Centrale des syndicats du Québec v. Quebec (Attorney General)

Per Abella J. (with McLachlin C.J. and Moldaver, Karakatsanis and Gascon JJ.): Section 38 of the Act infringes s. 15(1) of the Charter.

Per Côté J. (with Wagner, Brown and Rowe JJ.): Section 38 of the Act does not infringe s. 15(1) of the Charter. The appeal should fail at this stage of the analysis.

Per Abella J. (with Moldaver, Karakatsanis and Gascon JJ.): The infringement of s. 15(1) is justifiable under s. 1 of the Charter.

Per McLachlin C.J.: The infringement of s. 15(1) is not justifiable under s. 1 of the Charter.

____________________________

Per Abella, Moldaver, Karakatsanis and Gascon JJ.: Section 38 of the Pay Equity Act violates s. 15(1) of the Charter. The limitation draws a distinction on the basis of sex. All pay equity legislation demonstrably and unarguably creates distinctions based on sex because its goal is to recognize and remedy the discrimination that women have suffered in the way they are compensated in the workforce. This is systemic discrimination premised on the historic economic and social devaluation of women’s work compared to men’s work. The contrary view that the distinction created by the Act was not based on sex, but rather on the absence of male comparator groups in the enterprise, is formal equality, an approach expressly rejected by this Court in Andrews. It erases the sex‑based character of the legislative provisions and obscures the fact that the claimants disproportionately suffer an adverse impact because they are women. And the sex‑based character of the distinction drawn by s. 38 is inescapable. The two categories into which the Act sorts women — women in workplaces with male comparators and those without such comparators — expressly defined by the presence or absence of men in the workplace, are set up to address disparities in pay between men and women. Moreover, these categories single out for inferior treatment the group of women whose pay has, arguably, been most markedly impacted by their gender. So the categories set up by ss. 37 and 38 of the Act draw distinctions based on sex both on their face — that is, by their express terms — and in their impact.

The distinction is discriminatory. The women targeted for the delay suffer the effects of pay discrimination — without a remedy — for the period of the delay. The fact that the legislation did not create pay discrimination is irrelevant. The law has a discriminatory impact because it perpetuates the disadvantage of a protected group through a legislated “denial of access to remedial procedures for discrimination”. Nor does the fact that the Act was intended to help women, attenuate the fact of the breach. Purpose and intention are part of the s. 1 justification analysis. Determining whether there is a breach focuses on the impact of, not motive for, the law. And the impact is clear. Occupational segregation and low wages usually go hand in hand. There is no doubt that the claimants will experience a considerable economic impact as a result of the delay. And the very women singled out for differential and delayed access to pay equity under this scheme may be the very ones most likely to experience its effects disproportionately.

Section 15(2) has no application to this case. Section 15(2) is not a stand‑alone defence to any and all claims brought under s. 15(1). The purpose of s. 15(2) is to save ameliorative programs from the charge of “reverse discrimination”. Reverse discrimination involves a claim from someone outside the scope of intended beneficiaries who alleges that ameliorating those beneficiaries discriminates against him or her. It stands the purpose on its head to suggest that s. 15(2) can be used to deprive the program’s intended beneficiaries of the right to challenge the program’s compliance with s. 15(1). For the government to invoke a s. 15(2) defence, there must first be a claim by a person or group excluded from the program alleging that the exclusion is discriminatory. In this case, the complainants have been expressly included but in a manner they claim has a discriminatory impact. This is not a claim of “reverse discrimination”, it is a claim of discrimination.

The prima facie breach of s. 15(1) is however justified under s. 1 of the Charter. The pressing and substantial objective of the delay caused by s. 38 was to enact a scheme that created an effective remedy for systemic pay discrimination for women working in places where there are no male comparators. The issue was complex and required considerable research and analysis and there was scant policy experience elsewhere to draw on for inspiration. Therefore, the delay in developing and implementing a credible methodology is rationally connected to the objective of creating the possibility of an effective remedy. As to minimal impairment, where, as here, the government introduced, and gave effect to, an entirely new regime, a degree of delay is to be expected. But the government must demonstrate that, in the circumstances, it acted with reasonable diligence. The time frame must be calibrated to the nature and complexity of the issue, but it cannot be indefinite. While close to the line, the record supports the conclusion that Quebec was not unreasonable in the steps it took to keep the delay within reasonable bounds. The Pay Equity Commission in charge of administering the Act, was only created in 1996 and had to develop new strategies for implementation on multiple fronts. Although the delay was serious and regrettable, women in workplaces with no male comparators saw the creation of an effective, coherent remedy for systemic economic discrimination where none had previously existed. The ultimate advantages of Quebec’s expanded approach outweigh the harm. The proportionality test is met.

Per Wagner, Côté, Brown and Rowe JJ.: Section 38 of the Pay Equity Act is valid under s. 15(1) of the Canadian Charter of Rights and Freedoms . The distinction made in s. 38 is not based on sex, because the difference in compensation does not result from the fact that the affected employees are women. Certain of those who are meant to benefit from the Act — employees of enterprises that have no predominantly male job classes — wish to be paid in accordance with the same timetable that applies to others who are also meant to benefit from the Act — employees of enterprises that do have predominantly male job classes. An analysis of the evidence as a whole leads to the conclusion that the basis for the differential treatment affecting the employees in question lies instead in the lack of male comparators in their employers’ enterprises. It cannot be concluded that every distinction drawn by a pay equity statute is necessarily based on sex. Such a conclusion would deprive trial judges of any discretion in their assessment of the evidence and would make the first step in the s. 15(1) analysis irrelevant.

Moreover, the distinction made in s. 38 of the Act does not have a discriminatory impact. The four contextual factors identified in Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497 — (1) the nature of the affected interest, (2) a pre‑existing disadvantage, (3) a correspondence with actual characteristics and (4) the impact on other groups — are relevant to the determination of whether s. 38 is discriminatory.

The importance of the right at issue in this case is not in doubt. For the employees concerned, it is a question of being paid an amount that reflects the fair value of their work and receiving compensation based on a prejudice‑free evaluation that focuses on the objective value of their work. This right is all the more important because women working in enterprises with no male comparators have a significant pre‑existing disadvantage, one that is documented, is not in dispute and is in fact recognized from the outset in s. 1 of the Act. But this factor does not give rise to a presumption that a distinction is discriminatory.

Regarding the third factor, it can be seen that significant differences in compensation due to systemic gender discrimination already existed in the labour market and that these differences were maintained in the private sector. The Act and the time limits it establishes reflect the actual capacities and needs of the members of the affected group. The legislature set up a proactive scheme that, once completed, made it possible to redress differences in compensation due to systemic gender discrimination, including for employees working in enterprises with no male comparators. However, it was recognized from the outset that more time would be needed to develop an adequate method for calculating compensation adjustments. Although the Act may not be perfect, it has had an undeniably ameliorative effect on the employees in question. The decision to enact the Act quickly and to guarantee pay equity for a large number of employees working in more than 225,000 enterprises in Quebec obviously had a significant positive effect on those employees. It goes without saying that when a government develops a complex scheme such as that of pay equity legislation, it will not always be able to ameliorate the conditions of every member of a disadvantaged group at the same time and in the same way. An obligation of result cannot be imposed on the legislature in this regard, as that could cause the enactment of legislation to be postponed, to everyone’s detriment. Furthermore, the legislature was entitled to proceed by way of regulation to develop an innovative, simple and practical method for valuating differences in compensation for employers that have no predominantly male job classes. The different time limits provided for do not perpetuate prejudice or a stereotype. The impugned provision does not have a discriminatory impact, as it instead narrows the gap between these historically disadvantaged groups and the rest of society.

Given that s. 38 of the Act is valid, it is not necessary to go on to determine whether that section can be saved under s. 15(2). This silence should not be interpreted as an endorsement of Abella J.’s comments on that subject, however.

Per McLachlin C.J. (dissenting in the result): There is agreement with the majority that s. 38 of the Pay Equity Act violates s. 15(1) of the Canadian Charter of Rights and Freedoms . However, the breach cannot be justified under s. 1 of the Charter. It is questionable whether the government’s objective of ensuring employer compliance is pressing and substantial and therefore capable of justifying the breach of women’s right to equality. In any event, the infringement fails at the minimal impairment and balancing stages. The six‑year period was dictated not by the exigencies of working out what constituted equal pay in female‑dominated work‑places, where no male comparator groups were available, but in part by the government’s decision to negotiate with employers over a lengthy period in order to ensure that the scheme was one that employers would accept and with which they would comply. The government has not demonstrated that other less impairing options were not available. It has also not established that the denial of benefits to the affected and already marginalized women is proportionate to the public interest in denying them a remedy.

Cases Cited

By Abella J.

Applied: Kahkewistahaw First Nation v. Taypotat, 2015 SCC 30, [2015] 2 S.C.R. 548; Vriend v. Alberta, [1998] 1 S.C.R. 493; referred to: Syndicat de la fonction publique du Québec inc. v. Québec (Procureur général), [2004] R.J.Q. 524; Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61; Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; Bliss v. Attorney General of Canada, [1979] 1 S.C.R. 183; Withler v. Canada (Attorney General), 2011 SCC 12, [2011] 1 S.C.R. 396; Nova Scotia (Attorney General) v. Walsh, 2002 SCC 83, [2002] 4 S.C.R. 325; Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497; Alberta (Aboriginal Affairs and Northern Development) v. Cunningham, 2011 SCC 37, [2011] 2 S.C.R. 670; R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483; RJR‑MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624.

By Côté J.

Applied: Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497; distinguished: Vriend v. Alberta, [1998] 1 S.C.R. 493; referred to: Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381; R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483; Withler v. Canada (Attorney General), 2011 SCC 12, [2011] 1 S.C.R. 396; Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61; Alberta (Aboriginal Affairs and Northern Development) v. Cunningham, 2011 SCC 37, [2011] 2 S.C.R. 670; Kahkewistahaw First Nation v. Taypotat, 2015 SCC 30, [2015] 2 S.C.R. 548; Bliss v. Attorney General of Canada, [1979] 1 S.C.R. 183.

By McLachlin C.J. (dissenting in the result)

R. v. Oakes, [1986] 1 S.C.R. 103.

Statutes and Regulations Cited

Act respecting discrimination in employment, S.Q. 1964, c. 46.

Canadian Charter of Rights and Freedoms , ss. 1 , 15 .

Charter of human rights and freedoms, CQLR, c. C‑12, s. 19.

Charter of human rights and freedoms, S.Q. 1975, c. 6, s. 19.

Equal Pay Act, R.S.B.C. 1960, c. 131.

Equal Pay Act, R.S.S. 1953, c. 265.

Female Employees Equal Pay Act, S.C. 1956, c. 38.

Female Employees Fair Remuneration Act, 1951, S.O. 1951, c. 26.

Individual’s Rights Protection Act, R.S.A. 1980, c. I‑2.

Pay Equity Act, CQLR, c. E‑12.001, ss. 1, 10, 13, 31, 34, 37, 38, 50, 71, 114.

Pay Equity Act, R.S.O. 1990, c. P.7.

Pay Equity Act, S.Q. 1996, c. 43, ss. 1, 13, 37 [am. 2009, c. 9, s. 12], 38, 71, 114.

Regulation respecting pay equity in enterprises where there are no predominantly male job classes, (2005) 137 G.O. II, 976.

Regulation respecting pay equity in enterprises where there are no predominantly male job classes, CQLR, c. E‑12.001, r. 2.

Treaties and Other International Instruments

Convention concerning Equal Remuneration for Men and Women Workers for Work of Equal Value (No. 100), Can. T.S. 1973 No. 37, arts. 2, 3.

Convention on the Elimination of All Forms of Discrimination against Women, Can. T.S. 1982 No. 31.

International Covenant on Economic, Social and Cultural Rights, Can. T.S. 1976 No. 46.

Authors Cited

Canada. Pay Equity Task Force. Pay Equity: A New Approach to a Fundamental Right, Final Report. Ottawa, 2004.

Canada. Royal Commission on Equality in Employment. Report of the Commission on Equality in Employment. Ottawa: Supply and Services Canada, 1984.

Chicha, Marie‑Thérèse. L’équité salariale: Mise en œuvre et enjeux, 3e éd. Cowansville, Que: Yvon Blais, 2011.

Cornish, Mary. “Closing the Global Gender Pay Gap: Securing Justice for Women’s Work” (2007), 28 Comp. Lab. L. & Pol’y J. 219.

International Labour Office. Report of the Committee of Experts on the Application of Conventions and Recommendations, Report III (Part 1A), General Report and observations concerning particular countries. Geneva, 2007.

Martin, Sheilah L. “Persisting Equality Implications of the ‘Bliss’ Case”, in Sheilah L. Martin and Kathleen E. Mahoney, eds., Equality and Judicial Neutrality. Toronto: Carswell, 1987.

Oelz, Martin, Shauna Olney and Manuela Tomei. Equal Pay: An introductory guide, Geneva: International Labour Organization, 2013.

Pothier, Dianne. “Equality as a Comparative Concept: Mirror, Mirror, on the Wall, What’s the Fairest of Them All?” (2006), 33 S.C.L.R. (2d) 135.

Quebec. Assemblée nationale. Commission de l’économie et du travail. “Étude du projet de règlement sur l’équité salariale dans les entreprises où il n’existe pas de catégories d’emplois à prédominance masculine”, Journal des débats de la Commission de l’économie et du travail, vol. 38, no 35, 1re sess., 37e lég., 24 novembre 2004.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultation générale sur l’avant‑projet de loi sur l’équité salariale et modifiant certaines dispositions législatives”, Journal des débats de la Commission des affaires sociales, vol. 34, no 35, 1re sess., 35e lég., 6 février 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultation générale sur l’avant‑projet de loi sur l’équité salariale et modifiant certaines dispositions législatives”, Journal des débats de la Commission des affaires sociales, vol. 34, no 36, 1re sess., 35e lég., 7 février 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultation générale sur l’avant‑projet de loi sur l’équité salariale et modifiant certaines dispositions législatives”, Journal des débats de la Commission des affaires sociales, vol. 34, no 37, 1re sess., 35e lég., 8 février 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultation générale sur l’avant‑projet de loi sur l’équité salariale et modifiant certaines dispositions législatives”, Journal des débats de la Commission des affaires sociales, vol. 34, no 39, 1re sess., 35e lég., 15 février 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultations particulières sur le projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 32, 2e sess., 35e lég., 20 août 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultations particulières sur le projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 33, 2e sess., 35e lég., 21 août 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Consultations particulières sur le projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 34, 2e sess., 35e lég., 22 août 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Étude détaillée du projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 45, 2e sess., 35e lég., 7 novembre 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Étude détaillée du projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 46, 2e sess., 35e lég., 12 novembre 1996.

Quebec. Assemblée nationale. Commission des affaires sociales. “Étude détaillée du projet de loi no 35 — Loi sur l’équité salariale”, Journal des débats de la Commission des affaires sociales, vol. 35, no 47, 2e sess., 35e lég., 14 novembre 1996.

Quebec. Assemblée nationale. Journal des débats de l’Assemblée nationale, vol. 35, no 56, 2e sess., 35e lég., 21 novembre 1996.

Young, Margot. “Blissed Out: Section 15 at Twenty” (2006), 33 S.C.L.R. (2d) 45.

APPEAL from a judgment of the Quebec Court of Appeal (Vézina, St‑Pierre and Émond JJ.A.), 2016 QCCA 424, [2016] AZ‑51262412, [2016] J.Q. no 1991 (QL), 2016 CarswellQue 1852 (WL Can.), affirming a decision of Yergeau J., 2014 QCCS 4197, [2014] AZ‑51105800, [2014] J.Q. no 9467 (QL), 2014 CarswellQue 9184 (WL Can.). Appeal dismissed, McLachlin C.J. dissenting in the result.

Geneviève Bailargeon Bouchard and Denis Bradet, for the appellants Centrale des syndicats du Québec, Fédération des intervenantes en petite enfance du Québec (FIPEQ‑CSQ), Le syndicat des intervenantes en petite enfance de Montréal (SIPEM‑CSQ), Le syndicat des intervenantes en petite enfance de Québec (SIPEQ‑CSQ), Le syndicat des intervenantes en petite enfance de l’Estrie (SIPEE‑CSQ), Francine Joly, Nathalie Fillion, Louise Fréchette, Fédération du personnel de soutien de l’enseignement supérieur (FPSES) (CSQ), Syndicat des interprètes professionnels du Sivet (CSQ), Chantal Bousquet and Yannick François.

Jean Mailloux and Marilyne Duquette, for the appellants the Confederation of National Trade Unions (CNTU), Fédération de la santé et des services sociaux, Syndicat des travailleuses et travailleurs des CPE de la Montérégie, Syndicat des travailleuses des CPE de Montréal et de Laval, France Laniel, Ginette Lavoie and Danielle Paré.

Patrice Claude and Caroline Renaud, for the respondent.

Courtney Harris and S. Zachary Green, for the intervener the Attorney General of Ontario.

Fay Faraday and Janet E. Borowy, for the interveners the Equal Pay Coalition, the Women’s Legal Education and Action Fund and the New Brunswick Coalition for Pay Equity.

The reasons of Abella, Moldaver, Karakatsanis and Gascon JJ. were delivered by

[1] Abella J. — The issue in this appeal is whether Quebec’s approach to implementing pay equity in workplaces without male comparator job classifications violates the equality rights protected by s. 15 of the Canadian Charter of Rights and Freedoms . The complaint is not with Quebec’s proposed methodology, but with the six-year legislated delay in implementation.

[2] The first efforts to address the deep and persistent gap between women’s pay and men’s pay, referred to as “equal pay for equal work”, required employers to pay men and women the same wage for doing the same job, at the same workplace (see, e.g., Female Employees Equal Pay Act, S.C. 1956, c. 38; Equal Pay Act, R.S.B.C. 1960, c. 131; The Equal Pay Act, R.S.S. 1953, c. 265). Although this was an important first step, it failed to grapple with the systemic nature of pay discrimination. It became clear that “historical attitudes towards the role of women in society, along with stereotypical assumptions regarding women’s aspirations, preferences, capabilities, and ‘suitability’ for certain jobs, [had] contributed to occupational sex segregation in the labour market” (International Labour Office, Report of the Committee of Experts on the Application of Conventions and Recommendations (2007), at p. 271).

The result was that women doing work of equal value did not receive fair pay:

Female-dominated jobs . . . are generally less paid and less valued than male-dominated jobs. The lower rates of pay discourage men from entering these jobs. As a result women are concentrated in different jobs than men. This reinforces the view that low pay results from market factors and skill requirements rather than the under-valuation of women’s jobs. Women’s skills are often overlooked, as they are regarded as “natural” female characteristics rather than acquired through experience or training.

(Martin Oelz, Shauna Olney and Manuela Tomei, Equal Pay: An introductory guide (2013), at pp. 17-18)

The next responsive legislative step was a right to equal pay for work of equal value. Under this approach, methods were developed for measuring and remedying pay disparities between male-dominated jobs and female-dominated jobs within the same workplace or “establishment”. These methods were based on identifying job classes in an enterprise with similar characteristics and value, and then classifying them as male or female dominated (Final Report of the Pay Equity Task Force, Pay Equity: A New Approach to a Fundamental Right (2004), at p. 247).

In Quebec, this right to equal pay for work of equal value was adopted in 1975 through s. 19 of the Charter of Human Rights and Freedoms, S.Q. 1975, c. 6 (now CQLR, c. C-12). This provision states:

19. Every employer must, without discrimination, grant equal salary or wages to the members of his personnel who perform equivalent work at the same place.

A difference in salary or wages based on experience, seniority, years of service, merit, productivity or overtime is not considered discriminatory if such criteria are common to all members of the personnel.

. . .

It allowed for comparisons in the same workplace between predominantly male job categories and predominantly female categories even though the nature of the work performed was not the same. But the complaints-based regime by which female workers had access to s. 19 proved ill-suited to address systemic discrimination.[1] Since s. 19 required male comparator groups “at the same place”, female workers were excluded from access to pay equity where there were no male comparators.

Fixing this gap required a determination of how to measure pay discrimination against women who were not only in female dominated jobs, but whose workplaces contained no male dominated job with which to do a comparison. This appeal arises out of Quebec’s efforts to find — and adopt — a solution to that problem.

The Act recognized that systemic wage discrimination exists in the workforce whether or not there are male comparators in a particular workplace. Its purpose was to provide a remedy for this reality:

1. The purpose of this Act is to redress differences in compensation due to the systemic gender discrimination suffered by persons who occupy positions in predominantly female job classes.

Differences in compensation are assessed within the enterprise, except if there are no predominantly male job classes in the enterprise.

All employers with 10 or more employees were to conduct a pay equity exercise to identify and redress pay inequity through adjustments in compensation or the creation of a pay equity plan, depending on the size of the enterprise.

At the time the Act came into force, there was, however, no methodology for assessing pay equity adjustments where there was no male comparator. The 1995 reports on which the Act was based acknowledged the unique enforcement issues for such workplaces, but offered no precise methodology. Nor did any concrete proposals emerge as a result of the consultations leading to the implementation of the Act. The only clear consensus was the acknowledgment that the necessary male comparators would have to be found outside the particular workplace.

Regulatory authority was therefore given to the Pay Equity Commission under s. 114 to conduct the necessary research and to establish a methodology for identifying the appropriate male comparators. Section 114 states:

114. The Commission may make regulations

(1) for the purposes of the determination of adjustments in compensation in an enterprise employing fewer than 50 employees where there are no predominantly male job classes, determining typical job classes on the basis of job classes identified in enterprises in which adjustments in compensation have already been determined and prescribing standards or weighting factors to be applied to the valuation of differences in compensation between such job classes, with due regard, in particular, for the characteristics of enterprises whose job classes are to be so compared;

(2) for the purposes of the establishment of a pay equity plan in an enterprise where there are no predominantly male job classes, determining typical job classes on the basis of job classes identified in enterprises, in which a pay equity plan has already been completed, prescribing methods to be used to determine the value of those job classes and to valuate the differences in compensation between the typical job classes and the job classes in an enterprise and prescribing standards or weighting factors to be applied to such differences, with due regard, in particular, for the characteristics of enterprises whose job classes are to be so compared;

. . .

Regulations of the Commission are subject to the approval of the Government and may be amended by the Government upon approval.

No regulation of the Commission may be approved by the Government until it is examined by the appropriate committee of the National Assembly.

This section directed the Commission to develop a proposed methodology based on two criteria to bring it into line with s. 19 of the Quebec Charter: it was to draw comparisons with male job classifications found in enterprises where a pay equity exercise had already been completed; and it was to factor in the characteristics of the enterprises whose job classes were to be compared.

Section 38 of the Act sets out two possible deadlines by which those employers in workplaces with no male comparators were to determine adjustments or complete a pay equity plan:

38. In an enterprise where there are no predominantly male job classes, the adjustments in compensation must be determined or the pay equity plan must be completed either within the time limit set out in section 37 or within two years of the coming into force of the regulation made by the Commission under subparagraph 1 or 2 of the first paragraph of section 114, whichever time limit expires last.

But this first deadline for workplaces with no male comparators was illusory, since there could be no pay equity for those workplaces until a Regulation was passed. And, under s. 114, a regulation could not be passed until workplaces with male comparators had completed their first pay equity exercise on or before November 21, 2001.

The only realistically possible deadline under the legislation was the second one, namely within two years of the coming into force of the Regulation. The Pay Equity Commission did not settle on a methodology until 2003 and the Regulation respecting pay equity in enterprises where there are no predominantly male job classes, (2005) 137 G.O. II, 976 (now CQLR, c. E-12.001, r. 2), was not promulgated until May 5, 2005. The two-year grace period provided by s. 38 further postponed pay equity for workplaces without male comparators until May 5, 2007.

As a result, access to pay equity under the Act for women in workplaces without male comparators was delayed, in accordance with s. 38, by two years beyond the length of time it took to enact the Regulation under s. 114. By this time, almost six years had passed since women in workplaces that had male comparators first gained access to pay equity.

This six-year legislatively delayed access to pay equity resulting from s. 38 of the Act was challenged by several unions, who argued that it amounted to a breach of s. 15(1) of the Charter for women in workplaces without male comparators.

The trial judge found that the delay did not violate s. 15 (2014 QCCS 4197). In his view, a separate timeline did not engage s. 15 because the delayed access was not based on sex, but rather on the absence of a male comparator. And if there was a distinction based on sex, it was not discriminatory because it was not based on the premise that employees in female-dominated workplaces were [translation] “less capable or . . . worthy of recognition or value as human beings” (para. 244 (CanLII)). The Court of Appeal of Quebec dismissed the appeal (2016 QCCA 424).

Section 15(1) of the Charter states:

Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

When assessing a claim under s. 15(1), this Court’s jurisprudence establishes a two-step approach: Does the challenged law, on its face or in its impact, draw a distinction based on an enumerated or analogous ground, and, if so, does it impose “burdens or [deny] a benefit in a manner that has the effect of reinforcing, perpetuating or exacerbating . . . disadvantage”, including “historical” disadvantage? (Quebec (Attorney General) v. A, [2013] 1 S.C.R. 61, at paras. 323-24 and 327; Kahkewistahaw First Nation v. Taypotat, [2015] 2 S.C.R. 548, at paras. 19-20.)

The first question is therefore whether the limitation challenged by the unions — a six-year pay equity delay mandated by s. 38, for women employed in workplaces without male comparators — draws a distinction on the basis of sex.

In my view, it does. The goal of pay equity legislation is to recognize and remedy the discrimination that women have suffered in the way they are compensated in the workforce. This is systemic discrimination premised on the historic economic and social devaluation of “women’s work” compared to “men’s work” (Report of the Commission on Equality in Employment (1984), at p. 232; Final Report of the Pay Equity Task Force (2004), at pp. 25-27). Accordingly, pay equity legislation, including the Act at issue here, draws a distinction based on sex in targeting systemic pay discrimination against women. And, as explained later in these reasons, the specific provisions of the Act that target particular groups of women based on where they work — such as s. 38 — also necessarily draw a distinction based on sex.

The contrary view adopted by the trial judge, the Court of Appeal, and my colleague, is that the distinction created by the Act was based not on sex, but on the absence of male comparator groups in the enterprise. This is “formal equality”, an approach expressly rejected by this Court in Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143, where the Court refused to apply a rigid Diceyan analysis and declared, instead, that substantive equality is the premise underlying s. 15.

Bliss was decided under s. 1 (b) of the Canadian Bill of Rights , S.C. 1960, c. 44 . Its “similarly situated” “treating likes alike” approach — resulting in the conclusion that pregnancy-related discrimination is not based on sex because not all women are pregnant — was also rejected in Withler v. Canada (Attorney General), [2011] 1 S.C.R. 396:

. . . a formal analysis based on comparison between the claimant group and a “similarly situated” group, does not assure a result that captures the wrong to which s. 15(1) is directed . . . . What is required is not formal comparison with a selected mirror comparator group, but an approach that looks at the full context, including the situation of the claimant group and . . . the impact of the impugned law . . . . [para. 40]

(See also, Quebec v. A, at para. 346, rejecting the analysis in Nova Scotia (Attorney General) v. Walsh, [2002] 4 S.C.R. 325.)

In this case, the trial judge’s analysis falls into precisely the same error as Bliss by holding, in effect, that the distinction created by s. 38 is not based on sex because not all women are denied timely access to the scheme, only those in certain workplaces. The relevant distinction created by s. 38 of the Act is between male employees and underpaid female employees, whether or not those male employees are in the same workplace. The trial judge’s approach erases the sex-based character of the legislative provisions and obscures the fact that the claimants disproportionately suffer an adverse impact because they are women (Taypotat, at para. 21).

And the sex-based character of the distinction drawn by s. 38 is inescapable. That is because the two categories into which the Act sorts women — women in workplaces with male comparators, under s. 37, and those without such comparators, under s. 38 — are themselves inextricably related to sex. Only if we ignore the gender-driven bases for the two categories can it be said that the distinction is based only on workplace and not on sex. Not only are both categories expressly defined by the presence or absence of men in the workplace, but, more fundamentally, both categories are set up to address disparities in pay between men and women. Moreover, since women in workplaces without male comparators may suffer more acutely from the effects of pay inequity precisely because of the absence of men in their workplaces, these categories single out for inferior treatment the group of women whose pay has, arguably, been most markedly impacted by their gender. So the categories set up by ss. 37 and 38 of the Act draw distinctions based on sex both on their face — that is, by their express terms — and in their impact.

The second question is whether that distinction is a discriminatory one, that is, whether it imposes burdens or denies benefits in a way that reinforces, perpetuates, or exacerbates disadvantage.

The discriminatory impact of the delay is clear. The women targeted for the delay created by s. 38 (those in workplaces with no male comparators), suffer the effects of pay discrimination — without a remedy — for the period of the delay, in this case six years.

The fact that the legislation did not create pay discrimination is irrelevant. In Vriend v. Alberta, [1998] 1 S.C.R. 493, this Court held that “[i]t is not necessary to find that the legislation creates the discrimination existing in society” to find a s. 15 breach (para. 84 (emphasis in original)). There, as here, the law perpetuated the disadvantage of a protected group through a legislated “denial of access to remedial procedures for discrimination”, remedial procedures that “they so urgently need because of the existence of discrimination against them in society” (Vriend, at paras. 97 and 84).

The legislature chose to act to address pay discrimination against women, but denied access by delaying it for a group of women, leaving them, in comparison to male workers, paid less for longer. Whatever the motives behind the decision, this is “discrimination reinforced by law”, which this Court has denounced since Andrews (p. 172). The fact, then, that women in one type of workplace — with male comparators — received a remedy promptly is not an answer to the question of whether women in another type of workplace were also disadvantaged. It is no defence to a claim of discrimination by one group of women to suggest that another group has had its particular discrimination addressed.

There is no doubt that the claimants in this case will experience a considerable economic impact as a result of the delay. Occupational segregation and low wages “usually go hand in hand” (Final Report of the Pay Equity Task Force (2004), at p. 15). “[H]istorical attitudes towards the role of women” tend to “result in the undervaluation of ‘female jobs’ in comparison with those of men . . . when determining wage rates” (Report of the Committee of Experts on the Application of Conventions and Recommendations (2007), at p. 271). Put simply, “the more women are concentrated in a field of work, the less it pays” (Mary Cornish, “Closing the Global Gender Pay Gap: Securing Justice for Women’s Work” (2007), 28 Comp. Lab. L. & Pol’y J. 219, at pp. 224-25). And enterprises that employ mainly women tend to have lower pay rates (Oelz et al., at p. 18). So, as previously noted, the very women singled out for differential and delayed access to pay equity under this scheme may be the very ones most likely to experience its effects in a more pronounced way.

The fact that the Act was intended to help these women does not attenuate the fact of the breach. Purpose and intention are part of the s. 1 justification analysis. Determining whether there is a breach focuses on the impact of, not motive for, the law (Andrews, at pp. 181-82; Quebec v. A, at para. 333). As a result, the trial judge’s finding that the delay was not based on the assumption that employees in female segregated workplaces were “less capable or . . . worthy of recognition or value as human beings or as members of Canadian society” (Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497, at para. 99) cannot sanitize the discriminatory impact (Quebec v. A, at paras. 244, 327 and 330).

The impact of the delay created by s. 38, therefore, reinforced and perpetuated the historic economic disadvantage experienced by women in workplaces without male comparators, amounting to a prima facie breach of s. 15(1) of the Charter.

Quebec suggests that s. 15(2) could apply in this case to immunize s. 38 from the unions’ challenge under s. 15(1). In my respectful view, s. 15(2) has no application whatever to this case. It states:

Subsection (1) does not preclude any law, program or activity that has as its object the amelioration of conditions of disadvantaged individuals or groups including those that are disadvantaged because of race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

The purpose of s. 15(2) is to “save ameliorative programs from the charge of ‘reverse discrimination’” (Alberta (Aboriginal Affairs and Northern Development) v. Cunningham, [2011] 2 S.C.R. 670, at para. 41; R. v. Kapp, [2008] 2 S.C.R. 483). Reverse discrimination involves a claim from someone outside the scope of intended beneficiaries who alleges that ameliorating those beneficiaries discriminates against him. It stands the purpose on its head to suggest that s. 15(2) can be used to deprive the program’s intended beneficiaries of the right to challenge the program’s compliance with s. 15(1). And it would be equally inconsistent with the purpose to suggest that legislation that has a discriminatory impact on a scheme’s intended beneficiaries, can “serve” or be “necessary to” any ameliorative purpose in the sense intended by this Court in Cunningham.

Section 15(2) is not a stand-alone defence to any and all claims brought under s. 15(1). The purpose of s. 15(2) and the reasoning underlying this Court’s decision in Kapp dictate that, for the government to invoke a s. 15(2) defence, there must first be a claim by a person or group excluded from the program alleging that the exclusion is discriminatory. In this case, there is no such claim. The complainants in this case are women in workplaces without male comparators. They have not been excluded from the scheme. On the contrary, they have been expressly included but in a manner they claim has a discriminatory impact. Indeed, Quebec’s argument against a finding of discriminatory impact in this case is predicated on the fact that the scheme is intended to benefit women in workplaces without male comparators. And s. 1 of the Act leaves no doubt that the scheme aims to benefit “persons who occupy positions in predominantly female job classes”, without any qualifiers, and its second paragraph makes it clear that the claimants were always part of that target group.

Even if a pay equity statute could be characterized as an “affirmative action” or “employment equity” program anticipated by s. 15(2), a highly debatable proposition, the women who have brought the s. 15(1) claim in this case are the very beneficiaries of this scheme. They are not arguing that they have been excluded from the scheme, they are arguing that the timeline in the scheme applicable to them has a discriminatory impact. This is not a claim of “reverse discrimination”, it is a claim of discrimination, period.

The Act as a whole is, by its express terms, intended to apply to and benefit women in two kinds of workplaces — where male comparators exist, and where they do not. Different timelines have been set out for each. The claim of discrimination arises not from the difference in the timelines, which is not the focus of the complaint, but from the extent of the delay resulting from the timeline applicable to the claimants. To suggest that having different timelines means that one of those groups is not within the class of “intended beneficiaries” is, with respect, a misreading of the legislative scheme and of the concept of “reverse discrimination”.

This brings us to s. 1 of the Charter. Quebec bears the burden of justifying the prima facie infringement of s. 15(1) by identifying a pressing and substantial objective and showing that, in furtherance of that objective, it did not disproportionality interfere with the Charter right. The first step is to identify the objective of the infringing measure, namely the legislatively mandated delay in pay equity in workplaces where there are no male comparators. This delay occurred because of the combined effect of ss. 38 and 114 of the Act.

The essence of Quebec’s argument is that the pressing and substantial objective was to enact a scheme that created an effective remedy for systemic pay discrimination for women who were outside the existing pay equity scheme. The delay was, it argued, necessary to finding the right approach. I see no reason for refusing to accept this as the objective of the delay caused by s. 38.

The next question is whether the delay is rationally connected to the purpose. The Pay Equity Commission needed time to develop a methodology, and the resulting regulation had to move through the necessary stages of executive and legislative approval. The issue was complex and required considerable research and analysis. The 1995 reports commissioned by the government gave no guidance to Quebec on how to measure gender-based inequity in private enterprises lacking male comparators. Nor did any solution emerge as a result of the public consultations held during the legislative process. Moreover, there was scant policy experience elsewhere to draw on for inspiration. When the Act was adopted, there was little guidance from other Canadian jurisdictions for extending the benefits of pay equity legislation to female employees in workplaces without male comparators in the private sector. The closest equivalent in terms of coverage was Ontario’s Pay Equity Act, R.S.O. 1990, c. P.7, but it applied only to workplaces in the “public” sector. I therefore accept that the delay in developing and implementing a credible methodology is rationally connected to the objective of creating the possibility of an effective remedy.

As to minimal impairment, the question is whether the delay that resulted from the legislation impaired equality rights as little as reasonably necessary to create the possibility of an effective remedy.

Quebec was seeking a wide-ranging and pioneering approach to pay equity for women in workplaces without male comparators. Where, as here, the government introduced, and gave effect to, an entirely new regime, a degree of delay is to be expected. Minimal impairment requires the government to narrowly tailor limitations to its pressing and substantial objective, but, as McLachlin J. held in RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199:

The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement . . . . [para. 160]

In other words, governments enacting multi-faceted remedial regimes to protect constitutional rights should be given some degree of latitude to accomplish the objectives sought through these initiatives (RJR-MacDonald, at para. 135; Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927, at pp. 993-94; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624).

But the government must demonstrate that, in the circumstances, it acted with reasonable diligence. The time frame must be calibrated to the nature and complexity of the issue, but it cannot be indefinite.

The fact that so many years elapsed between the adoption of the Act and the promulgation of the Regulation is, on its face, troubling. The trial judge reviewed in considerable detail the reasons for the delay and found that it was not caused by a lack of diligence on Quebec’s part. There was ample evidence to support that factual finding. In this case, a significant cause of the delay was that the Pay Equity Commission, the agency in charge of administering the Act, was itself only created in 1996 and was tasked with numerous new duties. The move in 1996 from a complaints-based regime under s. 19 of the Quebec Charter to a proactive pay equity scheme meant that the Commission had to develop new strategies for implementation on multiple fronts.

As the trial judge found, the first proposal for a methodology to the Minister of Labour came in 2002 and was the result of considerable effort from the Pay Equity Commission. In the end, it was rejected because it did not conform to the conditions in s. 114. The Commission’s second proposal in 2004 to the Minister of Labour ultimately proved successful, and was followed closely by the promulgation of the Regulation in 2005. While close to the line, in my view the record supports the conclusion that Quebec was not unreasonable in the steps it took to keep the delay within reasonable bounds.

And I would make the same observation regarding the unions’ complaints about the two-year grace period given to employers under s. 38. Just as the scheme contemplated a regrettable but necessary period of delay while a viable methodology was developed and approved, it also contemplated a two-year period for employers to determine how to implement the new regime in their enterprises. Quebec struck an acceptable balance between the scheme’s objective in ensuring compensation, and the logistical complexity of implementing a new system. Again, although close to the line, in my respectful view the two-year grace period did not bring the length of the delay outside reasonable bounds.

The final step is proportionality, balancing the harm of the delay created by s. 38 against the benefit of waiting for the implementation of the regime.

The negative consequences of the delay are clear: access to pay equity was postponed until May 5, 2007. In the time that elapsed between the adoption of the Act and the end of the two-year grace period, women in workplaces without male comparators continued to experience pay inequity.

In my respectful view, however, the ultimate advantages of Quebec’s expanded approach outweigh the harm. Women in workplaces with no male comparators saw the creation of an effective, coherent remedy for systemic economic discrimination where none had previously existed. The delay was serious and regrettable, but had the long-term benefit of resulting in the Act’s meaningful ability to address pay discrimination for a previously excluded group of female employees.

Nonetheless, the unions argue that Quebec cannot satisfy the proportionality test because of the absence of retroactive payments covering the period of the delay. Quebec chose to make compensation available from the point in time at which a viable methodology had been developed. This represented a proper balancing of interests. On the one hand, the scheme was designed to make employers responsible for paying compensation once they had the tools to identify and measure pay inequity. On the other, employers would not be responsible for compensation before those tools were developed. As this Court held in RJR-MacDonald, proportionality does not require perfection, nor does legislation fail the proportionality test just because the claimants can “conceive of an alternative which might better tailor objective to infringement” (para. 160). The test is whether the balance Quebec struck falls within the range of “reasonable alternatives” (para. 160) and results in greater benefit than harm. In my view, it does.

The prima facie breach is therefore justified under s. 1.

I would dismiss the appeal with costs.

English version of the reasons of Wagner, Côté, Brown and Rowe JJ. delivered by

Côté J. —

As Binnie J. noted in Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381, at para. 30, and as the trial judge reiterated in the case at bar (2014 QCCS 4197, at para. 17 (CanLII)), “[p]ay equity has been one of the most difficult and controversial workplace issues of our times.” This appeal provides a good illustration of how difficult it is to achieve such an objective.

For many years, the situation of women in the workplace was characterized by significant inequality resulting from prejudice, stereotyping and discrimination. Such discrimination was described as systemic, because it [translation] “is cloaked in normality, so deeply imprinted on our attitudes and thinking is the gender‑based division of social roles” (Journal des débats de la Commission des affaires sociales, vol. 34, No. 35, 1st Sess., 35th Leg., February 6, 1996, at p. 1).

This problem was magnified by occupational segregation. Women were often concentrated in a narrow range of occupations with characteristics commonly associated with female job classes. The result was that these job classes were devalued and that such employees were often underpaid.

Conscious of this problem, the Quebec legislature enacted the Pay Equity Act, S.Q. 1996, c. 43 (now CQLR, c. E‑12.001) (“Act”), on November 21, 1996. The purpose of the Act, which is applicable to both the public sector and the private sector, is to “redress differences in compensation due to the systemic gender discrimination suffered by persons who occupy positions in predominantly female job classes” (s. 1).

An employer whose enterprise has 50 or more employees must establish, in accordance with the Act, a pay equity plan to redress differences in compensation due to discrimination (ss. 10, 31 and 50).

As a general rule, employers that were subject to the Act when it came into force on November 21, 1997 had to determine the required compensation adjustments and make the necessary payments within four years, that is, by November 21, 2001 (ss. 37 and 71). If an employer failed to do so within this time limit, the unpaid adjustments bore interest at the legal rate from that date (s. 71 para. 2).

At the time the Act was enacted, the Minister responsible for the Status of Women described it as an implementation statute (Journal des débats de l’Assemblée nationale, vol. 35, No. 56, 2nd Sess., 35th Leg., November 21, 1996, at p. 3306), because its purpose was to uphold the right to pay equity already provided for in the Charter of human rights and freedoms, S.Q. 1975, c. 6 (now CQLR, c. C‑12) (“Quebec Charter”):

19. Every employer must, without discrimination, grant equal salary or wages to the members of his personnel who perform equivalent work at the same place.

. . .

In enacting the Act, the legislature chose to go further than the Quebec Charter and to extend the application of the Act to employees working in enterprises in which it was impossible to make comparisons with predominantly male job classes. Indeed, such employees could not benefit from s. 19 of the Quebec Charter, since comparisons under that section were limited to employees “who perform equivalent work at the same place”. Although there were of course male employees in the enterprises in question, their compensation was also affected by systemic gender discrimination, because they worked in predominantly female occupations. A comparison with the compensation of male co‑workers would therefore be of no assistance in ensuring that the enterprise’s female employees were paid compensation that was free of discrimination.

Furthermore, under s. 19 of the Quebec Charter, an employee had to file a complaint and prove that his or her remuneration was discriminatory. This individual complaint scheme proved inadequate to address a systemic issue like that of pay equity. In the Act, a proactive scheme was therefore established under which the employer was required to review its own situation and ensure that the compensation it paid was free of prejudice and discrimination.

Quebec broke new ground when it enacted the Act, becoming the first province to include private sector enterprises without predominantly male job classes in the scope of pay equity legislation. However, a method for determining the adjustments required in such enterprises remained to be developed. One thing is clear from the reports of expert panels and numerous consultations: no concrete solution had been proposed for determining the compensation adjustments required in these enterprises.

The legislature thus opted to proceed by way of regulation, as had been recommended by most of the experts and stakeholders who were consulted. This meant that the Act could be passed quickly so as to guarantee pay equity for most employees. The Commission de l’équité salariale (“Commission”) (now the Commission des normes, de l’équité, de la santé et de la sécurité du travail) could then establish a solution by way of a regulation that would be submitted to the government to be examined and then made (s. 114 of the Act). The employers concerned would then have to determine and pay the required compensation adjustments within two years of the coming into force of the regulation (s. 38 of the Act).

It took just over five years for the solution in question to be developed and adopted. There were several reasons for this delay, as I will explain below. For one thing, the Act required the Commission to base its solution on job classes in enterprises in which a pay equity plan had already been completed. The Commission therefore had to wait for the completion of an initial pay equity exercise before it could start developing the regulation. Moreover, because of the complexity of the problem, the Commission had to appoint a consulting firm, produce documents and reports, and hold various consultations. The Regulation respecting pay equity in enterprises where there are no predominantly male job classes, (2005) 137 G.O. II, 977 (now CQLR, c. E‑12.001, r. 2), finally came into force on May 5, 2005, and the employees to whom it applied were entitled to compensation adjustments as of May 5, 2007.

The appellants represent employees working for enterprises in which there are no predominantly male job classes, childcare centres in particular. Under ss. 13, 38 and 114 of the Act, such employees had to wait until May 5, 2007 to obtain compensation adjustments, whereas employees of enterprises that had male comparators were entitled to adjustments as of November 21, 2001.

The appellants argue that, because the compensation adjustments were not retroactive to November 21, 2001, this delay was discriminatory, and contrary to s. 15(1) of the Canadian Charter of Rights and Freedoms (“Charter ”). They add that it cannot be justified under s. 1, because it did not have a pressing and substantial objective and was not proportional having regard to its effects.

The legislative provision the appellants are challenging is s. 38 of the Act, which provides that the payment of adjustments to employees working in enterprises with no predominantly male job classes would depend on the date when the regulation was made. The appellants submit that s. 38 is unconstitutional because the legislature did not provide that the compensation adjustments would be retroactive should the regulation’s coming into force date cause a delay in relation to the adjustments granted to employees working in enterprises with predominantly male job classes. Thus, what is in issue is the absence of retroactivity, not a failure on the government’s part to act promptly to make its regulation under s. 114.

The appellants ask this Court to declare s. 38 of the Act to be invalid and of no force or effect. The time limit provided for in s. 37 would therefore apply to the employees they represent, who could then obtain compensation adjustments retroactive to November 21, 2001.

The respondent submits that s. 38 of the Act does not create a discriminatory distinction based on an enumerated or analogous ground under s. 15(1) of the Charter. Even if it did, which the respondent denies, the distinction would be saved by s. 15(2) of the Charter. The respondent also argues that s. 38 is a reasonable limit prescribed by law that can be demonstrably justified under s. 1 of the Charter.

Following an exhaustive review of the evidence, Yergeau J. of the Quebec Superior Court, in a very well‑reasoned judgment, concluded that s. 38 of the Act should not be declared to be invalid, and dismissed the appellants’ application. He found on a balance of probabilities that the distinction created by that section was based not on sex, but on the fact that the enterprises in question had no predominantly male job classes that would allow for comparison. The delay had been caused by difficulties flowing from this reality, not by any negligence or failure to act on the government’s part (paras. 137‑38, 198 and 242).

Yergeau J. identified a number of factors that accounted for the delay in developing a method for calculating the necessary compensation adjustments for employees holding positions in enterprises with no male comparators:

[translation] A variety of factors, including (a) the delay in implementing equity in [enterprises with male comparators] as in the government itself, (b) the novel character of the application of equity to [enterprises with no male comparators], (c) a lack of Canadian legislative precedents, (d) the very structure of section 114 of the Act, (e) the scheme of the [Act], under which pay equity is not meant to become an exercise in levelling compensation by importing salary structures from one enterprise to another, (f) a lack of concrete suggestions from labour organizations and certified associations, and (g) the 2003 decision not to amend section 114 of the Act for strategic reasons, caused the Regulation respecting pay equity to be delayed. [para. 137]

Yergeau J. also found that s. 38 of the Act did not have a discriminatory impact, as it did not perpetuate prejudice or a stereotype. On the contrary, the legislature had recognized from the outset the systemic discrimination that affected employees holding positions in enterprises with no male comparators, and had chosen to confer the protection of the Act on this group that previously had no remedy under s. 19 of the Quebec Charter. He argued that this was the opposite of prejudice.

Section 38 of the Act was therefore valid and consistent with s. 15(1) of the Charter.

The Quebec Court of Appeal, per Vézina, St‑Pierre and Émond JJ.A., upheld the trial judge’s decision and expressed complete agreement with his reasons (2016 QCCA 424).

The main issue is whether, because the compensation adjustments were not to be paid retroactively, the additional time that was needed to establish a method of calculation to be applied by enterprises with no male comparators in order to implement pay equity violated the right to substantive equality guaranteed by s. 15(1) of the Charter.

As McIntyre J. noted in Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143:

. . . the greatly expanded role of women in all forms of industrial, commercial and professional activity led to much inequality and many forms of discrimination. [p. 172]

Women historically faced disadvantages in the labour market, both in hiring and in the performance of their duties. These disadvantages were reflected in part in their conditions of employment and in their remuneration, which was often less than that of their male co‑workers: this constituted systemic discrimination. Thus, less value was attached to work associated with women than to work associated with men even where, objectively, the work was of equal value.

In 1951, in an effort to remedy this problem, the International Labour Organization adopted the Convention concerning Equal Remuneration for Men and Women Workers for Work of Equal Value (No. 100), Can. T.S. 1973 No. 37, which laid down the fundamental principle of equal pay for work of equal value (art. 2). In this convention, each member state undertook to take measures to ensure the objective appraisal of all jobs on the basis of the work to be performed (art. 3) rather than of prejudice. Canada ratified the convention in 1972, and it is also a party to several other international instruments that reaffirm this principle, including the International Covenant on Economic, Social and Cultural Rights, Can. T.S. 1976 No. 46, adopted by the United Nations General Assembly (accession in 1976) and the Convention on the Elimination of All Forms of Discrimination against Women, Can. T.S. 1982 No. 31 (ratification in 1981).

In Canada, Ontario was the first province to enact pay equity legislation, which it did in 1951 with The Female Employees Fair Remuneration Act, 1951, S.O. 1951, c. 26. Quebec followed suit in 1964 in enacting the Act respecting discrimination in employment, S.Q. 1964, c. 46, which provided for an individual remedy in cases involving discrimination in hiring or in conditions of employment.

Those statutes addressed the most obvious cases of discrimination, but they did not create a right to equal pay for work of equal value. Quebec would be the first province to act in this regard when it enacted the Quebec Charter on June 27, 1975:

19. Every employer must, without discrimination, grant equal salary or wages to the members of his personnel who perform equivalent work at the same place.

A difference in salary or wages based on experience, seniority, years of service, merit, productivity or overtime is not considered discriminatory if such criteria are common to all members of the personnel.

. . .

However, it quickly became clear that wage discrimination was a systemic problem that could not be resolved by means of the individual remedy provided for in the Quebec Charter:

[translation] Although the Charter can be used to rectify certain situations, its overall impact on the achievement of pay equity is thus limited.

(Comité d’élaboration du projet de loi, Document d’orientation: Une loi proactive sur l’équité salariale (1995) (“Policy Paper”), reproduced in A.R., vol. III, at p. 11.)

[translation] . . . this path is inappropriate in the face of a problem that is systemic, that is, that reflects widespread and enduring practices and behaviours in the labour market.

(Comité de consultation en regard à la loi proactive sur l’équité salariale, Rapport final: Une loi proactive sur l’équité salariale (1995) (“Final Report”), reproduced in A.R., vol. III, at p. 84.)

In addition, although s. 19 of the Quebec Charter made it possible for an employee to file an individual complaint, such a complaint involved a process that was slow, costly and largely inaccessible to women who were not represented by a union (Journal des débats de la Commission des affaires sociales (February 6, 1996), at pp. 1‑2). This process had the perverse effect of entailing individual confrontations between the employer and the employee concerned. Pay equity legislation would shift the burden and would place the employer under an obligation to pay non‑discriminatory compensation to everyone without a complaint or a remedy being needed.

In 1992, following a public consultation, the Commission des droits de la personne acknowledged the limitations of the complaint scheme under the Quebec Charter and recommended the enactment of proactive pay equity legislation (Journal des débats de la Commission des affaires sociales (February 6, 1996), at p. 2). The Secrétariat à la condition féminine then set up a committee of experts for the preparation of a bill. In June 1995, this committee submitted the Policy Paper to the responsible minister; it, too, recommended the enactment of proactive pay equity legislation and identified the major elements such legislation should include. The Policy Paper’s content formed the basis for the current Act.

At the time, similar proactive legislation was already in force in several Canadian provinces, including Manitoba (1985), Ontario (1987), Nova Scotia (1988), Prince Edward Island (1988), Newfoundland (1988), New Brunswick (1989) and British Columbia (1995). Ontario’s pay equity statute was the only one of these that applied to the private sector as the committee proposed in the Policy Paper. However, the committee noted that Ontario, on whose experience it was drawing because of that province’s pioneering role in pay equity, had preferred to exclude private businesses with no male job classes from the scope of its legislation.

The authors of the Policy Paper asserted that the Act should have the broadest possible scope. They therefore recommended going further than the Ontario legislation and the Quebec Charter and targeting enterprises with no male comparators. The committee of experts acknowledged that this objective would be difficult to achieve, though:

[translation] On this subject, it is important to note from the outset that in an economy like ours, the general determination of compensation is a decentralized process. Each enterprise is free to fix the compensation that will be paid to its employees, and even when enterprises are unionized, conditions of employment are negotiated with each employer separately. This is in fact why the first legislative provisions on pay equity confined this exercise to the enterprise itself. But it goes without saying that in the context of female job ghettos, pay equity cannot be achieved within the enterprise. Therefore, unless a completely different solution is applied, for example raising the minimum wage, it must be acknowledged that achieving pay equity in such enterprises requires going outside the enterprise.

(Policy Paper, at p. 33)

While some possible solutions were touched on, they were not examined in detail, and the committee of experts proposed no concrete solution in this regard in its recommendations. The committee therefore arrived at the following final recommendation, which it is important to reproduce in full:

[translation]

5‑ that, in the case of private sector enterprises in which there are no, or not enough, male jobs to serve as comparators for all the female jobs, the responsible authority recommend formulas for compensation adjustments for those jobs in a report to be submitted to the responsible minister, which must be the subject of consultations with the enterprises concerned, and that the appropriate formulas then be prescribed by a government regulation, which may also set the time limit for the enterprises in question to make the first compensation adjustments on this basis;

(Policy Paper, at p. 64)

In short, if the goal was for the Act to apply to enterprises in which there were no predominantly male job classes, additional time would be needed in order to consider and develop a method of comparison for such enterprises.

A second committee held a public consultation in which about 30 women’s organizations, unions, employer organizations, representatives of cultural communities, and compensation experts participated. The Final Report was submitted to the responsible minister on December 1, 1995. In it, the provisions the committee believed should be included in the legislation were set out and explained. The recommendation concerning enterprises without internal male comparators remained essentially unchanged:

[translation] [Recommendation No. 23] In the case of private sector enterprises in which there are no, or not enough, predominantly male job titles to serve as comparators for all the predominantly female jobs, the Commission de l’équité salariale will recommend formulas for compensation adjustments for those jobs in a report to be submitted to the responsible minister. That report must be the subject of consultations with enterprises, employees or their representatives, and women’s groups. The appropriate formulas will then be prescribed by a government regulation, which may also set the time limit for the enterprises in question to make the first compensation adjustments on this basis.

(Final Report, at pp. 123 and 170)

Once again, although certain solutions were mentioned, it was expressly acknowledged that there was a need for additional research and analysis, which would inevitably cause delays:

[translation] In any event, this issue of a lack of male comparators is very important, and the examples given above illustrate its complexity. Identifying possible answers in this regard would require in‑depth studies and analyses, and the committee had neither the time nor the resources to conduct them. The committee is therefore of the view that one of the first tasks of the Commission de l’équité salariale should be to do research or have research done on the issue. For now, the committee can only outline some possible options, such as the ones mentioned above.

(Final Report, at pp. 125-26 (emphasis added).)

Two important conclusions can be drawn from the Policy Paper and the Final Report. First of all, the importance of including employers without predominantly male job classes in the scope of the Act was recognized. Including them would guarantee pay equity and a discrimination‑free workplace for as many employees as possible. Next, the authors of the Final Report also recognized the importance of comparing compensation strictly within an enterprise. Indeed, this was one of two fundamental principles that were recommended in that report:

[translation] The proactive legislation is thus centred around two fundamental principles:

(1) What must be compared is the value of the jobs, not the value of the individuals who perform them. To avoid gender bias, remuneration must be based on the content of a job, not on the gender of those who hold it.

(2) Predominantly female jobs and predominantly male jobs must be compared within a single employer’s enterprise. The purpose of pay equity between men and women is not to establish uniform rates of remuneration for all enterprises, as, for example, the minimum wage decree does, but to equalize the remuneration of equivalent predominantly female jobs and predominantly male jobs within a given enterprise. Pay equity thus requires an internal comparison exercise.

(Final Report, at p. 89)

It is important to note that this second principle seemed to be essential to reconciling the interests at stake. For most of the groups representing employees, it was important that the legislation have a broad scope, that it be as close to universal as possible. They were therefore in favour of having the Act apply to employers that did not have male comparators. Some of them also asked that the provision of the draft bill be amended to require the Commission to make a regulation, as opposed to merely authorizing it to do so. In addition, they wanted there to be a time limit for making the regulation. The legislature did not adopt these last two proposals, however (s. 114 of the Act).

Employer representatives were reluctant to have the calculation of the adjustments based on external comparators. As was noted in the Final Report, it is each employer’s prerogative to fix the compensation it offers or might offer in its enterprise, provided that the compensation is not discriminatory.

The transcripts of the consultations show that, once again, few concrete solutions were proposed. Indeed, as Yergeau J. stated, [translation] “the issue of how to go outside the enterprise to find male comparators without importing external wage structures remained unresolved at the end of the general consultation and for a long time afterwards” (para. 63).

In the end, when the legislation was passed, it reiterated, with very few changes, the recommendation from the Policy Paper and the Final Report regarding the development of compensation adjustment formulas for enterprises with no male comparators:

114. The Commission may make regulations

(1) for the purposes of the determination of adjustments in compensation in an enterprise employing fewer than 50 employees where there are no predominantly male job classes, determining typical job classes on the basis of job classes identified in enterprises in which adjustments in compensation have already been determined and prescribing standards or weighting factors to be applied to the valuation of differences in compensation between such job classes, with due regard, in particular, for the characteristics of enterprises whose job classes are to be so compared;

(2) for the purposes of the establishment of a pay equity plan in an enterprise where there are no predominantly male job classes, determining typical job classes on the basis of job classes identified in enterprises, in which a pay equity plan has already been completed, prescribing methods to be used to determine the value of those job classes and to valuate the differences in compensation between the typical job classes and the job classes in an enterprise and prescribing standards or weighting factors to be applied to such differences, with due regard, in particular, for the characteristics of enterprises whose job classes are to be so compared;

. . .

Regulations of the Commission are subject to the approval of the Government and may be amended by the Government upon approval.

No regulation of the Commission may be approved by the Government until it is examined by the appropriate committee of the National Assembly.

Once the Act came into force, therefore, the Commission had to develop a method for determining the required compensation adjustments for enterprises in which there were no predominantly male job classes. This promised to be difficult, as there was at the time no legislative precedent in Canada it could use as a model.

It should first be mentioned that the Act did not give the Commission carte blanche in preparing the regulation. The method the Commission adopted for determining typical job classes had to be based on “enterprises in which adjustments in compensation have already been determined” (s. 114 para. 1(1)).

Although some preliminary work was started, the Commission had to wait until adjustments had been completed by other employers before it could begin preparing the regulation, which, in principle, was to be completed by November 21, 2001. The president of the Commission confirmed this in a letter dated April 1999:

[translation] . . . the Commission has, in the past few months, begun conducting studies on various workplaces in which there are no predominantly male job classes. This complex work cannot be completed quickly, because identified job classes in enterprises in which compensation adjustments have already been determined must be taken into account. Nevertheless, the necessary research must lead as quickly as possible to the drafting of a regulation.

(Reproduced in A.R., vol. VI, at p. 5.)

Thus, priority had to be given to the pay equity exercise in enterprises with male comparators. However, that exercise was not completed within the time provided, and as of November 21, 2001, fewer than 50 percent of the enterprises had finished it (trial judge’s reasons, at paras. 95‑96).

In October 2001, the Commission hired a consulting firm, Gestion‑Conseil Loran inc. (“Loran”), to develop a method that would allow enterprises to valuate differences in compensation where there are no male comparators. Loran undertook to complete this work by December 15, 2001. The Commission thus expected to receive Loran’s report less than a month after the other enterprises with 10 or more employees had completed the pay equity exercise.

Loran finally submitted its report on February 18, 2002. It analyzed several possible solutions and ultimately proposed a six‑step method. However, the proposed method was judged to be [translation] “far too complex . . . as . . . Martine Bégin testified when examined at length on this point” (trial judge’s reasons, at para. 108).

The Commission therefore wanted to explore other possible options. On July 18, 2002, it published a consultation paper entitled Proposition d’approches en l’absence de comparateur masculin in which it proposed three approaches. At a meeting of the consultation committee on August 30, 2002, two of the proposed approaches were rejected, and the committee agreed to confine itself to the third approach, that of the assignment of comparators (segmentation approach).

A new consultation paper that focused solely on the assignment of comparators method was then published on October 16, 2002. The disadvantage of this approach was that it was incompatible with the Commission’s power to make regulations, as provided, and would therefore require an amendment to s. 114 of the Act.

The question was submitted to the National Assembly’s Standing Committee on Labour and the Economy for a general consultation on February 11 and 12, 2003, from which a consensus emerged: the stakeholders did not want the Act to be amended. Yergeau J. summed up the situation as follows:

[translation] The transcript of those two days of consultations shows that what happened on that occasion was a dialogue of the deaf between the government, which suggested amending section 114 of the Act in order to expedite the making of the regulation, and the unions, which asked the government to make a regulation first and then determine afterwards whether the Act needed to be amended. Bottom line, the [Act] was not amended, and the method developed by the Commission with the support of the consultation committee was discarded. [para. 117]

Thus, the contemplated solution was ruled out, and right after the general consultation, the Commission had to start all over again (trial judge’s reasons, at para. 119).

In the summer of 2003, the Commission’s consultation committee turned to an entirely new approach that entailed using generic predominantly male job classes to make the comparison. According to this method, the employer was to establish wage curves for the compensation it would give a maintenance worker and a foreman if these job classes existed in its enterprise. These fictitious classes would then be used for comparison with the existing predominantly female job classes in the enterprise. This method thus made it unnecessary to resort to external comparators, and it was compatible with the power to make regulations under the Act.

In September 2004, this approach was adopted and a proposed regulation was published in the Gazette officielle du Québec. The Commission held a public consultation on November 2, 3 and 4, and on November 24, the proposed regulation was submitted to the Standing Committee on Labour and the Economy, which approved it. The regulation was finally made on May 5, 2005, and it came into force on that same day. The enterprises concerned then had two years to determine the required compensation adjustments and begin paying them in accordance with the Act (s. 38).

My analysis under s. 15 of the Charter is guided by the social context and legislative history set out above. Section 15 reads as follows:

15. (1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

(2) Subsection (1) does not preclude any law, program or activity that has as its object the amelioration of conditions of disadvantaged individuals or groups including those that are disadvantaged because of race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

This Court considered s. 15(1) of the Charter for the first time in Andrews, in which McIntyre J., writing for the majority on this issue, defined discrimination as follows:

I would say then that discrimination may be described as a distinction, whether intentional or not but based on grounds relating to personal characteristics of the individual or group, which has the effect of imposing burdens, obligations, or disadvantages on such individual or group not imposed upon others, or which withholds or limits access to opportunities, benefits, and advantages available to other members of society. Distinctions based on personal characteristics attributed to an individual solely on the basis of association with a group will rarely escape the charge of discrimination, while those based on an individual's merits and capacities will rarely be so classed. [pp. 174‑75]

In Andrews, the Court considered several possible approaches to the analysis required by s. 15(1) (pp. 178‑83). McIntyre J. began by expressly rejecting an approach according to which every distinction drawn by law constitutes discrimination and must be considered under s. 1 of the Charter, arguing that to adopt it “would be to trivialize the fundamental rights guaranteed by the Charter and deprive the words ‘without discrimination’ in s. 15(1) of any content and, in effect, to replace s. 15(1) with s. 1” (p. 162). Moreover, the Charter cannot require that the legislature justify every distinction under s. 1 (p. 181). McIntyre J. also rejected an approach involving a consideration of the reasonableness and justifiability of the legislation under s. 15(1), which would leave no role for s. 1. In the end, he adopted a third approach:

The analysis of discrimination in this approach must take place within the context of the enumerated grounds and those analogous to them. The words “without discrimination” require more than a mere finding of distinction between the treatment of groups or individuals. Those words are a form of qualifier built into s. 15 itself and limit those distinctions which are forbidden by the section to those which involve prejudice or disadvantage. [pp. 180‑81]

Andrews is the seminal case on the interpretation of s. 15(1), and the principles laid down in it by McIntyre J. have been confirmed and clarified in several subsequent decisions of the Court (Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497, at para. 22; R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483, at para. 14; Withler v. Canada (Attorney General), 2011 SCC 12, [2011] 1 S.C.R. 396, at paras. 29 et seq.; Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61, at para. 319). What can be drawn from those cases is that s. 15(1) guarantees for every person substantive equality as opposed to mere formal equality under the law (Andrews, at pp. 163‑65; Kapp, at paras. 16 and 20; Withler, at para. 2; Alberta (Aboriginal Affairs and Northern Development) v. Cunningham, 2011 SCC 37, [2011] 2 S.C.R. 670, at para. 38; Quebec v. A, at para. 325; Kahkewistahaw First Nation v. Taypotat, 2015 SCC 30, [2015] 2 S.C.R. 548, at para. 17). This guarantee is rooted in a recognition of the fact that persistent systemic disadvantages have operated to limit the opportunities available to members of certain groups in society. The purpose of the Charter provision is thus to prevent conduct that perpetuates those discriminatory disadvantages (Taypotat, at para. 17; Quebec v. A, at para. 320).

The main issue under s. 15(1) is whether the impugned law violates this Charter guarantee of substantive equality (Withler, at para. 2). It is therefore necessary to apply the two‑step analytical framework laid down by the Court and ask the following two questions: (1) Does the law create a distinction based on an enumerated or analogous ground? and (2) Does the distinction create a discriminatory disadvantage by, among other things, perpetuating prejudice or stereotyping? (Kapp, at para. 17; Withler, at paras. 30 and 61; Quebec v. A, at para. 324). In this analysis, the main consideration must be the impact of the law (Andrews, at p. 165; Quebec v. A, at para. 319).

At the first step of the analysis, it must be determined whether, on its face or in its impact, a law creates a distinction on the basis of an enumerated or analogous ground (Taypotat, at para. 19). Accordingly, the courts will address only those grounds of distinction that were intended to be prohibited by the Charter (Withler, at para. 33).

As the trial judge indicated, s. 38 of the Act clearly makes a distinction by postponing the payment of compensation adjustments to May 5, 2007:

[translation] In the case at bar, however, section 38 of the Act, by allowing the payment of compensation adjustments to female employees in enterprises with no male comparators to be put off until May 5, 2007, had the effect of treating such employees differently, for some time, than those working in enterprises with comparators. The plaintiffs argue that, had it not been for section 38, these employees would, the belated coming into force of the Regulation respecting pay equity notwithstanding, have received an adjustment retroactive to December 21, 2001.

. . .

Furthermore, this was an adverse distinction for them, insofar as the making of the Regulation respecting pay equity was a long time coming, thereby depriving them of more than five years of retroactive compensation adjustments. [Emphasis deleted; paras. 192‑93.]

The question of the basis for that distinction requires that it be determined whether the impugned law has a disproportionate effect on employees working in enterprises without predominantly male job classes that is “based on [their] membership in an enumerated or analogous group” (Taypotat, at para. 21 (emphasis added)).

By establishing different time limits, the Act had an adverse impact on employees working in enterprises in which comparisons with predominantly male job classes are impossible. This group consists mostly of women and is at a particular disadvantage in the labour market, which may suggest that these women are being treated differently because of their sex. However, to resolve this issue, we must go further and ask what the basis for this differential treatment is.

In the case at bar, I am of the opinion that the distinction is not based on sex, because the differential treatment does not result from the fact that the affected employees are women. In this regard, an analysis of the evidence as a whole leads to the conclusion that the basis for the differential treatment affecting the employees in question lies in the lack of male comparators in their employers’ enterprises.

Contrary to what my colleague Abella J. asserts, this does not amount to adopting a formal equality approach as in Bliss v. Attorney General of Canada, [1979] 1 S.C.R. 183. Indeed, it has since been established that s. 15(1) grants every person a guarantee of substantive equality, not just of formal equality, under the law.

In Bliss, a woman was denied unemployment insurance benefits to which she would have been entitled had she not been pregnant. Her claim was rejected by the Court then on the basis that the distinction was based not on sex, but on pregnancy. This approach has since been specifically rejected in Andrews, in which McIntyre J. commented as follows regarding the analysis that is required in order to determine the basis for a distinction:

I would also agree with the following criticism of the similarly situated test made by Kerans J.A. in Mahe v. Alta. (Gov't) (1987), 54 Alta. L.R. (2d) 212, at p. 244:

. . . the test accepts an idea of equality which is almost mechanical, with no scope for considering the reason for the distinction. In consequence, subtleties are found to justify a finding of dissimilarity which reduces the test to a categorization game. Moreover, the test is not helpful. After all, most laws are enacted for the specific purpose of offering a benefit or imposing a burden on some persons and not on others. The test catches every conceivable difference in legal treatment.

For the reasons outlined above, the test cannot be accepted as a fixed rule or formula for the resolution of equality questions arising under the Charter. Consideration must be given to the content of the law, to its purpose, and its impact upon those to whom it applies, and also upon those whom it excludes from its application. The issues which will arise from case to case are such that it would be wrong to attempt to confine these considerations within such a fixed and limited formula. [Emphasis added; p. 168.]

This passage reiterates the importance of taking the context into consideration in order to determine the real reason for the distinction. After thoroughly reviewing the evidence, Yergeau J. concluded that the distinction created by s. 38 of the Act is based not on sex, but on the fact that the enterprises to which it applies do not have predominantly male job classes that would allow for comparison.

My colleague Abella J. instead focuses in her reasons on the basis for the distinction drawn by the Act as a whole (paras. 24‑25). But the provision being challenged by the appellants is s. 38 of the Act. It is thus the reason for the distinction established in that section, not the reason for the distinction the Act is intended to eliminate as reflected by its purpose, that must be identified at this step. My colleague does identify the real reason for the distinction in stating that s. 38 targets a particular group on the basis of where they work (para. 24). Her reasoning can lead to only one conclusion: that every distinction in a pay equity statute is necessarily based on sex.

The alleged violation is based on the principle that the legislature is, when it acts proactively, under a constitutional obligation not to be discriminatory in how it treats those who are meant to benefit from the law. But the distinction in this case is not based on sex, as the comparator group is not that of male employees, to whom the Act does not apply. Certain of those who are meant to benefit from the Act — employees of enterprises that have no predominantly male job classes — wish to be paid in accordance with the same timetable that applies to others who are also meant to benefit from the Act — employees of enterprises that do have predominantly male job classes. Hence, the distinction resulting from s. 38 cannot be rooted in the gender‑based discrimination that the Act is as a whole intended to remedy.

A pay equity statute grants rights to a group consisting essentially of women. Like any other statute, it may make distinctions, and such distinctions can be more advantageous for one group consisting of women than for another group also consisting of women. However, we cannot conclude that every distinction drawn by a pay equity statute is necessarily based on sex. Such a conclusion would deprive trial judges of any discretion in their assessment of the evidence and would make the first step in the s. 15(1) analysis irrelevant. On the contrary, it must be possible to show on a balance of probabilities that the distinction in question is based not on sex, but on another, perfectly legitimate ground.

McIntyre J. stated the following in Andrews:

It is not every distinction or differentiation in treatment at law which will transgress the equality guarantees of s. 15 of the Charter. It is, of course, obvious that legislatures may — and to govern effectively — must treat different individuals and groups in different ways. Indeed, such distinctions are one of the main preoccupations of legislatures. The classifying of individuals and groups, the making of different provisions respecting such groups, the application of different rules, regulations, requirements and qualifications to different persons is necessary for the governance of modern society. As noted above, for the accommodation of differences, which is the essence of true equality, it will frequently be necessary to make distinctions. What kinds of distinctions will be acceptable under s. 15(1) and what kinds will violate its provisions? [pp. 168‑69]

In the instant case, I agree with Yergeau J.’s analysis and accordingly conclude that s. 38 does not create a distinction based on sex. It therefore should not be declared to be invalid under the Charter.

This conclusion is in and of itself sufficient to dispose of the appeal. In light of the evidence before him, however, Yergeau J. preferred to discuss the second step of the s. 15(1) analysis, and I will therefore also consider this step, which, once again, confirms the validity of s. 38.

In theory, we must conduct the analysis under s. 15(2) of the Charter before turning to the second step of the s. 15(1) analysis (Kapp, at para. 40; Cunningham, at para. 44). But given my conclusion that s. 38 of the Act is valid, it is not necessary to go on to determine whether that section can be saved under s. 15(2).

The distinction made in s. 38 of the Act does not have a discriminatory impact.

A statute is not necessarily invalid because it makes distinctions (Andrews, at p. 167; Kapp, at para. 28; Withler, at para. 31). On the contrary, drawing distinctions is essential to the effective operation of our governments. That is why s. 15(1) guarantees the equal benefit of the law to every individual “without discrimination”. This qualifier prohibits only those distinctions that are discriminatory. The claimant must therefore prove that the law creating a distinction has a discriminatory impact in that it perpetuates a disadvantage faced by the group in question (Andrews, at p. 181; Kapp, at para. 28; Withler, at paras. 31‑34; Quebec v. A, at para. 322; Taypotat, at para. 20):

. . . the claimant’s burden under the Andrews test is to show that the government has made a distinction based on an enumerated or analogous ground and that the distinction’s impact on the individual or group perpetuates disadvantage.

(Quebec v. A, at para. 323)

The words “without discrimination” require more than a mere finding of distinction between the treatment of groups or individuals. Those words are a form of qualifier built into s. 15 itself and limit those distinctions which are forbidden by the section to those which involve prejudice or disadvantage.

(Andrews, at pp. 180-81)

What must be done at this stage of the analysis is to consider the situation of the group’s members and the law’s impact on them. “The analysis is contextual, not formalistic, grounded in the actual situation of the group and the potential of the impugned law to worsen their situation” (Withler, at para. 37; Quebec v. A, at para. 331). State conduct that widens the gap between a historically disadvantaged group and the rest of society will be found to be discriminatory (Quebec v. A, at para. 332).

In Law, four contextual factors were identified that might be relevant in determining whether a law is discriminatory: (1) the nature of the affected interest, (2) a pre‑existing disadvantage, (3) a correspondence with actual characteristics and (4) the impact on other groups. It is not always necessary to expressly canvass every one of these factors, and there will be cases in which other factors might also be pertinent (Law, at para. 62; Withler, at para. 66; Quebec v. A, at para. 331).

The importance of the right at issue in the case at bar is not in doubt. For the employees concerned, it is a question of being paid an amount that reflects the fair value of their work and receiving compensation based on a prejudice‑free evaluation that focuses on the objective value of their work. The Court has said the following in this regard:

Work is an important part of life. For many people what they do for a living, and the respect (or lack of it) with which their work is regarded by the community, is a large part of who they are. Low pay often denotes low status jobs, exacting a price in dignity as well as dollars. As such, the interest affected by the Act was of great importance.

(N.A.P.E., at para. 49)

This right is all the more important because women working in enterprises with no male comparators have a significant pre‑existing disadvantage, one that is documented, is not in dispute and is in fact recognized from the outset in s. 1 of the Act. The fact that there is pre‑existing prejudice is relevant to the s. 15(1) analysis (Withler, at para. 38; Quebec v. A, at para. 327; Taypotat, at para. 21) and requires that particular attention be paid to the distinctions that affect the group. But it does not give rise to a presumption that a distinction is discriminatory (Law, at para. 67).

What must be asked, therefore, is whether the Act and the time limits it establishes reflect the actual capacities and needs of the members of the affected group or, instead, impose a burden on them or deny them a benefit in a manner that has the effect of reinforcing, perpetuating or exacerbating their disadvantage (Taypotat, at para. 20; see also Law, at para. 70).

In this case, it can be seen that significant differences in compensation due to systemic gender discrimination already existed in the labour market and that these differences were maintained in the private sector. The legislature, which was not required to intervene under s. 15(1) of the Charter, chose to do so, after extensive consultations, to redress these differences. In the Act, it established a mechanism for achieving pay equity and set various applicable time limits. It is important to bear in mind that the systemic discrimination at issue in this case was not caused by the legislature’s actions. On the contrary, the Act has an ameliorative effect and does not have the effect of perpetuating that systemic discrimination.

My colleague McLachlin C.J. asserts that when the legislature enacted the Act, it sent to the members of the group represented by the appellants a message that the systemic discrimination in their jobs “was their problem” (para. 156). I disagree with this assertion, which in my view disregards the effect the Act has actually had on that group.

In enacting the Act, the legislature was responding directly to the needs of the members of the group represented by the appellants. It set up a proactive scheme that, once completed, made it possible to redress differences in compensation due to systemic gender discrimination. In so doing, it narrowed the gap that had existed between the employees in question and the rest of society for far too long. In order to extend that right to as many people as possible, the legislature chose to include employees working in enterprises with no male comparators. In the Act, it recognized the discrimination faced by the members of that group. When the Act was enacted in 1996, Quebec became the first province to tackle this problem. An initiative such as this that is designed to enhance substantive equality should be encouraged and praised. This is all the more true given that the Act continues today to be one of the most ambitious in the country.

However, it was also recognized from the outset that, if the intention was for the Act to apply to the group in question, more time would be needed to develop an adequate method for calculating compensation adjustments. Proceeding by way of a regulation thus made it possible to develop a simple and innovative method for redressing differences in compensation in enterprises with no male comparators.

In sum, the Act has an undeniably ameliorative effect on the employees in question. My colleagues’ analysis leads to the conclusion that any ameliorative measure adopted by the legislature that does not result in perfect equality would infringe s. 15(1). On the contrary, given that the problem existed and persisted in the private sector, and given that it would have been perfectly valid from a constitutional standpoint for the legislature not to intervene, any measure that ameliorates the conditions of this group will necessarily respond to its needs. Even if a solution is not necessarily perfect and does not totally eliminate the disadvantage suffered by the group, only an intervention that would damage or exacerbate the group’s situation would be discriminatory.

Lastly, an essential factor to consider is the significant positive effect of the Act on many other employees who also have pre‑existing disadvantages:

Where the impugned law is part of a larger benefits scheme . . . the ameliorative effect of the law on others and the multiplicity of interests it attempts to balance will also colour the discrimination analysis [under s. 15(1)].

(Withler, at para. 38)

When the Act was being drafted, the groups representing female employees sent the legislature a clear message: it had to act quickly. The legislature therefore chose to enact the Act quickly, thus guaranteeing pay equity for a large number of employees working in more than 225,000 enterprises in Quebec (trial judge’s reasons, at para. 46). The decision to do so obviously had a significant positive effect on those employees. At the same time, it was decided that the case of employers that have no predominantly male job classes would be dealt with by way of a regulation. Very few stakeholders objected to this approach. The comments of the Conseil du statut de la femme on this point are instructive:

[translation] There have also been calls for the legislation to apply to enterprises with fewer than 10 employees as well as for including a solution to the specific case of job ghettos. But everyone agrees that these two issues will not be easily resolved. We instead believe that rather than depriving a large majority of female workers of the benefits of such a statute, it is preferable to proceed right away. As indicated in the draft bill, however, we will have to go straight to work to address this serious deficiency.

(Journal des débats de la Commission des affaires sociales, vol. 34, No. 39, 1st Sess., 35th Leg., February 15, 1996, at pp. 2-3)

It would have been disadvantageous and unfair for female Quebec employees as a whole had the enactment of the Act been postponed for several years in order to identify a method for calculating compensation adjustments that could apply to employers with no male comparators. It goes without saying that when a government develops a complex scheme such as that of pay equity legislation, it will not always be able to ameliorate the conditions of every member of a disadvantaged group at the same time and in the same way. That is why an obligation of result cannot be imposed on the legislature in this regard. Such an obligation would cause it to show extreme caution and even, in some instances, to postpone the enactment of legislation, to everyone’s detriment.

The decision to proceed by way of regulation enabled the legislature to take the necessary time to develop an innovative, simple and practical method for valuating differences in compensation for employers that have no predominantly male job classes. The adoption of that method meant that it was not necessary to have recourse to external comparators or to import a salary structure that was foreign to the enterprise.

The legislature was entitled to proceed as it did. The different time limits provided for in the Act do not perpetuate prejudice or a stereotype. On the contrary, the legislature recognized the existence of discrimination and did what was necessary to rectify it. The situations of some of the groups concerned were different and therefore required different methods. The result of this reality may very well be that the Act provides for different time limits or even different procedures. Nevertheless, the Act does not have a discriminatory impact, as it instead narrows the gap between these historically disadvantaged groups and the rest of society.

In this case, as the Court put it in Withler, differential treatment was required in order to ameliorate the actual situation of the group represented by the appellants, as well as that of the other groups to which the section applies (para. 39). That differential treatment also enabled Quebec to be the first Canadian province to guarantee pay equity in private sector enterprises that had no male comparators.

My colleague Abella J. maintains that the reasoning adopted in Vriend v. Alberta, [1998] 1 S.C.R 493, should apply in the instant case. In Vriend, the Individual’s Rights Protection Act, R.S.A. 1980, c. I‑2, had in 1990 prohibited discrimination based on any of the following grounds: race, religious beliefs, colour, gender, physical disability, mental disability, age, ancestry or place of origin. The statute had subsequently been amended to add other grounds to the existing ones, namely marital status, source of income and family status. The Individual’s Rights Protection Act established a remedy for all members of society so that they could avail themselves of any of the grounds of discrimination provided for in the statute. However, the Alberta government had not deemed it necessary to include sexual orientation in the statute as a prohibited ground. Individuals who were discriminated against on the basis of their sexual orientation were therefore excluded from the protection of the statute and found themselves worse off than they had been before it came into force, given that they were not protected by the statute as were other disadvantaged groups. As a result, the gap between their situation and that of the rest of society widened and their marginalization increased. In short, society had moved forward, but their situation remained unchanged. Unlike a general protection scheme, such as the one established by the Individual’s Rights Protection Act, the scheme established by the Act in the case at bar deals with only one ground of discrimination, namely sex, and applies to only a small group of employees in Quebec, those who hold positions in predominantly female job classes in enterprises with 10 or more employees. The analogy my colleague is trying to draw cannot therefore be accepted, nor is it even desirable; and it is irrelevant. Moreover, the situation of the employees in question is alleviated by the fact that they belong to the group benefiting from the Act that is challenging its effects. This means that it cannot be argued that they do not receive “the equal protection and equal benefit of the law” (s. 15(1) of the Charter). Also, unlike the statute that was at issue in Vriend, the Act has an undeniably ameliorative effect for the employees the appellants represent, and it helps narrow the gap between them and the rest of society.

I therefore conclude that the impugned provision does not create a disadvantage by perpetuating prejudice or by stereotyping and does not have a discriminatory impact. The second condition has accordingly not been met. Although my conclusion with respect to s. 15(1) of the Charter means that I need not deal with s. 15(2) , my silence should not be interpreted as an endorsement of my colleague Abella J.’s comments on s. 15(2) .

In conclusion, I am of the opinion that the distinction made in s. 38 is based not on sex, but on whether there are predominantly male job classes in an enterprise for the purpose of comparing compensation. Furthermore, this distinction does not have a discriminatory impact on the group of women represented by the appellants. Section 38 of the Act is therefore valid under s. 15(1) of the Charter, and the appeal should be dismissed with costs.

The following are the reasons delivered by

The Chief Justice (dissenting in the result) — I agree with Abella J. that s. 38 of the Pay Equity Act, CQLR, c. E-12.001, breaches s. 15(1).

The legislated delay imposed by the provision creates a distinction based on sex because it distinguishes between women in female-dominated workplaces — the focus of s. 38 — and male employees in general, whose presence is the lynchpin for higher wages in mixed workplaces. The substantive effect of the delay Abella J. describes is that women in female-dominated workplaces were unable to secure fair wages for their work, in contrast to men doing comparable work in other arenas. Where a distinction viewed in context is redolent of gender-based difference, as it is here, that cannot be ignored in the s. 15(1) analysis. Otherwise, the promise of substantive equality that animates s. 15(1) is betrayed.

This distinction is discriminatory because the delay excused employers of women in female-dominated workplaces from paying their employees fairly for the work they were doing. Quebec gave employers carte blanche to ignore pay inequity in their organizations during the delay. The scheme said to impacted women: this is your problem. And the legislated delay signaled that it was acceptable for employers to make decisions to the detriment of the autonomy and inherent dignity of women in female-dominated workplaces. In effect, the scheme bolstered the very power imbalance between employers and female employees that lies at the heart of gender-based pay disparities, thereby perpetuating systemic inequality. Finally, it goes without saying that the economic consequences of pay inequity for women are profound.

However, I depart from Abella J. on the question of justification under s. 1 of the Charter. In my view the Attorney General has not established that the breach is reasonable and justified under s. 1: R. v. Oakes, [1986] 1 S.C.R. 103. The Attorney General at the outset must establish a pressing and substantial objective for delaying the right and disentitling affected women from access to pay equity, exacerbated here by a lack of retroactive corrections under the Act. The government says it needed six years to come up with a formula to determine corrections for female-dominated workplaces due to the absence of male comparator groups and the need to ensure employer compliance. It also says that retroactive correction was rejected because it was problematic for employers. I agree that devising a satisfactory method of evaluating underpayment is a pressing and substantial objective capable of justifying a limit on the right to equal benefit of the law. It is more questionable whether the government’s objective of ensuring employer compliance by imposing a long delay that would be unmet with any retroactive correction, for example, could justify breaching employees’ rights. To agree would be to accept that obeying pay equity laws is an option that can be negotiated and that the very segment that perpetuates systemic pay inequities — the employers — should be able to perpetuate them as the price of accepting the law.

The Attorney General’s attempt to justify the infringement, in any event, fails at the minimal impairment and balancing stages of the Oakes test. The Attorney General says the six-year delay was necessary to work out what constituted equal pay in female-dominated workplaces, where no male comparator groups were available. But it has not demonstrated that the whole of the six-year period was required for this purpose. This period was dictated not by the exigencies of the matter, but at least in part by the government’s decision to negotiate with employers over a lengthy period, in order to ensure that the scheme was one that employers would accept and with which they would comply. Were there other less impairing options than extended negotiations? The government has not negated this. Similarly, the Attorney General says it agreed to deny retroactive correction of pay inequity because this would be difficult for employers to accept. Again, the government has not demonstrated that other options that impaired the right less were not available. It does not suggest it raised the possibility of partial redress, for example. It simply relies on what it did. Minimal impairment cannot be established simply by saying that a lengthy delay was required full-stop or that a lengthy delay could not be accompanied by retroactive correction because this is problematic to the employers who were perpetrating the inequality. Finally, the Attorney General has not established that the denial of benefits to the affected, already-marginalized women, is proportionate to the public interest in denying them a remedy.

I conclude that the breach of s. 15(1) has not been shown to be justified under s. 1 of the Charter, and therefore I would allow the appeal.