Corps de l’article

A little over a decade ago, in Health Services[1], the Supreme Court of Canada (SCC) held that the Charter of Rights and Freedoms’ guarantee of freedom of association “protects the capacity of members of labour unions to engage in collective bargaining on workplace issues.” This decision broke from the SCC’s previous view, which limited constitutional protection to the right to organize, but not the right to engage in core associational activities. However, the constitutional right to collective bargaining did not emerge fully formed from that judgment and the strength of the court’s commitment to constitutional labour rights was untested. Since that time, the SCC has issued five additional judgments evincing varying levels of support for and interpretations of constitutionalized labour rights. In its most recent case, British Columbia Teachers’ Federation[2], the Court had an opportunity to clarify the ambiguities arising from its previous judgments, but instead it chose to take a pass. In a rather sphinx-like, two sentence oral judgment from the bench, Chief Justice McLachlan declared: “The majority of the Court would allow the appeal, substantially for the reasons of Justice Donald. Justices Côté and Brown would dissent and dismiss the appeal, substantially for the reasons of the majority in the Court of Appeal.” Thus not only did the court fail to provide its own reasons, but it also limited what we could draw from the British Columbia Court of Appeal’s (BCCA) judgments [3] to which it referred, since we cannot know what parts of those judgments caused the majority and dissenting judgments to qualify their agreement as being “substantial” rather than complete.[4] The result leaves us to ponder which direction this court, with its rapidly changing membership, will go in future constitutional labour rights cases.[5]

The purpose of this brief intervention is not to make predictions, a fraught exercise in any event, but rather to explore the unresolved jurisprudential ambiguities, which, it is argued, are tied to deeper tensions embedded in our political economy and in industrial pluralism’s signature labour rights regime, Wagnerism. The Wagner model crystalized a set of labour laws and institutions in the post-World War II era that sought to contain class conflict through a set of compromises. It promised militant workers a fairer share of socially produced wealth by enabling the exercise of collective power to partially offset the unequal bargaining power written into the DNA of capitalist social relations. At the same time, it entrenched managerial control over investment and production decisions to facilitate the pursuit of profits, also an imperative of capitalism. Statutory labour laws and labour board and judicial jurisprudence shifted back and forth between these poles at different times and different places so that ambiguity and incremental change became permanent features of the regime.

This contradiction takes a special form in the context of public sector collective bargaining, where industrial pluralism itself is often under assault by neo-liberal governments’ growing intolerance of collective bargaining as they seek to impose austerity measures on their own workforces. Most constitutional labour rights cases arise in this context, often involving abrogations of collective agreements and restrictions on the freedom to strike. As a result, the court faces a sharp contradiction between the state’s evisceration of collective bargaining and worker bargaining power on the one hand, and Wagnerism’s commitment to partially offset unequal bargaining power on the other. In this context, it is not surprising that jurisprudential ambiguities abound.

We begin the analysis of these ambiguities with a brief discussion of the Health Services case, which planted its seeds. We then look at the twisted road subsequent cases have followed in Fraser[6], Mounted Police Association of Ontario[7]; Saskatchewan Federation of Labour[8]; and Meredith[9]; and then examine more closely the BCCA’s judgments in BCTF[10], which the SCC “substantially” adopted.

Health Services

The case arose out of the efforts of a neo-liberal British Columbia government to restructure the provision of health care through the privatization of health support services.[11] To do so, it needed to strip job security provisions out of an existing collective agreement and prohibit future bargaining on those issues. It enacted Bill 29 hastily and without any meaningful consultation with the affected unions.[12] The union challenged the legislation as being in violation of the constitutional guarantee of freedom of association. To the surprise of most observers, the SCC ruled Bill 29 unconstitutional. In doing so, the SCC reversed its earlier jurisprudence and held that freedom of association protected a right to collective bargaining. But the holding was not so simple.

First, the SCC created some level of ambiguity about what was being protected. On the one hand, it made clear that the constitutional right to collective bargaining could not be reduced to a mere right to make collective representations[13] and, indeed, made the duty to bargain in good faith and to make reasonable efforts to arrive at an acceptable contract a central feature of this constitutional right.[14] Tellingly, the court referenced its decision in Royal Oak Mines[15] as a key source for understanding the nature of the constitutional duty to bargain even though that case elaborated on the statutory duty to bargain in good faith as it exists within our Wagner Act model of labour law. On the other hand, earlier in its judgement, the court insisted that it was not constitutionalizing a particular model of labour relations or a specific bargaining model.[16] Because the case involved government abrogating collective agreement clauses and completely prohibiting future collective bargaining on those issues, without any bargaining or consultation, the court did not have to elaborate further on what other processes, if any, might pass constitutional muster.

The SCC layered a second level of ambiguity over the first when it also held that not every interference with the process of collective bargaining violated freedom of association. The Charter, it said, only protected against “‘substantial interference’ with associational activity.”[17] The court explained that the test for “substantial interference” was “whether the process of voluntary, good faith collective bargaining between employees and the employer has been, or is likely to be, significantly and adversely affected”, and that this was to be determined by looking at the importance of the matter affected and whether the government action respected the duty to consult and negotiate in good faith.[18] In part, this determination depended on the baseline understanding of the scope and content of the duty to bargain in good faith (the first ambiguity), but then required these two additional assessments, the second of which was the impact of the measure on the process of good faith bargaining and consultation. This raised the possibility that, although a government measure might abrogate a collective agreement, it could do so without substantially interfering with the process of collective bargaining by, for example, consulting with affected unions about its concerns prior to taking action. However, because the BC government acted peremptorily, without an opportunity for meaningful consultation in advance of enacting Bill 29, there was no need for further clarification.

The final layer of ambiguity arose in the section 1 analysis, where the government has the opportunity to demonstrate that the violation of constitutional rights is justifiable in a free and democratic society. In addition to showing that it was pursuing a pressing and substantial objective and that the means it chose were rationally connected to the achievement of its objective, the government must also show that the means chosen minimally impaired the right that was violated. Here again, the question arose as to the relevance of pre-legislative consultations with the union. On the one hand, the court said evidence of consultations was relevant in determining whether the government had considered less intrusive options[19], while on the other hand, it also implied that such consultations themselves reduced the level of impairment, in a way that was analogous to its substantial interference analysis. But in this case it had done neither. “The government also failed to engage in meaningful bargaining or consultation prior to the adoption of Bill 29 or to provide the unions with any other means of exerting meaningful influence over the outcome of the process (for example, a satisfactory system of labour conciliation or arbitration)”.[20] This formulation raised the possibility that consultations could save legislation that might otherwise be invalid either by reducing the level of interference below the constitutional threshold or by justifying the legislation as minimally impairing.

Underlying these doctrinal ambiguities was a lack of clarity about the rationale for constitutionalizing collective bargaining. The court provided three grounds for its judgment: Canadian labour history, Canada’s international law commitments and Charter values, including enhancing human dignity, liberty, autonomy, democracy and equality. In so doing, it recognized, as it had in previous employment law judgments, that employment is a structurally unequal power relation in which employees are subordinated to employers and that collective bargaining was a means to palliate that inequality.[21] However, it did not place the amelioration of power imbalances, or for that matter any other particular value, at the centre of its analysis. This pluralism was quite consistent with Wagnerism, and its unresolved internal tensions, including debates over the issue of whether good faith bargaining is merely procedural or has a substantive element responsive to the employer’s superior bargaining power. If the latter, then it would be necessary to examine the content of bargaining positions to determine whether one party was making substantively unreasonable demands. It was not surprising then that Health Services embodied a similar ambiguity about the scope of the constitutional duty of good faith bargaining.

Working the Ambiguities: From Fraser to Meredith

Health Services created ambiguities, but the case law that followed did little to resolve them. Indeed, in Fraser[22], the first case to reach the court after Health Services, the SCC amplified them. In particular, it used the SCC’s caution in Health Services that it was not constitutionalizing a particular process to step back from that judgment’s apparent embrace of good faith bargaining as understood in the context of existing statutory collective bargaining regimes. This paved the way for the SCC to allow collective bargaining regimes that required something less—“a good faith process of consideration by the employer of employee representations and of discussion with their representatives”[23]—to pass constitutional muster. Not only did Fraser seem to lower the constitutional baseline, but it also upped the ante with regard to the standard for determining whether the government’s interference with that thin process was unconstitutional. Instead of the “substantial interference” test from Health Services, the court in Fraser seemed to substitute a “substantial impossibility” test. “In every case, the question is whether the impugned law or state action has the effect of making it impossible to act collectively to achieve workplace goals”.[24] More generally, support for the project of constitutionalizing labour rights seemed to be slipping as two justices, Rothstein and Charron, neither of whom participated in Health Services, would have reversed that judgment, while a third, Deschamps, who had dissented in part in Health Services, would have achieved the same result through her interpretation of its effect.[25]

Given the tone and trajectory of Fraser, the three labour rights cases released by the SCC in January 2015 came as a surprise. Not only did they seemingly sideline (but not overrule) Fraser, they strengthened and extended the labour rights protected by the Charter guarantee of freedom of association without, however, resolving the legacy of ambiguity from Health Services.

Beginning with the first judgment, Mounted Police Association of Ontario[26], the court struck down the imposition of a non-union representation plan on RCMP members because the scheme deprived these employees of a sufficient degree of choice and independence to enable them to meaningfully pursue collective workplace goals. In so doing, the court added choice and independence as two requirements that governments needed to respect in constructing a constitutionally valid labour relations regime. In reaching that conclusion, the court found it necessary to address the principles that inform its interpretation of what a constitutional right to a meaningful collective bargaining process entails. Drawing heavily on the former Chief Justice Dickson’s dissent in Reference re Public Service Employee Relations Act (Alta.)[27], the court adopted a purposive approach that centred on the protection of individuals against more powerful entities and the promotion of equality. “Nowhere are these dual functions of s. 2(d) more pertinent than in labour relations. Individual employees typically lack the power to bargain and pursue workplace goals with their more powerful employers. Only by banding together in collective bargaining associations, thus strengthening their bargaining power with their employer, can they meaningfully pursue their workplace goals”.[28]

This approach had implications for resolving some of the Health Services ambiguities. First, the baseline process must not reduce employees’ negotiating power so that they are unable to meaningfully pursue their goals.[29] A process that merely provides for employee consultation, without providing a modicum of bargaining leverage would clearly fail to pass constitutional muster. The court’s embrace of the more robust interpretation of collective bargaining implicit in Health Services reflected a shift toward the equality-enhancing pole of industrial pluralism. This turn toward equality was also apparent in the court’s re-affirmation of the “substantial interference” test[30] and its focus on the impact of the state’s action on employee bargaining power. “A process that substantially interferes with a meaningful process of collective bargaining by reducing employees’ negotiating power is therefore inconsistent with the guarantee of freedom of association…”.[31] However, given the context of the case, the court did not have to address the question of whether government consultation prior to the enactment of legislation could render the interference less than substantial or make the action minimally impairing under a s. 1 analysis.

The emphasis in MPAO on a process that provides workers with a modicum of bargaining power to achieve their collective goals foretold the result of the Saskatchewan Federation of Labour case[32], which held, for the first time, that freedom of association protects the right to strike or an acceptable substitute. The court identified the right to strike as the “powerhouse” of collective bargaining, which was required to achieve “approximate equality”.[33] This finding further reduced two of the core ambiguities of Health Services. First, it clarified that a process that did not provide workers with the collective power to withdraw their services was constitutionally deficient. Mere consultation or consultation that only required good faith consideration of employee proposals could not pass constitutional muster. Second, the denial of the freedom to engage in a collective withdraw of services amounted to a substantial interference with collective bargaining since it deprived workers of the power needed to meaningfully pursue their workplace goals.[34]

The context of the SFL case did not require the court to consider whether consultation prior to the enactment of legislation depriving workers of freedom of association could save it. That said, given the court’s analysis, it would be hard to imagine consulting with unions prior to depriving them permanently of the right to strike would be curative since its effect would be to strip them of bargaining leverage on an ongoing basis. But what about ad hoc back-to-work legislation to end a lawful strike or lock-out, or legislation that peremptorily imposed an agreement prior to a strike or lock-out?

This brings us to Meredith[35], the middle case in the 2015 trilogy. The case squarely raised the question of whether government action rolling back three-years of agreed upon wage increases for RCMP members amounted to substantial interference with the bargaining process. The case was complicated by the fact that the court had just held in MPAO that the consultation process leading to these increases was itself constitutionally deficient. Nevertheless, the court was prepared to consider that substantial interference with that deficient process could violate freedom of association.[36] However, the majority of the court held that the federal Expenditure Restraint Act[37](ERA), which imposed wage restraint across the entire public service, did not substantially interfere with “the collective pursuit of the workplace goals of RCMP members”.[38] The SCC justified this result on the basis that the wage cap imposed by the statute was consistent with the increase negotiated with other bargaining units “and so reflected an outcome consistent with actual bargaining.”[39] The SCC also took into account that the law did not preclude discussion on other issues.[40]

We might view the case as an anomaly given its factual context, but the SCC’s reasoning ignored the equality promoting analysis that was so crucial to its holding in MPAO. First, the fact that the government was able to negotiate agreements with other bargaining units in the shadow of the ERA fails to acknowledge that the impending legislation substantially reduced the bargaining power of those bargaining units subject to it. If the MPAO analysis applied, surely such legislation must be “inconsistent with the guarantee of freedom of association”.[41] Second, even if the government had reduced the extent of its interference by consulting and bargaining with other bargaining units, RCMP members were completely deprived of any opportunity to consult even within the confines of their constitutionally deficient scheme. Can bargaining with Peter really satisfy your duty to bargain with Paul?

Overall, while MPAO and SFL went some distance toward resolving the ambiguities of Health Services, Meredith undermined their impact. It allowed government interference with an existing agreement without any consultation, and gave credence to agreements reached under the threat of wage restraint legislation that deprived workers of the freedom to strike—the powerhouse of collective bargaining—and thus substantially reduced their ability to bargain from “approximate equality.”

BC Teachers: The SCC Takes a Pass

The BCTF case provided the SCC with an ideal opportunity to revisit and clarify the ambiguities stemming from its Health Services judgment. After all, the case originated in the same neo-liberal assault on labour rights that launched the Health Services litigation.[42]Bill 28[43] stripped the teachers of negotiated contract protections in relation to class size, among other matters, and prohibited future collective bargaining over these matters, just as Bill 29 had stripped collective bargaining rights from health service workers. After years of delay, while awaiting the outcome of the Health Services case, in 2011 the BC Supreme Court found that Bill28 was unconstitutional. It substantially and unjustifiably interfered with the collective bargaining rights of teachers.[44] Justice Griffin issued a declaration of invalidity with the usual one-year suspension to allow the government time to devise a constitutionally valid response. A period of consultations/negotiations followed, without an agreement. The government then enacted Bill 22[45], which continued to override contractual limits on class size, but permitted future negotiations on the subject. The BCTF challenged the law, claiming that it too violated freedom of association by substantially and unjustifiably interfering with collective bargaining.

Justice Griffin also heard this case and upheld the BCTF’s challenge. Her judgment had two prongs. The first was a factual finding that the BC government bargained in bad faith by coming into the process with a closed mind, unwilling to consider contractual limits on class size. The second was a legal holding that pre-legislative consultations were not relevant to the determination of whether the legislation substantially interfered with constitutionally protected bargaining rights.[46] The BC government appealed and a majority of the BCCA (4 to 1) reversed Griffin’s judgment.[47] The court released its judgment three months after the SCC’s 2015 trilogy.[48]

The majority judgment, written by Chief Justice Bauman and Justice Harris, held that pre-legislative consultations are an important contextual factor in determining whether the government substantially interfered with collective bargaining rights. As well, they also rejected Griffiths’ finding that the government bargained in bad faith, largely on the basis that it was inappropriate to have inquired into the substantive reasonableness of a party’s negotiating position.[49] Justice Donald dissented. He agreed with the majority that pre-legislative consultations are relevant, although his reasons were different. However, he disagreed with their decision to overrule the trial judge’s finding of fact that the government bargained in bad faith, in part because of the deference that an appeal court owes the finder of fact, but more importantly because he disagreed with their interpretation of the constitutional test for good faith bargaining and its application in this case. However, as noted at the beginning of this comment, the majority of the SCC[50] reversed the BCCA, substantially for the reasons of Justice Donald, and so it is his judgment that bears scrutiny regarding the resolution of constitutional ambiguities.

The most fundamental constitutional ambiguity is with respect to the process freedom of association protects. As we saw, while Fraser embraced a thin process of consultation, MPAO and SFL had in mind a process of collective bargaining in which there was an approximate equalization of bargaining power, although Meredith seemed to ignore that consideration entirely. Donald J.’s judgment reflects rather than clarifies this ambiguity. On the one hand, he accepts the MPAO approach and applies it to pre-legislative consultations: “Pre-legislative consultation, then, can be seen as a replacement for the traditional collective bargaining process, only if it truly is a meaningful substitution. To be meaningful, the bargaining parties must consult from an assumed position of ‘approximate equality’”.[51] On the other hand, his application of the principle is deeply flawed. Under what conceivable arrangements can workers acting collectively “consult from a position of ‘approximate equality’” with the government, especially once we understand, as the SCC did in SFL, that the freedom to strike is the powerhouse of collective bargaining and thus the source of “approximate equality”? Political strikes have never been acceptable in Canada and are nearly always unlawful under current labour legislation, which prohibits workers from striking during the life of a collective agreement. Donald J. ignores this reality and, therefore, is able to affirm that the consultation process preceding the enactment of the ERA, endorsed by the SCC in Meredith, was an adequate substitute for collective bargaining. In the result, we are left to ponder the strength of the SCC’s commitment to the requirement of “approximate equality” and what balance must be reached for that standard to be met.

The other baseline ambiguity is with the “good faith” requirement. Recall that in Health Services the SCC seemingly embraced the statutory standard of good faith bargaining articulated in its Royal Oak Mines judgment as the constitutional standard, while simultaneously insisting that it was not constitutionalizing a particular model of bargaining. However, Fraser seemed to suggest it was sufficient to require that the employer listen to or read collective representations in good faith to pass constitutional muster. The issue was central to the BCTF litigation because the trial judge found that the government had entered the process with a closed mind. In part, this was a factual finding, but it also depended on a prior legal judgment about what good faith bargaining requires in a constitutional context. The majority of the BCCA cited labour board case law regarding the statutory duty to bargain in order to inform its interpretation of the constitutional duty, while also insisting it was not constitutionalizing a particular model. However, their interpretation of the statutory duty to bargain in good faith differed from that of the SCC. Based on BC labour board jurisprudence, they did not allow a review of the parties’ negotiating positions for substantive reasonableness.[52] Indeed, the BCCA ignored the SCC’s judgment in Royal Oak Mines.[53] By contrast, Donald J. cited Royal Oak and sought to reconcile the conflicting case law by re-grounding the good faith requirement in MPAO’s emphasis on “approximate equality.” According to Donald J.: “To summarize, good faith from a constitutional perspective, has been described by the Supreme Court of Canada as requiring the parties to meet and engage in meaningful dialogue where positions are explained and each party reads, listens to, and considers representations made by the other. Parties’ positions must not be inflexible and intransigent, and parties must honestly strive to find a middle ground”.[54] His judgment then turns to the issue of unequal bargaining power to justify an inquiry into the substantive reasonableness of the government’s bargaining positions in the constitutional context: “The government always has the power to unilaterally resolve impasse through legislation, or force workers to end a strike through constitutionally compliant back-to-work legislation. This is a huge power imbalance that fundamentally alters the calculus of how negotiations unfold”.[55]

Had the SCC agreed with Justice Donald’s judgment in its entirely, then we could be reassured that the limited duty of good faith expressed in Fraser has been rejected and that the constitutional standard of good faith bargaining requires the parties to make reasonable efforts to find an acceptable compromise. Even more importantly, we could be reassured that the amelioration of unequal bargaining power is at the heart of any interpretation of the requirement of freedom of association in the labour context. However, the SCC passed on the chance to provide that reassurance.

With regard to ambiguities around the level of government interference that will make its actions unconstitutional, there is no doubt that MPAO restored the “substantial interference” standard and that Fraser’s impossibility test has been set aside. The BCCA’s BCTF judgments do not delve into this part of the jurisprudence. Nevertheless, there is a very important issue lurking here. Assuming a constitutionally valid collectively bargaining regime must enable workers acting collectively to deal with their employers from a position of approximate equality, when is that achieved and when does government action (or inaction) substantially interfere with it?

Conclusion

It is not surprising that the SCC’s articulation of constitutionalized labour rights is a work-in-progress and replete with ambiguities. The construction of constitutional labour rights can neither escape the contradictions of the socio-economic structure in which they operate nor the statutory and jurisprudential labour rights regime that developed within it. It is extremely unlikely, for example, that the SCC will pursue its “approximate equality” analysis to the point of holding that existing collective bargaining regimes are unconstitutional because they now fail to meet that standard. However, neither is the constitutional labour rights regime bound to reflect and reinforce the compromises reached at a particular time and place, especially given their unstable normative, political, and economic foundations. Finally, for the same reasons, it is also likely that the constitutional labour rights regime will remain unstable and replete with unresolved ambiguities. So while it may be disappointing that the SCC took a school holiday in BCTF, it will, like school holidays, be short-lived. The court will inevitably be back on the job, grappling with the remaining ambiguities, and perhaps creating new ones, in respect of the constitutional limits on government imposed wage controls, collective bargaining restructuring or back-to-work legislation.[56]