether
deference would be appropriate on the Board’s application of law to the facts before it:
even questions of mixed law and fact are to be accorded some measure of
deference, but this is not so in every case. It would be particularly inappropriate to
defer to a tribunal like the Board, the expertise of which lies completely outside the
realm of legal analysis, on a question of constitutional interpretation. Questions of
this type must be answered correctly and are subject to overriding by the courts. It
seems reasonable to accept the proposition that courts are in a better position than
administrative tribunals to adjudicate constitutional questions. It is interesting to
note that this particular panel’s professional training was not in law. So, although
the question here was one of mixed law and fact, it follows that the Board was not
entitled to deference because of the nature of the legal question to be answered (at
para. 40).
However, Iacobucci and Major JJ went on to observe that the case turned not on the
“Board’s conclusions as to the different activities carried on by Westcoast” but on the
“constitutional effect” of the conclusions (at para. 42). In other words, there was no attack
on the Board’s findings of fact. And when it came to apply the law to those findings, the
correctness standard was appropriate. Accordingly, it is difficult to say that Westcoast
resolves the deference question one way or another.
In Northern Regional Health Authority v. Horrocks, 2021 SCC 42, at para. 9, albeit
addressing the ‘overlapping jurisdiction’ category of correctness review rather
‘constitutional questions’, Brown J was clear that the decision-maker had to be “correct”
on the factually suffused question of characterizing the essential character of the dispute
between the parties (as relating to labour relations or human rights adjudication). This
led me to comment that “when determining whether or not a decision about competing
jurisdictional boundaries was lawful, the decision-maker must be correct and the
reviewing court must satisfy itself, based on the record, that the decision-maker came to
the correct conclusion”.
The most prominent entry on the ‘deference’ side of the ledger is Mouvement laïque
québécois v. Saguenay (City), 2015 SCC 16, [2015] 2 SCR, where Gascon J commented at
para. 46 that deference is appropriate “where the Tribunal acts within its specialized area
of expertise, interprets the Quebec Charter and applies that charter’s provisions to the
facts to determine whether a complainant has been discriminated against” (see also
Saskatchewan (Human Rights Commission) v. Whatcott, 2013 SCC 11, [2013] 1 SCR
467). But the emphasis on expertise means that the authority of this statement has been
weakened by the downgrading of expertise in Vavilov.
Consider also Consolidated Fastfrate Inc. v. Western Canada Council of Teamsters,
2009 SCC 53, [2009] 3 SCR 407. Here, the issue was whether a company was subject to
provincial or federal regulation. A provincial labour relations board held that the
company was an interprovincial undertaking subject to federal labour relations
legislation. Rothstein J. observed that the board’s “constitutional analysis rested on its
factual findings”: “Where it is possible to treat the constitutional analysis separately from
the factual findings that underlie it, curial deference is owed to the initial findings of fact”
(at para. 26). At first glance, this might appear to be a pro-deference proposition, but on
closer inspection, it appears that deference is only appropriate where the constitutional
question is can be separated from the underlying factual findings: deference on facts in
constitutional cases, but only as long as deference does not influence legal determinations
as to constitutionality.
This issue was also discussed in Toussaint v. Canada (Attorney General), 2011 FCA 213,
[2013] 1 FCR 374, in the context of appellate review rather than judicial review. Whilst
deference is appropriate on “factual findings and exercises of discretion that are heavily
suffused with facts” (at para. 54), correctness review is often applied in constitutional
cases “probably… because of the centrality of the legal issues in such appeals, and the fact
that questions of constitutional law are often extricable from the questions of mixed fact
and law that arise” (at para. 55).
In Société des casinos, the Supreme Court was clear: when applying the correctness
standard, a reviewing court must take the findings of fact made by the decision-maker as
they are (as long as they are reasonable), but it is for the court to determine for itself the
legal effects of those findings of fact. Put another way, the legal characterization of the
facts as found by the decision-maker is a matter for the court.
Whatever about my scepticism of the law/fact distinction (see chapter 6 of A Theory of
Deference in Administrative Law), this is now definitively the law. And, to be fair, In its
post-Vavilov case law applying the correctness categories, this has certainly been the
Supreme Court’s approach. I noted Northern Regional Health Authority v.
Horrocks, 2021 SCC 42 above and would now add Sharp v. Autorité des marchés
financiers, 2023 SCC 29 (a constitutional question case): there, the decision-maker had
made findings of fact about a ‘pump and dump’ scheme run by out-of-province actors but
the application of a context-sensitive legal standard to those facts was done without any
deference to the decision-maker’s conclusion. In the correctness categories, then, pure
findings of fact — the who, what, when, where and why of adjudication — get deference
but everything else is ultimately for the court.
Of course, it is always open to a reviewing court to adopt the analysis of the decision-
maker (and perhaps this is now the best way to understand the decision in Law Society
of Saskatchewan v. Abrametz, 2022 SCC 29) but the judge retains the final word on
whether the legal standard has been met based on the facts as found.
(ii) Positive and Negative Rights
In some of the Supreme Court’s Charter jurisprudence it has distinguished (or seemed to
distinguish) between two different types of Charter claim: positive and negative. A
negative claim involves using the Charter as a protective shield against state action. A
positive claim involves wielding the Charter as a sword to slash away a barrier to
benefiting from an existing statutory regime (see Côté J’s concurring reasons at paras.
129-131).
In Canada v. Boloh 1(a), 2023 FCA 120, at para. 69 (a positive claim under s. 6 of
the Charter), the Federal Court of Appeal suggested that there is a general
positive/negative “framework” established by Supreme Court decisions such as Dunmore
v. Ontario (Attorney General), 2001 SCC 94, Baier v. Alberta, 2007 SCC 31, and Toronto
(City) v. Ontario (Attorney General), 2021 SCC 34. Dunmore was a freedom of
association case but the other two were freedom of expression cases.
Here, the majority did not accept that there is any general framework, or at least their
reasons are not compatible with the existence of a general framework. Jamal J accepted
that in respect of freedom of expression the positive/negative distinction is hardwired
into the Supreme Court’s jurisprudence (at para. 44) but sharply distinguished the
association and expression cases:
[I]n the context of claims under s. 2(b) of the Charter, the threshold for proving
positive freedom claims is substantial interference with freedom of expression
(Toronto (City), at para. 25). However, the threshold for negative rights claims
involving freedom of expression, as explained in Irwin Toy Ltd. v. Quebec
(Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927, is whether the
purpose or effect of the government action merely restrictsfreedom of expression
(Toronto (City), at para. 24, citing Irwin Toy, at p. 971, and Baier, at paras. 27-28
and 45).
In the freedom of association context…the threshold for both “positive” and
“negative” freedom of association claims is the same: substantial interference
(see Dunmore, at para. 25; Health Services, at paras. 19 and 90; Fraser, at paras. 2
and 47; Mounted Police, at para. 72; Meredith, at paras. 4 and 24-
25; Saskatchewan Federation of Labour, at paras. 2 and 25). There is not a more
stringent threshold for positive rights claims under s. 2(d). For freedom of
association claims, the “elevated threshold in the second Dunmore factor”
(Toronto (City), at para. 25) of substantial interference already applies to all claims
involving both positive and negative duties (Fudge, at pp. 545-46 and 550) (at
paras. 41-42).
The concurring judges took a different view but for my purposes the important point that
emerges from the majority reasons is that the positive/negative distinction is not a
concept of general application when interpreting the Charter. Ease of access to a
regulatory regime or to other forms of state action will not depend on a prior
characterization of a claim as positive or negative in nature. What matters is the claim
that is made and the Charter provision that is invoked.
This is not to say that the positive/negative distinction might not have purchase in
particular cases (as was the case with the mobility rights claimed in Boloh) but whether it
does or not will depend on the proper interpretation of the Charter right or freedom at
issue, in light of the claim. This is an important point for those charged with designing
regulatory regimes and for those seeking to leverage Charter rights or freedoms to gain
access to them.
(iii) Tribunals and Remedies
The only member of the Supreme Court to address the remedial jurisdiction of the
decision-maker (here, the Quebec Administrative Labour Tribunal) was Côté J.
She noted that the managers had decided to bring a claim for certification rather than to
seek a declaration of unconstitutionality in superior court and suggested that
“[p]roceeding before a superior court is preferable insofar as such a court has the power
to make a formal declaration of unconstitutionality and to suspend the declaration in
order to give the legislature all the latitude it needs to enact a particular regime that meets
the minimum constitutional requirements of s. 2(d)” (at para. 156).
In fairness to the managers, they may well have brought the claim in the tribunal to avoid
being met in superior court by the counter-argument that they should first have sought a
remedy from the tribunal (Okwuobi v. Lester B. Pearson School Board, 2005 SCC 16, at
paras. 38-45). And there is surely no doubt that the Supreme Court’s consideration of the
issues was enriched by the tribunal’s detailed analysis of the facts (see also Forest Ethics
Advocacy Association v. Canada (National Energy Board), 2014 FCA 245, at paras. 42-
47; Denton v. British Columbia (Workers’ Compensation Appeal Tribunal), 2017 BCCA
403, at paras. 51-52; Campisi v. Ontario, 2017 ONSC 2884, at para. 13).
It is true that the tribunal could not suspend any declaration of inoperability (at para.
157), but it did not seek to do so. It was the courts that suspended the effect of the
tribunal’s decision, on the theory that on the standard of correctness the judges were
stepping into the shoes of the tribunal (cf. Okwuobi at para. 45). I remain sceptical: the
advantage of the decision could only apply to the particular managers who brought the
certification application so the need for broad legislative consideration of the regulatory
regime (the usual justification for suspending a declaration of unconstitutionality) is not
particularly keenly felt. Côté J correctly noted this point, but I do not think it can mean
that the tribunal should not have been seized of the matter in the first place.
For my part, I have sympathy for the managers’ choice of forum and, given the need for
extensive fact-finding, I suspect the fact that they first went to the tribunal ultimately
facilitated the Supreme Court’s comprehensive analysis of the important freedom of
association issues.
In the end then, this is an important decision on standard of review and regulatory design,
with interesting points made too about forum choice in cases arising at the intersection
of administrative law and constitutional law.
III. Justiciability
A. The Public-Private Divide
In its 2018 decision in Highwood Congregation of Jehovah’s Witnesses (Judicial
Committee) v. Wall, 2018 SCC 26, [2018] 1 SCR 750, the Supreme Court of Canada sought
to clarify the approach to the scope of judicial review. Rowe J wrote, for a unanimous
court, that “[j]udicial review is only available where there is an exercise of state authority
and where that exercise is of a sufficiently public character” (at para. 14).
On a plain reading of this sentence, two criteria must be met for a court to consider an
application for judicial review: the decision-maker at issue must be ‘exercising state
authority’ and the exercise must be ‘of a sufficiently public character’. It is not clear,
however, what “state authority” means, nor is the difference between “state” and “public”
explained.
In a comment on the Wall decision, I noted that the Supreme Court’s references to “state”
and “public” were conclusory and warned that “future courts will struggle to draw the line
between “public” and “private” decision-making and identify those exercises of “state”
authority that ought to be subject to judicial review remedies” (“Right and Wrong on the
Scope of Judicial Review: Wall v Highwood Congregation” (2018) 31 Canadian Journal
of Administrative Law & Practice 339, at p. 343).
In my view, there is really only ever one relevant question in cases about the scope of
judicial review of administrative action: is the decision at issue sufficiently public in
character to justify court oversight? (See further Daly, Understanding Administrative
Law in the Common Law World (OUP, 2021), at p. 203.) In answering this question, it is
helpful in my view to have regard to the factors set out by Stratas JA in Air Canada v.
Toronto Port Authority, 2011 FCA 347, [2013] 3 FCR 605, at para. 60:
i. the character of the matter for which review is sought;
ii. the nature of the decision-maker and its responsibilities;
iii. the extent to which a decision is founded in and shaped by law as opposed to
private discretion;
iv. the body’s relationship to other statutory schemes or other parts of
government;
v. the extent to which a decision-maker is an agent of government or is directed,
controlled or significantly influenced by a public entity;
vi. the suitability of public law remedies;
vii. the existence of a compulsory power; and
viii. an “exceptional” category of cases where the conduct has attained a serious
public dimension.
These factors can help us sort the state/public from the private. The difficulty is that,
in Wall, the Supreme Court sought to minimize the role of the Air Canada factors (at
para. 21). Two recent appellate authorities have reinvigorated the Air Canada factors and
both involve interesting applications of the principles about the scope of judicial
review: Khorsand v. Toronto Police Services Board, 2024 ONCA 597; and Nova Scotia
Health Authority v. Finkle and West, 2024 NSCA 87.
In Khorsand, Fairburn ACJ thoroughly reviewed the post-Wall case law and academic
literature. Adopting Professor McKee’s terminology, she distinguished between the
institutional (state) and functional (public) aspects of the decision in Wall and concluded
as follows: “Wall does not preclude reference to the Air Canada factors in teasing out
why, at a minimum, the functional criterion is or is not met when determining whether a
decision is public or not” (at para. 73).
In my view, this statement must be correct. Personally, I would go further, merge the
institutional and functional questions and use the Air Canada factors to provide an
answer to one single question I posed above: is the decision at issue sufficiently public in
character to justify court oversight?
In terms of applying the Air Canada factors, Fairburn ACJ makes two other astute points.
First, she adopts Groberman JA’s analysis from Strauss v. North Fraser Pretrial Centre
(Deputy Warden of Operations), 2019 BCCA 207, at para. 42:
In the end, the factors are merely guidelines in deciding whether a decision made
by a public official or tribunal has a sufficiently public character to be amenable to
judicial review. Some will be applicable and important in particular contexts while,
in those contexts, others may be irrelevant and unhelpful.
This seems entirely correct to me.
Second, she observes that in Wall, the Supreme Court was concerned to eliminate the last
factor on Stratas JA’s list of factors in Air Canada, as the judges were evidently concerned
that it was being used to justify judicial review in areas from which the courts should
generally steer clear. Citing paras. 20-21 of Wall, where Rowe J held that a series of lower-
court decisions expanding the scope of judicial review had been wrongly decided, she
observes:
This passage makes clear that it is wrong to apply the Air Canada factors to
transform the decision of a private actor – such as a church, sports club, or other
voluntary association – into a public decision. In my view, the passage also
cautions against characterizing a decision of a public body as public in function
simply because a broad segment of the public may be interested in or impacted by
it. For instance, a government decision to enter into a contract to purchase
property may be of significant interest to, and have an impact on, a broad segment
of a community; however, that would not transform the contractual decision into
a public one. In other words, it is important to distinguish between “public” in the
generic sense and “public” in the sense that the legality of state decision making is
at play (at para. 76).
As this passage suggests, it is important to set Wall in context. This was the key concern
of the Supreme Court. The use of an institutional criterion and a functional criterion was
designed to ensure that the scope of judicial review was not enlarged to include matters
inappropriate for court oversight. But this also means, in my view, that if the Air
Canada factors (minus the last one) can respond to the Supreme Court’s key concern,
they should be used. And, in my view, they can indeed be applied in a way that is sensitive
to that concern.
Khorsand might be a case in point. K wanted to become a special constable with the
Toronto Community Housing Corporation. To do so, K needed to pass a security
screening conducted by the Toronto Police Service (or more specifically, its Talent
Acquisition Unit). He failed to pass: neither the TPS nor the TCHC provided any reasons;
documents disclosed after a freedom of information request revealed that there had been
9 interactions between K and the TPS (none of which identified any criminal behaviour)
and that K had been described as “Brown”, “Middle Eastern”, or “Persian”. Reading
between the lines, K suspected he had been discriminated against on the grounds of race.
K sought judicial review on the grounds of procedural unfairness and unreasonableness.
But Fairburn ACJ held that the screening decision was not amenable to judicial review.
First, although the power to appoint someone as a special constable is statutory, the pre-
screening decision here was at one remove from the appointment power. The decision
was taken under a Memorandum of Understanding between the TPS and TCHC, the
provisions of which:
oblige the TCHC – the actual employer of TCHC special constables – to satisfy itself
of a community patrol officer applicant’s suitability before it can advance that
person for appointment as a special constable. Indeed, the TCHC must already
have hired the person, trained them, and ensured that “all screening tests” have
been conducted “by or on behalf of the TCHC” at its own expense, before it may
even put the individual forward “to the Board for appointment as a special
constable.” Under s. 15 of the MOU, the TCHC “shall not put forward to the Board
an applicant for appointment as Special Constable” where the TCHC is not
satisfied with “the results of the background investigations.” In other words, the
pre-screening decision was made by the TAU at the request of the TCHC as part of
its hiring process (at para. 81).
The hiring decision was, therefore, to be made before the TPS became involved (at para.
82).
Second, the majority of the court below had expressed concerns about systemic
discrimination and saw a role for judicial review as a matter of public importance.
Following Wall, however, “these concerns cannot transform a discretionary employment-
related decision into a public decision” (at para. 90). Fairburn ACJ was also not convinced
that the references to K’s ethnicity in the disclosed documents were, without more,
enough to demonstrate racial discrimination (at paras. 91-97).
Third, public law remedies were not suitable here. Judicial review meant disclosure,
“subject only to an undefined process to protect sensitive law enforcement information”
(at para. 103). But this risked undoing the carefully calibrated statutory regimes for
disclosure (at para. 101). Furthermore, judicial review, and concomitant disclosure, could
have a chilling effect:
The focus of the majority’s reasons below is on the disclosure of pre-existing
records contained in police databases, even those over which the police service has
no control. It is important to keep in mind though that background investigations
can extend well beyond information contained in police databases. Indeed, those
applying for a position with TCHC must sign an authorization allowing the TPS to
conduct a background investigation that extends well beyond police databases. For
instance, in this very case, Mr. Khorsand signed an authorization that permitted
any person or organization in receipt of it to provide disclosure of information
about him to the Toronto Police Service, including the conveying of “opinions” that
they may have.
This is akin to a more classic job reference check and something that may be
recorded in the notes of the person conducting the check. If such notes informed
the reason for Mr. Khorsand’s failure, the majority’s reasons dictate that the notes
would have to be disclosed. In my view, that would place a chill on receiving honest
and objective feedback about a potential TCHC community patrol officer (at paras.
107-108).
For these Air Canada-type reasons (see iii and iv above), judicial review was not
appropriate.
The Nova Scotia Court of Appeal reached the opposite conclusion in Finkle and West.
Here, the issue related to disciplinary action taken against doctors by the Authority. The
Authority argued that the relationship was contractual and thus not subject to judicial
review. However, Fichaud JA found that the matter was subject to judicial review. To
begin with, there was state action as the Authority used the doctors’ privileges — set out
in legislation and regulations — as “leverage” to impose disciplinary measures (at para.
79). This provided a “legislative nexus” (at para. 70) and led to the conclusion that the
Authority “exercised state authority” (at para. 77). These can easily be related to the Air
Canada factors about the nature of the decision-maker, its relationship to government
and the exercise of compulsory authority.
In addition, Fichard JA applied the Air Canada factors (at para. 85, adopting Khorsand)
to determining whether this exercise of state authority was public:
In this case, the motions judge discussed those factors from Air Canada’s list that
the judge determined to be applicable to the circumstances (paras. 91-94). Justice
Keith found (1) this was not a private matter, as the NSHA exercised authority
further to subordinate legislation under the Health Authorities Act, rather than
applying contractual or business considerations, (2) the NSHA’s power was
compulsory for physicians with privileges and (3) a public law remedy was suitable.
The judge’s analysis displays no error of law (at para. 86).
The bottom line, then, is that the Air Canada factors were useful to both appellate courts
in determining the scope of judicial review. Although Fichaud JA analyzed the existence
of “state” authority without reference to the factors, they can be understood to have
influenced his analysis, and he clearly used them on the question of ‘publicness’. Again,
in my view, only one question need be answered, and it can be answered by reference to
the Air Canada factors. This was the path Fairburn ACJ followed in Khorsand.
Regardless of the question posed, the answers given here are very interesting and quite
instructive about the difficulties that arise on the edges of the public/private divide.
B. Reviewability of Guidance
Two recent Canadian cases have dealt with the reviewability of soft law instruments and,
in both instances, the courts came out against judicial review.
In Harold the Mortgage Closer Inc. v. Ontario (Financial Services Regulatory Authority,
Chief Executive Officer), 2024 ONSC 4464, the applicants challenged guidance issued by
the Authority. Under the guidance, the Authority publishes details of enforcement action
on its website. A licensee subject to enforcement action may resist the action in a de novo
hearing before the Financial Services Tribunal. Here, the applicants complained, first,
that publication of the notice of enforcement action caused reputational harm and,
second, that the Authority should have posted the licensees’ response on its website.
Backhouse J held that the guidance was not justiciable. The Authority was not exercising
a statutory power of decision in adopting the guidance and the guidance did not affect any
of the applicants’ legal rights and obligations:
FSRA is not specifically required or empowered by statute to issue the
Transparency Guidance or publish the NOP. In this case, the Decisions were to
provide a non-binding guidance document on FSRA’s administrative processes
and to publish (or not publish) documents on FSRA’s website. Section 3 of
the FSRA Act provides FSRA’s statutory objects: the goals FSRA strives to achieve.
Section 3 does not confer any jurisdiction, authority, or a statutory power of
decision upon FSRA. Section 6 provides FSRA’s natural person powers, empowers
FSRA to administer and enforce legislation, and prohibits FSRA from establishing,
acquiring, or dissolving subsidiary corporations. While the Transparency
Guidance states that the policy achieves FSRA’s statutory objects, neither ss. 3 nor
6 confer any authority or obligation on FSRA to publish NOPs and FSRA does not
rely on either section to do so.
…
Although the applicants have an interest in their reputation, the publication of
allegations by the regulator does not give rise to a right to judicial review. The
Decisions here do not affect the legal rights, interests, property, privileges, or
liberty of the applicants. The Transparency Guidance issued by FSRA simply
describes when and how FSRA will publish documents related to its enforcement
proceedings. Reputational damage in the circumstances of this case does not give
rise to a right of judicial review (at paras. 70, 75).
Note that I provided some consultancy services to the Authority in respect of this matter.
The Federal Court of Appeal arrived at the same conclusion in Air Passenger Rights v.
Canada (Attorney General), 2024 FCA 128. The issue here related to the publication by
the Canadian Transportation Agency of a statement on its website at the outset of the
COVID-19 pandemic. There was a wave of flight cancellations and significant concern in
the airline industry about the economic consequences of shutting down international
travel. In relevant part, the statement read:
While any specific situation brought before the CTA will be examined on its merits,
the CTA believes that, generally speaking, an appropriate approach in the current
context could be for airlines to provide affected passengers with vouchers or credits
for future travel, as long as these vouchers or credits do not expire in an
unreasonably short period of time (24 months would be considered reasonable in
most cases).
The applicant here recognized that the statement was not a “decision” but nonetheless
argued that judicial review was appropriate because “(a) the Statement is a pre-judgment
by the CTA of air passengers’ rights to refunds for cancelled flights, and (b) the CTA acted
in response to improper third-party influence in formulating and posting the Statement
contrary to its Code of Conduct, giving rise to reasonable apprehension of bias” (at para.
14).
Walker JA held that judicial review was unavailable. She distinguished the statement
from cases addressing the lawfulness of guidelines that, though notionally non-binding,
had a mandatory character (at para. 23) and held that the statement itself did
not cause any affect on rights and obligations. First, it was the actions of third parties,
such as credit card companies, that affected passengers’ rights and obligations:
At its core, APR’s argument that the Statement is justiciable is based on the
premise that the actions of third parties (airlines and credit card companies) taken
in reliance on the Statement prejudicially affected air passengers’ rights and access
to refunds for cancelled flights in circumstances where refunds should arguably
have been available to them. APR insists that the Statement had the practical effect
of facilitating the airlines’ retention of passengers’ money without providing
services.
APR’s evidence and arguments are not persuasive. The actions of third parties are
not the actions of the CTA, nor is the CTA responsible for the decisions taken by
airlines and credit card companies. APR’s evidence demonstrates only that third
parties used the Statement to justify refund and credit card chargeback refusals.
The prejudicial effects asserted by APR flow not from the Statement or the conduct
of the CTA but from the interpretation and use of the Statement by third parties.
APR asks the Court to consider the Statement from the public’s perspective but
there is little evidence in the record of that perspective outside of a limited number
of email chains in which frustrated air travellers vented their dissatisfaction with
the Statement. In any event, the public’s possible interpretation of the Statement
does not establish prejudicial effect or justiciability (at paras. 29-30).
Second, the statement itself was non-binding:
Third-parties’ mischaracterization of the Statement, whether as a ruling or
approval, was not endorsed by the CTA and does not transform the Statement into
a mandatory pronouncement. The Statement is written in simple language and
conveys a possible way forward in unprecedented circumstances, subject to the
adjudication of each case on its own merits. It is drafted using permissive language
and addresses one topic. It does not purport to provide a detailed overview of the
state of Canadian legislation and jurisprudence regarding the right to refunds, nor
does the Statement alter an air passenger’s legal entitlement to a refund for certain
cancelled flights. Although APR asserts that the Statement misinforms the
travelling public about their refund rights, it has pointed to no requirement that
the CTA reference the relevant refund legislation, tariff and case law when issuing
an interim statement that makes clear reference to travellers’ ability to file a
complaint despite the guidance in the Statement (at paras. 31-32).
The analysis here is strikingly similar to that of Backhouse J in the Ontario case.
However, the applicant had another string to its bow, in the form of an argument that
judicial review is always available where procedural fairness is put in issue, especially
where a reasonable apprehension of bias is alleged. This too was rejected by Walker JA:
This Court does not have plenary jurisdiction to intervene in the conduct of a
federal board, commission or tribunal based on allegations of misconduct or
perception of bias absent a matter in respect of which a remedy is available.
Essentially, APR is asking the Court to censure the CTA regardless of the legal
effects of its conduct. This is not the Court’s role. At the admitted risk of repetition,
for a remedy to be available a matter must “affect legal rights, impose legal
obligations, or cause prejudicial effects” (Democracy Watch at para. 29). The
Statement does not do so and it is not otherwise amenable to judicial review (at
para. 44).
There is something to be said for the applicant’s argument on this point. Allegations of
bias have sometimes been given special treatment as far as reviewability is concerned (see
e.g. Fundy Linen Service Inc. v. Workplace Health, Safety and Compensation
Commission, 2009 NBCA 13). This point merits further consideration. However, if an
artful pleader were able to allege a reasonable apprehension of bias in order to circumvent
the well-established principles set out by Backhouse J and Walker JA, I am not sure this
would be a desirable development in the law of judicial review.
IV. Constitutional Issues
A. Constitutional Foundations of Judicial Review
This year, the Supreme Court of Canada handed down its much-anticipated decision
in Yatar v. TD Insurance Meloche Monnex, 2024 SCC 8. As expected (by me at any rate),
the Court reversed the approach below and (correctly, in my view) described the role of
discretion in judicial review proceedings.
In reasons written by Rowe J, the Court thereby confirmed that the fact that a right of
appeal is limited to questions of law does not prevent an individual from judicially
reviewing factual and other issues that would not fall within the scope of the right of
appeal. Moreover, although the Court did not grapple with some of the other issues
related to limited rights of appeal, such as privative clauses, its overall analysis gives me
the impression that, when confronted with these issues, the justices will favour judicial
review.
At root, Yatar is a case about accident benefits. There is a right of appeal on questions of
law from decisions of the Licence Appeal Tribunal. But Yatar wanted to raise an issue of
fact or mixed fact and law. Accordingly, she sought judicial review. Readers will recall that
the Court of Appeal (and the Divisional Court) held that given the existence of a limited
right of appeal, and the evident desirability of efficient resolution of accident benefits
claims, judicial review should only be permitted in rare cases.
This was the central issue on appeal to the Supreme Court of Canada, which resoundingly
restated the importance of judicial review. One of the “first principles” of Canadian public
law (at para. 45) is the “importance” of judicial review (at para. 46). Accordingly, “[w]hile
there is discretion to hear the application on the merits and deny relief, this discretion
does not extend to decline to consider the application for judicial review (at para. 49)”.
Judicial review is always available:
When an applicant brings an application for judicial review, a judge must consider
the application: that is, at a minimum, the judge must determine whether judicial
review is appropriate. If, in considering the application, the judge determines that
one of the discretionary bases for refusing a remedy is present, they may decline to
consider the merits of the judicial review application (Strickland, at paras. 1, 38
and 40; Matsqui, at para. 31). The judge also has the discretion to refuse to grant
a remedy, even if they find that the decision under review is unreasonable (Khosa,
at para. 135; Strickland, at para. 37, quoting Minister of Energy, Mines and
Resources, at p. 90) (at para. 54).
An interesting question here is whether some grounds for the exercise of discretion might
preclude consideration of the merits of a judicial review application (see Budlakoti v.
Canada (Citizenship and Immigration), 2015 FCA 139, at para. 28(1)). Rowe J’s analysis
would suggest not, though there may be some cases where it is plain and obvious that a
remedy could not possibly be available.
Beyond this, a limited right of appeal does not, on its own, communicate any legislative
intention to restrict access to the courts:
The Court of Appeal erred by holding that the limited right of appeal reflected an
intention to restrict recourse to the courts on other questions arising from the
administrative decision, and that judicial review should thus be
rare. Thelegislative decision to provide for a right of appeal on questions of law
only denotes an intention to subject LAT decisions on questions of law to
correctness review. The idea that the LAT should not be subject to judicial review
as to questions of facts and mixed facts and law cannot be inferred from this (at
para. 58).
This is very much in keeping with the narrow approach to legislative intent (under the
rubric of “institutional design”) in Vavilov. Where a legislature uses certain magic words,
like “appeal”, or “patent unreasonableness”, courts must give effect to them in the way
prescribed by Vavilov. But more thoroughgoing contextual analyses of legislative intent
are forbidden (see also Mason, discussed here).
Rowe J accepted that, where there is an adequate alternative remedy, a judge may exercise
discretion to refuse to grant relief in judicial review proceedings. For judicial review to be
ousted, however, there must be “an appropriate alternative forum or remedy” (at para.
61). Here, there was no such appropriate alternative. The right of appeal was limited to
questions of law, making it impossible for Yatar to put in issue the factual questions and
mixed questions of fact and law she wished to put in issue. And the possibility of an
internal reconsideration was not an alternative either:
The access to internal reconsideration cannot be an adequate alternative remedy,
as the reconsideration decision itself is the subject of the review. Alternatives do
exist where internal review processes have not been exhausted or where there is a
statutory right to appeal that is not restricted, such that questions of law, fact, and
mixed fact and law could be considered on appeal. But, that is not so here (at para.
63).
This is, surely, absolutely right. A final decision is always reviewable regardless of the
quality or quantity of internal processes of reconsideration or review (albeit, of course,
that when these function well they might weed out unreasonable or procedurally unfair
decisions).
To the argument that judicial economy provided a good reason for exercising discretion
not to entertain applications for judicial review except in rare cases, Rowe J had a firm
response:
Judicial economy is a legitimate concern. However, the countervailing
consideration is to ensure that those whose interests are being decided by a
statutory delegate have a meaningful and adequate means to challenge decisions
that they consider to be unreasonable having regard to their substance and
justification, or were taken in a way that was procedurally unfair (at para. 65).
The thread running through this analysis and culminating in the passage just quoted is
that judicial review is a constitutional fundamental. It is critically important that
individuals have the ability to challenge administrative decisions that are alleged to be
unreasonable or procedurally unfair.
Now, the Court left for another day the proposition that a privative clause might, in
combination with a limited right of appeal, preclude judicial review. But the emphasis on
judicial review as a fundamental feature of the Canadian public law landscape suggests
that the presence of a privative clause — unconstitutional, let us not forget, per the Court’s
decision in Crevier v. A.G. (Québec) et al., [1981] 2 SCR 220 — will not change the analysis
in any meaningful way. As I have demonstrated, privative clauses were no absolute bar to
judicial review historically and there is no reason today to deviate from tradition or the
first principles asserted in Crevier. Moreover, the fact that Rowe J poured cold water on
the judicial economy rationale for restricting judicial review would strongly suggest that
the concerns underlying the enactment of privative clauses should not sway Canadian
courts either.
For similar reasons, I would suggest that the Court’s analysis bodes ill for other
restrictions on judicial review, like s. 18.5 of the Federal Courts Act. This provision has
been invoked in the context of economic regulation of telecommunications and
transportation but might now also bar access to judicial review to judges disciplined by
the Canadian Judicial Council (see here). The question there will be the extent to which
the Federal Court of Appeal’s important decision in Canadian National Railway
Company v. Scott, 2018 FCA 148 is consistent with Vavilov and Yatar. (Note that my
client in Yatar took the position that s. 18.5 does have to be revisited in light of Vavilov.)
On the merits, the decision was unreasonable and the matter remitted to the Tribunal:
However, the LAT adjudicator failed to have regard to the effect of the
reinstatement of the IRBs between February and September. The LAT adjudicator
did not consider earlier tribunal decisions, some of which had held that when an
applicant’s benefits are reinstated, the limitation period can only be triggered when
they are validly terminated again (see Veldhuizen v. Coseco Insurance Co., 1995
ONICDRG 144 (CanLII); Rudnicki v. Certas Direct Insurance Co., 2001
ONFSCDRS 60 (CanLII)).
It is not in question that Ms. Yatar initiated mediation in September 2012. The
mediation took place between June 18, 2013 and January 14, 2014. On January 14,
2014, the mediator released his report. However, s. 281.1(2)(b) of the Insurance
Act and s. 51(2) of the SABS (as they existed at the time) do not trigger a 90-day
limitation period from the release of the mediator’s report. Rather, they provide
for an extension of the two-year limitation period from the mediator’s report. In
other words, it is arguable that there still needed to be a valid denial of the IRBs to
start the clock running. I do not purport to decide this question; it is one properly
to be decided by the LAT (at paras. 74-75).
Rowe J described this as a breach of “legal constraints” (at para. 76), which might prompt
one to wonder whether the matter should have been treated as falling within the appeal
clause in any event. For my part I do not think anything turns on
whether Vavilov’s contextual constraints are described as legal or factual (other than to
facilitate analysis): the key point here was that the adjudicator’s decision lacked
justification, intelligibility and transparency on factually suffused questions.
The privative clause issue left unresolved in Yatar was the subject of a thoughtful set of
reasons in Democracy Watch v. Canada, 2024 FCA 158, de Montigny CJ engaged
extensively with my writings on the subject but took a very different view from me. I do
not propose to attempt to produce a comprehensive response (or maybe it is a reply or
sur-reply at this stage!), as I know from my incoming correspondence that you are all well
able to make up your own minds when issues have been fully argued on both sides. Please
do read de Montigny CJ’s reasons, especially paras. 58-78, regardless of the fact that they
are obiter (see the concurring judges at para. 96).
It is worth highlighting a couple of points, however, as these go to the core of the
disagreement between those with competing views of the constitutional core minimum of
judicial review of administrative action.
Consider, first, the permissible scope of legislative intervention to limit judicial review. de
Montigny CJ is of the view that legislation could eliminate reasonableness review in its
entirety:
This is not only consistent with the various dicta of the Supreme Court with respect
to the role of judicial review (most explicitly in Crevier and Dunsmuir) and with
its insistence on respect for institutional design choices in Vavilov. It is also
aligned with the underlying rationale for judicial review in a parliamentary
democracy, which is that all exercises of delegated authority by the executive
branch must find their source in the law and be respectful of the Constitution (at
para. 73).
He goes on to ask, “As long as courts have the ability to intervene in cases where an
administrative decision-maker steps out of bounds and impermissibly oversteps its lawful
authority, how can it be said that the rule of law is threatened by the insertion of a
privative clause in a statute?” But this question begs the question. Given the significant
changes wrought by Vavilov, how can one say that a court can determine when a decision-
maker has ‘stepped out of bounds’ or ‘overstepped its lawful authority’ without applying
the reasonableness standard? Indeed, in Vavilov, the majority of the Supreme Court
remarked that “proper application of the reasonableness standard will enable courts to
fulfill their constitutional duty to ensure that administrative bodies have acted within the
scope of their lawful authority…” (at para. 67; see also at para. 109). To my mind, this
passage ties lawful authority to reasonableness review.
Second, there is the issue of respecting legislative intention. de Montigny CJ also sees the
downgrading of privative clauses in Vavilov as problematic because it fails to give
sufficient weight to parliamentary supremacy:
[P]rivative clauses are downgraded from an important factor in determining the
applicable standard of review (as in Dunsmuir) to a mere contextual factor in
determining the parameters of a reasonable decision. In light of the high degree of
deference to which administrative decision makers are entitled when their
decisions are subject to the reasonableness standard, it is not readily apparent
what extra protection from judicial scrutiny a privative clause would confer (at
para. 66).
As I have suggested previously (and see Koebisch v. Rocky View (County), 2021 ABCA
265, at para. 24), the answer is that a privative clause — full or partial — forms part of the
‘governing statutory scheme’ envisaged by Vavilov. As the majority of the Supreme Court
noted there, “where the legislature chooses to use broad, open-ended or highly qualitative
language — for example, “in the public interest” — it clearly contemplates that the
decision maker is to have greater flexibility in interpreting the meaning of such language”
(at para. 110). A privative clause would be an indication that “greater flexibility” is
appropriate, with a partial privative clause carrying less weight and a full privative clause
weighing heavily in the balance. This could be particularly significant in a situation where
an administrative decision-maker has been tasked with interpreting “precise and narrow
language” (Vavilov, at para. 110): in such circumstances, a privative clause would instruct
the court to give “greater flexibility” in applying the reasonableness standard even though
the language being interpreted is precise and narrow.
And now to the facts of the case! Alleging errors of fact and errors of law, the applicant
sought judicial review of a report of the Conflict of Interest and Ethics Commissioner
concluding that the Prime Minister had not violated conflict of interest legislation by
participating in two decisions involving a controversial charitable organization. For de
Montigny JA, the partial privative clause excluding judicial review for errors of law and
errors of fact meant that the application was doomed to failure; but the concurring judges
disagreed, given that previous panels of the Federal Court of Appeal had taken a different
view on the constitutional core minimum of judicial review (at para. 96).
All three judges agreed, however, that political oversight can be an adequate alternative
remedy to judicial review. de Montigny CJ noted that the Ethics Commissioner is “an
independent Officer of Parliament, and the position he occupies is firmly within the
legislative branch of government” (at para. 80). In addition, he found that the statutory
scheme suggests that Parliament intended political oversight of the Ethics Commissioner
to be a central feature:
It is very clear from subsections 44(7), 44(8), 45(3) and 45(4) of the COIA, which
require the Commissioner to provide his reports to the Prime Minister, that it is
for the Prime Minister to decide how to give effect to the Commissioner’s
determination, and for the House of Commons to hold the government to account.
The sanction is meant to be political, not judicial. This conclusion is reinforced by
the fact that the report of the Commissioner is to be made available to the public,
and that its conclusions are not determinative of the measures to be taken (at para.
81).
There are “dual supervisory roles” (at para. 82), but the courts’ task is limited by the
partial privative clause to ensuring the Commissioner respects their jurisdiction, grants
procedural fairness to affected parties and does not act fraudulently. Furthermore, for de
Montigny CJ, “Courts should be loath to perceive judicial remedies as the only effective
recourse in every instance where an aggrieved party raises an alleged illegality” (at para.
84, citing Canada (Auditor General) v. Canada (Minister of Energy, Mines and
Resources), [1989] 2 SCR 49). Ultimately, “courts should not be drawn in disputes raising
purely legal or factual issues within the jurisdiction of the Ethics Commissioner” (at para.
88).
As de Montigny CJ notes, this approach is consistent with that taken in provincial
superior courts where judicial review has been sought of decisions of officers of the
legislative assembly: McIver v. Alberta (Ethics Commissioner), 2018 ABQB 240 at paras.
70-77; Democracy Watch v. British Columbia (Conflict of Interest Commissioner), 2017
BCSC 123 at paras. 35-37.
For my part, I have always been wary of these decisions. If an officer of a legislative
assembly occupies a statutory office, with powers and functions accorded by law and —
by definition — subject to limitations set out in its parent statute, to my mind it does not
obviously follow that the officer’s decisions are non-justiciable.
I appreciate that if someone seeks to raise a political issue before the courts arising from
an officer’s decision judicial intervention might very well not be appropriate. However, if
as here the issue is whether the officer correctly or reasonably interpreted statutory
concepts, excluding judicial review seems quite strange as it means that the legislator has
the final word on the interpretation of its own statutes. After all, “[j]udicial review is the
means by which the courts supervise those who exercise statutory powers, to ensure that
they do not overstep their legal authority” (Dunsmuir v. New Brunswick, 2008 SCC 9,
[2008] 1 SCR 190, at para. 28). Why would it be any different when an officer of the
legislature is administering a statutory scheme?
In that regard, it is notable that the partial privative clause at issue in Democracy
Watch expressly preserves judicial review on ‘jurisdictional’ issues: but at what point does
an error of law or fact become a potential jurisdictional issue? I would say that the
difficulty of drawing this line is, in and of itself, a good reason to take a broad view of the
constitutional core minimum of judicial review.
In all events, this is a very interesting decision of which both obiter and ratio will repay
careful reading both on the constitutional foundations of judicial review and the adequacy
of political oversight as a remedy.
B. Amenability to the Charter
In two cases this year, the Supreme Court of Canada revisited the scope of application of
the Charter of Rights and Freedoms. In York Region District School Board v.
Elementary Teachers’ Federation of Ontario, 2024 SCC 22, the Court held that
the Charter applies to Ontario school boards. And in Dickson v. Vuntut Gwitchin First
Nation, 2024 SCC 10, a majority of the Court concluded that self-governing Indigenous
nations are subject to the Charter. I will outline the basic principles of s. 32, describe the
decisions in these cases and conclude with some thoughts about what they portend for
the future of s. 32, with particular reference to the application of the Charter to
universities.
Let us begin with some background. The application clause in s. 32 of
the Charter specifies the entities to which the Charter is applicable. Unsurprisingly,
the Charter applies to Parliament and the provincial legislative assemblies. But it also
applies to the “government” of Canada and of each province. The consequence of the
application of s. 32 is that any entity considered to be “government” will have to comply
with the Charter.
In Eldridge v. British Columbia (Attorney General), [1997] 3 SCR 624 the Supreme Court
explained that there are two ways in which an entity can be considered to be “government”
for the purposes of the Charter. An entity can be government in all of its functions, or can
be government in some functions that involve the implementation of government policy:
First, it may be determined that the entity is itself “government” for the purposes
of s. 32. This involves an inquiry into whether the entity whose actions have given
rise to the alleged Charter breach can, either by its very nature or in virtue of the
degree of governmental control exercised over it, properly be characterized as
“government” within the meaning of s. 32(1). In such cases, all of the activities of
the entity will be subject to the Charter, regardless of whether the activity in which
it is engaged could, if performed by a non-governmental actor, correctly be
described as “private”. Second, an entity may be found to attract Charter scrutiny
with respect to a particular activitythat can be ascribed to government. This
demands an investigation not into the nature of the entity whose activity is
impugned but rather into the nature of the activity itself. In such cases, in other
words, one must scrutinize the quality of the act at issue, rather than the quality of
the actor. If the act is truly “governmental” in nature — for example, the
implementation of a specific statutory scheme or a government program — the
entity performing it will be subject to review under the Charter only in respect of
that act, and not its other, private activities (at para. 44).
There are, therefore, two branches to the Eldridge test: entities can be inherently
governmental (a branch that has two aspects (or twigs?), either by virtue of the nature of
the entity or control by government); or entities can be governmental because they
undertake governmental actions.
In Eldridge, hospitals came within s. 32 because of their role in implementing a
government programme, the second branch of the test. This year’s cases focused more on
the first branch.
In York Teachers, the Supreme Court held that school boards in Ontario are inherently
governmental entities that are always “government” for the purposes of s. 32. Rowe J
offered two reasons in support of this conclusion, covering both aspects of the first branch
of the Eldridge test.
First, school boards exercise statutory authority under provincial legislation and are
subject to the “extensive powers” of the Minister of Education, such that they are “in
effect, an arm of government” (at para. 79). Second, public education “is inherently a
governmental function” (at para. 81), as is manifest from the constitutional guarantee of
denominational education in the province of Ontario in s. 93 of the Constitution Act,
1867.
Ultimately, “[a]ll actions carried on by Ontario public school boards are subject
to Charter scrutiny” (at para. 82). In a nutshell, the statutory authority they have is not
exercised for reasons internal to the school boards themselves but seeks to further an
external public purpose (at para. 80). Under both aspects of the first branch of
the Eldridge test — governmental by very nature and governmental control — school
boards qualified.
In Dickson, Kasirer and Jamal JJ held that the Vuntut Gwitchin First Nation was
“government” for the purposes of s. 32. They began by distinguishing the two different
ways in which an entity might be said to be “government” under the first branch of
the Eldridge test: by its very nature or by virtue of being controlled by government. Here,
the VGFN was not controlled by government:
Through the Self-Government Agreement, the VGFN exercises self-government
powers “by and for the first nation” (Yukon First Nations Self-Government Act, s.
2, “self-government agreement”)…Its essential function is to advance the interests
of the Vuntut Gwitchin in accordance with their “traditional decision-making
structures” and “to support and promote” their “contemporary and evolving
political institutions and processes” (Self-Government Agreement, preamble). To
this end, the Self-Government Agreement provides that the VGFN “shall have the
exclusive power to enact laws” in relation to an agreed upon list of matters,
including the administration of the VGFN’s “affairs and operation and internal
management” (Self-Government Agreement, s. 13.1). As a result, the VGFN
operates autonomously from, and is not substantially controlled by, either the
federal or the Yukon government (at para. 76).
However, the VGFN was nonetheless government because of its resemblance to the
characteristics set out as indicia of government in Godbout v. Longueuil (City), [1997] 3
SCR 844. It is a democratically elected entity that has law-making authority (including
the power to raise revenue) within a defined territory that is ultimately attributable at
least in part to the federal government and Yukon legislative assembly (at paras. 79-91).
Kasirer and Jamal JJ also found that the VGFN fell under the second branch of
the Eldridge test. Here, the issue was whether a residency requirement to be able to
participate in VGFN elections was consistent with the Charter right to equality. In
implementing this residency requirement, the VGFN was, in their view, implementing a
government programme:
The residency requirement under the VGFN’s Constitution was adopted at least in
part under federal statutory authority (even assuming it also reflects the exercise
of an inherent right to self-government). The residency requirement involves the
exercise of a statutory power of compulsion because it imposes legal restrictions
on who may serve as a VGFN Chief or Councillor. It has the force of law because it
forms part of the VGFN Constitution, adopted under the Self-Government
Agreement, which was itself approved and given effect by the federal and territorial
implementing legislation. As stated succinctly in s. 2 of the First Nations (Yukon)
Self-government Act: “The Self-Government Agreement is hereby approved and
has the force of law” (see also Yukon First Nations Self-Government Act, s. 5(1)
(bringing into effect the Self-Government Agreement), and s. 8(1)(b) (providing
that a Yukon First Nation such as the VGFN “shall, in a manner consistent with its
self-government agreement”, adopt a constitution that provides for, among other
things, “the governing bodies of the first nation and their composition,
membership, powers, duties and procedures”)). The federal legislation thus gives
the VGFN Constitution the force of federal law under the Constitution Act, 1867,
even if it already had the force of law as an Indigenous law (at para. 95).
Martin and O’Bonsawin JJ wrote separately on the s. 32 issue, reaching the same
conclusion as the majority but through a different route. They were not satisfied that
existing categories adequately captured Indigenous self-government and expressed
particular concern about the notion that the VGFN was exercising authority delegated by
“another level of government” (at para. 260):
Looking for a form of delegated authority by the federal or a provincial government
does not respect the historic and integral role of Indigenous societies in Canada
and fails to account for the modern context of self-governing Indigenous nations
(at para. 263).
They preferred a more expansive, purposive approach to s. 32. Whereas Kasirer and
Jamal JJ reasoned by analogy to Godbout, a case involving the application of s. 32 to
municipalities, Martin and O’Bonsawin JJ sought to ground their analysis in first
principles relating to the raison d’être of the Charter:
The reference to [Parliament and the provincial legislatures in s. 32] signals an
intention that the full legislative field — “all matters within the authority” of
Parliament and the provincial legislatures — be covered by the Charter. Otherwise,
individuals would be deprived of critical protections in situations of power
imbalance between the governed and those who govern. This would amount to
sanctioning Charter-free zones (at para. 269).
In the end, however, Martin and O’Bonsawin JJ agreed that the Charter applies to the
VGFN and relied on a similar set of factors to the majority:
For example, the VGFN council is democratically elected, it has the authority to
levy taxes, and it possesses extensive lawmaking powers — all of which were
described by La Forest J. in Godbout as non-exhaustive indiciaof government
(para. 51; see also VGFN Final Agreement (1993), ch. 24; VGFN Self-Government
Agreement, s. 13.0; Yukon First Nations Self-Government Act, s. 11 and Sch. III).
The VGFN also has the exclusive authority to govern its internal affairs (VGFN
Self-Government Agreement, s. 13.1). This includes the enactment of the residency
requirement itself — a legislative action that impacts individual members’ rights
by restricting democratic participation. Further, the VGFN can enact laws
applicable in the Yukon in relation to a wide array of matters touching the everyday
lives of its citizens, such as health care, education, adoption and child welfare,
solemnization of marriage, and estate administration (s. 13.2). It is empowered to
legislate on matters of a local or private nature on the settlement land, which the
VGFN Self-Government Agreement defines as including the administration of
justice, land use, expropriation of land, protection of fish, wildlife, and habitat,
environmental protection, vehicles, curfews and the prevention of disorderly
conduct, public health, public order, and other “matters coming within the good
government of Citizens on Settlement Land” (s. 13.3) (at para. 278).
The real difference between the two sets of judges was the role of delegation. It was
important for the majority (see e.g. at para. 86, albeit not centrally so: at para. 83) but not
for Martin and O’Bonsawin JJ, for whom the VGFN’s authority was “a different modality
for the exercise of legislative power” (at para. 279). First Nations are not, one might say,
like school boards or municipalities.
The majority went on to conclude that VGFN was shielded from the rigours of
the Charter’s equality guarantee by s. 25 of the Constitution Act, 1982, which states that
the Charter shall not “be construed so as to abrogate or derogate from any aboriginal,
treaty or other rights or freedoms that pertain to the aboriginal peoples”. Martin and
O’Bonsawin JJ dissented on this point.
Rowe J dissented on the application of the Charter to the VGFN. In his view,
the Charter was “designed by the federal and provincial governments, for those
governments” (at para. 437, emphasis original). A “separate” set of mechanisms was
created in the Constitution Act, 1982 to regulate the relationship between Indigenous
peoples and the federal and provincial governments (at para. 443). Given these features
of the 1982 constitutional settlement, judicial oversight of Indigenous legal orders
for Charter compliance would be inappropriate: “Whatever the nuances, their
approaches [Kasirer and Jamal JJ and Martin and O’Bonsawin JJ] amount to a means
for courts to apply Charter rights (or “values”) to the operation of Indigenous self-
government” (at para. 450).
As for the s. 32 jurisprudence, the application of the Charter has always turned on the
relationship between the entity in question and the federal and provincial governments
identified in s. 32 (at para. 461):
The jurisprudence simply recognizes that the scope of s. 32(1), while capable of
embracing extensions of the federal and provincial governments in the modern
administrative state, is not so broad as to encompass entities that bear no
significant connection to the particular governments enumerated in s. 32(1) itself
(at para. 463).
There is thus no need to shoehorn Indigenous legal orders into s. 32 because the fear of
creating Charter-free zones for government action — a consideration that underpins the
second branch of the Eldridge test — does not have any purchase in the context of self-
governing entities:
The jurisprudence simply recognizes that the scope of s. 32(1), while capable of
embracing extensions of the federal and provincial governments in the modern
administrative state, is not so broad as to encompass entities that bear no
significant connection to the particular governments enumerated in s. 32(1) itself
(at para. 468).
Ultimately, for Rowe J:
the VGFN is not, by its nature, part of the apparatus of the federal government in
the sense required under s. 32(1). Rather, its internal governance structures are
rooted in the VGFN’s own laws, customs, and practices. The VGFN Arrangements
give the VGFN’s unique governance structures effect within contemporary Canada
through mutually binding commitments between the VGFN and the Crown (at
para. 490).
Rowe J essentially sees First Nations as autonomous entities, somewhat like private
organizations created to further the interests of their members and which ought to be free
from court oversight. This is of a piece with his decision (for a majority of the Court)
in Highwood Congregation of Jehovah’s Witnesses (Judicial Committee) v. Wall, 2018
SCC 26, [2018] 1 SCR 750 (see especially at para. 36). More generally, Rowe J returns to
first principles just as Martin and O’Bonsawin JJ does. His reasons recall the point made
trenchantly by La Forest J in McKinney v. University of Guelph, [1990] 3 SCR 229 that
the Charter does not apply to private ordering:
The exclusion of private activity from theCharter was not a result of
happenstance. It was a deliberate choice which must be respected. We do not
really know why this approach was taken, but several reasons suggest
themselves. Historically, bills of rights, of which that of the Unites States is the
great constitutional exemplar, have been directed at government. Government is
the body that can enact and enforce rules and authoritatively impinge on individual
freedom. Only government requires to be constitutionally shackled to preserve the
rights of the individual. Others, it is true, may offend against the rights of
individuals. This is especially true in a world in which economic life is largely left
to the private sector where powerful private institutions are not directly affected
by democratic forces. But government can either regulate these or create distinct
bodies for the protection of human rights and the advancement of human dignity.
Moving forward, it is interesting to reflect on whether these cases portend any further
new developments in the scope of s. 32.
The application of the Charter to universities has long been a bone of contention.
In McKinney, the Court held that s. 32 does not cover universities, but the point has been
contested in subsequent jurisprudence (Pridgen v. University of Calgary, 2012 ABCA
139; BC Civil Liberties Association v. University of Victoria, 2016 BCCA 162; UAlberta
Pro-Life v. Governors of the University of Alberta, 2020 ABCA 1), especially under the
second branch of the Eldridge test (see Koshan and Hamilton). At this point in time, the
question is very much on the public agenda given the regulation of protests on university
property (see e.g. University of Toronto (Governing Council) v. Doe et al., 2024 ONSC
3755). Further litigation would not be surprising, especially in light of York
Teachers and Dickson.
Taken together, the focus in the majority reasons in York Teachers and Dickson on
delegated authority and public functions (including the elaboration of binding norms) is
likely to animate arguments for Charter application in university settings. For the most
part, Canadian universities exercise delegated authority under statute, make binding
rules, take highly consequential decisions and perform public functions in educating
students and distributing government money. These considerations chime with those
relied upon in York Teachers and Dickson. Even the minority reasons in Dickson, with
the emphasis on protecting individual rights and not creating Charter-free zones, point
in the direction of extending the application of s. 32 to universities.
Only Rowe J’s plaintive appeal to autonomy — the key factor, in my view, leading the
Court in the McKinney decision to hold that universities are not covered by s. 32 — gives
any comfort to university administrators who would like to ward off court oversight
for Charter compliance. But that appeal was in dissent in Dickson. Ultimately, if even
self-governing First Nations are subject to the Charter, by what rationale could self-
governing universities not be?
Conclusion
This has been a busy year for the Supreme Court of Canada in the administrative law field
and adjacent areas.
In Part I, I discussed the Court’s jurisprudence on reasonableness review. I have some
doubts about the appropriateness of its analysis in the Mandate Letters case, though it
may prove to be a once-off. I am more sanguine about the decision in CSFTNO, which
helps to establish a clear framework for judicial review of Charter issues.
The Charter allowed me to segue from Part I on reasonableness review to Part II on
correctness review, where the York Teachers case involved a difficult choice between
correctness and reasonableness. York Teachers at least brought clarity, however, as is also
true – in my view – of the decision in Société des Casinos.
In Part III, I addressed lower-court jurisprudence on justiciability, an issue that is quite
likely to return to the Supreme Court in the near future.
And I rounded off this ‘year in review’ in Part IV with a consideration of the Supreme
Court’s 2024 decisions on the application of the Charter.
To return to where I began, we are at the 5-year mark post-Vavilov. Not every possible
question has been answered – how could it?! – but it seems fair to say that Vavilov is
doing much better than its predecessor Dunsmuir at this point in its life cycle.
