Ontario (Attorney General) v. Fraser

This appeal raises anew the issue of the constitutionality of the labour relations regime that applies to farm workers in Ontario. Most Canadian provinces have brought the farming sector under their general labour relations laws, with some exceptions and restrictions. Except for a very short period of time, Ontario has always excluded farms and farm workers from the application of its Labour Relations Act (currently Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A) (“LRA”). In the present appeal, our Court must determine whether Ontario’s latest attempt to frame a separate labour relations regime for the farming sector respects the constitutional guarantee of freedom of association, or violates it by failing to safeguard the exercise of collective bargaining rights. The Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16 (“AEPA” or “Act”), was a response to this Court’s decision in Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, which found that the previous legislative scheme violated s. 2(d) of the Canadian Charter of Rights and Freedoms .

We are of the view that the AEPA has not been shown to be unconstitutional. Section 2(d) of the Charter protects the right to associate to achieve collective goals. Laws or state actions that substantially interfere with the ability to achieve workplace goals through collective actions have the effect of negating the right of free association and therefore constitute a limit on the s. 2(d) right of free association, which renders the law or action unconstitutional unless it is justified under s. 1 of the Charter . This requires a process of engagement that permits employee associations to make representations to employers, which employers must consider and discuss in good faith.

The law here at issue, the AEPA, properly interpreted, meets these requirements, and is not unconstitutional. We would therefore allow the appeal.

Before moving to the analysis of the issues raised by the appeal, it will be useful to review the factual background and the jurisprudential developments that gave rise to this case.

Prior to 1994, indeed since 1943, farm workers had been excluded from the general labour relations regime established by the LRA. In 1994, the Ontario legislature enacted the Agricultural Labour Relations Act, 1994, S.O. 1994, c. 6 (“ALRA”), which extended trade union and collective bargaining rights to agricultural workers. A year later, the legislature repealed the ALRA in its entirety and again excluded farm workers from the labour relations regime set out in the ALRA (Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1 (“LRESLAA”)). The LRESLAA was challenged on the basis that it infringed the guarantees of freedom of association under s. 2(d) and equality under s. 15 of the Charter . In Dunmore, a majority of this Court found a breach of s. 2(d). It held that the claimants were substantially unable to organize without protective legislation, and declared the law to be unconstitutional. This had the effect of nullifying the exclusion of farm workers from the LRA, but this Court suspended the declaration of invalidity for 18 months. The majority concluded it was not necessary to deal with the s. 15 challenge.

In response, the Ontario legislature enacted the AEPA in 2002, which came into force on June 17, 2003. In brief, the AEPA excluded farm workers once again from the LRA, but crafted a labour relations regime for farm workers in Ontario. It granted them the rights to form and join an employees’ association, to participate in its activities, to assemble, to make representations to their employers through their association on their terms and conditions of employment, and the right to be protected against interference, coercion and discrimination in the exercise of their rights (s. 1(2)). The employer must give an association the opportunity to make representations respecting terms and conditions of employment, and it must listen to those representations or read them (s. 5). The AEPA tasks a tribunal, the Agriculture, Food and Rural Affairs Appeal Tribunal, with hearing and deciding disputes about the application of the Act (ss. 2 and 11). After limited efforts to use the new protections of the AEPA, the respondents mounted a constitutional challenge to its validity.

The respondents argue that three more protections are required to meet the requirements of s. 2(d) of the Charter : (1) statutory protection for majoritarian exclusivity, meaning that each bargaining unit is represented by a single bargaining agent; (2) an LRA-type statutory mechanism to resolve bargaining impasses and interpret collective agreements; and (3) a statutory duty to bargain in good faith. The respondents argue that the Court’s recent decision in Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391, entitles them to laws offering these protections.

The four individual farm workers in this case (three of them respondents and an affiant) worked at Rol-Land Farms Ltd., a large industrial-type mushroom farm in Kingsville, Ontario. In 2002, after this Court’s decision in Dunmore, Xin Yuan Liu and other workers at Rol-Land approached the United Food and Commercial Workers Union Canada (“UFCW”) to represent them and bargain on their behalf. By the spring of 2003, 70 percent of the workers at Rol-Land had joined the union.

Rol-Land refused to recognize the UFCW as the employees’ representative. The UFCW then filed an application for certification with the Ontario Labour Relations Board, under the LRA. In the ensuing vote, the workers voted 132 to 45 in favour of certification. The resolution of the application has been kept on hold, pending the outcome of this appeal. After the vote, the UFCW wrote to Rol-Land requesting a meeting to commence negotiations toward a contract for the workers. Rol-Land did not respond to the letter. The respondents assert that the owner of Rol‑Land told a meeting of workers that the union would never be recognized (R.F., at para. 58).

The UFCW also attempted to bargain collectively on behalf of employees at Platinum Produce, an industrial greenhouse operating in Chatham, Ontario. While the employer expressed doubt that the UFCW could be an employees’ association under the AEPA, it gave the union an opportunity to make brief presentations. The meeting lasted approximately 15 minutes. The employer’s position was that the company was not required to bargain with the union and the meeting was not to be considered collective bargaining towards a collective agreement.

The UFCW subsequently presented Platinum Produce’s counsel with a draft collective agreement setting out proposed terms. That meeting lasted approximately five minutes. The employer has not responded to the proposals or to other proposed meeting dates. There have been no further meetings or communications about terms and conditions of work.

The respondents did not attempt to pursue remedies under the AEPA. Specifically, no recourse was made to the Tribunal set up under the Act to deal with complaints. Rather, the respondents sought a declaration that s. 3(b.1) of the LRA, which provides that the LRA does not apply to farm workers, and that the AEPA as a whole were unconstitutional. In brief, they submitted that these laws breached s. 2(d) of the Charter by failing to provide effective protection for the right to organize and bargain collectively and violated s. 15 by excluding farm workers from the protections accorded to workers in other sectors.

The chambers judge, Farley J., heard the application before the judgment of this Court in Health Services. He proceeded on the assumption that s. 2(d) did not protect collective bargaining. He dismissed the application on the ground that the AEPA met the minimum constitutional requirements necessary to protect the freedom to organize. He began by adopting the comments made by Sharpe J. at the trial level of Dunmore and finding that agricultural workers “are ‘poorly paid, face difficult working conditions, have low levels of skills and education, low status and limited employment mobility’” (paras. 23 and 33). But he was of the view that the AEPA did not prevent them from attempting to form employees’ associations. He stated:

There is nothing in the AEPA which would prevent the UFCW or any other union from attempting to organize agricultural workers into an employees’ association, recognizing that such an employees’ association would not thereby automatically have the right to strike nor the right to bargain collectively. See discussion in Peter W. Hogg, Constitutional Law of Canada, looseleaf, vol. 2 (Toronto: Carswell, 1997) at p. 41-5/6 (2002). The AEPA provides that the employees’ association may make representations to an employer concerning the terms and conditions of employment (s. 5 AEPA). These representations may be made by someone who is not a member of the association (s. 5(2)) so that a “union staffer” could perform that function. The representation may be made orally or in writing (s. 5(5)). One must read s. 5(6) and (7) in a purposive way in context. Thus while the employer need only give the association a written acknowledgment that the employer has read the written representations (s. 5(7)), it is implicit in the making of an oral representation that the recipient is hearing the oral representations as the employer has a duty to listen and the association speaker will have the opportunity then and there to enquire whether the recipient has heard the representations. As well the concept of listening and reading respectively involves the aspect of comprehending and considering the representations. Perhaps unfortunately there is no specific requirement that the employer respond to the substance of the representations; however, it should be noted that this would then involve the parties in a form of collective bargaining. [para. 19]

With respect to the particular statutory protections in the AEPA, Farley J. found that they met the minimum required standards. He found that they confer the power to organize (s. 1); protection against denial of access to property (s. 7); protection against employer interference with trade union activity (s. 8); protection against discrimination (s. 9); protection against intimidation and coercion (s. 10); protection against alteration of working conditions during the certification process (ss. 9-10); protection against coercion of witnesses (s. 10); and removal of Board notices (s. 10). He allowed that it would have been preferable to have mirrored the provisions of the LRA more precisely “to eliminate possible fears” that employers might alter working conditions to hinder associational activities (para. 18). However, he felt that the answer to these concerns lay with the Tribunal, which had not been asked to deal with the workers’ complaints. He stated:

If the Tribunal felt that it was for some jurisdictional reason constrained from negatively sanctioning such activity, then one would presume that the applicants or others of a like mind together with the UFCW would have a strong case to bring back in this regard. One would think it better to see how the Tribunal operates in fact before condemning it as powerless to deal with such abuses. This wait and see pragmatic approach is desirable with respect to possible concerns about lack of labour relations expertise/experience on the part of the specified panel roster of the Tribunal. There has been no use of the mechanics of the AEPA as to bringing a case before the Tribunal; the applicants stated that it would be fruitless to bring a useless application before a useless Tribunal. I am of the view that this condemnation is premature. A successful application would do one of several things: be effective positively as to action; or morally give the wrongdoing employer a “bloody nose”; or if truly an empty process, it would demonstrate the need for strengthening by legislative amendment. See also Danson v. Ontario (Attorney General), [1990] 2 S.C.R. 1086, . . . at p. 1099 . . . where Sopinka J. for the court stated: “This Court has been vigilant to ensure that a proper factual foundation exists before measuring legislation against the provisions of the Charter , particularly where the effects of impugned legislation are the subject of the attack.” [para. 18]

Farley J. also dismissed the discrimination claim brought under s. 15 of the Charter .

The Court of Appeal allowed the appeal and declared the AEPA to be constitutionally invalid. It rendered its decision after the judgment of our Court in Health Services. Winkler C.J.O., for the court, held that the AEPA substantially impaired the ability of agricultural workers to meaningfully exercise the right to bargain collectively, which was protected by s. 2 (d) of the Charter . The Act protected the right to organize, and it was premature to conclude that the statutory protections against employer intimidation were inadequate. However, it did not provide the “minimum” statutory protections required to enable agricultural workers to exercise their right to bargain collectively in a meaningful way, namely: “(1) a statutory duty to bargain in good faith; (2) statutory recognition of the principles of exclusivity and majoritarianism; and (3) a statutory mechanism for resolving bargaining impasses and disputes regarding the interpretation or administration of collective agreements” (para. 80). The court noted that the “primary difficulty has been that the union has been unsuccessful in engaging employers, who have no statutory duty to bargain in good faith” (para. 98). Having found that the AEPA infringes s. 2(d) of the Charter , the Court of Appeal considered whether the infringement could be saved as “reasonable and demonstrably justified” under s. 1 of the Charter , and found it could not. The legislation impaired the right more than necessary, despite the challenges facing legislators in the agricultural domain, which is a complex mix of family farms and larger industrial operations. It concluded that “the wholesale exclusion of agricultural employees from a collective bargaining scheme is not adequately tailored to meet the objective of protecting the family farm” (para. 129).

The Court of Appeal ordered the government “to provide agricultural workers with sufficient protections to enable them to exercise their right to bargain collectively, in accordance with these reasons” (para. 138). It suspended this order for 12 months to give the legislature an opportunity to respond. The Court of Appeal dismissed the claim under s. 15 of the Charter . Its judgment was appealed to this Court.

The issue is whether the failure of the Ontario government to enact a positive statutory framework for agricultural workers modelled after the Ontario LRA violates s. 2(d) of the Charter in a manner that cannot be justified by s. 1. If so, the AEPA is invalid under s. 52 of the Co nstitution Act, 1982 , and the Ontario legislature is obliged to bring the Act into harmony with the Charter . The respondents have also raised the issue of whether the AEPA violates their right to equality under s. 15 of the Charter . As the main question in this appeal remains the interpretation and application of s. 2(d), we will first consider this issue and then turn to s. 15.

In view of the conflicting approaches to the guarantee of freedom of association in the labour context put before us, it may be useful to canvas the jurisprudence and set out the principles that guide the analysis of s. 2(d). The respondents’ claim largely turns on the interpretation of our Court’s judgments in Dunmore and Health Services. The ultimate question is whether s. 2(d), properly understood and applied, requires the Ontario legislature to provide a particular form of collective bargaining rights to agricultural workers, in order to secure the effective exercise of their associational rights. To resolve this question, we will first consider the development of this Court’s jurisprudence in this area of the law.

(1) Freedom of Association in the Labour Context: The Jurisprudential Background

(a) The Early Cases

The first set of cases to consider s. 2(d) of the Charter , known as the Trilogy, were Reference re Public Service Employee Relations Act (Alta.), [1987] 1 S.C.R. 313 (the “Alberta Reference”); PSAC v. Canada, [1987] 1 S.C.R. 424; RWDSU v. Saskatchewan, [1987] 1 S.C.R. 460. The majority of the Court held that s. 2(d) did not protect the right to strike, the issue in the cases. In arriving at this conclusion, members expressed a number of views on the guarantee of freedom of association.

McIntyre J. stated that “like most other fundamental rights”, the right to freedom of association has no single purpose or value. Rather, reflecting the social nature of human beings, it protected the right to associate with others “both to satisfy [their] desire for social intercourse and to realize common purposes” (Alberta Reference, at p. 395). In the same case, Dickson C.J. (dissenting) stated: “What freedom of association seeks to protect is not associational activities qua particular activities, but the freedom of individuals to interact with, support, and be supported by, their fellow humans in the varied activities in which they choose to engage” (p. 366).

Three of the six Justices sitting on the Trilogy opined in obiter that s. 2(d) does not protect collective bargaining (Alberta Reference, at p. 390, per Le Dain J.; PSAC, at p. 453, per McIntyre J.). The reasons given included that collective bargaining is a “modern right”, that its recognition would go against the principle of judicial restraint, that s. 2(d) protects only individual rights, and that s. 2(d) was not intended to protect the goals or objects of organizations (Alberta Reference, at p. 391, per Le Dain J.; pp. 397 and 407, per McIntyre J.).

The Trilogy was endorsed in Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), [1990] 2 S.C.R. 367 (“PIPSC”). Sopinka J., stated in his reasons which appeared to be accepted by the other judges on this point: (1) “s. 2(d) protects the freedom to establish, belong to and maintain an association”; (2) “s. 2(d) does not protect an activity solely on the ground that the activity is a foundational or essential purpose of an association”; (3) “s. 2(d) protects the exercise in association of the constitutional rights and freedoms of individuals”; and (4) “s. 2(d) protects the exercise in association of the lawful rights of individuals” (p. 402).

Nine years later, in Delisle v. Canada (Deputy Attorney General), [1999] 2 S.C.R. 989, the Court once again revisited s. 2(d). The issue was whether exclusion of RCMP members from public bargaining associations, as opposed to their own association, violated s. 2(d). The majority of the Court, per Bastarache J., held it did not, on the ground that s. 2(d) does not give the right to belong to a particular group. Bastarache J. added that there is no general obligation for the government to provide a particular legislative framework for employees to exercise their collective rights, i.e. a different framework than already provided for RCMP members.

In dissent, Cory and Iacobucci JJ. pointed out that the appeal was not concerned with the right to strike or to bargain collectively (Delisle, at para. 51). The only issue was whether the statutory framework interfered with the right to associate with other public servants in pursuance of their mutual interests. They were of the view that s. 2(d) was violated because the impugned laws in purpose and effect interfered with the formation of employee associations. They noted that employees are a vulnerable group in our society, and their ability to form and join an employee association is crucially linked to their economic and social well-being (paras. 67-68).

In summary, the early cases affirmed that the core protection of s. 2(d) focusses on the right of individuals to act in association with others to pursue common objectives and goals. There was some suggestion (Sopinka J.’s fourth point in PIPSC) that only individual goals were protected. (This proposition, as we shall see, was rejected in Dunmore.) While three judges in the Trilogy expressed the view that s. 2(d) did not protect collective bargaining, the only question at issue in those cases was whether individuals had a right to strike, and the question of collective bargaining was not conclusively resolved by a majority of the Court.

(b) Dunmore

This Court’s decision in Dunmore marked a new stage in the development of s. 2(d) jurisprudence in the field of labour relations. It raised the question of whether s. 2(d) requires the government to provide a legislative framework which enables employees to associate to obtain workplace goals in a meaningful process.

The Ontario legislature had repealed legislation which gave farm workers a measure of protection, leaving them entirely outside the Province’s labour relations scheme. The evidence established that attempts to organize were persistently frustrated by employers. The farm workers came to court seeking protection of their basic right to associate. They sought the right to organize into employee associations. For this, they contended, they needed legislation that endorsed their constitutional right to associate and protected them from employer interference.

Bastarache J., for the majority of the Court, began the analysis in Dunmore by emphasizing the need for a purposive approach to s. 2(d) — “one which aims to protect the full range of associational activity contemplated by the Charter and to honour Canada’s obligations under international human rights law” (para. 13). After a full review of the jurisprudence, he stated:

. . . the activities for which the appellants seek protection [association for the purposes of achieving workplace goals in the labour relations context] fall squarely within the freedom to organize, that is, the freedom to collectively embody the interests of individual workers. [para. 30]

Bastarache J. went on to hold that in order to realize the purposes of s. 2(d), the right to organize must extend to “the exercise of certain collective activities, such as making majority representations to one’s employer”. He explained:

These activities are guaranteed by the purpose of s. 2(d), which is to promote the realization of individual potential through relations with others, and by international labour jurisprudence, which recognizes the inevitably collective nature of the freedom to organize. [para. 30]

The affirmation that s. 2(d) protection extends to collective activities that only a group can carry out, required rejection of Sopinka J.’s fourth proposition in PIPSC, which suggested that s. 2(d) only protected the right to further individual goals. Bastarache J. pointed out that certain activities are, when performed by a group, “qualitatively” different from those activities performed solely by an individual. He recognized that “trade unions develop needs and priorities that are distinct from those of their members individually” (Dunmore, at para. 17). As a result “certain collective activities must be recognized if the freedom to form and maintain an association is to have any meaning” (ibid.).

In the result, Bastarache J. concluded that the absence of legislative protection for farm workers to organize in order to achieve workplace goals made meaningful association to achieve workplace goals impossible and therefore constituted a substantial interference with the right to associate guaranteed by s. 2(d) of the Charter . He found that the absence of legislative support discredited the organizing efforts of agricultural workers and had a chilling effect on their constitutional right to associate. He concluded that farm workers in Ontario were substantially incapable of exercising their fundamental freedom to associate without a protective regime (para. 35). Quoting L’Heureux-Dubé J. in Delisle, Bastarache J. affirmed that

the right to freedom of association must take into account the nature and importance of labour associations as institutions that work for the betterment of working conditions and the protection of the dignity and collective interests of workers in a fundamental aspect of their lives: employment. [Emphasis deleted; para. 37.]

After Dunmore, there could be no doubt that the right to associate to achieve workplace goals in a meaningful and substantive sense is protected by the guarantee of freedom of association, and that this right extends to realization of collective, as distinct from individual, goals. Nor could there be any doubt that legislation (or the absence of a legislative framework) that makes achievement of this collective goal substantially impossible, constitutes a limit on the exercise of freedom of association. Finally, there could be no doubt that the guarantee must be interpreted generously and purposively, in accordance with Canadian values and Canada’s international commitments.

It is worth pausing at this juncture to summarize the propositions that led the majority of the Court in Dunmore to these conclusions.

Ÿ Section 2(d), interpreted purposively, guarantees freedom of associational activity in the pursuit of individual and common goals.

Ÿ The common goals protected extend to some collective bargaining activities, including the right to organize and to present submissions to the employer.

Ÿ What is required is a process that permits the meaningful pursuit of these goals. No particular outcome is guaranteed. However, the legislative framework must permit a process that makes it possible to pursue the goals in a meaningful way.

Ÿ The effect of a process that renders impossible the meaningful pursuit of collective goals is to substantially interfere with the exercise of the right to free association, in that it negates the very purpose of the association and renders it effectively useless. This constitutes a limit under s. 2(d) which is unconstitutional unless justified by the state under s. 1 of the Charter . (This is an application of the settled rule that a law or government act that in purpose or effect constrains exercise of a right constitutes a limitation for purposes of s. 1: see Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927.)

Ÿ The remedy for the resultant breach of s. 2(d) is to order the state to rectify the legislative scheme to make possible meaningful associational activity in pursuit of common workplace goals.

(c) Health Services: Its Impact

Dunmore established that claimants must demonstrate the substantial impossibility of exercising their freedom of association in order to compel the government to enact statutory protections. It did not, however, define the ambit of the right of association protected by s. 2(d) in the context of collective bargaining. Relying on Dunmore, the majority of the Court in Health Services, per McLachlin C.J. and LeBel J., held that legislation and government actions that repealed existing collective agreements and substantially interfered with the possibility of meaningful collective bargaining in the future constituted a limit on the s. 2(d) right of freedom of association.

The claimants were various unions and their members working in the health services industry of British Columbia. The industry was highly unionized and had negotiated collective agreements regarding salaries, benefits and working conditions. The government, directly or indirectly, was the employer. The government wanted to reduce costs by changing the structure of its employees’ working arrangements in ways that would have been impermissible under the existing collective agreements. It chose to do so, not through collective bargaining to the end of altering those collective agreements, but by the simple expedient of legislation. In short, the government used its legislative powers to effectively nullify the collective agreements to its benefit, and to the detriment of its employees. The legislation not only conflicted with existing collective agreements, but also precluded collective bargaining in the future on a number of issues and conditions of employment. (See R. K. Basu, “Revolution and Aftermath: B.C. Health Services and Its Implications” (2008), 42 S.C.L.R. (2d) 165, at p. 177; see also M. Coutu, L. L. Fontaine and G. Marceau, Droit des rapports collectifs du travail au Québec (2009), at p. 144.)

The unions responded by bringing an action claiming that the government had breached s. 2(d) by legislatively interfering with freedom of association. They further claimed that the government had done so in circumstances that could not be justified under s. 1 of the Charter . Health Services thus put directly in issue the right to collective bargaining. The claimants did not seek the enactment of associational protections. Rather, they asserted that s. 2(d) protected a right to collective bargaining and that the government had violated the constitutional guarantee of freedom of association by legislating to both overturn existing contracts and preclude effective collective bargaining in the future. The unions lost at trial and on appeal but succeeded in this Court.

While Health Services concerned the actions of a government employer nullifying collective bargaining arrangements with unions representing its own employees, the Court rested its decision on a more general discussion of s. 2 of the Charter . Applying the principles of interpretation established in Dunmore, a majority of the Court held that s. 2(d) includes “a process of collective action to achieve workplace goals” (para. 19). This process requires the parties to meet and bargain in good faith on issues of fundamental importance in the workplace (para. 90). By legislating to undo the existing collective bargaining arrangements and by hampering future collective bargaining on important workplace issues, the British Columbia government had “substantially interfered” with the s. 2(d) right of free association, and had failed to justify the resultant limitation on the exercise of the right under s. 1 of the Charter (paras. 129-61).

The decision in Health Services follows directly from the principles enunciated in Dunmore. Section 2(d), interpreted purposively and in light of Canada’s values and commitments, protects associational collective activity in furtherance of workplace goals. The right is not merely a paper right, but a right to a process that permits meaningful pursuit of those goals. The claimants had a right to pursue workplace goals and collective bargaining activities related to those goals. The government employer passed legislation and took actions that rendered the meaningful pursuit of these goals impossible and effectively nullified the right to associate of its employees. This constituted a limit on the exercise of s. 2(d), and was thus unconstitutional unless justified under s. 1 of the Charter .

While the majority decision in Health Services sits firmly within the principles the Court had earlier set out in Dunmore, in its discussion of the s. 2(d) right the Court went on to explain in greater detail what the government must permit in order to avoid the charge of substantial interference with the s. 2(d) right in the context of collective action in pursuit of workplace goals. In Dunmore, Bastarache J. stated that “the effective exercise of these freedoms may require . . . the exercise of certain collective activities, such as making majority representations to one’s employer” (para. 30). It remained uncertain what other collective activities might be protected.

The majority of the Court in Health Services affirmed that bargaining activities protected by s. 2(d) in the labour relations context include good faith bargaining on important workplace issues (para. 94; see also paras. 93, 130 and 135). This is not limited to a mere right to make representations to one’s employer, but requires the employer to engage in a process of consideration and discussion to have them considered by the employer. In this sense, collective bargaining is protected by s. 2(d). The majority stated:

Thus the employees’ right to collective bargaining imposes corresponding duties on the employer. It requires both employer and employees to meet and to bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation. [para. 90]

By way of elaboration on what constitutes good faith negotiation, the majority of the Court stated:

Ÿ Section 2(d) requires the parties to meet and engage in meaningful dialogue. They must avoid unnecessary delays and make a reasonable effort to arrive at an acceptable contract (paras. 98, 100-101);

Ÿ Section 2(d) does not impose a particular process. Different situations may demand different processes and timelines (para. 107);

Ÿ Section 2(d) does not require the parties to conclude an agreement or accept any particular terms and does not guarantee a legislated dispute resolution mechanism in the case of an impasse (paras. 102-3);

Ÿ Section 2(d) protects only “the right . . . to a general process of collective bargaining, not to a particular model of labour relations, nor to a specific bargaining method” (para. 91).

The Court in Health Services emphasized that s. 2(d) does not require a particular model of bargaining, nor a particular outcome. What s. 2(d) guarantees in the labour relations context is a meaningful process. A process which permits an employer not even to consider employee representations is not a meaningful process. To use the language of Dunmore, it is among those “collective activities [that] must be recognized if the freedom to form and maintain an association is to have any meaning” (para. 17). Without such a process, the purpose of associating in pursuit of workplace goals would be defeated, resulting in a significant impairment of the exercise of the right to freedom of association. One way to interfere with free association in pursuit of workplace goals is to ban employee associations. Another way, just as effective, is to set up a system that makes it impossible to have meaningful negotiations on workplace matters. Both approaches in fact limit the exercise of the s. 2(d) associational right, and both must be justified under s. 1 of the Charter to avoid unconstitutionality.

In summary, Health Services applied the principles developed in Dunmore and explained more fully what is required to avoid interfering with associational activity in pursuit of workplace goals and undermining the associational right protected by s. 2(d). Its suggestion that this requires a good faith process of consideration by the employer of employee representations and of discussion with their representatives is hardly radical. It is difficult to imagine a meaningful collective process in pursuit of workplace aims that does not involve the employer at least considering, in good faith, employee representations. The protection for collective bargaining in the sense affirmed in Health Services is quite simply a necessary condition of meaningful association in the workplace context.

(d) The Issue on This Appeal

Against this background, we return to the issue in this case. The Court of Appeal held that Health Services constitutionalizes a full-blown Wagner system of collective bargaining, and concluded that since the AEPA did not provide such a model, absent s. 1 justification, it is unconstitutional. The court appears to have understood the affirmation of the right to collective bargaining in Health Services as an affirmation of a particular type of collective bargaining, the Wagner model which is dominant in Canada.

With respect, this overstates the ambit of the s. 2(d) right as described in Health Services. First, as discussed, the majority in Health Services unequivocally stated that s. 2(d) does not guarantee a particular model of collective bargaining or a particular outcome (para. 91).

Second, and more fundamentally, the logic of Dunmore and Health Services is at odds with the view that s. 2(d) protects a particular kind of collective bargaining. As discussed earlier, what s. 2(d) protects is the right to associate to achieve collective goals. Laws or government action that make it impossible to achieve collective goals have the effect of limiting freedom of association, by making it pointless. It is in this derivative sense that s. 2(d) protects a right to collective bargaining: see Ontario (Public Safety and Security) v. Criminal Lawyers’ Association, 2010 SCC 23, [2010] 1 S.C.R. 815 (“CLA”), where the right to access government information was held to be “a derivative right which may arise where it is a necessary precondition of meaningful expression on the functioning of government” (para. 30). However, no particular type of bargaining is protected. 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.

It follows that Health Services does not support the view of the Ontario Court of Appeal in this case that legislatures are constitutionally required, in all cases and for all industries, to enact laws that set up a uniform model of labour relations imposing a statutory duty to bargain in good faith, statutory recognition of the principles of exclusive majority representation and a statutory mechanism for resolving bargaining impasses and disputes regarding the interpretation or administration of collective agreements (C.A. reasons, at para. 80). What is protected is associational activity, not a particular process or result. If it is shown that it is impossible to meaningfully exercise the right to associate due to substantial interference by a law (or absence of laws: see Dunmore) or by government action, a limit on the exercise of the s. 2(d) right is established, and the onus shifts to the state to justify the limit under s. 1 of the Charter .

The resolution of this appeal does not rest on stark reliance on a particular conception of collective bargaining. Rather, it requires us to return to the principles that underlie the majority rulings in Dunmore and Health Services. The question here, as it was in those cases, is whether the legislative scheme (the AEPA) renders association in pursuit of workplace goals impossible, thereby substantially impairing the exercise of the s. 2(d) associational right.

Justice Deschamps adopts a narrow interpretation of the majority reasons in Health Services, stating that they merely recognized “that freedom of association includes the freedom to engage in associational activities and the ability of employees to act in common to reach shared goals related to workplace issues and terms of employment” (para. 308). In her view, it was unnecessary for the majority in that case to consider the duty to negotiate in good faith, and consequently argues that the passages of the majority judgment that discussed this duty were in obiter.

However, such a narrow interpretation of the majority reasons in Health Services would not support the holding in that case. If s. 2(d) merely protected the right to act collectively and to make collective representations, the legislation at issue in that case would have been constitutional. The legislation in that case violated s. 2(d) since it undermined the ability of workers to engage in meaningful collective bargaining, which the majority defined as good faith negotiations (para. 90). The majority underlined that

the right to bargain collectively protects not just the act of making representations, but also the right of employees to have their views heard in the context of a meaningful process of consultation and discussion. This rebuts arguments made by the respondent that the Act does not interfere with collective bargaining because it does not explicitly prohibit health care employees from making collective representations. While the language of the Act does not technically prohibit collective representations to an employer, the right to collective bargaining cannot be reduced to a mere right to make representations. [Emphasis added; para. 114.]

In our view, the majority decision in Health Services should be interpreted as holding what it repeatedly states: that workers have a constitutional right to make collective representations and to have their collective representations considered in good faith.

Justice Rothstein argues that Health Services represents a radical departure from previous jurisprudence and was wrongly decided.

The central argument of our colleague is that s. 2(d) of the Charter does not protect collective bargaining. He understands the majority decision in Health Services to have constitutionalized collective bargaining. That, he says, is wrong in principle, inconsistent with the Court’s prior jurisprudence, and unworkable in practice.

Our colleague appears to interpret Health Services as establishing directly or indirectly a Wagner model of labour relations. The actual holding of Health Services, as discussed above, was more modest. Health Services affirms a derivative right to collective bargaining, understood in the sense of a process that allows employees to make representations and have them considered in good faith by employers, who in turn must engage in a process of meaningful discussion. The logic that compels this conclusion, following settled Charter jurisprudence, is that the effect of denying these rights is to render the associational process effectively useless and hence to substantially impair the exercise of the associational rights guaranteed by s. 2(d). No particular bargaining model is required.

Rothstein J. calls for Health Services to be overturned. But his views also imply overturning Dunmore, on which Health Services rests. Rothstein J. states that “the essence of freedom of association is that it enables individuals to do in association what they could do as individuals” (para. 197). This echoes the model of s. 2(d) adopted by the plurality in PIPSC, and rejected by Bastarache J. in Dunmore. For the reasons that follow, we remain of the view that Dunmore and Health Services represent good law and should not be overruled.

(i) The Caution Required in Overturning Precedent

Our colleague correctly recognizes at the outset of his reasons that overturning a precedent of this Court is a step not to be lightly undertaken. We would note that as we understand the law (see above), rejection of Health Services implies rejection of Dunmore as well, since the two cases rest on the same fundamental logic.

The seriousness of overturning two recent precedents of this Court, representing the considered views of firm majorities, cannot be overstated. This is particularly so given their recent vintage. Health Services was issued only four years ago, and, when this appeal was argued, only two years had passed.

Rothstein J. suggests that since Health Services deals with constitutional law, the Court should be more willing to overturn it (paras. 141-43). In our respectful view, this argument is not persuasive. The constitutional nature of a decision is not a primary consideration when deciding whether or not to overrule, but at best a final consideration in difficult cases. Indeed, the fact that Health Services relates to a constitutional Charter right may militate in favour of upholding this past decision. As Binnie J. stated on behalf of a unanimous Court in R. v. Henry, 2005 SCC 76, [2005] 3 S.C.R. 609, “[t]he Court should be particularly careful before reversing a precedent where the effect is to diminish Charter protection” (para. 44). Justice Rothstein’s proposed interpretation of s. 2(d) of the Charter would diminish the scope of the s. 2(d) right.

We note as well that, while the Court in this case was asked to clarify the ambit of Health Services, it was not asked to overrule it. British Columbia, the respondent in Health Services, stated explicitly that it was “not here contesting this Court’s conclusion that s. 2(d) protects a process of collective bargaining” (Factum of the Attorney General of British Columbia, at para. 18). Absent notice to the profession and interested persons, overruling Health Services seems to us procedurally inappropriate.

In our view, the arguments advanced by our colleague against Health Services do not meet the high threshold for reversing a precedent of this Court.

(ii) The Arguments on Jurisprudence

Justice Rothstein argues that Health Services represents a marked departure from prior jurisprudence. We do not agree.

Our colleague argues that the recognition of a constitutional right to collective bargaining in Health Services is not supported by the purpose of s. 2(d), because it improperly assigns a collective dimension to individual rights. The collective dimension of individual rights was recognized by Dickson C.J., dissenting in Alberta Reference, stating that s. 2(d) protects group activity for which activity there is “no analogy involving individuals” such as the right to bargain collectively (pp. 367-70). The Court in Dunmore modified the fourth proposition in the earlier case of PIPSC. As Bastarache J. there stated, “certain collective activities must be recognized if the freedom to form and maintain an association is to have any meaning”.

Consistent with this framework, the majority decision in Health Services framed s. 2(d) as an individual right (“the right of employees”, para. 87 (emphasis added)) that may require the protection of group activity (see also paras. 19 and 89). The fundamental inquiry is whether the state action would substantially impair the ability of “union members to pursue shared goals in concert” (para. 96 (emphasis added)). As in Dunmore, the majority concluded that the realization of the individual right required a capacity to act in common, which may give rise to a need to protect group activities and, as a consequence, to recognize group rights.

In summary, Health Services was consistent with the previous cases on the issue of individual and collective rights. It recognized, as did previous jurisprudence, that s. 2(d) is an individual right. But it also recognized, as did previous cases, that to meaningfully uphold this individual right, s. 2(d) may properly require legislative protection of group or collective activities.

Rothstein J. also emphasizes that “[i]ndividuals who are not members of an association . . . have no constitutional right to oblige their employers to bargain” (paras. 179 and 187). In our view, this outcome is not anomalous. It follows logically from the fact that collective bargaining is a derivative right, a “necessary precondition” to the meaningful exercise of the constitutional guarantee of freedom of association: see CLA, at para. 30. Where there is no reliance on freedom of association, there is no derivative right to require employers to bargain.

(iv) The Argument That Section 2(d) Is a Freedom, Not a Right

Our colleague argues that by requiring a process that allows for meaningful dialogue on workplace matters, Health Services wrongly converts a negative freedom into a positive right. This bright line between freedoms and rights seems to us impossible to maintain. Just as freedom of expression implies correlative rights, so may freedom of association. The freedom to do a thing, when guaranteed by the Constitution interpreted purposively, implies a right to do it. The Charter cannot be subdivided into two kinds of guarantees — freedoms and rights.

The majority in both Dunmore and Health Services held that freedom to associate may require the state to take positive steps. Bastarache J. in Dunmore underlined that “it may be asked whether the distinction between positive and negative state obligations ought to be nuanced in the context of labour relations” (para. 20). He further noted that

history has shown, and Canada’s legislatures have uniformly recognized, that a posture of government restraint in the area of labour relations will expose most workers not only to a range of unfair labour practices, but potentially to legal liability under common law inhibitions on combinations and restraints of trade. . . . In this context, it must be asked whether, in order to make the freedom to organize meaningful, s. 2(d) of the Charter imposes a positive obligation on the state to extend protective legislation to unprotected groups. [para. 20]

This Court has consistently rejected a rigid distinction between “positive” freedoms and “negative” rights in the Charter . For example, it recently held that s. 2(b) may require the government to disclose documents to the public in order to enable meaningful discourse: CLA, at para. 37. As stated by L’Heureux-Dubé J. in Haig v. Canada, [1993] 2 S.C.R. 995:

The distinctions between “freedoms” and “rights”, and between positive and negative entitlements, are not always clearly made, nor are they always helpful. One must not depart from the context of the purposive approach articulated by this Court in R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295. Under this approach, a situation might arise in which, in order to make a fundamental freedom meaningful, a posture of restraint would not be enough, and positive governmental action might be required. This might, for example, take the form of legislative intervention aimed at preventing certain conditions which muzzle expression, or ensuring public access to certain kinds of information. [p. 1039]

A purposive protection of freedom of association may require the state to act positively to protect the ability of individuals to engage in fundamentally important collective activities, just as a purposive interpretation of freedom of expression may require the state to disclose documents to permit meaningful discussion.

With respect, we also do not agree that the structure of the Charter reflects a rigid distinction between freedoms and rights. Rothstein J.’s reasons state that “[w]hen the Charter uses the term ‘right’, as it does in ss. 7 to 12 , either a positive entitlement is introduced, or a right to be free of some restriction or prohibition (i.e. a freedom) is introduced” (para. 192).

In fact, many of the rights in ss. 7 to 12 do not entitle individuals to any form of state action. Rather these provisions guarantee a mixture of negative and positive rights. For instance, s. 9 protects “the right not to be arbitrarily detained or imprisoned” and s. 12 protects “the right not to be subjected to any cruel and unusual treatment or punishment”. But s. 10 also protects a right to counsel and imposes a corresponding duty on police officers to facilitate the exercise of this right. See also Charkaoui v. Canada (Citizenship and Immigration), 2007 SCC 9, [2007] 1 S.C.R. 350, at para. 107, referring to “the s. 12 guarantee of freedom from cruel and unusual treatment” (emphasis added).

It may also be observed that Health Services does not impose constitutional duties on private employers, but on governments as employers and parliaments and legislatures as law makers, in accordance with s. 32 of the Charter . Rather, the majority held that individuals have a right against the state to a process of collective bargaining in good faith, and that this right requires the state to impose statutory obligations on employers. As held by Cory and Iacobucci JJ. in Vriend v. Alberta, [1998] 1 S.C.R. 493, one must “distinguish between ‘private activity’ and ‘laws that regulate private activity’. The former is not subject to the Charter , while the latter obviously is” (para. 66). If workers are incapable of exercising their right to collective bargaining, they may only bring a Charter claim against the government and not their employer, and they must show a state action.

Our colleague argues that the effect of Health Services is to privilege some associations over others, by interpreting s. 2(d) in a way that is not content-neutral. Broadly put, the argument appears to be that, by considering the goals of a particular association, one moves beyond pure associational activity into a court-based selection of what goals are acceptable and what goals are not.

Yet consideration of goals cannot be avoided. One of the basic principles of Charter interpretation is that rights must be interpreted in a purposive way — having regard for the purposes, or goals, they serve. Thus, in the Alberta Reference, McIntyre J. described the core of s. 2(d) protection as being association “both to satisfy [a] desire for social intercourse and to realize common purposes” (p. 395 (emphasis added)). Dunmore resolved the issue, not by saying that the s. 2(d) right must be content-neutral, but by asserting that it must be interpreted in conjunction with Canada’s values and international human rights and labour law commitments. In our view, this is the preferable approach. A content-neutral right is too often a meaningless right.

(vi) The Argument That Health Services Gives Contracts Priority Over Statutes

Our colleague argues that Health Services gives constitutional status to contracts, privileging them over statutes. The argument is based on the view that Health Services holds that breach of collective agreements violates s. 2(d). In fact, as discussed above, this was not the finding in Health Services. The majority in Health Services held that the unilateral nullification of significant contractual terms, by the government that had entered into them or that had overseen their conclusion, coupled with effective denial of future collective bargaining, undermines the s. 2(d) right to associate, not that labour contracts could never be interfered with by legislation.

(vii) The Argument That Health Services Removes Judicial Deference to the Legislation

Our colleague argues that Health Services undercuts the judicial deference courts have paid in the past to the legislature in labour relations. We observe at the outset that this argument rests on the premise — repeatedly rejected in Health Services — that the Court was constitutionally enshrining the Wagner model of labour relations. In fact this is not what Health Services held.

As stated in Health Services, “[i]t may well be appropriate for judges to defer to legislatures on policy matters expressed in particular laws” (para. 26). What Health Services rejected was a judicial “no go” zone for an entire right on the ground that it may involve the courts in policy matters: creating such a Charter-free zone would “push deference too far” (ibid.). This Court reached a similar conclusion in U.F.C.W., Local 1518 v. KMart Canada Ltd., [1999] 2 S.C.R. 1083, at paras. 62-63.

The approach to deference advanced in Health Services is consistent with this Court’s general jurisprudence. Deference should inform the determination of whether Parliament’s scheme satisfies the requirements of the Charter , as articulated by the courts. See P. Macklem, “Developments in Employment Law: The 1990-91 Term” (1992), 3 S.C.L.R. (2d) 227, at pp. 239-41. Conversely, the courts should not rely on deference to narrow the meaning of Charter rights in the first place. Doing so would abdicate the courts’ duty as the “final arbiters of constitutionality in Canada” (Nova Scotia (Workers’ Compensation Board) v. Martin, 2003 SCC 54, [2003] 2 S.C.R. 504, at para. 31).

In R. v. Advance Cutting & Coring Ltd., 2001 SCC 70, [2001] 3 S.C.R. 209, the reasons of both LeBel J. and Bastarache J. (dissenting) rejected the view that deference to Parliament on labour relations created a Charter-free zone of legislative action, although there was disagreement on the scope of s. 2(d) and the application of the Charter . The reasons of LeBel J., written on behalf of a minority of the Court in finding no violation of s. 2(d), acknowledged the existence of a jurisprudential policy of non-intervention in labour relations. However, LeBel J. also stated that “the jurisprudence of this Court has never held that labour laws are immune to Charter review” (para. 162).

Rothstein J. argues that courts should consider deference to Parliament in determining the scope of s. 2(d). This approach is inconsistent with this Court’s decision in Dunmore. While Rothstein J. adopts the approach of McIntyre J. in the Alberta Reference, this Court has since distanced itself from this view. In Dunmore, Bastarache J. referred to McIntyre J.’s discussion of deference under s. 1, rather than in outlining the scope of s. 2(d): para. 57; see also Delisle, per Cory and Iacobucci JJ., dissenting; KMart Canada Ltd., at paras. 62-63, in which Cory J. referred to this passage from McIntyre J.’s reasons under s. 1 in a freedom of expression case. Deference to legislatures properly plays a part, not in defining the nature and scope of a constitutional right, but within the margin of appreciation that the Oakes analytical process acknowledges, particularly at the minimal impairment stage.

Rothstein J. argues that Health Services is unworkable and therefore must be overturned (para. 256).

The short answer to this argument is that unworkability has not been established. Winkler C.J.O. speculates that more will be required to make Health Services work, and academics have weighed in with great passion, some in favour of the decision, some against it. But there is no concrete evidence that the principles enunciated in Dunmore and Health Services are unworkable or have led to intolerable results. It is premature to argue that the holding in Health Services, rendered four years ago, is unworkable in practice. In Henry, in holding this Court’s decision in R. v. Mannion, [1986] 2 S.C.R. 272, to be unworkable 19 years after it was delivered, Binnie J. noted that the unworkability of that decision “only emerged over time as the courts have struggled to work with the distinction between impeachment of credibility and incrimination” (para. 45).

Rothstein J. argues that the distinction drawn in Health Services between substantive and procedural rights is unworkable. Again, we must disagree. In our colleague’s view, the procedural right to collective activity under s. 2(d) would impinge on the substantive right to a concluded collective agreement rejected in Health Services. However, substantive impact does not invalidate a procedural right. All procedures affect outcomes, but that does not mean that all procedural rights are unworkable. The Charter may protect collective bargaining and not the fruits of that process.

Rothstein J. also suggests that more is required to transform the principles in Health Services into a full-blown labour relations scheme. This, however, does not establish unworkability. It is not the role of this Court to specify in advance precisely which model of labour relations the legislature should adopt. Instead, its role is to outline the boundaries within which the legislature must operate, and to assess if the scheme developed by legislators satisfies this test.

Justice Rothstein argues that academic criticism supports the view that Health Services should be overturned. While he agrees that criticism of a judgment is not sufficient to justify overruling it, he asserts that it is reason for the Court to “take notice”.

The first point to note is that the decisions that Rothstein J. relies on, the Trilogy and PIPSC, were themselves the subject of intense academic criticism (see, e.g., Macklem, at p. 240: “the combined effect of the Labour Trilogy and P.I.P.S. is a national embarrassment”; see also D. Beatty and S. Kennett, “Striking Back: Fighting Words, Social Protest and Political Participation in Free and Democratic Societies” (1988), 67 Can. Bar Rev. 573; A. C. Hutchinson and A. Petter, “Private Rights/Public Wrongs: The Liberal Lie of the Charter ” (1988), 38 U.T.L.J. 278; R. Elliot, “Developments in Constitutional Law: The 1989-90 Term” (1991), 2 S.C.L.R. (2d) 83). The real question is whether the academic criticism raises concerns not identified in Health Services that would justify overruling it.

Moreover, as our colleague acknowledges, there was a range of opinions expressed about the decision in Health Services. As is often the case, some commentators agree, while others disagree in whole or in part. A number of comments approved of the shift away from the Trilogy in favour of a broader and more contextual understanding of freedom of association: see, e.g., C. Brunelle, “La liberté d’association se porte mieux: un commentaire de l’arrêt Health Services”, in Conférence des juristes de l’État 2009: XVIIIe Conférence (2009), 237; P. Verge, “L’affirmation constitutionnelle de la liberté d’association: une nouvelle vie pour l’autonomie collective?” (2010), 51 C. de D. 353. Indeed, some commentators fault this Court’s decision for not going far enough in protecting collective bargaining and related issues: see, e.g., P. Verge, “La Cour suprême, le ‘droit’ à la négociation collective et le ‘droit’ de grève” (2006), 66 R. du B. 391; P. Verge, “Inclusion du droit de grève dans la liberté générale et constitutionnelle d’association: justification et effets” (2009), 50 C. de D. 267; J. Cameron, “The Labour Trilogy’s Last Rites: B.C. Health and a Constitutional Right to Strike” (2009-2010), 15 C.L.E.L.J. 297; J. Cameron, “Due Process, Collective Bargaining, and s. 2(d) of the Charter : A Comment on B.C. Health Services” (2006-2007), 13 C.L.E.L.J. 233.

Rothstein J. takes issue with the discussion of Canadian labour history in Health Services, pointing out that hostility to collective bargaining is part of Canadian labour law history. We agree with this obvious fact, which was largely true until the Second World War, which is indeed referred to in the majority reasons in Health Services.

The relevant question from the perspective of interpreting s. 2(d) of the Charter is not whether courts in the past have undermined collective bargaining, but rather whether Canadian society’s understanding of freedom of association, viewed broadly, includes the right to collective bargaining in the minimal sense of good faith exchanges affirmed in Health Services. Whether that right has been consistently guaranteed by the legal system does not resolve the issue before us, the content of the s. 2(d) guarantee. Charter guarantees must be given a generous and purposive interpretation. While the practice of courts pre-Charter may assist in interpreting Charter guarantees, it does not freeze them forever in a pre-Charter vice.

Rothstein J. takes issue with the majority’s conclusion in Health Services that international law supports a finding that s. 2(d) includes a right to collective bargaining.

The majority in Health Services discussed both “Canada’s current international law commitments and the current state of international thought on human rights” (para. 78 (underlining added)). Charter rights must be interpreted in light of Canadian values and Canada’s international and human rights commitments. In Dunmore, Bastarache J. emphasized the relevance of these in interpreting s. 2(d) in the context.

The fundamental question from the perspective of s. 2(d) is whether Canada’s international obligations support the view that collective bargaining is constitutionally protected in the minimal sense discussed in Health Services. The majority in Health Services relied on three documents that Canada has endorsed: the International Covenant on Economic, Social and Cultural Rights, 993 U.N.T.S. 3, the International Covenant on Civil and Political Rights, 999 U.N.T.S. 171, and the International Labour Organization’s (“ILO’s”) Convention (No. 87) concerning freedom of association and protection of the right to organise, 68 U.N.T.S. 17 (“Convention No. 87”).

The decision rendered by the ILO Committee on Freedom of Association (“CFA”), in the conflict between the employees of the B.C. health services and the government of British Columbia, concerned the very conflict that formed the factual background of the decision in Health Services. After applying Convention No. 87 and noting that Canada had not ratified Convention (No. 98) concerning the application of the principles of the right to organise and to bargain collectively, 96 U.N.T.S. 257 (“Convention No. 98”), the CFA concluded that the action of the government of British Columbia violated the employees’ right to freedom of association. It stated that the unilateral cancellation of collective agreements “may have a detrimental effect on workers’ interests in unionization, since members and potential members could consider it useless to join an organization the main objective of which is to represent its members in collective bargaining, if the results of bargaining are constantly cancelled by law” (Report No. 330 (2003), vol. LXXXVI, Series B, No. 1, at para. 304).

Rothstein J. argues that Convention No. 98 (which is not binding on Canada) does not support “a version of collective bargaining that includes a duty to bargain in good faith” (para. 249). While voluntariness is a component of the international model of collective bargaining — as noted by the majority in Health Services (para. 77, citing B. Gernigon, A. Odero and H. Guido, “ILO principles concerning collective bargaining” (2000), 139 Intern’l Lab. Rev. 33, at pp. 51-52) — the ILO Committee of Experts has not found compulsory collective bargaining to be contrary to international norms. The 1994 Report of the Committee of Experts discussed the domestic schemes that compelled employers to bargain with unions, listing Canada, and approvingly stated that such schemes illustrated “the principle that employers and trade unions should negotiate in good faith and endeavour to reach an agreement” (Committee of Experts on the Application of Conventions and Recommendations, Freedom of Association and Collective Bargaining (1994), at para. 243). This is precisely the general principle that Health Services endorses.

Rothstein J. argues that the majority in Health Services erred in relying on the underlying values of the Charter when interpreting the scope of s. 2(d) rather than on the text of the Charter itself (paras. 252-54). We can only respond that a value‑oriented approach to the broadly worded guarantees of the Charter has been repeatedly endorsed by Charter jurisprudence over the last quarter century.

Notwithstanding the comprehensive reasons of our colleague, we conclude that Health Services is grounded in precedent, consistent with Canadian values, consistent with Canada’s international commitments and consistent with this Court’s purposive and generous interpretation of other Charter guarantees. In our view, it should not be overturned.

(2) Application: Have the Respondents Established a Breach of Section 2(d)?

The essential question is whether the AEPA makes meaningful association to achieve workplace goals effectively impossible, as was the case in Dunmore. If the AEPA process, viewed in terms of its effect, makes good faith resolution of workplace issues between employees and their employer effectively impossible, then the exercise of the right to meaningful association guaranteed by s. 2 (d) of the Charter will have been limited, and the law found to be unconstitutional in the absence of justification under s. 1 of the Charter . The onus is on the farm workers to establish that the AEPA interferes with their s. 2(d) right to associate in this way.

As discussed above, the right of an employees’ association to make representations to the employer and have its views considered in good faith is a derivative right under s. 2(d) of the Charter , necessary to meaningful exercise of the right to free association. The question is whether the AEPA provides a process that satisfies this constitutional requirement.

Under the AEPA, the right of employees’ associations to make representations to their employers is set out in s. 5. The relevant sections are ss. 5(1), (5), (6) and (7):

5. (1) The employer shall give an employees’ association a reasonable opportunity to make representations respecting the terms and conditions of employment of one or more of its members who are employed by that employer.

. . .

(5) The employees’ association may make the representations orally or in writing.

(6) The employer shall listen to the representations if made orally, or read them if made in writing.

(7) If the representations are made in writing, the employer shall give the association a written acknowledgment that the employer has read them.

Sections 5(6) and (7) are critical. They provide that the employer shall listen to oral representations, and read written representations, and acknowledge having read them. They do not expressly refer to a requirement that the employer consider employee representations in good faith. Nor do they rule it out. By implication, they include such a requirement.

Three considerations lead us to conclude that any ambiguity in ss. 5(6) and (7) should be resolved by interpreting them as imposing a duty on agricultural employers to consider employee representations in good faith.

The first consideration is the principle that a statute should be interpreted in a way that gives meaning and purpose to its provisions. This requires us to ask what the purpose of the requirements in ss. 5(6) and (7) is. There can only be one purpose for requiring the employer to listen to or read employee representations — to assure that the employer will in fact consider the employee representations. No labour relations purpose is served merely by pro forma listening or reading. To fulfill the purpose of reading or listening, the employer must consider the submission. Moreover, the employer must do so in good faith: consideration with a closed mind would render listening or reading the submission pointless.

The second consideration is that Parliament and legislatures are presumed to intend to comply with the Charter : Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038, at p. 1078, per Lamer J. (as he then was), dissenting in part; R. v. Sharpe, 2001 SCC 2, [2001] 1 S.C.R. 45, at para. 33. At the time the AEPA was adopted Dunmore had pronounced that the Charter requires meaningful exercise of the right to associate in pursuit of workplace goals. Since Health Services, it has been clear that this requires employers to consider employee representations in good faith. Any ambiguity in the AEPA should be resolved accordingly.

The third consideration is the expressed intention of the Minister in debates on the legislation. When introducing the legislation, she stated:

The government is advised that the Supreme Court of Canada decision regarding Dunmore versus Ontario obligates the government to extend legislative protections to agricultural workers. It obligates us to do this to ensure that employees have the right to form and join associations, as well as have the protection necessary to ensure that the freedom of association is meaningful. The government of Ontario will meet these obligations.

(Legislative Assembly of Ontario, Official Report of Debates (Hansard), No. 46A, 3rd Sess., 37th Parl., October 22, 2002, at p. 2340 (emphasis added))

The government must, on the words of its Minister, have intended the legislation to achieve whatever is required to ensure meaningful exercise of freedom of association. As discussed above, meaningful exercise of the right to free association in the workplace context requires good faith consideration of employee representations. As pointed out by the respondents, the Minister also stated that the AEPA was not intended to “extend collective bargaining to agricultural workers”. However, this may be understood as an affirmation that the AEPA did not institute the dominant Wagner model of collective bargaining, or bring agricultural workers within the ambit of the LRA, not that the Minister intended to deprive farm workers of the protections of collective bargaining that s. 2(d) grants.

These considerations lead us to conclude that s. 5 of the AEPA, correctly interpreted, protects not only the right of employees to make submissions to employers on workplace matters, but also the right to have those submissions considered in good faith by the employer. It follows that s. 5 of the AEPA does not violate s. 2(d) of the Charter .

It is argued that the record thus far under the AEPA gives little reason to think that the AEPA process will in fact lead to good faith consideration by employers. The evidence shows that the respondents attempted to engage employers in collective bargaining activities on a few occasions. On each occasion the employer ignored or rebuffed further engagement. The employers have refused to recognize their association and have either refused to meet and bargain with it or have not responded to the demands of the respondents.

This history, scant as it is, does not establish that the AEPA violates s. 2(d). Indeed, the union has not made a significant attempt to make it work. As just discussed, properly interpreted, it does not violate s. 2(d). Moreover, the process has not been fully explored and tested. The AEPA, as Farley J. noted, contemplates a meaningful exercise of the right of association, and provides a tribunal for the resolution of disputes.

Farley J. expressed cautious hope that the Tribunal created by the Act would prove efficacious and that the relief claimed might be granted:

One would think it better to see how the Tribunal operates in fact before condemning it as powerless to deal with such abuses. . . . I am of the view that this condemnation is premature. A successful application would do one of several things: be effective positively as to action; or morally give the wrongdoing employer a “bloody nose”; or if truly an empty process, it would demonstrate the need for strengthening by legislative amendment. [para. 18]

Farley J. accordingly found that the complaint was premature:

. . . it would seem to me to be a premature and unfair complaint that the Tribunal charged under the AEPA with dealing with complaints — namely the Agriculture, Food and Rural Affairs Appeal Tribunal — is bereft of expertise in labour relations given its bipartite composition of labour and agricultural experienced personnel. That Tribunal should be given a fair opportunity to demonstrate its ability to appropriately handle the function given to it by the AEPA. [para. 28]

Section 11 of the AEPA specifically empowers the Tribunal to make a determination that there has been a contravention of the Act, and to grant an order or remedy with respect to that contravention. The Tribunal may be expected to interpret its powers, in accordance with its mandate, purposively, in an effective and meaningful way. Labour tribunals enjoy substantial latitude when applying their constituent statutes to the facts of a given case. As stated by the Ontario Labour Relations Board in Adams Mine, Cliffs of Canada Ltd. v. United Steelworkers of America (1982), 1 C.L.R.B.R. (N.S.) 384:

The Act does not spell out each and every right and obligation of labour and management. This Board is left with the task of applying the Act’s general language in the light of an infinite variety of circumstances which may arise. A rigid scheme of regulation is avoided and flexibility is provided although all within the limitations necessary to effectuate the dominant purpose of the Act. [pp. 399-400]

We conclude the AEPA does not breach s. 2(d) of the Charter . It is therefore unnecessary to consider the s. 1 arguments that the respondents’ demands for full LRA protections would be inappropriate because of the diverse nature of the agricultural sector, ranging from small family operations to larger commercial establishments.

As an alternative to their claim under s. 2(d), the respondents contend that Ontario has violated their equality rights under s. 15 of the Charter by excluding them from the statutory protections accorded to workers in other sectors. They argue that status as an agricultural worker is analogous to the enumerated grounds of discrimination in s. 15(1) because their occupation is a fundamental aspect of their identity.

Farley J., writing in 2006, found that the situation of farm workers had not changed appreciably since Dunmore, in 2001, where this Court wrote with sympathy of the vulnerable position of these workers and the need for greater labour protections. The Ontario legislature attempted to respond to the concerns expressed in Dunmore by enacting the AEPA.

The s. 15 discrimination claim, like the s. 2(d) claim, cannot succeed on the record before us. It is clear that the regime established by the AEPA does not provide all the protections that the LRA extends to many other workers. However, a formal legislative distinction does not establish discrimination under s. 15. What s. 15 contemplates is substantive discrimination that impacts on individuals stereotypically or in ways that reinforce existing prejudice and disadvantage: Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483, at para. 17. The AEPA provides a special labour regime for agricultural workers. However, on the record before us, it has not been established that the regime utilizes unfair stereotypes or perpetuates existing prejudice and disadvantage. Until the regime established by the AEPA is tested, it cannot be known whether it inappropriately disadvantages farm workers. The claim is premature.

The decision that we render today is another step in the resolution of the issues surrounding the organizational challenges faced by farm workers in Ontario. We hope that all concerned proceed on the basis that s. 2(d) of the Charter confirms a right to collective bargaining, defined as “a process of collective action to achieve workplace goals”, requiring engagement by both parties. Like all Charter rights, this right must be interpreted generously and purposively. The bottom line may be simply stated: Farm workers in Ontario are entitled to meaningful processes by which they can pursue workplace goals.

We would allow the appeal and dismiss the action. We would answer the constitutional questions as follows:

1. Does the Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16, infringe s. 2(d) of the Canadian Charter of Rights and Freedoms ?

No.

2. If so, is the infringement a reasonable limit prescribed by law as can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

3. Does s. 3(b.1) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A, infringe s. 2(d) of the Canadian Charter of Rights and Freedoms ?

No.

4. If so, is the infringement a reasonable limit prescribed by law as can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

5. Does the Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16, infringe s. 15 of the Canadian Charter of Rights and Freedoms ?

No.

6. If so, is the infringement a reasonable limit prescribed by law as can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

7. Does s. 3(b.1) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A, infringe s. 15 of the Canadian Charter of Rights and Freedoms ?

No.

8. If so, is the infringement a reasonable limit prescribed by law as can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

In the circumstances, we order no costs on this appeal or in the courts below.

The reasons of Charron and Rothstein JJ. were delivered by

Rothstein J. —

I. Introduction

I have had the opportunity to read the reasons of the Chief Justice and LeBel J. in this appeal. I agree with them that the appeal should be allowed and the action dismissed. However, I disagree with their interpretation of s. 2(d) of the Canadian Charter of Rights and Freedoms with respect to collective bargaining.

The Chief Justice and LeBel J. have accurately set out the background, the constitutional challenge and the judicial history in this case. While I agree with what they have written in these respects, I respectfully disagree with the reasons for their decision.

The reasons of the Chief Justice and LeBel J. are based upon the majority decision in Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391 (“Health Services”). The majority in Health Services found that s. 2(d) of the Charter confers constitutional status on collective bargaining. It concluded that collective bargaining as protected by s. 2(d) “requires both employer and employees to meet and to bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation” (para. 90). It further found that the requirement to bargain in good faith imposes a duty on employers to meet with employees and make a “reasonable effort to arrive at an acceptable contract” (para. 101).

Following the reasons in Health Services, the Chief Justice and LeBel J. in this case say that s. 2(d) protects a right to collective bargaining, which includes “a process of engagement that permits employee associations to make representations to employers, which employers must consider and discuss in good faith” (para. 2). According to them, there is no doubt that because of s. 2(d) employers “must engage in a process of meaningful discussion” because “the effect of denying these rights is to render the associational process effectively useless” (para. 54).

The term collective bargaining may have different meanings in other contexts, which I discuss in further detail below. For the sake of clarity, throughout these reasons, I will use the term to refer to the entitlements and obligations that the Chief Justice and LeBel J. view as being encompassed by s. 2(d), as quoted in paras. 121 and 122 above.

I respectfully disagree with the Chief Justice and LeBel J. that collective bargaining enjoys constitutional status under the s. 2(d) freedom of association. I do not agree that s. 2(d) requires the state to impose a complex set of statutorily defined reciprocal rights and duties on employers and workers associations, including a duty to bargain in good faith.

In my view, s. 2(d) protects the liberty of individuals to associate and engage in associational activities. Therefore, s. 2(d) protects the freedom of workers to form self-directed employee associations in an attempt to improve wages and working conditions. What s. 2(d) does not do, however, is impose duties on others, such as the duty to bargain in good faith on employers.

A constitutionally imposed duty to bargain in good faith strengthens the position of organized labour vis-à-vis employers. I express no opinion on the desirability of such an outcome for agricultural employees in Ontario. My point is only that courts are ill-suited to determine what is a matter of labour relations policy. Such policy decisions require a balancing of differing interests rather than an application of legal principles. Courts do not have the necessary expertise, or institutional capacity, to undertake a process which should involve consulting with and receiving representations from the various interested stakeholders and coming to an informed decision after balancing the necessary policy considerations. The decision to impose a duty of collective bargaining should be made by the legislature, and not by the court.

Since the majority reasons are an application of the findings in Health Services to the circumstances of this appeal, the initial question that is raised is whether Health Services was correctly decided. As I have already suggested, and as I will explain below, I would find that Health Services was not correctly decided, to the extent that it constitutionalizes collective bargaining.

In my opinion, overruling Health Services would dispose of the constitutional challenge in this case. The respondents’ (Fraser’s) argument that the Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16 (“AEPA”), violates the Charter because it does not protect a right to collective bargaining would have no basis. I therefore turn first to the question of whether it would be appropriate that Health Services be overruled.

The authorities are abundant that this Court may overrule its own decisions, and indeed it has done so on numerous occasions: see R. v. Bernard, [1988] 2 S.C.R. 833, at p. 849, per Dickson C.J. in dissent; R. v. Chaulk, [1990] 3 S.C.R. 1303, at p. 1353, per Lamer C.J. for the majority; R. v. B. (K.G.), [1993] 1 S.C.R. 740; R. v. Robinson, [1996] 1 S.C.R. 683.

However, in order to overrule its own precedent, the Court must be satisfied, based upon substantial reasons, that the precedent was wrongly decided. It is not appropriate simply because of a change in the composition of the Court that precedent should be overturned, because of the views of newly appointed judges. There must be compelling reasons to justify overruling: R. v. Salituro, [1991] 3 S.C.R. 654, at p. 665; Hamstra (Guardian ad litem of) v. British Columbia Rugby Union, [1997] 1 S.C.R. 1092, at paras. 18-19; R. v. Henry, 2005 SCC 76, [2005] 3 S.C.R. 609, at para. 44.

This Court’s most recent pronouncement on the question of overruling was in Henry. Writing for the Court, Justice Binnie first noted at para. 44 that the Court’s practice is against departing from precedent unless there are compelling reasons to do so. However, he also recognized that “while rare, departures [from precedent] do occur”. He further noted that constitutional decisions, including Charter decisions, are not immutable and may be overruled, though he held that “[t]he Court should be particularly careful before reversing a precedent where the effect is to diminish Charter protection.”

The values of certainty and consistency, which are served by adherence to precedent, are important to the orderly administration of justice in a system based upon the rule of law. Therefore, judges must proceed with caution when deciding to overrule a prior decision. The caution and care with which a judge must approach the question of overruling was articulated well by Gibbs J. of the High Court of Australia in Queensland v. Commonwealth (1977), 139 C.L.R. 585, at p. 599:

No Justice is entitled to ignore the decisions and reasoning of his predecessors, and to arrive at his own judgment as though the pages of the law reports were blank, or as though the authority of a decision did not survive beyond the rising of the Court. A Justice, unlike a legislator, cannot introduce a programme of reform which sets at nought decisions formerly made and principles formerly established. It is only after the most careful and respectful consideration of the earlier decision, and after giving due weight to all the circumstances, that a Justice may give effect to his own opinions in preference to an earlier decision of the Court.

What the courts are doing when deciding whether to overrule a precedent is a balancing exercise between two important values: correctness and certainty. A court must ask whether it is preferable to adhere to an incorrect precedent to maintain certainty or to correct the error. Indeed, because judicial discretion is being exercised, the courts have set down, and academics have suggested, a plethora of criteria for courts to consider in deciding between upholding precedent and correcting error.

In Bernard, Dickson C.J. in dissent, identified four reasons for overruling an earlier precedent, at pp. 850-61:

1. Decisions that predate the Charter and fail to reflect Charter values;

2. Subsequent developments in the law that undermine the validity of the precedent;

3. A prior decision that creates uncertainty contrary to the underlying values of clarity and certainty that lie behind stare decisis;

4. A prior decision that operates against the accused by expanding the scope of criminal liability beyond its normal limits.

These factors were subsequently adopted by the majority in Chaulk as a non-exhaustive list of considerations relevant to deciding whether to overrule an earlier precedent (p. 1353).

More recently, in Henry, Binnie J. identified the following reasons for overturning an earlier precedent, at paras. 45-46:

1. The prior decision departed from the purpose of a Charter provision as articulated in an earlier precedent;

2. Experience shows that the prior decision is unworkable as its application is unnecessarily complex and technical;

3. The prior decision is contrary to sound principle;

4. The prior decision results in unfairness.

The Supreme Court of the United States has also grappled with identifying the appropriate principles governing when courts should overrule precedent. In Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833 (1992), O’Connor, Kennedy and Souter JJ., writing for a majority of the court on this point, held that when the court considers whether to overrule an earlier case, “its judgment is customarily informed by a series of prudential and pragmatic considerations designed to test the consistency of overruling a prior decision with the ideal of the rule of law, and to gauge the respective costs of reaffirming and overruling a prior case” (p. 854). To this end, they articulated four factors to be considered in deciding whether to overrule precedent, at pp. 854-55:

1. Has the rule proved to be intolerable because it defies workability?

2. Is the rule subject to a reliance that would lend a special hardship to the consequences of overruling and add inequity to the cost of repudiation?

3. Have related principles of law developed as to have left the old rule no more than a remnant of abandoned doctrine?

4. Have facts so changed, or come to be seen so differently, as to have robbed the old rule of significant application or justification?

Professor Debra Parkes has summarized eight criteria suggested by Professor B. V. Harris as follows:

1. Can the precedent be distinguished?

2. Was it decided per incuriam?

3. Is the precedent unworkable?

4. Are new reasons advanced not considered in the earlier case?

5. Does the law now view the precedent to be wrong?

6. Do the values underlying error correction or doing justice outweigh the values of adherence to stare decisis?

7. Would error be swiftly corrected by the legislature in non-constitutional cases?

8. Are foundational principles of human and civil rights involved?

See D. Parkes, “Precedent Unbound? Contemporary Approaches to Precedent in Canada” (2006), 32 Man. L.J. 135, at p. 149, citing B. V. Harris, “Final Appellate Courts Overruling Their Own ‘Wrong’ Precedents: The Ongoing Search for Principle” (2002), 118 L.Q.R. 408.

If a precedent has overruled prior cases, two sets of precedents exist, an original precedent and a new precedent, although one has been overruled. In such cases it will be more important to carefully scrutinize the new precedent to determine if it has strayed from sound prior decisions and whether it would be preferable to return to the original, and more sound, decisions. In Adarand Constructors, Inc. v. Pena, 515 U.S. 200 (1995), O’Connor J. confronted this type of situation. She stated, at p. 231:

Remaining true to an “intrinsically sounder” doctrine established in prior cases better serves the values of stare decisis than would following a more recently decided case inconsistent with the decisions that came before it; the latter course would simply compound the recent error and would likely make the unjustified break from previously established doctrine complete.

Thus, where there exist earlier precedents from which the precedent at issue itself departed, it may be justifiable, based on the values underlying stare decisis, for the Court to return to the earlier precedents.

The criteria discussed in the above cases and articles may, depending on the circumstances of each case, be relevant in deciding whether overruling is appropriate. However, these criteria do not represent an exhaustive list of considerations or requirements. Rather, such criteria function as “guidelines to assist th[e] Court in exercising its discretion”: Chaulk, at p. 1353, per Lamer C.J. Fundamentally, the question in every case involves a balancing: Do the reasons in favour of following a precedent — such as certainty, consistency, predictability and institutional legitimacy — outweigh the need to overturn a precedent that is sufficiently wrong that it should not be upheld and perpetuated?

In the case of Health Services, I am of the opinion that the following considerations are relevant and justify overruling.

First, the error in Health Services concerns a question of constitutional law. Thus, not only does it go to one of the foundational principles of our legal system, but it is not susceptible to being corrected in a lasting way by the legislative branch. While s. 33 of the Charter may allow Parliament or the legislatures to suspend, temporarily, the force of this Court’s ruling, history over the last two decades demonstrates that resort to s. 33 by legislatures has been exceedingly rare. Health Services will, if left to stand, set out abiding principles of constitutional law. Only the Court may correct this error in fundamental principle. As noted in Planned Parenthood, it is “common wisdom that the rule of stare decisis is not an ‘inexorable command,’ and certainly it is not such in every constitutional case” (p. 854). The jurisprudence of this Court contains similar observations. Because the Charter involves the most fundamental principles underlying our law, it is particularly important that its provisions be correctly interpreted.

The Chief Justice and LeBel J. say that the constitutional nature of Health Services should only be a final consideration with respect to overruling difficult cases (para. 58). In my respectful view, and as my reasons will endeavour to demonstrate, there are no shortage of reasons to believe that Health Services is problematic on other grounds.

Relying on Henry, my colleagues also warn that this Court should be wary of overruling Health Services because doing so might have the potential to diminish Charter protection (para. 58, citing Henry, at para. 44). They say that this consideration “militate[s] in favour of upholding” Health Services (para. 58). However, the Court cannot be oblivious to errors in prior decisions. When considering overruling, the Court must balance correctness and certainty. If there is a potential diminishment arising from correcting prior error, that is a reason to be cautious, not a reason to forego correcting prior error altogether. Arguably, as Health Services itself strayed from prior precedent, returning to those prior precedents would promote certainty. However, even if certainty would favour retaining Health Services, in this case the need for a constitutionally correct answer is paramount.

Second, as I have indicated, Health Services strayed significantly from earlier sound precedents with respect to the purpose of Charter protection for freedom of association. The constitutional guarantee of freedom of association is premised on the recognition that individuals may be better able to secure their interests and achieve their goals if they may join with others in their attempt to do so. From this, two propositions necessarily follow: (a) that s. 2(d) was intended to secure the individual’s freedom to coordinate his or her actions with others and enjoy the benefits that flow naturally from that coordination; and (b) that s. 2(d) was not intended to promote or guarantee the outcomes for which the association was formed. The ruling in Health Services contradicts both of these central tenets. By constitutionalizing collective bargaining, Health Services created a group right that vests in the employee association rather than individual workers, and confers substantive outcomes for which the association was formed. It has therefore moved away from the sound principles established by earlier precedents of this Court.

Third, the constitutionalization of collective bargaining, as envisaged in Health Services and by the Chief Justice and LeBel J. in this case, is unworkable. On a practical level, the right to collective bargaining asserted by my colleagues is not workable without other elements of modern labour legislation in place. As Winkler C.J.O. recognized at the Court of Appeal, if it is to be effective, the right to collective bargaining will be hard pressed to perform its function without reinforcement from the other aspects of labour legislation that he identified (2008 ONCA 760, 92 O.R. (3d) 481). As a matter of principle, the distinction between substance and process on which the ruling in Health Services (and of the Chief Justice and LeBel J. in this case) is premised cannot be sustained. The process is itself a desirable outcome for the association of workers, and will result in substantive concessions by the employer. Thus, both principle and practicality militate against sustaining the ruling in Health Services.

Fourth, there has been intense academic criticism of Health Services, including by Professor Eric Tucker, who was himself cited by the majority in support of its decision. See B. Etherington, “The B.C. Health Services and Support Decision — The Constitutionalization of a Right to Bargain Collectively in Canada: Where Did It Come From and Where Will It Lead?” (2009), 30 Comp. Lab. L. & Pol’y J. 715, at pp. 734-39. I recognize that there is also academic commentary agreeing with the results of Health Services, as pointed out by the Chief Justice and LeBel J. (at para. 88). The existence of such commentary is not unexpected given the highly contentious and polarizing nature of labour relations. However, as I will explain, while some agree with the result, the academic criticism of concern here targets the reasoning of the majority in Health Services.

Even some of the authors who support the results of Health Services and who are cited by my colleagues are critical of aspects of the reasoning employed by the majority of the Court to achieve those results. For example, Professor Jamie Cameron, in “Due Process, Collective Bargaining, and s. 2 (d) of the Charter : A Comment on B.C. Health Services” (2006-2007), 13 C.L.E.L.J. 233, while supportive of the results of Health Services, criticized the decision because “B.C. Health Services relied on a concept of entitlement that was so heavily and exclusively contextualized to collective bargaining that the decision lost contact with the underlying values which have anchored s. 2(d) since the Labour Trilogy” (p. 262; see also pp. 240 and 259).

I reiterate that in light of such academic criticism, it is appropriate for this Court to take notice and acknowledge the errors that have been identified.

My colleagues say that it is “procedurally inappropriate” to overrule Health Services because none of the parties have expressly asked this Court to do so (para. 59). However, the substance of the arguments of the appellants and interveners supporting them are in effect a claim that Health Services should be overruled. The appellants and interveners supporting them all say that in this case there should be no obligation on agricultural employers to engage in compulsory collective bargaining. In my opinion, it is not possible to agree that there is no such obligation without overruling Health Services.

Further, as Deschamps J. observes, “an employer’s duty to bargain in good faith was not even raised” in Health Services (at para. 297), and while the parties in Health Services “recognized that under most Canadian labour law statutes, employers had an obligation to bargain in good faith, the claimants were not seeking a declaration characterizing this obligation as a constitutional one” (para. 304). Despite this issue not having been raised in Health Services, the majority in that case did not find it “procedurally inappropriate” to find that “collective bargaining imposes corresponding duties on the employer” and “requires both employer and employees to meet and to bargain in good faith ” (para. 90).

Lastly, while the foregoing factors all support overruling Health Services, as the Chief Justice and LeBel J. point out, it is fundamental that it be demonstrated that Health Services was decided in error. The balance of these reasons endeavours to explain why Health Services was erroneously decided.

Prior to explaining why Health Services erred in finding that s. 2(d) of the Charter protects collective bargaining, I will briefly refer to Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, which was consistent with this Court’s jurisprudence prior to Health Services on the contours of s. 2(d) in the labour law context. I do this to underline the explicit break from that jurisprudence in Health Services. This break came when a majority of this Court found that s. 2(d) required that government legislate to facilitate collective goals which an association was formed to pursue, rather than protecting the freedom of association itself. This finding went beyond the rule in Dunmore which mandated legislative protection only where such protection was necessary to the freedom to associate.

The Chief Justice and LeBel J. present an alternative interpretation which suggests that Health Services “follows directly from the principles enunciated in Dunmore” (para. 38). With respect, I do not agree with this interpretation because it does not follow from the words and findings in Dunmore. In Dunmore, the requirement that government provide legislation to protect workers was anchored in the proposition that certain workers could not associate without government intervention. This concept was embodied by the idea that the lack of legislation was a “substantial interference” to the ability to form an association (para. 25 (emphasis deleted)). Deschamps J. describes Dunmore as holding that agricultural workers “were substantially unable to exercise their constitutional right [to associate] without the support of a legislative framework” (para. 307). This characterization is in line with the words of Bastarache J. who himself noted the distinction between forming an association and enhancing and facilitating the goals of that association:

. . . a group that proves capable of associating despite its exclusion from a protective regime will be unable to meet the evidentiary burden required of a Charter claim. In such a case, inclusion in a statutory regime cannot be said to safeguard, but rather to enhance, the exercise of a fundamental freedom. [para. 39]

Health Services, in contrast, was not focussed on ensuring that government did not interfere with the right of individuals to form an association. Indeed the action was brought by existing associations. Instead, the decision in Health Services centred on the purported need to constitutionalize collective bargaining in order for the association to be “meaningful”. The majority concluded that without a legislated right to collective bargaining, and without constitutional protection of terms of the collective agreement in that case, the formation of an association was meaningless.

This sentiment was an express break from Dunmore because the majority in Health Services focussed on the goals of an association and the enhancement of those goals, rather than the ability of the claimants to associate (which they already had done). Constitutionalizing collective bargaining therefore cannot be said to safeguard the ability to associate, but instead is concerned with “enhanc[ing] the exercise of a fundamental freedom”, which cannot form the basis of a Charter claim (Dunmore, at para. 39). This express break from Dunmore is found in the reasons of the Chief Justice and LeBel J., where they suggest that providing associational protections in the AEPA without also enshrining good faith bargaining would “render the associational process effectively useless” (para. 54). This break from Dunmore is also recognized in the reasons of my colleague Abella J., where she observes that Health Services resulted in “creating a completely different jurisprudential universe” from that found in the previous “Dunmore ‘right to organize’ template” (paras. 324-25).

It is this shift from protecting what is necessary to exercise the freedom to associate, to constitutionalizing the goals of an association — that is, negotiating a collective agreement — which results in Health Services being inconsistent with the ruling in Dunmore. The majority in Health Services found that the freedom of association is meaningless unless the government also imposes a duty on employers to bargain in good faith, and protects the fruits of that bargaining process. The focus shifted in Health Services from protecting the right to associate to enhancing the goals of the association. This was an express break from Dunmore.

This Court first examined the scope of the Charter ’s guarantee of freedom of association in a series of three cases that came to be known as the “labour law trilogy” (“Trilogy”): Reference re Public Service Employee Relations Act (Alta.), [1987] 1 S.C.R. 313 (the “Alberta Reference”); PSAC v. Canada, [1987] 1 S.C.R. 424; and RWDSU v. Saskatchewan, [1987] 1 S.C.R. 460. The specific issue in those cases was whether s. 2(d) protected the right to strike. However, the members of the Court took the opportunity to consider, at length, the purpose and scope of the protections afforded by s. 2(d). The bulk of the substantive analysis is found in the Alberta Reference.

While the Court was divided on the result in those cases, there was nonetheless agreement on a number of core principles concerning the purpose and scope of s. 2(d). With respect to the purpose of granting constitutional protection to freedom of association, the following comment by McIntyre J. reflects a general consensus amongst the members of the Court:

While freedom of association like most other fundamental rights has no single purpose or value, at its core rests a rather simple proposition: the attainment of individual goals, through the exercise of individual rights, is generally impossible without the aid and cooperation of others. “Man, as Aristotle observed, is a ‘social animal, formed by nature for living with others’, associating with his fellows both to satisfy his desire for social intercourse and to realize common purposes.” [p. 395]

In addition, the Court was agreed that s. 2(d) protects an individual right to enter into an association and does not create group rights that vest in the association itself. In the words of Dickson C.J. (dissenting):

What freedom of association seeks to protect is not associational activities qua particular activities, but the freedom of individuals to interact with, support, and be supported by, their fellow humans in the varied activities in which they choose to engage. But this is not an unlimited constitutional license for all group activity. The mere fact that an activity is capable of being carried out by several people together, as well as individually, does not mean that the activity acquires constitutional protection from legislative prohibition or regulation. [p. 366]

McIntyre J. struck a similar note:

In considering the constitutional position of freedom of association, it must be recognized that while it advances many group interests and, of course, cannot be exercised alone, it is nonetheless a freedom belonging to the individual and not to the group formed through its exercise. [p. 397]

In Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), [1990] 2 S.C.R. 367 (“PIPSC”), the core areas of agreement in the Alberta Reference were later summarized by Sopinka J. in a passage that has since been frequently cited:

Upon considering the various judgments in the Alberta Reference, I have come to the view that four separate propositions concerning the coverage of the s. 2(d) guarantee of freedom of association emerge from the case: first, that s. 2(d) protects the freedom to establish, belong to and maintain an association; second, that s. 2(d) does not protect an activity solely on the ground that the activity is a foundational or essential purpose of an association; third, that s. 2(d) protects the exercise in association of the constitutional rights and freedoms of individuals; and fourth, that s. 2(d) protects the exercise in association of the lawful rights of individuals. [pp. 401-2]

These four propositions were later endorsed by majorities of the Court in Canadian Egg Marketing Agency v. Richardson, [1998] 3 S.C.R. 157, and Delisle v. Canada (Deputy Attorney General), [1999] 2 S.C.R. 989. In the latter case, a majority of the Court ruled that s. 2(d) does not entitle workers to any particular set of statutory protections for their associative activities, such as those providing for a right to collective bargaining.

The proposition that s. 2(d) does not confer a right to collective bargaining was once again endorsed by a majority of the Court in Dunmore. However, in that case the Court recognized that Sopinka J.’s four propositions, while valid, might not exhaust the entire scope of protection afforded by s. 2(d) (para. 16).

Bastarache J. explained that the core of s. 2(d) protection is to prohibit the state from interfering with an activity because of its associational nature. Relying on comments made by Dickson C.J. in the Alberta Reference, he noted that Sopinka J.’s fourth proposition, which protects “the exercise in association of the lawful rights of individuals”, suffers from a potential weakness. That proposition will, in general, serve as a useful test in determining whether the state has targeted the associational aspect of an activity rather than the activity itself. If the state has outlawed an activity at both the individual and the group level, it is likely that this is because the activity itself is deemed to be harmful or problematic. By contrast, if the state permits an individual to engage in an activity but has outlawed the performance of that activity in concert with others, this will generally indicate that the state has targeted the activity solely because of its associational nature. Thus, Sopinka J.’s fourth proposition assists in isolating the true intent or effect of a measure when determining whether it infringes a person’s freedom of association (paras. 16-18).

However, there may be cases in which the state has directly targeted the associational aspects of an activity which are not captured by Sopinka J.’s fourth proposition. An activity performed on a group level may be “qualitatively” different from what an individual can undertake in isolation, such that no direct comparison or analogy is possible. If one were to interpret the fourth proposition as requiring a strict analogy between the collective activity and its individual counterpart, a state restriction on such “qualitatively” different activities would not be considered to inhibit freedom of association and would therefore pass muster under s. 2(d). Recognizing this, Dunmore attenuated the requirement of an individual analogue. If it can be demonstrated that a restriction on a group activity is an attack on the associational nature of the activity, a s. 2(d) claim may yet succeed even if no direct analogy can be made between the group activity and a lawful individual counterpart.

This observation about how “qualitatively” different activities exist was aimed at explaining why certain activities which did not have an individual analogue must be protected in order to protect the freedom to form an association. The disposition in Dunmore turned on whether the lack of protection for agricultural employees was a “substantial interference” to the ability of workers to form an association (paras. 22-23). To suggest, in hindsight, that “[a]fter Dunmore, there could be no doubt that [s. 2(d)] extends to realization of collective, as distinct from individual, goals” as the Chief Justice and LeBel J. say at para. 32 (emphasis added), is to substantially overstate the holding in Dunmore.

While the basic framework set down in the Trilogy had stood for some 20 years, the Court decided to break with this line of precedent in Health Services. In that case, the majority held that s. 2(d) protects a right to collective bargaining and imposes a duty on employers to bargain in good faith. Such a right could not be accommodated within the framework set down by the Trilogy and followed in subsequent cases, and so the majority opted to overturn that line of precedents altogether.

The overarching reason advanced for rejecting the Trilogy’s interpretation of s. 2(d) was that it reflected a “decontextualized” rather than “purposive” approach to Charter interpretation: Health Services, at para. 30. The majority in Health Services found that insufficient attention had been paid to the close connection between freedom of association and labour relations (specifically trade unionism), and that the intimate ties between the two implied that s. 2(d) should be held to protect a broader set of entitlements than the Trilogy’s approach could support. Under this view, the main problem with the Trilogy’s approach was that it did not extend a right to “collective bargaining”, the protection of which was, in the majority’s view, a central purpose of freedom of association (para. 86).

In extending constitutional protection to collective bargaining, the majority in Health Services viewed this constitutional right as including an obligation on parties to bargain in good faith. The majority in Health Services described the protection afforded under s. 2(d) as follows, at para. 90:

. . . the state must not substantially interfere with the ability of a union to exert meaningful influence over working conditions through a process of collective bargaining conducted in accordance with the duty to bargain in good faith. Thus the employees’ right to collective bargaining imposes corresponding duties on the employer. It requires both employer and employees to meet and to bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation.

In fact, the majority went so far as to say that “the duty to consult and negotiate in good faith” is “the fundamental precept of collective bargaining” (para. 97).

The decision in Health Services purported to impose two limitations on this right. First, the right was said not to cover all aspects of “collective bargaining”, as that term is understood in the statutory labour relations regimes based upon the Wagner model that are in place across the country. The Wagner model refers to Canadian variants of the National Labor Relations Act, 49 Stat. 449 (1935) (the “Wagner Act”), which was enacted into law in the United States during the Depression. By the end of the 1930s, most Canadian provinces had passed legislation incorporating the main objectives of the Wagner Act. The Wagner model has four legislative hallmarks: explicit recognition of the right of employees to belong to a trade union of their choice; protections against employer coercion or interference with organizing activities, known as unfair labour practices provisions; a duty upon employers to bargain in good faith with their employees’ unions; and a dispute resolution mechanism for resolving impasses: see G. W. Adams, Canadian Labour Law (2nd ed. (loose-leaf)), vol. 1, at p. 1-11.

Second, it was said not to be aimed at securing a particular outcome in a labour dispute, or guaranteeing access to any particular statutory scheme: see Health Services, at para. 19. Nonetheless, the majority held that the process of good faith collective bargaining demands that unions and employers engage with each other and “make a reasonable effort to arrive at an acceptable contract”: Health Services, at para. 101.

The Chief Justice and LeBel J. say that Health Services “follows directly” from the finding in Dunmore, because the government action in that case “rendered the meaningful pursuit of [workplace goals and collective bargaining activities] impossible and effectively nullified the right to associate of its employees” (para. 38). However, as I have discussed above, this conflates two arguments. The first is that restrictions on the ability to associate, either directly or because the government interfered with an activity because of its “associational nature”, are unconstitutional, which was the finding in Dunmore. The second is that governments are required to provide legislation which enhances the ability of an existing association to pursue its goal of negotiating a collective agreement, which was the finding in Health Services, but was contrary to Dunmore. As noted above, an application of the actual holding in Dunmore would have asked only if the government substantially interfered with the ability to associate.

I now turn to the fundamental question in this case: namely, whether Health Services was wrong to constitutionalize collective bargaining. In my respectful view, Health Services was indeed wrong. The problems relating to this aspect of Health Services can be grouped into three categories.

First, the collective bargaining right recognized in Health Services is inconsistent with the purpose of s. 2(d). As I will explain, the interpretation of s. 2(d) adopted by the majority in Health Services is unsound in principle for a number of reasons, and the correct conceptual framework for s. 2(d) is that established by the Trilogy and applied in subsequent cases. Section 2(d) does not protect a right to collective bargaining.

Second, the reasons advanced in Health Services for protecting collective bargaining under s. 2(d) do not support that conclusion. The majority in Health Services found that the history of Canadian labour law, international law, and Charter values all pointed in favour of extending the guarantee of freedom of association to include collective bargaining. While all of those factors support constitutional recognition of the freedom of workers to associate, they do not support the right to collective bargaining.

Third, the approach to collective bargaining in particular, and s. 2(d) in general, articulated in Health Services raises significant problems relating to workability. In my view, the framework established in Health Services is both inherently unstable and is a vehicle for the imposition of judicial policy preferences.

I now address each of these problems in turn.

There are five reasons why the collective bargaining right recognized by Health Services is inconsistent with the purpose of s. 2(d). First, the analysis in Health Services improperly assigned collective dimensions to an individual right. Second, Health Services assigned positive obligations to the essentially negative freedom of association. Third, the reasons in Health Services replaced a content-neutral approach to s. 2(d) freedom and adopted an approach to s. 2(d) which privileges certain associations over others. Fourth, Health Services elevated contracts — collective bargaining agreements — above statutes and disrupted the ordinary hierarchy of laws. Fifth, the analysis in Health Services departs from a long-standing principle of judicial deference in the field of labour relations.

(1) Section 2(d) Protects Individual Interests, Not Group Interests

(a) Individual Freedoms Versus Collective Rights

First, Health Services reinterpreted an individual freedom as giving rise to collective rights with no individual rights foundation. This reinterpretation of the scope of s. 2(d) was a departure from previous jurisprudence that is not justified by the purpose of the Charter guarantee. The series of cases beginning with the Alberta Reference established that the freedom of association is an individual freedom which is intended to prevent the government from interfering with associations by treating groups differently than it treats individuals.

Health Services expanded s. 2(d) to protect collective rights which meant that individuals who are members of specific groups now enjoy greater constitutional rights than those who are not. In particular, following Health Services, workers’ associations enjoy a robust right to bargain because employers are constitutionally obligated to bargain with their association. Individuals who are not members of an association, on the other hand, have no constitutional right to oblige their employers to bargain.

While the Charter may protect certain collective rights, freedom of association does not fall into that category. Where a particular Charter guarantee extends greater rights to a group than to an individual, that effect is made clear in the text of the particular guarantee. For example, the right to minority language education in ss. 23(1) and 23(2) is subject to there being a sufficient number of eligible children to warrant public expenditures on minority language education (s. 23(3)). In this way, the guarantee in s. 23 is predicated on the existence of a group. Minority language educational rights thus have a “unique collective aspect even though the rights are granted to individuals” (Doucet-Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3, at para. 28, per Iacobucci and Arbour JJ.) Similarly, s. 35(1) of the Constitution Act, 1982 recognizes the “existing aboriginal and treaty rights of the aboriginal peoples”. Treaty rights, being rights established in a treaty between a group of aboriginals and the Crown, undoubtedly have a collective dimension to them insofar as they vest rights in a particular group. See, e.g., R. v. Powley, 2003 SCC 43, [2003] 2 S.C.R. 207, at paras. 10-11. By contrast, s. 2(d) provides that “[e]veryone has . . . freedom of association”. This language supports an interpretation of s. 2(d) as an individual freedom. See P. W. Hogg, Constitutional Law of Canada (5th ed. Supp.), vol. 2, at pp. 37-1 and 37-2.

(b) Qualitative Differences Between Individuals and Associations Do Not Change the Individual Nature of Freedom of Association

The Chief Justice and LeBel J. say that Health Services was “consistent with the previous cases on the issue of individual and collective rights” (para. 65). I must respectfully disagree. McIntyre J. in the Alberta Reference made the point, which is the point that I make in my reasons, that freedom of association protects the right of groups to engage in activities that are lawful or constitutionally protected for individuals. The reliance on a purposive approach to interpretation advanced by the majority in Health Services opens the door to s. 2(d) being interpreted as a generalized group right as opposed to an individual right, which is an interpretation that is not consistent with this Court’s prior jurisprudence. My colleagues’ reference to the interpretation of s. 2(d) advanced by Dickson C.J. in the Alberta Reference, was not a position accepted by the majority and, as my reasons point out, by constitutionalizing the right to bargain collectively Health Services departed from the position of the majority in that case.

As I will now explain, although in Dunmore, s. 2(d) was found to protect group activity that was “qualitatively” different from individual activity, s. 2(d) cannot be used to give groups greater constitutional protection than individuals.

The majority in Health Services, in rejecting the view that s. 2(d) only protects those activities that may be lawfully pursued on the individual level, said that Dunmore had overtaken the notion that freedom of association applies only to activities capable of performance by individuals. Dunmore cannot be interpreted in such a manner. In Dunmore, Bastarache J. explained that to limit s. 2(d) to activities that are performable by individuals might render futile certain fundamental initiatives, since some collective activities may, by their very nature, be incapable of being performed by an individual. As I explained above, in the view of Bastarache J., certain activities are, when performed by a group, “qualitatively” different from those activities performed solely by an individual, and this qualitative difference may merit constitutional protection for the collective activity (Dunmore, at para. 17). With respect, Dunmore does not stand for the proposition that such qualitative differences open the door to the notion of group rights entirely unconnected to individual rights.

I accept that there may be qualitative differences between individuals acting alone and individuals acting in concert. Professor Langille refers to the example of choir singing. See B. Langille, “The Freedom of Association Mess: How We Got into It and How We Can Get out of It” (2009), 54 McGill L.J. 177 (“Freedom of Association”), at p. 185. While he ultimately believes that the choir metaphor should not apply to determine the scope of s. 2(d) rights, in my opinion the metaphor is apt in explaining the limited type of qualitatively different group activities that may be protected by s. 2(d). A group of individuals singing together in a choir can produce musical effects, such as harmonies and counterpoint, that an individual singing alone cannot. Such effects are not just an accumulation of individual voices singing in exactly the same way. Rather, they produce a musical effect that is fundamentally different from what a lone individual can produce. In other words, the whole is not merely the sum of its parts. The ability to create such effects through the coordination of individual action is one reason why freedom of association is protected. I believe that Dunmore found that s. 2(d) may protect voluntary collective activity which is “qualitatively” different from individual activity in this limited way.

In my view, the question is not whether the activity is susceptible of being performed, in exactly the same manner, by an individual acting alone. A choir singing harmony may produce sound that is qualitatively distinct from an individual voice, but it is nonetheless produced by a group of individuals voluntarily singing together. One individual is free to sing in one octave, another is free to sing in another octave. A harmony may result if they choose to perform these individual activities in concert with one another. And if the state were to outlaw harmony, it would be attacking the individuals’ ability to do side by side at the same time what each may do apart. Thus, while harmony arising out of choir singing may have no direct individual analogue, a legislative attack on harmony would be an attack on the association itself, and should not be permitted under s. 2(d). Section 2(d) may recognize qualitative differences between individual and group activity, without altering the individual nature of freedom of association.

While s. 2(d) protects “qualitatively” different activities which are an emergent result of free individual organization, it does not impose a constitutional obligation on government to create out of whole cloth a set of qualitative differences that make the group more powerful than it otherwise would be, even if such differences would enhance its ability to achieve its goals.

Health Services did not purport to give constitutional force to a right to individuals to compel employers to bargain in good faith with them. Individuals who are not members of an association have no such right. Thus, the constitutional right to compel employers to bargain in good faith with associations must be a unique group right, not an activity emergent from an individual right.

(2) Section 2(d) Protects Freedoms Rather Than Rights

Second, the majority concluded in Health Services that s. 2(d) imposes a duty to bargain in good faith. It explained this conclusion at para. 90 finding that “the employees’ right to collective bargaining imposes corresponding duties on the employer” requiring “both employer and employees to meet and to bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation”.

Thus, Health Services stands for the proposition that the right to collective bargaining includes an entitlement to have “meaningful” influence over working conditions. In other words, the right includes an assurance that a real dialogue will take place between an employee association and the employer. Such assurance can only be provided if the employer is under a duty to engage with representations made by the association and make a good faith attempt to bargain to pursue the goal of accommodation. In the absence of such a duty, the employer would be free to refuse to negotiate with the employee association. Thus, according to Health Services, if s. 2(d) protected only the ability of workers to make collective representations and did not impose a duty on the employer to bargain in good faith, it would fail to protect the right to collective bargaining.

In my view, this proposition suffers from an important defect: it transforms s. 2(d) from a freedom into a “positive” right by imposing an obligation to act on third parties (i.e. the employer). There is a difference between saying that an individual has the freedom or liberty to do something, on the one hand, and saying that he or she has the right to do it, on the other. A person is free or at liberty when there is an absence of obstacles or impediments. Nothing further is required beyond this absence. However, to say that a person has a “right” is to imply something further. It suggests that a claim can be made on someone else in order to be provided with the object of the right.

It is possible to describe a freedom in terms of rights, but this may only serve to blur the distinction between what is being protected in either case. A freedom exists to protect a sphere of autonomy, an area within which the individual will encounter no obstacles. A right, on the other hand, exists to provide an individual with a claim to some specific thing. Generally, a freedom can be described as a right only if it is recognized that the right is “negative” in character, that is, only if it is described as an entitlement to be free of restriction or prohibition.

An example of how a freedom can sometimes be described using the word “right” can be found by examining ss. 7 to 12 of the Charter . When the Charter uses the term “right”, as it does in ss. 7 to 12 , either a positive entitlement is introduced, or a right to be free of some restriction or prohibition (i.e. a freedom) is introduced. For the positive rights, an individual is given a right to some form of state action, e.g. to be advised of a right to counsel upon arrest. For the negative rights, the individual is given a right to be free from some form of restriction or prohibition, e.g. a right not to be arbitrarily detained or imprisoned. As discussed above, the right to be free of a restriction or a prohibition is a description that encompasses the “negative” character of the right, and is simply another way to describe a freedom.

The Chief Justice and LeBel J. suggest that my reasons seek to maintain the “consistently rejected . . . rigid distinction” between freedoms and positive rights (para. 69). Referring to Ontario (Public Safety and Security) v. Criminal Lawyers’ Association, 2010 SCC 23, [2010] 1 S.C.R. 815 (“CLA”), by way of analogy, they say that in some circumstances a Charter freedom, such as freedom of expression under s. 2(b), “may require the government to disclose documents to the public in order to enable meaningful discourse” (para. 69). Therefore, they say s. 2(d) of the Charter “may require the state to act positively to protect the ability of individuals to engage in fundamentally important collective activities” (para. 70). While I would agree that in “exceptional circumstances” (Dunmore, at para. 21; Delisle, at para. 33) a Charter freedom may require positive state action, I do not agree that CLA is an apt analogy.

Creating positive rights in exceptional circumstances does not dilute the coherence of the distinction between freedoms and rights. This is because those positive rights will only be granted when they are genuinely derivative of a freedom. A derivative right is one that is necessary to allow individuals to exercise a fundamental freedom. In CLA, the right to access government information was considered to be “a derivative right which may arise where it is a necessary precondition of meaningful expression on the functioning of government” (para. 30 (emphasis added)). This Court has found that a derivative right must be “inextricably tied to” (Canadian Broadcasting Corp. v. New Brunswick (Attorney General), [1996] 3 S.C.R. 480 (“CBC”), at para. 23) a “necessary precondition” (Dunmore, at para. 42) and “clearly within the ambit of the freedom” (CBC, at para. 23). However, the core of any derivative right is that without that right individuals will not be able to exercise their Charter freedom.

By its very nature, a derivative right must be necessary to the exercise of a freedom, not constitute a stand-alone right itself. Therefore the ability to exercise the freedom itself must be looked to in order to determine whether any proposed right is genuinely derivative. For example, in R. v. National Post, 2010 SCC 16, [2010] 1 S.C.R. 477, news organizations were arguing that they should have s. 2(b) protection for confidential sources. In rejecting this argument, this Court said that this position was “built on the premise that protection of confidential sources should be treated as if it were an enumerated Charter right or freedom” but that “is not so” (National Post, at para. 38). In CBC a positive right of access to courts was sought in order to allow journalists to report on court proceedings. La Forest J., writing for the Court, stated that “courts . . . must be open to public scrutiny and to public criticism of their operations” (CBC, at para. 20). Without the derivative right of access to the courts, it would be impossible for the public (and journalists) to offer criticism in the exercise of their s. 2(b) freedom of the press.

In Dunmore, a right was derived from s. 2(d) which required positive governmental action that was “necessary for [employees] to exercise their constitutional freedom to form and maintain associations” (para. 67 (emphasis added)).

A right to collective bargaining is not a derivative right in the sense in which it has been recognized in CLA, CBC and Dunmore. As discussed above, the essence of freedom of association is that it enables individuals to do in association what they could do as individuals. Dunmore is consistent with this principle, in the manner that I have just described. By contrast, collective bargaining is not emergent from an activity that an individual alone could do. Rather, as described in Health Services, it imposes a duty on employers to meet with employees and make a “reasonable effort to arrive at an acceptable contract” (para. 101). No individual employee has a right to require an employer to meet and make a reasonable effort to arrive at an acceptable employment contract, which is the right for an employee association created by Health Services.

The Chief Justice and LeBel J. say that collective bargaining is a derivative right because it is a “necessary precondition” to make their choice to associate meaningful. Understood this way, compulsory collective bargaining does not enable association. Rather, it is entirely concerned with enhancing the ability of employee associations, once formed, to pursue their goals and provide them with a more favourable bargaining position.

There is no reason to think that individual employees would not have the exact same desire for a more favourable bargaining position. Every day in Canada there are individuals who enter into employment contracts with their employers. Those individuals might also benefit if their employers were compelled to negotiate their employment contracts with them. However, individuals do not have a constitutionally mandated right to compel their employers to negotiate over an employment contract with them simply because no such right exists in the Charter . Because no individual has the right to compel an employer to negotiate in good faith over an employment contract, collective bargaining is not emergent from an activity that an individual alone could do.

Accordingly, a right to collective bargaining is not derivative of a freedom as described in CLA, CBC and Dunmore. It is a stand-alone right created by the Court, not by the Charter .

Thus, to grant a right to collective bargaining under s. 2(d) purportedly as derivative of the freedom of association would not be consistent with the approach taken by this Court in its derivative rights jurisprudence in relation to the Charter . Such a derivative right will arise only where it is a “necessary precondition” to the exercise of the freedom (CLA, at para. 30). This careful approach is necessary to adhere to the distinction between Charter rights and freedoms and prevents transforming freedoms into rights.

Viewed in this light, it is clear that s. 2(d) is intended to protect a sphere of individual autonomy or liberty, and not to enhance by state action the capacity of individuals to do a particular activity more effectively or to guarantee that any particular endeavour for which association might take place will succeed.

(3) Section 2(d) Does Not Privilege Some Associations Over Others

A third error in the approach to s. 2(d) in Health Services is that it conceives of s. 2(d) as privileging some associations over others. I cannot agree with an approach to s. 2(d) which requires this Court to decide which associations and associational objectives are worthy of constitutional protection and which are not.

Health Services rejected the grounds advanced in earlier decisions for excluding collective bargaining from the Charter ’s guarantee of freedom of association on the basis that they did not “withstand principled scrutiny” (para. 22). The majority in Health Services held that the “overarching” problem with these earlier decisions, particularly the Alberta Reference and PIPSC, was that they had pursued a “decontextualized”, as opposed to a “purposive”, approach to freedom of association that had ignored the differences between organizations and treated all organizations in the same way: “Whatever the organization — be it trade union or book club — its freedoms were treated as identical” (para. 30). Having decided that the objects of trade unions were meritorious of protection, the majority decided that the protection of trade unions’ objects required the recognition of the duty to bargain in good faith.

These earlier cases did indeed exhibit a content-neutral approach to freedom of association in the sense that they did not claim to privilege particular associations. Health Services erred in saying that these approaches were not purposive.

The purposive approach to Charter interpretation was explained by Dickson J. (as he then was) in R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295 (“Big M”), at p. 344:

The meaning of a right or freedom guaranteed by the Charter [is] to be ascertained by an analysis of the purpose of such a guarantee; it [is] to be understood, in other words, in the light of the interests it [is] meant to protect.

In my view this analysis is to be undertaken, and the purpose of the right or freedom in question is to be sought by reference to the character and the larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concepts enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter . The interpretation should be, as the judgment in [Hunter v. Southam Inc., [1984] 2 S.C.R. 145] emphasizes, a generous rather than a legalistic one, aimed at fulfilling the purpose of the guarantee and securing for individuals the full benefit of the Charter ’s protection. At the same time it is important not to overshoot the actual purpose of the right or freedom in question, but to recall that the Charter was not enacted in a vacuum, and must therefore, as this Court's decision in Law Society of Upper Canada v. Skapinker, [1984] 1 S.C.R. 357, illustrates, be placed in its proper linguistic, philosophic and historical contexts. [Emphasis in original.]

The kind of “context” to which the reasons in Health Services refer is very different from that being discussed here. Health Services suggested that a “generic” approach to defining freedom of association is inappropriate because different groups must have different freedoms: the needs of a book club are not the same as those of a trade union, and assuming them to be entitled to precisely the same thing under s. 2(d) would be a mistake (para. 30). However, the “context” that is relevant to a purposive interpretation of Charter freedoms is not the context of the individuals who happen to be exercising that freedom in a given case. Rather, a purposive interpretation of s. 2(d) requires that one place freedom of association in its linguistic, philosophic and historical contexts. The origins of the concept, the words used to describe it, and the philosophical principles on which it relies will define the scope of s. 2(d) protection. The extent of that protection should not change depending on the particular factual context or circumstances in which s. 2(d) is being applied.

In the Alberta Reference, both Dickson C.J. and McIntyre J. did in fact adopt a purposive approach to interpreting the Charter ’s guarantee of freedom of association (pp. 363 and 393-94). It was not their failure to apply such an approach that led to the conclusion that all associations must receive identical freedoms under s. 2(d). Rather, it was the application of that approach that correctly led them to recognize that a guarantee protecting a fundamental freedom to associate must be interpreted in a content-neutral fashion as between different associations.

The protection of fundamental freedoms should not involve the Court in adjudicating the relative values of the way in which individuals exercise those freedoms. Just as this Court has not adjudicated on the relative value of a religion or its tenets under s. 2(a) or assessed the relative value or content of a given exercise of freedom of expression under s. 2(b), so too should this Court not privilege some associations over others under s. 2(d): Syndicat Northcrest v. Amselem, 2004 SCC 47, [2004] 2 S.C.R. 551, at para. 50.

Another example of a content-neutral approach can be found in Big M where this Court found:

In my view, however, as I read the Charter , it mandates that the legislative preservation of a Sunday day of rest should be secular, the diversity of belief and non-belief, the diverse socio-cultural backgrounds of Canadians make it constitutionally incompetent for the federal Parliament to provide legislative preference for any one religion at the expense of those of another religious persuasion. [Emphasis added; p. 351.]

Big M considered the purpose of the Charter , recognized the Christian underpinning of the impugned law, and concluded that by imposing “a sectarian Christian ideal, the Lord’s Day Act works a form of coercion inimical to the spirit of the Charter and the dignity of all non-Christians” (p. 337). Dickson J.’s words in Big M expressly say that there is no constitutional basis to prefer one religion over others.

In an article critical of the Health Services decision, Professor Langille describes as “chilling” the suggestion that the Court should “weig[h] the harm of banning book clubs as compared to banning collective bargaining and relegat[e] the former to a lower level of concern”: “Freedom of Association”, at p. 185. He goes on to state:

We should begin with a reminder that this is, after all, a constitution that is being interpreted. It is an entrenched bill of rights and freedoms. The Charter value of freedom of association is a basic one. It applies to all Canadians, and the role of the Court is to interpret it in a principled way. There is, in my view, not one freedom of association for Nova Scotia and another for Ontario, nor one for students and one for tenants, nor one for the service sector and another for the manufacturing sector. If there were, this freedom would not be a matter of fundamental justice. It would be, to use the Court’s word, “contextual”. [p. 202]

Like Professor Langille, I question whether the approach advocated in Health Services accords with a purposive interpretation of Charter rights. In Health Services, the majority appeared to be inquiring into the purpose of an activity to see if it merits constitutional protection. This approach requires judges to select among a range of objects and activities on the basis of their general “importance” to society rather than their connection to the freedom to associate. It is inappropriate for the Court to engage in this sort of inquiry in defining the scope of a constitutional right. It would be assessing whether, as a matter of policy, a given activity merits constitutional protection. In my view, the purpose of s. 2(d) is to protect associational activity against precisely such value judgments. A “contextual” approach of the sort proposed in Health Services would in fact be contrary to the purpose of s. 2(d) as it requires the judiciary to engage in these value judgments itself: see R. K. Basu, “Revolution and Aftermath: B.C. Health Services and Its Implications” (2008), 42 S.C.L.R. (2d) 165, at pp. 186-87. In R. v. Advance Cutting & Coring Ltd., 2001 SCC 70, [2001] 3 S.C.R. 209, LeBel J. recognized that such a role would be inappropriate for the Court and would take the Court beyond the scope of s. 2(d):

If this constitutional guarantee were to apply to the widest range of associations with the most diverse objects and activities, extending constitutional protection to a legislative creation like collective bargaining might have unforeseeable consequences and widen the sphere of constitutional protection to undefined and unknowable activities, well beyond the proper domain of s. 2(d). [para. 180]

My colleagues say that the “consideration of goals [of a particular association] cannot be avoided” as Charter rights “must be interpreted in a purposive way — having regard for the purposes, or goals, they serve” (para. 75). They say that a “content-neutral right is too often a meaningless right” (ibid.). Based on my colleagues’ approach, meaningful freedom of association requires the state to coerce employers into negotiating in good faith with employee associations.

However, as I have just discussed, their approach would require the courts to focus on the purposes of a particular association rather than the purpose of freedom of association found in the Charter itself. Their approach diverges from the purposive approach to Charter interpretation explained in Big M. Rather than focussing on the linguistic, philosophic and historical contexts of the right itself, their approach focusses on the particular aims and goals of the association in question. The inquiry shifts from a consideration of the purpose and context of the Charter , to a consideration of what activities the courts believe are normatively and subjectively more important. The result of the approach in Health Services was that the goal of employee associations — imposing the obligation of collective bargaining on employers — is constitutionally entrenched, while the goals of other associations were not. It is difficult to see how this result can be anything other than a judicial endorsement of the importance of collective bargaining over other unconstitutionalized associational activities.

A content-neutral approach does not allow for constitutional protection of collective bargaining and no constitutional protection of the aims and objects of other associations. Short of protecting all aims and objects of associations, s. 2(d) cannot be interpreted in a fashion which is neutral as between different associations while imposing a duty of collective bargaining on employers and groups of employees.

(4) Section 2(d) Does Not Give Constitutional Status to Contracts

A fourth difficulty with the collective bargaining right in Health Services is that it places contracts above statutes in the traditional hierarchy of laws. Although Health Services purported to constitutionalize the process of collective bargaining rather than its fruits, it in fact granted constitutional protection to the collective agreements on the basis that they were the fruits of that process. In Health Services, the challenged legislation had the effect of invalidating portions of existing collective agreements and consequently “undermining the past bargaining processes that formed the basis for these agreements” (para. 113). This was found to violate s. 2(d) (para. 136).

In response to this argument, the Chief Justice and LeBel J. write that Health Services did not hold that “labour contracts could never be interfered with by legislation” (para. 76). However, if as Health Services holds, it is unconstitutional for a statute to legislatively nullify “significant contractual terms”, then it must logically follow that those “significant contractual terms” have been elevated above statutes. The actual effect of Health Services is quite the opposite from what my colleagues assert it to be.

My view is consistent with that of Professor Hogg who observed that “[t]his ruling elevated collective agreements above statutes in the hierarchy of laws, and granted them virtually the same status as the provisions of the Charter itself” (p. 44-9). Indeed, in the Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), [1990] 1 S.C.R. 1123 (the “Prostitution Reference”), at p. 1171, Justice Lamer (as he then was) explicitly rejected the idea that a constitutional guarantee safeguarding freedom of contract was included under the Canadian Charter .

(5) Courts Have Afforded the Legislature Significant Deference in the Application of Section 2(d) to the Field of Labour Relations

A final difficulty with the approach to s. 2(d) taken in Health Services is that it explicitly rejected judicial deference by judges towards the legislature in labour relations. Indeed, the majority indicated that Courts had previously taken an “overbroad view” of judicial deference (para. 26). While judicial deference has its limits, the general approach of judicial deference in the field of labour relations is well supported by precedent and is sound in principle. I am of the view that the reasons for judicial deference strongly militate against constitutionalizing the right to collective bargaining. In my respectful opinion, the majority erred in Health Services by removing decision-making power on this question from Parliament and the provincial legislatures.

For nearly twenty years between the Trilogy and Health Services, a majority of this Court was consistently of the view that judges should defer to legislators on labour relations matters. As discussed by LeBel J. at paras. 156-62 of Advance Cutting & Coring, this position stemmed from a recognition that the management of labour relations requires a delicate exercise in reconciling conflicting values and interests and that the political, social and economic considerations that this exercise raises lie largely beyond the expertise of the courts. This position was also in line with history. The law of collective bargaining, as it has developed in Canada since the Depression and the Second World War, as well as union and employer conflicts like strikes and lockouts, have been subject to legislative control based on government policy rather than judicial intervention.

Beginning with the decisions in the labour Trilogy and continuing through PIPSC, Lavigne v. Ontario Public Service Employees Union, [1991] 2 S.C.R. 211, Delisle, and Advance Cutting & Coring, this Court has consistently deferred to elected legislatures on issues of labour relations. In Advance Cutting & Coring, Justice LeBel, writing for himself and Gonthier and Arbour JJ., described the Court’s deferential approach as a “non-intervention policy” (para. 160). LeBel J. upheld the legislative regime at issue in the appeal:

The question at stake in this appeal should thus be left to the political process. Such a solution would be consistent with the jurisprudential attitude of the Court that was summarized above. It retains a balance in the application of the Charter . It leaves the legal management of labour relations to Parliament and legislatures as well as to the parties to labour agreements, as the majority of the Court has held consistently since the labour law trilogy of 1987. . . . This limited and prudent approach to court interventions in the field of labour relations reflects a proper understanding of the functions of courts and legislatures. [Emphasis added; para. 239.]

The principle of deference provides a reason for choosing a more restrained version of s. 2(d) when the Court is faced with competing visions of what s. 2(d) protects.

My colleagues imply that my view on deference creates a “judicial ‘no go’ zone” or creates a “Charter-free zone” for labour relations (paras. 78-80). That is not a correct understanding of my position. Clearly if legislatures, for example, chose to enact legislation that permitted discrimination in labour relations or precluded the ability to form an employee association, that legislation would be subject to judicial review for being non-compliant with the Charter . In matters of labour relations, the Charter still applies. If my colleagues believe that my view on deference creates a “Charter-free zone”, they have misunderstood my reasons.

In my opinion, the principle of judicial deference in the field of labour relations is rooted in two underlying concerns. The first of these is that the Court is ill-equipped to carry out the requisite balancing of interests in the labour relations context. Since McIntyre J.’s comments in the Alberta Reference, this Court has recognized that labour relations are an “extremely sensitive subject” premised on “a political and economic compromise between organized labour — a very powerful socio-economic force — on the one hand, and the employers of labour — an equally powerful socio-economic force — on the other” (p. 414). While the courts are responsible for safeguarding the ability of individuals to do collectively that which they have the right to do as individuals, the judiciary is ill-equipped to engage in fine adjustments to the balance of power between labour and management.

The Chief Justice and LeBel J. write that this Court has “distanced itself” from the approach of McIntyre J. in the Alberta Reference, and now deals with deference under s. 1, rather than in outlining the scope of s. 2(d) (para. 81). With respect, I think this point is debatable. While that may have been the approach adopted by the majority in Dunmore, there are a number of examples of this Court dealing with deference at the s. 2(d) stage of the analysis. For example, LeBel J. dealt with deference at the s. 2(d) stage of the analysis in Advance Cutting & Coring. Another example is Delisle, where Bastarache J. for the majority (writing on behalf of Gonthier, McLachlin (as she then was), and Major JJ.), dealt with the concept of deference under the s. 2(d) stage of the analysis. Bastarache J. wrote that he shared the opinion of McIntyre J. in the Alberta Reference, writing that “[f]reedom of association does not include the right to establish a particular type of association defined in a particular statute; this kind of recognition would unduly limit the ability of Parliament or a provincial legislature to regulate labour relations” (Delisle, at para. 33).

The second underlying concern justifying judicial deference in the area of labour relations is that courts should avoid extending constitutional protection to a particular statutory model of labour relations. Different statutory models of labour relations may be appropriate in different socio-economic contexts, and the courts should avoid defining as a matter of constitutional law the particular framework according to which all Canadian labour relations must be structured for the indefinite future. Parliament and the provincial legislatures should not be precluded from fashioning appropriate labour relations regimes that are responsive to the relevant socio-economic contexts.

In Health Services, this Court departed from a long-standing course of judicial deference in the field of labour relations. The majority intervened to adjust aspects of the balance of power between unions, employers and individual employees, and, as a consequence, constitutionalized prominent features of the Wagner model under s. 2(d) of the Charter , namely collective bargaining.

Although the majority insisted that it was not enshrining a particular model of labour relations, and the Chief Justice and LeBel J. say that such a conclusion was “repeatedly rejected” (para. 77), I believe that such a conclusion is inescapable. As my colleague Deschamps J. observes, the duty to negotiate in good faith enshrined by Health Services “is one of the hallmarks of the Wagner model and that inevitably entails a number of statutory components” (para. 304). Professor Hogg writes:

The majority . . . claimed that it was not constitutionalizing “a particular model of labour relations”. But that is exactly what it was doing: North American labour relations regimes are based on the American Wagner Act of 1935. The Wagner model of compulsory collective bargaining with a single union compulsorily representing all members of a bargaining unit has not been adopted outside the United States and Canada, and, even in the United States, compulsory arbitration or other wage-setting mechanisms often replace collective bargaining in the public sector. Presumably, only compulsory collective bargaining on the Wagner model will now pass muster in Canada. The majority even claimed that the Court had been wrong in the past to exercise “judicial restraint in interfering with government regulation of labour relations”. But, without any clear prescription in the Charter , there is much to be said for leaving the regulation of labour relations to elected legislative bodies (and the sanction of the ballot box). [p. 44-8]

As Professor Langille points out, examination of the different labour relations regimes of the roughly 180 that compose the International Labour Organization (“ILO”), reveals a range of ways a government might choose to structure labour relations: see B. Langille, “Why Are Canadian Judges Drafting Labour Codes — And Constitutionalizing the Wagner Act Model?” (2009-2010), 15 C.L.E.L.J. 101, at p. 107. By way of illustration, bargaining between employers’ associations and trade unions is entirely voluntary in Germany; specifically, there is no obligation on employers’ association to bargain in good faith, as there is in North American labour relations regimes: see generally K. G. Dau-Schmidt, “Labor Law and Industrial Peace: A Comparative Analysis of the United States, the United Kingdom, Germany, and Japan Under the Bargaining Model” (2000), 8 Tul.

More fundamentally, the fact that the Wagner model of collective bargaining is currently the dominant mode of resolving labour relations issues today does not mean that this will always be the case. Peter A. Gall sounded this note of caution in the early years of the Charter :

Collective bargaining is extremely important in our society and has been for some time now. But will it always be so? Can we confidently predict that 50 or even 20 years from now collective bargaining will still be the primary activity of trade unions? Or will we have adopted some other technique for setting terms and conditions of employment, such as full-scale interest arbitration or greater reliance on legislated standards. If we cannot reject this out of hand, and I do not think we can, then we must seriously question whether collective bargaining is the kind of activity that warrants constitutional status. The Charter enshrines the fundamental principles of individual liberty. The activities of man may change over time, but these principles remain constant. Collective bargaining does not have this same timeless quality, and, accordingly, we should be leery of giving it constitutional protection under the concept of freedom of association.

(“Freedom of Association and Trade Unions: A Double-Edged Constitutional Sword”, in J. M. Weiler and R. M. Elliot, eds., Litigating the Values of a Nation: The Canadian Charter of Rights and Freedoms (1986), 245, at p. 248)

For these reasons, I am of the view that recognizing a constitutional right to collective bargaining, as did Health Services, represents an imprudent departure from the course of judicial deference.

In the previous section, I pointed to five reasons why the approach to s. 2(d) adopted in Health Services is inconsistent with both precedent and principle relating to the purpose of s. 2(d). In this section, I address the reasons advanced by the majority in Health Services for providing s. 2(d) protection to collective bargaining. As I understand Health Services, these reasons were advanced to demonstrate that collective bargaining is a fundamental right that justifies it being constitutionalized. With respect, I do not think these reasons withstand scrutiny.

Health Services rests its conclusion that s. 2(d) of the Charter contains a right of collective bargaining on four propositions, which the majority outlines at para. 20:

First, a review of the s. 2(d) jurisprudence of this Court reveals that the reasons evoked in the past for holding that the guarantee of freedom of association does not extend to collective bargaining can no longer stand. Second, an interpretation of s. 2(d) that precludes collective bargaining from its ambit is inconsistent with Canada’s historic recognition of the importance of collective bargaining to freedom of association. Third, collective bargaining is an integral component of freedom of association in international law, which may inform the interpretation of Charter guarantees. Finally, interpreting s. 2(d) as including a right to collective bargaining is consistent with, and indeed, promotes, other Charter rights, freedoms and values. [Emphasis added.]

In this section each of these contentions is addressed.

(1) The Continuing Validity of Past Precedents on the Scope of Section 2(d)

I have already dealt with the substance of the first proposition in the course of explaining why Health Services was wrong to overrule the approach to s. 2(d) of the Charter embodied in the prior jurisprudence, and I need not discuss it further here.

(2) Canadian Labour History Does Not Support Constitutionalizing Collective Bargaining Rights

The second reason advanced by the majority in Health Services is that collective bargaining has historically been recognized in Canada as an integral component of freedom of association (para. 25). The view that a right to collective bargaining which includes a duty on employers to bargain in good faith is a pre-statutory feature of Canadian labour law contradicts established accounts of the history of labour relations in Canada.

The labour history offered in Health Services in support of this argument is inconsistent with a number of historical accounts of the development of labour law in Canada and has recently been the subject of intense academic criticism. See E. Tucker, “The Constitutional Right to Bargain Collectively: The Ironies of Labour History in the Supreme Court of Canada” (2008), 61 Labour 151; Langille, “Freedom of Association”, at pp. 191-92; and Etherington, at pp. 726-27. For a historical account, see, e.g., J. Fudge and E. Tucker, Labour Before the Law: The Regulation of Workers’ Collective Action in Canada, 1900-1948 (2001); J. Fudge, “‘Labour is Not a Commodity’: The Supreme Court of Canada and the Freedom of Association” (2004), 67 Sask. L. Rev. 425.

Professor Tucker, whose account of the development of labour relations was invoked in support of the judicial history in Health Services, characterizes the majority’s historical claims as “flawed” and describes the use of his and other historians’ work as “ironic”, in the sense that it was used “to support a narrative that is inconsistent with the interpretation of that work (without acknowledging that difference of view)” (p. 168). He writes that the majority’s historical analysis fails to support the historical proposition that it seeks to defend, “namely that ‘Association for purposes of collective bargaining has long been recognized as a fundamental Canadian right which predated the Charter ’” (p. 166).

One of the problems in Health Services is that the term “collective bargaining” is used throughout Health Services without it being acknowledged that, from a historical perspective, there are two meanings that can be ascribed to the term. On the one hand, the term “collective bargaining” refers to self-directed activities engaged in by workers without the benefit of statutory enhancements or protection, i.e. workers organizing in order to attempt collective engagement with their employer in the hopes of obtaining better pay and working conditions. See Tucker, at p. 166.

On the other hand, the term is used as shorthand for a particular kind of statutorily enabled activity. “Collective bargaining”, in the age of the Wagner model, refers to a complex package of reciprocal rights and duties imposed on employers and workers alike.

The dominance of the Wagner model of labour relations in Canada has caused the term “collective bargaining” to become virtually synonymous with a specific set of legislatively imposed obligations, including the requirement of a duty to bargain in good faith. However, collective bargaining, in its original and most basic form, did not involve or require the existence of statutory obligations.

In its decision in Health Services, the majority did not distinguish between the Wagner form of collective bargaining and the non-statutory form when stating that the “right” to collective bargaining had a long history in Canada. This is manifested in the majority’s statement, for example, that “the duty to consult and negotiate in good faith” is “the fundamental precept of collective bargaining” (para. 97). While the duty to bargain in good faith may be a fundamental precept of the Wagner model of collective bargaining, it is not a fundamental precept of the concept of collective bargaining as it was understood before the introduction of the Wagner Act or as it is still understood today in many parts of the world.

It is true that there is a long-standing practice of Canadian workers associating for the purpose of bargaining collectively with their employers. Likewise, it is true that at least since the Trade Unions Act of 1872 workers enjoyed a legal freedom to associate for the purpose of bargaining collectively with their employers without being prosecuted or sued for simply doing so.

However, the legal rights of organization, which imposed duties of non-interference on employers, and collective bargaining, which imposed good faith duties of negotiation on employers, did not exist prior to their enactment in statutes:

The establishment of a legal right for workers to associate for the purposes of forming a trade union, in the sense that employers are subject to a concomitant duty not to interfere with their organizing, however, can only be traced to the freedom of trade union association legislation passed in the 1930s, while the legal right for workers to bargain collectively, in the sense that employers have a positive duty to participate in a process of good faith negotiation with their workers’ chosen representatives, first appeared in British Columbia and Nova Scotia statutes enacted in 1937, but only became generalized for private sector workers in the 1940s and for public sector workers in the 1960s and 70s.

Thus while the court is on firm historical ground when it states in paragraph 66 that collective bargaining (understood here as a social practice) has long been recognized in Canada (in the sense that it could neither be repressed nor ignored) and that “historically it emerges as the most significant collective activity through which freedom of association is expressed in the labour context,” its further claim that a procedural right to collective bargaining has long been recognized as fundamental in Canada prior to 1982 is deeply problematic as a statement of historical fact.

(Tucker, at p. 166 (emphasis in original))

Professor Langille also takes the position that Canada’s labour history does not reveal an acceptance by our common law courts or our legislatures of a concept of freedom of association that included an obligation on the part of employers to engage in collective bargaining: “Beyond any doubt, there was no duty imposed on an employer to bargain with a union — even if, contrary to all legal indications, there was an effectively protected right to belong to a union and to participate in a strike” (“Freedom of Association”, at p. 191).

Not only did courts not recognize such a bargaining right in the period before the adoption of the Wagner model in Canada, but they often issued injunctions against labour’s attempts to bargain collectively. Professor Etherington writes:

. . . prior to the adoption of statutes in Canada modeled on the Wagner Act, the best our unions could hope for was a laissez-faire attitude that would allow them to use strikes to force employers to bargain. More often than not during that period they were even disappointed in that hope by courts that were too willing to use their injunctive powers at common law to prevent unions from taking collective economic action against employers to compel them to bargain collectively. In that context, it may be a hollow claim to even assert that our law or society recognized access to collective bargaining in any sense as a fundamental right or freedom, but it is clearly not accurate to assert that it recognized a legal right to engage in collective bargaining that included an obligation on the part of employers also to engage in bargaining when approached by unions. [Emphasis added; p. 727.]

The Chief Justice and LeBel J. take issue with my focus on whether, historically, the right to collective bargaining was consistently guaranteed by the legal system, noting that the question should instead be “whether Canadian society’s understanding of freedom of association, viewed broadly, includes the right to collective bargaining in the minimal sense of good faith exchanges affirmed in Health Services” (para. 90 (emphasis in original)). With respect, this bare assertion, without any evidence or explanation as to what Canadian society’s understanding of freedom of association actually is, does not rehabilitate the flawed historical analysis in Health Services.

In light of the consistent academic criticisms, I cannot accept the majority’s assertion in Health Services that the Wagner model statutes did not create a modern right to bargain collectively but only “afforded it protection” (para. 25). While the legal freedom to enter into voluntary collective negotiations may have been a fundamental freedom prior to legislation based on the Wagner Act, these statutes did in fact constitute a substantial innovation over the status quo ante with respect to various labour rights, including the duty of employers to bargain in good faith.

(3) International Law Does Not Support Constitutionalizing Collective Bargaining Rights

The third proposition the majority relied on in Health Services was that collective bargaining is an integral component of the freedom of association under international law. The majority relied in particular on ILO Convention (No. 87) concerning freedom of association and protection of the right to organise, 68 U.N.T.S. 17 (“Convention No. 87”), in support of the position that collective bargaining is protected under international law. In doing so, it committed two errors.

First, in discussing protection for collective bargaining under international law, the majority conflated two distinct ILO Conventions. While Canada has ratified ILO Convention No. 87, that Convention deals with freedom of association and does not at any point specifically discuss collective bargaining. The majority in Health Services cites an extended passage from an article by B. Gernigon, A. Odero and H. Guido, “ILO principles concerning collective bargaining” (2000), 139 Intern’l Lab. Rev. 33, to elaborate on the scope of protection for collective bargaining under international law. However, in that article the authors are actually discussing the scope of ILO Convention (No. 98) concerning the application of the principles of the right to organise and bargain collectively, 96 U.N.T.S. 257 (“Convention No. 98”), which deals more specifically with collective bargaining. As my colleagues acknowledge, Canada has not ratified Convention No. 98. This means that Canada has no obligations under that Convention as outlined in the ILO Constitution: see Constitution of the International Labour Organisation, 15 U.N.T.S. 40, Art. 19(5)(e); B. A. Langille, “Can We Rely on the ILO?” (2006-2007), 13 C.L.E.L.J. 273. It is therefore inappropriate to interpret the scope of Canada’s obligations on the basis of that Convention.

Second, even if Convention No. 98 were applicable to Canada, the majority in Health Services would still have erred in relying on that Convention to constitutionalize a version of collective bargaining that includes a duty to bargain in good faith. While Convention No. 98 provides protection for a process of collective bargaining, it conceives of collective bargaining as being a process of “voluntary negotiation” that is fundamentally distinct from the model of collective bargaining incorporated in the Wagner model: see ILO Convention No. 98, Art. 4. More specifically, Convention No. 98 does not contemplate the imposition of a duty on parties to bargain in good faith: Langille, “Can We Rely on the ILO?”, at pp. 291-92. Indeed, Gernigon et al. express this point in the article relied on by the majority in Health Services:

The voluntary nature of collective bargaining is explicitly laid down in Article 4 of Convention No. 98 and, according to the Committee on Freedom of Association, is “a fundamental aspect of the principles of freedom of association” (ILO, 1996a, para. 844). Thus, the obligation to promote collective bargaining excludes recourse to measures of compulsion. During the preparatory work for Convention No. 154, the Committee on Collective Bargaining agreed upon an interpretation of the term “promotion” (of collective bargaining) in the sense that it “should not be capable of being interpreted in a manner suggesting an obligation for the State to intervene to impose collective bargaining”, thereby allaying the fear expressed by the Employer members that the text of the Convention could imply the obligation for the State to take compulsory measures (ILO, 1981, p. 22/6).

The Committee on Freedom of Association, following this line of reasoning, has stated that nothing in Article 4 of Convention No. 98 places a duty on a government to enforce collective bargaining with a given organization by compulsory means, and that such an intervention by a government would clearly alter the nature of bargaining (ILO, 1996a, para. 846).

It cannot therefore be deduced from the ILO’s Conventions on collective bargaining that there is a formal obligation to negotiate or to achieve a result (an agreement). [Emphasis added; pp. 40-41.]

The majority in Health Services was in error when it concluded that international law pointed to compulsory collective bargaining (paras. 69-79). My colleagues say that international norms are not inconsistent with compulsory collective bargaining (para. 95). While this is true, it does not assist with the interpretation of s. 2(d). Many positions — including a freedom of association which includes voluntary collective bargaining — are equally, if not more, consistent with international norms. However, the majority in Health Services said more than this. It said that Canada’s obligations and those international norms imply compulsory collective bargaining more than they imply voluntary associations (para. 72). With respect, international law does not support that conclusion.

(4) Charter Values Cannot Be Invoked to Support Constitutionalizing Collective Bargaining Rights

In its fourth proposition, the majority maintained that the recognition of a good faith collective bargaining right is consistent with and promotes other Charter rights, freedoms and values: namely, human dignity, equality, liberty, respect for the autonomy of the person and the enhancement of democracy: see Health Services, at para. 81. The majority said that the right promotes human dignity, liberty and autonomy of workers by giving them the opportunity to influence the establishment of workplace rules and thereby to gain a measure of control over a major aspect of their lives, that it enhances equality because it palliates the historical inequality between employers and employees, and that it achieves a form of workplace democracy and ensures the rule of law in the workplace by giving workers a voice to influence the establishment of rules that control major aspects of their lives (paras. 82-85).

A duty to bargain in good faith may achieve those ends. However, either the Charter requires something or it does not. The Chief Justice and LeBel J. say that a “value-oriented approach . . . has been repeatedly endorsed by Charter jurisprudence over the last quarter century” (para. 96). That may be so, however this value-oriented approach is a means by which courts interpret the Charter — a process, as I will now explain, that must begin with the words of the Charter itself and must be bound by the normal constraints of legal reasoning and analysis. As Mr. Justice Robert J. Sharpe and Professor Kent Roach say, “[t]he task of Charter interpretation has structure and discipline. The first source is obvious — the language of the Charter itself” (The Charter of Rights and Freedoms (4th ed. 2009), at p. 59). The role of the Court is to determine what the Charter requires and what it does not and then apply the requirements it finds to the case before it. It is not to simply promote, as much as possible, values that some subjectively think underpin the Charter in a general sense.

I agree with the words of Iacobucci J. in Bell ExpressVu Limited Partnership v. Rex, 2002 SCC 42, [2002] 2 S.C.R. 559, at para. 62, where he wrote “to the extent this Court has recognized a ‘Charter values’ interpretive principle, such principle can only receive application in circumstances of genuine ambiguity, i.e., where a statutory provision is subject to differing, but equally plausible, interpretations” (emphasis deleted). The Court cannot employ a Charter values argument to interpret the Charter itself so broadly that the interpretation is no longer plausible. As Dickson J. observed, “it is important not to overshoot the actual purpose of the right or freedom in question” (Big M, at p. 344). This means, as Professor Hogg says, that even though this Court has adopted a progressive, purposive approach to interpreting the Constitution, courts are not liberated from the “normal constraints of interpretation” (p. 15-50).

Section 2(d) protects the freedom to associate. It does not purport to guarantee the “collective goals” (reasons of the Chief Justice and LeBel J., at para. 46) of the association once formed. The majority’s interpretation in Health Services is not plausible because it created a free-standing right to the objectives of employee associations; it created a right which “requires both employer and employees to meet and to bargain in good faith, in the pursuit of a common goal of peaceful and productive accommodation” and make a “reasonable effort to arrive at an acceptable contract” (paras. 90 and 101). To suggest that s. 2(d) protects the right to equal bargaining power with one’s employer or “a form of workplace democracy” takes it far outside its linguistic, philosophical and historical context (para. 85).

I agree that one’s work is fundamental to one’s identity and well-being and that exerting control over one’s working conditions is a desirable goal. However, even assuming that such considerations militate in favour of legislative intervention to empower workers and employee associations, the fact remains that our Constitution leaves the determination as to whether, and to what extent, such intervention is appropriate to Parliament and the provincial legislatures. Section 2(d) is silent on such matters and this Court may not intervene into questions of economic and social policy in the absence of a legislative or constitutional grant of authority.

Beyond Health Services’ errors as to the nature of s. 2(d) and its reasons for constitutionalizing a duty to bargain in good faith, the collective bargaining right itself is unworkable for two reasons. First, as found by the Ontario Court of Appeal, the right to collective bargaining imposed by Health Services is unworkable unless it is supported by at least two additional elements of Wagner model collective bargaining. Second, the attempt to draw a distinction between the process of good faith bargaining and the fruits of such bargaining is unworkable, as Health Services itself demonstrates.

(1) The Problem of Constitutionalizing One Part of the Wagner Model

In the court below in this appeal, an experienced and eminent labour lawyer and now Chief Justice of Ontario, Winkler C.J.O., took the view that a constitutional right to meaningful collective bargaining must extend to two additional aspects: the principle of majoritarian exclusivity and a mechanism for resolving bargaining impasses and disputes regarding the interpretation and administration of collective agreements. Accordingly, he ordered legislation that would extend the missing protections to agricultural workers:

If legislation is to provide for meaningful collective bargaining, it must go further than simply stating the principle and must include provisions that ensure that the right can be realized. At a minimum, the following statutory protections are required to enable agricultural workers to exercise their right to bargain collectively in a meaningful way: (1) a statutory duty to bargain in good faith; (2) statutory recognition of the principles of exclusivity and majoritarianism; and (3) a statutory mechanism for resolving bargaining impasses and disputes regarding the interpretation or administration of collective agreements. [Emphasis added; para. 80.]

Abella J. in her reasons finds that a right to collective bargaining for agricultural workers must include an enforcement and compliance mechanism to resolve bargaining disputes (para. 339) and the statutory recognition of majoritarian exclusivity (para. 343).

Winkler C.J.O. explained that these elements were necessary to a workable system of good faith collective bargaining. He justified adding the mechanism for resolving bargaining impasses by explaining that the bargaining process would be “jeopardized” if the parties had no recourse to a dispute resolution mechanism when faced with fruitless bargaining (para. 82).

Indeed, when a duty is imposed on a party by law, it must be accompanied by sanctions or means of enforcement if there is non-compliance with the duty. Without sanctions or means of enforcement, compliance with the duty would be, to all intents and purposes, voluntary. This would hardly meet the requirement, according to the Chief Justice and LeBel J., that collective bargaining be mandatory.

Winkler C.J.O. also explained that a collective bargaining process that lacks the feature of majoritarian exclusivity would be “impractical” and lead to “chaos”:

It is impractical to expect employers to engage in good faith bargaining discussions when confronted with a process that does not eradicate the possibility of irreconcilable demands from multiple employee representatives, purporting to simultaneously represent employees in the same workplace with similar job functions. It is not overstating the point to say that to avoid chaos in the workplace to the detriment of the employer and employees alike, it is essential that a representative organization be selected on a majoritarian basis and imbued with exclusive bargaining rights. [para. 92]

Winkler C.J.O.’s concerns present a significant problem that the Chief Justice and LeBel J. do not address in their reasons. They limit constitutionalization to collective bargaining imposing a duty on employers to bargain in good faith. The reasons of the Chief Justice and LeBel J. provide no explanation for why Winkler C.J.O. is wrong. As the majority in Doucet-Boudreau said in discussing minority language educational rights: “A purposive approach to remedies in a Charter context gives modern vitality to the ancient maxim ubi jus, ibi remedium: where there is a right, there must be a remedy” (para. 25). I cannot agree that a right can be workable without the imposition of an appropriate remedy.

The Chief Justice and LeBel J. say that “[i]t is premature to argue that the holding in Health Services, rendered four years ago, is unworkable in practice” (para. 83). They say that it takes time before the unworkability of a decision emerges. I disagree. Winkler C.J.O.’s conclusion that a constitutional right to meaningful collective bargaining must include constitutionalizing elements of the Wagner model provides strong support for the proposition that, without these protections, compulsory collective bargaining is unworkable.

(2) The Untenable Distinction Between Substance and Process

Unworkability also arises from the majority’s instruction in Health Services to protect the process of collective bargaining without also protecting its substantive fruits. In Health Services, the majority posited that this distinction was entirely possible (para. 29). In my view, this distinction is unworkable because it is impossible to divorce the process of collective bargaining from its substantive outcomes. There are three reasons.

First, as I have already discussed, Health Services itself did not respect this distinction since the majority granted constitutional protection to “significant” terms of the collective agreements at issue in that very case. The majority found that the challenged B.C. legislation breached s. 2(d) not just by limiting future bargaining but also by invalidating existing collective agreements and consequently undermining the past bargaining process that formed the basis for these agreements. Therefore, the application of the collective bargaining right in Health Services had the result of protecting the substance of those agreements.

Second, the duty to bargain in good faith cannot be described as only a “procedural” guarantee, as the Chief Justice and LeBel J. do in this case and as the majority did in Health Services. Recognizing an employee association and requiring the employer to engage in collective bargaining are themselves substantive outcomes for which workers organize. In a labour context, as in other contexts, certain “procedures” are favoured because they are more likely to produce a certain outcome.

The very requirement that the parties engage in collective bargaining tips the economic balance between parties in favour of the workers and, as such, constitutes a particular outcome. Consequently, the act of engaging in the process itself constitutes a concession on the part of the employer. Were it not so, organized labour would have little reason to demand constitutional protection for the “right” to engage in a process of collective bargaining.

In addition to providing the substantive benefit of requiring employers to meet with workers, the duty to bargain in good faith brings other, more specific substantive benefits. As the term is understood in Canadian labour law, the duty to bargain in good faith prohibits an employer from flat out refusing to bargain with the union or from only going to a few cursory meetings: Adams, vol. 2, at p. 10-122. However, it also goes much beyond that. Depending on the circumstances, the duty to bargain in good faith can prohibit an employer from refusing to include or discuss the inclusion of standard industry terms in a collective agreement or, conversely, insisting on the inclusion of a term to the point of impasse: see Adams, vol. 2, at pp. 10-111 to 10-112; Royal Oak Mines Inc. v. Canada (Labour Relations Board), [1996] 1 S.C.R. 369, at para. 45, per Cory J. It also obligates employers to disclose material information to unions in advance of negotiations: Adams, vol. 2, at pp. 10-124 to 10-128. All of these aspects of the duty to bargain in good faith change these measures and constrain the range of negotiating positions available to the employer and thus have a substantive impact on the terms of employment.

Finally, for a duty to bargain in good faith not to be an illusory benefit, there must be both a way of dealing with bargaining impasses as well as an effective remedy for persistent breaches of a duty to bargain in good faith. The first requires that there be some default mechanism for resolving the dispute in case an impasse is reached — such as striking or binding arbitration — while the second may require, in extreme circumstances, the imposition by an arbitrator of particular terms of a collective agreement: W. B. Rayner, Canadian Collective Bargaining Law (2nd ed. 2007), at pp. 349-55. Each of these goes well beyond a mere process and results in the protection of a particular substantive outcome.

The majority’s inability to separate substance and process, and the consequent constitutionalization of collective bargaining terms demonstrates the unworkability of the distinction between substance and process asserted in Health Services. This unworkability is further underlined by the fact that the collective bargaining itself is an outcome for which parties organize and does affect substantive outcomes. For these reasons, as well as the unenforceability of the bare right to good faith bargaining, the ruling in Health Services is unworkable.

As I have explained through these reasons, I do not accept that s. 2(d) protects a right to collective bargaining. I am, however, of the view that s. 2(d) does protect a voluntary association of workers who wish to use their associational freedoms to come together and attempt to improve their wages and working conditions.

Under Canadian law, an individual is generally free to bargain with an employer over terms of employment. Because such individual bargaining is generally lawful, it necessarily follows that the decision of individuals to band together to approach their employer must necessarily be protected. The free decisions of individuals to do in association what they can lawfully do alone lies at the very heart of s. 2(d) protection, and it therefore follows that s. 2(d) must protect the decision of individuals to come together, to form a bargaining position and to present a common and united front to an employer.

However, s. 2(d) does not provide greater legal protection to individuals acting in concert than is afforded to individuals acting alone. While greater economic clout or political power may flow from the very act of association in a way that makes the associational activity “qualitatively” different from the individual activity, the legal rights and freedoms granted to individuals acting in association under s. 2(d) are nonetheless limited to the same rights and freedoms afforded to individuals acting alone.

While s. 2(d) protects the ability of workers to come together and to organize with a view to engaging in collective bargaining with an employer, s. 2(d) does not impose any obligation on an employer to actually negotiate with a group of employees. In the individual case, there is generally no legal obligation on an employer to negotiate with the employee. It is entirely permissible for an employer, in the course of negotiating a new contract with an employee, to make a “take it or leave it” offer to an employee, which the employee may then accept or reject. Such individual agreements are generally left to voluntary negotiation in accordance with the law of contract, subject only to requirements set out in employment standards legislation and other statutes. Thus, just as an employer can decline to meet or negotiate with an individual, so can an employer decline to meet or negotiate with a group of employees.

In my view, a proper application of s. 2(d) provides protection for voluntary associations of workers, but such protection does not involve the constitutionalization of a duty on employers to engage in collective bargaining. Such an approach is, in my view, consistent with the purpose and scope of s. 2(d), the principle of judicial deference in labour relations, and Canada’s labour history and international obligations.

Given the length of these reasons thus far, I now provide a summary of the principles discussed above before proceeding to apply these principles in the present case:

1. This Court may overrule its own precedents, but it should only do so where there are compelling reasons for doing so. In this case, such compelling reasons exist. Health Services involves Charter rights that are not susceptible to legislative correction, overruled a line of prior sound decisions, is unworkable and has been the subject of intense academic criticism.

2. Health Services erred for three reasons in concluding that s. 2(d) protects collective bargaining and obliges parties to bargain in good faith:

a. First, Health Services departed from sound principles established in this Court’s precedents on the nature and scope of s. 2(d); specifically, it departed from the following five characteristics of s. 2(d):

i. The purpose of s. 2(d) is to protect individuals rather than groups per se.

ii. Section 2(d) protects freedoms not rights.

iii. Section 2(d) does not empower the Court to privilege certain associations over others.

iv. Section 2(d) does not afford constitutional protection to contracts.

v. Section 2(d) is to be interpreted in such a way as to afford deference to the legislative branch in the field of labour relations.

b. Second, the reasons advanced in Health Services for protecting collective bargaining under s. 2(d) — Canadian labour history, Canada’s international obligations, and Charter values — do not support conferring a constitutional right to collective bargaining and imposing a duty on employers to engage in collective bargaining.

c. Third, the majority’s approach to collective bargaining in particular and s. 2(d) in general articulated in Health Services is unworkable. It extends constitutional protection to the duty to bargain in good faith without importing other aspects of the Wagner framework, and by purporting to protect the process of collective bargaining without also protecting its fruits, neither of which is tenable.

3. Section 2(d) protects the ability of individuals to form associations and to do in association what they can lawfully do alone. Because individuals are generally free to bargain with their employer individually, it follows that s. 2(d) must protect the decision of individuals to come together, to form a bargaining position and to present a common and united front to their employers. However, just as an employer is not obliged to bargain with an individual employee, s. 2(d) does not oblige an employer to bargain with a group of employees.

I agree with the conclusions of the Chief Justice and LeBel J. that the AEPA does not violate s. 2(d) of the Charter , but for the reasons I have given. Section 2(d) does not confer a right of collective bargaining; nor does it impose a duty on employers to meet with employees and “consider employee representations in good faith” (para. 104). I agree with Farley J. ((2006), 79 O.R. (3d) 219) that the AEPA satisfies all of the concerns raised in Dunmore.

On a plain reading of the provisions of the AEPA it provides all of the protections which were imposed by this Court in Dunmore, but goes no further. It does not provide any right to collective bargaining, or other incidents of Wagner Act collective bargaining. Indeed up to this point, the parties and the courts have all proceeded on the basis that the AEPA did not include a duty of collective bargaining. The claimants chose to bring this case because in their view, the AEPA did not include provisions to enforce a duty of collective bargaining on agricultural employers. Based on this Court’s ruling in Dunmore that s. 2(d) did not create a right of collective bargaining, Farley J. ruled that the AEPA did not violate the Charter . Health Services subsequently expanded the scope of s. 2(d) to constitutionalize the right to collective bargaining. Thus the Court of Appeal was obliged to and did find that the AEPA was no longer Charter compliant. Both of these conclusions were entirely consistent with the text of the AEPA and the parties’ understanding that the AEPA did not include a duty of collective bargaining on agricultural employers.

By enacting the AEPA, the legislature precisely addressed this Court’s ruling in Dunmore. The text, context and purpose of the AEPA clearly demonstrate that the legislature intentionally opted not to include a duty on employers to engage in collective bargaining with employee associations.

Nonetheless, the Chief Justice and LeBel J. say that s. 5 of the AEPA can be read as imposing a duty to bargain in good faith (para. 107), which would render the statute constitutional. They argue that the words of s. 5 are ambiguous and that the interpretive tools of purposive interpretation, the presumption of consistency with the Charter , and reference to legislative debates lead to this conclusion. Like my colleagues Deschamps J. and Abella J., I cannot agree.

The words of s. 5 are unambiguous. The relevant portions of s. 5 are subsections (1), (5), (6) and (7).

5. (1) The employer shall give an employees’ association a reasonable opportunity to make representations respecting the terms and conditions of employment of one or more of its members who are employed by that employer.

. . .

(5) The employees’ association may make the representations orally or in writing.

(6) The employer shall listen to the representations if made orally, or read them if made in writing.

(7) If the representations are made in writing, the employer shall give the association a written acknowledgment that the employer has read them.

These words could not be clearer: they provide employee associations the opportunity to make representations to an employer. The only obligation on an employer is to provide the employee association with the opportunity to make representations and to listen if they are oral or read and acknowledge them if they are written.

The words “listen to” or “read” and “give the association a written acknowledgment” are not ambiguous. This Court’s approach to statutory interpretation has long held that “the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament” (Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27, at para. 21, citing Elmer Driedger in Construction of Statutes (2nd ed. 1983)). To say that the words “listen to” or “read” and “give the association a written acknowledgment” are ambiguous would be to ignore the grammatical and ordinary sense of the words, and the purpose of the AEPA, and would manufacture ambiguity where none exists.

Professor Sullivan observes that “[i]t is presumed that the ordinary meaning of legislation is the most appropriate or ‘intended’ meaning” unless there is a reason to reject that meaning (Sullivan and Driedger on the Construction of Statutes (4th ed. 2002), at p. 34). As Professor Hogg, A. A. Bushell Thornton and W. K. Wright, write in “Charter Dialogue Revisited — Or ‘Much Ado About Metaphors’” (2007), 45 Osgoode Hall L.J. 1, at p. 12:

A broader general rule, requiring the courts to stretch the plausible interpretation of a statute in order to bring it into conformity with the Charter , “would wrongly upset the dialogic balance.” When a statute is unambiguous, courts should give effect to the clearly expressed legislative intent, even if it leads to the conclusion that the statute was unconstitutional and should be struck down for breach of the Charter .

In this case, there is nothing in the AEPA that suggests that there is a reason to depart from the ordinary and grammatical sense of the words.

It is true, as my colleagues say, that the words “listen to” and “read” and “give the association a written acknowledgment” neither impose nor preclude a duty to bargain in good faith. However, this does not lead to ambiguity. A duty to bargain in good faith, as discussed above, is a term of art in labour law that carries with it a complex series of reciprocal rights and obligations (see Adams, vol. 2, at pp. 10-111 to 10-112, 10-122, and 10-124 to 10-128). Where good faith bargaining is protected by statute it is explicitly included: see, e.g., s. 17 of the Ontario Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A, which requires that parties “shall bargain in good faith and make every reasonable effort to make a collective agreement”. To simply imply the existence of a set of statutory rights by the absence of a well-known term of art stretches the interpretive exercise beyond its breaking point.

The Chief Justice and LeBel J. say that “[t]here can only be one purpose for requiring the employer to listen to or read employee representations — to assure that the employer will in fact consider the employee representations” (para. 103). They argue that this leads to the conclusion that s. 5 includes a duty on employers to engage in collective bargaining with employee associations.

The purpose of the AEPA is set out expressly in s. 1:

1. (1) The purpose of this Act is to protect the rights of agricultural employees while having regard to the unique characteristics of agriculture, including, but not limited to, its seasonal nature, its sensitivity to time and climate, the perishability of agricultural products and the need to protect animal and plant life.

(2) The following are the rights of agricultural employees referred to in subsection (1):

1. The right to form or join an employees’ association.

2. The right to participate in the lawful activities of an employees’ association.

3. The right to assemble.

4. The right to make representations to their employers, through an employees’ association, respecting the terms and conditions of their employment.

5. The right to protection against interference, coercion and discrimination in the exercise of their rights.

Nothing in the explicit s. 1 purpose supports the view that agricultural employees have a right to require agricultural employers to engage in collective bargaining.

My colleagues ask what purpose there could be to requiring employers to listen to or read employee representations if not to respond to them. The answer is that the purpose is what the words say it is — to give employees the opportunity to more effectively put forward their representations by allowing them to do so collectively, rather than acting individually. Given the unique nature of the agricultural industry as recognized in s. 1, an employer is at liberty to respond or not. With respect, my colleagues interpretation of the words, “listen to” or “read” or “give the association a written acknowledgment” as including a duty on employers to engage in collective bargaining does not accord with the purpose as expressed by the plain language of the AEPA.

Finally, the Chief Justice and LeBel J. say that when the government of Ontario introduced the AEPA it intended the legislation to provide protection for collective bargaining. They base their view on a statement made by the Minister (at the time) that the Act was meant to meet the obligations set by this Court in Dunmore. They say because the Minister used the word “meaningful” she intended that the AEPA would protect collective bargaining, as the majority of this Court deemed necessary in Health Services. They say this despite the fact that Health Services had not yet been written or even argued before this Court. They say this despite an explicit statement made by the Minister, which they quote at para. 106, that stated that the AEPA was not intended to “extend collective bargaining to agricultural workers”. They suggest that the Minister was only disclaiming Wagner Act collective bargaining, rather than collective bargaining as they frame the term (ibid.).

As with the words of the AEPA, I read the words of the Minister plainly as presented. The comments quoted by the Chief Justice and LeBel J. indicate that the AEPA was intended to meet the obligations in Dunmore, which did not include an obligation on employers to engage in collective bargaining. Given the absence of any requirement for collective bargaining in either Dunmore or the AEPA the Minister’s comments support a plain reading of s. 5 as imposing only a duty to “listen to” or “read” the representations and “give the association a written acknowledgment” if the representations are made in writing.

As the Chief Justice and LeBel J. are of the view that agricultural employers in Ontario have a duty of collective bargaining, the appropriate remedy would have to be a declaration that the AEPA is unconstitutional in its present form and expressly reading in words empowering the Agriculture, Food and Rural Affairs Appeal Tribunal to order employers to engage in collective bargaining. With respect, my colleagues’ approach goes beyond the normal constraints of statutory interpretation; it amounts to an implied reading into the AEPA the duty of collective bargaining without declaring the Act unconstitutional. The remedial approach of the Chief Justice and LeBel J. is, in my respectful view, entirely novel and unprecedented.

As I have explained earlier, the proper judicial approach in matters of labour relations law is deference to the legislature. The imposition of a duty to bargain in good faith, like many other aspects of labour relations law, has the potential to reshape the economic landscape of entire industries by strengthening the position of organized labour. Such an outcome may be desirable, but the courts are not well suited to determining whether or not it is. Decisions of this kind require a balancing of interests rather than the application of legal principles, and they are best made after having consulted with and receiving representations from the various stakeholders whose livelihoods and economic interests are likely to be affected. Courts do not have the expertise or the institutional capacity to undertake such a process and thus are not well equipped to make an informed decision. If a duty to bargain in good faith is to be imposed, it should be by the legislature and not the court.

The Chief Justice and LeBel J. say that the freedom to engage in a coordinated attempt to negotiate with one’s employer is “meaningless” if it is not backed up by a reciprocal duty on the part of the employer. I cannot agree. The right to make representations in association is not meaningless. It is meaningful because of the increased persuasive weight carried by collective representations rather than individual representations. Indeed political parties are formed on this precise premise. As Le Dain J. noted in the Alberta Reference, at p. 391:

. . . the freedom to work for the establishment of an association, to belong to an association, to maintain it, and to participate in its lawful activity without penalty or reprisal is not to be taken for granted. . . . It is a freedom that has been suppressed in varying degrees from time to time by totalitarian regimes.

Canadians are accustomed to sound government and the respect for our personal liberties. For this reason, basic freedoms that are essential to the preservation of an open and democratic society may come to be taken for granted and their constitutional protection thought of as meaningless. However, freedom, unconstrained by oppressive government, is, indeed, more than meaningful. It is invaluable.

Accordingly, on the matter of s. 2(d) of the Charter , I find that I cannot agree with the Chief Justice and LeBel J.

I am in agreement with the Chief Justice and LeBel J. as to their disposition of the issues under s. 15. On the record before this Court, the category of “agricultural worker” does not rise to the level of an immutable (or constructively immutable) personal characteristic of the sort that would merit protection against discrimination under s. 15.

For these reasons, I would dispose of the constitutional questions in the same way as the Chief Justice and LeBel J., allow the appeal and restore the judgment of Farley J.

The following are the reasons delivered by

Deschamps J. — Canadian labour law is not static. Over the years, some of the changes in this field have been reflected in judicial decisions, such as those on freedom of association under s. 2(d) of the Canadian Charter of Rights and Freedoms . Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391, fed expectations, but it also caused some bewilderment. More importantly, it generated an unnecessary debate about whether a duty to bargain in good faith has been imposed on employers. I will begin by demonstrating why, in my view, the case at bar can and should be resolved on the basis of the answers this Court actually gave to the questions raised in Health Services, in which the issue of an employer’s duty to bargain in good faith was not even raised. I will then briefly explain why I am of the view that the analytical framework articulated by the Court in Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, should be limited to the context of that case.

When the case at bar was heard by the Ontario Superior Court of Justice ((2006), 79 O.R. (3d) 219), this Court’s judgment in Health Services had not yet been released. The issue put to Farley J. at that time was whether the Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16 (“AEPA”), was consistent with this Court’s decision in Dunmore. Health Services was released after Farley J.’s judgment, but before the Court of Appeal heard the appeal in the case at bar (2008 ONCA 760, 92 O.R. (3d) 481). As a result of comments made by the majority in Health Services, the case proceeded down a completely different path in the Court of Appeal, where the issue was whether union exclusivity, majoritarianism and mechanisms for resolving bargaining impasses and disputes — all parts of the “Wagner model” on which Canadian labour law statutes are based — were required by the Charter .

At first glance, the Ontario Court of Appeal’s affirmative response to this question is so far removed from any conclusion reached in Health Services that it seems surprising. After all, the majority in Health Services made it crystal clear that no specific model of labour relations is protected by s. 2(d) of the Charter , as they said: “. . . the right is to a general process of collective bargaining, not to a particular model of labour relations, nor to a specific bargaining method” (para. 91). By so stating, the majority of this Court were indicating that the Wagner model is not enshrined in the Charter . However, considering the predominance of the Wagner model in Canadian labour law, it is easy to see how Winkler C.J.O. reached the conclusion that the majority in Health Services must have been contemplating exclusivity, majoritarianism and mechanisms for resolving bargaining impasses and disputes at the same time as they discussed the duty to bargain in good faith. In my view, the holding in Health Services does not have the broad scope being attributed to it by the majority in the case at bar and, in particular, does not extend to imposing a duty on employers to bargain in good faith. I find that the AEPA is consistent with this Court’s conclusion in Dunmore and would therefore allow the appeal, but for different reasons than the majority.

My reading of Health Services is that it represents a step forward in the recognition of collective activities: joining individual voices through collective bargaining to achieve common goals is protected by the Charter . In that case, I endorsed the view, which I still hold, that:

(1) 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;

(2) the right is to a process of collective bargaining — it does not guarantee a certain substantive or economic outcome or access to any particular statutory regime; and

(3) the right places constraints on the exercise of legislative powers in respect of the collective bargaining process. [para. 174]

This incremental interpretation of s. 2(d) of the Charter was sufficient to dispose of the questions raised in Health Services and is also sufficient to dispose of those raised in this appeal. It leaves it up to the legislatures to make the difficult policy choices that must be made in order to achieve economic balance in labour law. This interpretation is also consistent with the restraint courts show in resolving the issues raised by the parties before them. The approach should not differ in cases involving constitutional interpretation.

As the majority in the instant case note (at para. 80), in R. v. Advance Cutting & Coring Ltd., 2001 SCC 70, [2001] 3 S.C.R. 209, LeBel J. acknowledged the existence of a judicial policy of non-intervention in labour relations. He made the following comment in that case:

Looking back over nearly 20 years of the application of the Charter , it is clear that this Court has been reluctant to accept that the whole field of labour relations should fall under the constitutional guarantee of s. 2(d). The law of collective bargaining, as it has developed in Canada since the Depression beginning in 1929 and the Second World War, as well as union and employer conflicts like strikes and lockouts, have been left largely to legislative control based on government policy. Laws restricting the choice of a bargaining agent or forbidding strikes and lockouts were deemed not to engage the guarantee of freedom of association as such. The social and economic balance between employers and their collective unionized employees was viewed as a question of policy making and management of sharply conflicting interests. Thus, it was thought more appropriate to leave the resolution of such conflicts and the policy choices they required to the political process. [para. 156]

Where the economic balance is concerned, I share the view expressed by LeBel J. in Advance Cutting:

Legislatures are entitled to a substantial, though not absolute, degree of latitude and deference, to settle social and economic policy issues (RJR-MacDonald, at para. 134, per McLachlin J.). Courts should be mindful to avoid second-guessing legislatures on controversial and complex political choices (M. v. H., [1999] 2 S.C.R. 3, at para. 79, per Cory and Iacobucci JJ.). As discussed above, the jurisprudence acknowledges that legislative policy-making in the domain of labour relations is better left to the political process, as a general rule. [para. 257]

An approach to constitutional interpretation similar to the one I set out above would have ensured a softer landing for Health Services. I readily acknowledge that the commentary on that decision was not unanimous. The Chief Justice and LeBel J. refer to the authors whose comments were mostly favourable, while Rothstein J. refers to those who were mostly unfavourable. I disagree with the expansive approach taken by the majority in the case at bar and with the use they make of Health Services. A more prudent approach, one that would be consistent with this Court’s jurisprudence on s. 2(d) and with the issues the Court actually considered in that case, would be to restrict the ratio of Health Services to the questions actually raised and the answers actually given in that case.

In Health Services, the claimants asked this Court to declare that the government had interfered with their right to unite to achieve common goals. While they recognized that under most Canadian labour law statutes, employers had an obligation to bargain in good faith, the claimants were not seeking a declaration characterizing this obligation as a constitutional one. Neither the British Columbia Supreme Court nor the Court of Appeal dealt with a duty on employers to bargain in good faith, because this subject was quite simply not raised. Indeed, it was in its legislative capacity — not as an employer — that the government had interfered with the employee’s rights. Therefore, the majority in Health Services did not need to comment on or make findings in respect of whether the government, as an employer, had a duty to negotiate in good faith. There was thus no need to impose a Charter-based duty to bargain on employers. A fortiori, there was no need to import, together with this duty, the good faith element that is one of the hallmarks of the Wagner model and that inevitably entails a number of statutory components. I cannot therefore agree with the majority in the case at bar that Health Services imposes constitutional duties “on governments as employers” (para. 73).

All that was required by the questions raised in Health Services was a finding that since the employees had a constitutional right to engage in associational activities and act in common to reach common goals, the legislature could not interfere with their right (i) by prohibiting them from addressing certain issues in the collective bargaining process; and (ii) by cancelling negotiated provisions in the agreements and thereby rendering the process meaningless. By enacting legislation that had prohibited the negotiation of certain issues in the course of Charter-protected associational activities and rendering useless the efforts expended to achieve a negotiated agreement on certain subjects, the legislature had interfered with their right. This conclusion did not depend on the employer’s being under a duty to bargain in good faith.

In the case at bar, the issue is similar to the one in Health Services in that it concerns legislative action, but it is not, as the Chief Justice and LeBel J. put it, whether the “AEPA provides a process that satisfies” “the right of an employees’ association to make representations to the employer and have its views considered in good faith” (para. 99). Although the right of employees to have their views considered in good faith may well flow from certain comments made in Health Services, they do not flow from the issues raised in that case. The duty to act in good faith is part and parcel of a web of statutory components. It should not be found to be a constitutional requirement in the instant case.

To frame the issue in this case, the AEPA must be situated in its context. The AEPA is the response of the Ontario legislature to this Court’s decision in Dunmore. In that case, agricultural workers had been excluded from the general statutory regime and had suffered from the statute’s underinclusiveness. The expanded definition of “freedom of association” that resulted from Health Services has no bearing on the protection the Ontario legislature must provide to agricultural workers. The reason is that Dunmore purported to impose on the Ontario legislature an obligation to provide agricultural workers with more than what had until then been considered to be included in the scope of the constitutionally protected right to associate. Indeed, the decision in Dunmore, which was consistent with Delisle v. Canada (Deputy Attorney General), [1999] 2 S.C.R. 989, at para. 10, was based on the premise that s. 2(d) “exists independently of any legislative framework”. However, it was held that agricultural workers should be afforded greater protection because they were vulnerable and were substantially unable to exercise their constitutional right without the support of a legislative framework.

As I explained above, in my view, the effect of Health Services is that freedom of association includes the freedom to engage in associational activities and the ability of employees to act in common to reach shared goals related to workplace issues and terms of employment. This delineation of the scope of freedom of association does not entail a more expansive protection than the legislative framework mandated by Dunmore for the agricultural workers. Therefore, if the AEPA complies with Dunmore, it will necessarily comply with the Charter . To answer the question in the case at bar, there is no need to import a duty to bargain in good faith. I cannot therefore agree with the statement of the majority in the case at bar that, “[s]ince Health Services, it has been clear that [a meaningful exercise of the right to associate] requires employers to consider employee representations in good faith” (para. 104).

I would be remiss were I not to mention the observation of the majority in Health Services (at para. 88) that, owing to s. 32 of the Charter , “a private employer is not bound” by s. 2(d) seems to have been lost in the case at bar. The requirement created by the majority in the instant case (at para. 73) that the legislature “impose statutory obligations on employers” to conduct good faith negotiations cannot be found in Health Services. As a result of this new requirement, Parliament and the legislatures will now, subject to justification under s. 1 of the Charter , have a constitutional obligation to ensure that their Wagner-based labour legislation includes all employees. The irony of this result is that no such obligation was sought by the claimants either in Dunmore or in Health Services, the very cases on which the majority now rely to support their statement in this appeal.

If Parliament and the legislatures are required to impose, in their statutory schemes, a duty on employers to bargain in good faith, this duty will apply to all public and private sector employees. Dunmore, which concerned the exclusion of a group of employees who required assistance to exercise their constitutional right to associate, will now apply to all Canadian employees regardless of whether they need such support to exercise their s. 2(d) right (see Delisle). This being so, the Court will be making a policy decision in the place of Parliament and the legislatures. I would prefer to exercise restraint in such a case.

Because my interpretation of Health Services is based on the questions that were actually raised in that case and the answers the Court actually gave to those questions, I am of the view that in the instant case, the AEPA has not been proven to violate the employees’ right to associate. Section 1 of the AEPA lists the rights of agricultural employees as follows: (1) the right to form or join an employees’ association; (2) the right to participate in the lawful activities of an employees’ association; (3) the right to assemble; (4) the right to make representations to their employers, through an employees’ association, respecting the terms and conditions of their employment; and (5) the right to protection against interference, coercion and discrimination in the exercise of their rights. Farley J. pointed out that Dunmore did not require the “legislation . . . to incorporate a complete panoply of collective bargaining rights” (para. 22) and found that the impugned legislation met the standards established in Dunmore. I agree with Farley J., and I respectfully disagree with the Court of Appeal’s interpretation.

Since Dunmore remains central to this appeal, I must comment briefly on the approach taken in that case.

In Health Services (at para. 176), I voiced concerns about the majority’s adoption of a criterion used in Dunmore to determine whether the government had infringed a Charter right. Even though both cases are based on compelling facts, principles should not be imported from one context into another that is not analogous to it. As I explained above, a similar unwarranted importation of principles can be observed in the case at bar. But there is more.

It is helpful to recall that, as noted by Judy Fudge, “[a]t issue in Dunmore was the total exclusion of agricultural workers in Ontario from any form of labour legislation that protected them against employer retaliation from joining and participating in a trade union. The union did not ask for collective bargaining rights” (“The Supreme Court of Canada and the Right to Bargain Collectively: The Implications of the Health Services and Support case in Canada and Beyond” (2008), 37 Indus. L.J. 25, at p. 30).

Dunmore was obviously a difficult case. At its heart was the economic inequality being suffered by agricultural workers. While labour law is to some extent always about economic inequality, this issue does not occupy the forefront of every battle. Health Services was not primarily about economic inequality — it concerned legislative interference with both existing and future collective agreements. Although economic inequality has the potential to undermine the peaceful foundations of democratic societies, economic equality is not an “equality right” for the purposes of s. 15 of the Charter . In addition, even though labour law provides tools that help reduce economic inequality, not all aspects of labour law are protected by the Charter (see Health Services, at para. 19). Finally, employment status is not, at least not at this time, regarded as an analogous ground for the purposes of s. 15 of the Charter .

Dunmore was based on the distinction between positive and negative rights. In my view, using this distinction as a basis for finding that s. 2(d) has been infringed involves some difficulty. Both the commentary and the case law provide sufficient justification for exercising caution before adopting an approach that relies on the positive-negative distinction, particularly when the claim concerns state action or inaction. Stephen Holmes and Cass R. Sunstein express this eloquently in The Cost of Rights: Why Liberty Depends on Taxes (1999), at pp. 43-44:

Individuals enjoy rights, in a legal as opposed to a moral sense, only if the wrongs they suffer are fairly and predictably redressed by their government. This simple point goes a long way toward disclosing the inadequacy of the negative rights/positive rights distinction. What it shows is that all legally enforced rights are necessarily positive rights.

. . . That is to say, personal liberty cannot be secured merely by limiting government interference with freedom of action and association. No right is simply a right to be left alone by public officials.

This brings to mind Cory and Iacobucci JJ.’s response in Vriend v. Alberta, [1998] 1 S.C.R. 493, at para. 56, to an argument that was analogous to the positive-negative rights dichotomy:

It is said, however, that this case is different because the challenge centres on the legislature’s failure to extend the protection of a law to a particular group of people. This position assumes that it is only a positive act rather than an omission which may be scrutinized under the Charter . In my view, for the reasons that will follow, there is no legal basis for drawing such a distinction.

Distinguishing between the freedom to exercise a right without state interference and the right to exercise a freedom unhampered by state action or inaction diverts the discussion from the substance of the actual protection afforded by the Charter . In our society, government activity is pervasive and unavoidable: see S. Bandes, “The Negative Constitution: A Critique” (1990), 88 Mich. L. Rev. 2271, at p. 2285, and Greater Vancouver Transportation Authority v. Canadian Federation of Students — British Columbia Component, 2009 SCC 31, [2009] 2 S.C.R. 295, at para. 34.

Dunmore was concerned with economic inequality. It was based on the notion that the Charter does not ordinarily oblige the government to take action to facilitate the exercise of a fundamental freedom. Recognition was given to the dichotomy between positive and negative rights. To get around the general rule, a somewhat convoluted framework was established for cases in which the vulnerability of a group justified resorting to government support. I agree with B. Langille, “The Freedom of Association Mess: How We Got into It and How We Can Get out of It” (2009), 54 McGill L.J. 177, that this detour appears to have been an artifice designed to sidestep the limits placed on the recognition of analogous grounds for the purposes of s. 15.

To redress economic inequality, it would be more faithful to the design of the Charter to open the door to the recognition of more analogous grounds under s. 15, as L’Heureux-Dubé J. proposed in Dunmore. Such an approach is preferable to relying on a distinction that does not rest on a solid foundation. This, of course, would entail a sea change in the interpretation of s. 15 of the Charter . The majority in the instant case resist such a change, referring to “Canadian values” and to the need to take a “generous and purposive” approach when interpreting Charter rights (at paras. 32, 90, 92 and 97), but to ensure consistency with the approach of the majority in Health Services (at paras. 81-96), they refer to equality in the s. 2(d) context without mentioning s. 15. My point here is not that each Charter protection should be interpreted in a formalistic manner. Rather, it is that if the law needs to move away from Dunmore’s distinction between positive and negative rights, this should not be accomplished by conflating freedom of association with the right to equality or any other Charter right that may be asserted by a litigant. An analysis based on principles grounding the protection of rights and freedoms offers a better prospect of judicial consistency than one based on the more amorphous notion of “Canadian values”.

For these reasons, I would allow the appeal and restore Farley J.’s judgment.

The following are the reasons delivered by

Abella J. (dissenting) — I fully endorse the Chief Justice and LeBel J.’s discussion of Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391. I agree with them that by including protection for the process of collective bargaining, Health Services enhanced the scope of s. 2(d) of the Canadian Charter of Rights and Freedoms beyond the formalism assigned to it by this Court’s 1987 labour Trilogy (Reference re Public Service Employee Relations Act (Alta.), [1987] 1 S.C.R. 313; PSAC v. Canada, [1987] 1 S.C.R. 424; and RWDSU v. Saskatchewan, [1987] 1 S.C.R. 460). I am also in agreement with their criticisms of Rothstein J.’s decision to reconsider the correctness of Health Services on his own motion, in the absence of a request from any of the parties that he do so, and without an opportunity for them to address the issue.

With the greatest respect, however, I do not agree that the Agricultural Employees Protection Act, 2002, S.O. 2002, c. 16 (“AEPA”), meets the new Health Services standard. I have great difficulty with stretching the interpretive process in a way that converts clear statutory language and express legislative intention into a completely different scheme. The AEPA does not protect, and was never intended to protect, collective bargaining rights.

Background

The AEPA was enacted in 2002 to respond to this Court’s 2001 decision in Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, which held that s. 2(d) protected the right to organize. Dunmore was decided in accordance with the labour Trilogy, the then operative s. 2(d) paradigm. The Trilogy was widely taken as standing for the proposition that s. 2(d) did not include protection for collective bargaining (Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), [1990] 2 S.C.R. 367). The Trilogy was not challenged in Dunmore, and Bastarache J., writing for the majority, was explicit that he was not addressing whether collective bargaining was protected under s. 2(d). What was protected, in his view, was the following:

. . . I conclude that at minimum the statutory freedom to organize in . . . the [Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A] ought to be extended to agricultural workers, along with protections judged essential to its meaningful exercise, such as freedom to assemble, to participate in the lawful activities of the association and to make representations, and the right to be free from interference, coercion and discrimination in the exercise of these freedoms. [Emphasis added; para. 67.]

It is not surprising, therefore, that the 2002 AEPA contains no reference to a protection which made no appearance on the constitutional stage until 2007. Or that the trial judge’s decision in 2006 in the case before us, applied the Dunmore “right to organize” template and found the legislation compliant with s. 2(d) ((2006), 79 O.R. (3d) 219).

But by the time the Court of Appeal heard this case in 2008, Health Services had been decided, creating a completely different jurisprudential universe. That was the new s. 2(d) universe Winkler C.J.O. applied to the AEPA (2008 ONCA 760, 92 O.R. (3d) 481). He found the legislation wanting. I agree with him.

Analysis

In granting constitutional protection to the process of collective bargaining under s. 2(d), Health Services found the duty to consult and negotiate in good faith to be a “fundamental precept” (para. 97). This does not guarantee that a collective agreement will be achieved, but good faith bargaining does require that the parties meet, engage in a meaningful dialogue, and make reasonable efforts to arrive at a collective agreement (paras. 90 and 101). Health Services confirmed that this involves not only the employees’ collective right, as confirmed in Dunmore, to organize and make representations, but also a corollary duty on the part of employers to meaningfully discuss, consult, and consider these representations:

. . . the right to bargain collectively protects not just the act of making representations, but also the right of employees to have their views heard in the context of a meaningful process of consultation and discussion. . . . [T]he right to collective bargaining cannot be reduced to a mere right to make representations. [Emphasis added; para. 114.]

(See also para. 101.)

This requirement of meaningful dialogic consultation has long been recognized in collective bargaining regimes: Royal Oak Mines Inc. v. Canada (Labour Relations Board), [1996] 1 S.C.R. 369, at para. 41; U.E.W. and DeVilbiss Ltd., [1976] 2 C.L.R.B.R. 101 (Ont.); George W. Adams, Canadian Labour Law (2nd ed. (loose-leaf)), vol. 2, at paras. 10.1710 and 10.1870-10.1920; Donald D. Carter et al., Labour Law in Canada (5th ed. 2002), at paras. 621-22; Wesley B. Rayner, Canadian Collective Bargaining Law (2nd ed. 2007), at pp. 333-34; Elisheva (Elika) Barak-Ussoskin, “Collaboration in the Tripartite System: The Right to be Consulted and the Duty to Consult”, in A. Höland et al., eds., Employee Involvement in a Globalising World: Liber Amicorum Manfred Weiss (2005), 439, at p. 445.

If we then turn to the relevant language of the AEPA and its description of what is required of an employer, we find the following:

5. (1) The employer shall give an employees’ association a reasonable opportunity to make representations respecting the terms and conditions of employment of one or more of its members who are employed by that employer.

. . .

(5) The employees’ association may make the representations orally or in writing.

(6) The employer shall listen to the representations if made orally, or read them if made in writing.

(7) If the representations are made in writing, the employer shall give the association a written acknowledgment that the employer has read them.

The process created by these provisions is the following: an employees’ association is entitled to make representations, either orally or in writing, about the terms and conditions of employment. If the representations are made orally, the employer is required to “listen” to them. If they are made in writing, the employer must “read” them and give the “employees’ association” a written acknowledgment that the representations have been read. That is the full extent of the employer’s duties — to listen, to read, and to acknowledge receipt. No response is required.

If we compare these duties under the AEPA to the linguistic markers set out in Health Services, we find that the following language is missing in action: “negotiate”, “meet”, “good faith”, “engage”, “exchange”, “dialogue”, “consultation”, “discussions”, “consideration”, “accommodation” and “union”. Nor does the key word “bargaining” appear.

Noting the absence in the AEPA of Health Services’ collective bargaining vocabulary is not a criticism of the government’s motives. The AEPA was the government’s good faith — and, as the trial judge found, successful — implementation of how Dunmore had defined the scope of s. 2(d) in 2001. This does not, however, assist in determining whether it complies with the revised scope described in Health Services. The Ontario government obviously cannot be held responsible for the redefinition of s. 2(d) that intervened between the enactment and appellate review of the AEPA, but neither can courts disregard the applicable law because of its infelicitous timing. Since the applicable law for s. 2(d) is now found in Health Services, the AEPA must be scrutinized for compliance with its principles. And since, on its face, no bargaining or consultation is required by the AEPA, let alone the good faith bargaining Health Services set out as a minimal constitutional protection, the AEPA violates s. 2(d) of the Charter .

Not only is there clarity of language, there is also clarity of purpose. The government’s intentions to exclude collective bargaining were forthright. The then Minister of Agriculture and Food, the Honourable Helen Johns, was unequivocal when she introduced the legislation in confirming that the legislation included no right to collective bargaining:

However, I need to make one thing very clear here. While an agricultural employee may join an association that is a union, the proposed legislation does not extend collective bargaining to agricultural workers.

(Legislative Assembly of Ontario, Official Report of Debates (Hansard), No. 46A, October 22, 2002, at p. 2339 (emphasis added))

This was based on the legislative goal of complying only with the rights required by Dunmore, rights which, as the Minister correctly noted, addressed only the “right to associate”, not the “right to collectively bargain”:

I’d like to say that the Supreme Court was very clear. They said that agricultural workers across the province had the right to associate. They did not say that they had the right to collectively bargain.

(Legislative Assembly of Ontario, Official Report of Debates (Hansard), No. 43A, October 16, 2002, at p. 2128 (emphasis added))

Judging from their conduct, the parties involved in this appeal seem to have accepted there were no protections for the process and enforcement of collective bargaining in the AEPA. The United Food and Commercial Workers Union Canada represented workers at Rol-Land Farms Ltd. After a vote in which an overwhelming majority voted in favour of certification, the union wrote to Rol-Land requesting a meeting to begin negotiations. The owner of Rol-Land Farms did not respond to the letter and refused to recognize the union. The same union also represented employees at Platinum Produce, where the employer gave the union the opportunity to make brief oral representations, but said it had no obligation to bargain towards a collective agreement. The meeting lasted 15 minutes.

In the years since the AEPA was enacted in 2002, there is no evidence of a single successfully negotiated collective agreement or even of any negotiations. I appreciate that statutory interpretation does not draw on the perceptions of the statute’s intended consumers, but where, as here, there is perfect harmony between statutory language, legislative intention, and public perception, the usual interpretative tools are vindicated.

In addition to finding a violation of s. 2(d) based on the explicit failure, by text and by design, to include even a hint of a process of collective bargaining, let alone a duty to engage in meaningful and good faith efforts to arrive at a collective agreement, I also agree with Winkler C.J.O. that for agricultural workers, the absence of a statutory enforcement mechanism and of majoritarian exclusivity is an infringement of s. 2(d).

Health Services recognized that s. 2(d) of the Charter obliged the state, either as employer or as legislator, to protect the process of collective bargaining (para. 88). The content of that protection will of course mean different things in different contexts. The determinative question will inevitably be, as Bastarache J. said in Dunmore, what protections are “essential” to the “meaningful exercise” of the right.

The right at issue in Dunmore was the right to organize. Bastarache J. concluded that this required ancillary protection for the freedom to assemble, to participate in the lawful activities of the “employees’ association” and to make representations, along with the right to be free from interference, coercion and discrimination in the exercise of those freedoms (para. 67). All of these protections found their way into the AEPA, which is why the trial judge gave it his stamp of constitutional approval.

Now, as a result of Health Services, we are dealing with a right to a process of good faith collective bargaining and consultation. What protections are essential for the meaningful exercise of this right for agriculture workers?

For a start, there is no point to having a right only in theory. Unless it is realizable, it is meaningless. There must therefore be an enforcement mechanism not only to resolve bargaining disputes, but to ensure compliance if and when a bargain is made.

At the moment, there is in fact a statutory mechanism in place for the enforcement of the AEPA — the Agriculture, Food and Rural Affairs Appeal Tribunal. But the fact that this Tribunal exists is, by itself, of no consequence if it cannot address the rights constitutionally guaranteed by Health Services.

Section 11 of the AEPA gives the Tribunal authority to grant a remedy for a contravention of the AEPA. But it is not a contravention of the AEPA to refuse to engage in a good faith process to make reasonable efforts to arrive at a collective agreement. It is therefore not part of the Tribunal’s mandate. No mandate, no jurisdiction; no jurisdiction, no remedy.

It strikes me as fundamentally contrary to our jurisprudence to invite the Tribunal to interpret its home statute in a way that contradicts the clear statutory language and legislative intent. If, on the other hand, the AEPA had included the protections set out in Health Services, the Tribunal would certainly have the authority to address and remedy any bargaining disputes and would therefore comply with what is required by s. 2(d).

This brings us finally to whether the process of good faith bargaining for agricultural workers requires that the employer bargain only with the union selected by a majority of the employees in the bargaining unit. This is known as the principle of majoritarian exclusivity, a routine protection in Canada’s labour laws. In the context of this case, and given the unique vulnerability of agriculture workers, I agree with Winkler C.J.O. that statutory recognition of such exclusivity is essential for them to exercise their bargaining rights meaningfully.

As long ago as 1944, when labour ministers from across Canada agreed to the principles which found their way into the model The Industrial Relations and Disputes Investigation Act, S.C. 1948, c. 54, majoritarian exclusivity was a central protection. Most provinces quickly aligned their legislation with these principles (Adams, vol. 1, at paras. 1.240-1.250).

With the exception of specific public services and the construction industry in Quebec (An Act respecting labour relations, vocational training and workforce management in the construction industry, R.S.Q., c. R-20), majoritarian exclusivity has remained a defining principle of the Canadian labour relations model (Rayner, at p. 16; Carter et al., at para. 574).

The reason for the protection is grounded in common sense and the pre-1944 experience. A lack of exclusivity allows an employer to promote rivalry and discord among multiple employee representatives in order to “divide and rule the work force”, using tactics like engaging in direct negotiations with individual employees to undercut “the credibility of the union . . . at the bargaining table” (Paul Weiler, Reconcilable Differences: New Directions in Canadian Labour Law (1980), at p. 126; see also Adams, vol. 1, at para. 3.1750).

Rol-Land Farms, for example, unrestrained by the legal requirement to bargain only with one bargaining agent, sponsored its own “employee association” in direct competition with the union that had the workers’ majority support. That is precisely the kind of conduct that Bora Laskin identified in 1944 as the flaw in Canada’s then existing labour legislation, namely that “it neither compelled employers to bargain collectively with the duly chosen representatives of their employees nor did it prohibit them from fostering company-dominated unions” (“Recent Labour Legislation in Canada” (1944), 22 Can. Bar Rev. 776, at p. 781). It also led Canada’s labour ministers that same year to include exclusivity among what were considered to be indispensable protections for collective bargaining rights.

The inevitable splintering of unified representation resulting from the absence of statutory protection for exclusivity is particularly undermining for particularly vulnerable employees. Professor David M. Beatty vividly observed that agricultural workers are “among the most economically exploited and politically neutralized individuals in our society”:

Because they are heavily drawn from a migrant and immigrant population, these workers face even more serious obstacles to effective participation in the political process. . . . Denying agricultural workers the benefits of [collective bargaining] means that the legal processes which enable much of the rest of our workforce to be involved in decision-making at the workplace in a realistic way are unavailable to the farm workers. Thus a group of workers who are already among the least powerful are given even less opportunity than the rest of us to participate in the formulation and application of the rules governing their working conditions.

(Putting the Charter to Work: Designing a Constitutional Labour Code (1987), at p. 89)

See also Task Force on Labour Relations, Canadian Industrial Relations: The Report of Task Force on Labour Relations (1968) (the “Woods Report”), at p. 86.

These conclusions were echoed by the trial judge in Dunmore, Sharpe J., whose observations were endorsed in this Court by Bastarache J.:

Distinguishing features of agricultural workers are their political impotence, their lack of resources to associate without state protection and their vulnerability to reprisal by their employers; as noted by Sharpe J., agricultural workers are “poorly paid, face difficult working conditions, have low levels of skill and education, low status and limited employment mobility” . . . . [para. 41]

The conditions of singular employment disadvantage for workers in the agricultural sector, as the trial judge in this case acknowledged, remain operative today. Permitting multiple representatives of disparate individuals or groups in such a workplace effectively nullifies the ability of its workers to have a unified and therefore more cogent voice in attempting to mitigate and ameliorate their relentlessly arduous working conditions.

I acknowledge that different models of labour relations exist globally, some of which do not recognize the principle of majoritarian exclusivity (Clyde W. Summers, “Exclusive Representation: A Comparative Inquiry into a ‘Unique’ American Principle” (1998), 20 Comp. Lab. L. & Pol’y J. 47; Roy J. Adams, “Prospects for Labour’s Right to Bargain Collectively After B.C. Health Services” (2009), 59 U.N.B.L.J. 85). These models, however, have been developed in entirely different historical contexts and systems of collective bargaining and have yet to be seriously road-tested in the Canadian context outside of the construction industry in Quebec. This is not to say that there is no room for innovation in the modalities of the Canadian labour relations model. But to “innovate” by eliminating a fundamental protection for the most vulnerable of workers is nullification, not innovation.

Can the absence of these statutory protections be justified under s. 1 of the Charter ? In my view they cannot.

Chief Justice Winkler found that the relevant objectives of the rights limitation — the failure to provide agricultural workers with the necessary statutory protections to exercise the right to bargain collectively — were “to protect the family farm and farm production/viability” (para. 122).

These were found by this Court to be pressing and substantial objectives in Dunmore and are conceded to reach the necessary threshold in our case.

Even assuming that there is a rational connection between at least the second objective and the limitation, I see the minimal impairment branch of the Oakes test as being determinative. Under this step we ask whether there are “less harmful means of achieving the legislative goal” (Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, at para. 53). The AEPA has an absolute exclusion of any protection for a process of collective bargaining: all agricultural workers, in all sectors of agriculture, no matter the size and nature of the agricultural enterprise, are precluded from exercising their s. 2(d) rights. If the government has impeded those rights more than is reasonably necessary to achieve its stated objectives, then this absolute exclusion is not constitutionally justified.

The first governmental objective of the absolute exclusion is the protection of family farms. Is a one-size-fits-all exclusion responsive to protecting family farms? It seems to me clear that less harmful means than outright exclusion are readily available to achieve the objective. Two provinces, for example, Quebec and New Brunswick, have specific exemptions for farms employing less than three (Labour Code, R.S.Q., c. C-27, s. 21) or five (Industrial Relations Act, R.S.N.B. 1973, c. I-4, s. 1(5)(a)) workers.

It is also worth remembering that we are dealing with a highly diversified sector, only some of which consists of family farms. As Bastarache J. noted, there is an “increasing trend . . . towards corporate farming and complex agribusiness” (Dunmore, at para. 62). Rol-Land Farms, for example, is a mushroom farm that employs between 270-300 workers. The nature of this kind of farm, as described in Wellington Mushroom Farm, [1980] O.L.R.B. Rep. 813, does “not differ in any material respect from a typical manufacturing plant” (para. 29). The description in the reasons of Vice-Chairman R. O. MacDowell is telling:

There is no close involvement with the family farm. The production process is not seasonal, but rather, resembles a production cycle. The labour force is neither casual nor transitory. The operation is of considerable size, employing close to 200 employees in a single location with a “factory atmosphere”; and the company is much less economically vulnerable than many other employers to which The Labour Relations Act applies. [para. 25]

Preventing all agricultural workers from access to a process of collective bargaining in order to protect family farms, no matter their size or character, is the antithesis of minimal impairment. Such a limitation harms the s. 2(d) right in its entirety, not minimally.

The other government objective is more general — the protection of the viability of farms and agricultural production. It is instructive to consider the 1992 recommendations of the Ontario government’s consultative Task Force on Agricultural Labour Relations, composed of representatives from the agricultural community, organized labour, farm workers and government (Report to the Minister of Labour (June 1992); Second Report to the Minister of Labour (November 1992)). The Task Force’s recommendations in its two reports are germane not because they should be seen as binding, but because they demonstrate that there are “less harmful means” than an absolute exclusion to achieve the government’s objective of protecting agricultural production and viability.

The Task Force considered whether — and how — agricultural workers should be entitled to bargain collectively, given the unique characteristics of the agricultural sector. It concluded that “all persons employed in agriculture and horticulture” should be able to engage in collective bargaining, including those on family or smaller farms, but in accordance with a separate labour relations scheme that is “sufficiently modified” to reflect the “particular needs” of the agricultural sector (First Report, at pp. 7-8).

The “single most critical issue” raised by farm owners before the Task Force was the “threat of work stoppage” (First Report, at p. 3). In response to this and many other submissions, the Task Force recommended that all forms of work stoppage be prohibited and replaced by a dispute resolution process that:

• emphasizes the preference for negotiated settlements between the parties.

• provides a conciliation and mediation service to assist the parties in reaching a negotiated settlement.

• provides an arbitration process for the final and binding resolution of all outstanding matters between the parties following exhaustion of the negotiation process. [First Report, at p. 10]

It also recommended that there be an Agricultural Labour Relations Act, to be administered by a separate Board (Second Report, at p. 17).

The government adopted these recommendations in the Agricultural Labour Relations Act, 1994, S.O. 1994, c. 6 (“ALRA”). The ALRA included protection for collective bargaining, including exclusivity, but prohibited work stoppages (ss. 3, 10 and 11). The inherent compromise in that legislation is reflected in its preamble:

It is in the public interest to extend collective bargaining rights to employees and employers in the agriculture and horticulture industries.

However, the agriculture and horticulture industries have certain unique characteristics that must be considered in extending those rights. Those unique characteristics include seasonal production, climate sensitivity, time sensitivity, and perishable nature of agriculture and horticulture products, and the need for maintenance of continuous processes to ensure the care and survival of animal and plant life.

The ALRA was repealed in 1995 (c. 1, s. 80). Thereafter, agricultural workers were left only with their pre-existing exclusion from the Ontario Labour Relations Act, 1995, S.O. 1995, c. 1, Sch. A, creating the spark that ignited Dunmore.

And since s. 1 of the Charter directs us to compare how other democratic governments limit a particular right, it is also helpful to look at how other Canadian jurisdictions deal with agricultural workplaces. Except in Alberta, agricultural workers in every province have the same collective bargaining rights as other employees, including exclusivity (Labour Relations Code, R.S.B.C. 1996, c. 244, s. 1; The Trade Union Act, R.S.S. 1978, c. T-17, s. 2; The Labour Relations Act, R.S.M. 1987, c. L10, s. 1; Labour Code (Que.), s. 21; Industrial Relations Act (N.B.), s. 1(5)(a); Trade Union Act, R.S.N.S. 1989, c. 475, s. 2(1); Labour Act, R.S.P.E.I. 1988, c. L-1, s. 7; Labour Relations Act, R.S.N.L. 1990, c. L-1, s. 2(1); Labour Relations Code, R.S.A. 2000, c. L-1, s. 4(2)(e)).

Reviewing the consequences of the near-universality of extending bargaining rights to Canadian agricultural workers, the 1992 Task Force concluded that the availability of the right to bargain collectively in these provinces has not “had a significant negative impact on farm economics” (First Report, at p. 3). This state of national affairs clearly does not preclude the government from offering a s. 1 justification unique to Ontario, but it has not, and perhaps realistically cannot, explain why Ontario’s farming interests are so different as to warrant a complete exclusion rather than less intrusive means of achieving its objectives.

The agricultural sector undoubtedly faces significant economic challenges, but so do many others, in none of which are employees deprived of access to a process of collective bargaining.

The government has therefore not justified why achieving protection for agricultural viability and production requires so uniquely draconian a restriction on s. 2(d) rights. The limitation is, in fact, like its relationship to protecting family farms, not even remotely tailored to meet the government’s objective in a less intrusive way. It is, in fact, not tailored at all. As Winkler C.J.O. concluded:

. . . the legislature made no attempt to engage in a line-drawing exercise . . . to tailor a collective bargaining system that recognizes the challenges facing the agricultural sector. [para. 135]

On the other hand, it bears repeating that the AEPA was designed before Health Services was decided. The government could hardly be expected to tailor its legislation in accordance with a bargaining regime it had neither a duty nor an intention to implement at the time. Nonetheless, the fact is that Health Services intervened and changed the microscope under which the AEPA was scrutinized. And under the new lens, the complete absence of any statutory protection for a process of collective bargaining in the AEPA cannot be said to be minimally impairing of the s. 2(d) right.

I would therefore dismiss the appeal without costs.