By: Jennifer Koshan and Jonnette Watson Hamilton

PDF Version: The End of Law: A New Framework for Analyzing Section 15(1) Charter Challenges

Case Commented On: Ermineskin Indian Band and Nation v Canada, 2009 SCC 9

After the Supreme Court of Canada handed down its decision in R v Kapp, 2008 SCC 41 in June of 2008 there were questions about whether the Court had changed the legal framework for analyzing challenges brought under section 15(1) of the Charter. Kapp had clearly changed the approach to section 15(2), granting it independent status to protect ameliorative laws, programs and activities. However, on the topic of section 15(1), the Court had sent mixed signals about its intended approach. The message sent by the Court’s February 13, 2009 decision in Ermineskin Indian Band and Nation v Canada is much clearer; the legal framework for analyzing section 15(1) claims will be very different than it has been for the past decade.

In Kapp, the Court left no doubt that Andrews v Law Society of British Columbia, [1989] 1 S.C.R. 143, is henceforth to be considered the leading case on section 15(1). Its decision in Law v Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497- the leading decision for the past decade – was relegated to a supporting role. Andrews, the Court asserted in Kapp (at para 17), “set the template for this Court’s commitment to substantive equality.” The ascendancy of Andrews and its two-part test is clearest in par. 17 in Kapp:

The template in Andrews, as further developed in a series of cases culminating in Law v Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497, established in essence a two-part test for showing discrimination under s. 15(1): (1) Does the law create a distinction based on an enumerated or analogous ground? (2) Does the distinction create a disadvantage by perpetuating prejudice or stereotyping? These were divided, in Law, into three steps, but in our view the test is, in substance, the same.

Note the Court’s assertion that the three-step test in Law is the same in substance as the two-part test in Andrews: Law had merely divided Andrews‘ first part into two questions, separating out the distinction drawn from the ground on which it was drawn.

The Court in Kapp had little to say about Andrews‘ first step. It did note (at para 22) that Law had been criticized for allowing “the formalism of some of the Court’s post-Andrews jurisprudence to resurface in the form of an artificial comparator analysis focussed on treating likes alike.” Aside from this acknowledgement of the criticism, however, the Court said nothing about comparator groups. This left up in the air the issues surrounding comparator groups that had been raised in the academic literature the Court cited. What would be the role of comparator groups henceforth?

The Court in Kapp had much more to say about Andrews‘ second step. This step – analyzing whether the distinction based on an enumerated or analogous ground identified in step one amounts to discrimination – is usually the most difficult part of a section 15(1) analysis. In Andrews, according to Kapp (at para 18), discrimination was defined by two concepts:

(1) the perpetuation of prejudice or disadvantage to members of a group on the basis of personal characteristics identified in the enumerated and analogous grounds; and

(2) stereotyping on the basis of these grounds that results in a decision that does not correspond to a claimant’s or group’s actual circumstances and characteristics.

However, in Law, as the Court in Kapp acknowledged (at para 19), discrimination had been defined “in terms of the impact of the law or program on the ‘human dignity’ of members of the claimant group, having regard to four contextual factors: 1) pre-existing disadvantage, if any, of the claimant group; (2) degree of correspondence between the differential treatment and the claimant group’s reality; (3) whether the law or program has an ameliorative purpose or effect; and (4) the nature of the interest affected . . .” In Kapp (at para 21, emphasis in the original) the Court acknowledged the difficulties created by “the attempt in Law to employ human dignity as a legal test.” Human dignity, while still “an essential value” underlying section 15 (at para 21), was admitted (at para 22) to be “an abstract and subjective notion” that was “confusing and difficult to apply” and, worse, “an additional burden on equality claimants.” That is all the Court had to say about human dignity in Kapp however. Its role in future section 15(1) jurisprudence was left unsettled.

The Court in Kapp was a little more specific about the four contextual factors set out in Law by which impact on human dignity was to be assessed. Those four factors, the Court asserted (at para 23) “are based on and relate to the identification in Andrews of perpetuation of disadvantage and stereotyping as the primary indicators of discrimination.” Law‘s factors one and four went to Andrews‘ perpetuation of disadvantage and prejudice, it was said, while Law‘s second factor dealt with stereotyping. The third factor – the ameliorative purpose or effect of a law or program – was now subsumed in the section 15(2) analysis, but might still have a role in determining whether the effect of the law or program is to perpetuate disadvantage. With this amalgamation in Kapp of Law‘s four contextual factors to Andrews‘ two concepts of discrimination, Law appeared to continue to have a role in future analyses under section 15(1).

Indeed, all of the lower courts which dealt with challenges under section 15(1) in the 7 ½ months between the Kapp and the Ermineskin decisions used the two-part test set out in Andrews, as restated in Kapp. See Hartling v Nova Scotia (Attorney General), 2009 NSSC 2 at para 17; C.C.-W. (Litigation guardian of) v Ontario (Health Insurance Plan, General Manager), [2009] O.J. No. 140 (SCJ) at para 104; Withler v Canada (Attorney General), 2008 BCCA 539 at para 155; Confédération des syndicats nationaux v. Québec (Procureur général), 2008 QCCS 5076 at paras 326-27; and Downey v Nova Scotia (Workers’ Compensation Appeals Tribunal), 2008 NSCA 65. Most also used the concept of human dignity and Law’s four contextual factors in the second step. Even though Withler (at paras 159-160) and Hartling (at para 19) noted that the Supreme Court in Kapp had moved away from Law‘s insistence that discrimination be defined in terms of the impact of the law or program on human dignity, they still applied Law‘s four contextual factors, as did the court in Downey.

In all, the lower court decisions after Kapp suggested that, aside from formulating the test as a two-part test, instead of a three-part test, Kapp was having little impact on analyzing section 15(1) claims.

The Supreme Court of Canada’s February 13, 2009 decision in Ermineskin Indian Band and Nation v Canada will change all that. Our colleague Nigel Bankes’ recent post “The Crown has neither the power nor the duty to invest Indian monies: The use of legislation to limit trust duties” sets out the facts of Ermineskin. Our focus in this comment is on how the Court articulates its approach to section 15(1) in Ermineskin, and how it applies that approach (or rather fails to do so in a satisfactory way) to the facts.

The analytical approach to section 15(1) is indeed changed in Ermineskin. The Supreme Court did not overrule Law in Kapp; it did not even say the test set out in Law was wrong. But Law is gone; there is no reference to it in Ermineskin. The only equality cases the Court relies upon in Ermineskin are Andrews and R v Turpin, [1989] 1 S.C.R. 1296.

The phrase “human dignity” is never mentioned in Ermineskin. None of the four contextual factors from Law are used; rather context is now the larger social, political and legal context of the impugned legislation: Ermineskin at para 193, citing Turpin. The Ermineskin decision is therefore a strong signal to lower courts and tribunals that the Law framework for analyzing an equality challenge should no longer be used.

Ermineskin makes it clear that the second part of the test from Andrews as re-stated by Kapp – the question of whether the distinction creates a disadvantage by perpetuating prejudice or stereotyping – is to be answered by considering “the broader context of a distinction”: Ermineskin at para 194. That is really all the guidance the Court offers on how a lower court or tribunal should conduct that analysis. This should certainly do away with the tendency towards formalism that Law‘s more specific four contextual factors approach had engendered. But what will it be replaced with? Ermineskin does not have the strong statements of the Court’s commitment to substantive equality that Kapp had (at paras 14-16). All the Court has to say about substantive equality in Ermineskin (at para 194) is that its “statement in Turpin signals the importance of addressing the broader context of a distinction in a substantive equality analysis.”

This aspect of Ermineskin is troubling in at least a couple of respects. First, discrimination seems to have been reduced to a consideration of prejudice and stereotyping. As noted above, the Court began to focus on these particular harms of discrimination in Kapp, tracing this understanding of discrimination back to Andrews. While the Law decision was rightly criticized, one thing it did do relatively well was to set out a broader range of harms flowing from discrimination – not just prejudice and stereotyping, but also vulnerability, powerlessness, oppression, stigmatization, marginalization, devaluation, and disadvantage more broadly (Law at paras 29, 42, 44, 47, 53, as cited in Women’s Legal Education and Action Fund, Intervener Factum in Newfoundland (Treasury Board) v NAPE at para 17 (available here)).

Second, by failing to provide sufficient doctrinal guidance, the Court has made it likely that lower courts will fixate on the words “prejudice and stereotyping” as a complete statement of the harms that substantive equality prohibits. It will be up to counsel for equality rights litigants to educate lower courts on a more fulsome definition of substantive equality. This is not a new hurdle – but it is frustrating that the Court decided to revisit its approach to section 15(1) of the Charter in cases where there were no interveners expert in equality rights – something it was also criticized for doing in Law.

What of the first part of the Andrews test as restated by Kapp, the question of whether the law creates a distinction based on an enumerated or an analogous ground? The Court said very little about this part of the test in Kapp, but it says even less in Ermineskin. In the latter case, the Court merely states (at para 188) that the first part of the test requires a determination of whether the law creates a distinction based on an enumerated or analogous ground. There is no elaboration.

The lack of direction, given the acknowledged criticism about comparator groups in Kapp, is disappointing. Even more concerning, however, is the way the Court applies the first part of the test from Andrews, as restated by Kapp, to the facts in Ermineskin. According to the Court (at para 185) the appellants had argued that, as Indians, they had been deprived by the Indian Act of the rights available to non-Indians whose property is held in trust by the Crown. The Court acknowledges (at paras 189 and 201) that the first requirement of the test is satisfied because the impugned legislation “creates a distinction between Indians and non-Indians because the legislation only applies to Indians.” It might seem as though the Court “refined” the comparator group from non-Indians whose property is held in trust by the Crown to non-Indians, period. However, in para. 195, the Court does contrast “the management of Indian moneys” with “other trust relationships where risk and financial returns are generally the only considerations, and where there is little concern with the trustee having complete control and discretion. . .”. The use of a comparator group that did not have to satisfy the Crown that management and investment of its funds was in its best interests might have affected the Court’s determination in para. 201 that the “distinction between Indians and non-Indians . . . is not discriminatory.” Even taking the narrow definition of discrimination – disadvantage created by prejudice and stereotyping – this paternalistic approach to management of First Nations’ money should surely meet the definition.

At one time, comparators were said to be relevant at each stage of the analysis: Hodge v Canada (Minister of Human Resources Development), 2004 SCC 65, [2004] 3 S.C.R. 357. Instead of using comparator groups in its analysis of whether the distinction drawn between those groups and the appellants created disadvantage, however, the Court in Ermineskin seems to focus on two other comparisons. The first is a comparison between the Crown investing the bands’ oil and gas royalties in a diversified portfolio and the Crown paying interest on the amounts: see paras. 191 and 196. The second is a comparison between bands who invest Indian moneys themselves and those who do not: see paras 200-201. Is the latter a relevant comparator group? For what purpose is the comparison drawn? As noted by Nigel Bankes, the Court in its section 15(1) analysis failed to focus on “the question of whether the Crown was in breach of a duty to transfer capital monies to the band for them to invest” (at p. 8, emphasis added). Had the Court focused on this question, the relevant comparator might have been beneficiaries who are not obliged to convince the Crown that they have a financial management plan before a transfer will be made. Arguably, the failure to identify and apply an appropriate comparator contributes to the Court’s finding that there was no discrimination here.

The Samson Nation argued that three possible comparator groups were relevant: (a) all other trust beneficiaries who are not Indians or Bands of Indians; (b) all others for whom the federal Crown acts as statutory fiduciary or trustee; and (c) all other beneficiaries of a federally-regulated fund managed by a fiduciary or trustee (Factum of the Samson Appellants in Ermineskin at para 280). They further argued that members of these comparator groups had the right to have their moneys invested or earn a maximum rate of return, without an undue risk of loss. By failing to give attention to the claimant’s choice of comparators, the Court is also distancing itself from previous case law, where the claimant’s framing of the comparator was to be the starting point for the comparator analysis (see Law at para 58).

We’ve gone on at some length about comparators because the Court does not use them the way it has said a court should. This emphasis troubles us as we don’t wish to reify this part of the section 15(1) analysis. Although we believe that appropriate attention to comparators would have resulted in a finding of discrimination in this case, we don’t mean to assert that this will always be the case. There will be other cases where other factors are better at bringing out the disadvantage. Here, however, it seems clear that extra obligations are being imposed on Indian beneficiaries that are not being imposed on others, and this should lead to a finding of discrimination (in spite of the Court’s rhetoric about “Aboriginal self determination and autonomy” at para 195).

A related problem is that there is no mention of grounds in Ermineskin, whether enumerated or analogous. A grounds based approach to section 15(1) is one of the few constants in the jurisprudence from Andrews to Law to Kapp. Although a number of problems have been noted with the concept of grounds, grounds do provide “the necessary history and context of discrimination”, and focusing on why something counts as a ground is a reminder of why discrimination is not allowed: see Dianne Pothier, “Connecting Grounds of Discrimination to Real People’s Real Experience” (2001) 13 Canadian Journal of Women and the Law 37 at 41. The parties in Ermineskin characterized the grounds at issue as “race, national or ethnic origin” (Factum of the Samson Appellants, para 279). By failing to acknowledge that these grounds were the basis of the differential treatment in Ermineskin, the Court glosses over the “social, political and legal context” of the claim that Turpin requires attention to (again).

There was quite a bit of positive “buzz” around the Supreme Court’s return to Andrews in Kapp. Andrews, however, was fairly confusing about what a substantive equality analysis would look like. It would be a contextual analysis, but what else it would be was not clear. We must also recall that there were controversies in the application of Andrews, and its approach to equality rights was restated even before Law (see the 1995 trilogy Miron v Trudel, [1995] 2 S.C.R. 418; Egan v Canada, [1995] 2 S.C.R. 513, Thibaudeau v Canada, [1995] 2 S.C.R. 627). It might feel good to be optimistic about the new direction in analyzing section 15(1) challenges now that Andrews is back. But there is little in Ermineskin‘s analysis of the section 15(1) challenge in that case to make us feel hopeful.