Developing Spousal Support Guidelines in Canada: Beginning the Discussion
 R.S.C. 1985, c. 3 (2nd Supp.). Although the explicit focus of this paper is on spousal support determinations under the federal Divorce Act, similar conclusions could be drawn about determinations of spousal support under provincial legislation. In general, despite some differences in statutory language, the interpretation of provincial spousal support statutes has been guided by the same basic principles, as articulated by the Supreme Court of Canada, that guide determinations under the Divorce Act.
 Moge v. Moge,  3 S.C.R. 813.
 Bracklow v. Bracklow,  1 S.C.R. 420.
 For elaboration of this theme see Carol Rogerson, "Spousal Support Post-Bracklow: The Pendulum Swings Again?" (2001), 19 Can. Fam. L. Q. 185 and D.A. Rollie Thompson, "Everything is Broken: No More Spousal Support Principles?" unpublished paper prepared for the Continuing Legal Education Society of British Columbia Family Law Conference, July 12-13, 2001.
 In addition to uncertainty, another perceived problem with child support awards, to which the guidelines were a response, was a sense that they were in general too low. As will be discussed further below, this "fairness" concern is more complex in the spousal support context, with concerns in the current environment both of awards that are in some cases too high and in others, too low.
 See the recent five-year review of the Federal Child Support Guidelines: Department of Justice, Canada, Children Come First: A Report to Parliament Reviewing the Provisions and Operation of the Federal Child Support Guidelines (May, 2002), which concludes that the guidelines have achieved their goal of making child support awards more consistent and predictable. The fairness of awards under the guidelines is a more complicated matter to assess. The Department of Justice concludes that the awards are perceived as being fair, but this remains a more contentious issue. Indeed, as will be discussed further below, some of the failures of child support in covering the full range of costs associated with the rearing of children create spillover effects in spousal support.
 See for example the study prepared for the Department of Justice, Canada by Danreb Inc., Spousal and Child Support Guidelines , October, 1988 (principal researcher: Julien Payne). This study was discussed and relied upon by the Alberta Law Reform Institute in recommending against adoption of a "fixed formula" for the determination of spousal support; see Alberta Law Reform Institute, Family Law Project: Spousal Support (Report for Discussion No. 18.2), October 1998, available at http://www.law.ualberta.ca/alri/ . See also another study prepared for the Department of Justice, Canada by the Canadian Research Institute for Law and the Family, Options for Reform of the Law of Spousal Support Under the Divorce Act, 1985 (May, 1991) (principal researcher: M.L.(Marnie) McCall).
 For an elaboration of these arguments see D.A. Rollie Thompson, "And ‘Average Justice' for All: Status and Stereotype in Support Law" unpublished paper prepared for the Law Society of Upper Canada and Ontario Bar Association—Family Law Section Continuing Legal Education Programme, "Child and Spousal Support Revisited" Toronto, May 2-3, 2002.
[8a] Andrews v. Andrews (1999), 50 R.F.L. (4th) 1 (Ont. C.A.).
 American Law Institute, Principles of the Law of Family Dissolution: Analysis and Recommendations (LexisNexis, 2002). The recommendations with respect to spousal support are found in chapter 5, "Compensatory Spousal Payments."
 The exception is the ALI proposals, which involve a blueprint for legislative reform—although the proposals do leave room for some shaping of the principles to reflect local norms.
 See Rogerson, supra note 4 and Thompson, supra note 4.
 Exceptions can, of course, be found. Some judges, who continue to give primacy to the compensatory framework, would deny any entitlement to spousal support based simply upon drop in standard of living if both spouses have maintained full employment during the marriage and there is thus no claim based on career loss. For judges who take this approach, non-compensatory support would be confined to cases where there is an inability to meet basic needs and could not be claimed by a spouse who is able to sustain a reasonable standard of living. See Leet v. Leet (2002), 25 R.F.L. (5th) 302 (N.B.Q.B.) and Graves v. Graves (2001), 20 R.F.L. (5th) (B.S.S.C.) for recent examples of cases denying entitlement based solely upon income disparity, without basic need. There are also some cases in which entitlement has been denied even when the claimant has basic need. In some cases conduct appears to be an unspoken factor, in others a concern to maximize resources for children in the payor's custody. For a review of cases see Rogerson, supra note 4.
 Keller v. Black,  O.J. No. 79, 182 D.L.R. (4th) 690 (S.C.J.).
 For a recent example see Grant v. Grant (2001), 22 R.F.L. (5th) 294 (Ont. S.C.J.).
 For a recent case clearly rejecting any principle of equalization see Cook v. Cook (2002), 27 R.F.L. (5th) 12 (N.S.S.C.).
 See Rogerson, supra note 4.
 The Ontario case law is reviewed extensively in Rogerson, supra note 4. For a recent decision from outside of Ontario adopting this approach see Weisner v. Weisner, 2002 CarswellAlta 1213.
 See Thompson, supra note 4.
 Andrews v. Andrews (1999), 50 R.F.L. (4th) 1(Ont. C.A.).
 Adams v. Adams (2001), 15 R.F.L. (5th) 1 (Ont. C.A.).
 See Thompson, supra note 4 and Professor Jay McLeod's annotation to the decision at 50 R.F.L. (4th) 1. On Linda Silver Dranoff's calculations, Andrews gave the wife 39% of the income that remained after child support, rather than 50%. See Linda Silver Dranoff, "Is there an evolving Spousal Support Formula? And does Need matter?" The Six-Minute Lawyer, Law Society of Upper Canada, Dec. 3, 2001.
 In a passage cited by Justice L'Heureux-Dubé in Moge, Ira Ellman argued that "Even crude approximations of theoretically defensible criteria are probably better than intuitive estimates of what is "fair" under a system lacking established principles of "fairness" in the first place." ("The Theory of Alimony" (1989), 77 Calif. L. Rev. 3 at 99)
 In practice, however, awards were often at a more modest level, reflecting more of a concern with basic provision and saving the public purse, rather than with fully protecting the wife's expectation interest.
 This discussion of early spousal support formulas draws on summaries provided in the Alberta Law Reform Institute, supra note 7 at 69 and McCall, supra note 7 at 7-8.
 For an excellent discussion of the centrality of fault to expectation-based models of spousal support see June Carbone, "The Futility of Coherence: The ALI's Principles of the Law of Family Dissolution, Compensatory Spousal Payments" (2002), 4 J. of Law and Family Studies 43.
 See for example, Tyerman v. Tyerman,  B.C.J. No. 2327 (S.C.) in which the parties had been married for only two years. The wife was 61 at time of marriage and husband 71. The wife had operated a hair-dressing salon prior to marriage, but ceased after marriage because the husband was adamant that she no longer work. The wife was unable to work after separation and the husband was ordered to pay support of $1,325 per month until his death. The court found that the marriage was entered into on the basis that wife's sole source of financial support would be from husband, with that support to continue for rest of wife's life. Although it is possible to justify the award in Tyerman on compensatory grounds, given that the wife gave up her employment because of the marriage in circumstances where she could never regain it, the actual analysis in the case is based on fulfilling the wife's expectation interest. See also the cases discussed below under the heading "Basic Social Obligation and Income Security."
 Law Reform Commission of Canada, Maintenance on Divorce: Working Paper 12 (Ottawa: Information Canada, 1975). It is somewhat difficult to slot the Commission's recommendations into a theoretical pigeon-hole. While much of the emphasis in the paper is on the rehabilitative and transitional aspect of spousal support, the working paper also places primary emphasis on the notion of spousal support as a response to "needs created by the marriage," a concept which can also be linked to compensatory theories, discussed below. The working paper also recognized that there would be some situations, following a long marriage, where self-sufficiency would not be possible and permanent support would be in order.
 Consisting of Pelech v. Pelech,  1 S.C.R. 801, Richardon v. Richardson,  1 S.C.R. 857 and Caron v. Caron,  1 S.C.R. 892.
 Ira Ellman, "The Theory of Alimony" (1989), 77 Calif. L. Rev. 3, reprinted in Canada at (1989), 5 Can. Fam. L. Q 1. Ellman provides further elaboration on his theory in a subsequent article, "Should the Theory of Alimony Include Nonfinancial Losses and Motivations?"  B.Y.U. L. Rev. 259. Another proponent of the compensatory theory is Allan Parkham, No Fault Divorce: What Went Wrong? (Boulder, Col: Westview Press, 1992).
 Many of the compensatory theories are grounded in concerns about economic efficiency and creating incentives for sharing behavior in families that will maximize marital gains.
 Under Ellman's theory, earning capacity loss claims would be confined to cases either where the earning capacity loss was incurred to further the economic advancement of the other spouse, such as a move to facilitate one spouse's career, or where it resulted from child care responsibilities. Earning capacity losses incurred for lifestyle reasons would not give rise to claims.
 Supra note 27.
 See Carbone, supra note 25.
 In his original article, supra note 29, Ellman acknowledged the difficulties of implementing the compensatory principle and talked about reliance on proxy measures of loss based on statistical evidence of average outcomes in such cases. He recognized that in the end precision is not obtainable and that the determination of alimony claims will rest upon the rough justice of trial judge discretion. But he asserted, in the passage cited at note 22, supra, that we are better off knowing the principle and what we should be doing, even if cannot do it perfectly. As will be discussed further below, as chief reporter for the ALI project on the Principles of Family Dissolution, Ellman endorsed a proxy measure for loss that arguably involves significant compromises of principle—a measure based on income disparity between the spouses at the point of marriage breakdown.
 In some cases the principle for long marriages has been expressed as providing similar lifestyles for each of the spouses. The use of need and standard of living as proxy measures for loss of opportunity is discussed by Bastarache J. A. (as he then was) in Ross v. Ross (1995), 16 R.F.L. (4th) 1 (N.B.C.A.) where he stated (at 7):
"It is in cases where it is not possible to determine the extent of the economic loss of the disadvantaged spouse that the Court will consider need and standard of living as the primary criteria, together with the ability to pay of the other spouse."
[35a] The gist of this assumption is that the wife, had she not assumed primary responsibility for child-rearing, would likely have had the same income as the husband.
 See Jana Singer, "Alimony and Efficiency: The Gendered Costs and Benefits of the Economic Justifications for Alimony" (1994), 82 Georgetown Law Journal 2423.
 Jana Singer, "Divorce Reform and Gender Justice" (1989), 67 N.C. L. Rev. 1103; also Singer ibid.
 See for example, Cynthia Starnes, "Divorce, and the Displaced Homemaker: A Discourse on Playing with Dolls, Partnership Buyouts, and Dissociation Under No-Fault" (1993), 60 U. Chic. L. Rev. 67. Starnes does not focus on income equalization, but rather on identifying increases in income over the course of the marriage and dividing the difference according to a sliding scale based on the number of years married. See also Robert Kirkman Collins, "The Theory of Alimony Residuals: Applying an Income Adjustment Calculus to the Enigma of Alimony" (2001), 24 Harv. Women's L.J. 23. Collins adopts a principle of "the equitable sharing of the residual economic benefits from work done during marriage" and proposes a scheme of post-divorce income sharing declining from 50% to zero over five equal periods pegged to the length of the marriage. He also recognizes as a beneficial side-effect of his theory that it would result in equal absorption of the economic shock of the separation. This suggests an alternative set of concerns with reliance and transition, rather than with compensation for work done during marriage.
[38a] See for example, Caratun v. Caratun (1993), 42 R.F.L. (3d) 113 (Ont. C.A.).
 Marinangeli v. Marinangeli (2001), 16 R.F.L. (5th) 326 (Ont. S.C.J.), now on appeal to the Ontario Court of Appeal. At issue in Marinangeli was the wife's entitlement to an increase in spousal support based on a post-divorce increase in the husband's income. The trial judge ruled that she was so entitled because she had contributed to his earning capacity by caring for the children during the marriage, thus freeing him to work. For other examples see Waterman v. Waterman (1996), 16 R.F.L. (4th) 10 (Nfld. C.A.) in which the Court re-conceptualized compensatory support as a form of sharing a marital asset (i.e. income) rather than as compensation for loss. For other recent cases where the spousal support analysis has recognized the wife's contribution to the husband's financial success through assumption of responsibility for the home and child care see Merritt v. Merritt,  O.J. No. 1732 (S.C.J.); Schmuck v. Reynolds Schmuck (1999), 50 R.F.L.(4th) 429 (Ont. S.C.J.); Lyttle v. Bourget,  N.S.J. No. 298 (S.C.); and Weir v. Weir (2000), 11 R.F.L. (5th) 233 (B.C.S.C.).
 Jane Ellis, "New Rules for Divorce: Transition Payments" (1993-94), 32 U. of Louisville J. of Fam. L 601. She writes, for example, that a ten year marriage would have two years of equal sharing, followed by three years in which the sharing went from 50% to zero. She suggests that a table with multipliers could be developed so that the amounts could be computed quickly and easily without professional help.
 Stepen D. Sugarman, "Dividing Financial Interests Upon Divorce" in Divorce Reform at the Crossroads, Stephen D. Sugarman and Herma Hill Kay, eds. (New Haven: Yale University Press, 1990).
 Milton C. Regan, Jr., Family Law and the Pursuit of Intimacy ( New York and London: New York University Press, 1993) and Alone Together: Law and the Meanings of Marriage (New York: Oxford University Press, 1999). His proposal is therefore similar to that of Singer, supra note 37, although justified in a different way.
 Supra note 41.
 At 160.
 The obligation to pay non-compensatory support to meet a former spouse's needs is said, by the Court, not to flow from the fact of marriage, per se, but rather from how the spouses have organized their lives. Specifically, the obligation is said to arise from a pattern of economic interdependence which developed during the marriage. Marriage is seen, at para. 30, to create interdependencies which cannot easily be unraveled. Non-compensatory support is said , at para. 31, to recognize "the reality that when people cohabit oaver a period of time in a family relationship, their affairs may become intermingled and impossible to disentangle neatly."
 See Bracklow v. Bracklow (1999), 3 R.F.L. (5th) 179 in which , after an eight year relationship, a five-year time-limit was imposed on the support obligation despite Mrs. Bracklow's on-going need. When interim support is taken into account, the length of the support period is roughly equivalent to the length of the relationship.
[46a] Even in the fault-based era, where the principles of alimony would have suggested that a wife guilty of a matrimonial offense would be completely disentitled from spousal support, in practice public policy concerns prevailed to justify the awarding of some support to keep the wife from becoming a public charge.
[46b] For a recent example of a case awarding spousal support on this basis see Skoreyko v. Skoreyko (2002), 28 R.F.L. (5th) 440 (B.C.S.C.) (both parties employed during 15 year childless marriage; wife loses sight after marriage breakdown; husband ordered to pay support).
[46c] Moge, supra note 2 at 865.
 At para. 31. See also M. v. H.,  2 S.C.R. 3 where, in the course of justifying the extension of the spousal support obligation to same-sex cohabiting couples, the Supreme Court of Canada recognized that one of the purposes of spousal support is to "alleviate the burden on the public purse" by shifting the obligation to provide support for needy persons to spouses.
 See June Carbone, "Income Sharing: Redefining the Family In Terms of Community" (1994), 31 Houston L. Rev. 359.
 For cases recognizing this rationale, which is reflected in s. 15.2(6)(c) of the Divorce Act, see Brockie v. Brockie (1987), 5 R.F.L. (3d) 440 (Man. Q.B.), aff'd (1987), 8 R.F.L. (3d) 302 (Man. C.A.), Kennedy v. Kennedy (1994), 98 B.C.L.R. (2d) 287 (B.C.S.C.), and Weisner, supra note 17.
 Joan Williams, "Is Coverture Dead? Beyond a New Theory of Alimony" (1994), 82 Georgetown Law Journal 2227. For an earlier version of such a proposal see Jane Rutherford, "Duty in Divorce: Shared Income as a Path to Equality" (1990), Fordham Law Rev. 539 which proposed equal division of income on a per capita basis, with the sharing to be indefinite in duration. Division of income on a per capita basis would fail to take into account the cost savings from sharing of household costs. Both John Eekelaar and Mary Ann Glendon have long advocated a "children-first" principle which would govern the distribution of all financial resources upon marriage breakdown. Such a principle would make provision for the basic or average needs of children's household a first claim on all financial resources. What this "second wave" of income-sharing proposals does, by drawing on a norm of equalization of household standards of living, is to provide for more generous income-sharing than that proposed by Eekelaar or Glendon in cases where there are sufficient resources to go beyond meeting basic or average needs.
 Andrews, supra note 19.
 Adams, supra note 20.
 See Cynthia Lee Starnes, "Victims, Breeders, Joy, and Math: First Thoughts on the Compensatory Spousal Payments under the Principles" (2001), 8 Duke Journal of Gender, Law and Policy 137.
 Approved by the National Conference of Commissioners on Uniform State Laws in 1970 as a model act proposed for adoption by individual states. See discussion in Grace Ganz Blumberg, "The Financial Incidents of Family Dissolution" in Sanford N. Katz, John Eekelaar, and Mavis Maclean, eds., Cross-Currents: Family Law and Policy in the United States and England (Oxford: Oxford University Press, 2000).
 Recall McLachlin C.J.'s comments in Bracklow, at para. 54, with respect to the inter-relationship between quantum and duration—a modest support order of indefinite duration can be collapsed into a more substantial lump-sum payment.
 See Superior Court of the State of California, County of Santa Clara, Rules of Court, Rule 3.3 (c), "Temporary Spousal Support Formula" found at http://claraweb.co.santa-clara.ca.us/sct/rules/summary.htm.
 See George H. Norton, "Support Schedules in California: Selected Custody and Spousal Support Issues" (1987), 4 Calif. Fam. Law Mthly 57, discussed in Appendix A to the Alberta Law Reform Institute report, supra note 7.
 See Humbolt County Trial Court Rules (2001), Appendix 9.7—Spousal Support Schedule. While this formula generates the average award, the rules also set a minimum award using 30% of the higher earner's income in the formula and a maximum award using 40%. In Humboldt County, the formula percentages are also lower in cases where there minor children and a concurrent child support award is in place. In such cases the percentages of the higher earner's income used in the formula are 30% (average award), 25% (minimum award) and 35% (maximum award). The Humboldt guidelines also make clear that no spousal support will generally be paid if the lower earner spouse has a net income of 60% or more of that of the higher earner.
 Supra note 19; discussed in Part II, above.
 "Permanent" simply means final determinations of support at trial (in contrast to temporary or interim awards), and does not necessarily imply orders for permanent or indefinite support.
 Of this Norton writes:
"This arbitrary limitation on spousal support would answer the difficult question of how long a spouse who cannot or will not earn remains the responsibility of his or her former spouse. Marriage is not an insurance policy. There is a time when society, rather than the former spouse, should bear this burden if a spouse cannot or will not earn."
 Norton writes at 71:
"Remarriage shall terminate spousal support, except when an order is made that support should not terminate, on a motion by a party intending to remarry and if good cause is shown. If a party remarrying is subsequently divorced, he or she may request reinstitution of support from a prior spouse, if support would have otherwise continued until the time of the motion and the term of the remarriage was less than five years or half the length of the prior marriage, whichever is less. If support is reinstituted, the court may consider changes in circumstances, but it may not award support to a point in time later than previously could have been ordered. This reflects a policy of the state to encourage remarriage without undue risk or penalty to the remarrying spouse."
 For marriages of ten years or less, the period would be not less than half the length of the marriage (calculated in months). For marriages of 10 to 20 years, the minimum period would be determined by the formula of months married times months married/ 240.
 Pa. R. Civ. P., sections 1910.16-1 to 1910.16-4.
 They also apply to applications for "spousal support" which would appear to refer to permanent support orders applicable during the period of separation prior to a formal divorce. See Rule 1910.16-1. Orders for alimony pendente lite are often in place for two or three years.
 See Domestic Relations Act, 23 Pa. C.S., section 3701.
 Supra note 54, section 1910.16-1(d).
 The formula is found in section 1910.16-4.
 See Marie Gordon, "Spousal Support Guidelines and the American Experience: Moving Beyond Discretion" paper presented at the National Family Law Program, Kelowna, B.C., July 14-18 2002.
 Domestic Relations Act, supra note 56, section 3701(c) allows a court to determine the duration of the order, "which may be for a definite or indefinite period of time which is reasonable under the circumstances."
 See "Family Law Guidelines" for Family Law Practice in Johnson County, Kansas, Johnson County Bar Association, Family Law Bench Bar Committee, revised February 2001, available at http://www.jocobar.org/practicetools.htm. The guidelines for maintenance are found in Section V.
 See s. 5.6. This formula applies to a difference of up to $50,000 per year. For differences in excess of $50,000, the applicable percentage is 22. The Shawnee County Family Law Guidelines also use 25% of the income difference, reduced to 22% to the extent the difference exceeds $50,000.
 See s. 5.7. In cases involving minor children the Shawnee County Family Law Guidelines use 20% of the income difference, reduced to 17% to the extent the difference exceeds $50,000.
 The guidelines for duration were crafted in light of provision in the Kansas spousal support legislation precluding a trial court from awarding maintenance for a period longer than 121 months (i.e. 10 years and 1 month), although allowing for the possibility of a judicial extension in exceptional cases. See K.S.A. 60-1610(2).
 American Law Institute, Principles of the Law of Family Dissolution: Analysis and Recommendations (LexisNexis, 2002). The recommendations with respect to spousal support are found in chapter 5, "Compensatory Spousal Payments."
 As with the Pennsylvania guidelines, the ALI proposes that its guidelines be presumptive, but that departures from the presumptions be allowed when a trial court makes written findings that establish that the presumption's application to the case before the court will yield a "substantial injustice." See sections 5.04(4) and 5.05(6) and the comments thereon.
 This claim is set out in s. 5.04.
 Justifications based on contract and expectation damages are also rejected given no-fault divorce.
 The rationale offered in the comment (c) on s. 5.04 is :
The obligation recognized by this section thus does not arise from the marriage ceremony alone, but takes longer to develop. As a marriage lengthens the parties assume roles and functions with respect to one another. When adults share enough of their lives together, they may mold one another as surely as parents affect their child. Eventually the molds harden. … The obligation assumed by this section thus assumes no blameworthiness for the marital failure, just as the obligation to support one's child assumes no blameworthiness for the child's conception. It is enough to recognize that the parties' situation at the end of the marriage is a consequence of both their acts to conclude that it is their joint responsibility.
 See supra note 43 and accompanying text. The Reporter's notes made explicit reference to Stephen Sugarman.
 The rule explicitly refers to expected incomes at the time of dissolution to recognize that spouses are expected to realize their earning potential after dissolution, even if they were not employed during the marriage. When parties do not realize their earning potential, income may be imputed. See comment (f) on s. 5.04.
 Although leaving leeway to the states to establish minimum duration, the Principles suggests that somewhere in the range between 5 and 10 years would be consistent with the rationale underlying these claims.
 As will be seen below, these are the requirements adopted in Maricopa County.
 Basically states are advised to begin by specifying the maximum value of the durational factor and the duration at which it is reached and then working backwards. Thus a determination that a maximum value of the duration factor would be .4 (which would mean sharing 40% of the income difference) and would be reached after a 40 year marriage would set the durational factor at .01 times the years of marriage.
 It is contemplated that the presumption could be rebutted and an award made for a shorter period if it is shown that the loss will be ameliorated more quickly because of anticipated changes in the financial position of parties.
 See comment (e) on s. 5.05.
 See comment (f) on s. 5.04.
 See s. 5.07.
 See. s. 5.09.
[89a] The child support recommendations are found in chapter 3. Through use of a supplemental percentage, the ALI child support formula makes some adjustment for parental income disparity. The "compensatory payment" (i.e. spousal support) is determined and transferred first. Initially child support is calculated by a "base percentage" and then a "supplemental percentage" of the payor's net income. As the recipient parent's income (in excess of a self-support reserve) rises, the supplemental percentage is reduced, reaching zero when parental incomes are equal (and even the base percentage can be reduced where the recipient parent's income exceeds that of the payor parent). The net effect of the supplement is to reduce, but not eliminate, disparities in household living standards.
 See comment (e) on s. 5.05 discussing implementing measure for earning capacity loss.
 See the Maricopa County guidelines, discussed below, which implemented a simplified version of the ALI guidelines that eliminated the separate primary care-giver claim.
 For example, see symposium issues in (2001), 8 Duke Journal of Gender, Law and Policy, (2002), J. of Law and Family Studies, [200l] Brigham Young U.L.R and (2002), 4 J. of Law and Family Studies.
[92a] June Carbone, "The Futility of Coherence: The ALI's Principles of the Law of Family Dissolution, Compensatory Spousal Payments" (2002), 4 J. of Law and Family Studies 43. She summarizes her paper as follows:
In this paper, I will explore the tradeoffs that underlie the ALI's proposed system of compensatory spousal payments in light of the tortured history of alimony. I will maintain, first, that the ALI is certainly correct that the existing law, with its emphasis on need, is incoherent at best, or insulting or archaic at its worst. Second, I will agree that the idea of compensation provides the alternative to need that has the best hope of bringing a measure of coherence to the existing cases. Third, however, I will show that at the core of the provision for compensatory payments lies a fundamental dilemma; the refusal to recognize fault necessarily limits the provision for compensation, not just with respect to the non-financial losses the ALI principles acknowledge, but with respect to some of the financial concerns at the heart of the proposals. I will nonetheless conclude that bringing fault back into the system is too costly to contemplate for all kinds of reasons, but that the failure to acknowledge it directly will fuel resistance to some of the ALI proposals.
 See for example J. Thomas Oldham, "ALI Principles of Family Dissolution: Some Comments,"  U. Ill. L. Rev. 801. However, Oldham's critique is complicated. While critical of some aspects of the expansion of spousal support under the ALI Principles, he also takes the position that insufficient compensation is provided to primary care-givers in short and intermediate duration marriages who have actually sacrificed earning capacity and are in need of economic rehabilitation. For these cases he would provide a more generous remedy. This aspect of his critique will be discussed further below.
 See for example Penelope Eileen Bryan, "Vacant Promises?: The ALI Principles of the Law of Family Dissolution and the Post-Divorce Financial Circumstances of Women"(2001), 8 Duke Journal of Gender, Law and Policy 167; Marsha Garrison, "The Economic Principles of Divorce: Would Adoption of the ALI Principles Improve Current Outcomes?" (2001), 8 Duke Journal of Gender, Law and Policy 119; and Cynthia Lee Starnes, "Victims, Breeders, Joy, and Math: First Thoughts on the Compensatory Spousal Payments under the Principles" (2001), 8 Duke Journal of Gender, Law and Policy 137.
 Bryan, ibid.
 Bryan, ibid and Garrison, supra note 94.
 Starnes, supra note 94, suggests using as a model the Uniform Probate Code formula for determining a spouse's elective share of an augmented estate based on the length of the marriage, which would produce a compensatory spousal payment of 15% of income disparity after 5 years, 30% of any disparity after 10 years, and 50% of any disparity after 15 years.
 Bryan and Starnes, both supra note 94. Starnes sees the loss principle as based on a "victimization" model of spousal support.
 Bryan and Starnes, both supra note 94.
 Starnes, supra note 94 is concerned about distinctions being drawn between "breeder" and "non-breeder" women.
 See Bryan and Garrison, both supra note 94, and Tonya L. Brito, "Spousal Support Takes on the Mommy Track: Why the ALI Proposal is Good for Working Mothers" (2001), 8 Duke Journal of Gender, Law and Policy 151.
 See Oldham, supra note 93. Oldham is very clear about the need to impose durational limits on such awards, given the frequency of remarriage, suggesting a maximum of 5 years. He, like the ALI, would make an exception and provide for indefinite support if a marriage with children exceeded a certain duration and the claimant exceeded a certain age.
 See Starnes, supra note 94.
[103a] See Federal/Provincial/Territorial Family Law Committee's Report and Recommendations on Child Support (January, 1995),Recommendation 10.3.2: "The Family Law Committee recommends that the non-financial contribution of custodial parents toward their children not be compensated within the child support formula at this point in time" (p. 47). The discussion supporting this recommendation indicated that these contributions "can already be compensated by way of spousal support under the Divorce Act as well as under some provincial legislation."
 Superior Court of Arizona, Maricopa County, Family Court Department, Spousal Maintenance Guidelines. The full text of the guidelines is not yet available on the web. A summary can be found at http://www.thefinancialexpert.com/leftpanel2.html. An extensive discussion of the Maricopa Guidelines and the process of their creation can be found in Ira Ellman, "The Maturing Law of Divorce Finances: Towards Rules and Guidelines" (1999), 33 Family Law Quarterly 801.
 The following "caution" appears at the beginning of the guidelines:
These guidelines contain a mathematical formula for calculating spousal maintenance. The formula should be used only after a threshold determination of eligibility for spousal maintenance is made under A.R.S. s. 25-319(A)(1), (2), or (4). The guidelines are simply intended to provide the court and parties with a starting point for discussion, negotiation or decision-making. They do not change or create public policy. They do not constitute a presumption. Most importantly, they are not intended to replace the trial court's obligation to consider specific evidence, as well as all applicable statutory factors.
 See A.R.S. para 25-319 where the grounds of entitlement include that a spouse lacks sufficient property; is unable to support himself or herself by appropriate employment; or has had a long marriage and is of an age which precludes employment. These grounds reflect those in the UMDA, supra note 54.
 See Gordon, supra note 69.
 Calculated or rounded to the nearest whole number of years.
 The durational factor is to be calculated or rounded to the nearest hundredth.
 Once again calculated or rounded to the nearest whole number of years.
 An award of indefinite duration may specify that it shall terminate at such time as the payor retires.
 Correlations between the formula and existing practice was based upon a sample of approximately 160 cases from 1996 to 1998, involving both contested cases and consent orders.
 The correlation with quantum was 0.75.
 See Linda Silver Dranoff, "Suggested Formula for Determining Spousal Support," paper presented at the Canadian Bar Association-Ontario, 2000 Institute of Continuing Legal Education, Toronto, January 28, 2000; the formula is further discussed in Linda Silver Dranoff, "Is there an evolving Spousal Support Formula? And does Need matter?" The Six-Minute Lawyer, Law Society of Upper Canada, Dec. 3, 2001. In the latter paper Dranoff also discusses the results of a survey she conducted amongst members of the family law bar to determine if they had used her suggested formula, or any other formula, and if so, whether it had been accepted by a court. Of 36 respondents, 14 had used a formula and reported a fairly high acceptance rate. Some used her formula or variants of it, others used a variety of other formulas.
 Supra note 19.
 This way of understanding the methodology implicit in Andrews is also supported by McLeod and Thompson, as discussed supra at note 21.
 In a case where 50% is used the results will be the same whether the rule is "leave the wife with 50% of the remaining available income" or "give the wife 50% of the income difference between the spouses." But if the percentage is, for example, 40%, the result under Dranoff's formula, which will be to give the wife sufficient spousal support so that she is left with 40% of the remaining available income, will be different than the result that would be achieved by calculating spousal support as 40% of the income difference between the spouses.
 See John Syrtash and Karen Freiday, "Opposing a Spousal Support Formula and Alternatives" (May 2002), 13 (6) Matrimonial Affairs 15.
 Canada, Department of Justice, Evaluation of the Divorce Act: Phase II: Monitoring and Evaluation (Ottawa, 1990).
 The discrepancy might be explained by, at least in part, by the fact that the men and women interviewed were not corresponding spouses.
 Canadian Centre for Justice Statistics, Child and Spousal Support: Introduction to the Maintenance Enforcement Survey ( Statistics Canada, 2002).
 Provincial orders barely had any type of spousal support orders--.04% with respect to each category.
 With respect to orders under provincial legislation, 1.3% of all support orders related to spousal support only and 2.5% related to combined child and spousal support orders.
 The database contains data from 21 selected courts in all provinces and territories except Quebec and Nunavut. The total number of cases, including original orders and variations, collected between November 1998 and February 2002 was 33,240.
 The American studies are reviewed by the ALI, supra note 9. The highest incidence found in any American study was 30% in a large sample of divorcing parents in California with at least one child under 16: Maccoby and Mnookin, Dividing the Child (Cambridge: Harvard University Press, 1992) at 123-4.
 Supra note 119.
 For an extremely thoughtful analysis of the impact of the discretionary nature of the support entitlement on the dynamic of bargaining see Craig Martin, "Unequal Shadows: Negotiation Theory and Spousal Support Under the Canadian Divorce Act" (1998), 56 U.T.Fac.L.Rev. 135.
 The source for this data is Statistics Canada, Health Statistics Division, Divorces 1999 and 2000, Cat. No. 84F0213XPB (December 2002), Table 15.
 Although given that a significant number of couples cohabit for some period of time before marriage, the median for relationship duration may bump back up to 11 years or even slightly higher.
  Thompson, supra note 4, provides a more detailed breakdown of the 1998 divorce data. Attempting to classify marriages into three categories of "short," "medium" and "long," he breaks the 1998 divorce data into thirds. His conclusions were: "short" marriages would be those under 7 years, while "long" marriages would be anything 16 years or longer. "Medium" length marriages would range from 8 to 15 years.
 See supra note 128.
 See California, where the Santa Clara County guidelines are now found in the Rules of Court, and Pennsylvania, where informal guidelines originally created in Allegheny County were eventually adopted state-wide through legislation.
- Date modified: