(Un)Constitutionality of the York University Labour Disputes Resolution Act

Last Monday, Daniel Flynn, the Ontario Minister Labour tabled An Act to resolve labour disputes between York University and Canadian Union of Public Employees, Local 3903 (the York University Labour Disputes Resolution Act). The legislation (which died on the order paper due to the election) would have put an end to the CUPE 3903 strike and, ultimately, forced the parties to resolve their dispute by binding arbitration. As is always the case with Canadian back-to-work legislation, the Act contains punitive provisions that set out stiff penalties for the union and for individual workers in the event that they refuse to return to work. In light of the Supreme Court’s decision in Saskatchewan Federation of Labour and the emerging jurisprudence on the contours of the constitutional protection of the right to strike, there are credible – though not dispositive – reasons to doubt the constitutional validity of the Act. This raises the tricky question of how unions ought to respond to back-to-work legislation that they reasonably believe to be unconstitutional.

Background

Workers at York University, and in particular those represented by CUPE 3903, have a long history of using strike action to advance their collective interests. This history of labour militancy is attributable to a variety of distinct but mutually reinforcing factors. One of these factors is that strikes at York often work; that is, they translate into tangible gains for workers. Obviously, past victories make subsequent strike action more appealing. A related factor is the particular organizing model used by CUPE 3903, described by Clarice Kuhling as “bargaining from below” and encapsulated in the union’s slogan Strike to Win!

Modelled on David McNally’s conception of “socialism from below”, CUPE 3903’s strategy is to build on the self-activity and consciousness of workers themselves to develop the solidarity and combativeness necessary for successful job action. This is put into practice by structures marked by a level of transparency and democracy rarely seen in Canadian unions, which are generally encouraged by the legal landscape to function as bureaucracies where membership input is largely limited to formal votes required at specific moments during the collective bargaining process and to the annual meeting at which policies are adopted and officers are elected.

CUPE 3903’s structures facilitate and encourage massive membership participation and direct accountability of the bargaining team to rank and file workers. Furthermore, bargaining is coordinated for three different units (graduate student teachers, graduate student researchers and contract faculty). Though the Ontario Labour Relations Act requires the bargaining team of each unit to “bargain in good faith and make every reasonable effort to make a collective agreement” applicable only to its members, there is significant wiggle room to pace bargaining in such a way as to ensure that members strike together, thereby increasing their collective strength.

The Strike

CUPE 3903 and York University have been bargaining since October of 2017. The union tabled 110 proposals, largely centred around three issues described as resistance to “the neoliberal university”: protecting graduate students from an unpredictable and potentially arbitrary “fellowship” model of funding, increasing job security for contract faculty, and strengthening gender equity measures.

The strike began on March 5, 2018, three days after the membership voted to reject the University’s “final offer”. Five weeks into the strike a new offer was put to the members, who massively rejected it. This led the Minister of Labour to appoint an Industrial Inquiry Commission, whose sole member was William Kaplan, an experienced labour mediator. Kaplan found that it was unlikely that the parties would resolve their dispute and concluded that “[f]ailing consensual interest arbitration, and assuming the continuation of this dispute, legislative intervention imposing interest arbitration will almost certainly be necessary.”

Kaplan directly addressed CUPE 3903’s unique “bargaining parameters and culture” and made the rather surprising suggestion that there may be too much union transparency and democracy, which “might not enhance collective bargaining”. He even went so far as to suggest that “the union might usefully reconsider its general approach”. These remarks show a tension between collective bargaining conceived of as a fundamental right and the current Canadian labour relations regime. In Health Services, Chief Justice McLaghlin described the scope of Charter protection of collective bargaining in terms of democratic association: “the constitutional right to collective bargaining concerns the protection of the ability of workers to engage in associational activities, and their capacity to act in common to reach shared goals related to workplace issues and terms of employment.”  Kaplan’s view – which I would venture is shared by most professionals in the field as well as by most legislatures – is that the ultimate end of our labour relations regimes is not to protect the associational activity of workers, but to secure collective agreements and – ultimately – “industrial  peace”.

Constitutionality of Back-to-Work Legislation

The Supreme Court of Canada held in Saskatchewan Federation of Labour that the right to strike is a constitutionally protected exercise of the freedom of association. Though the majority judgment clearly links striking to collective bargaining, the justification for extending constitutional protection to the right to strike puts workers’ self-organization – rather than the need to come to an agreement – front and centre:

[54]  The right to strike is essential to realizing [the] values [of human dignity, equality, liberty, respect for the autonomy of the person and the enhancement of democracy] and [the] objectives [of empowering and protecting marginalized groups] because it permits workers to withdraw their labour in concert when collective bargaining reaches an impasse. Through a strike, workers come together to participate directly in the process of determining their wages, working conditions and the rules that will govern their working lives. The ability to strike thereby allows workers, through collective action, to refuse to work under imposed terms and conditions. This collective action at the moment of impasse is an affirmation of the dignity and autonomy of employees in their working lives.

The Court goes on to affirm (at para. 57) that though striking is important to meaningful collective bargaining, “[s]trike activity itself does not guarantee that a labour dispute will be resolved in any particular manner, or that it will be resolved at all.”

The application of these principles to legislation curtailing or abrogating the right to strike depends on the well-known proportionality test set out in R. v. Oakes. Once a prima facie violation of workers’ right to strike is established, the State must demonstrate that the means employed are rationally connected to a pressing and substantial objective and that they impair “as little as possible” the workers’ freedom of association.

Evidently, ensuring the uninterrupted provision of essential services over which the State has a monopoly constitutes a pressing and substantial objective. But the Court was quite clear in pointing out that this only applies to cases where the withdrawal of services would endanger the life, personal safety or health of the whole or part of the population. Indeed, the majority judgment specifically states that “[i]n some circumstances, the public may well be deprived of a service as a result of strike action without being deprived of any essential service at all that would justifiably limit the ability to strike during negotiations.”

Constitutionality of the York University Labour Disputes Resolution Act

Clearly, the Act curtails the freedom of association of the members of CUPE 3903. Through a democratic process, they voted twice to strike and, by all accounts, participated in large numbers in collective action. The Act overrides these democratic decisions and, if adopted, would put an immediate end to the strike (s. 3(3)) and require CUPE 3903 members to “without delay, resume the performance of the duties of [their] employment” (s. 3(4)). The Act prohibits both the union and its members from threatening, authorizing, encouraging or supporting future strike action (s. 4) and provides for fines of up to $2,000 (in the case of an individual) and $25,000 (in the case of the union) for “each day of a contravention or failure to comply” (s. 7).

The question is thus not whether the Act violates workers’ freedom of association, but whether, all things considered, the violation is justified.

It is fairly obvious that universities are not an essential service in the usual sense of the term. However inconvenient the interruption of teaching and research may be, it does not pose a threat to the life, personal safety or health of the whole or part of the population. Accordingly, I think it would be clearly unconstitutional to simply abrogate ab initio the right to strike of university workers. What is less clear is whether ending a prolonged strike in a public service that is not essential could constitute a pressing and substantial objective justifying legislative intervention.

In Canadian Union of Postal Workers the Ontario Superior Court of Justice found that though the postal service is not an essential service in the strict sense, it provides a vital service to vulnerable and rural Canadians and that – given the “margin of appreciation” that must be afforded to legislatures in matters of political, social and economic policy – ensuring its continuation constitutes a pressing and substantial objective (paras. 197-201).

On the other hand, the Supreme Court has repeatedly opined that significant costs to the public must be tolerated if the right to strike is to be meaningful. For instance, in its unanimous opinion in Pepsi-Cola, the Court stated:

[25] Labour disputes may touch important sectors of the economy, affecting towns, regions, and sometimes the entire country.  The cost to the parties and the public may be significant. Nevertheless, our society has come to see it as justified by the higher goal of achieving resolution of employer-employee disputes and the maintenance of economic and social peace. The legally limited use of economic pressure and the infliction of economic harm in a labour dispute has come to be accepted as a legitimate price to pay to encourage the parties to resolve their differences in a way that both can live with.

Given these two arguments, it would be hazardous to predict how the courts would see the effects of a prolonged strike at a university. The drafters of the Act included a justification in the preamble that appears tailored to demonstrate that it responds to a pressing and substantial objective:

 Post-secondary education serves a critical public function. A lengthy extension or loss of an academic year has significant personal, educational, social and financial implications for students and their families as well as serious organizational and economic impacts on employers, the University and the broader public. These negative consequences may be long-term in nature and the repercussions could extend beyond the parties, the students and their families. The continuation of these disputes and the resulting disruption in education and its corresponding effects give rise to serious public interest concerns. The interests of students, families and the broader community require that these disputes be resolved.

Taken to its logical conclusion, this would mean that every time a strike by university workers threatens a lengthy extension or loss of an academic year, the legislature would be entitled to intervene. But that would amount to requiring university workers to “fight with one hand tied behind their backs”, since the dynamic of public sector labour relations requires significant disruption to public services for strikes to be successful. It could thus be argued that – if the preamble of the Act were taken at face value – university workers only have the right to strike when it is ineffective.

If we accept the hypothesis that putting an end to a prolonged strike in the university is a pressing and substantial objective, it seems obvious that it is rationally connected to the means typically found in back-to-work legislation.

The final step of the constitutional analysis requires the State to demonstrate that the right to strike is impaired as little as possible. In both Saskatchewan Federation of Labour and Canadian Union of Postal Workers it was found that, in order to pass constitutional muster, legislation that abrogates the right to strike must replace it with a meaningful alternative mechanism for resolving bargaining impasses, such as arbitration. At a minimum, the arbitrator’s jurisdiction must not be so restricted as to tilt the balance in favour of one of the parties.

On its face, the Act appears to meet this standard, since it provides for a system of mediation-arbitration (ss. 10-17). However, the arbitrator is required to take into account a series of factors, including “[t]he employer’s ability to pay in light of its fiscal situation”, “[t]he extent to which services may have to be reduced, in light of the award, if current funding and taxation levels are not increased” and “[a] comparison, as between the employees and comparable employees in the public and private sectors, of the nature of the work performed and of the terms and conditions of employment” (s. 15(2)).

The requirement that the arbitrator take these factors into account is problematic for two reasons.

First, in the public sector, the necessity of assuming “current funding and taxation levels” tilts in favour of the employer. Unfettered collective bargaining in the public sector may result in the reallocation of government resources or increased taxation, just as it may result in a decrease in capital’s relative share of output value in the private sector. Insisting that public sector workers’ compensation be determined on the basis of current funding and taxation levels is analogous to saying that private sector workers’ compensation must be determined on the basis of current levels of profit.

Second, the comparison requirement tilts in favour of the employer in cases where the workers have an industry-leading collective agreement. In many respects, CUPE 3903 is just such a case. Because of their history of successful strikes, York University contract faculty and teaching assistants have made gains that are, in many respects, above what workers at other universities have managed to extract. Indeed, CUPE 3903’s collective agreements have often set the standard for bargaining by unions in other institutions (this was the case, for example, with the tuition indexation clause that the union won in 2001). Where an arbitrator is required to engage in comparative analysis of similar workplaces, it is doubtful that the award will be more generous than the industry standard. CUPE 3903 is thus unlikely to make the kind of gains through arbitration that it might achieve through collective bargaining backed by strike action.

I admit that these are relatively novel arguments. In Canadian Union of Postal Workers, the arbitrator’s powers were much more restricted than in the instant case. If the government of Ontario could convince a judge that the objectives of the Act were pressing and substantial, then it would be an uphill battle for the union to argue that the means chosen do not pass the “minimal impairment” test.

What Is to Be Done?

The dissolution of the Ontario Legislative Assembly on May 9 has given CUPE 3903 members a reprieve. However, depending on the outcome of the election and of the bargaining process, they may be faced with back-to-work legislation as early as this summer. This raises the question of how to respond.

The potential unconstitutionality of back-to-work legislation raises tricky questions for unions. Even a successful court challenge may take years to work its way through the courts and it is unclear that a useful remedy can be provided.

This is less of a problem with general regimes that restrain or abrogate the right to strike. In such cases, a declaration of unconstitutionality on the basis of s. 52(1) of the Constitution Act, 1982 will likely benefit workers going forward. But back-to-work legislation is by definition time-sensitive and workers and unions might rightfully wonder what can be gained by being told by a court, several years after the fact, that their constitutional right to strike was violated. Though damages could be awarded under s. 24(1) (see Ward), it is doubtful that they could fully compensate for the effects of the violation, given the dynamic nature of collective bargaining.

One obvious strategy would be to disobey the law. Canadian public sector unions have a long history of refusing to comply with back-to-work legislation and have often made significant gains through illegal strikes. In these cases, the slogan adopted by their American counterparts is apposite: There is no illegal strike. Just an unsuccessful one.

In principle, a declaration of invalidity applies retroactively and the impugned legislation is void ab initio. Thus, workers or unions charged with violating the penal provisions could raise the unconstitutionality of the legislation as a defence. But this puts workers and unions in a difficult spot. They must either obey a law that they believe to be unconstitutional and thereby give it full effect (since subsequent declaratory relief is unlikely to be useful) or risk punishment for disobeying a law that turns out to be valid.