Canada (Attorney General) v. Lesiuk

Canada (Attorney General) v. Lesiuk Court (s) Database Federal Court of Appeal Decisions Date 2003-01-08 Neutral citation 2003 FCA 3 File numbers A-281-01 Notes Reported Decision Decision Content Federal Court Reports Canada ( Attorney General ) v. Lesiuk [2003] 2 F.C. 697 Date: 20030108 Docket: A-281-01 Neutral citation: 2003 FCA 3 CORAM: LÉTOURNEAU J.A. ROTHSTEIN J.A. MALONE J.A. BETWEEN: THE ATTORNEY GENERAL OF CANADA Applicant and KELLY LESIUK Respondent and WOMEN'S LEGAL EDUCATION AND ACTION FUND ("LEAF"), INCOME SECURITY ADVOCACY CENTRE Interveners Heard at Edmonton, Alberta, on November 19 and 20, 2002 Judgment delivered at Ottawa, Ontario, on January 8, 2003 REASONS FOR JUDGMENT BY: LÉTOURNEAU J.A. CONCURRED IN BY: ROTHSTEIN J.A. MALONE J.A. Date: 20030108 Docket: A-281-01 Neutral citation: 2003 FCA 3 CORAM: LÉTOURNEAU J.A. ROTHSTEIN J.A. MALONE J.A. BETWEEN: THE ATTORNEY GENERAL OF CANADA Applicant and KELLY LESIUK Respondent and WOMEN'S LEGAL EDUCATION AND ACTION FUND ("LEAF"), INCOME SECURITY ADVOCACY CENTRE Interveners REASONS FOR JUDGMENT LÉTOURNEAU J.A. [1] This application for judicial review against a decision (CUB 51142) of an Umpire rendered on March 22, 2001 raises the following questions: Did the Umpire err when he concluded that subsections 6(1) and 7(2) of the Employment Insurance Act, S.C. 1996, c. 23 as later amended (Act), which determine the eligibility requirements to unemployment benefits, violate subsection 15(1) of the Canadian Charter of Rights and Freedoms (Charter) and, therefore, should be ignored as if the provisions had never been enacted? These provisions mandate a regime based on hours worked for minimum eligibility requirements (MERs) which, at the time of the Umpire's decision, were 700 hours for general and special maternity benefits. On December 31st, 2000, the minimum requirement for maternity benefits was lowered to 600 hours. According to counsel for the applicant, the legislative change came as a result of a study of the operation of the special benefits program: see Human Resources Development Canada, 1999 Monitoring and Assessment Report, Application Record, vol. 10, tab D, page 7804. [2] Subsections 6(1) and 7(2) read at the time: 6. (1) In this Part, .... "major attachment claimant" means a claimant who qualifies to receive benefits and has 700 or more hours of insurable employment in their qualifying period; ... 6. (1) Les définitions qui suivent s'appliquent à la présente partie. .... « _prestataire de la première catégorie_ » Prestataire qui remplit les conditions requises pour recevoir des prestations et qui a exercé un emploi assurable pendant au moins 700 heures au cours de sa période de référence. ... 7. (2) An insured person, other than a new entrant or a re-entrant to the labour force, qualifies if the person (a) has had an interruption of earnings from employment; and (b) has had during their qualifying period at least the number of hours of insurable employment set out in the following table in relation to the regional rate of unemployment that applies to the person. 7. (2) L'assuré autre qu'une personne qui devient ou redevient membre de la population active remplit les conditions requises si, à la fois_: a) il y a eu arrêt de la rémunération provenant de son emploi; b) il a, au cours de sa période de référence, exercé un emploi assurable pendant au moins le nombre d'heures indiqué au tableau qui suit en fonction du taux régional de chômage qui lui est applicable. TABLE ========================================= Regional Rate of Required Number of Hours Unemployment of Insurable Employment in Qualifying Period ----------------------- 6% and under 700 more than 6% but not more than 7% 665 ..... TABLEAU ========================================= Taux régional de chômage Nombre d'heures d'emploi assurable requis au cours de période de référence ----------------------- 6_% et moins 700 plus de 6_% mais au plus 7_% 665 .... [3] For reasons that I will discuss after a brief summary of the facts and proceedings leading to the present application for judicial review, I believe the learned Umpire misapprehended and misconstrued the law and that his decision cannot stand. Facts and procedure [4] The respondent, Kelly Lesiuk, worked part-time from 1993 to 1998 as a registered nurse at the Brandon General Hospital. In mid-August, 1997, her husband accepted employment in Winnipeg, a two-hour drive from Brandon. It was decided that he would stay in Winnipeg during the week and return to Brandon on weekends so that she would be able to pick up some shift work. [5] In March of 1998, after their house in Brandon was sold, the respondent and her 3-year-old child rejoined her husband in Winnipeg. Shortly after arriving, she applied for regular employment insurance ("EI") benefits. The EI Commission advised her that she did not qualify for benefits. She had worked 667 hours in her qualifying period. Qualification, as per subsection 7(2) of the Act, required 700 hours. She was 33 hours short. [6] In April, 1998, the respondent's doctor recommended that she not work as she was now pregnant with a second child and the substantial weight increase would not be suitable to the heavy physical labour required of registered nurses. Qualification for special maternity or parental benefits were also contingent on 700 hours worked in the qualifying period. [7] The respondent appealed the Commission's rejection of her benefits to the Board of Referees. On November 19, 1998, the Board applied the provisions of the Act and denied her appeal. [8] The respondent appealed the Board's decision to the Umpire. Before the Umpire, it was argued that the eligibility requirements of the Act discriminate against women who, historically, have had the primary responsibility for childcare. On March 22, 2001, the Umpire rendered a decision that subsections 6(1) and 7(2) of the Act, the MERs, are unconstitutional violations of the respondent's section 15 equality rights under the Charter. He referred the matter back to the Board for a rehearing, but this time as if the provisions in question had never been enacted. [9] The Applicant, Attorney General of Canada, seeks judicial review of this decision pursuant to sections 18.1 and 28 of the Federal Court Act. He seeks to have the Umpire's decision set aside and the decision of the Board of Referees restored. The decision of the Umpire [10] After a review of the facts and the history of employment insurance legislation, the Umpire proceeded to apply the approach adopted by the Supreme Court of Canada in Law v. Canada, [1999] 1 S.C.R. 497 when analysing potential breaches of section 15 of the Charter. This process led him to the following conclusions: a) Subsections 6(1) and 7(2) of the Act impose differential treatment between the claimant and others. The minimum eligibility requirements for benefits set out in subsections 6(1) and 7(2) impose differential treatment, based on one or more personal characteristics, between Kelly Lesiuk and some other persons or groups to whom she may properly be compared. On the evidence presented, Kelly Lesiuk has established on balance that the MERs for benefits in subsections 6(1) and 7(2) draw a distinction between her and other workers. An eligibility standard that is based on the average number of hours worked per week by Canadian employees has a differential impact on those who, because of parental and family commitments, find it difficult, if not impossible to meet that standard. The average woman is disadvantaged because men work, on average 39 hours per week in paid employment, while women work, on average, only 30 hours. The EI requirements are based on 35 hours per week, thus disadvantaging women. Moreover, the evidence noted that women continue to perform two-thirds of unpaid labour, leaving them with fewer hours to devote to paid employment (see paragraph 50 of his decision). The system based on hours worked disproportionately affects women by increasing the number of the hours of work required to qualify and lowering the number of weeks of benefits if they are unable to increase their hours for work. b) The differential treatment is based on an enumerated or analogous ground. I reproduce paragraph 59 of his decision: On the evidence presented, I am satisfied that the status of being in a parent and child relationship is a ground analogous to those enumerated in s. 15(1). Parenthood is central to one's identity and personhood: it is a status that is immutable. It is true that the status will change when the children are no longer in need of a caregiver, but that does not change the fact that their status is immutable until that time comes. The appellant's [Kelly Lesiuk's] status as a parent and primary caregiver is one that the government has no legitimate interest in expecting her to change to receive equal treatment under the law. When a mother works part-time because of her part-time parental responsibilities, she should not receive inferior employment insurance coverage on that account. c) The purpose or effect is substantively discriminatory. The eligibility requirements demean the essential dignity of women who predominate in the part-time labour force because they must work for longer periods than full-time workers in order to demonstrate their work force attachment. By basing the threshold for eligibility and the definition of work attachment on the average work week of 35 hours, the eligibility requirements fail to take into account the fact that women's average work week in the paid labour force (30 hours) is 85% of men's average work week (35 hours). I note in passing that this actually appears to be a mis-statement of the evidence since 35 hours is the average for all workers. Since women continue to spend approximately twice as much time doing unpaid work as men, women are predominantly affected. The definition of major work force attachment under the Act violates Kelly Lesiuk's dignity even more directly since the entrance requirement measures and defines who is worthy of employment insurance benefits. Although a woman, under the present legislation, who works part-time year round has a strong attachment to the workforce, her attachment to the workforce is less than that of a man who is able to easily accumulate employment insurance benefits by working 700 hours in the summer and collecting employment insurance benefits every winter. Since the appellant's attachment to the work force is deemed inadequate, not because she lacks regular employment, but because her weekly hours of paid employment are limited by her unpaid obligations, the dignity of the part time working mother with children is more directly violated (see paragraph 66 of the decision). If, as in the case of the appellant, shift availability or child-care availability prevents her from reaching the 700-hour threshold, she faces a complete denial of benefits if her employment is interrupted and receives no protection from the employment insurance premiums she has paid. This has the effect of promoting the view that a woman is less capable, or less worthy of recognition or value as a human being or as a member of Canadian society (see paragraph 67 of the decision). [11] The Umpire reviewed the impugned provisions in light of the principles applicable to an analysis under section 1 of the Charter. He concluded that they could not be saved by that section. In his view, the legislation was not pressing and substantial. He expressed himself in the following terms at paragraph 70 of his decision: I am not satisfied on balance that the legislation is pressing and substantial. I have difficulty accepting that the objective of the variable (420-700) minimum eligibility requirements is to ensure that individuals who receive regular benefits have some work force attachment, relative to the rate of unemployment in their area of residence. Nor does the system discourage, as it was contended, misuse. As noted earlier, the premise that work force attachment can be measured by a single criterion of hours worked in a year is unique to Canada. Even Dr. Nakamura, who gave evidence on behalf of the respondent, accepted that a person could work just two hours a week, year after year and be strongly attached. In my view, the present eligibility requirements reward those who have the freedom of working long intensive hours at the expense of the family caregiver. Nor does the system discourage misuse. [12] In addition, he ruled that the MERs found in the impugned provisions were not rationally connected to the objective of ensuring that the recipients of benefits have a minimum level of work force attachment. Moreover, the lowering of the number of hours required to receive benefits would not detract the government objectives of encouraging work force attachment, and deterring misuse and dependence. [13] Finally, as he did not have the power to issue declarations of invalidity as a result of the decision of this Court in Nishri v. Canada (1997), 223 N.R. 376, he concluded that an appropriate remedy was to send the matter back to a Board of Referees with instructions to rehear the respondent's appeal as if the provisions of subsections 6(1) and 7(2) had never been enacted. He found support for this remedy in the decision of our Court in Zwarich v. Canada (Attorney General), [1987] 3 F.C. 253 (F.C.A.). Analysis of the decision [14] Counsel for the applicant submits that the MERs in subsections 6(1) and 7(2) of the Act are consistent with section 15 of the Charter in that they entail no differential treatment, involve no enumerated or analogous ground and reveal no evidence of discrimination. The respondent takes a diametrically opposite view. The statutory scheme [15] The Umpire did an extensive review of the statutory scheme. I need not repeat what he said. However, I want to emphasize that the EI system is a contributory scheme which provides social insurance for Canadians who suffer a loss of income as a result of a loss of their employment or who are unable to work by reason of illness, pregnancy and childbirth or parental responsibilities for a newborn or newly-adopted child. [16] It is in the context of such a scheme that the respondent seeks, in effect, either the complete elimination or a substantial lowering of the eligibility requirements without providing adequate or alternative parameters for the future. As will become evident when further discussing the respondent's contentions, other groups such as visible minorities, immigrants or disabled workers who fail to meet the MERs and are denied benefits could be justified in making similar claims. Indeed, in case of simply lowering the MERs, members of the respondent's group or of these other groups who would not meet the new lowered threshold would still be entitled to make the same claim on the same basis. Challenges could be made by the remaining members of these groups until, in the end, there is no threshold. [17] In a sense, the situation on this last point is not dissimilar to that of the claimant in Granovsky v. Canada, [2000] 1 S.C.R. 703 where Mr. Granovsky applied for, and was denied, a permanent disability pension. Although not attacking the eligibility requirements to a pension as discriminatory under section 15, he was, in effect, seeking a change in these requirements by challenging the contribution requirements imposed by Parliament. As in the present instance, he admitted that Parliament could, without infringement of the Charter, create a particular type of benefit targeted at particular individuals (in Granovsky, those who have a recent work force attachment - in our case, individuals who have a commitment to the work force), but that Parliament drew the line in the wrong place. In our case, the respondent complains that Parliament either drew the wrong line or drew it at the wrong place. [18] In his review of a legislative scheme similar to ours, before dismissing Mr. Granovsky's appeal, Binnie J., for a unanimous Court, expressed at paragraphs 9 and 14 views and concerns that are apposite here: The CPP [Canada Pension Plan] was designed to provide social insurance for Canadians who experience a loss of earnings owing to retirement, disability, or the death of a wage-earning spouse or parent. It is not a social welfare scheme. It is a contributory plan in which Parliament has defined both the benefits and the terms of entitlement, including the level and duration of an applicant's financial contribution. [...] I note at the outset that the appellant seeks an extension of the s. 15(1) principles laid down in the decided cases, which is understandable, but he does so in circumstances that provide no clear boundaries for the future. If he succeeds in having the "permanence" requirement of the CPP test rewritten, for example, will courts next be asked to dilute the CPP requirement that the disability be severe? The less severely disabled will no doubt argue that their interests are no less worthy of protection than those whose disabilities are more severe. Is the legislature then precluded from targeting the permanently disabled for special programs or services (special paratransit public bus facilities for example) without making the same services and programs available to those whose disabilities are temporary, and if so, how temporary would still be sufficient to qualify? (emphasis added) Whether the impugned provisions result in a substantively differential treatment on the basis of one or more personal characteristics [19] Under a subsection 15(1) analysis, the Court has to determine whether the impugned provisions impose a substantively differential treatment on the basis of one or more personal characteristics. This determination is fraught with difficulties. Although the Charter guarantees equality rights to individuals, a proper determination of the issue often requires a defining of the appropriate comparator groups against which the alleged difference in treatment will be measured: see Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497, at paragraph 66. In the case at bar, counsel for the respondent took an inter-sectional approach, whereby two possible freestanding personal characteristics are combined to define the group, and eventually found an analogous ground of discrimination: see Corbière v. Canada, [1999] 2 S.C.R. 203; Falkiner v. Ontario (2002), 59 O.R. (3d) 481 (Ont. C.A.). He defined the group to which Kelly Lesiuk belongs as "women in a parental status" and elected "all men" as a basis for comparison. To put it another way, women in a parental status, in his view, receive under the Act, as a result of the MERs, a treatment substantively different from men, whether in a parental status or not. [20] At the hearing, counsel for the intervener, Women's Legal Education and Action Fund, wanted to submit a comparator group different from the one advanced by the respondent. We refused to entertain her submission as a change of comparator group could work prejudice and unfairness on both parties who, before the Umpire, proceeded under a different assumption and adduced evidence accordingly. I am sure this is the type of situation that Iacobucci J. had in mind in Law v. Canada, [1999] 1 R.C.S. 497 when, at paragraph 58, he wrote: When identifying the relevant comparator, the natural starting point is to consider the claimant's view. It is the claimant who generally chooses the person, group, or groups with whom he or she wishes to be compared for the purpose of the discrimination inquiry, thus setting the parameters of the alleged differential treatment that he or she wishes to challenge. However, the claimant's characterization of the comparison may not always be sufficient. It may be that the differential treatment is not between the groups identified by the claimant, but rather between other groups. Clearly a court cannot, ex proprio motu, evaluate a ground of discrimination not pleaded by the parties and in relation to which no evidence has been adduced: see Symes, supra, at p. 762. However, within the scope of the ground or grounds pleaded, I would not close the door on the power of a court to refine the comparison presented by the claimant where warranted. Therefore, I shall utilize the comparator groups identified by the respondent. [21] By definition, laws granting social benefits entail a differential treatment. In determining categories of beneficiaries and eligibility requirements, they treat differently the persons who are excluded from their scope of application and, as a result, are denied benefits. I do not think that one can argue that these persons are not subject to a substantively differential treatment. The question is whether this occurs on the basis of one or more personal characteristics. The answer to that question is not obvious in the context of the evidence that was tendered by the parties. [22] Having regard to the relevant comparator groups, it is apparent that there are significant deficiencies in the statistical evidence which the parties have relied upon to support their positions. The applicant has relied upon studies that it admits suffer from the deficiencies pointed out by the respondent: - some data collected in 1997 did not reflect actual experience as arbitrary adjustments or assumptions were made during the transitional period arising from the ending of the Unemployment Insurance Act qualifications and the commencement of the Employment Insurance Act qualifications; - studies reflecting job separation rather than unemployment; - limitations of the Canadian Out of Employment Panel data base; and - misinterpretation of the data by one of the applicant's witnesses. [23] The parties did agree that data in tables in the 1999 Employment Insurance Coverage Survey and Women in Canada 2000, a gender-based statistical report, could be relied upon. However, this data also presents difficulties. [24] Although some information about a female parent in a parent/child relationship is available, that information is limited. For example, the statistics indicate that 69 percent of all women with children less than 16 years of age living at home were part of the employed work force in 1999. However, there is no data for this group identifying what proportion would be part-time workers only, what proportion of those who worked part-time in the past 12 months were unemployed in 1999 or what proportion of that group was ineligible for employment insurance benefits because of an inability to meet the minimum number of hours required to qualify for employment insurance. Information for the comparator group is similarly deficient. Without such information, it is difficult or impossible to determine the adverse effect of the minimum qualifying hours requirement on the respondent's group, as compared to the comparator group. [25] An inference may be drawn from the statistical data that some members of the respondent's group were ineligible for employment insurance in 1999. It appears that 179,000 individuals unemployed in 1999 had worked part-time in the previous 12 months and, of that group, some were ineligible for employment insurance because they did not attain the minimum number of qualifying hours to entitle them to be benefits. However, the breakdown between men and women is not shown and there is no breakdown for women in a parent/child relationship. [26] The data shows that for women in the 25 to 44 age group, 32.5 percent worked part-time in 1999 due to "caring for children". For all women who worked part-time in 1999, only 15.6 percent did so due to "caring for children". This may be a reasonable proxy for one portion of the respondent's group. However, it does not indicate what proportion of the respondent's group worked full-time or, indeed, what proportion was unemployed in 1999 but also worked part-time in the previous 12 months or what proportion of that group was ineligible because of the inability to meet the minimum threshold for qualifying hours. [27] Without adequate statistical data, it would be speculative to make the type of comparisons the respondent is asking the Court to make, i.e., that the respondent's group suffers a substantively differential treatment due to the qualifying hours requirement for employment insurance benefits as compared to the comparator group. [28] That does not mean that some generalizations cannot be made. For example, the parties agreed that those affected by the qualifying hours requirement was quite small as a portion of the total number of females in the parent/child relationship in the work force. In other words, the vast majority of those in the respondent's group are unaffected by the minimum requirement for qualifying hours: see the testimony of Dr. Nakamura to that effect, Applicant's Record, vol. 1, page 378. Small numbers alone, however, do not excuse discrimination when the individuals affected are excluded from receiving benefits. Nonetheless, the magnitude of the problem is put somewhat into perspective. This is not a case where an entire group is excluded from benefits or, indeed, where a significant portion of the group is excluded. In this respect, this case is unlike most section 15 social benefit cases: see for example Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624; Egan v. Canada, [1995] 2 S.C.R. 513; Collins v. Canada, [2000] 2 F.C. 3 (F.C.T.D.), affirmed on appeal [2002] 3 F.C. 320 (F.C.A.). [29] In her testimony before the Umpire, Dr. Nakamura, who is a professor in the Department of Finance and Management Science at the University of Alberta, and who conducted a study of the impact of the new EI system on beneficiaries, confirmed that, in dealing with the respondent and her group, we are dealing with a very small subset that may have lost coverage under the new system, but that additional measures were taken to alleviate the situation. It should be recalled that the old system used weeks and hours worked instead of merely hours as eligibility requirements. Here is what she said in answer to a question put by the Umpire who was referring to part-time female parents who are between 25 and 54 years of age and who have to juggle their timetable in order to work: THE COURT: Let us assume that the evidence establishes that women in that category, vis a vis men in that category, are being harmed. Is it fair to say that they can fall by the wayside so that we can protect the other three examples you've given? THE WITNESS: Some people did fall by the wayside. Most of the people who are not covered under EI weren't covered under UI. There's a very - it's a very special, small subset that lost coverage in going from UI to EI, as all of the studies show. However, in addition to that, there were some things done, explicitly to try and help that sort of situation. A person in that situation almost had to have another income from somewhere. One of the things that was done was to bring in a supplement to the child tax credit. All families which were covered for that, then got a supplement in addition to that. A second thing that was done was to try to look at the provisions for the collection of monies from an absent spouse, for instance. There were in fact a great deal of worrying about that and a great deal of looking at other aspects, other programs, to try and find some way of bringing some more help. I guess that another key issue there, was the feeling that if you were going to take money out of the pockets of some families and give it to other families, the families that you were taking it from should be in some way - should not be worse off than the families that you are giving it to. That was a concern under both UI and EI. (emphasis added) (Applicant's Record, vol. 1, pages 362-363) At page 378 of the Applicant's Record, dealing with persons who work less than 700 hours in a work year of 2000 hours, she went on to say: The total number of people in this country, who, over a period of five or ten years do not work more than 700 hours every single year, is a very small subset of people. Including a small subset of working women. [30] In a study released in 2000 on the impact of the new system based on hours requirements, Opinion Report, Applicant's Record, vol. 5, at page 3957, Dr. Nakamura concluded that it was "still too early to be able to definitively evaluate many of the novel features of the program, since these features were phased in over time and also the experience rating procedures will not have full effect until individual program participants have had a chance to build up benefit use histories for EI". She added that "nevertheless, certain patterns seem to be emerging". At pages 3957-3958, she wrote: So far the indications are that the net effects of EI on the availability of work are quite small but may have resulted in somewhat shorter average unemployment spells. Careful study is needed to distinguish between program effects and the effects of changed economic conditions. Nevertheless, at this point I believe that the switch to the EI system led to a net increase in eligibility and that the numbers of workers who would have qualified under UI and failed to qualify under EI are small too. The monthly labour force surveys conducted by Statistics Canada also suggest that it is generally true that a new job created after EI is less likely to be under 15 hours per week and more likely to be over 30 hours. Full evaluation results are needed to determine the extent to which these and other observed changes reflect the new eligibility conditions for some groups of workers or the switch from a weekly to an hourly unit of account for eligibility. Full evaluation results are also needed to establish whether the effects of EI are as anticipated with respect to workers in different types of families. (emphasis added) [31] In her testimony before the Umpire, she stressed some of the difficulties involved in a tinkering with the MERs for a group of workers like the respondent's group. At pages 363-365 of the Applicant's Record, vol. 1, the following exchange with the Umpire speaks to these difficulties, an important one being that there is no single group which is uniformly disadvantaged by the eligibility requirements: THE COURT: - I agree with you, there have to be trade offs. But if the trade offs are such that they're harmful to a particular group and they fail to address them, isn't that where the legislation can fail? THE WITNESS: How do you define this group, though? I cannot see a group there that is uniformly harmed. Within the group that I understood to be the group, 25 to 44, female, parent and with their work behaviour conditioned on their parental responsibility, there were ones helped and there were ones harmed by the legislation, that's a first thing. Second of all, I don't know law, you know law, but I would think that in divining such a law that you were going to pass, you would have to say, how do I know if a person is in this group? Am I going to make a special bar for the people who are working short hours because they are a parent? Am I going to investigate, to ascertain if that's why they are working short hours as opposed to they are simply working short hours and they are also relatively high income and they have a nanny who takes care of their child? There are many people who work short hours for very many reasons. Would we investigate to try and find out why they are working short hours? In welfare, we do investigate people. We do that not because anybody feels good about that, but because if you are taking money away from some and giving it to others in a way that has nothing to do with insurance principles, those who we take it from would like to know that those it's going to are less well off than them. We haven't wanted to have that type of investigation in UI because if we have it in UI, it's intrusive, first of all. We need this to be a program for the whole society. If we were investigating all of us, it's exceedingly expensive, it's exceedingly onerous. But in addition to that, it's time consuming. (emphasis added) [32] In conclusion, it is fair to say that there is a dearth of probative statistical evidence with respect to the alleged differential treatment suffered by the respondent's group on the basis of personal characteristics. The evidence shows that there is no group which is uniformly adversely affected. In addition, it reveals that the new EI system has led to a net increase in eligibility and that the numbers of workers who would have qualified under the old system and failed to qualify under the new is small. It also shows that the respondent's group is a very special, small subset of these small groups of workers who have lost in the change to the new system and yet not everyone in that subset has lost because some qualified who would not have done so under the old system: see excerpt from testimony in paragraph 30. [33] Notwithstanding that, as Iaccobuci J. said in the Law case, at paragraph 66, "an infringement of section 15(1) may be established by other means, and may exist even if there is no one similar to the claimant who is experiencing the same unfair treatment". I accept that there are other persons who share the same characteristics as the respondent and who were denied the benefits. Although I have doubts that there exists a causal relationship between the denial of benefits and the alleged characteristics, I am also willing to accept that the respondent experienced differential treatment on account of her gender and parental status. [34] Before leaving this issue, I need to address an additional argument raised by counsel for the applicant. [35] The claim made by the respondent is a claim made in her capacity as a "woman in a parental status who is working part-time". Indeed, the Umpire concluded at paragraph 59 of his decision that a mother who works part-time because of her unpaid responsibilities should not receive inferior employment insurance coverage on that account. Counsel for the applicant submitted that, as the statement reveals, the Umpire changed the focus from parental status alone to a combination of parental status and employment status. In so doing, he submitted that the Umpire erred in law because employment or occupational status is not an immutable personal characteristic: see Delisle v. Canada (Deputy Attorney General), [1999] 2 S.C.R. 989, per Bastarache J. at paragraph 44; Corbière v. Canada, supra; Canada (Attorney General) v. Taylor (1991), 81 D.L.R. (4th) 679 (F.C.A.). [36] With respect, I do not think that the Umpire's reference to part-time workers was meant to be a reference to a personal characteristic of the respondent. I believe such reference merely establishes the circumstances under, or the moment at, which the respondent and members of her group are discriminated. To put it differently, it is simply another way of saying that the respondent as a woman in a parental status is discriminated under the Act when she works part-time. Whether the differential treatment suffered by the respondent is based on enumerated or analogous grounds [37] The finding of the Umpire on the issue of analogous grounds, as these terms are used in section 15 of the Charter, is found in paragraph 59 of his decision. I have previously cited that paragraph. There is no doubt that the Umpire concluded that parental status is an analogous ground within the meaning of section 15. I believe it is fair to say, on a careful reading of his decision, that he combined it with gender to form one analogous ground: women in a parental status. I agree with him that that status is immutable until the time comes when the children are no longer in need of a caregiver. The period of immutability is, in my view, sufficiently long to satisfy the criteria for identifying new analogous grounds because the recognized ground in this case is "based on characteristics that we cannot change or that the government has no legitimate interest in expecting us to change to receive equal treatment under the law": see Corbière v. Canada, supra, at paragraph 13 per McLachlin and Bastarache JJ. [38] Consequently, I am satisfied that the differential treatment in the case at bar is based on an analogous ground. This brings me to the last stage of the subsection 15(1) analysis, namely whether the differential treatment discriminates. Whether the differential treatment amounts to discrimination under subsection 15(1) [39] The purpose of section 15 is to "prevent the violation of essential human dignity and freedom through the imposition of disadvantage, stereotyping, or political or social prejudice, and to promote a society in which all persons enjoy equal recognition at law as human beings or as members of Canadian society, equally capable and equally deserving of concern, respect and consideration": see Law v. Canada, supra, at paragraph 51. [40] It is the third stage of the subsection 15(1) analysis that is most directly involved with the concept of human dignity. The overarching question here is "whether the impugned law or provision perpetuates the view that certain persons are less capable or less worthy of recognition or value as human beings or as members of Canadian society". A claimant must show that her human dignity and/or freedom is adversely affected: see Lavoie v. Canada, [2002] S.C.J. No. 24, 2002 SCC 23, at paragraphs 42 and 47. The Supreme Court has indicated that it is to be an objective-subjective one, the objective part of which is to be understood in light of various contextual factors. As explained by Iacobucci, J. in Law, supra at paragraph 61: Equality analysis under the Charter is concerned with the perspective of a person in circumstances similar to those of the claimant, who is informed of and rationally takes into account the various contextual factors which determine whether an impugned law infringes human dignity, as that concept is understood for the purpose of s. 15(1). [41] In Lovelace v. Ontario, [2000] 1 S.C.R. 950, at paragraph 55, Iacobucci, J., again for a unanimous Court, re-iterated the directed nature of the dignity inquiry: In appreciation of the depth and complexity of the human dignity interest, the discrimination inquiry demands a full contextual analysis. However, this contextual analysis is a directed inquiry; it is focused through the application of contextual factors which have been identified as being particularly sensitive to the potential existence of substantive discrimination... The need for a contextual analysis was once again stressed in Gosselin v. Quebec (Attorney General) 2002 SCC 84, at paragraphs 24 per McLachlin C.J. At paragraph 25, she writes for the majority: The need for a contextual inquiry to establish whether a distinction conflicts with s. 15(1)'s purpose is the central lesson of Law. This is understandable because the contextual analysis brings a measure of objectivity in what could otherwise be an unfair exercise in subjectivity. [42] The Supreme Court identified four contextual factors relevant to the third stage of the discrimination analysis: i) pre-existing disadvantage, stereotyping, prejudice, or vulnerability; ii) the correspondence, or lack thereof, between the ground(s) on which the claim is based and the actual needs, capacity or circumstances of the claimant or others; iii) the ameliorative purpose or effects of the impugned law, program or activity upon a more disadvantaged person or group in society, and; iv) the nature and scope of the interest affected by the impugned government activity: see Law and subsequently Lovelace, supra, at paragraphs 68-69. In Gosselin, supra, at paragraph 29, the Chief Justice indicated that these four factors, which are not exhaustive, must be considered although none of these factors is a prerequisite for finding discrimination and not all factors will apply in every case. [43] In the Lovelace case, at paragraph 57, Iacobucci, J. stated that "the purposive s. 15(1) inquiry requires a claimant to advert to factors which contextually establish a breach of the human dignity interest". In its consideration of the factors advanced, the Court clearly emphasized the requirement that these be established. For example, in that case, the appellant Aboriginal groups were found to "have most certainly established pre-existing disadvantage, stereotyping, and vulnerability" in general terms, but had "failed to establish that the First Nations Fund functioned by device of stereotype": see paragraph 73 of the decision. The need for contextual factors to be established is in keeping with the claimant's basic onus under section 15 of demonstrating a discriminatory effect. As per Iacobucci, J.'s dictum in Law, supra at paragraph 67, "[t]here is no principle or evidentiary presumption that differential treatment for historically disadvantaged persons is discriminatory". [44] The respondent's arguments, both at the hearing and in the written pleadings, do not involve an attempt to establish the presence of these contextual factors which are undoubtedly relevant in the present instance. Very little evidence relating to these factors is provided. Similarly, the Umpire failed to pay much attention, directly or indirectly, to the aforementioned factors, except perhaps the second one as we shall see. As for the other three, to the extent that it can be said that he did pay some attention to them, he certainly did not tie them to any evidence. I will, therefore, proceed to my own consideration of these contextual factors as the Supreme Court did in the Gosselin case. [45] The first of the contextual factors asks us to consider the pre-existing disadvantage, stereotyping, prejudice, or vulnerability experienced by the individual or group. While it may be true that women have historically faced barriers to their entry into the workforce and that these barriers are rooted in stereotypes and prejudices, the respondent has not established that, in the context of unemployment insurance, there was a past and long history of disadvantages, stereotyping, vulnerability and prejudice caused by the MERs under the old system. Indeed, as previously mentioned, the respondent would like to return to the old system. Nor has she established that the new MERs function by device of stereotype or prejudice. They do not layer disadvantage and vulnerability onto the respondent or persons in situations similar to hers. The evidence establishes that the vast majority of employed women with young children work more than the minimum eligibility requirements in their qualifying period and are unaffected by those requirements. These requirements do not create or reinforce a stereotype that women should stay home and care for children. Nor do these requirements affect the dignity of women by suggesting that their work is less worthy of recognition. Anyone who works the requisite number of hours in their qualifying period will qualify. It would stretch reason to imagine that reasonable persons in the respondent's situation would feel themselves any less valuable as a worker or as a member of society by the mere fact of having narrowly fallen short of qualifying for EI benefits in a given year. Rather, I would imagine that a reasonable person would simply feel that they had narrowly missed qualifying because of an unfortunate confluence of events. [46] The correspondence between the ground(s) and the actual needs, capacity or circumstances of the claimant and others is arguably the only contextual factor that the Umpire adverted to in his reasoning. He felt that, by basing the threshold for eligibility and the definition of work attachment on the average work week of 35 hours, the eligibility requirements failed to take into account the fact that women's average work week in the paid labour force is lower than this average. He also referred to evidence that women perform more unpaid obligations than men. [47] In my opinion, these considerations do not provide a sufficient basis for the Umpire to be able to ground a finding that the impugned requirements do not correspond to the actual needs of the claimant and persons like her. In fact, there is much evidence in the record that suggests the opposite. Part of the rationale for moving to a system based on hours worked was to foster a flexibility that is reflective of modern and diverse work arrangements. In this light, it can be said that the legislation takes into account the traits and circumstances of women who are in a parental status by allowing them to structure their paid employment more flexibly so that more of them are able to qualify for benefits under the new system, whereas they would not have been able to under the old system. In her testimony, Dr. Nakamura referred to an identified working pattern for some married women with children according to which they work two hours a day in the morning, five days a week, and an eight-hour shift on Saturdays. By working 36 weeks on week days and 44 weeks on Saturdays, these workers accumulate 712 hours and now qualify for general and maternity benefits as well as training while they would have been excluded under the previous system: see Applicant's Record, vol. 1, pages 360-361. [48] In addition, Parliament was conscious that some women in a parental status would not qualify for EI benefits and that income would be needed from some other sources to offset the potential harm resulting from the exclusion. Looking at other ways of doing that, it provided for supplementary child tax credits and took measures to improve collection of monies from absent spouses as Dr. Nakamura pointed out in her testimony: see excerpt in paragraph 29. [49] The third contextual factor involves the consideration of the ameliorative nature of the legislation and is chiefly relevant only in respect of situations of so-called reverse discrimination. As explained by Iacobucci, J. in Law, supra, at paragraph 88, an ameliorative purpose or effect which accords with the purpose of subsection 15(1) of the Charter will likely not violate the human dignity of more advantaged individuals where the exclusion of these more advantaged individuals largely corresponds to the greater need or the different circumstances experienced by the disadvantaged group being targeted by the legislation. The present case does not involve a claim of discrimination by an "advantaged" person. It is true, however, that the legislation does have a broadly ameliorative purpose. It aims to provide income replacement to significant contributors who find themselves temporarily out of work. Moreover, it does so in a manner that is sensitive to regional rates of unemployment. The legislation's purpose is consistent with the equality values enshrined in subsection 15(1) of the Charter. [50] The fourth contextual factor asks us to consider the nature and scope of the interest affected by the impugned law. The more severe and localized the consequences of the legislation for the affected group, the more likely that discrimination will be founded. In the present case, the consequences are neither severe nor overly localized for women who are in a parental status. In fact, the evidence does not bear witness to much localization at all. The differential treatment is as between those who work at or above the threshold requirement for hours and those who fall short of this threshold. It is not localized on women and/or women in parental status in any statistically significant manner: see the testimony of Dr. Nakamura, previously cited in paragraph 31. [51] On the basis of the foregoing, I find that the respondent has not discharged her onus of establishing one or more of the contextual factors which would support her contention that her human dignity has been demeaned within the context of subsection 15(1) of the Charter. I acknowledge Iacobucci, J.'s comments in Law, supra, at paragraph 88, that it is not always necessary for the claimant to adduce evidence in order to show a violation of human dignity and that there will be cases where such a determination can be made on the basis of judicial notice and logical reasoning. However, I believe I would be stretching the concept of judicial notice beyond recognition and acting unfairly in this case if, on that sole basis, I were to conclude that a violation of human dignity exists in light of the historical, social, political and legal context of the respondent's and in light of the new eligibility requirements for unemployment benefits. The acts underlying the respondent's context as well as the rationale for the enactment of the new eligibility requirements are too controversial to lend themselves to an exercise of such judicial power. It is not at all plainly obvious that the distinction at issue has the effect of demeaning the human dignity of the respondent or persons like her. The eligibility requirements are not a manifestation of a lack of respect or loss of dignity. They are an administratively necessary tool tailored to correspond to the requirements of a viable contributory insurance scheme. Whether subsections 6(1) and 7(2) of the Act are saved by section 1 of the Charter [52] If I am wrong in my conclusion and, consequently, subsections 6(1) and 7(2) of the Act do violate the equality rights conferred by section 15 of the Charter, I believe that these provisions are saved by section 1. Is there a pressing and substantial objective? [53] The Umpire found that the legislation was not pressing and substantial. In these proceedings for judicial review, counsel for the respondent conceded that the Act is pressing and substantial in that the intended purpose of providing replacement income when an employee loses his or her job, or is unable to work because of illness, childbirth or parental responsibilities, is a pressing and substantial objective. He also agreed that entrance requirements are legitimate, pressing and substantial objectives, but submitted that these objectives could be better achieved by means other than the ones embodied in subsections 6(1) and 7(2). It is not disputed that the MERs' purposes are to eliminate fiscal incentives for employers to create part-time jobs with less than 15 hours per week; to ensure that recipients of benefits have a minimum degree of attachment to the workforce; to reduce the overall level of dependency on employment insurance benefits; and to provide a plan fiscally viable in which benefits are proportional to contributions. However, counsel for the respondent argued, for example, that it is quite possible to change the actual concept of entrance requirement by having a different mix without altering the fiscal impact on the regime. I shall come back to this issue when I discuss the concept of minimal impairment of the equality rights. [54] Therefore, I conclude that both the Act and the impugned provisions pursue valid substantial and pressing objectives. Is there a rational connection between Parliament's objectives and the impugned provisions? [55] Counsel for the respondent contended that there is no rational connection between a MER scheme that focuses exclusively on hours worked and the objective of ensuring workforce attachment. The Umpire agreed with this contention because, in his view, commitment to the workforce is better represented and attained by the duration of the commitment and MERs' which reflect this notion of duration. Thus, a person could work only two hours a week for years and have a strong attachment to the workforce. Indeed, counsel for the respondent made it clear that he wants to resort back to the old system which had a duration component in terms of weeks along with an hour requirement. [56] With respect, I fail to see how it can be said that the quantity of hours worked is unconnected or not rationally connected to the workforce attachment. The hours requirements involve a notion of duration which can be transposed in days and weeks. It is somewhat ironic that the respondent readily sees a rational connection to the workforce attachment in weeks requirements, but no such connection in a system based on hours worked when, for all practical purposes, she complains that the minimum requirement of 700 hours is too long a duration. The quantity of hours may not be the best indicator of attachment. It is certainly not the one that the respondent wishes, but it still constitutes an indicator of attachment with some rational connection to the workforce attachment. [57] Counsel for the respondent avers that the MERs, as they now stand, do not discourage misuse of the Act. Yet, the respondent would like a lowering of the threshold. However, a minimum threshold at some level aims at preventing situations where employees do not exhibit much attachment to paid work and habitually opt for unemployment benefits instead of continuing that work. Once again, with respect, I see a rational connection between a minimum threshold requirement and the objective of preventing a misuse of the Act. [58] In addition, it appears to me indisputable that the system based on hours worked is also rationally connected to two other important objectives of the Act. The first one was the elimination of fiscal incentives for employers to create part-time jobs with less than 15 hours per week, thereby avoiding their obligation to pay contributions to the plan and leaving the employees uninsured. The second aimed at implementing a fiscally viable plan in which benefits are proportional to contributions. Indeed, a contributory insurance scheme requires that a line be drawn somewhere in order to function effectively and be viable. The MERs in subsections 6(1) and 7(2) draw this necessary line and are undoubtedly rationally connected to this objective. [59] Finally, it was a stated objective of the new Act to keep the employment insurance system in tune with changing labour market trends: see the evidence of P. Phillips, Applicant's Record, vol. 4, pages 3281-3282. As it appears from Parliament's intent, there was a need and a desire to introduce greater flexibility in the system in order to better accommodate the growing variety of historically non-typical work arrangements, including the rise in part-time work and a higher participation of women in the new economy. [60] Parliament, after long and substantial studies, consultations and debates thought that a requirement based on hours worked provided that flexibility as every hour of work, whether resulting for a single or more than one part-time employment, would count towards the worker's contributions to the system. I am satisfied that the MERs are rationally connected to this objective as well. [61] With respect, I believe the learned Umpire overlooked some fundamental objectives of the Act and the impugned provisions in assessing the rational connection of the MERs. Had he not done so, I think he would have found a rational connection between the requirements based on hours worked and the objectives of the Act and those of subsections 6(1) and 7(2). Do the provisions minimally impair the section 15 equality rights? [62] The Supreme Court has warned against courts adopting an overly exacting approach to the issue of minimal impairment. In RJR-MacDonald Inc. v. Canada, [1995] 3 S.C.R. 199, at paragraphs 159 and ff., McLachlin J. (as she then was) described in the following terms the approach to be taken: The impairment must be "minimal", that is the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement... On the other hand, if the government fails to explain why a significantly less intrusive and equally effective measure was not chosen, the law may fail. (emphasis added) [63] Therefore, an important consideration in the minimal impairment analysis is the extent to which the Court should grant deference to the legislature. The Act is a complex insurance program subject to rigid requirements. The requirement of a minimum number of qualifying hours is an integral aspect of the insurance scheme. The minimum number of qualifying hours is necessary to ensure the fiscal integrity of the program and to prevent abuses. The number of qualifying hours varies according to geographical location based upon estimates of the opportunity for finding employment in specified regions. The absence of a threshold would change the nature of this program into a social assistance program. [64] The respondent's case is instructive. Ms. Lesiuk worked 667 hours in the relevant 12-month period. Had she been unemployed in Brandon where she had worked, she would have qualified for employment insurance because, in Brandon, the minimum number of qualifying hours at the time was 665. However, her claim for employment insurance was based on her decision to move to Winnipeg to join her husband. She was told she would have to apply for employment insurance in Winnipeg because that is where she was residing and where she would find opportunities for employment. However, because jobs were more readily available in Winnipeg than in Brandon, the minimum number of qualifying hours in Winnipeg was 700. She was, therefore, 33 hours short and not entitled to employment insurance. [65] Ms. Lesiuk's case is an example of the complexity of the employment insurance program. Hers, and many other cases regularly heard by umpires, are sympathetic and I understand the frustration that Ms. Lesiuk must have felt, having met the minimum number of qualifying hours in Brandon but not in Winnipeg. Perhaps this argues for an adjustment to the Regulations applying the minimum number of qualifying hours where the hours were earned rather than in the location to which an individual moves. Undoubtedly, there are counter-arguments also. [66] What this points out is that in a complex program such as employment insurance, the Court, with no expertise in the area, is not well equipped to propose changes to such programs. In my respectful opinion, this is not a case in which the Court should superimpose upon an already complex program additional required adjustments. I do not say that a differential between minimum threshold requirements as between the respondent's group and men may not be desirable. However, such a change will obviously give rise to other considerations. Should the threshold for the respondent's group be lowered or should the threshold for men be increased? If a differential is material, should there be other minimum requirements for the respondent's group and what should they be? Are there other groups who can justify lower requirements? What groups should they be compared to and what additional requirements should be imposed upon those groups, if any? How is this Court to assess and minimize the amount of disruption to the economies in different regions?: see the testimony of Dr. Nakamura where she mentions that this was an important concern of the government when it moved from the old to the new system. Applicant's Record, vol. 1, page 70. [67] Whatever the minimum entrance requirement, there will always be persons or groups who will not be able to qualify. The following two comments made by the majority and found at paragraph 55 of the Gosselin case are appropriate in the present case: Perfect correspondence between a benefit program and the actual needs and circumstances of the claimant group is not required to find that a challenged provision does not violate the Canadian Charter. The situation of those who, for whatever reason, may have been incapable of participating in the programs attracts sympathy. Yet the inability of a given social program to meet the needs of each and every individual does not permit us to conclude that the program failed to correspond to the actual needs and circumstances of the affected group. As Iacobucci J. noted in Law, supra, at paragraph 105, we should not demand "that legislation must always correspond perfectly with social reality in order to comply with subsection 15(1) of the Charter". Crafting a social assistance plan to meet the needs of young adults is a complex problem, for which there is no perfect solution. No matter what measures the government adopts, there will always be some individuals for whom a different set of measures might have been preferable. The fact that some people may fall through a program's cracks does not show that the law fails to consider the overall needs and circumstances of the group of individuals affected, or that distinctions contained in the law amount to discrimination in the substantive sense intended by subsection 15(1). Our role in such a situation is to ensure that the legislator has been reasonable in choosing the place where the line is to be drawn. I am satisfied that Parliament's choice in this case falls within a range of reasonable alternatives. The appropriate remedy for the respondent and the two interveners in this case, Women's Legal Education and Action Fund and Income Security Advocacy Centre, is with the legislature, not the Courts. The proportionality test: does the abridgment of the right outweigh the obtainment of the legislative goal? [68] The last step of the section 1 analysis involves a balancing act: do the infringing effects of the impugned provisions outweigh the importance of the objective sought. As Bastarache J. said in Lavoie, supra, at paragraph 70, this stage: ...examines the nature of the infringement and asks whether its costs outweigh its benefits. The implication of finding a violation at the fourth stage is that even a minimum level of impairment is too much: the costs to the claimant so outweigh the benefits that no solace can be found in the fact that the legislation violates the Charter "as little as reasonably possible". Moreover, if the costs of the legislation are significant enough, and the legislation only partially achieves its objectives, greater evidence of its benefits may be necessary in order to survive s. 1. [69] As previously mentioned, the EI Act is a fundamental insurance plan which provides coverage to workers and other special beneficiaries for their loss of income. The benefit of such a law is very substantial. Under the Act, the risk is pooled and the cost of high-risk participants is borne by low-risk participants. Claimants receive benefits which exceed the amount of the premiums they pay. Meaningful insurance coverage is provided to employees with a minimal work force attachment. The MERs set the limits of this minimal work force attachment. They are essential to the proper functioning and effectiveness of the insurance scheme. Their importance in determining the extent of the risk assumed, the costs of operating the system and the level of benefits paid cannot be doubted. Alterations of the MERs can induce a greater risk and result in an increase in the premiums or a reduction in the level of benefits. [70] I do not wish to diminish the deleterious effects that the MERs have on employees like the respondent who cannot meet the threshold. She lost the benefits as a result of her failure to qualify by very few hours. In terms of access to the benefits, the respondent's right to be treated equally with men, whether in a parental status or not, is also fundamental. However, I cannot conclude that the negative effects resulting from the exclusion of some workers from eligibility for benefits overwhelmingly outweigh the positive social effects of the employment insurance scheme. To paraphrase the words of Bastarache J., I cannot say that "the costs to the claimant so outweigh the benefits that no solace can be found in the fact that the legislation violates the Charter "as little as reasonably possible". [71] For these reasons, I would allow the application for judicial review with costs, set aside the decision of the Umpire and refer the matter back to the Chief Umpire or his designate for determination on the basis that the respondent's appeal from the decision of the Board of Referees should be dismissed. "Gilles Létourneau" J.A. "I agree Marshall Rothstein J.A." "I agree B. Malone J.A." FEDERAL COURT OF CANADA APPEAL DIVISION NAMES OF SOLICITORS AND SOLICITORS ON THE RECORD COURT FILE NO.: A-281-01 STYLE OF CAUSE: The Attorney General of Canada v. Kelly Lesiuk PLACE OF HEARING: Edmonton, Alberta DATE OF HEARING: November 19 and 20, 2002 REASONS FOR JUDGMENT BY:LÉTOURNEAU J.A. CONCURRED IN BY: ROTHSTEIN J.A. MALONE J.A. DATED: January 8, 2003 APPEARANCES: Mark Kindrachuk Rochelle Wempe FOR APPLICANT Byron Williams Evelyn Braun FOR RESPONDENT Joanne M. Boulding FOR INTERVENER, INCOME SECURITY ADVOCACY CENTRE Kerri Froc FOR INTERVENER, WOMEN'S LEGAL EDUCATION AND ACTION FUND SOLICITORS OF RECORD: Morris Rosenberg FOR APPLICANT Deputy Attorney General of Canada Ottawa, Ontario Public Interest Law Centre FOR RESPONDENT Winnipeg, Manitoba Income Security Advocacy Centre FOR INTERVENER, INCOME SECURITY AND Toronto, Ontario ADVOCACY CENTRE Balfour Moss FOR INTERVENER, WOMEN'S LEGAL Regina, Saskatchewan EDUCATION AND ACTION FUND