An Overview of the Comparator Group Analysis in Human Rights Jurisprudence

Prepared for the Council of Canadians with Disabilities CURA "Disabling Poverty/Enabling Citizenship" by Nathan Irving (LL.B. student, University of Manitoba), September 6, 2009

I. Introduction

This report was prepared as a backgrounder for a larger future research project that intends to explore the impact of the human rights principle of the "duty to accommodate" on persons with disabilities. It is one of five reports prepared for further research purposes.

The report addresses a relatively recent trend in Canadian human rights jurisprudence: the importation of the comparator group analysis from the section 15 Charter[1] context into the analytical framework for statutory human rights adjudication. The comparator group analysis, which requires claimants to prove that they have been treated differently than a specific group which mirrors their characteristics save for the alleged ground of discrimination, has been sharply criticised for being antithetical to substantive equality. The details of this argument and the pitfalls of the comparator group analysis are provided elsewhere,[2] and are not the focus of this report. Suffice it to say that this development is particularly troubling for persons with disabilities and other equality-seeking groups who have spent years fighting for judicial recognition that equality is not always achieved by same treatment, but rather often requires that true differences be taken into account and accommodated so as to render society inclusive and accessible.

II. The Role of the Comparator Group Analysis

The topic of this report is the role that the comparator group analysis is playing in the statutory human rights jurisprudence since the Supreme Court of Canada issued its rulings in two landmark human rights cases, Meiorin[3] and Grismer.[4] Specifically, the questions addressed include the following: Where has the comparator group analysis appeared? How has it been applied by courts, tribunals, boards, and labour arbitrators? Has it played a significant role in the jurisprudence? Has it been a serious obstacle to complainants who have alleged discrimination in employment and in the provision of services?

To answer these questions, I reviewed all post-Meiorin and post-Grismer court cases which involved statutory human rights claims and at least some analysis of discrimination and/or the duty to accommodate, totalling 118 court decisions.[5] In addition, I reviewed tribunal and labour arbitration cases for the years 2008 and 2009 which involved both allegations of discrimination on the basis of disability and at least some analysis of discrimination and/or the duty to accommodate, totalling eighty-one tribunal and fifty-five arbitral decisions.[6]

The comparator group analysis came up in only a minority of these cases. Of the 118 court decisions reviewed, the comparator group analysis was mentioned, discussed, and/or applied in just twenty. Of the eighty-one tribunal decisions, the analysis was mentioned, discussed, and/or applied in just two. And of the fifty-five arbitral decisions, only four mentioned, discussed and/or applied the comparator group analysis.

Of the twenty court cases where the comparator group analysis was raised, fourteen involved allegations of discrimination in employment,[7] while seven involved claims of discrimination in the provision of services.[8] Thus, 15% of the employment court cases (14/91) referred to the comparator group analysis, while 25% of the service cases (7/28) made some mention of the analysis. At the tribunal level, these numbers were much smaller. Of the two tribunal decisions invoking the comparator group analysis, one involved allegations of discrimination in employment[9] and one was a service case.[10] Thus, less than 2% of employment-related tribunal decisions (1/61) referred to the comparator group analysis, while 5% of service-related tribunal decisions (1/20) mentioned the analysis. Finally, just over 7% of the arbitral decisions (4/55) made reference to the comparator group analysis.[11]

To get a better sense of where the comparator group analysis cropped up and how it was applied in different contexts, I divided the cases into the following three categories: discrimination in services; discrimination in employment (termination); and discrimination in employment (benefits). The latter category encompassed claims of discrimination in the provision of an assortment of employee benefits, such as bonuses, severance pay, and disability benefits. Many of the latter cases involved allegations of discrimination when an employee did not receive the same benefits as other employees because of disability-related absences.

Claimants who challenge the termination of their employment as a human rights violation can take some comfort in knowing that the comparator group analysis has not been an obstacle to complainants in cases of this kind. None of the four court cases,[12] one tribunal case,[13] or one arbitration case[14] which referred to the comparator group analysis and fell within this category was defeated by the comparator group analysis (although some of these cases were unsuccessful for other reasons). This was the only category of cases which did not include a single instance where a claim was defeated by the comparator group analysis. Moreover, the decision-makers in these cases who referred to the comparator group analysis downplayed its significance, rejected its application, or simply skirted the analysis by deciding the case on other grounds. An example of a decision that downplayed the significance of the comparator group analysis is Ottawa (City) v. Canada (Human Rights Comm.) (No. 2).[15] In this case, the claimant was dismissed for chronic innocent absenteeism. On the matter of the comparator group analysis, the court concluded that "counsel correctly devoted very little attention to this issue. Whichever comparator group was chosen, the conclusion that there was differentiation established in the treatment of the complainant would not be affected."[16] Another example of a court decision which downplayed the significance of the comparator group analysis is Ottawa (City) v. Ottawa-Carleton Public Employees' Union, Local 503,[17] a decision of the Ontario Divisional Court which addressed the termination of an employee who had a mental disability. In response to the parties' arguments regarding the correct application of the comparator group analysis, the Court simply concluded that the employer's taking into account of days missed due to disability was "prima facie discrimination on its face, when measured against any proposed comparator group."[18]

Other courts which addressed allegations of discriminatory termination went even further. They outright rejected the application of a comparator group analysis in this context. In ADGA Group Consultants Inc. v. Lane,[19] a case which arose after an employee with bipolar disorder was terminated following a manic episode, the Ontario Divisional Court agreed with the Commission that a comparator group analysis is inappropriate in cases where a person with a disability seeks accommodation of his or her needs. The Court went on to state the following:

[i]n cases of disability in the employee termination context, it is not necessary or appropriate to have to establish a comparator group. Disability cases bring with them particular and individualized situations. Once it is established that the termination of the employee was because of, or in part because of, the disability, the claimant has established a prima facie case of discrimination. The onus then shifts to the employer to establish that it met its duty of procedural and substantive fairness to the point of undue hardship.[20]

In Kemess Mines Ltd. v. I.U.O.E., Local 115,[21] the British Columbia Court of Appeal also rejected a comparator group analysis as unnecessary, but on the basis of the language employed in the human rights statute. After reviewing the relevant provisions, the Court concluded that "there is therefore no need to conduct a comparative analysis. It is, by definition, prima facie discriminatory for an employer to refuse to employ or continue to employ a person because of a physical or mental disability."[22]

By contrast, in employment cases involving allegations of discrimination in the distribution of employment benefits, where the comparator group analysis was raised by the adjudicator, it acted as a much greater obstacle for complainants. Of the nine court cases which touched upon the comparator group analysis and fell within this category,[23] the comparator group analysis defeated three.[24] The comparator group analysis also defeated one of the three arbitral decisions involving employment benefits where the comparator group was raised.[25] In the instances where the comparator group analysis defeated a claim in this category of cases, not one decision maker selected persons without disabilities or the able-bodied norm as the appropriate comparator group. In two cases, the "correct" comparator group was comprised of other persons with disabilities.[26] Because these groups were treated no differently (i.e. no worse) than the claimant, the claims were defeated. Furthermore, the comparator groups that were selected were described at a high level of specificity, often as a result of decision-makers' insistence that the comparator group be selected "in light of the underlying rationale of the benefit at issue."[27] A good illustration of this is British Columbia Government and Service Employees' Union v. British Columbia (Public Service Employee Relations Commission).[28] This case involved a challenge to the discontinuation of LTD benefits following six months of unsuccessful job search assistance from a committee. The majority of the British Columbia Court of Appeal agreed with the arbitrator that

the purpose of the Rehabilitation Committee is to return to work, as soon as possible, those employees who have been injured or ill, and that the appropriate comparator group is between employees with disabilities who cannot return to their own job but are placed elsewhere within six months, and those who cannot return to their own job and are not placed elsewhere within six months.[29]

In employment benefit cases which involved some mention of comparator groups but were not defeated by this analysis, decision makers approached the comparator group analysis in a myriad of ways. In the majority of these cases, a comparator group analysis was conducted but nothing turned on it because the claims were decided on other grounds. However, in a recent 2009 employment benefits case heard by the British Columbia Court of Appeal, the Court ruled that the comparator group analysis was unnecessary. The case was Communications, Energy and Paperworkers' Union of Canada, Local 789 v. Domtar Inc.[30] While the Court ultimately ruled against the complainant, the decision may set a precedent with regard to the need (or lack thereof) to perform a comparator group analysis in the employment benefits context:

The LTD employees were clearly suffering from a disability. The question for the arbitrator was whether they had received any adverse treatment. He concluded that they did not. It was not necessary to his conclusion that he identify a "comparator group", and on my reading of his reasons, he did not seek to do that.[31]

The final category of cases is the service cases. Of the seven service court cases which made reference to the comparator group analysis, only one was defeated by it.[32] This goes against suspicions that the comparator group analysis would be particularly problematic in this context. While a greater proportion of service court cases than employment court cases referred to the comparator group analysis (25%:15%), the proportion of service cases which were defeated by the comparator group analysis was nearly equal to the proportion of employment cases defeated by this analysis (3.6%:3.3%). This suggests that rights claimants, lawyers, and researchers should not focus on the analysis within service cases to the exclusion of the employment context.

That being said, the comparator group analysis has done its fair share of damage within the service context, and it clearly commands the attention of rights claimants, lawyers, and academics. Perhaps the most glaring illustration of its potential to thwart claimants in the service context is a 2008 decision of the British Columbia Supreme Court in Moore.[33] In this decision, the Court's re-definition of the service in issue as well as the "correct" comparator group ultimately defeated the claim to equality in education. As in certain aforementioned employment benefit cases, the Court re-defined the comparator group as being comprised of other students with disabilities. Since there was no evidence upon which to find that other students were treated better than the claimant (to be more exact, the claimant's son), the Court did not find that a case of discrimination had been made out.

As with the employment benefit cases, the outcomes of the remaining service cases were determined by other factors. I should also note that two of these cases alluded to possible changes in the legal landscape in the aftermath of a 2009 Supreme Court of Canada decision, R v. Kapp.[34] In Tranchemontagne, the Ontario Superior Court had this to say:

The [Supreme Court] it seems, has responded to the criticisms it hinted at in Kapp, including academic criticism about the judicial tool of a "comparator group" becoming a barrier as a result of its inflexible application. This tool had been used in discrimination analysis in a way that suggested that there might be only one appropriate comparator group, or that the entire case could stand or fall based on a single comparison. As is the case with respect to the dignity requirement in the third Law step, the Supreme Court is moving away from the requirement to identify a specific comparator group against which the discriminatory effect of impugned legislation is to be assessed. At the very least, it would appear the court is moving away from the necessity for a formalistic comparator-group analysis. The Supreme Court's analysis in Kapp and Ermineskin should not be taken as an indication that comparisons are irrelevant to equality inquiries. Indeed, the court did make a comparison in Ermineskin. What the court did not do, though, was conduct a strict comparator analysis as it had done previously on numerous occasions. It would seem from the court's most recent pronouncements on the issue that a discrimination claim does not turn on identification of a single correct comparator group. However, comparison may nonetheless be useful, for example in the initial stage of discrimination analysis, as evidence of the fact of distribution of benefits or burdens among different individuals or groups.[35]

It remains to be seen, of course, whether this change in language will have practical ramifications. Contrary to what the above passage might suggest, the Court in Tranchemontagne proceeded to identify the "correct" comparator group, which as in so many cases cited above, was comprised of other persons with disabilities.

III. Conclusions regarding Disability Claims

In conclusion, while the comparator group analysis has created obstacles in the employment and service contexts, and while courts in a number of disability rights cases have continued to define and re-define narrow, highly specified, comparator groups—which frequently consist of other persons with disabilities thereby forestalling a substantive equality approach to claims—there is some indication that the courts might start a shift away from a rigid comparator group analysis.

• [1] Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.)., 1982, c. 11 [Charter].
• [2] See e.g. Daphne Gilbert & Diana Marjury, "Critical Comparisons: The Supreme Court of Canada Dooms Section 15" (2006) 24 Windsor Y.B. Access Just. 111; Diane Pothier, "The Conundrum of Comparators" in S. McIntyre & S. Rodgers, eds., Strategizing Systemic Inequality (Toronto: Irwin Law); and Andrea Wright, "Formulaic Comparisons: Stopping the Charter at the Statutory Human Rights Gate" in Fay Faraday, Margaret Denike, & M. Kate Stephenson, eds., Making Equality Rights Real: Securing Substantive Equality Under the Charter (Toronto: Irwin Law, 2006).
• [3] British Columbia (Public Service Employee Relations Commission) v. British Columbia Government and Service Employees' Union, [1999] 3 S.C.R. 3, 35 C.H.R.R. D/257 [Meiorin].
• [4] British Columbia (Superintendant of Motor Vehicles) v. British Columbia (Council of Human Rights), [1999] 3 S.C.R. 868, 36 C.H.R.R. D/129 [Grismer].
• [5] This statement is qualified by the fact that I only reviewed the highest level court decision for each case.
• [6] I did not review decisions arising from preliminary motions, including motions to dismiss. While time did not allow for their review, it might be wise to review these decisions in future research, for it is possible that rights claims are being dismissed at the outset as a result of the comparator group analysis.
• [7] See Ottawa (City) v. Canada (Human Rights Comm.) (No. 2), 2005 FCA 311, 54 C.H.R.R. D/462 [Ottawa]; Canada Safeway Ltd. v. Alberta (Human Rights and Citizenship Comm.), 2003 ABCA 246, 47 C.H.R.R. D/220 [Safeway]; British Columbia v. Hutchinson (No. 2), 2005 BCSC 1421, 54 C.H.R.R. D/468 [Hutchinson]; ADGA Group Consultants Inc. v. Lane (2008), CHRR Doc. 08-524 (Ont. Div. Ct.) [ADGA]; Canada (Human Rights Comm.) v. Canadian National Railway Co. (No. 2) (2000), 38 C.H.R.R. D/107 (F.C.T.D.) [CNR]; Saskatchewan (Dept. of Finance) v. Saskatchewan (Human Rights Comm.), 2004 SKCA 134, 50 C.H.R.R. D/457 [Saskatchewan]; Ottawa (City) v. Ottawa-Carleton Public Employees' Union, Local 503, [2007] O.J. No. 735 (Ont. Div. Ct.) [Ottawa-Carleton]; Kemess Mines Ltd. v. I.U.O.E., Local 115, 2006 BCCA 58, CHRR Doc. 06-906 [Kemess]; Communications, Energy and Paperworkers' Union of Canada, Local 789 v. Domtar Inc., 2009 BCCA 52, 2009 CarswellBC 283 [Communications]; Canada (Attorney General) v. Buffett, 2007 FC 1061, 78 Admin. L.R. (4th) 54 [Buffett]; Ontario Secondary School Teachers' Federation v. Upper Canada District School Board, [2005] O.J. No. 4057 (Ont. Sup. Ct.) [Secondary School]; British Columbia Government and Service Empoyees' Union v. British Columbia (Public Service Employee Relations Commission), 2005 BCCA 129, 251 D.L.R. (4th) 73 [British Columbia]; Pisoney v. London Life Insurance Co., 2003 BCSC 1116, 19 B.C.L.R. (4th) 91 [Pisoney]; and Real Canadian Superstore v. United Food and Commercial Workers, Local 1400, 1999 SKQB 196, 182 D.L.R. (4th) 223 [Superstore].
• [8] One case was considered to be both an employment and a service case. See Hutchinson, ibid. Also see Canada (Human Rights Comm.) v. M.N.R., 2003 FC 1280, 49 C.H.R.R. D/38 [M.N.R.]; Ontario (Attorney General) v. Ontario (Human Rights Comm.) (2007), 62 C.H.R.R. D/315 (Ont. Div. Ct.) [Ontario]; British Columbia (Ministry of Education) v. Moore (No. 2), 2008 BCSC 264, 62 C.H.R.R. D/289 [Moore]; Van der Smit v. Alberta (Human Rights and Citizenship Comm.), 2009 ABQB 121, CHRR Doc. 09-0522 [Van der Smit]; Ontario (Director, Disability Support Program) v. Tranchemontagne, [2009] O.J. No. 1613 (Ont. Sup. Ct.) [Tranchemontagne]; and Gwinner v. Alberta (Human Resources and Employment), 2002 ABQB 685, 217 D.L.R. (4th) 341 [Gwinner].
• [9] Merrick v. Ipsco Saskatchewan Inc. (No. 3), (2008) CHRR Doc. 08-954 (Sask. H.R.T.) [Merrick].
• [10] J. and J. on behalf of R. v. British Columbia (Children and Family Development) (No. 2), 2009 BCHRT 61, (2009) CHRR Doc. 09-0232 [J. and J.]
• [11] Toronto (City) v. Toronto Professional Fire Fighters' Assoc. (Policy Grievance), [2009] O.L.A.A. No. 251 [Toronto]; Providence Care v. Ontario Public Service Employees Union, Local 431 (Thomas Grievance), [2008] O.L.A.A. No. 198 [Providence]; Saskenergy Inc. v. Communications, Energy and Paperworkers Union of Canada, Local 649 (Zielenin Grievance), [2008] S.L.A.A. No. 10 [Saskenergy]; and British Columbia v. British Columbia Government and Service Employees' Union (Lenihan Grievance), [2008] B.C.C.A.A.A. No. 69 [Lenihan].
• [12] See Ottawa, supra note 7; ADGA, supra note 7; Ottawa-Carleton, supra note 7; and Kemess, supra note 7.
• [13] See Merrick, supra note 9.
• [14] See Lenihan, supra note 11.
• [15] Ottawa, supra note 7.
• [16] Ibid. at para. 16.
• [17] Ottawa-Carleton, supra note 7.
• [18] Ibid. at para. 68.
• [19] ADGA, supra note 7.
• [20] Ibid. at paras. 94-96.
• [21] Kemess, supra note 7.
• [22] Ibid. at para. 30.
• [23] See Safeway, supra note 7; CNR, supra note 7; Saskatchewan, supra note 7; Communications, supra note 7; Buffett, supra note 7; Secondary School, supra note 7; British Columbia, supra note 7; Pisoney, supra note 7; Superstore, supra note 7.
• [24] See CNR, ibid.; British Columbia, ibid.; and Superstore, ibid.
• [25] See Toronto, supra note 11. The other two arbitral decisions in this category were Saskenergy, supra note 11; and Providence, supra note 11.
• [26] See British Columbia, supra note 7; and Toronto, ibid.
• [27] Superstore, supra note 7. This case is a particularly good illustration of how complex and convoluted the analysis has become given the need to select a comparator group that mirrors the characteristics of the claimant in light of the rationale of the benefit sought.
• [28] British Columbia, supra note 7.
• [29] Ibid. at para. 17.
• [30] Communications, supra note 7.
• [31] Ibid. at para. 37.
• [32] See Moore, supra note 8.
• [33] Ibid.
• [34] R. v. Kapp, 2008 SCC 41, [2008] 2 S.C.R. 483 [Kapp]. See J. and J., supra note 10; and Tranchemontagne, supra note 8.
• [35] Tranchemontagne, ibid. at paras. 112-113.