Tranchemontagne — Statutory Challenges to Statutory Enactments: What is the Appropriate Standard?

The Tranchemontagne series of cases have received some previous attention on this site. In this note, however, I want to focus on one of the more significant, if understudied, aspects of the 2009 Ontario Disability Support Program v. Tranchemontagne Ontario Divisional Court decision. [1] In this decision the Court directly addresses a growing trend in equality jurisprudence – that of using the provincial human rights statutes to challenge the legislative content of governmental programs, rather than proceeding by way of a section 15 Charter equality claim. In this case the Divisional Court pronounced itself on an issue that has plagued such cases over the last few years: what is the appropriate legal test to be used in deciding human rights statutory challenges to other statutory provisions? Should it be the constitutional Law v. Canada (Minister of Employment and Immigration[2] and R. v. Oakes[3] test used for Charter equality claims, or the well established statutory human rights Ontario Human Rights Commission and O’Malley v. Simpsons-Sears Ltd. [4] and British Columbia (Public Service Employee Relations Commission) v. BCGSEU(Meiorin) [5] tests? Prior to this decision, the use of Law and Oakes in the non-constitutional context appeared to be spreading beyond challenges to legislated programs, in a manner that seemed to weaken human rights protections in Canada.

What legal test to apply in such cases is of profound significance for equality jurisprudence. Cases such as Tranchemontagne have arisen because of the increasing difficulties faced by equality-seeking groups in bringing social and economic claims under section 15 of the Charter, leading claimants to seek remedies elsewhere. Application of the Law and Oakes tests in the human rights context would stymie this option by importing a jurisprudential frame that imposes higher burdens on claimants than those specifically designed for the statutory context. Given the confusion in this area of law, the Divisional Court’s decision has been appealed to the Ontario Court of Appeal, which heard oral arguments in March of this year and is set to render its decision on this case in the upcoming term.

Facts

The 2009 Tranchemontagne decision is the latest in a decade-long saga that began with two claimants’ applications for benefits under the Ontario Disability Support Program (ODSP).  Robert Tranchemontagne and Norman Werbeski each separately applied for ODSP benefits and were denied on the basis that they both suffered from a ‘sole impairment’ within the meaning section 5(2) of the Ontario Disability Support Program Act (ODSPA), which bars claimants from benefit entitlement if their living restrictions are based solely on drug or alcohol addiction.  The claimants appealed the denial on the basis, inter alia, that section 5(2) of the ODSPA is contrary to the concept of disability in the Ontario Human Rights Code (OHRC), and therefore must give way to the Code’s primacy over other legislation. The case was appealed up to the Supreme Court on jurisdictional issues, and eventually came back to the Social Benefits Tribunal (SBT), which ruled on the merits that both claimants were ‘persons with disabilities’ under section 4(1) of the ODSPA, and that section 5(2) was in fact discriminatory and inconsistent with section 1 of the Ontario Human Rights Code. The issue of section 5(2)’s consistency with the OHRC was then appealed by the Director to the Ontario Divisional Court, which is the decision under consideration.

Issue

Before getting into the specifics of the decision, let us stop to consider how this question arose in the first place. While the answer may seem obvious to human rights practitioners, a moment’s pause reveals how strange these types of claims are. Why are the claimants using one statute, the Ontario Human Rights Code, to challenge the content of another statute, instead of bringing a section 15 Charter equality claim, arguing that section 5(2) of the OPSPA violated their constitutional equality rights? And how does the OHRC, concerned primarily (thought not exclusively) with relationships between private parties, apply to the dispute at all?

The Charter prohibits discrimination on the basis of enumerated and analogous grounds in regards to governmental action. The provincial human rights statutes also applies only to certain types of relationships, but instead of being primarily concerned with governmental action, the Codes more broadly prohibits discrimination within certain specified and mostly private relationships, and do so on specifically enumerated grounds.[6] One of the regulated relationships under the human rights statutes is in the provision of services (sometimes more specifically limited to ‘services customarily available to the public’). ‘Services’ have generally been considered to include legislated government benefit programs, although it is only recently that the definition of ‘service’ has received much jurisprudential consideration.[7] It is by way of this regulated relationship that human rights statutes have come to bear on legislation apportioning public benefits, and to the legislative mechanisms by which they are distributed. Thus, the human rights statutes are being applied not only to the actual delivery of services, but to the content of legislation which apportions the benefits or to the criteria used to determine service eligibility. While we ordinarily think that statutory enactments in a parliamentary democracy can only be challenged on the basis of constitutionality, all human rights statutes contain supremacy clauses (such as section 47(2) of the OHRC), which state that their terms prevail over other statutes that ‘purport to require or authorize conduct’ that is inconsistent with their terms. As such, the human rights statutes are considered quasi-constitutional. Statutory provisions that provide a service can thus be challenged where they require or authorize conduct inconsistent with the requirements of a provincial human rights statute. It is in this manner that the OHRC was argued to apply to the ODSPA in the Tranchemontagne case, because both parties agreed (without much discussion) that ODSPA income support constitutes a service within the meaning of the Code.

We may assume that the claimants chose the statutory path for a few reasons: firstly, most claimants taking the human rights route do so because of the perceived formality and formulaic nature of the section 15 analysis over the past few decades, as well as the time and cost involved, which has made a constitutional equality claim the last legal resort for many. The second reason, we might assume, is that addiction is more clearly defined as a mental health disability under the Code than under the Charter’s jurisprudence.

Why is the human rights route viewed as more friendly to complainants than a constitutional claim? First and foremost, because since at least the 1999 Law v. Canada Supreme Court decision, the legal tests for establishing discrimination under the Charter and under the Codes have been different, with the human rights test from O’Malley setting a standard that is arguably easier for claimants to meet. But, in the years since claims about the content of legislation began being brought to human rights decision-makers, this has begun to change.

Over the last years government lawyers have appeared before human rights adjudicators and argued that the constitutional Law test, and not O’Malley, is the appropriate case to apply when there is a human rights challenge to legislation. In addition, they have often argued that the Charter’s section 1 Oakes test should be applied to determine the justifiability of an infringement, rather than analyzing the existence of a statutory defence such as a bona fide occupational requirement (BFOR), ordinarily dealt with through the Meiorin case’s framework.  Human rights adjudicators, and judges hearing such cases on appeal, have often favoured the application of the Law/Oakes test, viewing that test as more appropriate when the respondent party is a government actor, rather than deciding based on the legal instrument under which the claim is brought. [8]

Prior to the 2009 Tranchemontagne decision, the intrusion of the Law/Oakes test into the human rights framework had begun to stray past claims against legislative benefits. In one case a respondent hockey association argued for the application of Law/Oakes in a gender discrimination claim for the exclusion of female hockey players from the association.[9] In another, the British Columbia Court of Appeal broadly applied the Law framework in a cursory fashion to reviewing a discrimination claim between a union and a public sector employer.[10] Unlike many decisions on the question, however, the 2009 Divisional Court decision, penned by Justice Bellamy, sets itself the task of explicitly deciding whether the same legal test exists for constitutional and statutory discrimination decisions, or whether their frameworks of analysis are indeed different.

Which Legal Test Applies?

Before the Tribunal the arguments around the appropriate legal test centred firstly, on whether Law or O’Malley was controlling in the human rights context, and secondly, if the Law test was controlling, whether its human dignity analysis had to be applied.

For the Divisional Court, Justice Bellamy had two things to say about the appropriate legal test: first and most importantly, she affirmed that O’Malley is the controlling test for the statutory discrimination analysis, which is not the same process as under section 15 of the Charter. She rightly points out that in its first consideration of constitutional equality in Andrews in 1989, the Supreme Court drew inspiration from existing human rights jurisprudence and the O’Malley test, but also outlined the differences between the two instruments. The Supreme Court has continued to use the O’Malley case to analyze statutory human rights claims since that time, despite there being numerous opportunities to merge the analysis under both instruments. For instance, the Court applied the O’Malley framework just six months after deciding the 1999 Law case in Meiorin, and again in Grismer some 9 months after Law, and has continued to do so since that time.

Secondly, despite reaffirming the difference in analytic structure under the two instruments, Justice Bellamy goes on to modify the existing O’Malley test in the case of legislated services and benefits by suggesting that it should be ‘informed’ by the constitutional Andrews decision. In effect, this appears to move O’Malley closer to the constitutional approach. Historically the human rights analysis has been relatively categorical. A claimant must establish a prima facie violation by showing a distinction based on a prohibited ground. The onus then shifts to the respondent to demonstrate that the violation is a bona fide requirement or qualification (BFOQ), the establishment of which renders a distinction non-discriminatory. But in Bellamy J.’s reformulation, in the alternative to demonstrating the existence of a BFOR/BFOQ, the respondent will now also have the opportunity to demonstrate that the “distinction does not create a disadvantage by perpetuating prejudice or stereotyping”. This addition to the analysis is drawn from constitutional equality jurisprudence. In particular, as Bellamy J. explains, it is based on the recent Supreme Court R. v. Kapp decision, in which the Supreme Court (arguably in obiter) moved away from the longstanding Law test and stated that the 1989 Andrews case once again provides the core understanding of constitutional equality.[11] Thus, despite Justice Bellamy’s statement that there is a different standard to be used in the constitutional and statutory contexts, and that O’Malley provides the framework for analyzing statutory discrimination, she appears in fact to have united the standards in the context of challenges to legislated services and benefits.

Where Does this Leave Equality Jurisprudence?

This decision (mostly) constitutes a victory for human rights claimants. The Divisional Court has affirmed the Supreme Court’s apparent abandonment in Kapp of the regressive elements of the constitutional equality jurisprudence in the statutory human rights context, and ensured that its problematic elements, such as a formulaic approach to ‘human dignity’ and the strict location of the ‘appropriate comparator group’, stop seeping into human rights jurisprudence. [12] The Court clearly states that despite having similar purposes, these two instruments are not the same, and neither are their analytic frames.

However, there is one thing that was not done in the decision Divisional Court’s Tranchemontagne decision: there is no explanation as to why the discrimination analysis regarding legislated services needs to be brought closer to Charter equality jurisprudence, or why such cases are different from other types of statutory discrimination claims. Underlying the Divisional Court’s decision, and much of the case law in the area, is an implicit sense that there is something special about claims against legislated benefit programs, setting them apart from other types of statutory discrimination claims, but what exactly that is is never explicitly addressed.

I would suggest that the ‘specialness’ of such claims emerges from the fact that by applying to both the substance and delivery of legislated governmental programs, the human rights statutes are being used to review governmental action in both its putatively “public” and “private” capacities outside of the constitutional context. By applying to the content of legislation and the manner by which governmental programs are apportioned, rather than solely their delivery, the Codes are being brought to bear on one of the core elements of the state’s public function – the determination of resource allocation – which is the paradigmatic situation that section 15 of the Charter regulates, and an area in which the courts are often loathed to interfere with.  In effect, these claims represent an attempt to bypass the closure of section 15 of the Charter to claims for economic and social rights. Thus, bringing these claims under the human rights statutes, and under the O’Malley and Meiorin framework, affords governmental action less deference than it would have received under the Charter. Whether or not this is a problem is something that courts should actively address.

One way to do so would be by undertaking a broad level analysis of the similarities, differences and relationship between section 15 of the Charter and the human rights codes. This would include answering some practical questions, such as: (a) whether the prohibition on discrimination under the Codes amounts to the same thing as the guarantee of equality provided by the Charter; (b) whether the defences under the Codes operate in the same manner as section 1 of the Charter; and (c) whether the concept of ameliorative programs under the Codes and under section 15(2) of the Charter are in line with each other. But it also requires some more abstract questions to be answered, such as: (a) whether the two instrument seek to achieve the same ends, and if so, (b) whether they seek to do so in the same manner; (c) whether state actors and private actors should be held to the same standard in regards to potentially discriminatory actions, and finally (d) what impact the practical realities and difficulties for equity seeking groups of bringing claims under each instrument should have on the formulation of the jurisprudential tests they need to meet. The Supreme Court of Canada began to address some of these questions in its Tranchemontagne 2006 decision, thinking through the differences and similarities between the supremacy clauses in human rights statutes and section 52 of the Charter.[13] The Court of Appeal should take this opportunity to not only affirm the Divisional Court’s decision, but also to go further with this broad level thinking and set its mind to clarifying some of the contours of the relationship between statutory and constitutional equality protection in Canada.

[2] [1999] 1 S.C.R. 497 [Law]

[3] [1986] 1 S.C.R. 103 [Oakes]

[4] [1985] 2 S.C.R. 536 [O’Malley]

[5] [1999] 3 S.C.R. 3 [Meiorin]

[6] It does explicitly apply to governmental action where the state acts as a party to one of the regulated relationships (such as services, contracting, tenancy, etc.). My argument is that, unlike the Charter, it does not regulate the state qua state, but instead has historically applied to governmental action primarily where it is acting as any other service provider, contracting party, landlord, etc. This, however, looks to be in the process of changing because of the types of cases under discussion here.

[7] Up until the last few years most case law concerning these provisions had focused on the meaning of ‘customarily available to the public’, instead of what constitutes a ‘service’ per se. However, since the Supreme Court’s 2004 Tranchemontagne decision ruled to allow all administrative decision-makers with authority to interpret questions of law the power to interpret and apply Human Rights Codes, the issue has arisen with a much greater frequency. It has become a question of pressing concern and some confusion for human rights adjudicators. See Ballieram v. WSIB, 2010 HRTO 712; cf., Dopelhamer v. WSIB, 2010 HRTO 765.

[8] See, for instance, Marakkaparambil, Sumitha v. Ontario (Health and Long-Term Care) 2007 HRTO 24; Ontario (AG) v. Ontario Human Rights Commission, 2007 CanLII 56481 (ON S.C.D.C.) at para. 47 [Brathwaite]; Ontario Secondary School Teachers’ Federation v. Upper Canada District School Board [2005] O.J. No. 4057; Alberta (Minister of Human Resources and Employment) v. Alberta (Human Rights, Citizenship and Multiculturalism Commission) 2005 ABQB 363; Pringle v. Alberta (Human Rights, Multiculturalism and Citizenship Commission) [2003] 4 W.W.R. 173; Mis v. Alberta (Human Rights and Citizenship Commission) [2003] 4 W.W.R. 173; BCGSEU v. Canada (Public Service Employee Relations Commission) 2002 BCCA 476 [BCGSEU]; Gwinner v. Alberta (Minister of Human Resources and Employment) (2002), 217 D.L.R. (4^th) 341 (Q.B.) at para. 98 [Gwinner]

[9] This argument was not accepted by the arbitrator or by the Manitoba Queen’s Bench. Manitoba High Schools Athletic Association Inc. v. Pasternak, [2008] M.J. No. 10 (QB).

[10] BCGSEU supra note 8.

[11] 2008 SCC 41[Kapp].

[12] Bellamy J. does not abandon the comparator group analysis however, and does use it in the statutory context in determining whether or not the impugned distinction perpetuates disadvantage or stereotype.  However, she also makes clear that the comparative approach to determining discrimination cannot be rigid and formulaic, but must instead think broadly about socioeconomic context.

[13] 2006 SCC 14.