Printer page

Printer page - Content

Source: 1994fca0280
[1994] 2 F.C. 189


Thibaudeau v. M.N.R.

A-1248-92

Suzanne Thibaudeau (Applicant)

v.

Her Majesty the Queen (Respondent)

Indexed as: Thibaudeau v. M.N.R. (C.A.)

Court of Appeal, Pratte, Hugessen and Létourneau JJ.A."Québec, February 28 and March 1; Ottawa, May 3, 1994.

Income tax " Income calculation " Application for judicial review of Tax Court decision dismissing appeal from assessment " Applicant separated custodial parent " Income Tax Act, s. 56(1)(b) requiring inclusion in income of child support payments made pursuant to divorce judgment (inclusion/deduction system) " Separated custodial parents subject to discrimination based on family status " Cannot rely on general obligation in family law to gross up maintenance payments to take account of tax consequences in Charter-based attack " Awarding of child support imprecise, discretionary " Not legitimate to look outside income tax system to correct injustice created thereby " If charge offending Charter, s. 15, logically indefensible to claim no breach because source can adjust to correct inequality.

Constitutional law " Charter of Rights " Equality rights " Income Tax Act, s. 56(1)(b) requiring inclusion in income of child support payments made pursuant to judgment " Not discriminating on ground of sex " Legislation adversely affecting men and women not discriminating on ground of sex solely because women (or men) affected more numerous " Focus not on numbers, but on nature of effect " Legislation discriminating on ground of sex only when some members of one sex more adversely affected than equivalent group of opposite sex " Neither phrase "separated custodial parent" nor any of its components constituting ground analogous to those enumerated in s. 15 " Separated custodial parents subject to discrimination based on "family status" " Discrete and insular minority historically suffering prejudice and in need of protection.

Constitutional law " Charter of Rights " Limitation clause " Income Tax Act, s. 56(1)(b), requiring inclusion in income of child support payments made pursuant to judgment, discriminating against separated custodial parents on ground of family status " Objective of inclusion/deduction system (tax subsidy allowing former spouses greater financial resources than when living together, compensating in part for lost economies of maintaining single household) sufficiently important to warrant Charter breach " S. 56(1)(b) and objective rationally connected " S. 1 justification failing because system not meeting minimum impact, proportionality tests " Evidence system frequently failing to benefit those intended to benefit, usually benefitting those not in need of assistance, and containing no corrective mechanisms to remedy problem.

Judicial review " Equitable remedies " Declarations " Court asked to delay invalidity declaration to allow for amendment of legislation " Court unable to delay declaration Income Tax Act, s. 56(1)(b) invalid on application for judicial review of Tax Court decision " Could only order Tax Court dispose of appeal as Court might have done originally " Rights of individuals guaranteed by supreme law of country not suspended without very strong reasons therefor.

This was an application for judicial review of the Tax Court's dismissal of the applicant's appeal from her 1989 tax assessment. The applicant is the custodial parent of two minor children. By the terms of a divorce judgment she was awarded $1,150 per month for the maintenance of the children. Income Tax Act, paragraph 56(1)(b) required the inclusion in income of any amount received pursuant to a judgment as an allowance payable on a periodic basis for the maintenance of children of the marriage, if the recipient was separated pursuant to a divorce from the spouse required to make the payment. Paragraph 60(b) permitted the applicant's former husband to deduct the maintenance payments in the computation of his income for tax purposes. Companion provisions provide for the similar inclusion and deduction of other payments of a similar nature. Before the Tax Court the applicant alleged that this system, known as the inclusion/deduction system, violated her equality rights as guaranteed by Charter, section 15. Notwithstanding its finding that the applicant had suffered a "tax cost" of $3,705 resulting from the inclusion/deduction system, the Tax Court dismissed the appeal because the Court which had awarded maintenance payments had taken into account the effects of income tax in fixing the amount of those payments, and in any event there was a general obligation in family law to gross up maintenance payments to take account of tax consequences. The applicant claims discrimination based on a ground analogous to those enumerated in section 15. The intervenor alleges that paragraph 56(1)(b) discriminates based on sex. The group to which the applicant claims to belong is separated, custodial parents receiving maintenance payments for their children.

Held (Létourneau J.A. dissenting), the application should be allowed.

Per Hugessen J.A. (Pratte J.A. concurring): Paragraph 56(1)(b) is facially neutral: it does not create distinctions based on sex; it is not a colourable stratagem whose provisions can apply only to members of one sex. Legislation that adversely affects both men and women is not discriminatory on the ground of sex solely because the women (or men) affected are more numerous. Such a mechanistic approach would be likely to defeat the purposes of the Charter. It is not because more women than men are adversely affected, but because some women, no matter how small the group, are more adversely affected than the equivalent group of men, that a provision can be said to discriminate on grounds of sex. Sex differs significantly from the other grounds enumerated in section 15. There is an almost infinite number of religions, races, nationalities etc. and no two subsets within any of those categories could properly be described as opposites. There are only two sexes. One excludes the other. Women who claim that a law discriminates on the basis of sex necessarily do so because it draws a distinction based on their shared characteristic of femaleness which it does not draw for those who have the opposite characteristic of maleness. Accordingly one cannot say that an otherwise neutral rule discriminates on the basis of sex simply because it affects more members of one sex than of the other. The focus is not on numbers but on the nature of the effect"on quality rather than quantity. If legislation which adversely affects women has the same adverse effect upon men, even though their numbers may be smaller or the likelihood of their suffering less, it cannot logically be said that the ground of discrimination is sex. In each case the impact of the legislation must be weighed; it is not enough simply to count the numbers affected. Paragraph 56(1)(b) impacts adversely on more women than men because mothers are far more likely to be custodial single parents than fathers. Since, however the legislation must also impact in exactly the same way on custodial fathers, although in very much smaller numbers, it does not discriminate on the basis of sex.

Paragraph 56(1)(b) draws an intentional distinction between the applicant and others based upon her being a separated custodial parent. The qualities of being separated and a parent are "personal characteristics". Second, the inequality created for separated custodial parents is discriminatory and imposes a burden on them not imposed on others. This may be tested by looking at each of the defining components of the group in turn. Thus a non-separated custodial parent is not required to include in income maintenance payments received from her or his spouse. A separated non-custodial parent is not required to include maintenance payments in income. Separated non-parents having custody of a child, such as a grandmother, are not obliged to include in their income payments received from either or both of the child's parents. The third and last stage of the section 15 analysis is to inquire whether the personal characteristics at issue constitute grounds analogous to those enumerated. The authorities require an examination of any alleged analogous ground in the broadest context with a view to determining whether such ground has historically served or today actually serves to single out or distinguish a disadvantaged group. Neither the phrase "separated custodial parents" nor any of its components constitute a ground analogous to those enumerated in section 15. The definition of the group must have some component which is included within the alleged ground, but it is unlikely to be the ground itself. The appropriate description of the ground of discrimination to which separated custodial parents are subject would be "family status". Such status has historically been, and is still, used as a basis for stereotyping. That family status or some similar expression figures as a prohibited ground of discrimination in most human rights statutes also confirms its analogous nature to the grounds enumerated in the Charter. Separated custodial parents are a discrete and insular minority which has historically suffered prejudice and has need of protection.

While the Tax Court Judge was correct in his view that courts should take account of income tax consequences and that the Court which granted the applicant's divorce did in fact do so, he was wrong to see that as an answer to a Charter-based attack on paragraph 56(1)(b). The family law system does not always, or even usually, operate to correct the inequality created by paragraph 56(1)(b). The awarding of child support is notoriously imprecise. It is virtually impossible to rely on the family law system as a corrective to any injustice created by the inclusion/deduction system. Second, it is not legitimate to look outside the income tax system to correct an injustice which that system has created. Income tax comes after income. When the Act creates a charge, it does so on the basis that there is a source of income upon which that charge can operate. If the charge itself offends the equality provision of the Charter, it is logically indefensible to claim that there is no breach because the source can adjust to correct the inequality. If the tax is in breach of section 15, calling the family law system in aid is putting the cart before the horse.

As to the section 1 analysis, the objective of the inclusion/deduction system (a tax subsidy, allowing former spouses greater financial resources than when living together, thereby compensating in part for the lost economies of maintaining a single household) is sufficiently important to warrant a Charter breach. There is a rational connection between the impugned provision and the legislative objective. But as the system does not meet the minimum impact and proportionality tests, the section 1 justification failed. The material produced did not establish a measured, proportionate response to a perceived problem, but demonstrated that the inclusion/deduction system frequently fails to give any benefit at all to those whom it is allegedly designed to assist, almost always benefits those who do not need assistance, and contains no corrective or control mechanisms designed to remedy the problem. The Government failed to satisfy the burden of proving section 1 justification.

The tax disadvantage suffered by separated custodial parents was not corrected by the equivalent to married credit, the dependant tax credit, and the child tax credit, which are independent of the inclusion/deduction system. In any event, the aggregate of all such credits was less than the amount of the tax loss actually suffered by the applicant as a result of the inclusion/deduction system.

The respondent argued that if paragraph 56(1)(b) were found invalid by reason of section 15, then paragraph 60(b) must also fail. Since paragraph 60(b) was not directly in issue in these proceedings, it would be most improper to make any finding with regard to it. Other considerations could result in a court finding that the paragraph could survive a Charter attack. Paragraph 60(b) played no role in the calculation of the applicant's income or her assessment.

As to delaying the declaration of invalidity to allow time for amendments to the legislation, as this was an application for judicial review of the decision in an income tax appeal, the Court could only ultimately order that the Tax Court dispose of the applicant's appeal in a way that the Court might have done originally. The Court was dealing with the rights of individuals which are guaranteed to them by the supreme law of the country. It would take very strong reasons to justify any suspension of those rights. None was suggested. Finally, it was unlikely that this Court would have the last word on the subject.

Per Létourneau J.A. (dissenting): The applicant was not subject to discrimination because of her civil status or social condition. While paragraph 56(1)(b) imposes a burden on the applicant and on the class of persons to which she belongs which it does not impose on other groups with different civil status, the difference in treatment thereunder does not create inequality and is not discriminatory. The Act sets up a whole set of distinctions and differences in treatment which take into account the economic reality that is a result of the taxpayer's civil status. It is not necessarily discriminatory per se to treat different civil statuses differently. On the contrary, it might be discriminatory to treat different civil statuses identically, when they create different duties and responsibilities for the people in question, and which consequently demand different treatment in social, political, economic and legal terms. It is necessary to examine the general context surrounding the Act to determine whether there is discrimination. The difference in treatment is not based on civil status (a ground that is analogous to those listed in section 15), but results from the different physical constraints and duties that fall upon taxpayers who live in different economic situations because of their different personal statuses. When read and taken literally in isolation, the measure may appear discriminatory, but it is not when it is placed in its socio-economic and socio-political context and the goal in mind is taken into account. This action questions the constitutional validity of a legislative measure designed to remedy the disadvantaged situation in which the applicant's group was placed before. It also questions the standard of effectiveness that such a measure must meet if it is to be constitutionally valid. The beneficiaries of the inclusion/deduction system are mostly women, a majority of whom, particularly of low-income women, receive an advantage and a benefit from the income-splitting permitted by the system. Paragraph 56(1)(b) therefore produces a beneficial effect in a majority of cases and helps to partially remedy the inequalities suffered by this group of taxpayers in the past. The remedy created by paragraph 56(1)(b) treats the applicant and the people in her group differently from other taxpayers because of their different social, political and economic circumstances which are the result of their different civil status, and despite its imperfections is not discriminatory.

The applicant argued that divorced women have a unique social condition because of their income and level of education and are in a disadvantaged position in society. The argument that the discrimination of which she was a victim was based on social condition was without merit. The applicant described the social condition in which she and the members of her group find themselves, or the result of discrimination, not the cause. There was no evidence that the alleged discrimination was dictated by the applicant's social condition. On the contrary, the legislative provision was designed to remedy the social condition in which taxpayers in her group find themselves because of divorce or separation as a result of the breakdown of the marriage.

statutes and regulations judicially considered

Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44], ss. 1, 15(1).

Charter of Human Rights and Freedoms, R.S.Q. 1977, c. C-12, s. 10 (as am. by S.Q. 1982, c. 61, s. 3).

Federal Court Act, R.S.C., 1985, c. F-7, s. 18.1(3) (as enacted by S.C. 1990, c. 8, s. 5).

Income Tax Act, S.C. 1970-71-72, c. 63, ss. 56(1)(b), 60(b), 63 (as am. by S.C. 1974-75-76, c. 26, s. 33; 1976-77, c. 4, s. 21; 1980-81-82-83, c. 109, s. 19; 1984, c. 1, s. 25; c. 45, s. 22; 1988, c. 55, s. 39; 1990, c. 39, s. 13), 70(6) (as am. by S.C. 1973-74, c. 14, s. 19; 1974-75-76, c. 26, s. 38; 1985, c. 45, s. 33), 73 (as am. by S.C. 1973-74, c. 14, s. 20.1; 1977-78, c. 32, s. 14; 1980-81-82-83, c. 48, s. 39; c. 140, s. 41; 1985, c. 45, s. 35; 1986, c. 6, s. 36; 1988, c. 55, s. 51), 74.1(2) (as am. by S.C. 1986, c. 55, s. 17), 118(1)(a) (as am. by S.C. 1988, c. 55, s. 92), (b) (as am. idem), (d) (as am. idem), 118.8 (as enacted idem), 122.2 (as enacted by S.C. 1978-79, c. 5, s. 4; as am. by S.C. 1984, c. 1, s. 65; 1988, c. 55, s. 97), 146(16)(b) (as am. by S.C. 1990, c. 35, s. 13), 171(1) (as am. by S.C. 1980-81-82-83, c. 158, s. 58, item 2), 251 (as am. by S.C. 1980-81-82-83, c. 140, s. 129; 1988, c. 55, s. 190).

Tax Court of Canada Act, R.S.C., 1985, c. T-2, s. 18.28 (as enacted by R.S.C., 1985 (4th Supp.), c. 51, s. 5).

cases judicially considered

applied:

Symes v. Canada, [1993] 4 S.C.R. 695; R. v. Oakes, [1986] 1 S.C.R. 103; (1986), 26 D.L.R. (4th) 200; 24 C.C.C. (3d) 321; 50 C.R. (3d) 1; 19 C.R.R. 308; 14 O.A.C. 335; Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; (1989), 56 D.L.R. (4th) 1; [1989] 2 W.W.R. 289; 34 B.C.L.R. (2d) 273; 25 C.C.E.L. 255; 10 C.H.R.R. D/5719; 36 C.R.R. 193; 91 N.R. 255; R. v. Swain, [1991] 1 S.C.R. 933; (1991), 75 O.R. (2d) 388; 71 D.L.R. (4th) 551; 63 C.C.C. (3d) 481; 5 C.R. (4th) 253; 3 C.R.R. (2d) 1; 125 N.R. 1; 47 O.A.C. 81; R. v. Turpin, [1989] 1 S.C.R. 1296; (1989), 48 C.C.C. (3d) 8; 69 C.R. (3d) 97; 39 C.R.R. 306; 96 N.R. 115; 34 O.A.C. 115; Gagnon v. The Queen, [1986] 1 S.C.R. 264; (1986), 25 D.L.R. (4th) 481; [1986] 1 CTC 410; 86 DTC 6179; 65 N.R. 321; 1 R.F.L. (3d) 113.

considered:

Brossard (Town) v. Quebec (Commission des droits de la personne), [1988] 2 S.C.R. 279; (1988), 53 D.L.R. (4th) 609; 10 C.H.R.R. D/5515; 88 CLLC 17,031; Foucher c. Québec (Procureur général), [1989] 1 R.J.Q. 703 (S.C.); Schachtschneider v. Canada, [1994] 1 F.C. 40 (C.A.).

referred to:

Moge v. Moge, [1992] 3 S.C.R. 813; (1992), 99 D.L.R. (4th) 456; [1993] 1 W.W.R. 481; 81 Man. R. (2d) 161; 30 W.A.C. 161.

authors cited

Canada. Department of Justice. Bureau of Review. Evaluation of the Divorce ActPhase II: Monitoring and Evaluation, Ottawa: Queen's Printer, 1990.

Canada. Report of the Federal/Provincial/Territorial Family Law Committee. The Financial Implications of Child Support Guidelines: Research Report, May 1992.

Dictionnaire de droit privé et Lexiques bilingues, 2e éd., Cowansville (Québec): Éditions Yvon Blais, 1991, "civil status".

Durnford, J. W. and S. J. Toope. "Spousal Support in Family Law and Alimony in the Law of Taxation" (1994), 42 Can. Tax J. 1.

Petit Larousse illustré, Paris: Librarie Larousse, 1984, "état civil" (civil status).

Vocabulaire juridique, publié sous la direction de Gérard Cornu, 2e éd., Presses Universitaires de France, Paris 1990, "état civil" (civil status).

Zweilbel, Ellen B. and Richard Shillington. Child Support Policy: Income Tax Treatment and Child Support Guidelines, Toronto: The Policy Research Centre on Children, Youth and Families, 1993.

APPLICATION for judicial review of the Tax Court's dismissal of the applicant's appeal of her 1989 tax assessment (Thibaudeau, S. v. M.N.R. (1992), 92 DTC 2111 (Eng.); 92 DTC 2098 (Fr.)). Appeal allowed.

counsel:

Michel C. Bernier and Richard Bourgault for appellant.

Guy Laperriere and Carole Johnson for respondent.

Mary Eberts and Steve Tenai for intervenor Support and Custody Orders for Priority Enforcement (SCOPE).

solicitors:

Bernier, Beaudry, Sainte-Foy, Quebec for applicant.

Deputy Attorney General of Canada for respondent.

Tory Tory DesLauriers & Binnington, Toronto, for intervenor Support and Custody Orders for Priority Enforcement (SCOPE).

The following are the reasons for judgment rendered in English by

Hugessen J.A.: This is an application for judicial review of a reported decision1*ftnote1 (1992), 92 DTC 2111 (T.C.C.). of the Tax Court of Canada dismissing the applicant's appeal from her 1989 tax assessment. Since the appeal in the Tax Court was governed by that Court's "informal procedure" the only recourse is by means of judicial review.2*ftnote2 No issue was taken in this Court as to the possible impact of s. 18.28 [as enacted by R.S.C., 1985 (4th Supp.), c. 51, s. 5] of the Tax Court of Canada Act, R.S.C., 1985, c. T-2, as amended, which provides that a judgment rendered under the informal procedure "shall not be treated as a precedent for any other case."

The applicant is a mother of two minor children of which she has custody. By the terms of the judgment granting the divorce between her and her former husband, the children's father, she was awarded no alimentary allowance for herself (she is gainfully but fairly modestly employed) but was awarded the sum of $1,150 per month for the maintenance of the children.

Under the provisions of the Income Tax Act3*ftnote3 S.C. 1970-71-72, c. 63, as amended. the payment received by the applicant for the maintenance of her children was required to be included in her income. Paragraph 56(1)(b), as it applied to the 1989 taxation year, required the inclusion in income of:

56. (1) . . . 

(b) any amount received by the taxpayer in the year, pursuant to a decree, order or judgment of a competent tribunal or pursuant to a written agreement, as alimony or other allowance payable on a periodic basis for the maintenance of the recipient thereof, children of the marriage, or both the recipient and children of the marriage, if the recipient was living apart from, and was separated pursuant to a divorce, judicial separation or written separation agreement from, the spouse or former spouse required to make the payment at the time the payment was received and throughout the remainder of the year.

If it were not for this text such alimony would not otherwise be taxable in the hands of the recipient under the general charging sections of the Act.

A counterpart to this provision, paragraph 60(b), permitted the applicant's former husband to deduct those same maintenance payments in the computation of his income for tax purposes. Companion provisions provide for the similar inclusion and deduction of other payments of a similar nature, for example those made by and to the respective parents of children born out of wedlock. The whole system is frequently referred to compendiously as the inclusion/deduction system.

Before the Tax Court the applicant's claim was in essence that the inclusion/deduction system violated her equality rights as guaranteed by subsection 15(1) of the Charter [Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44]]:

15. (1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

The Tax Court Judge dismissed that claim. Although there was little or no evidence before him on the point, he was prepared to take judicial notice of the fact that "the appellant is part of a group of which the great majority is separated or divorced women, who have a certain degree of financial self-sufficiency (in that they receive no alimony for themselves), who have custody of their children and who receive taxable alimony from their spouse for the benefit of the children."4*ftnote4 At p. 2118.

(It may be noted in passing that, as a result of the intervention by SCOPE, the absence of evidence which was remarked on by the Tax Court Judge has been more than amply remedied in this Court. Despite its great volume, however, the evidence does little more than confirm what the Tax Court Judge found as fact on this point.)

The Tax Court Judge then went on to say (at page 2118):

This group of people described earlier, of whom the appellant is one, is entitled, in my opinion, because of certain personal characteristics, to the guarantee set out in section 15 of the Charter, which prohibits all forms of discrimination based on the stated grounds or on analogous grounds.

The Tax Court Judge further found as a fact [at page 2120] that "the essential elements of the manner in which the tax impact on the appellant of including the alimony in her income" were "substantially" as they had been suggested by the expert witness Mr. Drouin, namely that in the year 1989, the applicant had suffered a "tax cost" of $3,7055*ftnote5 Reduced to $2,500 after taking account of the gross-up. resulting from the inclusion/deduction system.

Notwithstanding those findings, the Tax Court Judge dismissed the applicant's appeal because he was of the view that the Court which had awarded maintenance payments to the applicant had in fact taken account of the effects of income tax in fixing the amount of those payments and that there was, in any event, a general obligation in family law to gross-up maintenance payments to take account of tax consequences. He concluded (at page 2121):

I therefore conclude that, if the court takes into account the tax consequences on both the payer and the recipient of the alimony in determining the amount of the alimony to be paid for the support of the children, the parent who receives the alimony suffers no prejudice even if he or she must include those payments in his or her income. If a trial court fails to consider the tax consequences or assesses them incorrectly, the party concerned should exercise his or her right of appeal to obtain the adjustment to which he or she is entitled. Obviously, where there is an agreement between the parties, the party who receives the alimony must satisfy himself or herself that the alimony is grossed-up to a fair level, in cases where that alimony is taxable, of course. As a corollary arising from the propositions set out in the present paragraph, it is correct to say that the recipient of the alimony for the maintenance of the children is in turn entitled to take into account the tax implications following the receipt of such alimony and to withhold from the total amount of the alimony the appropriate portion of this amount which represents the additional income tax that he or she must pay by reason of the inclusion of the alimony payments in his or her income.

After proceedings in judicial review had been launched in this Court, and indeed after the date for hearing had been set, an application to intervene was made by "Support and Custody Orders for Priority Enforcement" (SCOPE), an interest group. Leave to intervene was granted and SCOPE duly filed a memorandum and voluminous materials to which reference has already been made.

The grounds of discrimination alleged by the applicant and by the intervenor are different. The intervenor claims that paragraph 56(1)(b) discriminates, and that the applicant is a member of a group that suffers discrimination, on the grounds of sex. The applicant on the other hand makes a different claim. In her notice of appeal to the Tax Court of Canada she alleges:

3. . . . 

(f) The provisions of the above-cited sections of the Act cause an injustice and an inequality before the Act affecting the taxpayer solely on the ground of her status as a "spouse" or "former spouse" or "parent", a person having the custody of children for whose benefit alimony is paid, as compared with all other persons having custody of children receiving alimony;

(g) The provisions of the above-cited sections of the Act cause an injustice and inequality of taxation affecting the taxpayer as compared with say other taxpayer whose children receive income which is not alimony . . . . [Notice of appeal, respondent's factum, at p. 5.]

I read those allegations as a claim of discrimination based on a ground analogous to those enumerated in section 15 of the Charter. The group to which the applicant claims to belong and which suffers discrimination can, I think, fairly be described as separated6*ftnote6 I use the word "separated" in these reasons as including those who are divorced. custodial parents receiving maintenance payments for their children.

It is convenient to examine these grounds separately.

The intervenor's claim of discrimination on the ground of sex requires a consideration of the recent decision of the Supreme Court of Canada in Symes.7*ftnote7 Symes v. Canada, [1993] 4 S.C.R. 695. In that case the Court had to deal with an attack on the child care allowance provided by section 63 of the Income Tax Act. The attack was based on a section 15 claim of discrimination on the grounds of sex. Iacobucci J., writing for the majority, concluded that the claim failed because it had not been shown that section 63 created a distinction on the basis of sex. That finding was made as a result of the first step of the three-stage analysis mandated for section 15 claims by Andrews8*ftnote8 Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143. and Swain9*ftnote9 R. v. Swain, [1991] 1 S.C.R. 933.

which Iacobucci J. summarized as follows (at page 761):

First, it must be determined whether s. 63 establishes an inequality: does s. 63 draw a distinction (intentionally or otherwise) between the appellant and others, based upon a personal characteristic? Second, if an inequality is found, it must be determined whether the inequality results in discrimination: does the distinction drawn by s. 63 have the effect of imposing a burden, obligation or disadvantage not imposed upon others or of withholding or limiting access to opportunities, benefits and advantages available to others? Finally, assuming that both an inequality and discrimination can be found, it must be determined whether the personal characteristic at issue constitutes either an enumerated or analogous ground for the purposes of s. 15(1) of the Charter.

Paragraph 56(1)(b) of the Income Tax Act, which is under attack here, like section 63 [as am. by S.C. 1974-75-76, c. 26, s. 33; 1976-77, c. 4, s. 21; 1980-81-82-83, c. 109, s. 19; 1984, c. 1, s. 25; c. 45, s. 22; 1988, c. 55, s. 39; 1990, c. 39, s. 13], is facially neutral: it does not, either in terms or by necessary implication, create distinctions based on sex; it is not a colourable stratagem whose provisions can, in fact, and despite the neutrality of the language, apply only to members of one sex (as would be the case, for example, in a text creating a distinction based upon being pregnant or having prostate cancer). That was not the end of the matter for Iacobucci J., however, and he went on to undertake a detailed "adverse effects" analysis of the impugned provision. Such an analysis is, of course, necessary to see if the law creates an indirect or unintentional distinction based on sex. It would also be appropriate in a case where it was alleged that the officials charged with administering the law did so in a discriminatory manner, although that was not suggested either in Symes or in the present case.

Sex was the only personal characteristic in issue in Symes and there was no claim based on any other enumerated or analogous ground.

In the course of his consideration of the first step (whether the section creates a distinction based on sex), Iacobucci J. said (at page 763):

. . . I have no doubt that women disproportionately incur the social costs of child care. Whether or not such costs are imposed by society upon women, however, is not the s. 15(1) issue. The s. 15(1) issue is whether s. 63 of the Act has an adverse effect upon women in that it unintentionally creates a distinction on the basis of sex. In my view, in order to establish such an effect, it is not sufficient for the appellant to show that women disproportionately bear the burden of child care in society. Rather, she must show that women disproportionately pay child care expenses. Only if women disproportionately pay such expenses can s. 63 have any effect at all, since s. 63's only effect is to limit the tax deduction with respect to such expenses. [Emphasis in original text.]

and again (at pages 766-767):

In another case, a different subgroup of women with a different evidentiary focus involving s. 63 might well be able to demonstrate the adverse effects required by s. 15(1). For example, although I wish to express no opinion on this point, I note that no particular effort was made in this case to establish the circumstances of single mothers. If, for example, it could be established that women are more likely than men to head single-parent households, one can imagine that an adverse effects analysis involving single mothers might well take a different course, since child care expenses would thus disproportionately fall upon women.

and again (at pages 767-768):

. . . to say that the s. 63 deduction is more often denied to women is to recognize a reality, namely, that when there are two supporting persons in a household, the woman is more often the lower income earner. Although both supporting persons contribute to the actual child care expenses incurred, the deduction will be denied to the woman as the lower income earner. In this sense, then, the woman is more often "affected" by s. 63. However, to describe s. 63 in this fashion is not to admit that s. 63 has an "adverse effect" which subordinates women. As I described above, to deny the deduction to women would only exaggerate a societal inequality if the woman in question actually paid more child care expenses. Since, as I have already indicated, proof is lacking on this point, the only obvious distinction is a parental one. And, as just noted, the appellant's focus has effectively excluded parental status arguments. [Emphasis in original text.]

Although it was not necessary for him to do so, in the light of his conclusion on the first step, Iacobucci J. also went on to examine the second step, namely, is the impugned provision discriminatory? In this connection he said (at pages 770-771):

 . . . the important thing to realize is that there is a difference between being able to point to individuals negatively affected by a provision, and being able to prove that a group or sub-group is suffering an adverse effect in law by virtue of an impugned provision. As already noted, proof of inequality is a comparative process: Andrews, supra. If a group or subgroup of women could prove the adverse effect required, the proof would come in a comparison with the relevant body of men. Accordingly, although individual men might be negatively affected by an impugned provision, those men would not belong to a group or subgroup of men able to prove the required adverse effect. In other words, only women could make the adverse effects claim, and this is entirely consistent with statements such as that found in Brooks, supra, to the effect that "only women have the capacity to become pregnant" (at p. 1242).

Looking at this point a different way, if s. 63 creates an adverse effect upon women (or a subgroup) in comparison with men (or a subgroup), the initial s. 15(1) inquiry would be satisfied: a distinction would have been found based upon the personal characteristic of sex. In the second s. 15(1) inquiry, however, the sex-based distinction could only be discriminatory with respect to either women or men, not both. The claimant would have to establish that the distinction had " the effect of imposing a burden, obligation or disadvantage not imposed upon others or of withholding or limiting access to opportunities, benefits and advantages available to others" (Swain, supra, at p. 992). The burden or benefit could not, as a logical proposition, fall upon both sexes. Likewise, to the extent that a court might undertake a broader search for "disadvantage that exists apart from and independent of the particular legal distinction being challenged" (Turpin, supra, at p. 1332), I cannot imagine how such disadvantage could be located for both men and women at the same time. [Emphasis in original text.]

These passages give me some difficulty. They are, of course, strictly speaking, obiter, but they are of great authority. They seem to suggest that a facially neutral, uncolourable legislative provision, which is impartially applied in accordance with its terms and affects both men and women, can nonetheless be discriminatory on the grounds of sex. It seems to me that such an assertion, which I, of course, accept, requires some elaboration.

Obviously there can now be no doubt that there will be many cases, both in Charter litigation and under the various human rights statutes, where an adverse effects analysis is necessary and will lead to the conclusion that there has been discrimination. It is also clear that it is no answer to such a claim to show that the impugned policy, action or legislation may also impact adversely on persons other than members of the affected group: a provision which discriminates against Hindus may likewise have an adverse effect upon Buddhists; a policy that discriminated against Blacks would not be saved by a showing that it was also harmful to aboriginal people. But surely it cannot be the case that legislation that adversely affects both men and women is discriminatory on the grounds of sex solely because the women (or men) in question are more numerous. Such a mechanistic approach would be likely to defeat the purposes of the Charter. Indeed, in my view it is not because more women than men are adversely affected, but rather because some women, no matter how small the group, are more adversely affected than the equivalent group of men, that a provision can be said to discriminate on grounds of sex.

In this connection it is important to recognize that sex differs significantly from the other enumerated grounds. There is an almost infinite number of religions, races, nationalities etc. and no two subsets within any of those categories could properly be described as opposites. There are only two sexes.10*ftnote10 I neither overlook nor discount the possibility that, in another case and other circumstances, it might be successfully argued that homosexuals or transsexuals constitute a third or even a fourth sex. That possibility has simply no relevance in the present context and even if a court were to accept such an argument, it seems to me that it would necessarily do so on the basis that each of the sexual categories so found was exclusive of all the others. One excludes the other. A male is always the opposite of a female and vice versa. Women or any group or subgroup of women who claim that a law discriminates on the basis of sex necessarily do so because it draws a distinction based on their shared characteristic of femaleness which it does not draw for those who have the opposite characteristic of maleness. It would be likewise if there were a claim of discrimination by a group or a subgroup of men.

Accordingly, it seems to me that one cannot logically say that an otherwise neutral rule discriminates on the basis of sex simply because it affects more members of one sex than of the other. Nor do I think that is what Iacobucci J. suggests in Symes when he talks of a law having a "disproportionately" adverse effect upon women or of women being "more likely" to suffer from it than men. The focus, surely, is not on numbers but on the nature of the effect; on quality rather than quantity. If legislation which adversely affects women has the same adverse effect upon men, even though their numbers may be smaller or the likelihood of their suffering be less, it cannot logically be said that the ground of discrimination is sex. This does not mean, of course, that there is no Charter breach, for in such circumstances it is very likely that another ground of discrimination will be in play.

To illustrate: it is a shameful truth that far more women in Canada suffer from poverty than men.11*ftnote11 See Moge v. Moge, [1992] 3 S.C.R. 813, at pp. 853-854.

Legislation which discriminated against the poor would therefore adversely affect more women than men. It could not be said, however, to discriminate on the grounds of sex unless it also drew a distinction against poor women which did not apply to poor men or unless it created a different effect on women than on men. Otherwise, the outcome of a section 15 attack on such legislation would turn on whether poverty was a ground analogous to those enumerated. On the other hand, legislation which imposed a physical test which could more easily be met by most men would have a disproportionate effect on women even if some men failed the test and some women, with difficulty, met it. The test would be qualitatively different even for those women who succeeded and would be vulnerable to attack as discriminatory on the grounds of sex. In each case the impact of the legislation must be weighed; it is not enough simply to count the numbers affected.

In the context of the present case I, like the Tax Court Judge, have simply no doubt that paragraph 56(1)(b) impacts adversely on more women than men. That is because mothers are far more likely to be custodial single parents than fathers. Since, however, the legislation must also impact in exactly the same way on custodial fathers, although in very much smaller numbers, I do not see how it can be said to differentiate or to discriminate on the basis of sex. In my view, the importance of the material showing the numerically disproportionate effect of paragraph 56(1)(b) on women must come in the context of a section 1 analysis.

And a section 1 analysis there must be. Just as much as the intervenor's claim of discrimination on the ground of sex seems to me to be problematical, so does the applicant's claim of discrimination on an analogous ground seem manifest.

Excluding those "non-personal" characteristics which are required by paragraph 56(1)(b) for the inclusion in a taxpayer's income of maintenance payments destined for her children, the applicant's claim to discrimination is based on her status as a separated custodial parent. When that claim is subjected to the three-step analysis mandated by Andrews, Swain and Symes, supra, the result seems to me to be starkly apparent.

First, paragraph 56(1)(b) draws an intentional distinction between the applicant and others based upon her being a separated custodial parent. There can be simply no doubt in my mind that the qualities of being separated and a parent are "personal characteristics", and I note that on several occasions in Symes, Iacobucci J. refers almost with regret to the fact that the appellant there expressly declined to assert her claim on her status as a parent. I have some doubt as to whether the fact of having the custody of a child can properly be described as a "personal characteristic", but in the light of the view I take of the other two components of the claim asserted by the applicant, it is not necessary to go further into that question. Simply put, there is no requirement that the grounds of discrimination and the definition of the affected group be coterminous; all that is necessary is that at least one of the characteristics of the latter be included within the former. Indeed, though the quality or attribute which is the ground may be one that is possessed by everybody (race, national origin) or by very few (mental or physical disability) the definition of the affected group will almost always be more narrowly focused (Haitian immigrants, blind taxpayers).

Second, the inequality created for separated custodial parents is discriminatory and imposes a burden on them not imposed on others. This may be simply tested by looking at each of the defining components of the group in turn.

Thus, a non-separated custodial parent is not required to include in income maintenance payments received from her or his spouse. This is so even in the rare, but not impossible, circumstance where the maintenance is paid pursuant to a court order or a written agreement but the spouses are living together.

Next, a separated non-custodial parent (for example a divorced father whose own parents pay him money to help with the upbringing or the education of their grandchildren) is not required to include maintenance payments in income.

Finally, separated non-parents having custody of a child, such as a grandmother or an uncle are not obliged to include in their income payments received from either or both of the child's parents.

The conclusion is inescapable: the law creates for the group as defined a burden which it does not impose on others.

The third and last stage of the section 15 analysis is to inquire whether the personal characteristics at issue constitute grounds analogous to those enumerated. In my view, they do.

The concept of the "analogous ground" as forming a basis for discrimination was adopted by McIntyre J. in Andrews, supra, where he said (at page 175):

The enumerated grounds in s. 15(1) are not exclusive and the limits, if any, on grounds for discrimination which may be established in future cases await definition. The enumerated grounds do, however, reflect the most common and probably the most socially destructive and historically practised bases of discrimination and must, in the words of s. 15(1), receive particular attention. Both the enumerated grounds themselves and other possible grounds of discrimination recognized under s. 15(1) must be interpreted in a broad and generous manner, reflecting the fact that they are constitutional provisions not easily repealed or amended but intended to provide a "continuing framework for the legitimate exercise of governmental power" and, at the same time, for "the unremitting protection" of equality rights. . . . 

He then went on (at page 182) to say that the " enumerated and analogous grounds' approach most closely accords with the purposes of s. 15" and to conclude (at page 183) by incorporating the American concept of a "discrete and insular minority".

The Andrews decision was expanded and commented upon by Wilson J. in R. v. Turpin12*ftnote12 [1989] 1 S.C.R. 1296. as follows (at page 1332):

McIntyre J. recognized in Andrews, that the " enumerated and analogous grounds' approach most closely accords with the purposes of s. 15 and the definition of discrimination outlined above" (p. 182) and suggested that the alleged victims of discrimination in Andrews, i.e., non-citizens permanently resident in Canada were "a good example of a discrete and insular minority' who came within the protection of s. 15" (p. 183). Similarly, I suggested in my reasons in Andrews that the determination of whether a group falls into an analogous category to those specifically enumerated in s. 15 is "not to be made only in the context of the law which is subject to challenge but rather in the context of the place of the group in the entire social, political and legal fabric of our society" (p. 152). If the larger context is not examined, the s. 15 analysis may become a mechanical and sterile categorization process conducted entirely within the four corners of the impugned legislation. A determination as to whether or not discrimination is taking place, if based exclusively on an analysis of the law under challenge is likely, in my view, to result in the same kind of circularity which characterized the similarly situated similarly treated test clearly rejected by this Court in Andrews.

Immediately following the passage from her reasons in Andrews which Wilson J. refers to in the above-quoted extract from Turpin, she made the following additional important observation (at pages 152-153):

I believe also that it is important to note that the range of discrete and insular minorities has changed and will continue to change with changing political and social circumstances. For example, Stone J. writing in 1938, was concerned with religious, national and racial minorities. In enumerating the specific grounds in s. 15, the framers of the Charter embraced these concerns in 1982 but also addressed themselves to the difficulties experienced by the disadvantaged on the grounds of ethnic origin, colour, sex, age and physical and mental disability. It can be anticipated that the discrete and insular minorities of tomorrow will include groups not recognized as such today. It is consistent with the constitutional status of s. 15 that it be interpreted with sufficient flexibility to ensure the "unremitting protection" of equality rights in the years to come.

As I understand these authorities they do not suggest timidity in the approach to the grounds of discrimination. Rather they require an examination of any alleged analogous ground in the broadest context with a view to determining whether such ground has historically served or today actually serves to single out or distinguish a disadvantaged group. This is a different analysis from that which requires a determination of whether an impugned provision draws a distinction or discriminates on the alleged ground. In particular, and while I have gone to some pains to indicate that I do not think that a mere numerical imbalance or disproportion in the effect of legislation will be enough to establish a distinction or discrimination based on the particular ground of sex, it seems to me that the opposite is the case when the question is to know whether some other ground is analogous to those enumerated in section 15. If one were to suppose, for example, that race had been left out of the enumeration, one would not, in determining whether it was "analogous", ask whether race was always a ground of discrimination. Clearly, it is not any more than are the other enumerated grounds. What would be important would be to establish that race frequently serves as a basis for the kind of prejudicial, stereotypical, mindless categorization that is the hallmark of discrimination. That is what makes all the enumerated grounds analogous to one another.

To put the matter another way, when determining whether or not a ground is analogous to those enumerated in section 15 it is legitimate to look to see if it has frequently served as a ground for discrimination in other circumstances but it is clearly not necessary that it should always do so: if "race" was not enumerated we would not have to show that humankind invariably discriminates against those of different races in order to establish that it is an analogous ground. By contrast, where the question is to know whether a law draws a distinction, imposes a burden or confers an advantage on the ground of sex the causal connection between the ground and the distinction, burden or advantage can only be established if none of the members of the opposite sex would feel those effects in the same way, assuming the law were properly applied.

This highlights a further important point to which I have already alluded: different considerations come into play where it is alleged (as more frequently happens in human rights claims) that a law or policy, whether or not it is discriminatory on its face, is applied in a discriminatory way; an uneven and unfair application on the ground of sex, even though it might affect some who were not members of the target group, would, of course, still be discriminatory. In such a case it is not so much the law which is attacked as the manner of its implementation. There is no suggestion here, however, that the Income Tax Act is applied otherwise than strictly in accordance with its terms.

I have identified the group to which the applicant claims to belong as separated custodial parents. Neither that phrase nor any of its components constitute as such a ground analogous to those enumerated in section 15 but that is hardly surprising. As previously indicated, the definition of the group must have within it some component which is included within the alleged ground but it is unlikely to be the ground itself. An individual or group does not suffer discrimination because they are "race" or "religion" but rather because they belong to a particular race or practice a particular religion. Or, approaching the matter from the other end, a Haitian or a Hindu who suffers discrimination does so not on the ground of being Haitian or Hindu but of race or religion.

The appropriate description of the ground of discrimination to which separated custodial parents are subject would, it seems to me, be "family status". I consider it to be almost self-evident that such status has historically been, and is still, used as a basis for stereotyping. One has only to call to mind such traditional expressions as the "happily married man", the "old maid", the "gay13*ftnote13 In the traditional sense of that word. bachelor" or the "merry widow" to find examples. Even the law is not above such stereotypes and the civilian counterpart of the reasonable man (itself a stereotype) is the "bon père de famille". And, of course, it is only in comparatively recent times that the shocking disabilities imposed by both common law and civil law on married women have been done away with; since such disabilities did not apply to single women they were not based only on sex but on family status as well.

The fact that family status or some similar expression figures as a prohibited ground of discrimination in most human rights statutes also serves to confirm its analogous nature to the grounds enumerated in the Charter.

Finally, the group to which the applicant belongs and which claims discrimination on the ground of family status, separated custodial parents, can readily be seen as a discrete and insular minority which has historically suffered prejudice and has need of protection.

Before concluding on this aspect of the matter it is necessary to say a few words about the reasons for which the Tax Court Judge dismissed the applicant's claim. As indicated, he was of the view that any adverse tax effects of the inclusion/deduction system on the applicant could and should have been corrected by the ordinary operation of the family law system.

While there can be no doubt that the Judge was right in his view that courts should take account of income tax consequences and that the court which granted the applicant's divorce did in fact do so, I think, with respect that he was wrong to see that as an answer to a Charter-based attack on paragraph 56(1)(b). My reasons are twofold.

In the first place, there is substantial material in this record to indicate that in practice the family law system does not always, or even usually, operate to correct the inequality created by paragraph 56(1)(b). In the second place, even if it did, it seems to me that it is simply not legitimate to look outside the income tax system to correct an injustice which that system has itself created.

As to the first point, I start from the proposition that the Income Tax Act and the family law system are almost polar opposites when it comes to their approach to the determination of sums of money. Income tax seeks to be precise, virtually to the last penny, and income tax liability is established by a code of rules whose detail and complexity is legendary. The awarding of child support on the other hand, somewhat like the evaluation of damages, is notoriously imprecise. There are a number of guiding principles such as the welfare of the children and the obligation of both parents to contribute in accordance with their means, but within those principles there is a very large range of judicial discretion. As one who has had the duty of fixing child support, I know that it never amounts to a simple mathematical calculation. In those circumstances, it is virtually impossible, in my view, to rely upon the family law system as a corrective to any injustice created by the inclusion/deduction system. It is simply too blunt an instrument for the job.

Furthermore, the evidence available in this record is strongly suggestive of a failure of the family law system to in fact correct the injustice created by paragraph 56(1)(b). In the Report of the Federal/ Provincial/Territorial Family Law Committee, The Financial Implications of Child Support Guidelines: Research Report, (May 1992), one finds the following extracts (at page 90):

Tax consequences should be considered by everyone involved in determining child support awards whether they are the result of agreements between the parties, negotiations between lawyers or determined by the courts. Although the tax consequences should be an element of every child support determination, there is evidence to suggest that these calculations may not always be made. (A survey of judges was conducted in February 1990 by Judge R. James Williams. 147 judges responded to this survey. The results indicate that only a minority of lawyers present income tax calculations to the courts where it would be appropriate to do so. Furthermore the survey shows that a majority of judges will not do their own tax calculations if they are not presented by counsel.

Judge R. James Williams, Child Support, An Update and Revision of Quantification of Child Support, (1989) 18 R.F.L. 234), May 2, 1990.) Moreover, where the tax consequences are considered, there are no formal guidelines showing how they should be taken into consideration.

and again (at pages 91-92):

Negotiations and settlements occur in the majority of support determinations in Canada. In cases where support awards have been negotiated and agreed upon by the parties, it is impossible to determine how or whether the income tax implications have been taken into account. Whether income tax implications have been taken into account or not support payors are allowed to deduct the total support amount paid.

If tax implications are to be taken into account there are a number of issues to consider. There is no guidance in the Divorce Act, nor provincial or territorial legislation as to how the calculations should be made or how the benefit of the deduction should be shared between the parties.

In most cases, where the court has arrived at a specific amount for the child's needs, that amount will then be divided in proportion to each parent's income. Then, the non-custodial parent's share is increased to compensate the custodial parent for the tax that he or she will have to pay. The support payer would then deduct the total amount paid from his or her income, which may represent a greater tax benefit than the amount he actually paid to cover the tax which the custodial parent would have to pay. Where the tax implications have been dealt with in this way, the tax subsidy is often not passed on to the custodial parent for the children.

Where it is obvious that the parties do not have the means to cover the child's needs, the benefit of the deduction to the non-custodial parent may be raised as an incentive to slightly increase the amount of child support. However, there is no consensus or guidance as to how this deduction should be shared between the parties.

In order to ensure that the tax benefits are taken into account fairly, two extra steps would have to be performed in each child support determination: (1) the benefit of the deduction to the non-custodial parent would have to be determined and (2) the benefit should be divided between the parties. However, even if these two steps were applied, there remains the problem of how the benefit should be allocated between the parties. Arguments could be made that the benefit should be totally allocated to the child or that it should be divided equally between the parties or in proportion to both parents' share of the child's costs.

For custodial parents who are taxable, the responsibility of including the amount of support received within their income is another issue which deserves to be addressed. For custodial parents, saving the portion of the support payment that must be paid in income tax at the end of the year can be a significant burden. Although some may argue that this burden should not be recognized, custodial parents should be saving the portion of the award that applies to tax. In some cases, it is not financially possible to save this money each month without depriving the children. It is important to consider this issue in the context in which it actually occurs; namely that two-thirds of Canadian women and children live in poverty following divorce. However, in certain cases, the Child Tax Credit and Goods and Services Tax Credit could offset the taxes that are payable.

Another important issue is that the tax implications of a support award may vary with the passage of time. Ideally, child support orders should be varied to reflect the parties' changing marginal rates as such changes occur. This would ensure that the parties continue to be responsible for the same proportion of the children's needs throughout the years. At present, the legal costs of seeking a variation of support orders could serve as a deterrent for such applications.

By the same token, in a paper entitled Child Support Policy: Income Tax Treatment and Child Support Guidelines, by Ellen B. Zweibel and Richard Shillington, following an analysis of the impact of the alleged "tax subsidy" given by the inclusion/deduction system, the authors comment (at page 17):

What about the 49% of cases in Table 2.3 (Evaluation awards) and the 48% of cases in Table 2.5 (Melson/Delaware awards) where the deduction/inclusion provisions did not produce a subsidy? How do family lawyers and judges respond to the conundrum presented by the Income Tax Act's effect on these households?

Family law determines child support based on the children's needs and the parent's relative abilities to meet those needs. In some cases, the child support is then adjusted for income tax. When the custodial mother's tax liability exceeds the father's tax savings, the tax adjustment becomes more problematic and less likely. The father's tax savings can no longer be used to persuade him to pay a fully grossed-up award. If the father indemnifies the mother for her tax liability, the effect on his disposable income is greater than he anticipated and arguably greater than he originally agreed to. But, if the support payment is not fully grossed up, then the effective value of the child support payment is considerably diminished. The custodial mother receives less child support than she originally anticipated and is left to make up any shortfall. The deduction/inclusion provisions do not work for these cases and the Finance Department's primary rationale supporting the current tax regime ignores them.

Under the right circumstances, the deduction/inclusion provisions can provide a beneficial subsidy to separated and divorced families. The payor's tax savings must exceed the recipient's tax liability. The payor and recipient must have a common goal of increasing the support available for the children and they must be assisted by accountants and lawyers.

However, circumstances are not always right. This research shows the real possibility that little or no subsidy is available to a sizeable number of separated and divorced families. The current policy also ignores the reality that child support is a contentious issue and that non-custodial fathers seeking to minimize their payments may not readily agree to either a gross-up or to a further sharing of any tax savings above the gross-up. The Finance Department's rationale also ignores the number of persons who settle their child support arrangements on their own, without the assistance of lawyers or accountants, the number of lawyers and judges who rely on rough estimates and the number of cases where, despite the custodial mother's lawyer's careful tax calculations, the "glass ceiling" moves in to reduce the award. [Emphasis in original text.]

These materials, in my view, greatly weaken any argument that the family law system can be relied upon to correct problems created by paragraph 56(1)(b).

My second objection to the position taken by the Tax Court Judge is more fundamental. It relates in effect to causation. Both in theory and in fact, income tax comes after income. If there is no income there is no tax. Thus, when the Income Tax Act creates a charge it necessarily does so on the basis that there is a source of income upon which that charge can operate. If the charge itself offends the equality provisions of the Charter, however, it is logically indefensible to claim that there is no breach because the source can adjust to correct the inequality. In concrete terms, if the family law system requires a gross-up of maintenance for income tax, that can only be because the Income Tax Act taxes maintenance payments made to the custodial single parent. If that tax is itself in breach of section 15 of the Charter, calling the family law system in aid is putting the cart before the horse.

Accordingly, I conclude that paragraph 56(1)(b) offends the rights of single custodial parents to equality before and under the law and to equal benefit of the law and that it can only apply to the applicant if it can be saved by section 1.

The various steps of a section 1 analysis are well known and have been many times restated since the landmark decision of the Supreme Court of Canada in R. v. Oakes.14*ftnote14 [1986] 1 S.C.R. 103.

The first stage requires an assessment of the objectives sought to be achieved by the impugned provision and a determination of whether they are important enough to warrant a Charter breach.

The objectives of the inclusion/deduction system are described as follows by Beetz J. in Gagnon v. The Queen15*ftnote15 [1986] 1 S.C.R. 264. (at page 268):

Thus, the amount deductible by the taxpayer under s. 60(b) is taxable in the hands of the recipient under s. 56(1)(b).

The purpose of these provisions, by allowing income splitting between former spouses or separated spouses, is to distribute the tax burden between them. As C. Dawe wrote in an article titled "Section 60(b) of the Income Tax Act: An Analysis and Some Proposals for Reform" (1979), 5 Queen's L.J. 153:

This allows the spouses greater financial resources than when living together, compensating in part for the lost economics of maintaining a single household.

Likewise, in the Report of the Federal/ Provincial/Territorial Family Law Committee, supra, the following appears (at pages 84-85):

There are a number of policy considerations for the deduction of support payments by the payer and inclusion of these amounts as taxable income to the recipient. The rationale for this policy has been provided by the Department of Finance. First, it is a principle of taxation that, where a deduction has been claimed by a payor in respect of a payment, the recipient of that payment should pay income tax on it. Second, by requiring support recipients to include the amount of child support within their income, the system recognizes the basic principle of fairness that tax payers with the same incomes from different sources should pay the same amount of tax. Third, the tax assistance offered by the deduction may provide an incentive for the payer to make regular and complete payments. (Despite this policy, Ontario recently announced that 75 per cent of support payers were in some default in the payment of support orders.) Fourth, the tax treatment provides a subsidy which benefits the children since it encourages higher support payments.

When the policy was developed, the majority of support payors were in a higher tax bracket than the recipients. Since a dollar of support payment gives rise to a greater tax reduction to the payer than tax increase to the recipient, the deduction lowers the combined federal plus provincial taxes of the separated couple and may provide an incentive for the payer to increase the support payments. However, since the number of tax brackets have been reduced, support creditors and debtors are not necessarily in different tax brackets although one may earn more than the other.

According to 1988 taxation statistics provided by Revenue Canada, about 60 per cent of support payers were in the top two tax brackets prior to the deduction of support payments. Almost 90 per cent of recipients were either non-taxable or in the lowest tax bracket. After the deduction/inclusion of child support, 50 per cent of payers remain in the top two brackets, while 80 per cent of recipients remain non-taxable or in the lowest tax bracket. Again, these figures demonstrate the great disparity of income between support recipients (who are mostly women with children) (according to the Divorce Act Evaluation, women were awarded sole custody of the children in three-quarters of the cases, and in 98 per cent of the cases, the direction of support is from the father to the mother) and support payers (who are mostly men). This evidence also suggests that, in the large majority of cases, there may be some need for sharing the tax savings from the deduction of support payments with the children.

Despite the strong indication in the last two paragraphs of the quoted extract that the "tax subsidy" objective (clearly the most important) is not being achieved (as to which I shall have more to say presently), I have little difficulty in concluding that such objective meets the required level of importance.

Likewise, the next step, which requires the Court to be satisfied that there is a rational connection between the impugned provision and the legislative objective, presents little difficulty. While it might, in some cases, be argued that the ineffectiveness of a provision is so great as to cast doubt on the rational connection between it and its objective, I do not think that can be said to be the case here.

The last two stages of the section 1 analysis, minimum impairment and proportionality, present far more difficulty, however. The last two paragraphs of the Federal/Provincial/Territorial Family Law Committee Report, just quoted, hint at the manner in which the inclusion/deduction system operates in an unduly intrusive and disproportionate manner in some cases.

By the same token, the passage from Zweibel and Shillington, quoted earlier, which is based on an analysis by those authors of data produced by the Government itself,16*ftnote16 Canada, Department of Justice, Bureau of Review, Evaluation of the Divorce ActPhase II: Monitoring and Evaluation (Ottawa, Queen's Printer, 1990). suggest that the inclusion/deduction system only produces a tax subsidy in slightly over 50 per cent of the cases.

The same authors, after examining other forms of "tax recognition" granted to single custodial parents such as tax credits, conclude as follows (at pages 22-23):

2.5.2.1 Custodial mothers receive greater tax recognition than the non-custodial fathers primarily when they qualify for the refundable child tax credit. Otherwise the non-custodial father is favoured.

The custodial mother received greater tax recognition of the cost of raising children than the non-custodial father in 56% of the cases (95 of 171 cases). However, 97% of these cases (92 of 95 cases) involved custodial mother's with taxable income (pre-child support) in the lowest tax bracket and 90% (85 of 95 cases) had incomes under $20,000. Thus, the custodial mother's tax recognition exceeded the non-custodial father's primarily when her income was low enough to qualify for the full refundable child tax credit.

It is important to note that not all custodial mothers in the lowest income tax bracket were favoured. Approximately one third of these mothers (45 of 137) had less tax recognition than the non-custodial fathers from whom they received payments.

2.5.2.2 There is a significant disadvantage for custodial mothers in the middle income tax bracket.

The picture changes dramatically once the custodial mother's income rises above the first income tax bracket. In 91% of these cases (31 of 34 cases), her tax recognition is less than the non-custodial father's. There were 38 cases (22% of total cases in the data base) where the custodial mother effectively received no recognition of the cost of raising children because her tax liability on the support payment exceeded her child related tax credits. These cases were concentrated in the middle income tax bracket where 65% of the custodial mothers were in this position.

2.5.2.3 The tax system contributes to custodial mothers downward trend in standard of living.

A snapshot of the relative standard of living of the non- custodial and custodial households, both in relationship to each other and to the family's pre-separation standard of living is provided in Table 2.7. The custodial mother's household standard of living always declines relative to the family's pre-separation standard of living. In stark contrast, 57% of the non-custodial fathers maintain or improve their standard of living.

The Justice Department Evaluation17*ftnote17 Supra, footnote 16. itself talks in similar terms (at page 132):

It is beyond the scope of the present evaluation to enter into the debate about the most appropriate model for apportioning the costs or raising children. The data suggest, however, that as in Phase I, the burden falls disproportionately on the custodial parent who is, usually, but not exclusively, the woman. Some indications of this are that, except in the very lowest income groups, men were ordered to pay about 18 percent of their gross income in support or about $250.00 per child per month. At the same time, the bare minimum cost of raising a preschool-child in a single parent family has recently been estimated at $350.00 without child care and $790.00 per month with child care. While some proportion of this cost will be borne by the custodial parent who will generally earn less than the non-custodial parent, the more telling statistics are that most men without custody have incomes after paying support which leave them considerably above the poverty line while a majority of women with custody of the children have, after receiving support, incomes putting them below the poverty lines for various family sizes.

The Government takes issue with the figures produced by Zweibel and Shillington, primarily on the basis that the data base used by them (which, as noted, was published by the Government itself) is too small and is also unreliable. The affidavit of Nathalie Martel18*ftnote18 Respondent's supplementary record, at p. 147.

produced by the respondent proposes use of a larger and more up-to-date data base also generated by the Department of Justice: "Child Support Award Database". Based on this, Ms. Martel develops her own figures to show how many couples enjoy a net tax gain and how many suffer a net tax loss as a result of the application of the present system. She divides the couples according to income of the custodial parent into income ranges of $10,000 (a classification different from that used in the Income Tax Act to establish marginal rates and relied on by Zweibel & Shillington). The results follow.

Only in the $0"$10,000 and $10"$20,000 ranges are the number of winners (90 and 279 respectively) greater than the number of losers (0 and 62). In the other income categories the majority are always net losers from an income tax perspective (i.e. there is a greater tax burden as a result of the inclusion/deduction system). These are the figures:

Income range Winners Losers

$20"$30,000 42 105

$30"$40,000 2 71

$40"$50,000 6 18

$50,000 + " 7

Overall, in 59% of the cases the parents pay less tax on the maintenance payments than they would if there was no inclusion/deduction requirement (winners); 37% of parents suffer a tax disadvantage under the present regime (losers) and in 4% of cases the result is neutral.

Because Ms. Martel was of the view that the figures reported in the data base which she had chosen were possibly skewed by alleged under-reporting (a proposition which in my view has not been demonstrated) she has massaged them so as to make them conform to the general profile of Canadian taxpayers. Assuming, for argument's sake only, that this is a legitimate exercise, the results are not very much more favourable to the Government's case:

Income range Winners Losers

$0 "$10,000 90 "

$10"$20,000 305 38

$20"$30,000 66 82

$30"$40,000 6 65

$40"$50,000 8 16

$50,000 + 1 6

Again the winners outnumber the losers only where the income of the custodial parent is in the two lowest ranges. Amongst such persons there must be a substantial number who, because of their very low incomes, pay no tax at all and are therefore bound to be net "winners" from the inclusion of maintenance payments in an income which attracts no tax. In all the other income ranges the losers outnumber the winners.

The respective percentages under the revised figures are 67% winners, 29% losers and 4% neutral.

Under either calculation, and even when a couple "wins", i.e. jointly pay less tax than they would without the inclusion/deduction system, the Income Tax Act imposes no requirement that the tax saving or "subsidy" thus created will be passed on to the children who are the intended beneficiaries. And, as we have seen, there is good reason to doubt that this actually takes place in many cases and that the family law system effectively corrects the problem. On the other hand, of course, the non-custodial parent is virtually always a winner by reason of having his taxable income reduced. Only in the very unusual circumstances where a) the payments are fully and accurately grossed up for tax consequences and b) the custodial recipient is in a higher marginal tax bracket than the non-custodial payer, would the latter end up being disadvantaged by the present system. While such a case may exist I have never heard of it.

The upshot of all the foregoing is that even on the Government's own figures the inclusion/deduction system, whose alleged purpose is to benefit single custodial parents and their children, cannot do so in at least one third of the cases. There is no guarantee that it actually does so in the remaining two thirds of the cases and there is evidence to suggest that it does not.

Since the burden of proving section 1 justification lies upon the Government, it is my view that this material falls lamentably short of the mark. Indeed, far from being a measured, proportionate response to a perceived problem, the inclusion/deduction system frequently fails to give any benefit at all to those whom it is allegedly designed to assist, almost always benefits those who do not need assistance, and contains no corrective or control mechanisms designed to remedy the problem. The system does not meet the minimum impact and proportionality tests and the section 1 justification fails.

Before concluding, there are a number of relatively minor and ancilliary points which require to be mentioned.

First, there was some suggestion by the respondent that the tax disadvantage suffered by separated custodial parents was offset in other parts of the Income Tax Act. Notably, reference was made to the equivalent-to-married credit (paragraph 118(1)(b) [as am. by S.C. 1988, c. 55, s. 92]), the dependant tax credit (paragraph 118(1)(d) [as am. idem]) and the child tax credit (section 122.2 [as enacted by S.C. 1978-79, c. 5, s. 4; as am. by S.C. 1984, c. 1, s. 65; 1988, c. 55, s. 97]). Those credits were fully and adequately dealt with by the Tax Court Judge who concluded that each of them was independent of the inclusion/deduction system and therefore could not be said to be a corrective to it. In any event, the aggregate of all such credits in 1989 was less than the amount of the tax loss actually suffered by the applicant as a result of the inclusion/deduction system.

Second, counsel for the respondent suggested that any finding that paragraph 56(1)(b) was invalid or inoperative by reason of section 15 of the Charter would necessarily carry with it a finding that the "deduction" counterpart in paragraph 60(b) must also fail and she asked us to so find. I am not prepared to accede to that request. In the first place, paragraph 60(b) has not been the focus of this litigation and it is not impossible that other considerations, or indeed other evidence, could result in a court finding that the paragraph could survive a Charter attack. Second, and more important, paragraph 60(b) plays no role in the calculation of the applicant's income or her assessment. Since it is not directly in issue in these proceedings it would be most improper for us to make any finding with regard to it.

Next, respondent's counsel asked that, in the event that we should allow the application, we delay any declaration of invalidity of paragraph 56(1)(b) for a period of time to allow the Government to introduce the necessary amendments to the Act. There are several reasons why this cannot be done. At a technical level, these proceedings take the form of an application for judicial review of an income tax appeal. In such a proceeding our powers as a reviewing Court are limited.19*ftnote19 See s. 18.1(3) of the Federal Court Act, R.S.C., 1985, c. F-7 [as enacted by S.C. 1990, c. 8, s. 5].

In particular, we can only ultimately order that the Tax Court dispose of the applicant's appeal in a way that that Court itself might have done originally; subsection 171(1) [as am. by S.C. 1980-81-82-83, c. 158, s. 58, item 2] of the Income Tax Act limits the power of the Tax Court to allowing or dismissing the appeal. On a more fundamental level, we are dealing here with the rights of individuals which are guaranteed to them by the supreme law of the country. It would take very strong reasons indeed to justify any suspension of those rights. None has been suggested. Lastly, and as a practical matter, the chances of this Court's having the last word on the subject are rather less than those of winning first prize in the lottery.

Finally, I would note the somewhat ironic fact that the applicant herself may well not emerge from this litigation as a winner. As previously indicated, the Judge who granted her decree of divorce took account, even if only in rather approximate terms, of the impact of income tax on both the applicant and her former husband. That being so, one may anticipate that the latter will now apply to have the payments varied to take account of the fact that the applicant will receive them free of tax.

I would allow the application, I would set aside the decision of the Tax Court and I would remit the matter to the Tax Court for redetermination on the basis that paragraph 56(1)(b) of the Income Tax Act violates the applicant's rights under section 15 of the Charter.

Pratte J.A.: I agree.

* * *

The following is the English version of the reasons for judgment rendered by

Létourneau J.A.: I have had the benefit of reading the opinion of my colleague Hugessen J.A., and I agree with him, for the reasons he gives, that there is no discrimination on the basis of sex in the present case. However, I believe that the applicant, contrary to what she contends, has not been subject to discrimination because of her civil status or social condition.

I agree with my colleague that paragraph 56(1)(b) of the Act imposes a burden on the applicant and on the class of persons to which she belongs which it does not impose on other groups with different civil status.20*ftnote20 My colleague refers to the applicant's "family status". This concept is a component of the broader concept of "civil status" which is recognized as a prohibited ground of discrimination in s. 10 of the Quebec Charter of Human Rights and Freedoms, R.S.Q. 1977, c. C-12 [as am. by S.Q. 1982, c. 61, s. 3]. See Brossard (Town) v. Quebec (Commission des droits de la personne), [1988] 2 S.C.R. 279, in which the Court included family relationships and filiation in the concept of "civil status". See also the Dictionnaire de droit privé et Lexiques bilingues, 2nd ed., Cowansville (Que.): Editions Yvon Blais Inc., 1991, at p. 233, where it is stated that the main element of the civil status of natural persons is status in the family, the point at which it is known whether the person is married or not; see the definition of civil status in the Petit Larousse illustré, Librairie Larousse, Paris, 1984, at p. 386, where reference is made to [translation] "the status of individuals in respect of family relationships, birth, marriage, death and so on"; G. Cornu, Vocabulaire juridique, 2nd ed., Presses Universitaires de France, Paris, 1990, at p. 327.

However, I cannot share his view that the difference in treatment that it creates is discriminatory on that basis.

First, the Income Tax Act sets up a whole set of distinctions and differences in treatment which take into account the economic reality that is a result of the taxpayer's civil status. This is the case, for example, and this is not an exhaustive list, for the married tax credit (paragraph 118(1)(a) [as am. by S.C. 1988, c. 55, s. 92]), the equivalent-to-married tax credit for a wholly dependent person (paragraph 118(1)(b)), the tax credit for children or dependants (paragraph 118(1)(d)), the rules for attribution of income between related persons (subsection 74.1(2) [as am. by S.C. 1986, c. 55, s. 17] and section 251 [as am. by S.C. 1980-81-82-83, c. 140, s. 129; 1988, c. 55, s. 190]), the transfer or attribution of property to a surviving spouse (subsection 70(6) [as am. by S.C. 1973-74, c. 14, s. 19; 1974-75-76, c. 26, s. 38; 1985, c. 45, s. 33]), the transfer of property accumulated in a registered retirement savings plan to the registered retirement savings plan owned by the spouse (paragraph 146(16)(b) [as am. by S.C. 1990, c. 35, s. 13]), child care expenses (section 63), the inter vivos transfer of property to a spouse (section 73 [as am. by S.C. 1973-74, c. 14, s. 20.1; 1977-78, c. 32, s. 14; 1980-81-82-83, c. 48, s. 39; c. 140, s. 41; 1985, c. 45, s. 35; 1986, c. 6, s. 36; 1988, c. 55, s. 51]) and the transfer of certain unused tax credits to a spouse (section 118.8 [as enacted by S.C. 1988, c. 55, s. 92]).

It may be unfortunate that the Act uses civil status, a category that is laden with discriminatory potential, to identify the various groups that are subject to different taxation. Because that taxation is based on varying income and capacity to pay, it could also, undoubtedly, be collected from those same groups by using other more neutral designations, so that the different treatment would then not be unfairly coloured with a discriminatory tint by the category used.

In my opinion, it is not necessarily discriminatory per se to treat different civil statuses differently. For example, it goes without saying that it costs more for a divorced couple to live under two roofs than to live together under the same roof. On the contrary, it might be discriminatory to treat different civil statuses identically, when they create different duties and responsibilities for the people in question, and which consequently demand different treatment in social, political, economic and legal terms. In R. v. Turpin, and later in Symes v. Canada,21*ftnote21 [1993] 4 S.C.R. 695, at pp. 756-757. the Supreme Court recalled, as follows, that we must examine the general context surrounding the Act, to determine whether there is discrimination:

In determining whether there is discrimination on grounds relating to the personal characteristics of the individual or group, it is important to look not only at the impugned legislation which has created a distinction that violates the right to equality but also to the larger social, political and legal context. McIntyre J. emphasized in Andrews (at p. 167):

For, as has been said, a bad law will not be saved merely because it operates equally upon those to whom it has application. Nor will a law necessarily be bad because it makes distinctions.

Accordingly, it is only by examining the larger context that a court can determine whether differential treatment results in inequality or whether, contrariwise, it would be identical treatment which would in the particular context result in inequality or foster disadvantage. A finding that there is discrimination will, I think, in most but perhaps not all cases, necessarily entail a search for disadvantage that exists apart from and independent of the particular legal distinction being challenged.

Similarly, I suggested in my reasons in Andrews that the determination of whether a group falls into an analogous category to those specifically enumerated in s. 15 is "not to be made only in the context of the law which is subject to challenge but rather in the context of the place of the group in the entire social, political and legal fabric of our society" (p. 152). If the larger context is not examined, the s. 15 analysis may become a mechanical and sterile categorization process conducted entirely within the four corners of the impugned legislation. A determination as to whether or not discrimination is taking place, if based exclusively on an analysis of the law under challenge is likely, in my view, to result in the same kind of circularity which characterized the similarly situated similarly treated test clearly rejected by this Court in Andrews.22*ftnote22 [1989] 1 S.C.R. 1296, at pp. 1331-1332.

[Emphasis added.]

To undertake only a purely textual analysis of the provisions of the Income Tax Act, which establishes a distinction that takes civil status into account, and then conclude that there is discrimination amounts to ignoring the social, political, legal and economic reality which this Act and its provisions inhabit, and which are experienced differently by taxpayers whose family situations differ. In my view, the difference in treatment created by paragraph 56(1)(b) does not create the inequality to which the Supreme Court referred in Turpin, and accordingly it is not discriminatory.

The Income Tax Act is essentially economic legislation, which may even be described as amoral,23*ftnote23 Even income received from sources such as prostitution or the unlawful sale of narcotics is taxed. its purpose being to trace income and tax it on the basis of the social and economic needs of the community, taking into account the reality of the taxpayer's economic situation, which may vary depending on whether or not he or she is married or supporting a family. To ignore this economic context, the reality that underlies it and the importance that the Government must necessarily place on it would mean that the numerous provisions of this Act which set up a distinction and impose different burdens based on different economic realities, because different civil statuses produce different needs, would be prima facie discriminatory.

For all practical purposes, the applicant's argument amounts to saying that the Income Tax Act is discriminatory, in its essence and its structure, because it takes into account the economic reality that results from, inter alia, different civil statuses, and that accordingly its constitutional validity must be justified under section 1 of the Charter. I cannot accept this argument precisely because, first, the difference in treatment accorded does not create inequality since Parliament treats different situations differently. Second, the difference is not based on civil status, and accordingly on a ground that is analogous to those listed in section 15, but rather results from the different physical constraints and duties that fall upon taxpayers who live in different economic situations because of their different personal statuses.

Moreover, we must take into account the actual nature of the impugned legislative provision. Before the amendments made by paragraph 56(1)(b), the manner in which the Act dealt with married and divorced couples was identical, ignoring the different economic circumstances in which these people lived and thereby aggravating their social and economic situations. Paragraph 56(1)(b), which is in issue here, is intended precisely as a remedy for the disadvantages that this group of taxpayers, to which the applicant belongs, had suffered at one time. In addition, married taxpayers complain of the favourable treatment accorded to the applicant and to divorced spouses since, unlike them, they cannot split their income and reduce their tax.24*ftnote24 See Foucher c. Québec (Procureur général), [1989] 1 R.J.Q. 703 (S.C.), in which the applicant, who was married and a father, was seeking leave to bring a class action on the ground that the Income Tax Act treats married people differently from separated or divorced people. The remedial measure necessarily creates a distinction by taking into account these people's civil status, since this is the group it is addressing, and this is the group that is living in a different and difficult economic situation as a result of the breakdown of the family unit. This distinction does not necessarily constitute discrimination. When read and taken literally in isolation, the measure may appear discriminatory, but it is not when it is placed in its socio-economic and socio-political context and the goal in mind is taken into account.

This action, which arises out of paragraph 56(1)(b), questions the constitutional validity of a legislative measure designed to remedy the disadvantaged situation in which the applicant's group was placed before. It also questions the standard of effectiveness that such a measure must meet if it is to be constitutionally valid.

In the case at bar, the beneficiaries of the deduction/inclusion system are in very large part women, and a majority of them, in particular women of low-income, receive an advantage and a benefit from the income-splitting permitted by the system.25*ftnote25 Evidence was given of a benefit of 275 million dollars created by the deduction/inclusion system in 1990 and borne by the federal and provincial governments. See the affidavit of Nathalie Martel, at p. 148 of the respondent's supplementary record. Paragraph 56(1)(b) of the Act therefore produces a beneficial effect in a majority of cases and helps to partially remedy the inequalities suffered by this group of taxpayers in the past.26*ftnote26 The evidence submitted by the applicant and the respondent points out a disagreement in respect of how effective the system is. While the applicant, referring to the study by Zweibel and Shillington, Child Support Policy: Income Tax Treatment and Child Support Guidelines, The Policy Research Centre on Children, Youth and Families, at pp. 5, 14, 15, 17 and 24, submits that the end result in 49% of the cases studied was an adverse tax impact, the respondent produced the results of an analysis prepared for the Department of Justice by Ms. Martel, an economist with the Personal Income Tax Division of the Department of Finance Canada. She concludes, after adjusting her study, that the effect of the system is adverse in 29% of cases and neutral in 4%. That analysis was conducted using a new and expanded data bank which is more reliable than the one used by Zweibel and Shillington, which contained a large number of incoherent variables (for example, one parent had custody of 15 children). I am satisfied that the result of these authors' data was an overestimate of the percentage of cases in which there was an adverse tax impact because a number of cases involving social assistance recipients were excluded and the number of cases studied was limited (only 171 cases). In my opinion, the percentage of cases where there was an adverse tax impact falls somewhere between the 29% submitted by the respondent and the 49% submitted by the applicant, and is probably closer to 29% than to 49%.

Of course, the remedy is not perfect and could be improved.27*ftnote27 See the article by J. W. Durnford and S. J. Toope entitled "Spousal Support in Family Law and Alimony in the Law of Taxation" (1994), 42 Can. Tax J. 1, at p. 75, in which the authors propose reforms to the deduction/inclusion system to improve its effectiveness and fairness. However, this does not mean that it is discriminatory in its present form. Section 15 of the Charter does not require that a remedy or mitigation of an earlier prejudice be 100% effective, flawless and without secondary effects if it is to be constitutionally valid. To impose such an obligation in terms of the result would have a paralysing effect on any initiative contemplated or taken to correct the prejudicial effects of a policy in the past.

In conclusion, I am of the opinion that the remedy created by paragraph 56(1)(b) treats the applicant and the people in her group differently from other taxpayers because of their different social, political and economic circumstances which are the result of their different civil status, and that despite its imperfections this measure is not discriminatory. As my colleague Linden J.A. stated in Schachtschneider v. Canada:

Treating both married and unmarried people fairly under the Income Tax Act without discrimination is no easy task. Past efforts to achieve this in the United States have led not to peace but only to an "uneasy truce". (See Druker v. C.I.R., 697 F.2d 46 (2nd Circ.); see also Bittker, "Federal Income Taxation and the Family" (1975), 27 Stan. L. Rev. 1389.) It is apparently virtually impossible to erect a scheme that is perfectly fair to all. Because of this, some latitude must be accorded to legisla-

tures in setting the terms of that truce. (Ibid., at p. 51, per Friendly J.).28*ftnote28 [1994] 1 F.C. 40 (C.A.), at p. 77.

The applicant also claims to have been the victim of discrimination on the basis of social condition. If I understand the argument properly as it is formulated, divorced women have a unique social condition because of their income and their level of education, and are in a disadvantaged position in society.

In short, she contends that she is a member of a disadvantaged group, a minority, and is a victim of stereotypes, historical disadvantages or social or political prejudice. At page 74 of their memorandum, counsel for the applicant write:

[translation] In the case at bar, we believe that sections 56(1)b, 56(1)c and 56(1)c.1 are discriminatory in their wording and in their application when they add to the income of a former spouse or a parent who has custody of a child for whom support payments are received; it is obvious in practice that 99% of the time mothers and former wives must bear the burden of the legal and physical custody of the children of the marriage with the person who is required to pay support and these provisions of the Income Tax Act operate to solidify their misery, with no justification whatsoever. [Emphasis added.]

This is therefore an argument based on a ground that is analogous to those listed in section 15 of the Charter.

The applicant's claim that the discrimination of which she is a victim is based on social condition is without merit, in my view. What the applicant is describing is the social condition or situation or, if you prefer, status in which she says she and the members of her group find themselves. She is describing the result of discrimination, if such there be, and not the cause: she is confusing cause and effect. Moreover, while it is possible that a group may be discriminated against because of its social condition, there is no evidence in the case at bar to establish that the alleged discrimination created by paragraph 56(1)(b) of the Act is in any way dictated by the applicant's social condition. It cannot be argued that this paragraph imposes a tax burden on the applicant because of her social condition. On the contrary, this legislative provision is designed to remedy the social condition in which taxpayers in her group find themselves because of divorce or separation as a result of the breakdown of the marriage.

For these reasons, the application for judicial review should be denied.