COURT OF APPEAL FOR ONTARIO

CITATION: Amalgamated Transit Union, Local 113 v. Ontario, 2024 ONCA 407

DATE: 20240523

DOCKET: COA-23-CV-0712

Nordheimer, Copeland and Dawe JJ.A.

BETWEEN

Amalgamated Transit Union, Local 113, Marvin Alfred and Kevin Morton; Canadian Union of Public Employees, CUPE Local 2 and Gaetano Franco

Applicants (Respondents)

and

The Crown in Right of Ontario as represented by the Attorney General of Ontario

Respondent (Appellant)

Rochelle Fox, Daniel Huffaker and Priscila Atkinson, for the appellant

Ian J. Fellows, Joanna Birenbaum, Kristen Allen, Emily Home and Adriana Zichy for the respondents Amalgamated Transit Union, Local 113, Marvin Alfred and Kevin Morton

Elizabeth Nurse and Devon Paul, for the respondents Canadian Union of Public Employees, CUPE Local 2 and Gaetano Franco

Heard: January 15, 2024

On appeal from the order of Justice William S. Chalmers of the Superior Court of Justice, dated May 8, 2023, reported at 2023 ONSC 3618.

Dawe J.A.:

[1] The Toronto Transit Commission (“TTC”) operates Toronto’s transit system, which includes subway, streetcar, and bus routes. The TTC has more than 12,000 employees, who are members of several different unions, including the respondents, Amalgamated Transit Union, Local 113 (“ATU Local 113”) and CUPE Local 2 (jointly the “respondents” or the “unions”).[^1]

[2] In 2011, the Ontario legislature passed the Toronto Transit Commission Labour Disputes Resolution Act, 2011, S.O. 2011, c.2 (the “TTC Act”). The TTC Act eliminates TTC workers’ right to engage in any form of strike activity, and also bars the TTC from locking out its employees. If the TTC and its unions are unable to resolve issues through collective bargaining, the TTC Act requires them to submit to binding interest arbitration.

[3] In 2015, four years after the TTC Act was enacted, the Supreme Court of Canada released a landmark labour law decision, Saskatchewan Federation of Labour v. Saskatchewan, 2015 SCC 4, [2015] 1 S.C.R. 245, (“SFL”). In SFL, the Court reversed a series of its previous decisions and found for the first time that the right to strike is an integral aspect of the right of freedom of association enshrined in s. 2(d) of the Canadian Charter of Rights and Freedoms.

[4] In the wake of SFL, the respondents applied to the Ontario Superior Court of Justice for a declaration that the TTC Act violated s. 2(d) of the Charter and could not be justified under s. 1. Their application took more than seven years to proceed to a hearing, as both sides assembled voluminous evidential records that included evidence from multiple expert witnesses.

[5] In May 2023, the application judge found the TTC Act unconstitutional and struck it down, effective immediately. He found that the TTC Act violated s. 2(d) of the Charter, and that the government had failed to meet its onus of justifying the breach under any of the branches of the Oakes s. 1 legal framework: see R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, at pp. 136-40.

[6] The appellant, the Crown in Right of Ontario (“Ontario”), appeals from this decision. Ontario argues that the application judge erred by finding that the TTC Act violates s. 2(d) of the Charter. In the alternative, Ontario contends that the application judge made multiple errors in his s. 1 analysis, and submits that he ought to have found that any s. 2(d) Charter breach was justified under s. 1.

[7] I would dismiss the appeal. I agree with the application judge’s conclusion that the TTC Act violates TTC employees’ s. 2(d) Charter rights, although I would arrive at this conclusion by a somewhat different analytic route. I accordingly agree with his finding that it was Ontario’s burden to justify the infringement under s. 1.

[8] I agree with Ontario that the application judge made some legal errors in his s. 1 analysis. In particular, he erred by concluding that the TTC Act did not have pressing and substantial legislative objectives, and also by finding that the means the legislature chose to achieve these objectives – namely, pre-emptively banning all TTC strikes and lockouts – was not rationally connected to the legislature’s goals.

[9] However, I am not persuaded that the application judge made any reversible errors when he found further that Ontario had not met its burden under the minimal impairment and proportionality prongs of the Oakestest. His conclusions on these two branches of the Oakesanalysis arise from his factual findings based on the evidence before him, which are entitled to appellate deference.

[10] I would accordingly dismiss the appeal.

A. Facts and evidence

[11] The evidence adduced on the application regarding the history of labour relations between the TTC and its unionized employees, in the years before and after the 2011 enactment of the TTC Act, is extensively summarized in the application judge’s reasons. I will provide only a brief overview.

(1) Events prior to the enactment of the TTC Act

[12] The TTC operates the largest transit system in Canada and the third-largest system in North America. At the time of the application, the system consisted of four subway lines,[^2] eleven streetcar routes, and approximately 140 bus routes. It has more than 12,000 employees, who occupy a wide array of job classifications.

[13] ATU Local 113 is the bargaining agent for approximately 11,320 of the TTC’s employees. It was established in 1899, and has been negotiating collective agreements with the TTC for more than 100 years. CUPE Local 2 represents approximately 700 TTC employees, who mostly work in signals, electrical and communications. It has been negotiating collective agreements with the TTC for more than 50 years.

[14] Before the TTC Act was enacted in 2011, labour relations between the TTC and its unionized employees were governed by the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A (“the LRA”),and its predecessor statutes. The LRA adopts what is often called the “Wagner model” of labour-management relations.[^3] As the application judge explained in his reasons:

The LRA allows unions to engage in legal strike activity after the collective agreement has expired, bargaining has reached an impasse, and certain statutory prerequisites are met, including a membership vote. Strike activity under the LRA can include a variety of measures including a full withdrawal of labour, a partial withdrawal, refusal to work overtime, or work to rule. The LRA regime also entitles employers to lock-out employees or unilaterally alter their terms and conditions of employment, after the collective agreement has expired.

[15] Between 1991 and 2008, a total of twelve full TTC service days were lost due to strikes by ATU Local 113, during which CUPE Local 2 workers either also went on strike or were locked out by TTC management.

[16] The longest strike, in 1991, was settled after eight days with the involvement of the Minister of Labour. In 1999 there was a strike that ended after two days. In 2006 there was a one-day “wildcat” strike that ended after the Labour Board ruled the strike to be unlawful and ordered the strikers back to work. In 2008 there was a strike that lasted for a day and a half before the legislature enacted back-to-work legislation.

[17] The 2008 strike was noteworthy because it occurred with very little notice to the public. The TTC had reached a tentative agreement with ATU Local 113 negotiators that required ratification by Local 113’s members, who voted against the deal. Although Local 113 had previously agreed to provide 48 hours’ notice of any strike action, after the failed ratification vote TTC employees walked off the job at 12:01 A.M. on a Saturday morning, giving the public only 90 minutes advance warning. The application judge noted that “[a]s a result, there were thousands of people stranded late at night, without transit.” This led the Ontario legislature to take the unusual step of convening the next day, on a Sunday afternoon, to pass back-to-work legislation.

[18] In the wake of the 2008 strike, Toronto City Council debated a motion that asked the province to declare the TTC an essential service, but the motion was defeated. However, City Council revisited the issue two years later, in late 2010, and this time the motion passed.

(2) The TTC Act

[19] On February 20, 2011, the government introduced Bill 150, which created the TTC Act. The legislation passed second reading on March 8, 2011 and was sent to committee. A few weeks later, on March 30, 2011, after two days of committee hearings, Bill 150 carried without amendment, and it received royal assent that same day. This also happened to be the day before the TTC’s collective agreement with ATU Local 113 expired.

[20] The application judge noted in his reasons that studies and stakeholder consultations prior to the enactment of Bill 150 were limited:

In answer to an undertaking, the Government stated that no other models were studied at the time the legislation was introduced. The Government argues that Bill 150 was introduced after consultations with the City, TTC, and Unions. The [unions] argue that based on the evidence, the consultations with the unions were very limited. There is no evidence that the province obtained a study or report to examine the impact of TTC strikes on the life, health or safety of the public.

[21] As the application judge explained, the TTC Act removes employees’ right to strike or take collective action of any kind, replacing that right with mandatory interest arbitration:

The TTC Act removes the right of TTC employees to engage in a strike. No collective action of any kind is permitted. The [removal of the] right to strike applies to all TTC employees regardless of the job. For example, no distinction is made between operators and customer service agents. The TTC Act also removes the employer’s right to lock out the employees and prohibits the employer from unilaterally changing the terms and conditions of employment.

Instead of the right to strike, the TTC Act provides for a mandatory binding interest arbitration process. In the event of an impasse in bargaining … the arbitrators are to be appointed by joint agreement. If the parties are unable to agree on an arbitrator, the Minister of Labour shall appoint one.

(3) Collective bargaining at the TTC since 2011

[22] The parties presented conflicting evidence about how the TTC Act has affected collective bargaining between the TTC and its unions. Both sides adduced opinion evidence from academic experts about how replacing the right to strike with binding interest arbitration generally affects the collective bargaining process. The parties also presented evidence about the history of collective bargaining at the TTC since 2011.

[23] In summary, the unions’ evidence was that the relationship between the TTC and the unions has worsened since the passage of the TTC Act, and the TTC’s management has become less respectful and reasonable. The application judge accepted this evidence, and adopted the comments of the arbitrator who conducted interest arbitrations in 2018 and 2021. The arbitrator noted in his 2021 award that since the passing of the TTC Act, “collective bargaining has been singularly, serially and completely unsuccessful.”

[24] Based on the expert evidence presented by both sides, the application judge concluded that “the loss of the right to strike adversely affects other critical components of the relationship between the employer and employees.” In particular, he accepted that “there has been an increase in grievances since the introduction of the TTC Act and a decrease in member involvement, which indicate the impact of the infringement of the right to strike on union democracy.”

[25] The application judge found further that “the removal of the right to strike has had a negative effect on the negotiating process.” He explained:

The manner in which negotiations were conducted after the enactment of the TTC Act supports the conclusion that employees have not been on an equal footing with the TTC. Mr. Kinnear, Mr. Morton and Mr. Franco’s evidence with respect to the approach taken by TTC negotiators in 2011 is particularly troubling.[^4] The negotiators were more confrontational and aggressive. The negotiators stated that they had less incentive to be conciliatory because the union did not have the right to strike. The Government did not file an affidavit from any TTC negotiators in 2011 to rebut this evidence.

[26] The application judge also found that “interest arbitration substantially interfered with the ability of the parties to reach voluntary settlements”, explaining:

As noted by Professor Hebdon,[^5] binding interest arbitration can result in “chilling” and “narcotic” effects. That has also been borne out by the evidence. The TTC and its largest union, Local 113 have been unable to reach voluntary agreements without assistance in three of the last four bargaining sessions.

Arbitrator Kaplan was the arbitrator appointed to settle the terms of the collective agreement between the TTC and Local 113, in 2018 and 2021. In his 2021 Award, he notes that although the parties met ten times between the end of February 2021 and the end of March 2021, there were no agreed items to be incorporated in the collective agreement. He states that there was “no possibility of finding common ground about anything”.

(4) Section 1 evidence

[27] Ontario adduced evidence from two expert witnesses who offered opinions about how a strike leading to a shutdown of the Toronto transit system would affect traffic congestion and air pollution. The first witness, Dr. Eric Miller, is an expert in transportation systems analysis and travel demand modelling. He gave his opinion about how motor vehicle traffic in Toronto would likely increase if there was a full transit system shutdown. The second witness, Dr. Marianne Hatzopoulou, is an expert in modelling road transport emissions and urban air quality. She ran a hypothetical emissions model, and gave opinion evidence predicting that air pollution would increase in the city as a result of increased motor vehicle traffic, and about the impact this would have on public health.

[28] In response, the respondents called Dr. Paul Villeneuve, an epidemiologist with expertise in the human health effects of air pollution. He reviewed air pollution and hospitalization data from the 2006 and 2008 strikes, and provided the opinion that there was no evidence that these strikes caused any adverse health consequences.

[29] On the issue of the economic impact of a TTC strike, Ontario relied on two reports from 2008, one written by Marilyn Churley and published by the transit unions, and the second written by the City’s Economic, Development, Culture and Tourism Division. The latter report estimated that the short-term economic impact of a TTC strike would be approximately $50 million per day.

[30] On the issue of the impact of a TTC strike on equity-seeking groups, Ontario adduced evidence from Dr. Steven Farber, a transportation geographer and spatial analyst. Dr. Farber offered the opinion that a TTC strike would have a disproportionate impact on “lower-income, visible minorities, younger, immigrants, and otherwise less affluent people”, because these people were most likely to rely on public transit.

B. Standard of review

[31] As Favreau J.A. recently observed in her majority reasons in Ontario English Catholic Teachers Association v. Ontario (Attorney General), 2024 ONCA 101 (“OECTA”), at para. 47, questions of constitutional validity are subject to a correctness standard:

The constitutional validity of [legislation] is a question of law to be decided on a standard of correctness. However, this court owes deference to the application judge’s findings of fact, including findings based on social and legislative evidence.

[32] She explained further, at para. 51, that findings of fact relevant to the legal question at issue are to be reviewed on a standard of palpable and overriding error:

Accordingly, the questions of whether the Act violates s. 2(d) of the Charter and, if so, whether it is saved by s. 1 of the Charter are to be reviewed on a standard of correctness. This inquiry includes consideration of what factors are relevant to deciding these issues. However, the trial judge’s findings of fact relevant to this assessment are to be reviewed on the palpable and overriding error standard of review.

[33] Accordingly, I review the application judge's determination regarding the constitutionality of the TTC Act on a correctness standard, but defer to his findings of fact, absent a palpable and overriding error.

C. Does the TTC Act violate s. 2(d) of the Charter?

[34] The application judge found that the TTC Act violated s. 2(d) of the Charter. This finding shifted the burden to Ontario to justify the infringement under s. 1. Ontario contends that the application judge erred by finding a s. 2(d) breach. In the alternative, Ontario submits that the application judge erred in his analysis under s. 1, and that any infringement of s. 2(d) is justified under s. 1.

[35] As a threshold matter, the parties disagree about the proper framing of the s. 2(d) Charter issue in this case. The application judge approached this question by asking whether the TTC Act’s elimination of TTC employees’ right to strike “substantially interferes” with meaningful collective bargaining. He then engaged in a detailed analysis of the evidence of how the TTC Act had affected collective bargaining between the TTC and its unions, and concluded that the “substantial interference” test had been met.

[36] Ontario takes the position that the application judge asked the right question in the s. 2(d) analysis, but reached the wrong answer on the evidence before him. In broad terms, Ontario contends that because the TTC Act replaces the right to strike with compulsory binding interest arbitration, it necessarily protects the collective bargaining process sufficiently to avoid breaching s. 2(d) of the Charter.

[37] The respondents, conversely, agree with the application judge’s conclusion that the TTC Act violates s. 2(d) of the Charter, but submit that his case-specific and evidence-based analysis was more complicated than it needed to be. They argue that after the majority decision in SFL, legislation that entirely removes the right to strike after the expiry of a collective agreement, as the TTC Act does, will necessarily violate s. 2(d) of the Charter.

[38] For reasons I will now explain, I agree with the respondents on this legal point.

[39] It is well-established that in general, the question of whether legislation infringes the s. 2(d) Charter right to collective bargaining must be determined by conducting a two-part inquiry into whether the law interferes with activities that fall within the scope of the s. 2(d) right and, if so, whether the legislation “substantially interferes” with the right: see e.g., Société des casinos du Québec inc. v. Association des cadres de la Société des casinos du Québec, 2024 SCC 13, at paras. 17-37. This ordinarily requires a contextual and fact-specific inquiry: Health Services and Support – Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391, at para. 92 (“Health Services”). This was the approach the application judge took in his s. 2(d) analysis.

[40] However, as I read Abella J.’s majority reasons in SFL, she held that because of the importance of the right to strike to the collective bargaining process,any complete ban on unionized workers’ ability to strike after the expiry of a collective agreement will invariably “substantially interfere” with their s. 2(d)-protected collective bargaining rights. This does not require a case-specific inquiry into precisely how eliminating the right to strike has affected the collective bargaining process in the particular circumstances. Moreover, while the question of whether interest arbitration serves as a constitutionally adequate substitute for the right to strike may be an important factor in the s. 1 justification analysis, it has no bearing on the threshold question of whether eliminating the right to strike violates s. 2(d).

[41] It follows that the application judge’s detailed assessment of the evidence about how the TTC Act has affected collective bargaining between the TTC and its unionized employees was unnecessary at the s. 2(d) stage of his Charter analysis. However, the time he devoted to this was not wasted, because he would still have had to consider this evidence once he got to his s. 1 analysis.

(1) The evolution of the s. 2(d) Charter jurisprudence on the right to strike.

[42] In a 1987 trilogy of early Charter labour law decisions, a majority the Supreme Court of Canada held that the s. 2(d) Charter right to freedom of association did not protect either the right to collective bargaining or the right to strike: see Reference re Public Service Employee Relations Act (Alta.), 1987 88 (SCC), [1987] 1 S.C.R. 313 (the “Alberta Reference”); PSAC v. Canada, 1987 89 (SCC), [1987] 1 S.C.R. 424; RWDSU v. Saskatchewan, 1987 90 (SCC), [1987] 1 S.C.R. 460.

[43] Dickson C.J.C. and Wilson J. wrote separate reasons in this trilogy in which they disagreed with the majority on both of these issues.[^6] As I will discuss, Dickson C.J.C.’s dissenting legal analysis is particularly important, because it has now been adopted by a majority of the Supreme Court of Canada.

[44] In his dissenting reasons in the Alberta Reference, which Wilson J. joined, Dickson C.J.C. explained that he would have interpreted s. 2(d) as protecting both the right of employees to collectively bargain and their right to strike in support of the collective bargaining process. As he explained at p. 371:

I am satisfied, in sum, that whether or not freedom of association generally extends to protecting associational activity for the pursuit of exclusively pecuniary ends—a question on which I express no opinion—collective bargaining protects important employee interests which cannot be characterized as merely pecuniary in nature. Under our existing system of industrial relations, effective constitutional protection of the associational interests of employees in the collective bargaining process requires concomitant protection of their freedom to withdraw collectively their services, subject to s. 1 of the Charter. [Emphasis added.]

[45] The three Alberta statutes at issue in the Alberta Reference were broadly similar to the TTC Act, in that they each prohibited strikes, and required at least some disputes to go to binding arbitration. Importantly, Dickson C.J.C. did not find it necessary to conduct any detailed case-specific analysis of how the prohibitions on striking in the three statutes had affected the collective bargaining process since their enactment. Instead, he merely stated at p. 372:

These provisions directly abridge the freedom of employees to strike and thereby infringe the guarantee of freedom of association in s. 2(d) of the Charter.

[46] In its 2007 decision in Health Services, the Supreme Court of Canada partially reversed the majority holdings in the 1987 labour trilogy, stating “that the holdings in the Alberta Reference and PIPSC excluding collective bargaining from the scope of s. 2(d) can no longer stand”, and that s. 2(d) of the Charter should now be understood as protecting “the capacity of members of labour unions to engage, in association, in collective bargaining on fundamental workplace issues”: at paras. 19, 36. However, citing the Court’s earlier decision in Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, the majority in Health Services added, at para. 90:

Section 2(d) of the Charter does not protect all aspects of the associational activity of collective bargaining. It protects only against “substantial interference” with associational activity….

This is the origin of the “substantial interference” test that now governs s. 2(d) Charter claims involving the right to collective bargaining: see Société des casinos du Québec inc., at paras. 17-37.

[47] The final step of the Supreme Court of Canada’s retreat from the 1987 labour trilogy came with the Court’s 2015 decision in SFL, where the majority reversed the second major holding of the 1987 trilogy and held that s. 2(d) of the Charter should now be understood as also protecting the right to strike.

[48] In her majority reasons in SFL, Abella J. explained at paras. 2-3 that the right to strike is essential to the rights protected by s. 2(d):

The question in this appeal is whether a prohibition on designated employees participating in strike action for the purpose of negotiating the terms and conditions of their employment amounts to a substantial interference with their right to a meaningful process of collective bargaining and, as a result, violates s. 2(d) of the Charter. The question of whether other forms of collective work stoppage are protected by s. 2(d) of the Charter is not at issue here.

The conclusion that the right to strike is an essential part of a meaningful collective bargaining process in our system of labour relations is supported by history, by jurisprudence, and by Canada’s international obligations. … The right to strike is not merely derivative of collective bargaining, it is an indispensable component of that right. It seems to me to be the time to give this conclusion constitutional benediction.

[49] Abella J. also explained that she was following the path that had first been laid out by Dickson C.J.C. in his dissenting reasons in the Alberta Reference, noting at para. 75 of her reasons that:

[The] historical, international, and jurisprudential landscape suggests compellingly to me that s. 2(d) has arrived at the destination sought by Dickson C.J. in the Alberta Reference, namely, the conclusion that a meaningful process of collective bargaining requires the ability of employees to participate in the collective withdrawal of services for the purpose of pursuing the terms and conditions of their employment through a collective agreement. Where good faith negotiations break down, the ability to engage in the collective withdrawal of services is a necessary component of the process through which workers can continue to participate meaningfully in the pursuit of their collective workplace goals. In this case, the suppression of the right to strike amounts to a substantial interference with the right to a meaningful process of collective bargaining. [Emphasis added.]

[50] As I read Abella J.’s reasons, she found that the ability to strike after the expiry of a collective agreement is so important to the collective bargaining process that eliminating it entirely will necessarily violate s. 2(d) of the Charter. While this still involves an application of the two-part Dunmore/Health Services “substantial interference” test, it does not require a case-specific inquiry into how the elimination of the right to strike in a particular case has actually affected collective bargaining. Since the right to strike is now recognized as an integral aspect of the s. 2(d) right to collectively bargain, any law that eliminates the right to strike entirely will “substantially interfere” with the affected workers’ s. 2(d) fundamental freedoms.

(2) The TTC Act violates s. 2(d) by eliminating the “right to strike” at the end of a collective agreement

[51] Ontario emphasizes that the Saskatchewan legislation that was at issue in SFL – the Public Service Essential Services Act, S.S. 2008, c. P-42.2(“the PSESA”) – did not make binding interest arbitration compulsory or provide any other method for resolving disputes between the government and the public service unions to whom the legislation applied. According to Ontario, this was why the PSESA violated s. 2(d) of the Charter, and why the prohibition on strikes in the TTC Act does not. As summarized in its factum, Ontario’s argument is that:

Precluding strikes without substituting a fair and effective alternative, such as arbitration, eviscerates union bargaining power, especially in cases of impasse. The failure to provide an alternative was the unjustified s. 2(d) breach in SFL. [Emphasis in original.]

[52] The core underlying premise of Ontario’s argument is its contention that by replacing strikes with compulsory binding interest arbitration, the TTC Act “equalizes bargaining power as both sides face the risk of being unsuccessful before the neutral arbitrator.” According to Ontario, binding interest arbitration “does not undermine employee bargaining power at all”, and therefore should be treated as if it were the functional equivalent of the right to strike, such that legislatures can freely swap one for the other without needing to justify the substitution under s. 1 of the Charter.

[53] In my view, Ontario’s argument that eliminating the right to strike does not violate s. 2(d) of the Charter as long as the right to strike is replaced with a “fair and effective substitute” is based on a misreading of Abella J.’s reasons in SFL. The failure of the Saskatchewan legislation to provide any “fair and effective alternative” to the right to strike was indeed what led Abella J. to find “an unjustified breach” of s. 2(d) of the Charter: that is, a s. 2(d) violation that could point not be justified under s. 1. However, her reasons make clear that the presence or absence of a fair and effective alternative to striking in the legislation had no bearing on the threshold s. 2(d) breach analysis, and only became significant at the s. 1 justification stage.

[54] In this regard, Abella J.’s majority reasons in SFLmust be read alongside Dickson C.J.C.’s dissenting reasons in the Alberta Reference, which Abella J. substantially adopted. The three statutes that were at issue in the Alberta Reference, unlike the legislation at issue in SFL, did provide for binding interest arbitration. For various reasons the arbitration scheme in the Alberta statutes was less extensive than the arbitration provisions in the TTC Act: the government had to take active steps to trigger the arbitration process, and the legislation made some issues non-arbitrable. It was these deficiencies that would have led Dickson C.J.C. to find that the Alberta legislation could not be justified under s. 1.

[55] For present purposes, however, the important point is that Dickson C.J.C. viewed the adequacy of the arbitration scheme as solely relevant to s. 1. As he explained at p. 372 of his dissenting reasons in the Alberta Reference, the legislative provisions at issue “directly abridge the freedom of employees to strike and thereby infringe the guarantee of freedom of association in s. 2(d) of the Charter” (emphasis added).

[56] Abella J. adopted this approach in her majority reasons in SFL. She explained at para. 60:

Alternative dispute resolution mechanisms … are generally not associational in nature and may, in fact, reduce the effectiveness of collective bargaining processes over time. Such mechanisms can help avoid the negative consequences of strike action in the event of a bargaining impasse, but as Dickson C.J. noted in RWDSU v. Saskatchewan, they do not, in the same way, help to realize what is protected by the values and objectives underlying freedom of association:

. . . as I indicated in the Alberta Labour Reference, the right to bargain collectively and therefore the right to strike involve more than purely economic interests of workers . . . . [A]s yet, it would appear that Canadian legislatures have not discovered an alternative mode of industrial dispute resolution which is as sensitive to the associational interests of employees as the traditional strike/lock-out mechanism . . . .

That is why, in the Alberta Reference, Dickson C.J. dealt with alternative dispute resolution mechanisms not as part of the scope of s. 2(d), but as part of his s. 1 analysis. [Citations omitted, emphasis added.]

Abella J. went on to characterize the right to strike as “the ‘irreducible minimum’ of the freedom to associate in Canadian labour relations”: SFL, at para. 61.

[57] Earlier in her reasons, at para. 46, Abella J. had observed that the suppression of the right to strike amounts to a substantial interference with collective bargaining:

[I]t should come as no surprise that the suppression of legal strike action will be seen as substantially interfering with meaningful collective bargaining. That is because it has long been recognized that the ability to collectively withdraw services for the purpose of negotiating the terms and conditions of employment — in other words, to strike — is an essential component of the process through which workers pursue collective workplace goals. [Emphasis added.]

After reviewing academic and judicial commentary, she then concluded, at para. 51 that “the right to strike is constitutionally protected because of its crucial role in a meaningful process of collective bargaining.”

[58] Later on in her reasons, at para. 78, Abella J. explained that while the “substantial interference” test from Dunmore and Health Services continues to govern, legislation that entirely eliminates the ability of workers to strike in support of their collective bargaining efforts will meet the substantial interference test:

The test, then, is whether the legislative interference with the right to strike in a particular case amounts to a substantial interference with collective bargaining. The PSESA demonstrably meets this threshold because it prevents designated employees from engaging in any work stoppage as part of the bargaining process. It must therefore be justified under s. 1 of the Charter. [Italics in original; underlining added.]

[59] Significantly, Abella J. did not suggest that the “the absence of a meaningful dispute resolution mechanism to resolve bargaining impasses” in the PSESA had any bearing on the s. 2(d) Charter analysis. Rather, she treated this as relevant only to the separate issue of whether the legislation could be justified under s. 1: see SFL, at para. 81.

[60] In summary, I am satisfied that a close examination of Abella J.’s reasons in SFL, including her adoption of Dickson C.J.C.’s dissenting reasons in the Alberta Reference,leads to the conclusion that any law that entirely eliminates employees’ right to strike after the end of a collective agreement will necessarily infringe their s. 2(d) Charter rights. The legislation may still be constitutional – that is, justified under s. 1 – if it provides for an alternative dispute resolution mechanism, such as binding interest arbitration. However, the adequacy of any legislative substitute for the right to strike only comes into play at the s. 1 justification stage of the Charter analysis.

[61] Significantly, the Court of Appeal for Québec recently reached this same conclusion in Alliance des professionnels et des professionnelles de la Ville de Québec c. Procureur général du Québec, 2023 QCCA 626, at para. 93-97, leave to appeal to SCC refused (April 11, 2024). This decision is not binding on me, but I find Mainville J.A.’s interpretation of this aspect of SFL persuasive.

[62] I would add further that the SFL majority’s repeated description of the “right to strike” as a newly-recognized component of the s. 2(d) Charter right of freedom of association would make little sense if legislatures were always free to replace the ability to strike with compulsory binding interest arbitration, without having to justify the substitution under s. 1. Among other things, this would mean that legislatures could abolish the right to strike without any need to demonstrate that they were acting in pursuit of a “pressing and substantial” legislative objective. If this were so, it would be hard to see how the ability to strike could properly be characterized as a “right”, let alone one that Abella J. described in SFL as “an essential part of a meaningful collective bargaining process”, and “an indispensable component” of the s. 2(d) right to collective bargaining: SFL, at para. 3 (emphasis added).

[63] Since the TTC Act entirely eliminates TTC employees’ ability to strike during the collective bargaining process, it necessarily follows that the legislation “substantially interferes” with their s. 2(d) collective bargaining rights to such an extent that these rights are infringed.

[64] It follows that it was not necessary at this stage of the Charter analysis for the application judge to make any case-specific inquiry into exactly how the TTC Act’s removal of the right to strike has affected collective bargaining since 2011. Rather, the application judge could have simply found a breach of s. 2(d) based on the TTC Act’s complete elimination of TTC employees’ right to strike, and then gone on to consider whether this breach could be justified under s. 1.

D. Is the breach of s. 2(d) justified under S. 1 of the Charter?

[65] Once it is determined that the TTC Act violates s. 2(d) of the Charter, the burden shifts to Ontario to justify the infringement under s. 1.

[66] The framework for deciding whether legislation that breaches a Charter right can be justified under s. 1 is well-established. As Karakatsanis and Martin JJ. explained in their majority reasons in R. v. Ndhlovu, 2022 SCC 38, 419 C.C.C. (3d) 285, at paras 118-19:

To meet its burden under s. 1, the Crown must show the infringement is “demonstrably justified”, which means the infringing measures must be justified based on a “rational inference from evidence or established truths.” Bare assertions will not suffice: evidence, supplemented by common sense and inference, is needed.

A breach of the Charter is justified under s. 1 when the challenged law has a “pressing and substantial object and . . . the means chosen are proportional to that object.” The law is proportionate where the means adopted are rationally connected to the law’s objective, minimally impairing of the right in question, and the law’s salutary effects outweigh its deleterious effects. [Citations omitted.]

(1) Pressing and substantial object

[67] The parties disagree about how the legislative object of the TTC Act should be characterized. According to Ontario, “the Act’s objective is preventing disruptions of TTC services.” The respondents disagree, arguing that this stated objective is overly broad and “conflate[s] the objective with its means.”

[68] I agree with the respondents on this point. As Cory J. explained in U.F.C.W., Local 1518, v. KMart Canada Ltd., 1999 650 (SCC), [1999] 2 SCR 1083, at para. 59, the legislative objective must be defined with precision:

For the purpose of the Oakes test, a legislative objective must not be overstated. It must be accurately and precisely defined so as to provide a clear framework for evaluating its importance, and to assess the precision with which the means have been crafted to fulfil that objective. If the objective is stated too broadly, its importance may be exaggerated and the analysis compromised.

(2) Proportionality

[97] The second stage of the Oakestest requires courts to assess whether “the means chosen [to achieve the legislature’s objectives] are reasonable and demonstrably justified”: Oakes, at p. 139. This involves “a form of proportionality test”, in which courts must consider: (i) whether the means chosen are rationally connected to the objectives; (ii) whether they impair the right at issue as little as possible; and (iii) whether the salutary effects of the legislation outweigh its deleterious effects. The party seeking to uphold legislation – usually the government, and here, Ontario – must succeed on all three branches of the proportionality test.

(1) Rational connection

[98] At the rational connection stage of the Oakes proportionality analysis, the government bears the burden of showing “a causal connection between the infringement and the benefit sought on the basis of reason or logic”: RJR-MacDonald Inc. v. Canada (Attorney General), 1995 64 (SCC), [1995] 3 S.C.R. 199, at para. 153. As McLachlin C.J.C. noted in Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, at para. 48, “[t]he government must show that it is reasonable to suppose that the limit may further the goal, not that it will do so.”

[99] In OECTA, at para. 187, Favreau J.A. explained that “[t]he evidentiary burden at this stage is ‘not particularly onerous.’ Direct proof of a causal relationship between the measure and the objective is not required” (Citations omitted.)

[100] The application judge found that Ontario had not met its burden at the rational connection stage. After noting that “[t]he ‘hallmarks’ of rational connection are ‘care of design’ and ‘a lack of arbitrariness’”, he concluded:

I am of the view that the evidence supports the conclusion that there was a lack of care taken by the Government when the TTC Act was enacted.

The application judge discussed the speed with which the TTC Act had been enacted, noting that there was “no evidence that the Government conducted extensive consultations or meaningful discussions with the TTC unions before the legislation was introduced.” He concluded:

It is my view that the lack of care is evident in the way in which the legislation prohibits the right to strike of all TTC employees regardless of whether the employee’s job has any connection with the “pressing and substantial” objective. For example, it is difficult to see how a strike by customer service agents or painters would affect the health and safety of the public.

[101] In my view, it was an analytic error for the application judge to take this approach at this stage of the Oakes inquiry. Instead, his focus should have been on whether it was reasonable for the legislature to suppose that its chosen means – namely, banning strikes and lockouts at the TTC – would advance its objectives. Whether the government could or should have engaged in more extensive consultations with the unions, and whether its goals could have been adequately achieved by a more carefully tailored strike ban, were both questions that properly fell to be considered later, at the minimal impairment stage of the Oakesanalysis rather than at the rational connection stage.

[102] I recognize that in OECTA,a majority of this court held that overly broad legislation can sometimes raise rational connection concerns. That case involved wage restraint legislation directed at “the responsible management of the province’s finances and the protection of sustainable public services.” The majority found that the legislature’s decision to include some workers in the energy and academic sectors was not rationally connected to this objective, since these workers were not paid directly by the province, and there was no reason to imagine that increasing their pay would indirectly cause the province to have to pay more money to their employers.

[103] The situation in the case at bar is meaningfully different. Even if strikes by certain TTC workers – such as customer service agents or painters, to use the application judge’s example – would not directly jeopardize public health and safety, the environment, or the economy, the legislature could still rationally conclude that its objectives would be advanced by a scheme under which all TTC bargaining disputes were resolved by binding arbitration, rather than by an approach that would treat different groups of TTC employees differently based on their job descriptions, even if they were all part of the same bargaining unit. The analytically distinct question of whether the legislature was justified in taking this broad-brush approach is better assessed at the minimal impairment stage of the Oakesanalysis.

[104] Significantly, this was what the Supreme Court of Canada majority did in SFL, where one of the objections to the PSESA was that it allowed the government to unilaterally designate which workers were considered “essential”. Abella J. found that the legislation was rationally connected to the legislative objective of preserving essential public services, although she then found that it failed the minimal impairment step of the Oakesanalysis.

[105] In summary, I conclude that the application judge erred in law by finding that Ontario had not established a rational connection between the means chosen and the legislature’s objectives. However, I agree with the respondents that this error was more one of classification than of substance, since the considerations that led the application judge to find a lack of rational connection were all ones he was entitled to consider at the minimal impairment stage of his analysis.

(2) Minimal impairment

(1) The legal framework

[106] In Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331, at para. 102, the Supreme Court of Canada provided the following summary of the minimal impairment stage of the Oakes proportionality analysis:

At this stage of the analysis, the question is whether the limit on the right is reasonably tailored to the objective. The inquiry into minimal impairment asks “whether there are less harmful means of achieving the legislative goal”. The burden is on the government to show the absence of less drastic means of achieving the objective “in a real and substantial manner.” The analysis at this stage is meant to ensure that the deprivation of Charter rights is confined to what is reasonably necessary to achieve the state’s object. [Citations omitted.]

[107] In OECTA, at para. 200, Favreau J.A. summarized the governing principles in the following terms:

In Libman v. Quebec (Attorney General), the Supreme Court emphasized that “the law must be carefully tailored so that rights are impaired no more than necessary.” While the court accords deference to the legislature’s choices, deference does not insulate the government from having to demonstrate that an impugned measure is minimally impairing and justified under s. 1. At the same time, however, legislators are not held to a level of perfection; the court should not find a law minimally impairing because it can “conceive of an alternative which might better tailor the objective to infringement.” [Citations omitted.]

[108] Whether legislation meets the minimal impairment requirement of Oakes is a question of law, reviewable on a correctness standard. However, to the extent that this question depends on findings of fact, these findings attract appellate deference and are reviewable on the palpable and overriding error standard.

(2) The application judge’s reasons

[109] Citing Hutterian Brethren, the application judge noted that at the minimal impairment stage of the Oakesanalysis, “the government is entitled to deference in formulating its objective and the manner in which it seeks to achieve the objective”, but that this deference is not absolute. He then found that Ontario had failed to meet its burden at this stage of the justification analysis.

[110] The application judge relied on the evidence establishing that no other Canadian municipal transit system employees have had their right to strike legislated away entirely. Referring to the example of Metrolinx, which “operates the GO trains which provide public transit into the city of Toronto”, he noted:

In the case of Metrolinx, the province did not ban all strike action, but instead adopted a tailored approach to preserve the right to strike as much as possible while protecting legitimate public health and safety interests.

Continuing, the application judge observed:

The TTC is the only municipal public transit service in Canada whose workers are completely prohibited from engaging in all forms of strike action at any time, and regardless of the job performed by the employee. I find the fact that the other transit systems do not have a complete ban on the right to strike to be most persuasive. I appreciate that the TTC is the largest and most complex transit system in the country, but transit disruptions in other cities would also impact the local economies, and likely affect equity-seeking groups disproportionately. The province has not removed the right to strike of transit workers in those jurisdictions but has developed other means to achieve their objectives.

[111] In addition, the application judge had earlier commented in his rational connection analysis on the haste with which the TTC Act was enacted. He returned to this point in his minimal impairment analysis, noting that the TTC Act “was enacted on the day before the first round of bargaining following the 2008 strike.” He concluded:

I am of the view that the Government failed to establish that the removal of the right to strike for all TTC employees is “minimally impairing”. There is no evidence that there was consultation or study to identify other methods to achieve the objectives without completely removing the right to strike. I find that the legislation is a blunt instrument and does not provide a tailored and nuanced approach to the issue.

[112] In particular, the application judge found as fact that the legislature chose to impose a complete strike ban accompanied by interest arbitration without first giving serious consideration to what he referred to as the “hybrid model”, namely:

[A] model by which all employees could technically strike when the agreement expires but in advance the parties have negotiated an essential services agreement which identifies those persons required to remain on the job during a work stoppage.

[113] I see no basis for interfering with the application judge’s factual conclusion that the Ontario legislature adopted a complete strike ban without first studying or seriously considering the alternative “hybrid model”. This finding is well-supported by the evidential record.

(3) Minimal Impairment Analysis

[114] Ontario contends that the application judge erred by not recognizing that the government’s objective was to prevent all TTC service disruptions, and that no other alternative approach, such as the “hybrid model” that governs Metrolinx and the transit systems in some other jurisdictions, including Montreal, would have achieved this objective. Ontario argues further that “[t]he evidence is that the hybrid model would not be suitable for the TTC”, because the Toronto transit system is too large and complex to function properly on a reduced-service basis. According to Ontario:

While [the application judge] also noted the hybrid labour models used for Metrolinx (GO Transit) and Montreal’s transit system, his reasons fail to meaningfully analyze whether these models would achieve the Legislature’s objective. The evidence was that the TTC’s view was that a hybrid model applied to its system would result in safety risks due to overcrowding and would not provide sufficient leverage to achieve an agreement at impasse. Ontario’s expert evidence demonstrated that traffic and pollution increased even when a partial strike (hybrid) was modelled and transit times were dramatically longer. The evidence also showed that impacts of a TTC strike will be greater than transit strikes in other jurisdictions.

[115] I accept the self-evident proposition that nothing short of a complete strike ban would permit the TTC to keep running at full capacity, with no service disruptions or delays whatsoever. However, as I have already discussed, I am not satisfied that the legislative objective here can properly be framed as preventing all service disruptions, as opposed to preventing the harms the legislature believed these disruptions would cause to public health and safety, the environment, and the economy.

[116] The first problem Ontario faces is that the application judge was not satisfied by the evidence before him that the legislature’s concerns were objectively justified. His factual conclusions are entitled to substantial appellate deference. Even though the application judge made these factual findings as part of his “pressing and substantial objective” analysis, which was a legal error, they are findings he was entitled to make and rely on at both the minimal impairment and proportionality stages of the Oakesanalysis.

[117] The second problem Ontario faces, closely related to the first, is that the application judge did not accept that the harms that would flow from a partial TTC system shutdown (such as might have resulted if the legislature had adopted a “hybrid” strike ban model) were so severe that they justified imposing a full strike ban. Put simply, he did not accept Ontario’s contention that the Toronto transit system is so radically different from the transit systems in other large Canadian cities that the legislative model that is used to regulate strikes elsewhere would not work in Toronto. This is also a factual finding that is entitled to deference.

[118] On the first point, the application judge concluded that the evidence left him unsatisfied that a TTC strike would in fact jeopardize public health or safety. He noted that Ontario had not presented any evidence that the 2008 TTC strike, which occurred on very short notice, had “resulted in actual harm to the public.” On the issue of whether a transit strike would result in “negative health effects because of an increase in air pollution”, he found “the evidence of the [unions’] expert, Dr. Villeneuve, to be more persuasive on this issue” than the evidence of Ontario’s two experts. The application judge explained:

[Dr. Villeneuve] reviewed the actual air pollution and hospitalization data during the strikes in 2006 and 2008 and found that there was “no evidence” that past TTC strikes “adversely impacted the health of residents” of Toronto. Although those work stoppages were only 1-2 days in length, I note that the history of strikes involving the TTC are usually of short duration. Dr. Villeneuve also noted that there may be health benefits to a strike including a shift to walking and cycling.

[119] The application judge concluded that he was not persuaded that a TTC strike would lead to the harms contemplated by Ontario:

I am satisfied that the Government failed to establish that the TTC meets the definition of “essential services” as set out by Dickson C.J. in Alberta Labour Reference. An “essential service” is “one the interruption of which would threaten serious harm to the general public or to a part of the population. … [and] ‘would endanger the life, personal safety or health of the whole or part of the population’”. There is no persuasive evidence before me that a TTC strike would endanger the life, personal safety, or health of the population. [Citations omitted.]

[120] On the question of whether a TTC strike would cause serious economic harm, the application judge found that this harm was not borne out on the evidence:

I find the evidence put forward by the Government to support its position that there are serious economic consequences of a TTC strike to be lacking. There is no current report that sets out a detailed analysis of the economic consequences of a strike. It relies on two reports prepared in 2008; one by the City of Toronto, and one on behalf of the union. The reports were prepared 15 years ago and provide only broad estimates of the economic consequences of a TTC strike, without any detailed analysis.

I do not find, on the evidence before me, that the effect of a TTC strike will result in “serious” and “especially injurious” economic consequences.

[121] As I have already discussed, the application judge held that Ontario could not rely on the protection of equity-seeking groups as a legislative objective, which was an analytic error. However, his conclusion that Ontario had not met its burden of establishing that TTC strikes would cause any serious harm to the public at large implies that he was also not satisfied that members of equity-seeking groups would suffer these harms disproportionately. I am not persuaded that the application judge’s factual conclusions on these issues are tainted by any palpable and overriding error. I am also satisfied that he did not make any reversible legal errors when he applied the minimal impairment test to his findings of fact.

[122] Ontario argues that the application judge erred by placing undue reliance on the evidence that at least some strike action is permitted in all other Canadian transit systems. According to Ontario:

While evidence that describes other jurisdictions’ labour relations models may be relevant to determining whether a measure is minimally impairing, it does not establish that these other models as applied to the TTC would be, in fact, less impairing of the right to strike while also achieving the government’s objective. Moreover, such descriptive information does not provide evidence of those models’ effectiveness in those other jurisdictions. [Footnotes omitted; emphasis in original.]

[123] This argument fails to account for the onus of proof. It was Ontario’s burden to affirmatively show that the legislative models that have historically been used in other jurisdictions would not work in Toronto, because they would not adequately achieve the government’s objectives. The application judge was not satisfied that the evidence demonstrated this, nor was he satisfied that the Ontario legislature had made a considered decision to reject these other models. Even accepting that Ontario was not required to show more than that its chosen approach fell “within a range of reasonable alternatives” (RJR-MacDonald, at para. 160; see also: Hutterian Brethren, at para. 53), the application judge was not satisfied that the Ontario legislature had made a reasoned decision to reject other established less rights-impairing alternatives. Put another way, he was not persuaded that a complete strike ban was a “reasonable alternative”. I see no basis for interfering with what were essentially fact-driven and evidence-based conclusions.

[124] In summary, Ontario has not established that the application judge made any reversible errors in concluding that it had failed to meet its burden under the minimal impairment branch of the Oakes proportionality analysis. While this conclusion is sufficient on its own for me to conclude that Ontario’s appeal must fail, I will for completeness go on to consider its arguments concerning the third branch of the Oakes proportionality test.

(3) Proportionality between salutary and deleterious effects

(1) General principles

[125] As Favreau J.A. explained in OECTA, at para. 215, the final stage of the Oakes analysis engages a balancing exercise:

This last branch of the s. 1 analysis asks whether there is proportionality between the overall effects of the Charter-infringing measure and the legislative objective. The court must turn its mind to the effects of the measure to determine, on a normative basis, whether the infringement of the right in question can be justified in a free and democratic society. This requires a balancing between the measure’s salutary and deleterious effects … [T]he Supreme Court explained the court’s task in the following terms:

It is only at this final stage that courts can transcend the law’s purpose and engage in a robust examination of the law’s impact on Canada’s free and democratic society “in direct and explicit terms”…. In other words, this final step allows courts to stand back to determine on a normative basis whether a rights infringement is justified in a free and democratic society. Although this examination entails difficult value judgments, it is preferable to make these judgments explicit, as doing so enhances the transparency and intelligibility of the ultimate decision. Further, as mentioned, proceeding to this final stage permits appropriate deference to Parliament’s choice of means, as well as its full legislative objective. [Citations omitted.]

[126] The normative and value-laden balancing of the relevant factors is a question of law that is reviewable on a correctness standard. However, to the extent that the assessment of the salutary and deleterious effects of legislation depends on factual findings, the findings made by the application judge must be respected unless they are tainted by palpable and overriding error.

(2) The application judge’s reasons

[127] The application judge relied on Dickson C.J.C.’s observation in RWDSU, at pp. 477-478, that in view of the importance of the right to strike as an aspect of the protected s. 2(d) Charter right:

[T]he relevant question [on proportionality], therefore, is whether the potential for economic harm to third parties during a work stoppage is so massive and immediate and so focused in its intensity as to justify the limitation of a constitutionally guaranteed freedom in respect of those employees.

[128] The application judge determined that the salutary effects of the TTC Act were minimal:

As I stated earlier in these reasons when considering the “pressing and substantial” objective, I am of the view that the evidence does not support a finding that a TTC strike would “threaten serious harm” or “endanger the life, personal safety or health” of the whole or part of the population.

With respect to the purported economic costs of a TTC strike, I find that the evidence relied on by the Government to be [sic] inadequate. As noted earlier, the Government did not put forward an expert to provide an opinion on the economic consequences of a strike. Instead, it relied on two reports prepared in 2008. Both reports provide broad statements about the economic effects of a strike, without any rigorous analysis. For example, there is no analysis as to whether the loss would be made up when the strike was over.

The reports are now 15 years old. More importantly, there is no report that deals with the options workers now have in the event of a TTC strike. I am prepared to take judicial notice of the fact that many office workers were able to work remotely during the pandemic.

[129] Although the application judge had previously rejected Ontario’s contention that the protection of equity-seeking groups was one of the underlying legislative purposes of the TTC Act, he nevertheless addressed this objective in his proportionality analysis, stating:

I agree with the submissions of the Government that equity-seeking groups will experience a TTC strike more keenly. However, I am also of the view that the measure proposed of eliminating the right to strike will have a negative effect on equity-seeking groups within the TTC. TTC workers also reflect the ethnic and gender makeup of Toronto. Those individuals may be at greater risk if the transit unions do not have a right to strike.

[130] After quoting comments by Abella J. in SFLabout the importance of the right to strike, he stated that equity considerations for workers within the unions were also relevant to the analysis:

When balancing interests, it is important to consider the equity seeking groups in the public who may be affected, as well as the equity-seeking groups within the union. Although a strike will adversely affect equity seeking groups who tend to be greater consumers of public transit, the removal of the right to strike will result in a loss of protection to marginalized or equity-seeking groups within the union and may have a negative impact on their ability to right imbalances in the workplace.

[131] The application judge considered the historical context of labour relations within the TTC, noting that the deleterious effects of strikes have been minimal, in part because “transit shutdowns are generally short lived”. He ultimately concluded that the salutary effects of the TTC Act did not outweigh its deleterious effect on workers’ s. 2(d) rights, noting:

I have weighed the benefits of the removal of the right to strike with the corresponding negative effects. I conclude, on the evidence, that the benefits that may be achieved by removing the right of TTC employees to strike do not outweigh the harm caused by the loss of the right to meaningful collective bargaining. The right to meaningful collective bargaining is to be interfered with only in the “clearest of cases” and when the harm to third parties is “massive and immediate”. I do not find on the evidence before me that this is the “clearest of cases”. I conclude that in the circumstances of this case, the balancing exercise results in a finding that the legislation is not demonstrably justified as a reasonable limit in a free and democratic society.

(3) Proportionality Analysis

[132] Ontario contends that the application judge failed to conduct the final-stage Oakes balancing exercise “appropriately”. Its first complaint is that the application judge erred by “reject[ing] cogent evidence on traffic congestion and air quality.” In essence, Ontario seeks to have this court reweigh this evidence and draw different conclusions. I am not satisfied that there is any basis for us to interfere with the application judge’s findings of fact in this manner. In my view, Ontario has not shown that the application judge’s assessment of the expert evidence on these issues was tainted by any palpable and overriding errors.

[133] Ontario’s second complaint is that the application judge “effectively neglected to give any weight to the avoidance of economic harms when weighing the Act’s salutary effects” (emphasis in original). This argument also amounts to an invitation to second-guess the application judge’s assessment of the economic harm evidence. He gave cogent reasons for discounting the economic reports that were tendered by Ontario, noting that they were both dated and lacked analytic rigour. I see no basis for interfering with his assessment of the weight this evidence deserved.

[134] Ontario’s third argument is that the application judge erred in his assessment of the impact of a TTC strike on vulnerable groups. Ontario objects that there was “no evidence below regarding the ethnic and gender makeup of TTC employees, nor was there any evidence regarding the effects of not striking or arbitration on equity-seeking groups among TTC employees” (emphasis in original).

[135] As I interpret the application judge’s comment that TTC workers “reflect the ethnic and gender makeup of Toronto”, he was not purporting to make any findings about the precise demographic composition of the TTC’s workforce. The TTC has more than 12,000 employees, and I do not think that the application judge needed specific evidence to be able to conclude, as he did, that at least some of them are members of “equity-seeking groups”.

[136] However, I do agree that it was an error for the application judge to seemingly treat the existence of these employees as a complete answer to Ontario’s contention that a TTC strike would disproportionally have a negative effect on marginalized or equity-seeking groups in the population at large. The Oakes balancing of salutary and deleterious effects did not automatically even out merely because there are persons from disadvantaged groups whose interests need to be included on both sides of the scale. To the extent that eliminating the right to strike adversely affects equity-seeking TTC employees, this deleterious effect might be justified by the salutary effect of preventing even graver harms to members of the public, including members of equity-seeking groups amongst the population.

[137] That said, I am not persuaded that this reasoning error undermined the application judge’s ultimate conclusion that Ontario had not met its burden under the final branch of the Oakestest. As I have already discussed, the application judge’s reasons as a whole make clear that he was not satisfied that Ontario had met its burden of demonstrating that any members of the public would be so seriously harmed by a TTC shutdown that their interests outweighed the competing Charter freedoms of TTC workers. This conclusion was based on his assessment of the evidence and his findings of fact, which are entitled to substantial appellate deference.

[138] Ontario’s fourth argument is that the application judge erroneously discounted the harms that would result from a TTC strike by assuming that any transit system shutdown would necessarily be brief. In its factum, Ontario argues that the application judge:

[M]isapprehended the Act’s salutary effects by emphasizing that in the 20 years before the Act’s enactment, TTC employees were on strike for a mere 12 days. In this, he failed to acknowledge his own finding that these strikes “only lasted a few days because the Government legislated TTC employees back to work”. History suggests that without legislative intervention, the brevity of transit strikes cannot be guaranteed. [Footnotes omitted.]

[139] Not all of the TTC strikes between 1991 and 2008 were ended by back-to-work legislation. Nevertheless, it is reasonable to imagine that both sides’ negotiation positions before 2011 would have been influenced by their knowledge that the Ontario legislature would probably not let a transit strike last indefinitely.

[140] To be clear, the constitutionality of hypothetical future back-to-work legislation after SFL is not before us in this appeal. Legislation that requires workers to return to work, and to submit outstanding bargaining issues to binding arbitration, raises different concerns than are presented by legislation that eliminates the right to strike from the outset. When (or if) a constitutional challenge to back-to-work legislation arises, it will have to be decided on its own particular facts and its own particular circumstances. Nothing I say here should be understood as expressing any opinion about these issues, none of which are before us.

[141] However, I accept that SFL has changed the legal landscape, and thus agree with Ontario’s observation that “[i]n the absence of the [TTC Act] and post-SFL, there is no guarantee that future labour disruptions will be brief”, although it equally cannot be assumed that they will necessarily be lengthy.

[142] I also agree with Ontario that the application judge seems to have assumed that because the TTC strikes before 2011 were brief, this would also be true of any future strikes. I agree further that this affected his assessment of the final Oakes proportionality test, since it led him to characterize the exercise as balancing the importance of TTC employees’ right to strike against “the deleterious effects of relatively brief transit shutdowns” (emphasis added). He found that the balance tipped in favour of upholding the right to strike. As a matter of common sense, any impact that a TTC service shutdown has on public health and safety, the environment, or the economy will probably increase the longer the shutdown lasts. The application judge did not consider whether the balance might eventually tip in the opposite direction if a TTC strike were to last for longer than a few days.

[143] However, I am not persuaded that the application judge’s failure to address the question of how SFL might affect the duration of future TTC strikes was a reversible error, let alone one that would allow us to overturn his fact-based conclusions and find that Ontario has met its burden on the final branch of the Oakestest.

[144] To reiterate, the narrow question on this appeal is whether Ontario has demonstrated a constitutionally sound justification for pre-emptively banning all TTC strikes. The separate question of what evidence might be needed to justify back-to-work legislation once a TTC strike is underway is not before us, and I express no opinion about this.

[145] Ontario’s argument seems to be that because it may now be more difficult after SFL for the legislature to constitutionally justify back-to-work legislation, future TTC strikes may last longer than they once did, making it more likely that the harms the legislature was trying to prevent by banning transit strikes will materialize.

[146] However, even if one were to assume that a future TTC strike could eventually cause such severe harms that back-to-work legislation would become justified under s. 1, it does not automatically follow that this gives the legislature a proper s. 1 justification to pre-emptively ban all TTC strikes before they start.

[147] Importantly, the evidence that the application judge accepted supports the conclusion that pre-emptive strike bans have a different and more severe effect on the collective bargaining process than ad hoc back-to-work legislation that is enacted only after a strike is in progress.

[148] As Professor Hebdon noted in his report, the right to strike has a significant impact on the efficacy of collective bargaining:

Collective bargaining works when the parties (labour and management) can strike or lockout. Strikes or lockouts can impose significant losses on both parties thus creating the pressure to settle. Where the right to strike exists, there are negotiated settlements without a strike in over 95 percent of the cases. [Citation omitted.]

In other words, the mere possibility that there may be a strike or lockout if negotiations fail puts pressure on both sides to reach a negotiated agreement. Even if labour and management both assume that the legislature will eventually intervene to end any strike or lockout, their shared uncertainty about whether and when this might happen can maintain some of this pressure. Former ATU Local 113 President Bob Kinnear made this point in his affidavit evidence, stating that:

[B]argaining with a right to strike, even where back to work legislation may be imposed, puts the union and its members in a much more powerful position than bargaining when the union has been stripped of the right to strike.

[149] During the pre-TTC Act negotiating era, both sides knew that there was a good chance that any TTC strike or lockout would, if necessary, be ended relatively quickly by back-to-work legislation. However, the application judge accepted the evidence of Mr. Kinnear and the other union officials that the collective bargaining process between the TTC and its unions deteriorated once the TTC Act was enacted in 2011 and the right to strike was eliminated entirely.

[150] Having regard to the application judge’s factual findings, I am not persuaded that his assumption that future strikes would be short undermines his overall conclusion that Ontario had not met its s. 1 burden of justifying a full pre-emptive strike ban. Once a TTC strike has started, Ontario may be able to present better evidence about the actual impact of the strike on public health and safety, the environment, and the economy. Depending on what this evidence shows, Ontario may in the future be constitutionally justified in preventing these harms from continuing by ordering TTC employees back to work. However, the possibility that Ontario might eventually be able to justify ending a future TTC strike with back-to-work legislation does not imply that the application judge was wrong to find that Ontario had not met its burden of justifying the TTC Act’s pre-emptive strike ban, based on his assessment of how the deleterious impact of the ban on TTC employees balanced against the government’s interest in avoiding the public harms that might flow from a transit shutdown.

(3) Conclusions on s. 1

[151] In summary, while I agree with Ontario that the application judge made some legal errors in his s. 1 Charter analysis, I am not persuaded that any of these errors fatally undermined his conclusion that Ontario had not met its burden of justifying the breach of TTC workers’ s. 2(d) Charter rights. The main thrust of the application judge’s reasons is that Ontario’s evidence did not persuade him that a full pre-emptive ban on TTC strikes was reasonably needed to protect public health and safety, the environment, and the economy. He concluded that the public harms the legislature feared would be caused by a full or partial TTC shutdown were speculative and unproven, whereas the impact of the strike ban on TTC employees’ s. 2(d) Charter rights was significant. Although the issue of whether legislation is justified under s. 1 of the Charter is a question of law, reviewable on a correctness standard, the application judge’s underlying findings of fact are entitled to appellate deference. I am not persuaded that his factual findings reveal any palpable and overriding errors on any essential points, nor am I persuaded that his conclusion that Ontario had not met its burden under s. 1, based on the facts as he found them, was wrong in law.

E. Disposition

[152] I would accordingly dismiss the appeal and uphold the application judge’s finding that the TTC Act violates s. 2(d) of the Charter in a manner that has not been justified under s. 1. I would also uphold his declaration that the Act is of no force or effect under s. 52 of the Constitution Act, 1982.

[153] If the parties are unable to agree on costs, they may make brief written submissions limited to five pages each. The respondents will file their submissions within 10 days of the date of these reasons and the appellant will file their submissions within 10 days thereafter. No reply submissions are to be filed.

“J. Dawe J.A.”

“I agree. J. Copeland J.A.”

Nordheimer J.A. (dissenting):

A. Charter s. 2(d)

[154] I am prepared to accept my colleagues’ conclusion that the TTC Act constitutes a breach of the respondents’ rights under s. 2(d) of the Charter. Some of the language used in SFL might allow for a conclusion that a prohibition on the right to strike, if coupled with a “meaningful alternative mechanism for resolving bargaining impasses”, does not constitute an infringement of s. 2(d). Nevertheless, I defer to my colleagues’ point that a fair reading of the decision as a whole does not allow for that conclusion. I also do not need to engage in the debate over s. 2(d) because, in my view, the issue in this case can be resolved by the application of s. 1.

[155] Reaching that conclusion means that I do not have to address the arguments advanced by the respondent respecting levels of grievances and impacts on union democracy. I will, however, say that I view it as a dubious proposition that one can attach motives to why grievances are filed. Presumably there are as many reasons for the filing of grievances as there are grievances filed. I consider it similarly dubious to ascribe motives to bargaining stances taken. There can be many reasons why either side could decide to take a harder position than it had in the past. In any event, it remains the fact that neither side, regardless of the hardness of their position, can predict with any realistic measure of assurance what the contents of any award from an arbitrator may be. Consequently, taking a hardline position may not lead to a positive result in the arbitrator’s award. Finally, the suggestion that interest arbitration weakens union democracy is an unproven one. I note, in that regard, that the experts proffered in this case did not agree on that proposition.

[156] In any event, and as I have said, the issue whether a meaningful alternative mechanism can ameliorate a s. 2(d) breach is properly determined under the s. 1 analysis, namely, whether the limit on the right is reasonable and demonstrably justified. It is on that issue that I part company with my colleagues.

B. Charter s. 1

[157] The test under the s. 1 analysis is set out in many cases but the central decision is R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103. That decision set out a two-part test. The first part (at para. 69) is whether: “[T]he objective, which the measures responsible for a limit on a Charter right or freedom are designed to serve, must be ‘of sufficient importance to warrant overriding a constitutionally protected right or freedom’ [citing R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, at p. 352].”

[158] If the objective is of sufficient importance, then the second part of the test (at para. 70) requires that “the means chosen are reasonable and demonstrably justified.” In making that determination, three factors are to be considered: (i) the means “must be rationally connected to the objective”; (ii) the means must minimally impair the right in question; and (iii) the means must be proportional to the objective.

(1) Sufficient importance

[159] In determining the first issue, the objective of the statute must be carefully defined. The parties disagree on the definition of the objective of the TTC Act. The appellant says that it is preventing disruptions of TTC services. The respondents agree with the application judge that the objective is the maintenance of an essential service.

[160] My colleagues draw a distinction between the question whether the TTC provides an essential service and whether “the harms that the government believed would arise if a TTC strike caused a transit system shutdown were so grave that the Legislature’s goal of preventing these harms can be seen as ‘pressing and substantial.’”

[161] It is not clear to me that there is a meaningful difference between these statements of the objective. Presumably the objective of preventing disruptions of TTC services and their consequent harms arose precisely because the TTC provides an essential service. While there may be tactical reasons why the appellant would prefer not to engage in the essential services debate, that does not change the fact that the contest over the proper enunciation of the objective may be more illusory than real. It seems to me that the essential services question is one that is unavoidable at the first stage of the s. 1 analysis.

[162] In any event, I do not see how the appellant could expect to uphold the legislation absent a finding that the TTC is an essential service. A prohibition on the right to strike that would be Charter compliant, outside of an essential service, would appear to be a difficult one to maintain in light of SFL. The opposite is true if the legislation deals with an essential service. This view would appear to be consistent with the reasoning in SFL where, at para. 79, the court said that the maintenance of essential public services is “self-evidently a pressing and substantial objective”. It thus falls to be determined whether the TTC is an essential service.

[163] In my view, in light of the preamble to the legislation, and the purpose of the legislation as expressed when it was introduced in the Legislature, it is clear that its objective was to maintain the operations of the TTC because it is an essential service. I disagree with the application judge’s conclusion to the contrary. I also disagree with my colleagues’ view that this error “was more one of form than of substance”. In my view, it was a fundamental error in the application judge’s reasoning and one that infected the balance of his analysis.

[164] The preamble to the TTC Act states that “work stoppages” at the TTC and the “resulting disruption of transit services give rise to serious public health and safety, environmental, and economic concerns.” It also refers to the public interest requiring a different approach.

[165] Similarly, in the introduction to the legislation, and in subsequent debates on it in the Legislature, numerous references are made to health and safety concerns, and the impact on the vulnerable, that arise from disruptions to TTC services. Economic concerns are also mentioned.

[166] In his analysis, the application judge keyed on the wording of the preamble but did not make any reference to the legislative debates. It is unclear why he did not do so since they were before him. More importantly, in assessing the purpose of legislation, extrinsic evidence such as Hansard or minutes of parliamentary debates may be considered: R. v. Sharma, 2022 SCC 39, 165 O.R. (3d) 398, at para. 88; Canadian Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3, at para. 27.

[167] On the issue of essential service, much emphasis was given by the application judge, and by the respondents, to the definition of an essential service as set out by Dickson C.J. in Reference Re Public Service Employee Relations Act (Alberta), 1987 88 (SCC), [1987] 1 S.C.R. 313. In that decision, Dickson C.J. relied heavily on the definition of essential services enunciated by decisions of the Freedom of Association Committee of the International Labour Office. Those decisions defined an essential service as one "whose interruption would endanger the life, personal safety or health of the whole or part of the population”: Reference re Public Service Employee Relations Act (Alberta), at p. 375.

[168] In my view, it is important not to engage in too rigid an application of that particular definition to the issues that confront us here. For one, Dickson C.J. did not suggest that the I.L.O. decisions were conclusive or binding in any way. It is obvious that they could not be. Rather, he referred to them as “helpful and persuasive”: at p. 375. I would also note the salient and helpful point made by McIntyre J. in that case when he observed, at p. 419: “None of these issues is amenable to principled resolution. There are no clearly correct answers to these questions. They are of a nature peculiarly apposite to the functions of the Legislature.”

[169] I appreciate that Abella J. in SFLadopted Dickson C.J.’s consideration of the definition of essential service. That does not change the fact that, fairly read, Dickson C.J. was not proposing a definition of essential service for all purposes. In that regard, it is important to remember exactly what Dickson C.J. said. While he referred to the I.L.O. decisions and found them helpful, he said, at p. 374-75: “The logic of s. 1 in the present circumstances requires that an essential service be one the interruption of which would threaten serious harm to the general public or to a part of the population.”

[170] Harm can take many forms. For the purposes of this analysis, I do not see anything in the existing authorities that stipulates that, in considering what is an essential service, one should take an unduly restrictive or narrow approach to the definition of serious harm. In particular, I believe it goes too far, in this context, to determine that serious harm must meet the precise words used by the I.L.O. Indeed, other decisions of the Supreme Court of Canada appear to have applied a broader definition of the harm required. As alluded to by McIntyre J. in Reference Re Public Service Employee Relations Act (Alberta), serious harm should be considered in the context of the purposes of the legislation, an issue to which I now turn.

[171] The harm which the TTC Act is designed to prevent is the impact that a disruption of TTC services has to both the population as a whole, and to significant portions of that population. I begin with the latter point.

[172] The evidence before the application judge demonstrated that a TTC strike has a disproportionate effect on the most vulnerable members of our society. The application judge appeared to accept that this was the case, although, for reasons that are not evident, he expressed the effect as possible rather than certain. He did note, however, that persons in lower income jobs may not have the option of working from home and that persons who cannot access vehicles have fewer transit options.

[173] The evidence before the application judge was more definitive than he allowed. In his expert report, which was the only expert evidence on this point, Dr. Farber gave his opinion that “lower-income, visible minorities, younger, immigrants, and otherwise less affluent people, will have more of their work and daily activity trips disrupted by a TTC shutdown, compared to the trips of other residents of the City.” In a second report, Dr. Farber demonstrated “the necessary role transit plays in the lives of historically marginalized populations in Toronto”. He gave examples of this role, which he drew from research he had done about transit ridership during the COVID-19 pandemic.

[174] Dr. Farber’s research found the following:

Nearly 70 percent of respondents to his survey who kept riding transit did not have access to a private vehicle.
Two-thirds of essential workers continued to use transit.
72 percent of retail workers (e.g. grocery store workers) continued to use transit.
70 percent of other essential workers (manufacturing, food preparation, agriculture, mining, maintenance and construction) continued to use transit.
55 percent of healthcare and social assistance workers continued to use transit.
43 percent of respondents said that their most important destination requiring their continued use of transit was for groceries.

[175] Dr. Farber’s research also found that women were more likely than men to find that giving up transit made them feel less independent, made it harder to get food, and made it more difficult to look after family. Persons with disabilities also felt less independent, incurred greater expense, and found it harder to get food. Recent immigrants reported that giving up transit meant that they did not have culturally specific grocers and ethnically competent healthcare providers within walking distance. People with poorer health similarly reported that they did not have the right type of amenities in close proximity.

[176] In summary, Dr. Farber found that public transit remained especially essential to those who did not own a vehicle, those who are essential workers, those with lower incomes, and those who were non-white.

[177] The application judge dismissed all of this evidence on the basis that equity was not mentioned in the preamble to the TTC Act and that he viewed the appellant’s purported concern about equity as a “post-facto objective that did not cause the law to be enacted”. He also found that equity concerns “go both ways” in that there would be TTC employees who were in equity seeking groups.

[178] I fail to understand the basis for the application judge’s conclusion. First, equity issues were not an after-the-fact attempt to justify the legislation. Those concerns were expressly stated by the Minister of Labour, both when he introduced the TTC Act in the Legislature, and during the subsequent debates on the Bill. The application judge’s reliance on the discussion in Big M Drug Mart Ltd., at paras. 89-91, on the subject of shifting purpose, has no application here.

[179] Further, his reference to the issue going both ways is curious. It is also untethered to any principles surrounding the application of s. 1 of the Charter. The contrast that he attempts to draw, between the impact on members of the public generally and a subset of TTC employees, hardly represents comparable groups respecting the issue of whether the TTC is an essential service.

[180] I would have thought that the evidence of the extent to which some people were required to continue to use the TTC, during the lockdowns and other restrictions brought on by the COVID-19 pandemic, would resolve any debate over the importance of the TTC. It is difficult to believe that any person, who had other options, would have continued to use public transit during the onset of the pandemic, especially since, early on, the methods by which the virus could be transmitted were unclear. The fact that the application judge refused to consider this evidence in his determination whether the TTC is an essential service renders his conclusion on that issue fundamentally flawed.

[181] I would also not be as dismissive of the impacts on the public generally, arising from a disruption of TTC services, as was the application judge. While I accept that better evidence might have been led about the economic impacts of such disruptions, and the potential health effects of them, it should be self-evident that interfering with people’s ability to get to their employment, and to engage in consumer activities, will undoubtedly have economic impacts. Similarly, it does not require evidence to appreciate that, when the TTC is disrupted, more people drive. More cars on the roads means more pollution. Increased pollution is injurious to everyone’s health. That is a self-evident truth.

[182] Neither the health nor the economic issues may have had the evidentiary strength, on their own, to conclude that the appellant’s goal to avoid TTC disruptions was a pressing and substantial concern. However, these issues should not be viewed in isolation. They are part of the overall context in which the purpose of the TTC Act must be considered.

[183] I conclude that the TTC meets the appropriate standard to be considered an essential service as that term is properly understood in this context. The disruption of TTC services would clearly result in serious harm to a portion of the population – a portion that is particularly vulnerable. People who are prevented from getting to work and earning income, people who cannot buy food, and people who cannot get to medical treatments, are not instances of mere inconvenience. To portray them in that fashion does them a tremendous disservice. It is also unfair. The TTC Act therefore satisfies the first part of the two-part test from Oakes.

(2) Reasonably and demonstrably justified

(1) Rational connection

[184] Turning to the second part of the Oakes test, I begin with the first factor, that the means “must be rationally connected to the objective”. To establish a rational connection, the appellant must show “a causal connection between the infringement and the benefit sought ‘on the basis of reason or logic’": Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331, at para. 99, citing RJR-MacDonald Inc. v. Canada (Attorney General), 1995 64 (SCC), [1995] 3 S.C.R. 199, at para. 153. Further, the appellant need only show that it is “reasonable to suppose that the limit may further the goal, not that it will do so”: Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, at para. 48.

[185] It would seem incontrovertible that, if the objective of the appellant is to prevent disruptions of TTC services and thus avoid the harms caused by them, a prohibition on strikes is logically connected to that objective. The rational connection factor is thus met.

[186] The application judge instead incorporated a “care of design” component into his analysis of this factor. Further, he appears to have treated this component as a test for overbreadth. Neither consideration finds any support in the existing law as enunciated by the Supreme Court of Canada as it relates to the rational connection factor. Both only appear in that court’s consideration of the minimum impairment factor.

[187] In addition, I fail to see a sufficient evidentiary foundation for the application judge’s conclusion that there was a lack of care taken by the appellant when the TTC Act was enacted. His comment that the legislation was “rushed” is not made out by the facts. There were six weeks between the introduction of the legislation and its passing. The application judge also failed to consider, in this regard, that there was a request from Toronto City Council for the TTC to be deemed an essential service, that had been preceded by a request from the TTC Commissioners for such a designation. There was, therefore, much discussion and debate prior to the introduction of the TTC Act in the Legislature. None of this was taken into account by the application judge before he reached his conclusion that the legislation was “rushed”, nor was it taken into consideration before he made his finding that there was a lack of care taken by the appellant. Neither of those conclusions is warranted on the record.

(2) Minimal impairment

[188] I turn to the second factor under the second part of the Oakes test, that the means must minimally impair the right in question. It would seem that it is this factor that takes prominence in the reasons of my colleagues. It is important to remember the test that is applied to determine whether the legislation represents a minimum impairment. The test was stated by McLachlin J. in RJR-MacDonald, at para. 160:

As the second step in the proportionality analysis, the government must show that the measures at issue impair the right [at issue] as little as reasonably possible in order to achieve the legislative objective. The impairment must be "minimal", that is, the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement. [Citations omitted.]

[189] In my view, my colleagues have fallen into the error that McLachlin J. identifies at the end of the above quotation, that is, they have conceived of what they view as a better alternative than a complete prohibition on strikes. That is not the role of the court in assessing minimum impairment: Ontario English Catholic Teachers Association v. Ontario (Attorney General), 2024 ONCA 101, at para. 200.

[190] On this point, my colleagues have essentially adopted the approach taken by the application judge of contrasting the prohibition on strikes against the approach taken in other provinces with respect to their transit systems, as well as against the approach taken by the appellant in respect of other transit systems in Ontario. My colleagues justify their agreement with the application judge’s approach on the basis that “his factual conclusions are entitled to substantial appellate deference.”

[191] With respect, the usual deferential approach to factual findings is not applicable when dealing with a constitutional question. This point was recently reiterated by the Supreme Court of Canada in Société des casinos du Québec inc. v. Association des cadres de la Société des casinos du Québec, 2024 SCC 13. In her concurring reasons in that case, Côté J. enunciated the appropriate standard of review when dealing with a constitutional question. She said that when a constitutional question is involved, the standard of correctness applies to questions of law and questions of mixed fact and law. She noted, at para. 94: “‘Mixed’ findings are those that determine ‘whether the facts satisfy the applicable legal tests’” (citation omitted).

[192] She both reiterated and clarified this point at para. 97, where she said: “A reviewing court must show deference to findings of pure fact that can be isolated from the constitutional analysis” (emphasis added).

[193] The majority adopted Côté J.’s analysis of the standard of review at para. 45.

[194] In my view, the factual findings that the application judge made, and to which my colleagues defer, were not pure facts isolated from the constitutional analysis. Quite the contrary. The application judge’s findings were integral to his minimal impairment analysis. Those factual findings are therefore not entitled to deference from this court. Further, and as I will explain, some of the application judge’s factual findings reflect palpable and overriding error.

[195] In any event, the analytical approach that compares the TTC Act to other situations, is fundamentally flawed for four main reasons.

[196] First, it fails to take into account the unique nature of the TTC. The application judge accepted that the TTC is the largest and most complex transit system in Canada. It is the third largest in North America. The extent and importance of its operations cannot reasonably be compared to transit systems elsewhere in Ontario or elsewhere in Canada.

[197] My colleagues say that the application judge’s view, that the Toronto transit system was not so different from other transit systems in Canada to justify the TTC Act, is a factual finding “entitled to substantial appellate deference”. I do not agree. With respect, this was not a factual finding. It was an expression of the application judge’s opinion, an opinion for which he lacked the requisite expertise, and which was unsupported by any expert evidence. Indeed, the evidence that was before the application judge, as I noted earlier, was that the Toronto transit system is unique in Canada in terms of its size and complexity. Further, this opinion was directly connected to the application judge’s conclusion on the constitutional issue and thus draws a standard of review of correctness.

[198] Second, the appellant is entitled to take different approaches to an issue in different parts of the province. The appellant’s view that it is important to prevent any disruption of TTC services, given their importance, may not raise the same concerns when the appellant is considering the impact of transit disruptions in other municipalities. The appellant’s decision to take different approaches to these situations is a policy decision which is entitled to deference from the courts. Among other reasons, such deference accords the appellant the measure of “leeway” to which McLachlin J. said governments are entitled.

[199] Third, in terms of what occurs in other provinces, once again different provinces are entitled to adopt different approaches to the same problem. The fact that they do so does not render one approach right and another approach wrong. It simply reflects the fact that, for a variety of reasons, provinces may differ in terms of the approach that they take to the same problem. As LeBel J. observed in R. v. Advance Cutting & Coring Ltd., 2001 SCC 70, [2001] 3 S.C.R. 209, at para. 275: “In general, differences between legislative approaches to similar problems are part of the very fabric of the Canadian constitutional experience. Provincial differences must be factored into any proper analysis of the concept of minimal impairment, when assessing the validity of provincial legislation.”

[200] Fourth, in considering the situation elsewhere in Ontario and elsewhere in Canada, it is of some moment that we do not have any evidence regarding the labour relations histories in those other places, nor do we have any evidence as to the effectiveness of the varying approaches taken. On that point, it may be that a hybrid model works sufficiently well in Montreal. We simply do not know. Again, we also do not know the history of strikes in the Montreal transit system. We do know the history of strikes in the TTC and it is not a happy one. A brief summary of the history is sufficient to make the point:

Year	Result
1970	12 day strike
1974	23 day strike ended by legislation
1978	8 day strike ended by legislation
1984	Possible strike pre-empted by legislation
1989	45 day work to rule ended by legislation
1991	8 days
1999	2 day strike ended by threat of legislation
2006	1 day
2008	1 day strike ended by legislation

[201] In his minimal impairment analysis, the application judge did not consider this history. He also did not consider the lack of evidence as to the labour history and effectiveness of other approaches used elsewhere. He further did not give appropriate emphasis to the unique nature of the TTC in terms of its size, its complexity, and the importance of its functioning to the City of Toronto and the greater Toronto area.

[202] It is also of importance to note that, in terms of alternative approaches, the only evidence before the application judge was that the TTC did not favour a hybrid approach. In her affidavit filed on the application, the Executive Director of Human Resources said that a hybrid model was viewed by the TTC as “inappropriate”. Among other reasons, the TTC expressed concerns about the safety implications of such a model, including overcrowding during times when the system would operate.

[203] The application judge did not consider any of this in his minimal impairment analysis. His analysis, on this factor, is thus fundamentally flawed.

[204] Further, the application judge did not consider the provisions in the TTC Act that provide for compulsory arbitration. This marks a significant distinguishing factor between this case and the case that was before the Supreme Court of Canada in SFL. The application judge failed to recognize that important distinction before concluding that “the TTC Act is similar to the over broad legislation deemed unconstitutional in SFL.”

[205] I set out, at the beginning of my reasons on this factor, the application of the minimal impairment test as set out by McLachlin J. in RJR-MacDonald. As noted in that case and repeated in Libman v. Quebec (Attorney General), 1997 326 (SCC), [1997] 3 S.C.R. 569, the Legislature is entitled considerable leeway in deciding between various options. As the court put it in Libman, at para. 59:

This Court has already pointed out on a number of occasions that in the social, economic and political spheres, where the legislature must reconcile competing interests in choosing one policy among several that might be acceptable, the courts must accord great deference to the legislature's choice because it is in the best position to make such a choice.

[206] In my view, the appellant has met its burden to explain why there were not reasonable alternatives that could achieve the desired objective with less impairment of the right to strike. I have already alluded to those reasons above.

[207] One is the unique nature of the TTC, the number of people who rely on its services, the impact that a strike has on those people, as well as on the greater Toronto area as a whole. Another is the salient fact that the TTC did not favour a hybrid model because of safety concerns and the practical implications of attempting to run the TTC in a piecemeal fashion. Yet another is that the evidence shows that the operation of the TTC does not easily permit distinguishing between different categories of employees in terms of their importance to the overall operation of the system. For example, there is little point in having drivers available to run the subway cars if there are no ticket agents to allow patrons to enter the system, or electricians to keep the signaling system running, or maintenance people to ensure the tracks are clear, and so on.

[208] Similarly, given the nature of the TTC, it does not address the impacts of an interruption in services to say that the subway can run but the buses and street cars cannot. As the evidence demonstrates, the TTC (like most transit systems) is an integrated operation that requires each component to be up and running for the overall system to operate effectively.

[209] The application judge suggested two alternatives as being less impairing than a complete ban on strikes. One was the hybrid model, which I have already addressed. The other was a requirement for 48 hours’ notice of any strike action. With respect, notice does not accomplish the prime objective, which was to avoid any disruption in TTC services. Knowing that a strike is coming does not accomplish that objective unless, of course, the application judge intended the notice to give enough time for the Legislature to prohibit the strike. If that was the application judge’s reasoning, then he is simply allowing the Legislature to do through the back door what he says that they are not entitled to do through the front door.

[210] Recognizing that a ban on strikes has a significant impact on the collective bargaining process, the Legislature provided for “a meaningful dispute resolution mechanism”, that is, compulsory arbitration before a neutral arbitrator. Given the objective of preventing system disruptions, that dispute resolution mechanism provided TTC employees with a reasonable alternative to strike action to achieve their goals while at the same time avoiding service disruptions. It also reflects the practical reality of the labour relations history of the TTC.

[211] In the final analysis, the question under this factor was aptly put by McLachlin C.J. in Hutterian Brethren of Wilson Colony. She said, at para. 53:

Another way of putting this question is to ask whether there are less harmful means of achieving the legislative goal. In making this assessment, the courts accord the legislature a measure of deference, particularly on complex social issues where the legislature may be better positioned than the courts to choose among a range of alternatives.

[212] The legislative goal was to avoid disruptions in TTC services. None of the alternative means suggested by the application judge, or by the respondents, achieve that goal. The appellant was not required to compromise its goal just to achieve less impairment.

(3) Proportionality

[213] Turning to the third factor under the second part of the Oakes test, the means must be proportional to the objective. The question to be answered was set out by McLachin C.J. in Canada (Attorney General) v. JTI-Macdonald Corp., 2007 SCC 30, [2007] 2 S.C.R. 610, at para. 45:

The final question is whether there is proportionality between the effects of the measure that limits the right and the law's objective. This inquiry focuses on the practical impact of the law. What benefits will the measure yield in terms of the collective good sought to be achieved? How important is the limitation on the right? When one is weighed against the other, is the limitation justified? [Emphasis in original.]

[214] The legislation’s objective was to avoid disruptions in TTC services because of the various impacts those disruptions have, as I have set out above. Those impacts are serious. Also serious is the impact on the respondents’ right to strike. In considering the latter, however, it is important to remember that the legislation provides for compulsory arbitration. It is of some importance on this issue to recognize the reality of the labour relations history of the TTC. Arbitration has been the alternative to negotiated agreements between the TTC and the respondents for most of the past 30 plus years, due to the fact that the vast majority of the strikes have been ended by back-to-work legislation that included compulsory arbitration. In a very real sense, the TTC Act simply codifies the reality that has existed between the TTC and the respondents for decades.

[215] In his analysis on this factor, the application judge seems to have focussed almost entirely on the economic effects of a disruption in TTC services. Perhaps this is due to his emphasis on Retail, Wholesale and Department Store Union v. Saskatchewan, 1987 90 (SCC), [1987] 1 S.C.R. 460, where Dickson C.J. said, at p. 477: “In the meantime, in my view, legislatures are justified in abrogating the right to strike and substituting a fair arbitration scheme, in circumstances when a strike or lock-out would be especially injurious to the economic interests of third parties.”

[216] It is not, however, just the economic interests of third parties that fall to be considered under this factor. That was the key factor in RWDSUbut, in this case, there are many other interests involved, some more important in my view, especially the impacts on the vulnerable.

[217] On this point, the application judge once again compared equity seeking groups in the public generally with equity seeking groups within the TTC. Putting aside that there was virtually no evidence regarding the latter, I have already set out above why this is not an appropriate comparison.

[218] The application judge also held, on this factor, that the balancing had to take into account that “transit shutdowns are generally short lived.” That observation completely fails to take into account that the reason why TTC strikes are short lived derives from the fact that almost every time those strikes occur, they are ended by back-to-work legislation in favour of compulsory arbitration.

[219] Due to these errors, the application judge’s balancing under this factor is not entitled to deference.

[220] The result of the legislation leaves the respondents with “a meaningful alternative mechanism for resolving bargaining impasses”. Strikes coupled with back-to-work legislation manifestly do not accomplish the objective of avoiding disruptions in TTC services. The TTC Act achieves the “collective good” that comes from avoiding those disruptions. The benefits that arise from ensuring the continuation of TTC services outweigh the restriction on the overall bargaining process. Thus, the limit is proportional to the objective.

[221] Once again, my colleagues find that, notwithstanding the errors made by the application judge, his ultimate conclusion is salvaged based on deference. I repeat that deference is not applied in the same manner where a constitutional question is engaged. In any event, the errors made by the application judge clearly drove his analysis and those errors can be fairly described as palpable and overriding for the reasons I have set out above.

[222] Finally, I address my colleagues’ observations about the difference between the analysis to be undertaken here and the analysis that might be undertaken once a strike has occurred. My colleagues say that, once a strike has started, Ontario “may be able to present better evidence about the actual impacts on health and safety, the environment, and the economy.” With respect, neither the Legislature nor the public should have to wait for harm to occur before acting. There is no constitutional principle of which I am aware that requires a government to only react to a situation of harm rather than to be proactive in avoiding the harm.

C. Conclusion

[223] I would allow the appeal, set aside the decision below, and dismiss the application.

Released: May 23, 2024 “I.N.”

“I.V.B. Nordheimer J.A.”

[^1]: The individual respondents in this appeal are officers or former officers of the two respondent unions. [^2]: One of the subway lines, Line 3 Scarborough, has since been shut down. [^3]: The name derives from US legislation, the National Labor Relations Act of 1935, which is known as the Wagner Act after its author, US Senator Robert F. Wagner. [^4]: Bob Kinnear is the former president of ATU Local 113, and Kevin Morton is the former secretary-treasurer. Geatano Franco is the president of CUPE Local 2. [^5]: Professor Robert Hebdon is an emeritus professor at McGill University who was called by the unions to give expert opinion evidence in the area of industrial relations. [^6]: In the Alberta Reference, Dickson C.J.C. dissented in the result, with Wilson J. concurring. They would have found that the legislation at issue violated s. 2(d) and was not justified under s. 1. In PSAC v. Canada, Dickson C.J.C. dissented in part, finding that the legislation at issue in that case violated s. 2(d) and that some, but not all, of its provisions were justified under s. 1, while Wilson J. would have struck it down in its entirety. In RWDSU, Dickson C.J.C. wrote a separate concurrence in which he would have found a s. 2(d) violation but upheld the legislation under s. 1. Wilson J. dissented, and would have struck the legislation down as not justified under s. 1.

A. Facts and evidence

(1) Events prior to the enactment of the TTC Act

(2) The TTC Act

(3) Collective bargaining at the TTC since 2011

(4) Section 1 evidence

B. Standard of review

C. Does the TTC Act violate s. 2(d) of the Charter?

(1) The evolution of the s. 2(d) Charter jurisprudence on the right to strike.

(2) The TTC Act violates s. 2(d) by eliminating the “right to strike” at the end of a collective agreement

D. Is the breach of s. 2(d) justified under S. 1 of the Charter?

(1) Pressing and substantial object

(2) Proportionality

(1) Rational connection

(2) Minimal impairment

(1) The legal framework

(2) The application judge’s reasons

(3) Minimal Impairment Analysis

(3) Proportionality between salutary and deleterious effects

(1) General principles

(2) The application judge’s reasons

(3) Proportionality Analysis

(3) Conclusions on s. 1

E. Disposition

A. Charter s. 2(d)

B. Charter s. 1

(1) Sufficient importance

(2) Reasonably and demonstrably justified

(1) Rational connection

(2) Minimal impairment

(3) Proportionality

C. Conclusion

Judges

Counsel

Parties

Areas of Law

Fact Patterns

Expert Witnesses

Issues

Statutory Provisions 3

Legal Tests 2

Legislation 4