COURT OF APPEAL FOR ONTARIO

CITATION: Ontario (Health) v. Association of Ontario Midwives, 2022 ONCA 458

DATE: 20220613

DOCKET: C68946

Fairburn A.C.J.O., Roberts J.A. and Van Melle J. (ad hoc)

BETWEEN

Her Majesty the Queen in Right of Ontario as Represented by the Minister of Health

Applicant (Appellant)

and

Association of Ontario Midwives and Human Rights Tribunal of Ontario

Respondents (Respondents)

S. Zachary Green and Yashoda Ranganathan, for the appellant

Mary Cornish, Adrienne Telford, Lara Koerner-Yeo and Jackie Esmonde, for the respondent Association of Ontario Midwives

Jason Tam and Brian A. Blumenthal, for the respondent Human Rights Tribunal of Ontario

Reema Khawja and Raj Dhir, for the intervener Ontario Human Rights Commission

Heard: November 10, 2021 by video conference

On appeal from the order of the Divisional Court (Backhouse, Pattillo and Lederer JJ.), dated June 26, 2020, with reasons reported at 2020 ONSC 2839, dismissing an application for judicial review of the decisions of the Human Rights Tribunal of Ontario, dated September 24, 2018, with reasons reported at 2018 HRTO 1335, and February 19, 2020, with reasons reported at 2020 HRTO 165.

Fairburn A.C.J.O.:

overview

[1] Ontario midwives are almost exclusively women.[^1] In 2013, the Association of Ontario Midwives (the “AOM”) brought a human rights complaint on behalf of more than 800 midwives, alleging systemic gender[^2] discrimination by the Ministry of Health and Long-Term Care (the “MOH”), which funds Ontario’s midwifery program. The AOM challenged the MOH’s compensation practices back to 1994, when Ontario midwives were regulated, and sought compensation back to 1997.

[2] The Human Rights Tribunal of Ontario (the “Tribunal”) Adjudicator divided her decision into two periods: 1993 to 2005, and 2005 to 2013.

[3] She found that, in 1993, the parties agreed to equitable compensation principles that were designed to ensure that midwives’ compensation was not affected by harmful assumptions and stereotypes concerning the value of women’s work. Significantly, she found that these principles were connected, if not imbued, with gender. She was satisfied that the parties maintained a connection to the principles until 2005. Therefore, she concluded there was insufficient evidence of discrimination for the period from 1994 to 2005.

[4] However, the Adjudicator found that the situation changed after 2005, as the MOH gradually withdrew from the principles they had agreed to in 1993. Indeed, in 2010, after a joint compensation report recommended that midwives receive a 20% compensation increase, the MOH made it explicit that the principles would no longer inform the compensation of midwives and that they would not be replaced with a new methodology for determining appropriate and fair compensation. The Adjudicator found that this left the compensation of midwives exposed to the well-known effects of gender discrimination on women’s compensation. Considering the totality of the evidence and applying well-established jurisprudence, the Adjudicator concluded that sex was more likely than not a factor in the adverse treatment midwives experienced after 2005, including the significant compensation gap that developed between midwives and certain family physicians, who the Adjudicator found served as a male comparator. Accordingly, the MOH was liable for discrimination under Ontario’s Human Rights Code, R.S.O. 1990, c. H.19 (the “Code”).

[5] In a separate decision, the Adjudicator made remedial orders arising from her finding of liability, including orders granting a compensation adjustment of 20% back to 2011 and compensation for injury to dignity, feelings, and self-respect in the amount of $7,500 per eligible midwife, plus orders to promote ongoing compliance with the Code.

[6] The MOH applied to the Divisional Court for judicial review of both decisions. Its application was dismissed.

[7] This is an appeal from the Divisional Court’s decision. While the MOH raises a multitude of issues on appeal, those issues can be distilled into the following three overarching questions that call out for an answer:

(1) What is the standard of review of the Tribunal’s decisions post-Vavilov?

(2) Is the Adjudicator’s liability decision unreasonable? In particular:

• Do the Adjudicator’s reasons fail to reveal a rational chain of analysis?

• Was it unreasonable for the Adjudicator to find that gender was a factor in the compensation of midwives?

• Did the Adjudicator reverse the burden of proof?

• Did the Adjudicator unreasonably ignore the MOH’s expert evidence tendered to prove that gender was not a factor in midwives’ compensation?

• Did the Adjudicator unreasonably find that community health clinic (“CHC”) physicians remained appropriate comparators after they became predominantly female?

• Did the Adjudicator unreasonably impose a positive obligation on the MOH?

(3) Is the Adjudicator’s remedy decision unreasonable?

[8] As I will explain: (1) the Adjudicator’s decisions are reviewable on a reasonableness standard; and (2) both of the Adjudicator’s decisions are reasonable. I would dismiss the MOH’s appeal.

background

(1) Systemic Discrimination

[9] The AOM’s claim was one of adverse impact and systemic discrimination. I find it necessary to provide a brief overview of the nature of adverse impact and systemic discrimination at the outset of my reasons because, as will become apparent, an appreciation of these concepts is integral to understanding the Adjudicator’s decision.

[10] Adverse impact discrimination occurs when seemingly neutral rules, policies, procedures, systems, or structures have a disproportionate impact on disadvantaged groups: see Fraser v. Canada (Attorney General), 2020 SCC 28, 450 D.L.R. (4th) 1, at paras. 30-31. As Abella J. recognized in Fraser, an increased awareness of adverse impact discrimination has led to a shift away from a fault-based conception of discrimination towards an effects-based model. Identifying adverse impact discrimination involves critically examining systems and structures, recognizing that discrimination is “frequently a product of continuing to do things ‘the way they have always been done’”: Fraser, at para. 31, citing Fay Faraday, “One Step Forward, Two Steps Back? Substantive Equality, Systemic Discrimination and Pay Equity at the Supreme Court of Canada” (2020) 94 S.C.L.R. (2d) 301, at p. 310.

[11] In Canadian National Railway Co. v. Canada (Canadian Human Rights Commission), 1987 CanLII 109 (SCC), [1987] 1 S.C.R. 1114, Dickson C.J. defined systemic discrimination, at p. 1139, as follows:

[S]ystemic discrimination in an employment context is discrimination that results from the simple operation of established procedures … none of which is necessarily designed to promote discrimination. The discrimination is then reinforced by the very exclusion of the disadvantaged group because the exclusion fosters the belief, both within and outside the group, that the exclusion is the result of “natural” forces, for example, that women “just can’t do the job”. [Citation omitted.]

Therefore, in a claim of systemic discrimination, it can be difficult to identify one single rule, policy, or procedure that results in adverse impact discrimination. Often “there is no single identifiable ‘villain’, no single action identifiable as ‘discriminatory””: Fraser, at para. 35, citing Mary Eberts & Kim Stanton, “The Disappearance of the Four Equality Rights and Systemic Discrimination from Canadian Equality Jurisprudence” (2018) 38 N.J.C.L. 89, at p. 92. Rather, systemic discrimination may emerge from an “invisible structure, with its accompanying set of practices” or as a “complex web of seemingly neutral, systemic barriers”: Fraser, at para. 35, citing Eberts & Stanton, at p. 92; British Columbia (Public Service Employee Relations Commission) v. B.C.G.S.E.U., 1999 CanLII 652 (SCC), [1999] 3 S.C.R. 3, at para. 42.

[12] In this case, the Adjudicator, having “taken a step back … to consider the systemic nature and cumulative effects of policies and conduct on the compensation of midwives”, concluded that midwives were subjected to systemic gender discrimination after 2005. This had the effect of “perpetuat[ing] the historic disadvantage midwives have experienced as sex-segregated workers” and “undermin[ing] their dignity”.

[13] Before turning to a review of the history of the regulation of midwifery and how midwives’ compensation has been set over time, it is important to acknowledge that the midwife profession is the ultimate sex-segregated profession: women providing a service for women in relation to women’s health. And with this confluence of factors at work, combined with the fact that “[o]ccupational segregation and low wages ‘usually go hand in hand’”, and that jobs that are considered “female jobs” are often undervalued, there was an obvious risk that midwives would be under-compensated because they are women: see Centrale des syndicats du Québec v. Quebec (Attorney General), 2018 SCC 18, [2018] 1 S.C.R. 522, at para. 34.

(2) The History of Regulating Midwifery and Compensation-setting for Midwives

[14] The history of midwifery in Ontario, particularly the history of the regulation of midwifery and how midwives’ compensation has been set since 1994, is also integral to understanding the Adjudicator’s decision. Recognizing the importance of the history for understanding the Adjudicator’s decision, “Appendix A” to these reasons contains a detailed timeline of the various reports, committees, and working groups that considered midwives’ compensation from 1987 to 2013.

(a) The regulation of midwifery

[15] Midwifery has been a regulated health profession in Ontario since January 1, 1994.

[16] Prior to that date, midwives were officially excluded from Ontario’s health care system. Those who chose to practice prior to regulation did so in precarious circumstances, and up against attitudes that home births were unsafe and that midwives should be practising under the supervision of a physician.

[17] At the time of regulation, medicine was male-dominated and still strongly identified with men’s work. In contrast, midwifery was, and indeed continues to be, strongly identified with women’s work. The AOM has described midwifery as a “gender trifecta”: a service provided by women, for women, in relation to women’s reproductive health.

[18] In 1986,[^3] Ontario established the Task Force on the Implementation of Midwifery (the “Task Force”) to recommend a framework for the regulation of midwives and their integration into the health care system. In 1987, the Task Force released its report on the implementation of midwifery in Ontario.

[19] The Task Force recommended that regulated midwives be given a broad scope of practice and the autonomy to practice as primary caregivers for women with low-risk pregnancies, therefore allowing them to be “a true alternative to physician care for a proportion of women.”

[20] While compensation for midwives was not among the topics that the Task Force was asked to address, its report made clear that midwives should be paid at a fair and reasonable level that reflected their level of responsibility, the demands on their time, the difficulty of their work, the cost of participating in continuing education activities, and the cost of professional liability insurance. It also suggested positioning midwives somewhere between the starting salary for a nurse with a baccalaureate degree and the fees physicians were paid under OHIP for pregnancy, labour, birth, and postpartum care.

[21] Similarly, in 1992, the Interim Regulatory Council of Midwives (the “Council”), which had been established a couple years prior, also recommended an equitable formula for the compensation of midwifery determined by reference to objective factors such as skill, education, working conditions, and degree of responsibility. The Council also cited the Task Force recommendation that compensation for midwives fall between that of a senior nurse and a family physician.

[22] The Women’s Health Bureau (the “WHB”), the branch of the MOH that was responsible for developing the policy framework to support the public funding of midwifery, shared the view that midwives should be paid as primary care providers and that their compensation should fall between that of a family physician and a senior salaried nurse. A WHB “Options Paper” on compensation for midwives noted the “[n]ecessity to establish a fair and equitable pay level based on pay equity, reflecting responsibilities, working conditions and level of education.”

[23] A joint working group of the MOH and the AOM was eventually created to determine payment levels and develop a standard contract for payment of midwifery services. Robert Morton, a compensation expert, was retained to assist the working group in determining an “appropriate and fair” compensation level for midwives. His work included conducting an evaluation of the skill, effort, responsibility, and working conditions, or “SERW”, of midwives as compared to other healthcare professions. The “Morton Report” evolved out of that work. A fundamental principle that was accepted by the parties and established in the Morton Report was that the compensation for midwives would reflect the overlapping scope of practice they shared with primary care nurses and physicians working in CHCs.

[24] Throughout the decision, I refer to the compensation principles developed by the joint working group and established in the Morton Report as the “1993 principles”. The 1993 principles included an evaluation of the skill, effort, responsibility, and working conditions of midwives as compared to other healthcare professionals, specifically primary care nurses and family physicians working in CHCs, the latter of which the Adjudicator found to be a male comparator.

[25] Following the development and application of the 1993 principles, the parties agreed to an initial salary range for midwives of $55,000 to $77,000. This placed an entry-level midwife’s salary around the top salary of a CHC senior nurse, and the highest compensation level for a midwife at approximately 90% of the base salary of an entry level CHC family physician.

[26] That was the state of affairs upon regulation in 1994.

(b) Post-regulation – 1994-2005: 11 years of wage freezes

[27] From 1994 until 2005, midwives’ salaries were frozen, although they did not experience that freeze alone. CHC workers, including senior nurses and physicians, also had their salaries frozen from 1992 to 2003.

[28] During the period midwives experienced wage freezes, several events of note occurred.

[29] In 1999, the AOM and MOH agreed to move midwives from a salary-based compensation structure to a “course of care” compensation structure and to create an independent contractor model for midwives. While this resulted in a change to the compensation model for midwives, it did not result in compensation increases.

[30] Also in 1999, nurse practitioners received formal recognition with a new compensation level that was higher than the CHC senior nurses to which midwives had been compared in 1993.

[31] In 2004, CHC physicians obtained representation from the Ontario Medical Association (“OMA”), and for the first time they were included in the OMA’s negotiations with the MOH for physician compensation.

[32] In the meantime, in 2003, the AOM retained the Hay Group to support its negotiations with the MOH for a new funding agreement.[^4] In February 2004, the Hay Group issued a report confirming the continued appropriateness of the 1993 principles as a reasonable “equity structure”, as the Hay Group principal called it, for compensating midwives.

[33] Finally, in 2005, after 11 years of compensation restraint, the parties reached a 3-year agreement, effective April 1, 2005, which resulted in increases to midwife compensation, including a first-year increase of 20% to 29%, depending on experience level, and 1% to 2% increases in the remaining years of the contract.

(c) 2005-2013: A widening compensation gap

[34] The 2005 funding agreement expired in 2008, and the AOM and MOH commenced negotiations for a new funding agreement. By that time, the AOM had become concerned that a gender gap in compensation had developed between midwives and their CHC physician comparator.

[35] In 2009, the MOH and AOM reached a new three-year agreement, retroactive to April 1, 2008. Under this agreement, midwives received a raise of 2% per year plus an increase in benefits. The parties also agreed to a joint non-binding compensation review conducted by an independent third-party consultant to inform the next round of negotiations. In particular, the consultant was to be tasked with recommending an appropriate total compensation package for midwifery services. The parties agreed that, in recommending a total compensation package, the consultant was to consider available evidence, including:

• Comparable, relevant and historical compensation levels … of nurses, doctors and other relevant health care providers;

• Comparable and relevant midwifery compensation models in other jurisdictions; and

• The initial 1993 Morton compensation report and the February 2004 Hay compensation review report.

[36] As they had agreed, in 2010 the parties participated in a joint compensation study conducted by the Courtyard Group. The Courtyard Report is central to both the Adjudicator’s decision on liability and her decision on remedy.

[37] The Courtyard Report affirmed the ongoing relevance of the 1993 principles, including comparison with CHC physicians. The Courtyard Report found that nurse practitioners[^5] at the bottom end of their compensation range were now paid the same as level one midwives, and in some settings paid significantly more. Further, the compensation of CHC physicians was now “well above that paid to midwives.” At the time of the Courtyard Report, and as reflected in that Report, the highest paid midwife was paid $104,847 and the lowest paid CHC physician was paid $181,233. Therefore, the highest paid midwife had gone from being paid 90% to around 57% of the lowest level of pay of a CHC physician. The Courtyard Report recommended a 20% increase in compensation, referred to as an “equity adjustment”, effective April 1, 2011, to restore midwives to their historic position of being compensated at a level between senior CHC nurses (now nurse practitioners) and CHC physicians.

[38] Following the release of the Courtyard Report, the MOH raised concerns about the Report’s methodology and its recommendations, and, for the first time, advised the AOM that negotiations would be governed by the government’s policy of compensation restraint, consistent with the compensation restraint legislation that had been passed before Courtyard started its work. While the government’s compensation restraint legislation did not apply to the midwives as independent contractors, the MOH took the position that the government’s compensation restraint policy nonetheless applied.

[39] There were a series of negotiations and attempts at compromise. The MOH offered 0% increases in the first two years of the contract followed by a compensation increase in the third year of 2% with an additional 3% quality improvement incentive. This offer aligned with the compensation restraint legislation, which required no increases for the first two years and a prohibition on making up for any losses in year three of the contract. The AOM rejected the MOH’s offer.

[40] Ultimately, the AOM and MOH reached a funding agreement in 2013, but it was entered into on a without prejudice basis to the AOM pursuing legal action.

[41] In 2013, the AOM brought an application to the Tribunal under the Code, alleging that midwives had experienced systemic gender-based discrimination in compensation and seeking an increase in compensation retroactive to 1997.

analysis

(1) Standard of Review

(a) Overview

[42] On an appeal from a decision of the Divisional Court disposing of an application for judicial review, this court must “determine whether the Divisional Court identified the appropriate standard of review and applied it correctly”: Longueépée v. University of Waterloo, 2020 ONCA 830, 153 O.R. (3d) 641, at paras. 47-48. It is, in effect, a de novo review of the Tribunal’s decision: see Northern Regional Health Authority v. Horrocks, 2021 SCC 42, 462 D.L.R. (4th) 585, at para. 10; Canadian Federation of Students v. Ontario (Colleges and Universities), 2021 ONCA 553, 157 O.R. (3d) 753, at para. 20. This court steps into the shoes of the Divisional Court and focuses upon the Tribunal’s decision: see Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, at paras. 45-47. See also: Law Society of Ontario v. Diamond, 2021 ONCA 255, 458 D.L.R. (4th) 603, at para. 34; Ontario Nurses’ Association v. Participating Nursing Homes, 2021 ONCA 148, 154 O.R. (3d) 225, at para. 40, leave to appeal refused, [2021] S.C.C.A. No. 134; and B.L. v. Pytyck, 2021 ONCA 67, at para. 20.

[43] Accordingly, the first question is whether the Divisional Court identified the correct standard of review.

[44] Before the Divisional Court, the MOH submitted that the standard of review was reasonableness, based on Shaw v. Phipps, 2010 ONSC 3884, 325 D.L.R. (4th) 701 (Div. Ct.), aff’d 2012 ONCA 155, 347 D.L.R. (4th) 616, and Intercounty Tennis Association v. Human Rights Tribunal of Ontario, 2020 ONSC 1632, 446 D.L.R. (4th) 585 (Div. Ct.). In response, the Tribunal, together with the AOM, submitted that Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, 441 D.L.R. (4th) 1, had overruled Shaw v. Phipps on the question of standard of review, and that in accordance with s. 45.8 of the Code, the Adjudicator’s decision should not be set aside unless it was “patently unreasonable”.

[45] The Divisional Court accepted the MOH’s submission, concluding that Vavilov did not overrule Shaw v. Phipps. Accordingly, it applied a reasonableness standard.

[46] The parties renew their same arguments before this court. The MOH maintains the standard of review is reasonableness and argues for a robust application of the reasonableness standard in accordance with Vavilov. The AOM and the Tribunal submit that the standard of review is patent unreasonableness, and that the court “should accord the utmost deference” to the Tribunal’s decisions, given its considerable specialized expertise and the wording of s. 45.8 of the Code, which states that a Tribunal decision should not be set aside unless it is “patently unreasonable”. In the alternative, the AOM submits that a reasonableness review of the Tribunal’s decisions should be anchored “in judicial restraint and [respect for] the distinct role of administrative decision makers”: Vavilov, at para. 75.

[47] For the reasons that follow, I conclude that Vavilov does not undermine the reasoning in Shaw v. Phipps, which adopted a reasonableness standard of review for determinations of fact, the interpretation and application of human rights law, and remedial decisions. Before turning to that decision, it is helpful to first provide some statutory context, as it is relevant to the interpretation of s. 45.8 of the Code.

(b) Legislative context

[48] In December 2006, the Ontario legislature enacted significant amendments to the Code: see Human Rights Code Amendment Act, 2006, S.O. 2006, c. 30, s. 5.

[49] The amendments clarified that “the Tribunal is a specialized body whose sole task is to resolve human rights complaints” (emphasis added): Shaw v. Phipps, at para. 28. Specifically, s. 32(3), which outlines the selection process for Tribunal members, highlights the requisite expertise of Tribunal members. Section 32(3) requires candidates to be assessed on their “[e]xperience, knowledge or training with respect to human rights law and issues”, their “[a]ptitude for impartial adjudication”, and their “[a]ptitude for applying … alternative adjudicative practices and procedures”.

[50] Notably, the amendments also removed the right to appeal and added s. 45.8, which contains a privative clause and a legislated standard of review:

Subject to section 45.7 of this Act, section 21.1 of the Statutory Powers Procedure Act and the Tribunal rules, a decision of the Tribunal is final and not subject to appeal and shall not be altered or set aside in an application for judicial review or in any other proceeding unless the decision is patently unreasonable. [Emphasis added.]

[51] While the legislature enacted s. 45.8 in December 2006, the section, among others, was not proclaimed to come into force until June 30, 2008. Meanwhile, in March 2008, the Supreme Court of Canada decided Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190, which merged patent unreasonableness with reasonableness at common law. The result was that in the judicial review context there were now only two common law standards of review: correctness and reasonableness.

(c) Shaw v. Phipps

[52] In 2010, following both the enactment of s. 45.8 of the Code and the decision in Dunsmuir, the Divisional Court in Shaw v. Phipps was required to come to grips with the meaning of “patently unreasonable” under s. 45.8. The Divisional Court engaged in a complete analysis of the issue and ultimately concluded that the standard of reasonableness applied.

[53] The Divisional Court’s decision was appealed to this court. In contrast with the Divisional Court’s reasons on standard of review, this court dealt with the issue briefly in a single paragraph. It upheld the Divisional Court’s interpretation of s. 45.8, noting that all counsel agreed that the Divisional Court properly identified “reasonableness” as the appropriately deferential standard of review on an application for judicial review of the Tribunal’s decision: at para. 10. Accordingly, the analysis underpinning my conclusion on the meaning of “patently unreasonable” actually lies in the Divisional Court’s reasons in Shaw v. Phipps, not this court’s reasons.

[54] I will now review the Divisional Court’s reasons in some detail, as an understanding of them is necessary to understand my conclusion that Shaw v. Phipps is consistent with Vavilov. Importantly, as will be discussed, in interpreting s. 45.8 of the Code, the Divisional Court: (1) recognized that the reviewing court should apply the legislated standard; (2) purposively interpreted s. 45.8 of the Code; (3) considered the content of the standard in light of general principles of administrative law; and (4) addressed rule of law concerns that had weaved their way into the jurisprudence surrounding the difference between patent unreasonableness and unreasonableness.

[55] In Shaw v. Phipps, the Divisional Court was faced with two competing arguments on standard of review. The Tribunal and one of the applicants argued that the Tribunal’s decisions must not be set side unless they were “clearly irrational” or “evidently not in accordance with reason”: Shaw v. Phipps, at para. 32. The other parties submitted that a standard of review analysis under the Code should be undertaken in light of Dunsmuir, meaning that, despite the legislation using the term “patent unreasonableness”, the standard of reasonableness applied.

[56] The Divisional Court started its analysis by considering Supreme Court authority on the application of legislated standards of review. In particular, it recognized, based on R. v. Owen, 2003 SCC 33, [2003] 1 S.C.R. 779, at para. 32, that a legislature can specify the standard of review to be applied, and that absent a constitutional challenge, that standard should be applied.

[57] The Divisional Court also referred to the Supreme Court’s decision in Canada (Citizenship and Immigration) v. Khosa, 2009 SCC 12, [2009] 1 S.C.R. 339, which was decided after Dunsmuir. In Khosa, the Supreme Court considered the effect of s. 58(2)(a) of British Columbia’s Administrative Tribunals Act, S.B.C. 2004, c. 45, which provides that “a finding of fact or law or an exercise of discretion by the tribunal in respect of a matter over which it has exclusive jurisdiction under a privative clause must not be interfered with unless it is patently unreasonable”. The Divisional Court noted that Khosa, at para. 18, affirmed that “where the legislature has enacted judicial review legislation, analysis of that legislation is the first order of business”: Shaw v. Phipps, at para. 35. The Divisional Court further quoted para. 19 of Khosa, which notes that “[d]espite Dunsmuir, ‘patent unreasonableness’ will live on in British Columbia”, at least in some form.

[58] Indeed, the Divisional Court noted two British Columbia decisions that had affirmed that the legislated standard of patent unreasonableness continued to live on in British Columbia post-Dunsmuir: Manz v. Sundher, 2009 BCCA 92, 91 B.C.L.R. (4th) 219, at paras. 35-36; Victoria Times Colonist, a Division of Canwest Mediaworks Publications Inc. v. Communications, Energy and Paperworkers Union of Canada, Local 25-G, 2009 BCCA 229, 310 D.L.R. (4th) 367, at paras. 28-29.[^6] The Divisional Court also noted that in Victoria Times, the court applied the test of “irrationality” to determine whether a decision of the B.C. Labour Relations Board interpreting its constituent statute was patently unreasonable: Shaw v. Phipps, at para. 36.

[59] Against that backdrop, the Divisional Court went on to interpret s. 45.8 of the Code.

[60] Alive to the legislative history of s. 45.8, the Divisional Court recognized that at the time it was enacted (as distinct from when it was proclaimed into force), which was prior to Dunsmuir, patent unreasonableness was the most deferential standard of review according to common law principles: Shaw v. Phipps, at para. 37. However, the Divisional Court further recognized that the content of the legislated patent unreasonableness standard must be determined in light of general principles of administrative law, which Dunsmuir had changed, in accordance with para. 19 of Khosa:

Despite Dunsmuir, “patent unreasonableness” will live on in British Columbia, but the content of the expression, and the precise degree of deference it commands in the diverse circumstances of a large provincial administration, will necessarily continue to be calibrated according to general principles of administrative law. [Italics in original; underlining added.]

[61] To the Divisional Court, it was “obvious” that the legislative intent was to have the courts afford the highest degree of deference to the Tribunal in relation to its determinations of fact and the interpretation and application of human rights law because of its experience and expertise: see e.g. Code, s. 32(3). The Divisional Court explained: “It is obvious that when the Legislature enacted that standard in December 2006, the intent was to have the courts accord the same high degree of deference to the Tribunal that they accorded to other experienced and expert administrative tribunals”: Shaw v. Phipps, at para. 38. Despite recognizing that the legislative intent was for courts to give a “high degree of deference to the Tribunal”, the Divisional Court still had to consider what that meant. Specifically, the Divisional Court had to grapple with the suggestion that patent unreasonableness meant that only “clearly irrational” decisions – as opposed to merely “irrational” ones – were vulnerable to judicial reversal. In other words, the suggestion was that on judicial review, an Adjudicator’s irrational decision would have to be upheld, provided it was not clearly irrational in nature.

[62] Referring to Dunsmuir, the Divisional Court recognized two problems with the “clearly irrational” standard.

[63] First, the Divisional Court pointed to the illusory distinction between reasonableness and patent unreasonableness. As Bastarache and LeBel JJ. explained in Dunsmuir, “[l]ooking to either the magnitude or the immediacy of the defect in the tribunal's decision provides no meaningful way in practice of distinguishing between a patently unreasonable and an unreasonable decision”: at para. 41.

[64] Second, the Divisional Court recognized, relying on Dunsmuir, that “it would be inconsistent with the rule of law to uphold an irrational decision just because the irrationality of the decision was not clear enough”: Shaw v. Phipps, at para. 39, citing Dunsmuir, at para. 42.

[65] The Divisional Court acknowledged that the reasonableness standard does not contain a “spectrum or continuum of deference”: Shaw v. Phipps, at para. 40, citing Mills v. Ontario (Workplace Safety and Insurance Appeals Tribunal), 2008 ONCA 436, 237 O.A.C. 71, at para. 19. Nevertheless, as the Divisional Court further recognized, reasonableness takes its colour from the “context”. As the Divisional Court explained, that meant that the range of possible and acceptable outcomes expands or contracts depending on factors such as whether there is a privative clause, the nature of the question, and the decision-maker’s purpose and expertise: Shaw v. Phipps, at para. 40, citing Khosa, at para. 59, Mills, at para. 22, and Gerald P. Heckman, “Substantive Review in Appellate Courts Since Dunsmuir” (2009) 47:4 Osgoode Hall L.J. 751, at pp. 778-79.

[66] Ultimately, the Divisional Court emphasized the high degree of deference owed to the Tribunal, concluding as follows, at para. 41:

[R]eading the words of s. 45.8 of the Code purposively and in light of general principles of administrative law, it would follow that the highest degree of deference is to be accorded to decisions of the Tribunal on judicial review with respect to determinations of fact and the interpretation and application of human rights law, where the Tribunal has a specialized expertise. [Emphasis added.]

[67] In the case before them, this meant that “a high degree of deference [was] to be accorded to the Tribunal's determination” of liability under the Code and the appropriate remedy because “these are questions within the specialized expertise of the Tribunal”: Shaw v. Phipps, at para. 42. In other words, the decisions on liability and on remedy were to be respected “unless they [were] not rationally supported - in other words, they [were] unreasonable”: Shaw v. Phipps, at para. 42, citing Dunsmuir, at para. 42.

[68] Since Shaw v. Phipps, the Divisional Court has consistently applied reasonableness as the standard of review for decisions of the Tribunal, even post-Vavilov: see e.g. Stepanova v. Human Rights Tribunal of Ontario, 2017 ONSC 2386 (Div. Ct.), at para. 18, leave to appeal to Ont. C.A. refused, M47977 (January 19, 2018); Abbey v. Ontario (Community and Social Services), 2018 ONSC 1899, 408 C.R.R. (2d) 219 (Div. Ct.), at paras. 22, 30; Konesavarathan v. Middlesex-London Health Unit, 2019 ONSC 3879 (Div. Ct.), at para. 42, leave to appeal to Ont. C.A. refused, M50638 (November 26, 2019); Intercounty Tennis Association, at para. 45.

(d) Determining the standard of review under Vavilov

[70] In Vavilov, the majority of the court addressed two broad aspects of administrative law. First, the majority charted a new course for determining the applicable standard of review in the context of administrative decisions. Second, it provided additional guidance for reviewing decisions on a reasonableness standard: at para. 2.

[71] The majority in Vavilov reaffirmed the importance of legislative intent when determining the applicable standard of review, describing legislative intent as the “polar star” of judicial review: at para. 33. Under the Vavilov framework, the presumption is that reasonableness is the applicable standard in all cases, except where required by “clear indication of legislative intent or by the rule of law”: at para. 10. Therefore, the reasonableness standard may be rebutted where the legislature indicates an intention that a different standard should apply: at paras. 17, 32-35. In other words, Vavilov “requires courts to give effect to clear legislative direction that a different standard was intended”: at para. 32.

[72] Under the sub-heading “Legislated Standards of Review”, the majority in Vavilov devoted two paragraphs to the issue. First, at para. 34, the majority affirmed prior Supreme Court case law on the effect of legislated standards of review, including Khosa and Owen:

Any framework rooted in legislative intent must, to the extent possible, respect clear statutory language that prescribes the applicable standard of review. This Court has consistently affirmed that legislated standards of review should be given effect: see, e.g., R. v. Owen, 2003 SCC 33, [2003] 1 S.C.R. 779, at paras. 31-32; Khosa, at paras. 18-19; British Columbia (Workers' Compensation Board) v. Figliola, 2011 SCC 52, [2011] 3 S.C.R. 422, at para. 20; Moore v. British Columbia (Education), 2012 SCC 61, [2012] 3 S.C.R. 360, at para. 55; McCormick v. Fasken Martineau DuMoulin LLP, 2014 SCC 39, [2014] 2 S.C.R. 108, at para. 16; British Columbia (Workers' Compensation Appeal Tribunal) v. Fraser Health Authority, 2016 SCC 25, [2016] 1 S.C.R. 587, at paras. 8 and 29; British Columbia Human Rights Tribunal v. Schrenk, 2017 SCC 62, [2017] 2 S.C.R. 795, at para. 28.

[73] Second, at para. 35, the majority addressed the situation where a legislature has indicated that courts are to apply a correctness standard. The majority explained that where the legislature indicates the standard of review is correctness in reviewing specific questions, that standard must be applied.

[74] Accordingly, Vavilov reaffirmed that, subject to “limits imposed by the rule of law”, legislated standards of review must be respected: at para. 35. In other words, Vavilov did not change the law with respect to respecting legislated standards of review.

(e) Shaw v. Phipps is consistent with Vavilov

[75] The Tribunal and the AOM maintain that Shaw v. Phipps is now out of step with Vavilov. They argue that, by describing legislative intent as the “polar star” of judicial review, Vavilov has all but directed courts to give expression to the word “patently” in s. 45.8. While they do not definitively suggest where the light falls between unreasonableness and patent unreasonableness, the Tribunal and the AOM say there is light, and it must be respected.

[76] In my view, the Divisional Court’s approach to the interpretation of s. 45.8 in Shaw v. Phipps is entirely consistent with Vavilov. The Divisional Court in Shaw v. Phipps did exactly what Vavilov instructs us to do now.

[77] In Shaw v. Phipps, the Divisional Court engaged in a purposive interpretation of s. 45.8 that took into account legislative intent, recognizing that the legislature intended that the highest degree of deference be accorded to the Tribunal’s determination of facts, its interpretation and application of human rights law, and decisions on remedy. As Professor Paul Daly has noted in his article “Patent unreasonableness after Vavilov” (2021) 34 Can. J. Admin. L. & Prac. 167, at pp. 175-76, this interpretation of s. 45.8 can be said to respect legislative intent:

Ontario judges have considered since Dunsmuir that the goals of patent unreasonableness can be achieved through the application of the reasonableness standard. In this instance, it might be said that the assimilation of patent unreasonableness to reasonableness does no violence to legislative intent. [Emphasis added.]

[78] The Divisional Court also specifically considered case law on legislated standards of review, including Owen, and interpreted s. 45.8 in light of the general principles of administrative law, in accordance with Khosa. Notably, both Owen and Khosa were cited with approval in para. 34 of Vavilov.

[79] In addition, the Divisional Court took into account rule of law concerns in interpreting s. 45.8, which is consistent with the statement in para. 35 of Vavilov that courts are to respect legislated standards of review “within the limits imposed by the rule of law.”

[80] The Divisional Court further recognized that while reasonableness is a single standard, it takes its colour from its context. Vavilov affirmed this point, at para. 89, where the majority stated:

[T]he particular context of a decision constrains what will be reasonable for an administrative decision maker to decide in a given case. This is what it means to say that "[r]easonableness is a single standard that takes its colour from the context". [Citations omitted.]

[81] In my view, Shaw v. Phipps is also consistent with Vavilov’s stated desire to bring “greater coherence and predictability to this area of law”: Vavilov, at para. 10. It does this by avoiding the practical and theoretical difficulties of distinguishing between patent unreasonableness and reasonableness that were identified in Dunsmuir.

[82] As for the application of the reasonableness standard, both Shaw v. Phipps and Vavilov recognize that reasonableness must take into account the relevant “colour” or “constraints”, including the expertise of the Tribunal and the existence of a privative clause in s. 45.8: Vavilov, at para. 90. What this means in practice is that, when reviewing a decision from the Tribunal, judges are to apply reasonableness with the appropriate measure of judicial restraint that respects the distinct role of administrative decision-makers, in accordance with Vavilov, which includes the following guidance:

▪ “[R]easonableness review finds its starting point in judicial restraint and respects the distinct role of administrative decision makers”: at para. 75.

▪ Reviewing courts must not apply a standard of perfection when reviewing written reasons: at para. 91.

▪ Reviewing courts should pay respectful attention to the decision maker’s demonstrated expertise and application of specialized knowledge. Expertise may help explain an outcome that seems puzzling on its face: at para. 93.

▪ The history and context of the proceedings must inform the reviewing court’s reading of the reasons: at para. 94.

▪ To set aside a decision as unreasonable, “[a]ny alleged flaws or shortcomings must be more than merely superficial or peripheral to the merits of the decision.” Instead, they must be "sufficiently central or significant to render the decision unreasonable.” A decision with “sufficiently serious shortcomings” will not “exhibit the requisite degree of justification, intelligibility and transparency”: at para. 100.

▪ The reasoning must be rational and logical for the decision to be reasonable, but the analysis is not a “line-by-line treasure hunt for error”: at para. 102.

▪ Reasons should be read in light of the record and administrative regime in which they are given. Read holistically, reasons must reveal a rational chain of analysis to be reasonable. The conclusion must flow from the analysis undertaken and the record: at para. 103.

▪ Reviewing courts must not reweigh and reassess evidence; absent exceptional circumstances, the reviewing court should not interfere with factual findings of the decision maker: at para. 125.

▪ A decision maker’s failure to address key issues or central arguments may reflect a potential gap or flaw in the reasons. However, decision makers need not respond to every argument or make an explicit finding on every element leading to a conclusion. Reviewing courts cannot expect that they will: at para. 128.

[83] In sum, the standard of review of the Tribunal’s decisions remains reasonableness, although the application of the reasonableness standard is now informed by the guidance provided in Vavilov.

(2) Is the Adjudicator’s Liability Decision Unreasonable?

[84] While the MOH acknowledges that reasonableness is a deferential standard, it challenges the Adjudicator’s liability decision from every angle. The MOH argues that the Adjudicator engaged in illogical reasoning, failed to articulate a rational chain of analysis, made unreasonable factual findings, gave some evidence too much weight or too little weight, drew ungrounded inferences, ignored important expert and other evidence, reversed the burden of proof, and misinterpreted and misapplied the Code in imposing a positive obligation on the MOH.

[85] Ultimately, the question for this court is whether the Adjudicator’s decision as a whole is reasonable. While reasonableness review is not a “rubber-stamping” process and is a robust form of review, it finds its starting point in judicial restraint: Vavilov, at para. 13. As Vavilov instructs, at para. 85, “a reasonable decision is one that is based on an internally coherent and rational chain of analysis and that is justified in relation to the facts and the law that constrain the decision maker. The reasonableness standard requires that a reviewing court defer to such a decision” (emphasis added). To determine whether a decision is reasonable, “the reviewing court asks whether the decision bears the hallmarks of reasonableness – justification, transparency and intelligibility – and whether [the decision] is justified in relation to the relevant factual and legal constraints that bear on the decision”: Vavilov, at para. 99.

[86] In assessing the reasonableness of the Tribunal’s liability decision, and bearing in mind the multitude of complaints raised by the MOH, I have structured my analysis around the following key issues raised by the MOH:

(a) Do the Adjudicator’s reasons fail to reveal a rational chain of analysis?

(b) Was it unreasonable for the Adjudicator to find that gender was a factor in the compensation of midwives?

(c) Did the Adjudicator reverse the burden of proof?

(d) Did the Adjudicator unreasonably ignore the MOH’s expert evidence tendered to prove that gender was not a factor in midwives’ compensation?

(e) Did the Adjudicator unreasonably find that CHC physicians remained appropriate comparators after they became predominantly female?

(f) Did the Adjudicator unreasonably impose a positive obligation on the MOH?

[87] As I will explain, in my view the Adjudicator’s liability decision is reasonable.

(a) Do the Adjudicator’s reasons fail to reveal a rational chain of analysis?

(i) Overview

[88] The MOH argues that the Adjudicator’s reasons fail to reveal a logical chain of analysis. According to the MOH, the reasons are circular, internally incoherent, and draw peremptory conclusions. For instance, the MOH characterizes the conclusion that sex was a factor in the adverse treatment of midwives as a “peremptory conclusion that failed to reveal a rational chain of analysis.”

[89] It is helpful to start with a return to paras. 102-104 of Vavilov, where the majority explains the proper approach to assessing whether a decision is based on internally coherent reasoning. A reviewing court must not set out on a “treasure hunt” to identify missteps in the decision maker’s reasoning: Vavilov, at para. 102. Rather, the reviewing court must remain focussed on the task at hand, determining if the reasons are rational and logical by tracing the decision maker’s reasoning to see whether there are any fatal flaws in the overarching logic: Vavilov, at para. 102. A decision will be unreasonable if its reasons, “read holistically, fail to reveal a rational chain of analysis or if they reveal that the decision was based on an irrational chain of analysis”, or where “the conclusion reached cannot follow from the analysis undertaken … or if the reasons read in conjunction with the record do not make it possible to understand the decision maker’s reasoning on a critical point”: Vavilov, at para. 103. At the end of the day, the reasoning must “add up”: Vavilov, at para. 104.

[90] In my view, the Adjudicator’s reasoning does “add up”. Her reasons reveal a logical chain of analysis grounded in the record and the relevant jurisprudence in support of her key conclusion that sex was a factor in the adverse treatment that midwives experienced and the compensation gap that developed between midwives and CHC physicians after 2005.

[91] At this point, my focus is strictly on whether the Adjudicator’s reasons reveal a logical chain of analysis, not the factual or legal constraints that bear on her decision. Later in these reasons I come to a more detailed analysis of the evidence and law in support of the Adjudicator’s decision.

(ii) The discrimination claim

[92] In assessing the Adjudicator’s reasons, it is important to first understand what was and was not at issue before the Tribunal.

[93] The AOM claimed that the MOH had violated the right of midwives to equal treatment on the basis of sex under ss. 3 and 5 of the Code, which affirm that every person has the “right to contract on equal terms” and the “right to equal treatment with respect to employment” without discrimination.[^7]

[94] The AOM also relied on s. 9 of the Code, which prohibits a direct or indirect infringement of ss. 3 and 5, and ss. 11 and 12 of the Code, which prohibit “constructive discrimination”, which occurs when a neutral requirement, qualification, or factor results in discrimination, and discrimination because of association with a person identified by a prohibited ground of discrimination.

[95] In particular, the AOM claimed that the MOH violated those provisions by: (1) establishing and maintaining an inequitable compensation and funding system for midwives in Ontario; (2) providing unequal and discriminatory compensation and funding that undervalued midwives’ work; (3) failing to take proactive steps to prevent inequitable compensation and funding for a historically disadvantaged and almost exclusively female profession; (4) actively refusing to take any reasonable steps to investigate and remedy systemic gender discrimination when the AOM raised the issue; and (5) failing to address gendered integration barriers that midwives faced.[^8]

[96] Notably, the MOH’s response to the AOM’s claim was not that it had taken gender into account to ensure that midwives were not subjected to systemic gender discrimination. Rather, the MOH denied that gender was ever a factor in determining compensation for midwives.

[97] It is also notable that while the parties were at odds on many issues before the Tribunal, they largely agreed on the relevant legal principles to be applied, most importantly, the test for discrimination. The same is also true before this court.

[98] With that background in mind, as Vavilov demands, I turn to the Adjudicator’s reasons.

(iii) The Adjudicator’s application of the test for discrimination reveals a logical chain of analysis

[99] The Adjudicator, who has expertise in interpretating and applying the Code, recognized that it is to be given a broad, purposive interpretation to ensure that its purpose is fulfilled. She noted that the Code’s purpose is to remedy discrimination by focussing on the effect of the actions complained of rather than the intent of the party alleged to have discriminated.

[100] As an expert in human rights law, she was also alive to the nature of systemic gender discrimination in compensation. Referencing well-established jurisprudence, she described how deeply held attitudes about women’s work can lead employers and compensation-setters to give less value to their work, often without conscious decision-making. These unconscious attitudes are hidden and embedded in seemingly neutral compensation policies and practices.

[101] The Adjudicator articulated the well-established three-step test the complainant must meet to demonstrate prima facie discrimination, citing to the Supreme Court’s decision in Moore v. British Columbia (Education), 2012 SCC 61, [2012] 3 S.C.R. 360, at para. 33. To establish a prima facie case of gender discrimination, the claimant must show that:

(1) they are a member of a group protected by the Code;

(2) they have been subjected to adverse treatment; and

(3) their gender was a factor in the adverse treatment.

[102] The Adjudicator recognized that under the third prong of the test an applicant must only prove that there is a connection between the prohibited ground and the adverse treatment. The connection does not need to be causal. And the connection can co-exist with other non-discriminatory factors, meaning sex need not be the only or even the predominant factor. Further, the Adjudicator recognized that the applicant does not need to prove that the respondent intended to discriminate. In support of these propositions, the Adjudicator cited to Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Bombardier Inc. (Bombardier Aerospace Training Center), 2015 SCC 39, [2015] 2 S.C.R. 789, at paras. 43-52, and Peel Law Association v. Pieters, 2013 ONCA 396, 116 O.R. (3d) 81, at paras. 59-60.

[103] The Adjudicator also explicitly recognized that the burden to prove discrimination on the balance of probabilities lay with the AOM and that the case must be decided on the totality of the evidence. The Adjudicator then went on to consider whether the AOM had proven discrimination under the three-part test.

[104] The MOH conceded that the first part of the test was met, as midwives are almost exclusively women and therefore have a characteristic protected from discrimination under the Code. The Adjudicator noted that at the time of the hearing there was one male midwife.

[105] The Adjudicator also held that the second and third parts of the test were met for the period from 2005 to 2013. She ultimately concluded as follows:

I have concluded, on the balance of probabilities and on the totality of the evidence, that there is sufficient evidence from which to infer that midwives experienced adverse treatment and that sex is more likely than not a factor in the treatment they experienced and the compensation gap that has developed between midwives and CHC physicians since 2005. [Emphasis added.]

[106] In reaching these conclusions, the Adjudicator rejected the MOH’s submission that gender had never been a factor in setting midwives’ compensation.

[107] The Adjudicator recognized from the outset of her analysis that she could not presume a connection between gender and adverse treatment solely from context, citing Bombardier, at paras. 69, 88. At the same time, she was alive to the social context of this claim. As she stated, “[t]he negative effects of gender on the compensation of sex-segregated workers are well known.”

[108] Importantly, based on the evidence before her, the Adjudicator found as a fact that in 1993 the AOM and the MOH were both “aware of the pervasive nature of system[ic] discrimination in compensation, the stereotypes associated with women’s work and the necessity to ensure that women are paid by reference to objective factors like SERW.”

[109] Not only were the MOH and the AOM aware of the problem, but the Adjudicator found as a fact that the parties did something about their joint concerns in this regard: they agreed to the 1993 principles that were embodied in the Morton Report. Appropriate compensation of midwives was to be based on their relative skill, effort, responsibility, and working conditions as compared to other health care professionals. The Report recommended a relative positioning of midwives between primary care nurses and physicians working in CHCs, who the Adjudicator found served as a male comparator. The Adjudicator found that these principles were adopted to ensure that midwives’ compensation was not negatively affected by traditional assumptions and stereotypes about the value of women’s work.

[110] The Adjudicator recognized the importance of using CHC physicians as comparators. Indeed, she found that comparisons for midwives “were based on work historically done by men in order to ensure that their compensation corresponded with the work itself and not the gender of the person doing the work.”

[111] The fact was that, at the time of regulation, physicians were male-dominated and were associated with work historically done by men. While CHC physicians were more than 50% female by 2001, the Adjudicator found that CHC physicians remained a relevant comparator after that time. CHC physicians were family physicians who simply worked in a particular setting, a fact that was recognized by the MOH and the OMA, who worked to harmonize the compensation of CHC physicians with their peers. As one MOH witness testified, “it became important to be able to say that a primary care physician is a primary care physician is a primary care physician”. And, even by 2013 when the AOM brought its claim, family physicians remained male dominated.

[112] As noted by the Adjudicator, that was not to say that CHC physicians and midwives were the same or that the alignment between the compensation of the two groups could not change over time.[^9] Even so, the CHC physicians and the midwives continued to share an overlapping scope of practice that was comparable. The parties had agreed CHC physicians were an appropriate comparator at regulation, and this had been confirmed since by the Hay and Courtyard Reports. Indeed, the MOH continued to promote family physicians and midwives as comparable obstetrical providers, equally competent to care for women with non-high-risk pregnancies. Furthermore, the MOH had not produced a job evaluation to suggest that midwives and CHC physicians were no longer comparable for compensation purposes.

[113] For the Adjudicator, the adverse treatment experienced by midwives started after the 2005 agreement was concluded, as midwives gradually lost the connection to the 1993 principles. It culminated after the release of the Courtyard Report, when the MOH withdrew from the process and advised the AOM that the 2010 negotiations would be governed by compensation restraint. It was at that time that the MOH all together abandoned the 1993 principles that had informed negotiations with the AOM since 1993.

[114] The Adjudicator found that, while not every difference in treatment will amount to discrimination, the MOH walking away from the principles specifically designed to prevent gender discrimination, without replacing them with anything else, and imposing compensation restraint on the “sex-segregated workers” was clearly disadvantageous to them, resulting in “adverse treatment”. She found that this satisfied the second prong of the test.

[115] The Adjudicator found that the third prong of the test was also met, as there was sufficient evidence from which she inferred that sex was more likely than not a factor in the adverse treatment that midwives experienced after 2005.

[116] In particular, the Adjudicator found that the MOH’s response to the Courtyard Report constituted sufficient evidence from which to infer that gender was more likely than not a factor in the adverse treatment experienced by midwives. She was not satisfied that there were good reasons for the MOH to abandon the 1993 principles, which had informed the Courtyard Report, following the Report’s release.

[117] Importantly, the study had proceeded with the full cooperation of the parties and an active steering group made up of equal numbers of representatives of the MOH and the AOM. The parties were asked for input on a draft version of the Courtyard Report. Then, after the release of the Report, the MOH unilaterally concluded that CHC physicians were no longer an appropriate comparator, without conducting any study to validate its position. It walked away from the Courtyard Report – which meant it was also walking away from the 1993 principles – even though it had agreed in the last round of bargaining to conduct a joint compensation study that would inform the next round of collective bargaining. And, while the MOH raised criticisms of the final Courtyard Report, the Adjudicator found that they were minor and could easily have been remedied. As found by the Adjudicator, the fact is that it was only once the MOH saw the final Courtyard Report’s recommendation of a 20% compensation increase for midwives (an increase that still would have been less advantageous than what the 1993 principles called out for) that the MOH did an about face and said that compensation restraint policy would apply.

[118] The Adjudicator found that at this time the MOH was aware of the risk that the AOM might bring a claim of discrimination. As she pointed out, the Negotiations Branch, which was leading discussions on behalf of the MOH, adverted to an “outside risk” in a slide presentation that the AOM could bring an “equity issue forward” under the Code, although it noted that “the relationship between midwives and obstetricians was not clear.”

[119] The Adjudicator found that the Courtyard Report, born out of a joint effort between the MOH and the AOM, was sufficiently compelling for the MOH to realize that the AOM’s claim of gender discrimination might have some validity. She also found the MOH refused to take reasonable steps to respond and failed to monitor the compensation of midwives for the impact of gender discrimination, which undermined the MOH’s argument that gender was not a factor in their compensation levels.

[120] The Adjudicator concluded that, having walked away from the Courtyard Report (and, therefore, the 1993 principles) the MOH’s failure to take any steps to address concerns raised about discrimination in midwife compensation was a factor from which an inference of discrimination could be drawn. The Adjudicator rejected the MOH’s submission that differences in compensation paid to CHC physicians and midwives were based solely on occupational differences and labour market forces, and that gender was not a factor. There was no evidence that the compensation of physicians was tied to an analysis of their skill, effort, responsibility, and working conditions. Rather, CHC physicians were given increases because of recruitment and retention issues and after they gained representation from the OMA, leading to the harmonization of their compensation with other physicians.

[121] In summary, the Adjudicator concluded that gender was a factor in the adverse treatment of midwives for a number of reasons, including that the MOH abandoned the principles designed to safeguard against gender discrimination, after receiving a report confirming the ongoing relevance of the principles, and without a credible explanation for doing so or a new methodology to ensure that midwives were being paid appropriately.

[122] In my view, the Adjudicator’s reasons read holistically do logically “add up” and reveal a rational chain of analysis to support her conclusion that midwives were subjected to discrimination.

(b) Was it unreasonable for the Adjudicator to find that gender was a factor in the compensation of midwives?

[123] The MOH points to what it says are two interrelated errors in the Adjudicator’s reasons: (1) it was unreasonable for the Adjudicator to find that the 1993 principles were connected to or imbued with gender; and (2) it was unreasonable to draw an inference that gender was a factor from the Courtyard Report because the Report did not provide a gender-based analysis. In my view, neither argument can succeed, especially given Vavilov’s admonition to reviewing courts not to reweigh or reassess evidence considered by the decision maker.

[124] First, the MOH submits that there is no evidence to support the Adjudicator’s finding that “the original funding principles which were agreed on by the parties, followed by the joint working group and incorporated into the [Ontario Midwifery Program] framework, are connected if not imbued with gender.” Indeed, it says that none of the relevant documents fundamental to establishing the 1993 principles, including the work of the joint working group or the Morton Report, refer to “sex”, “gender”, “pay equity”, or a “male comparator”.

[125] The MOH made this same argument before the Divisional Court, which held that the submission overlooks the “overwhelming evidence” that the Adjudicator relied on to come to the conclusion that the 1993 principles were connected to or imbued with gender. I agree.

[126] The evidence before the Adjudicator included evidence that did reference pay equity and gender. It also included other evidence from which the Adjudicator could reasonably infer that the 1993 principles were designed to address the issue of systemic gender discrimination in the compensation of a highly sex-segregated group.

[127] For instance, Jane Kilthei, the AOM President at the time of regulation, testified that “the issue of equity, equity for women, equity for midwives was the water we swam in.” She also testified that while she knew that the 1993 joint working group was not engaged in a technical job evaluation under the Pay Equity Act, R.S.O. 1990, c. P.7, she understood it was engaged in a “pay equity exercise”. The term “pay equity exercise” was also how the joint working group process was described to the AOM’s members when they were asked to ratify the results of the process.

[128] Similarly, there was evidence from Margaret McHugh, who was hired by the MOH’s WHB in 1992[^10] as its Midwifery Implementation Coordinator, that the parties were concerned with pay equity at the time of regulation. She prepared an “Options Paper” on compensation for midwives, which was approved by her boss, Assistant Deputy Minister Jodey Porter. The paper noted the “[n]ecessity to establish a fair and equitable pay level based on pay equity, reflecting responsibilities, working conditions and level of education” (emphasis added). Ms. McHugh testified before the Tribunal that while they were not doing a formal pay equity assessment under the Pay Equity Act, she understood “pay equity” to mean it was necessary to ensure that midwives were not discriminated against in their compensation:

[W]omen had historically been underpaid and their work had been undervalued, and if we were going to establish a brand new, female exclusive-almost profession, that we had to ensure that that profession was not going to be discriminated against or that there wouldn't be bias against their payment method just by looking at other female-dominated professions and kind of going, "Oh, well, you know, you should be paid a small amount since you're women." So we had to make sure that that happened. It didn't necessarily mean that we were going to do a formal pay equity assessment under the [Pay Equity] Act. It meant that we were going to make sure that we were not underpaying midwives, that they were fairly and equitably paid according to their skills and experience and education, and not according to somebody's picking out something. It was going to be evidence-based.

[129] Ms. McHugh testified that she did not recall anyone “pushing back” on the issue of pay equity. She stated that there was focus on ensuring that midwives were not underpaid "simply because it was a female-dominated profession". She also testified more generally about her work on the use of gender analysis at the time in the development of health policy.

[130] The Adjudicator was alive to this evidence and the evidence of the context giving rise to the 1993 principles. At the time, the parties were aware of the history of discrimination against midwives, as set out in the Task Force’s report. She found that given this history, comparing midwifery with work historically done by men was significant to overcoming the stereotypes that would have “undoubtedly” affected midwives’ initial compensation levels.

[131] The Adjudicator also found that it was reasonable for the AOM to perceive the methodology leading to the 1993 principles – specifically an analysis of skill, effort, responsibility, and working conditions and the use of a male comparator – as a pay equity exercise because it corresponded with the Pay Equity Act.

[132] The Adjudicator ultimately found that it was reasonable for midwives to have operated from the perspective that they were engaged in a pay equity exercise. The fact that not every single person involved at the time shared that perspective did not undermine the effect of the principles, which, as the Adjudicator described, “worked against the prevailing stereotypes about midwifery work and its association with women.”

[133] In my view, the Adjudicator’s finding that gender was a factor in the development of the 1993 principles is reasonable in light of the evidence before her.

[134] Second, the MOH similarly argues that the inference drawn by the Adjudicator that gender was more likely than not a factor based on the Courtyard Report was unreasonable because “[t]he Courtyard report was not an investigation into whether midwives had experienced sex discrimination.” The MOH highlights that John Ronson, the author of the Courtyard Report, was not an expert in pay equity or gender-based analysis, that Mr. Ronson testified that his recommendation was based on a “generalized sense of fairness”, and that the Courtyard Report did not use equity in the sense of pay equity but rather “as a lawyer would use it … as in equitable remedies.” In the MOH’s submission, a recommended raise in the Courtyard Report is not an indicator of sex discrimination.

[135] The Adjudicator drew reasonable inferences based on the evidence. As the Divisional Court noted, the extraction of “one or two answers” from the cross-examination of Mr. Ronson had the effect of taking that testimony out of context. The fact is that, when considered in its proper context, and in the context of the evidence as a whole, it was open to the Adjudicator to find that the Courtyard Report affirmed that gender was a factor in midwives’ compensation.

[136] As noted above, the parties agreed in 2009 that a third-party consultant would conduct a review with the primary goal of suggesting an appropriate “total compensation” package for midwives. The parties agreed that the consultants would consider:

• Comparable, relevant and historical compensation levels … of nurses, doctors and other relevant health care providers;

• Comparable and relevant midwifery compensation models in other jurisdictions; and

• The initial Morton compensation report and the February 2004 Hay Compensation review report. [Emphasis added].

[137] The Courtyard Report affirmed the ongoing relevance of the 1993 principles, stating: “The compensation model principles established in the Morton Report of [1993], which have evolved somewhat since that time, appear to have served the public, the profession and the Ministry very well. There appears to be no appetite or need to change the fundamental model of compensation.” Importantly, it was not until after the release of the final Courtyard Report that the MOH suggested that the 1993 principles would no longer have any relevance.

[138] Having reasonably found that the 1993 principles were connected, if not imbued, with gender, it was open to the Adjudicator to find that the Courtyard Report, which affirmed those principles, indicates “that gender discrimination may be an operative factor in the compensation of midwives”. It was also open to the Adjudicator to reasonably find that the Courtyard Report was “sufficiently compelling for the MOH to realize that the AOM’s claim of gender discrimination may have some validity.”

(c) Did the Adjudicator reverse the burden of proof?

[139] The MOH argues that the Adjudicator reversed the burden of proof by requiring it to prove, with a job evaluation, that midwives and CHC physicians were “not comparable” for compensation purposes. The appellant accepts that it did not call an expert to, as the Adjudicator put it in her remedy decision, “validate, one way or the other, whether midwives remained appropriately paid despite increases paid to CHC physicians.” The MOH says that it was under no obligation to do so and to insist otherwise constitutes a reversal of the burden.

[140] The MOH rightly notes that it was the AOM’s responsibility to prove that midwives’ pay was discriminatory on the basis of sex. The AOM attempted to do so through the Durber Report, but the Adjudicator rejected that evidence for the purposes of determining liability. According to the MOH, having done so, she erred in finding discrimination on the basis that the MOH did not prove a lack of discrimination.

[141] The AOM submits there is no merit to the burden of proof argument, noting that the MOH could have easily tendered a gender-based study of its compensation practices to meet the evidential burden that it faced, but it failed to do so as it feared the results of doing another study. The AOM highlights an internal MOH draft note from a Labour Relations Steering Committee, from July 2011, that was before the Tribunal, in which the Ministry advised against another compensation review because “there is merit to the claim that midwives deserve a significant increase after several years of no or minimal compensation increases.” The draft note stated that “[a] second review will not likely achieve a much lower recommended amount. A second report carried the risk of another 20% recommendation”.

[142] In my view, there was no reversal of the burden of proof in this case. The suggestion to the contrary conflates the evidential burden with the ultimate burden.

[143] On multiple occasions, the Adjudicator made clear that she understood who held the ultimate onus in this case. For instance, she said that the onus of proving discrimination “lies with the AOM and that the standard of proof is the balance of probabilities”. She later reinforced her clear understanding that in a human rights case, “the burden of proof remains on the applicant throughout.” She also quoted from the Pieters decision, placing particular emphasis on the following sentence: “The question to be decided is whether the applicant has satisfied the legal burden of proof of establishing on a balance of probabilities that the discrimination has occurred”: Pieters, at para. 83.

[144] Not only did the Adjudicator understand that the burden to prove discrimination remained on the midwives throughout, but she kept it right there. All that shifted was the evidential burden.

[145] The Adjudicator found that the AOM had successfully established a prima facie case of discrimination. As previously described, that prima facie case was rooted in, but not limited to, the fact that, in 2010, the MOH essentially walked away from 1993 principles without replacing them with anything to ensure fair and appropriate compensation for midwives, in a situation where, as the Adjudicator found, both parties had previously agreed in 1993 that a comparator was critical to avoiding discrimination. The evidential burden then shifted to the MOH to provide a credible explanation for why sex was not “a factor” in the adverse treatment of the midwives. She explained:

The traditional analysis is often described in this way: the applicant has the evidential burden to prove a prima facie case; once a prima facie case is established, the evidential burden shifts to the respondent to prove a credible, non-discriminatory explanation which rebuts the prima facie case; the evidential burden shifts back to the applicant to prove that the respondent’s explanation is pre-textual.

[146] This statement of law accords with this court’s decision in Pieters. As noted by Juriansz J.A., at paras. 73-74:

In discrimination cases … the law, while maintaining the burden of proof on the applicant, provides respondents with good reason to call evidence. Relatively “little affirmative evidence” is required before the inference of discrimination is permitted. And the standard of proof requires only that the inference be more probable than not. Once there is evidence to support a prima facie case, the respondent faces the tactical choice: explain or risk losing.

If the respondent does call evidence providing an explanation, the burden of proof remains on the applicant to establish the respondent’s evidence is false or pretextual.

[147] While the MOH says that it did explain rather than risk losing, read as a whole, the Adjudicator’s reasons demonstrate that she found that the MOH failed to meet the evidential burden or, in other words, failed to provide a credible explanation for why sex was not “a factor” in the adverse treatment of midwives. The fact that the MOH failed to tender a study to validate its proposition that midwives’ compensation remained free from the effects of gender discrimination was simply one reason why the Tribunal found that the MOH’s evidence of non-discrimination was not persuasive to fully explain the compensation gap. This was not a reversal of the burden of proof.

[148] The MOH also submits, relying on Pieters, at paras. 72-74, that where a respondent calls evidence providing a non-discriminatory explanation for the treatment, the applicant then holds the burden of proving that the respondent’s evidence is “false or a pretext”. The MOH argues that in the absence of any finding that the evidence that Ontario led was incredible, unreliable, false or a pretext, it was unreasonable for the Adjudicator to discount the reasonable explanations it provided to explain the compensation gap between CHC physicians and midwives.

[149] The MOH’s argument cannot succeed. The Adjudicator recognized that gender as “a factor” can co-exist with other non-discriminatory factors. She found that while the MOH’s explanations were not unreasonable, the MOH failed to meet its evidential burden to rebut that gender was also “a factor”. Therefore, the MOH’s explanations did not fully explain the compensation gap. As noted above, the Adjudicator found that there was no evidence that the compensation of CHC physicians was tied to their skills, effort, responsibilities, and working conditions. Rather, they were given increases because of recruitment and retention issues and the harmonization of their compensation with other physicians. In a context where midwives had had their compensation set by comparison to CHC physicians, who served as a male comparator, the MOH could only defend in part what it paid to midwives by explaining why it gave increases to CHC physicians.

[150] As the Tribunal recognized, the connection between gender and the impugned treatment need not be an exclusive one.

[151] In summary, the Adjudicator knew that the ultimate burden remained on the AOM throughout, and she kept it there. All that shifted to the MOH was the evidential burden, which the Adjudicator found was not met.

(d) Did the Adjudicator unreasonably ignore the MOH’s expert evidence that was tendered to prove that gender was not a factor in midwives’ compensation?

[152] The MOH next argues that the Adjudicator erred by failing to engage with its expert evidence in her liability decision. That evidence is said to demonstrate that gender was not a factor in midwives’ compensation. In the MOH’s submission, the Adjudicator unreasonably ignored this evidence. I disagree.

[153] There was a mountain of evidence before the Adjudicator. The AOM relied on the evidence of 20 factual witnesses and 4 experts. The MOH relied on the evidence of 14 factual witnesses and 6 experts. The experts and witnesses testified over approximately 50 days. The complete record included thousands of pages of transcripts, affidavits, exhibits, and submissions by the parties.

[154] The appellant is right that the Adjudicator did not engage in any detailed way with the expert evidence in her reasons on liability. As she said in the liability decision, while the expert evidence would “very likely” be relevant to remedy, it was not necessary to rely on that evidence for purposes of liability. The Adjudicator explained:

I have decided this case on the facts that were presented to me, the application of the legal principles which govern human rights adjudications and pay equity decisions describing the historic factors which affect women’s compensation. I did not find it necessary to rely on any of the experts in coming to my decision on liability. The expert evidence will very likely be relevant to remedy. A number of experts agreed that a job evaluation should be undertaken.

The Adjudicator then went on to explain why she was not relying on the evidence of particular experts.

[155] The AOM tendered the evidence of Paul Durber, a pay equity specialist, to prove discrimination. He conducted a pay equity evaluation of midwifery compensation dating back to 1994. The Adjudicator found that she could not accept Mr. Durber’s methodology to support a finding of liability because the effect of doing so “would be to retroactively impose the statutory obligations under the Pay Equity Act onto the MOH.” It was reasonable for her to reject Mr. Durber’s report on that basis.

[156] The Adjudicator found that it was also unnecessary to rely on the evidence of the AOM’s expert Hugh Mackenzie at the liability stage. Mr. Mackenzie, an economist, was retained to determine the implications of Mr. Durber’s pay equity analysis for midwives’ compensation.

[157] Of the MOH’s six experts, three were retained to rebut Mr. Durber’s evidence. Robert Bass, a pay equity specialist, provided a detailed critique of Mr. Durber’s pay equity analysis, as did Dr. John Kervin, a sociology professor with expertise in wage discrimination. Dr. Richard Chaykowski, a professor with expertise in economics and industrial relations, critiqued both Mr. Durber’s and Mr. Mackenzie’s expert evidence. Having rejected Mr. Durber and Mr. Mackenzie’s evidence, the Adjudicator said she did not need to rely upon their evidence in reply.

[158] The Adjudicator also found it unnecessary to rely on the expert evidence of AOM expert Dr. Ivy Bourgeault, a medical sociologist with expertise in midwifery and gender, and the MOH’s fourth expert, Dr. Candace Johnson, a political science professor with expertise in maternal health policy and gender, who critiqued Dr. Bourgeault’s evidence. As the Adjudicator explained, the gendered history of midwifery and the history of regulation were described by the factual witnesses in the proceeding. Factual witnesses and the Task Force’s report addressed the “structural embeddedness of medical dominance” and the caring dilemma[^11] associated with midwifery work, the issues addressed by the reports.

[159] The Adjudicator also explained that it was unnecessary to rely on the MOH’s other two experts, Dr. David Price and Dr. Lisa Graves, who testified about the training and work of family physicians and the challenges they faced over the past 20 years. While she acknowledged that their evidence would be relevant to anyone conducting a job evaluation comparing CHC physicians to midwives for compensation purposes, that was not her role in determining whether gender was a factor in the adverse treatment of midwives. She also noted that no one disputed Dr. Price’s comment that there had been an explosion of medical knowledge.

[160] In my view, the Tribunal’s treatment of the MOH’s expert evidence was reasonable when understood in the context of the litigation and the decision.

[161] The Adjudicator declined to delve into evidence that was placed before her to critique the Durber and Mackenzie reports, which she declined to accept because of its flawed methodology. She also preferred to rely on factual evidence instead of expert evidence that spoke to the same issues. It was open to her to do so. And, most fundamentally, she did not ignore the evidence that the MOH says demonstrated non-discriminatory reasons for the gap in compensation between CHC physicians and midwives. For instance:

• The Adjudicator accepted that there was an explosion of medical knowledge, which increased the complexity of family medicine and the demands on family physicians.

• She discussed the shortage of CHC physicians and the related recruitment and retention issues.

• She recognized that bargaining strength was a factor in the midwife compensation, but noted that this too was a gendered issue. She explained that midwives’ bargaining strength “depends in large part on the MOH recognizing the connection between midwifery and gender and being informed about the effects of gender on the compensation of sex-segregated workers.”

• She noted that there was no dispute that midwives and physicians are different, which was recognized when the differences between them were valued in 1993, and she recognized that the exact alignment between the two could change over time, but the fact remained that their compensation should reflect the overlapping scope of their work.

[162] In the end, while the Adjudicator recognized these factors, she simply did not accept that gender was not also a factor, alongside the explanations provided by the MOH, in the adverse treatment of midwives.

[163] At the end of the day, the MOH has not pointed to any expert evidence that could explain away the central findings of fact that drove the conclusion of discrimination: (1) in 1993, the parties agreed to equitable compensation principles that were designed to ensure that midwives’ compensation was not affected by harmful assumptions and stereotypes concerning the value of women’s work; (2) the 1993 principles were connected, if not imbued, with gender; (3) the 1993 principles were affirmed in the Hay Report in 2004; (4) by 2010 there was a significant compensation gap between midwives and their comparators, inconsistent with the 1993 principles; and (5) despite the 1993 principles being reinforced yet again under the jointly commissioned Courtyard Report, the MOH turned its back on the principles and did not substitute them with a new methodology for determining the appropriate and fair compensation of midwives. Taking into account these findings and others, the Adjudicator concluded that gender constituted “a factor” in the adverse treatment experienced by midwives. The MOH did not point us to any evidence that fully rebutted these findings of fact.

[164] In conclusion, I am satisfied that the Adjudicator’s treatment of the MOH’s expert evidence that the MOH tendered to explain the reasons for CHC physician increases was reasonable.

(e) Did the Adjudicator unreasonably find that CHC physicians remained an appropriate comparator after they became predominantly female?

[165] The MOH challenges the Adjudicator’s finding that CHC physicians remained an appropriate comparator, given that the Adjudicator found that CHC physicians have been more than 50% female since at least 2001 and by 2004 were predominantly female.

[166] In my view, the Adjudicator’s finding that CHC physicians remained an appropriate comparator is reasonable.

[167] I have already summarized her reasoning on this point. To recap, while CHC physicians became female-dominant over time, they were family physicians who worked in a particular setting. After 2004 when the OMA started bargaining on their behalf, their compensation was harmonized with the compensation of the larger group of family physicians, who remained more than 50% male even in 2013. As previously noted, one witness testified that it became important to say that “a primary care physician is a primary care physician is a primary care physician” no matter what setting they were working in. Thus, in accordance with the “a physician is a physician is a physician” philosophy, CHC physicians became financially aligned with other family physicians, leaving a gulf between the compensation of midwives and CHC physicians. In other words, CHC physicians remained a male comparator in 2013, even though they were predominately women, because their pay had been aligned with a male dominated group.

[168] Further, as the Adjudicator recognized, the Code does not prescribe rules to determine the sex of an occupational group or any methodology for developing a compensation model that is Code-compliant. In this case, the parties chose CHC physicians as a male comparator and the Adjudicator reasonably found that they remained a male comparator until 2013.

(f) Did the Adjudicator unreasonably impose a positive obligation on the MOH?

[169] The MOH repeats many of the same arguments that it made before the Divisional Court in arguing that the Tribunal unreasonably imposed a positive obligation on the MOH to compare midwives’ compensation to that of CHC physicians. The MOH’s objections include that the Adjudicator misapprehended the evidence, referred to Ontario Human Rights Commission (“OHRC”) policies without citing any particular policies, failed to meaningfully grapple with the MOH’s arguments, and imposed a novel legal duty. The Divisional Court addressed the thrust of these arguments and concluded that the Adjudicator’s decision was reasonable. I agree and will dispense with the MOH’s key arguments briefly.

[170] The Adjudicator referred to the requirement to act proactively in two places in her liability reasons. First, under the heading “Failing to Resolve the ‘Flaws’ in Courtyard”, the Adjudicator found that the “[t]he adverse impact on midwives of losing the connection to the 1993 principles [was] compounded by a failure on the part of the MOH to take reasonable steps to respond to the AOM’s allegations that their compensation was falling behind based on the original funding principles.” It was in that context that the Adjudicator first noted the requirement to act proactively to monitor workplace systems and to take preventive measures to identify and remedy the adverse impacts of practices and policies that appear neutral.

[171] Later in her reasons, under the heading “Proactive Prevention”, the Adjudicator noted that the MOH admitted that it had taken no proactive steps to monitor the compensation of midwives for the impact of gender discrimination on the fairness of their compensation. By contrast, the MOH monitored the compensation of CHC physicians for evidence of recruitment and retention issues and to ensure their compensation was aligned with other physicians.

[172] While the Adjudicator found that the MOH was not required to engage in any particular strategy to monitor, identify, and redress discrimination in the compensation of midwives, it was required to take steps that were effective and proportional to its obligations under the Code to both prevent and remedy discrimination. The Adjudicator found that the lack of proactivity was most evident in the lack of regular negotiations between the AOM and the MOH, and the long gap between joint compensation studies. In the face of having agreed in 1993 that a comparator was necessary to fend off discrimination in the context of the gender trifecta in this case – women serving women in relation to women’s reproductive health – the failure to monitor, identify, and redress discrimination was “just one factor” from which the Adjudicator drew “an inference of discrimination”.

[173] Notably, the Adjudicator did not find that the failure to monitor was in and of itself a breach of the Code. Rather, it was one of many facts that informed her rejection of the MOH’s submission that gender was not a factor in midwives’ compensation. After all, the MOH could not point to any proactive attempts to monitor midwives’ compensation for the impact of gender discrimination and further failed to take reasonable steps to respond when the AOM raised the issue of discrimination. This was the case in spite of the fact that, as far back as the time of regulation, the MOH was aware of the risk of systemic gender discrimination creeping in, an acknowledgement that found expression in the 1993 principles. In the context of this case, it was reasonable for the Adjudicator to rely on the MOH’s failure to monitor as a basis for concluding that gender was a factor in the adverse treatment of midwives.

[174] To the extent that the Adjudicator made more general comments about proactive duties under the Code, nothing turns on them in this particular case, given her narrow factual findings. I would leave for another day how far the Code goes in imposing proactive obligations.

(3) Is the Adjudicator’s Remedy Decision Unreasonable?

[175] The MOH devoted little attention to the issue of remedy, either in its factum or in oral submissions. I intend to approach the matter in a commensurate manner.

[176] The MOH argues that, even if the Adjudicator reasonably found discrimination, the remedial orders were unreasonable because liability was based on a “novel obligation not previously recognized in any court or Tribunal decision applying the Code.” In these circumstances, citing to Canada (Attorney General) v. Hislop, 2007 SCC 10, [2007] 1 S.C.R. 429, at paras. 99-100, the MOH says “it was unjust to order retrospective remedies.”

[177] Even assuming for the purpose of argument that Hislop applies to a claim for retroactive wage adjustments found to be owing to workers as a result of discrimination under the Code, the first point has already been answered in the reasons above: the Adjudicator did not impose a novel duty on the MOH.

[178] The MOH also argues that it was unreasonable for the Adjudicator not to have discounted the monetary remedies she ordered to account for flaws in the Courtyard Report or the fact that the CHC physicians received salary cuts after 2012. Respectfully, this argument is misplaced.

[179] In the liability decision, the Adjudicator provided the parties with an opportunity to sort this matter out without further input from the Tribunal. Indeed, she strongly encouraged the parties to do so. Unfortunately, the parties failed to resolve the matter between themselves.

[180] In asking the Tribunal to determine the remedy, the parties declined to present further evidence from a compensation expert to inform the remedial award. Instead, they urged the Tribunal to determine a remedy on the evidence that was led at the hearing. This left the Adjudicator without a compensation study that addressed the many questions that were raised during the proceeding about how midwives should be compensated.

[181] In these circumstances, the Adjudicator was left to decide the matter on the best available evidence. She concluded that the best evidence of the consequences of the move away from the 1993 principles was the jointly commissioned Courtyard Report. According to the Adjudicator, implementing Courtyard would “[bring] the parties as close as possible” to the “place they would have been but for the discrimination”.

[182] In reaching that conclusion, she explained in detail why she rejected the MOH’s argument that a number of deficiencies rendered the Courtyard Report unreliable for the purposes of determining remedy. In addition to addressing the specific criticisms, the Adjudicator also fairly noted that the Courtyard review was an iterative process and the MOH had every opportunity to participate through the steering committee and to review draft reports. In addition, the MOH had an opportunity to provide additional evidence or commission other studies but had declined to do so.

[183] The Code provides the Tribunal with broad remedial discretion to order remedies that are fair, effective and responsive to the circumstances of the particular case. In exercising her remedial discretion, the Adjudicator fashioned a remedy based on the evidence that was before her. The MOH has not pointed to any legitimate basis for interfering with the Tribunal’s discretionary remedial decision.

conclusion

[184] I am satisfied that the Adjudicator’s decisions on both liability and remedy bear the hallmarks of reasonableness – justification, transparency and intelligibility. They are justified in relation to the relevant factual and legal constraints that bear on them. The reasons are also transparent and intelligible. Accordingly, I would dismiss the appeal.

[185] The parties may submit written argument in respect of the costs of the appeal. The submissions will not exceed five pages. The parties will file their submissions within ten days of the release of these reasons.

Released: “June 13, 2022 JMF”

“Fairburn A.C.J.O.”

“I agree. L.B. Roberts J.A.”

“I agree. Van Melle J. (ad hoc)”

Appendix A – Chronology of Key Events

1985-1986:

▪ The Task Force on the Implementation of Midwifery in Ontario is established.

▪ The Ontario government commits to the enactment of pay equity legislation.

1987:

▪ The Midwifery Task Force releases its report. The report notes that “[t]he remuneration paid to midwives should fairly reflect their level of responsibility, the demands on their time, the difficulty of their work, the cost of participating in continuing education activities, and the cost of professional liability insurance.” It suggests positioning midwives between the starting salary for a nurse with a baccalaureate degree and the fees physicians were paid under OHIP for pregnancy, labour, birth and postpartum care.

▪ Ontario’s Pay Equity Act is passed. It becomes effective January 1, 1988.

1989:

▪ The Interim Regulatory Council of Midwives (the “Council”) is appointed to develop standards of practice.

1991:

▪ The Ontario government introduces the Midwifery Act, 1991, S.O. 1991, c. 31, and the Regulated Health Professions Act, 1991, S.O. 1991, c. 18, which come into force on January 1, 1994.

1992-1993:

▪ Throughout 1992 and into 1993, the AOM develops its “Principles of Funding” to help guide its negotiations with the MOH. They include the principles that “[t]he funding arrangement must acknowledge midwives as autonomous practitioners. It must reflect midwives' level of skill and responsibility in the provision of primary care, her education at a baccalaureate level, the realities of working on call and the time intensive nature of midwifery care. In addition it must deal with the costs of running a practice.”

▪ In June 1992, the Council releases its report, entitled “Models of Payment and Practice Committee: Report and Recommendations”. The Report recognizes the importance of ensuring that midwives are paid equitably in keeping with their role as primary care providers. It notes that the fundamental measurements to determine dollar equity are skill, education, working conditions and degree of responsibility. It also notes that it had been suggested that midwife remuneration should fall between that of senior salaried nurses and family physicians.

▪ In the fall of 1992, the Women’s Health Bureau develops an internal “Options Paper” to assist in identifying appropriate options for public funding of midwifery. It notes the “[n]ecessity to establish a fair and equitable pay level based on pay equity, reflecting responsibilities, working conditions and level of education.” Around the same time, the MOH creates a document entitled “Principles of funding midwifery in Ontario”. It sets out the principles of funding that should be recognized, including that “financial compensation fall between the level of a family practitioner and a senior salaries nurse”.

▪ In December 1992, the Minister of Health announces that the Ontario government is committed to managing and funding midwifery services. The MOH decides that the program will be housed in the Ministry’s Community Health Branch, the branch responsible for community-based managed health care, including CHCs.

▪ A Joint Working Group, comprised of MOH and AOM representatives, is created to determine payment levels for midwifery services. It starts to meet in May 1993. As it works through funding issues, it also drafts the Ontario Midwifery Program Framework, which forms the basis for the Ontario Midwifery Program.

▪ The Joint Work Group enlists the help of Robert Morton and Associates, consultants, to assist in establishing an appropriate salary range. In July 1993, the final Morton Report is released. It uses the terms “appropriate” and “fair” as guiding principles. “Appropriate” is defined as setting a compensation range that reflects the relative skill, effort, responsibility and working conditions for midwives in comparison to related health care professions. The term “fairness” recognizes that compensation should be determined in relation to other professionals working in the same economic market. The Report notes the agreement on the relative positioning of midwifery in relation to primary care nurses and family practitioners in CHCs.

▪ The government adopts the 1993 Ontario Midwifery Program Framework. The framework is formally approved by the AOM in October 1993. The salary range for a midwife is $55,000 to $77,000. This places an entry-level midwife’s salary around the top salary of a CHC senior nurse, and the highest compensation level for a midwife at approximately 90% of the base salary of an entry level CHC family physician.

1994:

▪ Midwifery becomes a regulated health profession in Ontario on January 1, 1994.

1998-1999:

▪ Nurse practitioners receive formal recognition.

▪ A government report identifies a significant shortage of obstetricians and gynecologists.

▪ The AOM and the MOH enter into a second funding agreement. It changes the funding structure and creates the independent contractor model that still exists, but does not result in compensation increases for midwives.

2000-2001:

▪ The AOM asks for a compensation increase based on the cost-of-living allowance, but the request is denied.

▪ The MOH significantly increases the budget for midwifery services, but none of this money is budgeted for compensation increases.

▪ The MOH initiates a strategic review of the CHC program. It is recommended that the CHC program institute more competitive salaries.

2003:

▪ CHC staff receive their first compensation increases after 11 years of wage freezes.

▪ The MOH initiates an evaluation of the midwifery program, which affirms that it has been successful in terms of the services it provides.

▪ The AOM commissions a report from the Hay Group to support its negotiations with the MOH for a new funding agreement. The Hay Group Report affirms the comparison with CHC physicians and nurse practitioners.

2004:

▪ CHC physicians are included for the first time in the OMA’s negotiations with the MOH for physician compensation.

▪ The AOM initiates a campaign entitled “Because Storks Don’t Deliver Babies”. It warns the MOH that midwives are prepared to engage in job action.

2005:

▪ The AOM and MOH enter into a new funding agreement, the third such agreement between the parties. Midwives receive their first increase after 11 years of wage freezes. The total compensation increase is 20% to 29%, depending on experience.

2009:

▪ The AOM and the MOH reach a new funding agreement, the fourth such agreement between the parties. The parties agree to conduct a joint but non-binding compensation study to be completed before the next round of negotiations.

2010:

▪ In March, the government introduces compensation restraint legislation, which applies to public sector employees.

▪ In July, the Courtyard Group is retained to conduct a joint-compensation study – the first one since 1993. A joint steering committee is formed to support the work of the consultants.

▪ In October 2010, the final Courtyard Report is released. It recommends a 20% increase in compensation, described as a “one-time equity adjustment”. It notes that “[t]he compensation model principles established in the Morton Report of 199[3], which have evolved somewhat since that time, appear to have served the public, the profession and the Ministry very well.”

▪ At the time of the Courtyard Report, and as reflected in that Report, the highest paid midwife was paid $104,847 and the lowest paid CHC physician was paid $181,233. Therefore, the highest paid midwife had gone from being paid 90% to around 57% of the lowest level of pay of a CHC physician.

▪ After the release of the Courtyard Report, the MOH advises the AOM that compensation restraint will apply to negotiations with the AOM.

2013:

▪ The AOM and the MOH reach a funding agreement that accords with the compensation restraint policy. The agreement is subject to the AOM’s right to bring legal action.

▪ The AOM files an application with the Human Rights Tribunal of Ontario, alleging systemic gender discrimination in compensation.

[^1]: While midwives and their clients are referred to as “women” and “female” throughout this decision, this terminology should not be read as suggesting that all midwives and their clients identify as women.

[^2]: The AOM claimed discrimination on the prohibited ground of sex. In this decision, I have used the word “gender” as well as “sex” in describing the prohibited ground of sex.

[^3]: At para. 67 of the liability decision, the Adjudicator notes that the government established the Task Force in 1985. The “Report of the Task Force on the Implementation of Midwifery in Ontario” suggests that the Government announced its intention to establish the Task Force in 1986. Nothing turns on this fact.

[^4]: Like the Adjudicator, I use the term “funding agreement” rather than “compensation agreement” because the AOM negotiates with the MOH as independent contractors over the delivery of midwifery services in Ontario.

[^5]: As the Adjudicator notes, when nurse practitioners were regulated, they had a larger scope of practice than registered nurses and their salary scale was positioned above that of the CHC primary care nurses, which were the nurses that midwives had been compared to in 1993. However, the 1993 principles evolved slightly in 2004 when the Hay Group Report recommended midwives’ compensation fall somewhere between a nurse practitioner and CHC family physician (although closer to the physician), based on their independent review of the role descriptions of midwives, CHC nurse practitioners, and CHC family physicians.

[^6]: Courts in British Columbia have continued to apply the standard of patent unreasonableness. Examples after the time Shaw v. Phipps was decided include United Steel Workers, Paper and Forestry, Rubber, Manufacturing, Energy Allied Industrial and Service Workers International Union, Local 2009 v. Auyeung, 2011 BCCA 527, 345 D.L.R. (4th) 630, at paras. 65-73; and Pacific Newspaper Group Inc. v. Communications, Energy and Paperworkers Union of Canada, Local2000, 2014 BCCA 496, 68 B.C.L.R. (5th) 57, at paras. 44-48, leave to appeal refused, [2015] S.C.C.A. No. 60. Professor Paul Daly explains that pre-Vavilov jurisprudence from British Columbia applies the patent unreasonableness standard in a “highly deferential manner”. He also suggests that it is arguable that the statutory definition in the B.C. legislation (which gives direction as to how patent unreasonableness is to be applied in that context) and the common law definition have merged in B.C. Professor Daly further identifies principles that courts in B.C. have used in defining the standard, including that to be patently unreasonable a decision must be “evidently unreasonable”: “Patent unreasonableness after Vavilov” (2021) 34 Can. J. Admin. L. & Prac. 167, at pp. 169-72.

[^7]: As the Adjudicator noted, there was no dispute between the parties that the issues related to the compensation of midwives fall within the broad definition of “with respect to employment” despite midwives’ status as independent contractors.

[^8]: The Adjudicator found that she had insufficient evidence to consider the fifth allegation, which is not relevant to this appeal.

[^9]: Not only could change occur, but the Courtyard Report reflected that fact. Indeed, while the Courtyard Report reinforced the importance and appropriateness of the 1993 principles, the compensation recommendation arising from that Report would not have returned midwives to their exact historical position under those principles.

[^10]: While the Adjudicator states that Margaret McHugh was hired in the summer of 1993, her testimony and other evidence suggest she was hired in the summer of 1992. Nothing turns on this fact.

[^11]: As Dr. Bourgeault describes the caring dilemma, it refers to the “tension between providing high quality care for [midwives’] clients, particularly in accordance with the continuity of care element of the midwifery model of practice, and being able to maintain familial responsibilities.”