Egan v. Canada [1995] 2 S.C.R. 513: Old age security legislation providing for allowance for spouse of pensioner – Definition of "spouse" restricted to person of opposite sex

Present: Lamer C.J. and La Forest, L'Heureux-Dubé, Sopinka, Gonthier, Cory, McLachlin, Iacobucci and Major JJ.


Constitutional law -- Charter of Rights -- Equality rights -- Old age security legislation providing for allowance for spouse of pensioner – Definition of "spouse" restricted to person of opposite sex -- Whether definition of "spouse" infringes s. 15(1) of Canadian Charter of Rights and Freedoms -- If so, whether infringement justifiable under s. 1 of Charter -- Old Age Security Act, R.S.C., 1985, c. O-9, ss. 2, 19(1).

The appellants are homosexuals who have lived together since 1948 in a relationship marked by commitment and interdependence similar to that which one expects to find in a marriage. When E became 65 in 1986, he began to receive old age security and guaranteed income supplements under the Old Age Security Act. On reaching age 60, N applied for a spousal allowance under s. 19(1) of the Act, which is available to spouses between the ages of 60 and 65 whose combined income falls below a fixed level. His application was rejected on the basis that the relationship between N and E did not fall within the definition of "spouse" in s. 2, which includes "a person of the opposite sex who is living with that person, having lived with that person for at least one year, if the two persons have publicly represented themselves as husband and wife". The appellants brought an action in the Federal Court seeking a declaration that the definition contravenes s. 15(1) of the Canadian Charter of Rights and Freedoms on the ground that it discriminates on the basis of sexual orientation and a declaration that the definition should be extended to include "partners in same-sex relationships otherwise akin to a conjugal relationship". The Trial Division dismissed the action. The Federal Court of Appeal, in a majority decision, upheld the judgment.

Held (L'Heureux-Dubé, Cory, McLachlin and Iacobucci JJ. dissenting): The appeal should be dismissed. The definition of "spouse" in s. 2 of the Old Age Security Act is constitutional.

Per Lamer C.J. and La Forest, Gonthier and Major JJ.: The analysis under s. 15 of the Charter involves three steps: the first looks to whether the law has drawn a distinction between the claimant and others; the second questions whether the distinction results in disadvantage, and examines whether the impugned legislation imposes a burden, obligation or disadvantage on a group of persons to which the claimant belongs which is not imposed on others, or does not provide them with a benefit which it grants others; the third step assesses whether the distinction is based on an irrelevant personal characteristic which is either enumerated in s. 15(1) or one analogous thereto. The first step is satisfied in this case, since Parliament has clearly made a distinction between the claimant and others. The second step is also satisfied: while it may be true that the appellants have suffered no prejudice because by being treated as individuals they have received considerably more in combined federal and provincial benefits than they would have received had they been treated as "spouses", there is nothing to show that this is generally the case with homosexual couples. Sexual orientation is a deeply personal characteristic that is either unchangeable or changeable only at unacceptable personal costs, and so falls within the ambit of s. 15 protection as being analogous to the enumerated grounds. All that remains to be considered under the third step is whether the distinction made by Parliament is relevant. In assessing relevancy for this purpose one must look at the nature of the personal characteristic and its relevancy to the functional values underlying the law. A form of comparative analysis must be undertaken to determine whether particular facts give rise to inequality. This comparative analysis must be linked to an examination of the larger context, and in particular with an understanding that the Charter was not enacted in a vacuum, but must be placed in its proper linguistic, philosophic and historical contexts.

The singling out of legally married and common law couples as the recipients of benefits necessarily excludes all sorts of other couples living together, whatever reasons these other couples may have for doing so and whatever their sexual orientation. What Parliament clearly had in mind was to accord support to married couples who were aged and elderly, for the advancement of public policy central to society. Moreover, in recognition of changing social realities, s. 2 was amended so that whenever the term "spouse" was used in the Act it was to be construed to extend beyond legal married couples to couples in a common law marriage. Marriage has from time immemorial been firmly grounded in our legal tradition, one that is itself a reflection of long-standing philosophical and religious traditions. But its ultimate raison d'être transcends all of these and is firmly anchored in the biological and social realities that heterosexual couples have the unique ability to procreate, that most children are the product of these relationships, and that they are generally cared for and nurtured by those who live in that relationship. In this sense, marriage is by nature heterosexual. It would be possible to legally define marriage to include homosexual couples, but this would not change the biological and social realities that underlie the traditional marriage.

Many of the underlying concerns that justify Parliament's support and protection of legal marriage extend to heterosexual couples who are not legally married. Many of these couples live together indefinitely, bring forth children and care for them in response to familial instincts rooted in the human psyche. These couples have need for support just as legally married couples do in performing this critical task, which is of benefit to all society. Faced with the social reality that increasing numbers choose not to enter a legal marriage but live together in a common law relationship, Parliament has elected to support these relationships. Parliament is wholly justified in extending support to heterosexual couples like this, which is not to say, however, that it is obligated to do so and may not treat married and unmarried couples differently.

Neither in its purpose nor in its effect does the legislation constitute an infringement of the fundamental values sought to be protected by the Charter. None of the couples excluded from benefits under the Act are capable of meeting the fundamental social objectives thereby sought to be promoted by Parliament. While these couples undoubtedly provide mutual support for one another, and may occasionally adopt or bring up children, this is exceptional and in no way affects the general picture. Homosexual couples differ from other excluded couples in that their relationships include a sexual aspect, but this sexual aspect has nothing to do with the social objectives for which Parliament affords a measure of support to married couples and those who live in a common law relationship. The distinction adopted by Parliament is relevant here to describe a fundamental social unit to which some measure of support is given.

The impugned legislation, even had it infringed s. 15, would have been upheld for the reasons given in McKinney v. University of Guelph and for those mentioned in the discussion of discrimination in this case. Per Sopinka J.: The impugned legislation infringes s. 15(1) of the Charter, for the reasons given by Cory J. Such infringement, however, is saved under s. 1. Government must be accorded some flexibility in extending social benefits and does not have to be pro-active in recognizing new social relationships. It is not realistic for the Court to assume that there are unlimited funds to address the needs of all. A judicial approach on this basis would tend to make a government reluctant to create any new social benefit schemes because their limits would depend on an accurate prediction of the outcome of court proceedings under s. 15(1). This Court has recognized that it is legitimate for the government to make choices between is advantaged groups and that it must be provided with some leeway to do so. When the definition of "spouse" in the Old Age Security Act is measured against overall objectives of alleviation of poverty of elderly spouses, it should not be judged on the basis that Parliament has made this choice for all time. The history of the legislation shows an evolving expansion of the definition of the intended recipients of the benefits. The Attorney General of Canada has taken the position that the means chosen does not have to be necessarily the solution for all time. Hence, since the impugned legislation can be regarded as a substantial step in an incremental approach to include all those who are shown to be in serious need of financial assistance due to the retirement or death of a supporting spouse, it is rationally connected to the objective. With respect to minimal impairment, the legislation represents the kind of socio-economic question in respect of which the government is required to mediate between competing groups rather than being the protagonist of an individual. In these circumstances, the Court will be more reluctant to second-guess the choice which Parliament has made. There is also proportionality between the effects of the legislation on the protected right and the legislative objective. The proper balance was struck by Parliament in providing financial assistance to those who were shown to be in the greatest need of assistance.

Per Cory and Iacobucci JJ. (dissenting): In determining whether a s. 15(1) right to equality has been violated, the first step is to determine whether, owing to a distinction created by the questioned law, a claimant's right to equality has been denied. During this first step, the inquiry should focus upon whether the challenged law has drawn a distinction between the claimant and others, based on personal characteristics. The second step is to determine whether the distinction created by the law results in discrimination. In order to make this determination, it is necessary to consider first, whether the equality right was denied on the basis of a personal characteristic which is either enumerated in s. 15(1) or which is analogous to those enumerated, and second, whether that distinction has the effect on the claimant of imposing a burden, obligation or disadvantage not imposed upon others or of withholding or limiting access to benefits or advantages which are available to others. Any search for either equality or discrimination requires comparisons to be made between groups of people. Whether or not discrimination exists must be assessed in a larger social, political and legal context. The resolution of the question as to whether there is discrimination under s. 15(1) must be kept distinct from the determination as to whether or not there is justification for that discrimination under s. 1 of the Charter. This analytical separation between s. 15(1) and s. 1 is important since the Charter claimant must satisfy the onus of showing only that there exists in the legislation a distinction which is discriminatory. Only after the court finds a breach of s. 15(1) does the government bear the onus of justifying that discrimination.

Since the law challenged draws a clear distinction between opposite-sex couples and same-sex couples, this case presents a situation of direct discrimination. As a result of the definition of a common law spouse as a "person of the opposite sex", homosexual common law couples are denied the benefit of the spousal allowance which is available to heterosexual common law couples. This distinction amounts to a clear denial of equal benefit of the law. In addition to being denied an economic benefit, homosexual couples are denied the opportunity to make a choice as to whether they wish to be publicly recognized as a common law couple because of the definition of "spouse" set out in the Old Age Security Act. The public recognition and acceptance of homosexuals as a couple may be of tremendous importance to them and to the society in which they live. To deny homosexual couples the right to make that choice deprives them of the equal benefit of the law.

The distinction in the Act is based on a personal characteristic, namely sexual orientation. Sexual orientation is analogous to the grounds of discrimination enumerated in s. 15(1). The historic disadvantage suffered by homosexual persons has been widely recognized and documented. Sexual orientation is more than simply a "status" that an individual possesses: it is something that is demonstrated in an individual's conduct by the choice of a partner. Just as the Charter protects religious beliefs and religious practice as aspects of religious freedom, so too should it be recognized that sexual orientation encompasses aspects of "status" and "conduct" and that both should receive protection.

The distinction drawn by s. 2 of the Old Age Security Act on the basis of sexual orientation constitutes discrimination. The legislation denies homosexual couples equal benefit of the law not on the basis of merit or need, but solely on the basis of sexual orientation. The definition of "spouse" as someone of the opposite sex reinforces the stereotype that homosexuals cannot and do not form lasting, caring, mutually supportive relationships with economic interdependence in the same manner as heterosexual couples. The appellants' relationship vividly demonstrates the error of that approach. The discriminatory impact cannot be deemed to be trivial when the legislation reinforces prejudicial attitudes based on such faulty stereotypes.

The impugned legislation is not saved under s. 1 of the Charter. While the objective of the spousal allowance, which is geared toward the mitigation of poverty among "elderly households", is of pressing and substantial importance, the allowance in its present form is not rationally connected to its legislative goals. A program which included the appellants would better achieve the intended goal while respecting the Charter rights of gays and lesbians. Nor is the denial of the appellants' s. 15 rights through the ineligibility for receipt of the spousal allowance minimally impaired simply because the appellants' joint income would have roughly been the same because of N's receipt of provincial support supplementing his income for a completely unrelated reason. The provincial and federal programs are clearly not co-extensive, and even if they were part of the same overlapping legislative scheme, this is not sufficient to ground a s. 1 justification. Finally, the attainment of the legislative goal is outweighed by the abridgment of the right in this case. The importance of providing relief to some elderly couples does not justify an infringement of the equality rights of the elderly couples who do not benefit for constitutionally irrelevant reasons. The definition of "spouse" in s. 2 of the Act should be read down by deleting the words "of the opposite sex" and reading in the words "or as an analogous relationship" after the words "if the two persons publicly represented themselves as husband and wife".

Per L'Heureux-Dubé J. (dissenting): A return to the fundamental purpose of s. 15 of the Charter is necessary in order to reconcile the divergent approaches taken by this Court in recent jurisprudence, as well as in the present case and in Miron and Thibaudeau. At the heart of s. 15 is the protection of, and respect for, basic human dignity. Discrimination" must therefore be at the forefront of the court's analysis. In order for discrimination to be addressed and identified in all of its varied contexts and forms, it is preferable to focus on impact (i.e. discriminatory effect) rather than on constituent elements (i.e. the grounds of the distinction). Discriminatory effects must, moreover, be evaluated from the point of view of the victim, rather than from that of the state. Considerations of relevance are more properly viewed as justifications under s. 1 than as factors integral to the identification of discrimination in the first place.

The following factors must be established by a rights claimant before the impugned distinction will be found to be discriminatory within the meaning of s. 15 of the Charter: (1) there must be a legislative distinction; (2) this distinction must result in a denial of one of the four equality rights on the basis of the rights claimant's membership in an identifiable group; and (3) this distinction must be "discriminatory" within the meaning of s. 15. A distinction is discriminatory within the meaning of s. 15 where it is capable of either promoting or perpetuating the view that the individual adversely affected by this distinction is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration. The absence or presence of discriminatory impact should be assessed according to a subjective-objective standard -- the reasonably held view of one who is possessed of similar characteristics, under similar circumstances, and who is dispassionate and fully apprised of the circumstances. This determination is arrived at by considering two categories of factors: (1) the nature of the group adversely affected by the impugned distinction, and (2) the nature of the interest adversely affected by the impugned distinction. With respect to the first category, groups that are more socially vulnerable will experience the adverse effects of a legislative distinction more vividly than if the same distinction were directed at a group which is not similarly vulnerable. In evaluating the nature of the group affected by the impugned distinction, it is relevant to inquire into many of the criteria traditionally employed in the Andrews analysis, such as whether the impugned distinction is based upon fundamental attributes that are generally considered to be essential to our popular conception of `personhood' or `humanness', whether the adversely affected group is already a victim of historical disadvantage, whether this distinction is reasonably capable of aggravating or perpetuating that disadvantage, whether group members are currently vulnerable to stereotyping, social prejudice and/or marginalization, and whether this distinction exposes them to the reasonable possibility of future vulnerability of this kind. Membership in a "discrete and insular minority", lacking in political power and thus vulnerable to having its interests overlooked, is another consideration that may be taken into account. The absence or presence of some of these factors will not, however, be determinative of the analysis. However, awareness of, and sensitivity to, the realities of those experiencing the distinction is an important task that judges must undertake when evaluating the impact of the distinction on members of the affected group.

Similarly, the more fundamental the interest affected or the more serious the consequences of the distinction, the more likely that the impugned distinction will have a discriminatory impact even with respect to groups that occupy a position of advantage in our society. While the Charter is not a document of economic rights and freedoms, the nature, quantum and context of an economic prejudice or denial of such a benefit are important factors in determining whether the distinction from which the differing economic consequences flow is one which is discriminatory. The discriminatory calibre of a particular distinction cannot, however, be fully appreciated without also evaluating the constitutional and societal significance of the interests adversely affected. Tangible economic consequences are but one manifestation of the more intangible and invidious harms flowing from discrimination, which the Charter seeks to root out. In other cases, the prejudice will be to an important individual interest rather than to one that is economic in nature. Both categories of factors emphasize that it is no longer the "grounds" of the distinction that are dispositive of the question of whether discrimination exists, but rather the social context of the distinction that matters. An effects-based approach to discrimination is the logical next step in the evolution of s. 15 jurisprudence since Andrews.

Homosexual couples are denied the equal benefit of the law on the basis of the legislative distinction in s. 2 of the Old Age Security Act, which defines couples as relationships of "opposite sex". That the appellants are able to claim higher benefits as separate individuals does not alter the fact that they have been denied the benefits, both tangible and intangible, of filing for old age benefits as a couple. The impugned distinction excludes the rights claimants because they are homosexual. Consideration of both the nature of the group and the interest affected leads us to conclude that the distinction is discriminatory. Same-sex couples are a highly socially vulnerable group, in that they have suffered considerable historical disadvantage, stereotyping, marginalization and stigmatization within Canadian society. The distinction relates to a fundamental aspect of personhood and affects individuals who, in addition to being homosexuals, are also elderly and poor. Turning to the interest affected, the impugned legislation is a cornerstone in Canada's social security net, which is, in turn, a cherished and fundamental institution in our society.

The violation of s. 15(1) of the Charter cannot be salvaged by s. 1, as it is not relevant to a proportionate extent to a pressing and substantial objective. While the objective of the legislation is pressing and substantial, the means chosen to achieve this objective fails all three branches of the proportionality test. The legislation excludes couples who would fill all of the other criteria in the Act except the requirement that they are of the opposite sex. To find that this distinction is rationally connected to the objective of the legislation requires us to conclude that same-sex couples are so different from married couples that it would be unreasonable to make the same benefits available to both. At best, the government has only demonstrated that this is its assumption. The presumption that same-sex relationships are somehow less interdependent than opposite-sex relationships is, itself, a fruit of stereotype rather than one of demonstrable, empirical reality. Nor is s. 15 minimally impaired. A reasonable alternative remedy is available: the discriminatory effect would be eliminated without prejudice to the rights or interests of any other group by extending coverage to same-sex couples who otherwise fulfil all of the other non-discriminatory criteria required in the Act. Deference under this branch of the s. 1 test is not appropriate when there is a reasonable alternative that is readily available, that is not the subject of conflicting social science views, and that could not result in a concomitant prejudice to another group. Finally, the deleterious effects of the impugned distinction outweigh its salutary effects.

Per McLachlin J. (dissenting): The reasons of Cory and Iacobucci JJ. were substantially agreed with. On the basis of the principles outlined in Miron v. Trudel, released concurrently, the impugned legislation infringes s. 15(1) of the Charter and the infringement is not saved under s. 1.

| Return to Topic Menu | Return to Main Menu |