Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Understanding economic inequality for women in Canada's retirement income system: reform, restructuring… Barnsley, Paula Elizabeth 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0005.pdf [ 11.46MB ]
Metadata
JSON: 831-1.0077459.json
JSON-LD: 831-1.0077459-ld.json
RDF/XML (Pretty): 831-1.0077459-rdf.xml
RDF/JSON: 831-1.0077459-rdf.json
Turtle: 831-1.0077459-turtle.txt
N-Triples: 831-1.0077459-rdf-ntriples.txt
Original Record: 831-1.0077459-source.json
Full Text
831-1.0077459-fulltext.txt
Citation
831-1.0077459.ris

Full Text

UNDERSTANDING ECONOMIC INEQUALITY FORW O M E N IN CANADA'S RETIREMENT INCOME SYSTEM: REFORM, RESTRUCTURING A N D B E Y O N D BY PAULA ELIZABETH BARNSLEY B . A . (Ed.), Memorial University o f Newfoundland, 1969 M . E d . , Memorial University o f Newfoundland, 1973 L L . B . , Dalhousie University, 1991  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S OF T H E D E G R E E OF  M A S T E R OF L A W S in T H E F A C U L T Y OF G R A D U A T E STUDIES (Faculty o f Law)  We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH C O L U M B I A December 1998 © Paula Elizabeth Barnsley, 1998  In  presenting  degree  this thesis  in partial fulfilment  the  requirements  for  I further agree that permission for  of this thesis for scholarly purposes  department  or  by  his  or  her  representatives.  may be granted  permission.  Department of  JL fit-uj  The University of British Columbia Vancouver, Canada  DE-6 (2/88)  advanced  It  is  17  extensive  by the head of  understood  that  publication of this thesis for financial gain shall not be allowed without  Date  an  at the University of British Columbia, I agree that the Library shall make it  freely available for reference and study. copying  of  copying  my or  my written  -11-  ABSTRACT Gendered poverty among the elderly is a statistical fact. Previous studies have identified inequitable treatment o f women and insufficient income for unattached elderly women among the most serious shortcomings o f the retirement income system. Despite pension reform over the past decade, the gender gap has widened for elderly Canadians whose incomes fall below the poverty line. This thesis seeks to understand the relationship between the laws that govern Canada's retirement income system and the over-representation o f elderly women among Canada's poor, and to explore why the retirement income system continues to deliver benefits in a manner that, though expressed in gender neutral language, is systemically unfair to women.  The benefits o f Canada's retirement income system may be accessed through workforce participation and, in a more limited way, through a spousal relationship. Familial ideology is used as the theoretical framework to examine the role o f the laws that govern access to benefits in reinforcing and perpetuating assumptions about women that undermine their economic autonomy. This examination reveals that gendered economic inequality is embedded within Canada's retirement income system because it accepts the social and economic construction implicit in familial ideology o f women as economically subordinate to, and dependent upon, men. The relationship between gender inequality and the two modes o f delivery o f retirement income benefits, during retirement as pension benefits and prior to retirement as tax subsidies that enhance taxpayers' opportunities to accumulate retirement savings, is also explored. A tax expenditure analysis exposes the bias against the economically disadvantaged (mostly women) inherent i n delivering benefits as tax subsidies.  Additionally, familial, public/private and  -inrestructuring ideologies are used as methodological tools to interrogate the reform process which, although ignoring gender issues, paradoxically deepened and compounded the  systemic  inequalities for women that existed prior to reform.  The thesis concludes by offering suggestions for developing a progressive agenda for advancing gender equality within the retirement income system. The limitations o f legal action as a strategy for implementing this type o f agenda are discussed, and political action is designated as the most promising strategy for achieving progressive reform.  -iv-  TABLE OF CONTENTS  ABSTRACT  ii  T A B L E OF CONTENTS  iv  ACKNOWLEDGEMENT  INTRODUCTION  viii  1  CHAPTER O N E C A N A D A ' S RETIREMENT INCOME SYSTEM - A N OVERVIEW I  9  E A R L Y HISTORY: TO 1951 (a) The Government Annuities Act, 1908 (b) The Old Age Pensions Act, 1927  10  n  EVOLUTION OF THE THREE TIERED SYSTEM: 1951 TO 1970s (a) 1951: Old Age Security Act and Old Age Assistance Act Debate - 'universal' or 'contributory'? The 1951 Program (b) 1957: Tax Assisted 'Private' Plans RPPs RRSPs (c) 1965: Canada Pension Plan Financing Retirement pension benefits Eligibility for retirement pension benefits Survivor benefits Other benefits (d) 1966 and 1975: Additions to Tier 1 1966: Guaranteed income supplement 1975: Spouse's allowance (e) The Three Tiered System  17 17 18 19 22 22 23 27 29 30 31 32 33 33 33 35 36  HI  PROPOSAL FOR REFORM: GREEN PAPER ON PENSION REFORM (1982)  39  10 12  -V-  IV  THE'90s: PENSION REFORM (a) 1991: Tier HI Reform (b) 1998: Tier II Reform Financing Retirement pension benefits Eligibility for retirement pension benefits Survivor benefits Death benefit (c) Tier I Reform (proposed)  43 44 46 47 48 49 49 49 51  V  W H A T DOES THE T A X SYSTEM H A V E TO DO WITH PENSIONS?  56  CHAPTER TWO R E A L I T Y A N D M Y T H : T H E E C O N O M I C SITUATION OF E L D E R L Y W O M E N I  W O M E N AND THE RETIREMENT INCOME SYSTEM (a) Access through Workforce Participation (b) Access through Spousal Relationship Tier I: S P A Tier II: C P P survivor benefits Tier III: R P P survivor benefits Tier HI: Spousal R R S P s (c) Rationale for Spousal Provisions Spouses are women But not all women are spouses (d) The Longevity Factor (e) Neutral is not Enough  U  W O M E N : R E A L OR IDEAL? (a) What is Ideology? (b) The Ideology of the Family Familial ideology and dependency The family wage Dependency and family law Contradictory effects: family and welfare law The family - private safe haven? Marginalization of non-conforming families (c) Familial ideology and (Inequality (d) The Family and the Public/Private Divide.  62 62 63 72 73 73 74 75 75 75 78 79 81 83 83 88 90 90 92 94 96 97 103 105  -VI-  CHAPTER THREE F A M I L I A L I D E O L O G Y A N D (IN)INEQUALITY IN T H E R E T I R E M E N T I N C O M E S Y S T E M  110  I  DELIVERY OF RETIREMENT PENSION BENEFITS  111  (a) (b)  Ill 116 117 125 125 127 131  n  A s of Right A s a Spouse S P A (Tier I) Survivor benefits C P P (TierU) RPPs (Tier I I I ) Spousal R R S P s (Tier I I I )  DELIVERY OF T A X BENEFITS (a) A s of Right (b) A s a Spouse Spousal R R S P s (c) The Problem with Tax Benefits  133 133 134 135 138  CHAPTER FOUR RESTRUCTURING A N D (IN)EQUALITY IN T H E R E T I R E M E N T I N C O M E S Y S T E M  145  I  T H E RETIREMENT INCOME SYSTEM - REFORMED OR RESTRUCTURED? (a) The Reformed System (b) Reformandthe 1 9 8 2 Reform Agenda Guaranteed reasonable minimum income Fair opportunities and arrangements A v o i d serious disruption o f living standards (c) Reform and the Restructuring Agenda  147 147 150 150 151 153 155  n  (IN)EQUALITY AND THE RESTRUCTURED RETIREMENT INCOME SYSTEM  160  (a)  160 162 163 166 169 171 171 172 174 176 177  (b)  (c)  The Public/Private Divide The restructured 'private' Market regulation o f the economy 'Re'-privatizing social welfare The restructured 'public' Transformed Social Safety Net The new citizenship The 'ordinary'/'special' dichotomy Targeting The new senior citizenship Restructuring - The " N o Choice' Alternative?  -Vll-  CHAPTER FIVE E Q U A L I T Y FOR W O M E N IN T H E R E T I R E M E N T I N C O M E S Y S T E M : W H E R E TO F R O M H E R E ? .183 I  ALTERNATIVES TO THE REFORMED RETIREMENT INCOME SYSTEM (a) A Progressive Reform Agenda Strengthening the 'public' tiers Reduced role for the 'private' tier Suggestions for an equality enhancing agenda  184 185 186 187 188  n  B E Y O N D RESTRUCTURING - STRATEGIES FOR EQUALITY SEEKING REFORM (a) Legal Action The Charter Spousal provisions Economic equality International human rights agreements (b) Political Action  190 190 191 200 203 205 207  CONCLUSION  211  BIBLIOGRAPHY  215  -Vlll-  ACKNOWLEDGEMENT  I am deeply appreciative o f the respect, guidance, encouragement, and immeasurable patience extended to me by Professors Claire Y o u n g and Marlee K l i n e during the writing o f this thesis. I am grateful as well to the L a w Foundation o f N e w Brunswick for its financial support. For their moral support I thank my family, my friends, and my colleagues at Eddy Young Hoyt and Downs who encouraged me via email. A n d especially, many thanks to Roger for his patience and support.  -1-  UNDERSTANDING ECONOMIC INEQUALITY FOR WOMEN IN CANADA'S RETIREMENT INCOME SYSTEM: REFORM, RESTRUCTURING AND BEYOND  INTRODUCTION  Elderly women in Canada are twice as likely as elderly men to live in poverty.  1  Whether  above or below the poverty line , women's retirement incomes lag significantly behind those o f 2  men.  3  I am interested in understanding why elderly Canadian women are so much more  economically vulnerable than their male counterparts and what, i f anything, can be done to redress this inequity. The understanding I seek requires an examination o f Canada's retirement income system and the tax policies that support it. O f course tax policies do not exist in a vacuum but reflect the contemporary political climate.  The retirement income system, too, reflects the  particular contexts in which it evolved, and integral to these contexts are relations o f social, economic and political subordination based on such characteristics as gender, class, race, sexual orientation, disability and citizenship. Therefore, in order to understand the current structure o f the retirement income system and the inequalities within it, it is necessary to consider the historical social, economic and political context in which the system developed. A s the past informs the  'Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications  (Ottawa: The Caledon Institute of Social Policy, 1996), Appendix C - Pension Trends, graph entitled "Poverty Rates, Elderly Canadians, By Sex, 1980 - 1994". Although Statistics Canada goes to great pains to avoid the use of the term 'poverty line', and instead uses the term "low income cut-off (Statistics Canada, Women in Canada: A Statistical Report (3 ed.) (Ottawa: Statistics Canada, 1995) at 86), in this thesis I follow the practice of the National Council of Welfare and use Statistics Canada's low income cut-offs as indicators of poverty. References to 'poverty line' are references to Statistics Canada's low income cut-offs. 2  rd  'Revenue Canada, Tax Statistics on Individuals: 1995 Tax Year (Ottawa: Revenue Canada, Statistics Division, 1997) at 131-133 and 237.  -2present, so analysis o f the current social, economic and political context must ground the construction o f strategies through which to advance a reform agenda aimed toward redressing the disproportionate representation o f women among those who are unable to amass adequate resources for their economic security in retirement.  Necessarily, then, the inquiry into the relationship  between poverty o f elderly women and the retirement income system w i l l include political, as well as legal, analysis.  Currently Canada's pension system is undergoing significant reform. Concern with inadequacies in the pension system, which would only worsen as the percentage o f the population over 65 increased, motivated the federal government to create the Task Force on the Retirement Income System in 1979. The task force report, the Green Paper on Pensions published in 1982 , 4  outlined the federal government's proposals for improving the pension system and provided a springboard for "the great pension debate that continued throughout the 1980s".  5  In presentations to government committees following the publication o f the Green Paper, feminist and anti-poverty activists tenaciously pushed for a pension system that achieves substantive gender equality. The entrenchment o f the Charter in 1982 raised expectations for 6  Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Welfare Canada, 1982). 4  Monica Townson, Partnerships in Pensions: Who's Responsible Anyway (Ottawa: The Canadian Centre for Policy Alternatives, 1990) at 2. 5  ^Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11.  -3achieving gender equality in pension reform. After all, in a democratic country in which gender 7  equality is a constitutionally protected right, social policies should be expected to drive pension reform toward the goal o f ensuring adequate retirement incomes for all Canadians. This would necessarily require closing the standard o f living gap between men and women during their retirement years that the current pension system reinforces. However, Statistics Canada data show that, although the poverty rate for Canadians over the age o f 65 years has dropped over the years from 1980 to 1994, the gender gap has widened. While some progress has been made, the poverty 8  rate for Canada's elderly remains disturbingly high, especially for unattached women.  9  In this thesis I examine the relationship between Canada's pension and tax systems and the over-representation o f elderly women among Canada's poor. I explore why Canada's retirement income system continues to deliver benefits in a manner that, though it is gender neutral on its face, is systemically unfair to women. I consider the extent to which the reform process being phased in through the 1990s, and which w i l l undoubtedly continue into the first decade of the 2000s , w i l l 10  bring about a retirement income system that is gender neutral in substance, rather than merely in  See Andrea Seale, "Can the Canada Pension Plan Survive the Charter? Section 15(1) and Sex (In)Equality" (1985) 10 Queen's L J . 441 and Susannah Worth Rowley, "Women, Pensions and Equality" (1986) 10 Dalhousie L.J. 283. 7  "Caledon Institute of Social Policy, supra note 1, Appendix C. The graph entitled "Poverty Rates, Elderly Canadians, By Sex, 1980-1994 shows that 39.9% of women and 26.6% of men over the age of 65 had incomes below the poverty rate in 1980, while in 1994 25.8% of women and 10.7% of men over the age of 65 had incomes below the poverty rate. Ibid. The graph entitled "Poverty Rates, Elderly Unattached Individuals, By Sex, 1980-1994" shows that in 1994 31.8% of unattached men and 52.9% of unattached women over the age of 65 lived in poverty. 9  In 1991, after a decade of task forces, studies and debates over draft legislation, legislation reforming the private, tax subsidized part of the pension system came into force. Legislation effecting Canada Pension Plan reform passed third reading on December 4,1997 and came into force on January 1,1998. Although the Seniors Benefit that was intended to replace the Old Age Security and Guaranteed Income Supplement in 2001 has been shelved, reform of these programs in the foreseeable future is anticipated. The history, structure and reform of Canada's retirement income system are described and discussed in detail in Chapter One. l0  -4form. In this analysis o f recently implemented and proposed reform initiatives I looked for, but did not find, gender equality enhancing improvements which recognize that a system based on men's economic realities cannot achieve equality by treating women and men the same and thus ignoring women's economic realities. I argue that the political climate i n Canada i n the 1990s, based on a "neoliberal governing philosophy" which "celebrates the ideas of market-driven development and free enterprise"", is reflected in the downloading of responsibility for the economic well-being o f Canada's elderly from the public sphere (government) to the private spheres (the market and families). This trend can be seen in recent and proposed reforms to the retirement income system. Further, I shall demonstrate that this shift in the public/private divide increases the gendering o f poverty among the elderly. M y aim is to use tax expenditure analysis and feminist legal theory to expose the flaws in the reformed pension system that may look like they are responding (or, in the case o f proposed reforms, w i l l respond) to feminist demands for fairer treatment o f elderly w o m e n , and to inform recommendations for moving toward the objective o f achieving an 12  equitable pension system.  This thesis is organized into five chapters. Chapter One describes the historical evolution o f Canada's retirement income system from a needs based welfare program for destitute elderly to the primary source o f income for Canada's elderly. It discusses the important relationship between the pension and income tax systems, and summarizes recent as well as proposed reform o f the pension system.  '' Janine Brodie, Politics on the Margins: Restructuring and the Canadian Women's Movement (Halifax:  Fernwood Publishing, 1995) at 11. l2  Mary Janigan, "Making the Middle Class Pay", 110-39 Maclean's 42 (September 29, 1997).  -5Chapter T w o looks at the economic situation of elderly women in Canada, both statistically and ideologically. Analysis o f the economic security o f women in their senior years requires examination of the commonly held assumption that these women's economic needs w i l l be taken care o f within the private family, and i n particular, by a male spouse. The assumption that elderly women reside within traditional families where, though they may have limited independent resources, their economic needs w i l l be met by their husbands is a myth based on historical construction and bears little resemblance to today's reality. The related assumption that women's employment incomes are supplementary to family wages earned by their husbands is equally misguided.  13  In this chapter I develop the argument that it is necessary to look beyond the surface  of the law to its foundational underpinnings to understand how and why gender neutral legislation can produce a system that is gender biased in its impact. In particular, the ideology o f the family and its relationship to the division between public and private spheres of power and responsibility are examined i n detail.  In Chapter Three I examine the role of familial ideology in the delivery of pension benefits and the consequent systemic gender inequalities. This examination verifies that familial ideology severely restricts, i n some cases to the point o f exclusion, access by women to pension benefits. It also allows the state to download its responsibility for the economic well-being of elderly women to the private sphere where their economic security depends on their real lives fitting a too frequently "false" ideal. The discussion focuses on the impact upon women of primary access to  Jane Lewis, "Towards a Framework for Analysing the Position of Women under Income Security Policies" in E . Diane Pask, Kathleen E . Mahoney and Catherine Brown, eds., Women, the Law and the Economy (Toronto: Butterworths, 1984) 145 at 145; Maureen A . Maloney, "What Is the Appropriate Tax Unit for the 1990s and Beyond?" in Allan Maslove ed., Issues in the Taxation ofIndividuals (Toronto: University of Toronto Press, 1994) 116 at 125129. 13  -6retirement income through work force participation and secondary access through provisions in the pension and income tax legislation that allow women to access retirement benefits not in their own right but as spouses. Drawing on feminist theory, I examine the relationship between familial ideology and the rules in Canada's pension and tax legislation that govern access to retirement income, and identify links between those rules and the over-representation among Canada's poor o f elderly women who do not fit the traditional family 'norm'. This analysis is informed by the analysis i n Chapter T w o o f feminist literature on the ideology o f the family, the elastic boundary between public and private spheres, and the economic effects on elderly women o f relegating 'family' matters to the private sphere.  In Chapter Four I turn to the relationship between the contemporary political climate and the (non)responsiveness o f pension reform to gender inequalities. In particular I focus on the shift in the boundary between public and private associated with the transition from the post war welfare state as it existed in the late 1970s and early 1980s, to the political and economic restructuring o f the late 1980s through the 1990s, and how that shift shaped a reformed pension system that marginalizes the issues o f equality and economic security for women in their retirement years. The late 1980s marked the beginning o f what Janine Brodie refers to as the third period o f major social transformation in Canada, from the welfare state to the neo-liberal state characterized by a transnational capitalism as its governing orthodoxy.  14  Associated with this transformation is  shrinkage o f the public portion o f the public/private continuum and a major transformation o f the 15  social safety net "to make it fit with the market-based, self-reliant, and privatizing ideals o f the new  l4  Brodie, supra note 11 at 16-18.  "Ibid, at 49-63.  -7order."  Interests o f women, along with those o f others who are different from the "ordinary  16  Canadian" , have been marginalized in the neo-liberal state which filters women's issues through 17  reconstructed understanding of equality and o f what is public and what is private, argues Brodie.  18  Drawing on the work o f Brodie and other feminist scholars on the issue o f restructuring, in this chapter I observe that the trend in pension reform is: 1)  toward privatization by increased use o f tax subsidies for retirement savings and employer pension plans (plans that systemically advantage men); and  2)  away from universality of, and toward cutbacks to, public pension programs (programs that, because o f their economic situation, are necessary for the financial autonomy o f elderly women).  I argue that the neo-liberal agenda which has dominated Canadian politics in recent years has transformed feminist and anti-poverty activists' pleas for equal treatment o f women i n pension reform to fit a new social order. A s well, I assert that the new social order has weakened the effectiveness o f equality seeking groups to redress systemic discrimination against women in the retirement income system.  Chapter Five sets out my conclusions on the responsiveness of pension reform to gender equality and considers strategies for progressing towards the goal o f equal treatment for women in the pension system. I tentatively suggest possible next steps.to take to achieve that goal. The first part o f this chapter focuses on the mode o f delivery of pension benefits that would best serve  Ibid. at 57.  ,6  Ibid. at 71.  ]1  ™Ibid. at 65-74.  -8a feminist agenda. The current pension system delivers benefits both through the tax system and through social security programs. I explore how the trend in pension reform to increasing tax system delivery impacts on the economic well-being o f elderly women, and whether gender equality can be more readily achieved through social security programs or tax incentive programs, or a combination of these. The discussion then turns to how to implement the type of program that offers the best hope for progressing toward gender equality. Two types of responses are possible political action and legal action, and these may work either independently or in combination. However, as I shall discuss, neither o f these responses is without problems.  M y thesis is that the retirement income system will continue to be unfair to women, and women w i l l continue to be over-represented among Canada's poor, as long as measures directed at achieving equality fail to respect both the differences among women and the differences between men and women. Pension reform initiatives, like initiatives intended to 'improve' maternity benefits, too often "exacerbate existing avenues o f oppression, thereby contributing to the entrenchment o f patterns o f domination in the larger society" because they make assumptions 19  about all women notwithstanding that these assumptions do not reflect the real lives o f many women. Although some women may benefit, these initiatives are unacceptable because they operate in a way that defeats the feminist goal o f eliminating systemic discrimination which can only be achieved by enhancing equality for all women, not just those who fit a particular stereotype.  l9  Nitya Iyer, "Some Mothers Are Better Than Others: A Re-examination of Maternity Benefits" in Susan B.  Boyd, ed., Challenging thePublic/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto  Press, 1997) 168 at 170. 20  This test for evaluating gender 'equality enhancing' reform is derived from Iyer, ibid.  -9CHAPTER ONE CANADA'S RETIREMENT INCOME SYSTEM - AN OVERVIEW  Canada's retirement income system is supposed to perform two essential tasks. The first is to ensure that elderly people have incomes high enough to allow them to live in dignity no matter what their circumstances were during their working years. The second is to maintain a reasonable relationship between income before and after retirement so that old age does not bring a drastic reduction in a person's standard of living.'  The economic well being o f the vast majority o f Canada's elderly is acutely connected to Canada's retirement income system. For elderly women, whose access to economic resources throughout their lifetimes is more limited than that of men , the connection between economic well 2  being and the pension system is particularly profound. Accordingly, appreciation o f the obstacles 3  to economic autonomy for elderly women is impossible without an understanding o f the current federal pension system and the proposals for its reform. Since the current pension system was not  'National Council of Welfare (Canada), A Pension Primer (Ottawa: Supply and Services, 1996) (hereinafter "Pension Primer") at 1.  In 1980 the poverty rate of women aged 18 and older was 18 per cent - for men aged 18 and older it was 12.7 per cent. In 1995 the poverty rate for women aged 18 years and over was 18.2 per cent - for men it was 14.3 per cent. (National Council of Welfare (Canada), Poverty Profile 1995 (Ottawa: Supply and Services, 1997) (hereinafter "Poverty Profile") at 84.) For a general discussion of the lack of economic well-being of women as compared to men see Maureen Baker, Aging in Canadian Society: A Survey (Toronto: McGraw-Hill Ryerson, 1987) at 86-87. See also Pat Armstrong, "Women's Work: Women's Wages" in Greta Hofmann Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside, 1987) 354 and "Restructuring Public and Private: Women's Paid and Unpaid Work" in Susan Boyd, ed., Challenging the Public/Private Divide (Toronto: University of Toronto Press, 1997) 37. 2  Dennis Guest, in The Emergence ofSocial Security in Canada,!"* ed. (Vancouver: UBC Press, 1997) at 197, states that "[t]he chief problem [with Canada's pension system] for women is that pension schemes in both the public and private sectors were developed with men in mind." The effect, he points out, is twofold - the pension system does not reflect the fact that women's work patterns differ from men's and pension plans tend to view women as dependants of men. Guest concludes his brief discussion of women and pensions with a quote from the Canadian Advisory Council on the Status of Women, As Things Stand, (Ottawa: CACSW, 1983) at 89: "Whether a Canadian woman works in the labour force full-time or part-time, whether she leaves work to raise a family and subsequently returns, or whether she works as a homemaker, she can expect to be poor in her senior years. Pension reform in Canada is an absolute necessity." As noted in Poverty Profile, ibid, at 26, the problem has not disappeared: "Women in the labour force are paid much lower wages than men on average, and they have disproportionately more part-time jobs and fewer full-time jobs than men, so the size of their pensions tends to be small." 3  -10-  created i n its present three tiered form but evolved over time, the past informs the present. Accordingly, a description o f Canada's current pension system must begin with its history.  I  EARLY HISTORY: TO 1951  (a)  The Government Annuities Act. 1908  4  5  State involvement i n providing economic security for the elderly is a relatively recent phenomenon associated with industrialization, urbanization and the wage-labour system which set the stage for the organization, and consequent political empowerment, o f labour. The transition from rural to urban society meant that a family unit was no longer connected to a parcel o f land that could provide for its needs. In this new economy the conflicting interests between those with capital, whose return on investment fell as wages rose, and those who maintained themselves and their families by selling their labour, whose level of well being rose or fell with their wages, created tensions that the state was called upon to mediate. One o f these tensions was the responsibility for the well-being o f those who, because o f age or disability, could no longer sell their labour and pay for their needs. Wages were generally too low for workers to set resources aside for their old age, 6  a problem which was compounded by increasing life expectancies.  The discussion that follows is necessarily a simplification of the historical, economic, social and political events that it describes. The following sources from which I have drawn provide a more comprehensive analysis: Robert L. Brown, Economic Security in an Aging Population (Toronto: Butterworths, 1991); John Burbidge, Social Security in Canada: An Economic Appraisal (Toronto: Canadian Tax Foundation, 1987); Kenneth Bryden, Old Age Pensions andPolicy-Making in Canada (Montreal: McGill-Queen's University Press, 1974); Dennis Guest, ibid.; 4  Anna Shola Orloff, The Politics of Pensions: A Comparative Analysis of Britain, Canada, and the United States  (Madison: The University of Wisconsin Press, 1993); Jane Ursel, Private Lives, Public Policy: 100 Years of State Intervention in the Family (Toronto: Women's Press, 1992). 5  S.C. 1908, c. 5.  Burbidge, supra note 4 at 20-21; Bryden, supra note 4 at 42-43.  6  -11-  In its 1889 report on issues relating to conflicts between labour and capital, the Royal Commission on the Relations o f Labour and Capital appointed by the federal government in 1887 noted the problem o f poverty among the elderly in urban areas. Consistent with the beliefs o f the 7  privileged classes from which members o f Parliament and the Commissioners were drawn, that poverty in old age "could be avoided by the application of thrift", the Commission recommended 8  implementation of a federal annuity program to assist workers in saving for their old age. Although the only one o f the Commission's recommendations for improving the lives o f workers to be implemented was instituting Labour Day as a national holiday, a government sponsored not-forprofit annuities program to help the elderly poor was eventually passed by Parliament in 1908.  9  This program made little difference to the elderly poor, most o f whom did not have the financial resources to purchase an annuity, from the government or otherwise. In 1915 "labourers accounted for only 4 percent o f sales" and by 1927 when the first federal pension legislation was enacted, 10  "only 7,713 annuities were issued"", mainly to people from the lower paid professions such as  7  Bryden, ibid, at 22; Guest, supra note 3 at 21-26.  Guest, ibid, at 26, attributes the slow development of Canada's pension system to this belief: "...the possibility of saving for one's old age, given the low wages and the unemployment, particularly during the winter months, was an illusion that would impede progress toward old age pension legislation for years to come." Recent pension refonri, the subject of this thesis, shows this belief to be alive and well a century later. Legislators did not apply the same "thrift" standard to themselves, however. "In 1905 privy counsellors who had been head of a department for five years or over were granted a retirement pension of $3,500 per annum, and this legislation was made retroactive to include all former cabinet ministers with the required years in office." (Guest at 36.) 8  This scheme was based on the assumption that persons of limited financial means did not trust private insurance companies but would trust the federal government with their savings. Many of the deficiencies of privately purchased annuities were remedied and a better return on investment was offered by the not for profit government scheme. Contributions were voluntary, could be made only up to a fixed lifetime limit per individual, and, except in the case of death before maturity, could be taken out onlyafter the age offifty-fiveand in the form of an annuity. See Burbidge, supra note 4 at 21-22 and Bryden, supra note 4 at 51-53. It is interesting to note that the parliamentary debates which preceded the enactment of this legislation are surprisingly current. 9  l0  Bryden, supra note 4 at 52; Guest, supra note 3 at 37.  "Guest, ibid.  -12teachers, small businessmen and clerks.  12  These represented merely the tip o f the iceberg as far  as the number o f persons in need o f old age security was concerned , and demands by labour for 13  public pensions continued, through the Trades and Labour Congress o f Canada ( " T L C " ) . The T L C did not accept the Government Annuities Act, 1908 as a response to its demands and pressed continuously over the next two decades for an old age pension scheme.  The annuities program set up i n 1908 continued but subscriptions remained limited except for a brief period during the 1930s when its interest rate was more attractive than market interest rates.  14  Political pressure from the insurance industry i n the 1960s, which resented competition i n  selling annuities from a government program, resulted in legislative action that left the legislation in force but rendered it ineffective.  (b)  15  The Old A%e Pensions Act. 1927  u  In response to pressure from provincial and federal legislators who had taken up the cause o f the T L C on the pension issue, Prime Minister W . L . Mackenzie K i n g set up a special committee in 1924 which studied the subject and developed the pension policy reflected in Canada's first  Guest, ibid. Senator Ferguson, who did not oppose the legislation but argued for transfers to the elderly financed by a progressive income tax, or an annuity scheme that was compulsory and subsidized, predicted that "little or no advantage would be taken of [the program] by the people for whom it was mainly designed" because they could not afford to buy annuities no matter how favourable the terms. His prediction proved to be correct. (Bryden, supra note 4 at 52; see also Burbidge, supra note 4 at 22.) l2  l3  Guest, ibid.  l4  Bryden, supra note 4 at 53-59.  Ibid. See also Government Annuities Improvement Act, S.C. 1974-75-76, c. 83 which increased the rate of return on existing Government Annuity contracts but discontinued future sales. ]5  I6  S.C. 1926-27, c. 35.  -13public pension legislation, The Old Age Pensions Act, 1927 (the "1927 A c t " ) .  17  A t the time, a  number o f working models from jurisdictions that had pension programs in place were available for study, including means-tested, contributory and universal plans.  18  The model favoured by most  jurisdictions was a contributory one. Canada chose a means tested model. Bryden suggests that the committee was most influenced by the models in place in Britain, Australia and N e w Zealand, which were means-tested. In addition Bryden speculates that, in recognition that it was "breaking 19  new ground" politically, the committee was careful that its proposal 20  ...keep the estimated costs within limits which it hoped would be acceptable. That automatically ruled out a universal plan financed out of general revenues. ...A contributory plan appeared to be almost equally impractical. The administrative problems, which were complicated enough in a unified plan, appeared in the 1920s to be insuperable in a joint [federal/provincial] plan. 21  Burbidge points out the direct political advantage o f a non-contributory plan - it could take effect immediately while a contributory plan would mean a delay in commencing benefits.  22  Guest also  considers that the political climate was not ripe for a universal program at this time and attributes the choice o f a non-contributory plan to constitutional complications.  23  This is an extremely simplified summary. World War I and the social, economic and political forces and events that followed in its aftermath also played an important role. See footnote 4 for references to a more comprehensive account of events during this interesting period that led to Canada's first public pension legislation. l7  18  Bryden, supra note 4 at 75-78.  "In 1925, the year after the committee gathered its information about the pension plans in other countries, the United Kingdom added a contributory plan to its means-tested plan. See Bryden, ibid, at 75. 20  Bryden, ibid, at 77.  Ibid.  2,  22  Burbidge, supra note 4 at 23-24.  23  Guest, supra note 3 at 75-76.  -14There is little doubt that Canada's constitutional division of powers was a key element in shaping the type o f pension program embodied i n the 1927 A c t . Since pensions were generally regarded as falling under the "matters o f a merely local or private nature" , the 1927 A c t simply 24  "authorized the federal government to enter into an agreement with any province to reimburse that province for half the cost o f pensions paid under provincial legislation which met the requirements of the federal act and regulations."  25  Each province had to implement and administer its own plan,  subject to federal approval, which took the form o f federal-provincial agreements that could not be terminated by the federal government unless it gave the province 10 years notice. The province could only terminate the agreement by repealing its legislation.  26  To receive the federal grant, the provincial pension plan had to be restricted to persons aged 70 years or older, who were either British subjects or widows of aliens who were British subjects, and who had been resident in Canada for at least 20 years, the immediately preceding five years of which they must have been resident in the province paying the benefits. Indians as defined by the Indian Act were specifically excluded. F u l l participation o f the provinces was not achieved until 1936, which meant lower pensions in the case o f pensioners who had lived for a portion o f  Constitution Act, 1867 (U.K.), 30 & 31 Vict., c. 3, subs. 92(16). This categorization meant that pensions were considered to be within the legislative competence of the provinces, a view that was confirmed in In the Matter 2i  of a Reference as to Whether the Parliament of Canada had Legislative Jurisdiction to enact the Employment and Social Insurance Act, being Chapter 38 of the Statutes of Canada, 1935, [1936] S.C.R 427; affirmed sub nom Attorney-General for Canada v. Attorney-General for Ontario, [1937] A.C. 355 (P.C.). See Burbidge, supra note  4 at 23 note 33. 2S  Bryden, supra note 4 at 61. The federal contribution rate was increased to 75 per cent in 1931. (Bryden  at 74. See also An Act to Amend the Old Age Pension Act, S.C. 1931, c. 42, s. 1.) Ibid.  26  -15the qualifying 20 years in a non participating province.  27  In addition, recipients had to meet an onerous means test. Persons with income i n excess of $365 per year did not meet this test. Income included annuities, five per cent o f the value o f 28  real property whether or not such income actually existed, the amount of a government annuity that the value o f personal property could have purchased (a reasonable amount for clothing and furniture was excluded), and "other means o f livelihood in cash or kind which pensioners 'may reasonably be expected to receive."' Wealth o f the pensioner's children could be included in the 29  latter category, regardless o f whether the pensioner received income from that source.  30  For the  purpose o f computing income, a married pensioner was deemed to have received one-half o f the couple's combined income.  The amount o f the pension was $240 per year, reduced dollar for  dollar by any amount that the pensioner's yearly income, including certain types o f imputed income as noted above, exceeded $125. The provinces could include i n their plans provisions for recovery from deceased pensioners' estates o f up to the full amount o f the pension payments made plus interest at five per cent.  31  A paying province was required to pay pro rata to the number of qualifying years that the pensioner resided in the province and was entitled to pro rata reimbursement from any other province(s) in which the pensioner resided during the 20 year qualifying period. See Bryden, ibid.at 62. 27  According to Bryden, ibid., this figure was based on "an empirically unverified assumption prevalent in government circles that an old person required a minimum of $ 1.00 a day to live." This makes an interesting comparison with the $3,500 per annum pension for cabinet ministers referred to supra at note 8. Bryden at 117 notes that "[t]he 1927 plan offered no more than protection against complete destitution at an advanced age." 28  29  Bryden, ibid, at 63.  30  Burbidge, supra note 4 at 23.  Ibid.  }]  -16The 1927 Act, as amended periodically, was Canada's pension system until 1951. Although it was far from perfect,  32  Canada's first pension scheme was successful in shifting the focus on  social assistance for the elderly from a local concern to an issue o f national importance.  33  National  coverage was achieved by 1936 and the qualifying conditions for federal funding imposed a high degree o f national uniformity. Further, through conditions that required the provinces to enter into agreements with the federal government and which made those agreements difficult to amend, the scheme effected a degree o f stability that, in turn, reinforced the value o f a national pension program.  "Problems included adininistrative inconsistencies related to divided federal-provincial jurisdiction and difficult to meet qualifying conditions that excluded many who needed financial assistance. For general discussion see Bryden, supra note 4 at 81-104; Guest, supra note 3 at 77-79; Burbidge, supra note 4 at 23. Ursel, supra note 4 at 160-163 discusses the bureaucratic expansion and policing aspects of the means tested pension scheme embodied in the 1927 Act and their patriarchal underpinnings. The provinces administered the pension program along with other means tested welfare programs. Welfare officers, whose function was to determine who was or was not eligible for benefits, increased in number. They V  ...operated as investigators of family resources, locating husbands who could afford maintenance and adult children who could afford to support their elderly parents.... Built into the delivery system of these bureaucracies was a continued commitment to the sexual division of labour, which assigned women the primary role of child care and men the primary role of breadwinner. ... [W]elfare law evolved to provide support to those who, through no fault of their own, were without a breadwinner. The welfare officer became the 'provider' for recipient families, claiming for itself the same level of scrutiny and control accorded to the traditional patriarch. (Ursel at 164-165). "Bryden, ibid, at 81-104; Guest, ibid, at 77-79. According to Orloff, supra note 4 at 267 the 1927 Act "had paved the way for Canada's version of a modern welfare state." On the other hand Ursel, ibid, at 167 sees the 1927 Act as a very limited foray into the pension arena by the federal government whose "role was largely limited to paying the bill."  -17-  II  EVOLUTION OF THE THREE TIERED SYSTEM:  (a)  1951: Old Age Security Ac? and Old Age Assistance Ac? 4  1951 TO 1970s  5  M a n y o f those who had pushed for a federal pension scheme were unhappy with the 1927 A c t from the beginning and calls for change followed closely on the heels o f its enactment.  The  most pressing issues were that the age requirement o f 70 years was considered too high, the amount of the pension was too low, and the means test was both humiliating (in the scrutiny into personal affairs associated with its enforcement) and discouraging to "the spirit o f thrift and self reliance" in its dollar for dollar reduction o f an individual's efforts to provide for old age.  37  36  Through the  depression years more compelling matters moved pensions to the political background, but renewed interest by both the Liberal government and Conservative opposition as well as organized labour accompanied the revitalized economy during and after W o r l d War II.  Although  38  amendments to the 1927 A c t could respond to demands to increase pension amounts, changing 39  the age qualification and means testing requirements meant a restructuring o f the program.  34  S.C. 1951 (2 Sess.), c. 18.  35  S.C. 1951, c. 55.  36  Bryden, supra note 4 at 104.  nd  "Ibid. Bryden, ibid, at 109 attributes the post war interest in government responsibility for social security programs by all political parties to "the Keynesian revolution in economics" which showed promise for reconciling the competing interests between the market and "the needs of the urban-industrial society." 38  Though according to Bryden, ibid., such amendments merely adjusted the amounts to increases in the standard of living so there was no real increase in pension amounts. 39  -18-  Debate - 'universal' or 'contributory'? Means testing became so politically unpopular in the late 1940s that the pension reform debates were not over whether the means test should be abolished, but whether the reformed pension system should be 'universal' or 'contributory'.  40  These two types o f programs were  founded on very different principles o f social policy. Proponents o f universal pensions included the political party associated with the interests o f labour, the Cooperative Commonwealth Federation ( C C F ) , and the second largest trade union in Canada, the Canadian Congress of Labour ( C C L ) . They considered a basic level of social security to be "a fundamental human right" which 41  the federal government had a responsibility to provide "free from humiliating means tests".  42  Proponents o f contributory pensions, including both the governing Liberals, the Conservative opposition as well as some trade unions, held the more conventional view that a pension system that links benefits to contributions "stimulates, from an early age, the person who w i l l later benefit thereby, to economize and to take thought for his (sic) old age."  43  Although a universal pension  system would be fairer to women than a contributory system that links benefits as well as contribution levels to wages, gender issues had little, i f any, impact on the debate. In the end the choice was determined by the division o f powers in the Canadian constitution, which was as much o f an obstacle to contributory pensions in the early 1950s as it was i n 1927. While constitutional  Bryden notes that the terms 'universal' and 'contributory' are not used consistently by the proponents and opponents of the two types of plans. I follow Bryden in "classifying as 'universal' any non-means-tested plan in which contributions are not a condition of benefits..." regardless of whether the means of financing is from general revenues, ear-marked revenues or other means of financing, and in classifying as'contributory' any non-means-tested plan "...in which receipt of benefits is conditional on prior contributions." (Ibid, at 110.) 40  41  Bryden, ibid, at 113.  Ibid.  n  Ibid. at 108, quoting from the Quebec Social Insurance Commission, Fifth Report p. 149.  41  -19experts agreed that pensions financed by the federal government out o f general revenues did not infringe on provincial constitutional powers, contributory pension legislation w o u l d .  44  Since the  required constitutional amendment would delay pension reform and political pressure to repeal 45  the means test, which both 'universal' and 'contributory' proponents agreed was intolerable, was immediate, a stereotypical Canadian solution was reached - compromise.  46  T h e 1951 P r o g r a m The Old Age Security Act* ( " O A S Act") and the Old Age Assistance Act ( " O A A Act") 1  48  which were passed in 1951 and came into force on January 1, 1952 formed the legislative foundation o f the new pension program ("the 1951 Program"). The O A A responded to the demand  "Ibid, at 123.  As the British North America Act, 1867, 30 Victoria, c. 3 (the "BNA Act") did not contain a provision for its own amendment, constitutional amendments had to be enacted in the same way as the BNA Act itself, by the United Kingdom Parliament. By convention the United Kingdom Parliament could not enact an amendment to the BNA Act except at the request of Canada. For amendments that directly affected provincial powers, a practice developed that unanimous consent of the provinces was a pre-requisite. (See Peter W. Hogg, Constitutional Law of Canada, 3 ed. (Toronto: Carswell, 1992) at 61-64.) 45  rd  I do not mean to suggest that the constitutional impediment alone induced the giant step from a means-tested to a universal pension system. Obviously the compromise would not have been made if the political, social and economic climates were antagonistic to universality. Guest, supra note 3 at 137, attributes the enactment of the 1951 pension scheme to "intensive lobbying by pensioners' organizations and labour unions." Bryden, supra note 4 at 109110, sees the change rooted in the aftermath of the "greatest war in history, following hard after the greatest depression in history" and the "Keynesian revolution in economics" which viewed governments as having "a major potentiality and responsibility for promoting economic growth and full employment and for providing wide-ranging social security." Ursel brings a feminist perspective to history of the post war period during which the federal government, in its role as mediator between labour and capital at a time when the balance between these two interests approached equilibrium, took on the responsibility for satisfying the demands by labour for a "living wage" that capital resisted. Subsidies to families through such social programs as universal family allowances and old age pensions were emanations of state policies "that would at one and the same time cushion capital from the growing strength of labour and increase resources to the reproductive unit." (Supra note 4 at 226.) While an historical analysis of this period is beyond the scope of this thesis, the sources I have referred to show that the transition to a universal pension system resulted from the convergence of many factors, the influence of which are subject to many interpretations. 46  41  Supra note 34. This Act repealed the 1927 Act. Supra note 35.  1  -20for lower age qualification for pensions; it followed the model o f the 1927 A c t and provided for federal financial contributions, from general revenues, to provincially administered means tested pensions for qualifying persons between the ages o f 65 to 70 years. The federal cost sharing was reduced from 75 per cent under the 1927 A c t as amended in 1931 to 50 per cent, with the maximum federal contribution set at $40 per pensioner per month. The O A S effected the objective o f eliminating the means test by providing a basic pension o f $40 per month to all persons over the age o f 70 years who met the residency test, and, unlike its predecessor, the 1951 Program benefits 49  were extended to Indians as defined in the Indian Act.  The O A S component o f the 1951 Program was a marked departure from its predecessor pension plan. It was administered by the federal government and fully financed from a separate fund raised by "earmarked taxes" known as the "2-2-2 formula" because o f the two per cent increases in personal income tax, corporate income tax and federal sales tax imposed to offset the costs o f the universal pension.  50  The earmarked taxes represented another part of the compromise  that 'contributory' proponents, including the Liberal government o f the day, made to expedite a response to the hated means tested pension scheme. The St. Laurent government did not want Canadians to lose sight o f the financial impact on them o f the universal pension scheme, hoping to chill the climate for further demands to extend public pensions.  51  Although this solution had  some elements of a contributory plan, benefits were not tied to contributions and 'temporary loans'  Bryden, supra note 4 at 104-105. The residency test was substantially retained from the 1927 Act as amended in 1947. The 1947 amendment permitted absences during the 20 year residency period to be offset by prior residence of double the duration of the absence. The OAS added the requirement that an applicant must reside in Canada for the year prior to approval of an application for the universal pension. 49  Ibid. at 105; Burbidge, supra note 4 at 24.  50  51  Bryden, supra note 4 at 105; Guest, supra note 3 at 137.  -21from general revenues could be made to the special fund i f necessary.  The choice o f financing the scheme through an earmarked tax meant that the scheme fell within a grey area and was neither clearly within federal legislative competence as a scheme financed out o f general revenues would be, nor clearly within provincial competence as a contributory scheme would be. Constitutional experts advised that the federal government could not avoid a constitutional amendment, though a much narrower one than would be required for a fully contributory scheme.  53  Premier Duplessis o f Quebec objected to the original wording o f the  amendment but his concurrence was obtained with wording that assured provincial paramountcy to his satisfaction.  54  The key achievement o f the 1951 Program was that, for those over 70 years, it "removed once and for all the sense o f personal failure that accompanied acceptance o f an old age pension under the original act."  55  However, the O A A component meant that those between 65 and 70 years  were subjected to the same humiliating intrusions into their affairs that made the 1927 A c t so unpopular. Further, although the 1951 Program may have been the beginning o f a social safety net for elderly Canadians, as noted by Burbidge it provided only a very basic level o f income security and consequently resulted in little progress toward replacement income sufficient to maintain pre-  52  Bryden, supra note 4 at 105.  Ibid. at 123-124; Guest, supra note 3 at 137.  5i  Ibid. The resulting amendment was enacted by the British North America Act, 1951, 14-15 Geo. VI, c. 32  s4  and became s. 94A of the BNA Act. It was repealed and replaced by a subsequent amendment in 1964 (see note 78, infra). (The BNA Act is now the Constitution Act, 1867.)  "Guest, supra note 3 at 137.  -22retirement life styles for middle class retirees.  (b)  56  1957: Tax Assisted 'Private' Plans  RPPs W h i l e all Canadians who met the age and residency requirements could anticipate entitlement to a basic level o f retirement income from the 1951 Program, some were fortunate enough to be able to look forward to some measure o f income replacement on retirement through private employment pension plans, private savings or both. Employment pension plans "were created to further a company's corporate goals o f inspiring cooperation among employees, raising morale and efficiency, cutting labour turnover, and inducing the retirement o f older workers."  57  Starting in 1919 employer sponsored pension plans that met certain registration requirements prescribed by income tax legislation ("RPPs") were given favourable tax treatment. The usual 58  59  60  Burbidge, supra note 4 at 24. As noted in the Pension Primer, supra note 1, one of the two goals of a pension system is to provide for consistency between pre- and post- retirement standard of living. 56  "Brown, supra note 4 at 55, quoting from Morton and McCallum, Report: Task Force on Inflation Protection for Employment Pension Plans Vol. 2 (Toronto: Queen's Printer for Ontario, 1988). "Income tax was first legislated in Canada in 1917 by Thelncome War Tax Act, 1917, S.C. 1917, c. 28.S.C. 1917, c. 28. Employer sponsored registered retirement plans, RRPs, came in a variety of forms but are generally of two types. Defined benefit plans provide an employee with a contractually defined benefit on retirement, usually a percentage of the employee's average annual wages in the last few years of his or her working life. This was the most popular type of RPP until recent years. The second type, defined contribution plans, are characterized by employer and employee contributions in an amount contractually agreed upon, and the pension benefit the employee will get on retirement is the amount of pension annuity that can be purchased with the contributions that have accumulated during the employee's working life. A third type of employer sponsored retirement scheme, deferred profit sharing plans ("DPSPs") was given similar preferred income tax treatment in 1961. Contributions to DPSPs are made by employers only and contribution amounts are not fixed but fluctuate according to the employer's profit. The effect on retirement is similar to a defined contribution plan however; the pension amount depends on the amount of annuity that can be purchased with the accumulated contributions. 59  Peter W. Hogg and Joanne E. Magee, Principles of Canadian Income Tax Law, 2 ed. (Toronto: Carswell, 1997)at357. See also An Act to amendThe Income WarTaxAct, 1917, S.C. 1919, c. 55, subs. 2(4) which permitted a taxpayer to deduct from income for tax purposes "[a]ny part of the remuneration of a taxpayer retained by his employer in connection with an employee's superannuation or pension fund or plan". Payments to the employee out 60  nd  -23arrangement for these plans was that both the employer and employee made contributions; both contributions were tax deductible, employer contributions were not included in employee income as a taxable employment benefit, and investment income earned on contributions was not taxed while it remained i n the plan. The importance of RPPs grew as labour unions became increasingly interested in this employee benefit in both the public and private spheres during the 1940s and 1950s.  RPPs are obviously available only to work force participants. Since employers are not compelled to provide pension plans for employees, only a portion o f work force participants ' 6  receive this employment benefit, usually those employed either by large companies with a unionized labour force or in the public sector.  RRSPs It was not until almost forty years after tax incentives to encourage R P P s were legislated that the federal government implemented similar incentives for contributions to individual savings invested in registered retirement savings plans ( " R R S P " s ) .  62  In 1957, recognizing that only part  of the work force could access employer sponsored pension plans, the federal government amended  of the fund or plan had to be included in the employee's taxable income. 6  1983.  'In 1993 only 35 per cent of the paid labour force were RPP members, roughly the same percentage as in  (Statistics Canada, Canada's Retirement Income Programs: A Statistical Overview (Ottawa: Statistics  Canada 1996) at 11 (hereinafter "Statistical Overview").) "Employer sponsored pension plans are subject to complex regulation under either provincial or federal legislation, depending on whether the employment falls under federal or provincial jurisdiction. Although some of the features of RPPs will be noted in the discussion of this third tier of Canada's retirement income system, analysis of such pension plans in any detail is beyond the scope of this thesis and the discussion will focus on RRSPs that are similar to defined contribution pension plans in the type of benefit received on retirement.  -24the Income Tax Act (the " 1 9 5 7 I T A amendments") to provide workers not covered by employer 63  pension plans with the opportunity to achieve economic security in their retirement years.  64  The  concept was simple - through R R S P contributions taxpayers could distribute their incomes more evenly over their lifetimes by deferring a portion o f it from their working years to their retirement years. The 1957 I T A amendments set a formula for determining the portion o f income that could be contributed tax free to an R R S P and provided for tax free income earned on qualified R R S P investments.  65  The potential tax savings were significant. Contributions are generally made in  relatively high income earning years (ie. during the contributor's working years) when, because o f Canada's progressive rates o f taxation, they would be taxed at a relatively high rate, and withdrawals are generally made in low income earning years (ie. retirement years) when they w i l l be taxed at a lower rate. In addition, unlike other investment income, which is subject to income tax, income earned on R R S P investments is completely sheltered from taxation when it is earned and is taxed only when withdrawn. The effect o f deferral itself is a further tax advantage - the taxpayer gets the use o f money that would otherwise go to the government as tax, and when tax is eventually paid, assuming inflation over time, the payment w i l l be made with deflated dollars. In  Hereinafter the Income Tax Act, R.S.C. 1985, c. 1 (5 Supplement), whether current or predecessor legislation, is referred to as the "ITA". Amending legislation specifically referred to will be cited in full. 63  lh  An Act to Amend the Income Tax Act, S.C. 1957, c. 29, s. 17. This provision, as amended, is currently s. 146 of the ITA. Although the legislation did not limit the benefit of making RRSP contributions to workers not covered by employer pension plans, the formula for determining contribution limits was designed with retirement income security for this group in mind. (See Brown, supra note 4 at 73.) M  The formula for calculating a taxpayer's maximum RRSP contribution was, until recent reforms were enacted in 1991,20 per cent of earned income up to a prescribed dollar maximum that was increased regularly to 1976 and then frozen until 1986. The maximum contribution allowed from 1976-1986 was $5,500 for taxpayers who were not RPP members, and for RPP members it was $3,500 minus the amount of employee contributions made to the RPP. It should be noted here that the tax incentive formula for RRSP contributions introduced in 1957 and maintained until 1991 did not effect equality of potential retirement benefits between RPPs and RRSPs. In particular defined benefit RPPs were accorded most favourable tax treatment. 65  -25effect, the deferred tax is an interest free loan to the taxpayer from the government.  A taxpayer may choose to contribute to a spousal R R S P rather than an R R S P in her or his own name, although the taxpayer's contribution limit remains the same regardless o f where the contributions are directed. This provision is intended to encourage the higher income earning spouse to provide for retirement needs o f the dependant non or low income earning spouse. For taxpayers who are married or are in a common law heterosexual relationship, a spousal R R S P 66  offers an added tax advantage. It allows R R S P withdrawals to be allocated as income to the lower income spouse and taxed at a lower marginal rate.  In addition to tax savings to the R R S P contributor and a consequent lessening o f the potential political pressure to improve public pensions for retirees who do not have access to employer sponsored pensions, R R S P s served a second politically important function. The money invested i n R R S P s provided risk capital for economic development.  67  This dual function explains  why R R S P s , despite the name, were not restricted to retirement savings. R R S P investments can be 'cashed i n ' at any time, although when they are withdrawn the amount must be included in the taxpayer's income for the year o f withdrawal and is taxed accordingly.  The obvious limitation o f RPPs and R R S P s as vehicles for providing retirement income is their inaccessibility for a large portion o f the population. Employers are not required to provide  Subs. 252(4) of the ITA defines "spouse" as including "a person of the opposite sex who cohabits with the taxpayer in a conjugal relationship and (i) has so cohabited with the taxpayer throughout a 12-month period ending before that time, or (ii) is a parent of a child of whom the taxpayer is a parent..." 66  67  This is true of RPPs as well.  -26R P P s and only some do, and R R S P s are available only to those with sufficient disposable income to be able to make a contribution. Even among those who have money available to invest in R R S P s , higher income earners gain greater tax savings than lower income earners. This is because o f progressive tax rates. A $5,000 contribution results in a tax reduction o f $1,250 for a taxpayer whose low taxable income places her in a tax bracket that has a 25 per cent federal/provincial marginal rate, while the same contribution results in a tax reduction o f $2,000 for a taxpayer whose earnings place h i m  68  at a 40 per cent marginal tax rate. The higher earner thus has an additional  $750 left in his pocket.  Although RPPs and R R S P s are frequently referred to as private pension plans, strict statutory requirements must be met in order to qualify for the significant tax savings associated with the deductibility o f contributions and the sheltering from tax o f the income earned on the contributions. Accordingly, this third tier o f Canada's retirement income system is both publicly regulated and funded. Perhaps the 'private' designation derives from the following distinctions between this tier and the others: •  contributions are not compulsory but are made at the initiative o f individuals (for R R S P s ) or in compliance with privately negotiated collective agreements (RPPs);  •  no public funds are expended unless contributions are first made from private funds (ie. the state helps those who help themselves); and  •  the expenditure of public funds takes the form o f leaving money in the taxpayer's pocket, by reductions in the amount o f taxes that would otherwise be payable, rather than making  In keeping with the political nature of this thesis, at times my use of gendered pronouns (he/she, him/her) reflects a deliberate choice for its political effect. 68  -27a transfer from collected revenues. Regardless o f its origin, the 'private' label hides the fact that public monies are being applied to subsidize retirement savings for high income earners to an income replacement level that is much higher than available under other forms o f retirement assistance provided from public funds.  The tax incentives introduced in the 1957ITA amendments for registered individual savings plans made tax subsidized 'private' schemes theoretically available to all income earning Canadians, provided economic realities are ignored. Thus, the third tier o f Canada's retirement income program took its shape, and it continued i n essentially the same form until its reform by the 1990 amendments to the I T A .  (c)  69  1965: Canada Pension Plan  70  Although the 1951 Program was achieved through political compromise, the concept o f a universal pension plan and its corollary o f entitlement to a basic level o f income for the elderly quickly embedded itself into the Canadian identity. A s expected o f a compromise solution, there was considerable criticism.  Fiscal conservatives maintained their affection for contributory  programs while those who supported universality were critical o f the means test for the 65-70 age group and wanted the pension amounts increased. Towards the late 1950s, as the operation o f the 1951 Program continued to reveal its weaknesses, the idea of adding an earnings related component to the pension program gained general popularity and support from Canada's three mainstream  'Discussed below at pp. 44-46. 'Enacted by Canada Pension Plan, S.C. 1964-65, c. 51 and currently cited as R.S.C. 1985, c. C-8.  -28political parties.  71  Labour interests came to realize that those who earned incomes above a  subsistence level encountered a substantial decrease in standard o f living on retirement and that private pension plans could not be counted upon to alleviate this problem. O n the other hand, those who would have preferred a fully contributory pension scheme recognized that discarding a plan that guaranteed a minimum level o f financial security for the elderly would not be politically expedient.  This common interest in superimposing a contributory component on the existing  universal pension scheme precipitated a major restructuring of Canada's pension system in the m i d 1960s, effected by enactment o f the Canada Pension Plan in 1965 and revision o f the O A S A c t 72  in 1966 . Tiers I ( O A S and related programs) and II (Canada Pension Plan) o f the pension 73  program that exists today have their origins in the restructuring that occurred in the 1960s.  The Canada Pension Plan ("CPP") was established for the benefit o f members o f the paid labour force and their dependants. It "impose[d] contributions and providefd] old age, survivor, and disability pensions which are identifiable though not precisely related to income earned in employment."  74  It was widely accepted that maintenance o f a continuous standard o f living from  employment to retirement requires retirement income amounting to between 50 and 80 per cent o f  7l  Bryden, supra note 4 at 137-144.  Supra note 70. This Act, as amended from time to time, is referred to hereinafter as the "CPP Act". Quebec opted out of the CPP, wishing to emphasize its special status in Canada by enacting sister legislation implementing the Quebec Pension Plan (QPP) as Quebec's equivalent to the CPP. The CPP and QPP remained essentially the same until 1984. For reasons of simplicity and brevity, discussion of the contributory component of Canada's pension program in this paper is limited to the CPP. It is important to note however that Quebec's determination to have its own pension plan had a significant impact on the CPP. The federal government wanted to achieve national uniformity and accordingly negotiated with Quebec, resulting in both the CPP and QPP becoming "more adequate and comprehensive than initially envisioned by Ottawa or Quebec City." (Guest, supra note 3 at 143. See also Bryden, supra note 4 at 162-175 and Burbidge, supra note 4 at 17-18 and 25-26.) 12  An Act to amend the Old Age Security Act, S.C. 1966-67, c. 65.  n  74  Bryden, supra note 4 at 133.  -29pre-retirement income.  75  A t the time the C P P was enacted, Tier I ( O A S ) benefits amounted to  approximately 20 per cent o f the average industrial wage ( " A I W " ) .  76  Legislators deliberately  limited Tier II (CPP) retirement benefits to a maximum o f 25 per cent o f the A I W , leaving a gap to be filled by Tier III (RPPs and R R S P s ) .  77  The C P P program was phased in over a ten year  period; full benefits were not payable until 1976. Since the C P P provided more than retirement benefits, the 1951 constitutional amendment was not broad enough and a further constitutional amendment was required.  78  The following is a brief summary o f key elements:  Financing  79  The C P P was designed as a modified pay-as-you-go system. During the first ten transition years contributions exceeded the benefits paid out to build a surplus equivalent to approximately two years o f benefits. F r o m 1976, contributions plus interest earned on the reserve funds  7S  80  were  Brown, supra note 4 at 68-73.  Ibid. at 39. Statistics Canada calculates the average aggregate wage for workers in Canadian industries on a weekly basis. The average industrial wage is the average of these weekly statistics for a calendar year. In 1997 the annual industrial wage was approximately $35,400. (Department of Finance Canada, News Release 97-083, "Canada Pension Plan Tabled" (25 September 1997).) n  "Guest, supra note 3 at 199-200. Ibid. at 143. The 1951 constitutional amendment was limited to old age pensions; the CPP included survivor, death and disability benefits. The required amendment was enacted by the British North America Act, 1964 (U.K.), 12-13 Eliz. II, c. 73. It replaced s. 94A (added by the 1951 amendment - see supra, note 54) and enhanced the federal power to legislate in relation to old age pensions to include "supplementary benefits, including survivors, and disability benefits irrespective of age". Unlike the 1951 amendment, the 1964 amendment was not controversial and was obtained expeditiously. n  79  CPP Act, supra note 70, ss. 108-113.  ""Surpluses were available as loans to the provinces at attractive interest rates. (Bryden, supra note 4 at 136137; Guest, supra note 3 at 143.)  -30intended to maintain the two year equivalence i n surplus funds and to offset benefits paid. Contributions were levied as a special payroll tax and were mandatory on all employee income i n respect o f employees between the ages o f 18 and 65 years,  81  including self employed, above a  minimum level statutorily defined and referred to as the Year's Basic Exemption ( " Y B E " ) up to 8 2  a maximum statutorily defined by formula and referred to as the Year's M a x i m u m Pensionable Earnings ( " Y M P E " ) .  8 3  Employers and employees made equal contributions, with the self employed  paying both the employer and employee shares.  84  Contributions were deductible i n computing  taxable income, until 1988 when the employee subsidy was converted to a tax credit.  Retirement Pension Benefits  85  86  C P P benefits are based on a contributor's lifetime employment earnings and C P P contributions, and are intended to replace 25 per cent o f the contributor's earnings at retirement, up to a maximum that corresponds approximately to the A I W . Eligible persons who apply for C P P retirement pension benefits receive monthly payments o f 25 per cent o f their "average monthly pensionable earnings", an amount based on the pensioner's average lifetime earnings with various  Persons between the ages of 65 and 70 may, at their option, continue to make contributions. (Brown, supra note 4 at 42.) 81  CPP Act, supra note 70, s. 20. In 1966 the exemption was $600; in 1997 it was $3,500. (Brown, supra note 4 at 44-45; Finance Canada, News Release 97-083, supra note 76.) 82  83  CPP Act, supra note 70, s.18. The YPME was and is indexed to the AIW. (Ibid.)  CPP Act, supra note 70, ss. 13 and 14. In 1966 the contribution rate for each employee and employer was 1.8 per cent of the employee's YMPE; in 1997 it was 2.925 per cent. (Ibid.) 84  * An Act to amend the Income Tax Act, S.C. 1988, c. 55, s. 92, which added section 118.7 to the ITA. See also Pension Primer, supra note 1 at 19. A tax deduction gives a greater tax benefit to high income earners, whose marginal tax rates are higher than those of lower income earners. A tax credit gives the same tax benefit to all taxpayers regardless of level of income. s  86  CPP Act, supra note 70, ss. 44-53.  -31allowable adjustments, YMPE.  87  up to a maximum based on the average o f the previous three years'  Retirement pension benefits based on fewer than ten years o f contributions are reduced  8 8  proportionately.  89  Once entitlement to C P P retirement pension is established, the pension amount  remains static, except for annual cost o f living adjustments.  Eligibility for Retirement Pension Benefits  90  Initially the qualifying age for C P P retirement pension was 70 years, reducing annually by one year until 1970 when it became 65. U n t i l 1974 a retirement test was applied to pensioners between the ages o f 65 to 70, which had the effect o f reducing pension amounts payable for pensioners who earned annual income above $960. Currently the age requirement for full C P P retirement pension is 65 years, regardless o f whether or not the person has stopped working. Persons who want to retire early may, i f they are between the ages 60 and 65 and have stopped working, elect to take C P P pension benefits at rate reduced by 0.5 per cent per month o f early retirement. Alternatively late retirement may be taken between the ages o f 65 to 70, with monthly payments increased accordingly. There is no residence requirement; C P P benefits are payable to pensioners within or outside o f Canada.  Ineligible persons are those who have not earned  employment income equal to 10 per cent o f the average industrial wage, including persons who  "Provided that the adjustments will not reduce the number of months of pensionable income below 120 (ie. 10 years), up to 15 per cent of months worked before the age of 65 can be disregarded. Months of pensionable earnings after age 65 can either be substituted for before age 65 months or disregarded. Amendments in 1983 (retroactive to 1978) added special "child-rearing" dropouts - in addition to the 15 per cent dropout provision, a parent who stays home to care for a child under the age of seven years may exclude those years from the calculation of lifetime earnings. (Bryden, supra note 4 at 135-136 and Pension Primer, supra note 1 at 26-30.) The maximum monthly CPP retirement pension in 1997 was $736.81. (Finance Canada News Release 97083, supra note 76.) 88  89  Bryden, supra note 4 at 135.  90  CPP Act, supra note 70, ss 42-44 and definition of "contributor" in subs. 2(1).  -32were never or seldom employed, and those who worked but for very low wages.  Persons who work outside the paid labour force, the largest group o f whom are women who work in the home, are excluded from making contributions or receiving benefits other than benefits arising from their dependency relationship to a contributor.  Women are also over-  represented among part-time paid workers and those whose work force participation is interrupted, as well as among those whose wages fall below the A I W .  Thus, too often a woman's paid  working life falls short o f the standard working life, based on a male norm, that is used to calculate the amount o f C P P retirement benefits payable to a pensioner.  Survivor Benefits  91  A benefit o f 60 per cent o f a deceased C P P contributor's retirement pension is payable to his or her surviving spouse aged 65 years or above. If the surviving spouse is between the ages 92  o f 45 to 65, is disabled, or has dependent children, she or he is entitled to a benefit equal to a flat rate calculated by a formula prescribed under the C P P A c t and 37.5 per cent o f the retirement pension to which the deceased was entitled. If the surviving spouse is between the ages o f 35 and 45, is not disabled and has no dependent children, she or he is entitled to a reduced version o f the pension that a 45 year old surviving spouse would receive. A surviving spouse under the age o f 35 and without disability or dependant children is not entitled to a benefit until she or he reaches the age o f 65 years or becomes disabled. Initially benefits ceased on remarriage but this provision  9I  CPP Act, supra note 70, ss. 58 and 59.  92  From 1986 the meaning of "spouse" under the CPP Act was extended to include a common law spouse.  (An Act to amend the Canada Pension Plan and the Federal Court Act, S.C. 1986, c. 38, subs. 1(3).)  -33was repealed in 1986. If the surviving spouse receives C P P benefits o f her or his own, these are 93  integrated with the survivor benefits to a statutory ceiling.  Other Benefits Other C P P benefits include death benefits,  (d)  94  disability benefits  95  and child's benefits.  96  1966 and 1975: Additions to Tier I  1966: Guaranteed Income Supplement To complement the addition of a second tier to the pension system a number of amendments were made to the first tier, the O A S and O A A . The most significant o f these was in 1966 when a means tested component, the guaranteed income supplement ("GIS") was added to the basic universal pension.  97  The GIS was intended to supplement the O A S during the ten year period that  the C P P was being phased i n ,  98  the belief was that the O A S and C P P would provide adequate  retirement income and that when full C P P benefits became payable, the GIS would no longer be  Ibid., s. 30.  n  A lump sum death benefit, related to the amount of the contributor's retirement benefits, is paid to the estate of a deceased contributor. In 1997 the death benefit was equal to six months of retirement benefits to a maximum of $3,580. (Finance Canada News Release 97-083, supra note 76.) 94  Disability pensions are payable to contributors who meet a test of disability and unemployability. Benefits consist of a fixed amount plus an amount based on previous earnings. 95  Dependant children of a deceased or disabled contributor are entitled to a benefit fixed by formula and payable monthly until the later of when child turns 18 years, ceases to be a full time student or turns 25 years while remaining a full time student. 96  'An Act to amend the Old Age Security Act, S.C. 1963, c. 16 increased the OAS benefit to $75 per month.  91  98  Bryden, supra note 4 at 131; Guess, supra note 3 at 144.  -34necessary." The maximum GIS payable in a year was set at 40 per cent o f the O A S payment and only those whose entire income consisted o f the O A S and GIS were eligible for the GIS maximum.  100  Unlike O A S , GIS benefits must be applied for each year.  In 1971 marital status became a determining factor in the amount o f the G I S benefit well as the rate by which GIS is reduced when income exceeds a certain amount.  102  101  as  In spite o f its  name, over time the G I S has become more o f an integral part o f Canada's social security system than a 'supplement' to O A S . While initially the GIS was simply an additional 40 per cent o f the O A S benefit, amendments over subsequent years have altered the ratio o f GIS to O A S . In 1995 the maximum GIS benefit was $460.79 for a single person and $300.14 for a married person, while the O A S benefit was $387.74.  103  In 1994,40 per cent o f O A S recipients also received GIS benefits,  though for women the percentage was significantly higher at 65 per cent.'  04  "This belief was short-lived; in 1970 the OAS Act was amended to eliminate the phase out provision built into the original GIS amendment. Note however that "income" has a statutory definition and income from certain specified sources is excluded from the definition. ( Bryden, supra note 4 at 131.) l00  An Act to amend the Old Age Security Act, 1970-71-72, c. 9, ss. 5 and 6. Different rates were established for single and married pensioners. For example in 1971 a single pensioner could qualify for GIS up to $55 and a married couple could qualify for GIS up to $95. GIS rates periodically increased to the point that GIS exceeded OAS in 1981. In 1986 me maximum GIS payable monthly for a single pensioner was $338.95 and for pensioners married to each other it was $441.50 per couple. The maximum OAS in 1986 was $285.20 for each pensioner. (Burbidge, supra note 4 at 12-13.) m  An Act to amend the Old Age Security Act, ibid. For single pensioners and married pensioners whose spouses do not qualify for the OAS or the spouses' allowance (discussed below) the recapture rate is 50 per cent on income above the combined OAS and GIS benefits; for married pensioners the rate is 25 per cent on the couple's combined excess income for the year. (Burbidge, ibid, at 12.) m  Statistical Overview, supra note 61.  m  Ibid. at 110.  x  -351975: Spouse's Allowance Another component, the Spouse's Allowance ("SPA") was added to Tier I of the retirement income scheme by a 1975 amendment to the O A S A c t .  105  A t the same time the O A A component  of the 1951 program was eliminated and Tier I took its current three component form (OAS/GIS/SPA).  S P A was enacted to provide benefits equivalent to the O A S and G I S to needy people between the ages o f 60 and 65 who were married to GIS recipients. A residency requirement o f 10 years residence in Canada had to be met, and the qualification ceased on marital separation or divorce. O n the death o f the O A S / G I S recipient spouse, S P A benefits continued until the S P A recipient reached age 65 or remarried. A subsequent amendment to the O A S A c t in 1985 provided that any widow or widower between the ages o f 60 to 65 who met the residency requirement could, subject to need as measured by the G I S / S P A income test, qualify for the S P A benefit, regardless o f whether her or his spouse was an O A S / G I S recipient.  The effect o f S P A was that the family income o f a pensioner whose spouse was younger by five years or less would be the same as it would be i f both spouses were pensioners. S P A benefits are based on the married couple's couple's income and S P A is recaptured from excess family income at 75 per cent until the O A S equivalent is depleted, at 50 per cent until the GIS equivalent is reduced to a set minimum, and then at 25 per cent until the S P A is eliminated.  106  An Act to amend the Old Age Security Act, to repeal the Old Age Assistance Act and to amend other Acts  m  in consequence thereof, S.C. 1974-75-76, c. 58, s. 9. This amending legislation also eliminated the special fund and from this point all Tier I benefits were paid out of general revenues. As the name of the legislation indicates, the OAA component of Tier I ceased to exist at this point as well. l06  Burbidge, supra note 4 at 12-13.  -36Because of the connection to a marital relationship, many needy people in the qualifying age group are excluded from S P A benefits, including the never married, separated, divorced, lesbians and gay men.  107  For the excluded who are not in the paid labour force, welfare is often the only alternative,  and welfare rates generally fall below the S P A benefit.  (e)  The Three Tiered System Legislative initiatives during the 1960s which implemented the C P P and added the needs  based GIS as a supplement to the universal  108  O A S pension gave Canada's retirement income  system its three tiered structure comprised of public pensions intended to provide a basic income level for all Canadians o f retirement age ( O A S and GIS), mandatory earnings related public pensions intended to ensure some relationship between pre- and post-retirement income up to the A I W (CPP), and private arrangements partially funded by public subsidies (RPPs and RRSPs). A s the preceding historical overview shows, the three tiered structure was not the result o f a comprehensive notion o f what a pension system should be; rather it evolved in a somewhat piecemeal manner in response to continuing political pressure for improved economic security for the elderly. The tiers do not stack evenly on top of each other, but tidily interlock with the tier(s) above or below i n some places and leave gaping holes in others.  On the issue of eligibility of lesbians and gay men for SPA benefits see the discussion of Egan v. Canada, [1995] 2 S.C.R. 513, in Chapter Three, below. 107  By 1977 full OAS benefits were universally available to those aged 65 and older who lived in Canada for 40 years after the age of 18; persons with fewer than 10 years residency in Canada after age 18 did not qualify for OAS, unless they came from countries with international social security agreements with Canada. (See An Act to l08  amend the Old Age Security Act, S.C. 1976-77, c. 9, s. 1, and Pension Primer, supra note 1 at 5.) Partial OAS  benefits, on a prorated basis were available to those whose residency periods fell within 10 and 40 years after age 18.  -37Although one o f the goals o f Canada's retirement income system is "to ensure that elderly people have incomes high enough to allow them to live in dignity no matter what their circumstances were during their working years" , presently the basic income provided by 109  O A S / G I S / S P A (Tier I), though indispensable, unfortunately falls below the poverty level, and is especially low for single seniors among whom women are over represented.'  10  W h i l e it originally  provided an income o f 20 per cent o f the A I W ' " , by the end o f the 1970s the O A S income level dropped to 14 per cent o f the A I W , due to failure to fully index the benefit to the cost o f living, and has remained at that level. GIS increases the income level, though the recipients o f maximum GIS still fall far below the poverty l i n e . " GIS is clawed back dollar for dollar on income that exceeds 2  the O A S / G I S maximum. A l l three components o f Tier I are now paid out o f general tax revenues and benefits are indexed quarterly to the cost o f living.  For retired work force participants, C P P (Tier II) potentially tops up the Tier I benefit by 25 per cent o f the A I W , to 39 per cent o f the A I W , depending on the retiree's lifetime workforce  A Pension Primer, supra note 1 at 1.  m  "°The low level of Tier I income calls to mind the arbitrary basis for setting the benefit level of the old age pension in 1927 at $365 per year. See supra note 28. '"Before 1967 the OAS was not adjusted for inflation. An amendment in 1967 added a limited annual inflation increment to the OAS by incorporating into the OAS Act the "pension index" defined in the CPP Act. Since the pension index only partially reflected the consumer price index, the OAS, GIS and CPP fell further behind the cost of living each year. See Bryden, supra note 4 at 131-132 for a detailed description of the pension index. " Maximum GIS in 1995 was $460.79 per month or $5,529.48 for the year. The maximum OAS benefit at January 1, 1995 was $387.74 per month or $4,652.88 for the year. (Statistical Overview, supra note 61 at 109.) In 1995 the poverty line ranged from $16,874 (singles) and $21,092 (couples) for residents of cities with populations over 500,000 to $11,661 (singles) and $14,576 (couples) for residents of rural areas. (Caledon Institute of Social 2  Policy, Round Table on Canada Pension Plan Reform: Gender Implications (Ottawa: The Caledon Institute of Social  Policy, 1996), Appendix C - Pension Trends, graphs entitled "Maximum Old Age Security/Guaranteed Income Supplement versus Poverty Line for Different Areas, Single Senior, 1995" and "Maximum Old Age Security/Guaranteed Income Supplement versus Poverty Line for Different Areas, Elderly Couple, 1995".)  -38earnings.  This level o f income hovers around the poverty l i n e . "  3  In addition C P P provides  survivor benefits to spouses o f deceased retirees. Other benefits provided on the death o f a C P P member include benefits to dependent children and a one time death benefit intended to offset funeral expenses." C P P benefits are fully indexed to inflation and are paid out o f a special C P P 4  Fund that is financed by an earmarked payroll tax. Through its child care dropout provisions, C P P recognizes to some extent that women's work patterns vary from the male standard.  Under the present retirement income system, the key to maintaining a post-retirement income that provides a consistent standard o f living into retirement is access to Tier III. RPPs and R R S P s could potentially replace 70 per cent o f a taxpayer's earned income up to 2.5 times the AIW.  1 1 5  Though heavily publically funded through tax subsidies, participation in this tier depends  on private arrangements through employer-employee negotiations (RPPs) and personal savings (RRSPs). Thus accessibility to Tier III benefits is problematic for the economically disadvantaged. This limited access is especially problematic when one considers that in 1994 Tier LTT benefits cost the public purse roughly the same as Tier I benefits."  6  " The maximum CPP benefit at age 65 for 1995 was $713.19 per month or $8,558.28 for the year. (Statistical Overview, ibid, at 91 and 109.) The maximum OAS/CPP pension for 1995, then, was $ 13,211.16. 3  1l 4  As discussed previously, CPP also provides disability pensions to disabled members.  " Until tax reform in 1990 implemented mechanisms to equalize retirement savings potential of all Tier III vehicles, only defined benefit RPPs had this potential. See below, pp. 44-45. 5  " In 1994 the estimated expenditures were $22,780,000. (Department of Finance, Government of Canada Tax Expenditures (Ottawa: Department of Finance, 1997) at 28 (hereinafter "Tax Expenditures 199T').) That same year, Tier I benefits of roughly the same amount were paid out. (Statistical Overview, supra note 61 at 110.) 6  -39III  PROPOSAL FOR REFORM: GREEN PAPER ON PENSION REFORM (1982)  During the latter part o f the 1970s and early 1980s, a period o f rapid inflation, the replacement value o f O A S benefits slipped to 14 per cent o f the A I W " and employer sponsored 7  pension plans (RPPs) failed to fill government expectations that formed the rationale for capping the replacement value o f C P P benefits at 25 per cent o f the A I W . Indeed the private sector seems to have paid little attention to the expectation imposed on it - the percentage o f paid workers who were members o f employer sponsored pension plans increased from 30 per cent i n 1970 to only 32 per cent i n 1986.  118  Further, membership in a R P P was not necessarily a guarantee o f benefits on  retirement. A s Guest points out, "coverage was also restricted by lengthy and inflexible vesting provisions and a lack o f portability.""  9  Accordingly, by the late 1970s the second goal o f the  retirement income system, maintaining a consistent standard o f living into retirement, was out o f reach for a majority o f Canadians nearing or entering the end o f their working years; they faced retirement without the percentage o f replacement income needed to maintain their pre-retirement standard o f living.  While little attention was paid to gender issues as the pension system evolved into its three tiered form, during the late 1960s and into the 1970s the unfairness for women o f a pension system that was developed with men's lives in mind emerged as an issue in the struggle for women's equality. In 1970 the Royal Commission on the Status o f Women published its report, raising the  " Brown, supra note 4 at 39. 7  Pension Primer, supra note 1 at 41.  m  Supra note 3 at 200.  U9  -40question o f adequate pensions for women as one o f its concerns.'  20  Due to a number o f factors,  including the inadequacies o f their own pensions and the tenuousness o f their dependency on a man which was severable by divorce or death, elderly women were particularly vulnerable to a sudden plunge into poverty.  121  In response to intensifying public concern about the erosion o f pension benefits and failure o f the system to meet the needs o f the elderly, and particularly the needs o f elderly women, several studies o f Canada's pension system were published during the late 1970s and early 1980s.  122  In  1981 the federal government held a three day conference attended by "representatives o f the pension industry, business, labour, provincial governments, pensioners' groups, women's groups, and members o f Parliament and the Senate ...to share their ideas on pension reform."  123  later the federal government published what has come to be known as its Green Paper,  A year which  identified fundamental problems inherent in the pension system and put forward for public discussion proposals for reform.  124  The Green Paper recognized the following as "the most serious shortcomings" o f the  Report of the Royal Commission on the Status of Women in Canada (Ottawa: Queen's Printer, 28  m  September 1970). Ibid. at 37-40 and 325-328.  m  '"These are listed in Burbidge, supra note 4 at 78. Included are reports by a commission created by the province of Quebec, the Economic Council of Canada, the Ontario Legislative Assembly Select Committee on Pensions, Business Committee on Pension Policy, Canadian Labour Congress, and a 10 volume report of the Royal Commission on the Status of Pensions in Ontario. '"Guest, supra, note 3 at 199 '"Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Welfare Canada, 1982) (hereinafter "Green Paper")  -41-  retirement income system:  • Some elderly people who are single, particularly single women, have incomes that are insufficient. • Inflation has eroded the value of employer-sponsored pensions for many Canadians who are now retired and has also eroded the value of benefits accumulated by employees who are still at work. • Many employees will ultimately receive little pension income, even after many years of work, because they do not stay in one job long enough to acquire vested benefits. • Even if vested benefits are acquired, pensions are generally not "portable" when an employee changes jobs. • A significant number of working people are likely to experience declines in their living standards when they retire, in large measure because employersponsored pension plans are not universally available, because they lack sufficient opportunity to accumulate retirement savings, or because pension plan benefits are not sufficiently large.  • Women are inequitably treated throughout much of the retirement income system. (emphasis-added) 125  A s a result o f these findings, the Green Paper suggested that the focus for reform o f the 'public' components o f the retirement income system (Tiers I and II) should be on providing adequate income for all elderly people, but particularly single elderly women, and on equitable treatment for women. Inadequacies i n employer sponsored pension plans accounted for most o f the noted shortcomings i n the 'private' component (Tier III). Criticism o f the R R S P program included insufficient opportunity to accumulate savings and, like the rest o f the retirement income system, inequitable treatment o f women.  According to the Green Paper, the following principles provided the conceptual foundation  for Canada's retirement income system and should continue to inform and guide the reform  Hbid. at 2.  -42process:  • Elderly Canadians should be guaranteed a reasonable minimum income. • The opportunities and arrangements available to all Canadians to provide for their retirement should be fair. • Canadians should be able to avoid serious disruption of their pre-retirement living standards on retirement. 126  The identified shortcomings o f the pension system and the principles upon which reform should be based provide a meaningful frame o f reference for critique o f the reforms that eventually followed in the 1990s.  Various interest groups, including women's groups, labour unions, welfare organizations, provincial governments, business groups, the pension industry and the Economic Council o f Canada, responded to the Green Paper appeal for public discussion o f pension issues, publishing studies and making submissions to a parliamentary committee. While even a cursory discussion o f the decade long pension debate is too large an undertaking in the context of this thesis, it is both relevant and important to note that there was a comprehensive public debate on the issues and that interest groups in general, and women's groups in particular, were given the opportunity to speak to the issues that concerned them.  127  The weight given to the various voices, however, is another  matter.  Guest notes that the debate divided on ideological lines.  128  Those who considered  Ibid. at 11.  m  l27  See Guest, supra note 3 at 199-207 for an analytical summary of the pension debate.  Ibid.  m  -43retirement pensions to be primarily a social rather than individual responsibility favoured an institutional solution to pension problems; a common recommendation from those who shared this ideology was expansion o f the C P P . The competing ideology considered pensions to be the responsibility o f individuals and favoured a private market solution. Those who shared this ideology believed either that the weaknesses o f existing R P P s could be rectified by greater regulation o f the pension industry, or that employer sponsored pension plans should be mandatory. When one considers the underlying philosophical question - should individuals or society be responsible for providing for financial well-being o f Canada's seniors? - this debate is reminiscent o f the universal/contributory debate surrounding the 1951 program.  IV  THE '90s: PENSION REFORM  Not surprisingly, given their separate historical development and separate legislative regimes, reform o f the three tiers o f Canada's retirement income system was not implemented as a single package. The 'private', though tax subsidized, part o f the pension system (Tier III) was the first to be revamped, through amendments to the I T A which came into effect i n 1991 (the "1991 I T A amendments").  129  Tier II was next, with reforming legislation coming into force i n 1998.  Intended reform o f the first tier, scheduled to be implemented i n 2001, was outlined i n the 1996 federal budget.  130  This part o f the thesis sets out key elements of the reform (and proposed reform)  o f all three tiers.  An Act to amend the Income Tax Act and Related Acts, S.C. 1990, c. 35, s. 13.  m  See Department of Finance Canada, News Release 98-071, "Finance Minister's Statement on the Seniors Benefit" (28 July 1998). Although on July 28, 1998 Finance Minister Paul Martin announced that intended reform of Tier I will not proceed to implementation, the proposals outlined in the 1996 budget are relevant to the discussion of pension reform in this thesis. l30  -44-  (a)  1991: Tier III Reform The 1991 reform of the tax provisions relating to pensions was intended to accomplish two  things. First, it was to equalize the tax assistance available to all Tier III pension schemes, and secondly to allow flexibility for taxpayers who may not have the cash flow to make full contributions in some tax years. To achieve the first objective, the amendments significantly increased the amount o f tax deductible contribution limits to R R S P s and removed tax incentives that favoured employer-sponsored RPPs o f the defined benefit type over defined contribution pension arrangements, including all RPPs that are not defined benefit plans and all R R S P s . Providing a seven year carry forward of a taxpayer's allowable but unused R R S P contribution limit effected the second objective. The seven year limit on the carry forward was eliminated in the 1996 budget; unused R R S P contribution room may now be carried forward indefinitely.  The increased contribution limit revealed the government's hand with respect to pension reform.  Individual responsibility for retirement income is the clear message, either through  personal savings or negotiated employment pension plans. The amendments permit a taxpayer to contribute 18 per cent o f earned income to an R R S P , up to a maximum that w i l l increase to $ 15,500 and w i l l then be indexed to the A I W .  1 3 1  The 18 per cent figure is a rate reduction from the  previous 20 per cent, which reduced the maximum allowable deduction for lower income  The 1990 amendment raised the contribution limit from $7,500 to $11,500, to be further increased by $1,000 increments each year until 1995 when it would reach $15,500. The 1995 and 1996 budgets maintained the 1994 limit of $14,500 for 1995, reduced the limit to $13,500 for 1996 and froze it at that level until 2002 when the incremental increases will be revived for 2003 and 2004, at which time it will be up to $ 15,500 and thereafter will be indexed to the AIW. Department of Finance, "Tax Assistance for Retirement" in Budget Papers (Ottawa: Department of Finance Canada, 1995) (hereinafter "1995 Budget Papers") and "Measures Relating to Retirement Saving" in Budget Papers (Ottawa, Department of Finance Canada, 1996) (hereinafter "1996 Budget Papers"). See also the definition of "money purchase limit" in subs. 147.1(1) of the ITA, including its history. 13l  -45earners.  132  W h y was the 18 per cent rate chosen?  The eighteen percent limit is based on the "rule of nine" - the assumption that, on average, it is necessary to contribute nine dollars in the current year in order to obtain one dollar of annual pension. Therefore if one sets aside eighteen percent of one's income annually, one will receive a pension equal to two percent for every year of service. A two percent pension earned over a career of thirty to thirty-five years will replace an additional sixty to seventy percent of preretirement earnings in addition to what the public system delivers. 133  The 1991 I T A amendments equalized and integrated the treatment of all Tier HI retirement savings plans, whether employer sponsored or individual. If a taxpayer is a member of an employer sponsored plan, the 18 per cent R R S P limit is reduced by the value o f the contribution. Different methods of calculating the value are prescribed for the three general types o f employer-sponsored retirement savings plans - defined benefit plans, defined contribution plans and deferred profit sharing plans.  134  These adjustments reflect the intention that following the 1991 amendment the  same level o f retirement savings may be achieved through R R S P s as through RPPs. Prior to the 1991 I T A amendments workers who were members o f RPPs benefited from tax sheltering to a greater extent than those who were not R P P members and had to rely on R R S P s for their retirement savings.  The increased maximum contribution represents the federal government's commitment to  Prior to reform a person earning $37,500 could make a tax deductible RRSP contribution of $7,500 (20% of $37,500 = $7,500), but after reform the same taxpayer's limit was reduced to $6,750 (18% of $37,500 = $6,750). Only taxpayers with earned income over $41,667 (18% of $41,667 = $7,500) gained higher contribution limits as a result of the amendment. l32  '"Barbara Austin, "Policies, Preferences and Perversions in the Tax-Assisted Retirement Savings System" (1996) 41 McGill L.J. 57bt 575. 134  The distinctions between these types of plans are described supra at note 59.  -46provide a tax subsidy for contributions on earnings up to 2.5 times the A I W .  1 3 5  W h e n the  contribution limit reaches $15,500 in 2004, 18 per cent o f earned income up to $86,111 (18% o f $86,111 = $15,500) may be contributed to an R R S P .  This gives high income earners the  opportunity to accumulate tax subsidized retirement savings over a normal 35 year working life that w i l l replace 70 per cent o f an income o f up to $86,111 i n 2004 dollars. Thus public monies are being applied to subsidize retirement savings for high income earners to an income replacement level that is much higher than available under other forms of retirement assistance provided from public funds.  (b)  1998: Tier II Reform Remarks regarding B i l l C - 2 ,  136  the legislation that would reform the C P P , '  3 7  by the Minister  of Finance, the Honourable Paul Martin, to the House o f Commons Standing Committee on Finance and to the Senate Standing Committee on Banking, Trade and Commerce focussed primarily on transgenerational fairness and the need for restructured financing to achieve sustainability of the plan in the face of'baby boom' and 'baby bust' demographics.  138  Accordingly,  "75 per cent of the changes [to the C P P in B i l l C-2] are on the financing side, and only 25 per cent  Supra note 133 at 576 and 587.  ns  Bil\ C-2, An Act to establish the Canada Pension Plan Investment Board and to amend the Canada Pension Plan Act and the Old Age Security Act and to make consequential amendments to other Acts, I ' Sess., 36 m  s  Pari., 1996-97 ( S.C. 1997, c. 40). 137  Bill C-2 came into force on January 1, 1998.  '-""Department of Finance Canada, News Release 97-095, "Notes for Remarks by the Honourable Paul Martin Minister of Finance to the House of Commons Standing Committee on Finance on Bill C-2" (28 October, 1997); Department of Finance Canada, News Release 97-120, "Notes for Remarks by the Honourable Paul Martin Minister of Finance before the Senate Standing Committee on Banking, Trade and Commerce regarding Bill C-2" (17 December, 1997).  th  on the benefit side."  139  Indeed, as the discussion below illustrates, the changes to benefits could  be considered to fall on the financing side as they are cost cutting measures.  The following is a brief summary o f key elements o f C P P reform relating to retirement pensions  140  Financing  implemented by B i l l C-2:  141  The original modified pay-as-you-go system w i l l change to a more fully funded system. This change w i l l be implemented by: •  incrementally increasing contribution rates from the 1997 rate of 5.85 per cent (one-half of which is paid by each o f the employer and the employee) to 9.9 per cent in 2003 when the rate w i l l remain steady ; 142  •  increasing the return on investment o f C P P funds by professional investment o f the C P P fund " i n a diversified portfolio o f securities"  143  B i l l C-2 establishes an Investment Board  at arm's length from governments, though accountable to both the public and governments, and subject to legislated investment rules similar to those that apply to other pension funds.  ibid.  m  Although the discussion in this thesis is limited to retirement pensions, it should be noted that CPP disability pension benefits have been affected by the cost cutting measures enacted in Bill C-2. 140  141  CPP Act, supra note 70, ss. 108-113.1, as amended by S.C. 1997, c. 40, ss. 89-94.  Ibid. Recall that contributions are levied on income above the YBE and up to the YMPE, and that the rate represents the aggregate of equal contributions made by the employer and the employer. See discussion above at pp. 29-30. >42  Finance Canada, News Release 97-083, supra note 76. Recall that previously CPP fund investments were restricted to loans to provinces at attractive interest rates. (See supra note 80.) Under the amended legislation, provinces may still borrow from the fund, but after a transitional period such loans will be at market rates. (S.C. 1997, c. 40, s. 90). l43  -48•  freezing the amount o f earnings that are exempt from contribution (the Y B E  1 4 4  ) at $3,500  instead o f allowing it to increase proportionately each year with increases in the A I W .  1 4 5  Lower wage earners, among whom women are over-represented, w i l l be penalized by this change more severely than higher wage earners because a higher portion of their wages w i l l be subject to contribution. The maximum amount o f income subject to contribution (YMPE  1 4 6  ) w i l l continue to be indexed to wages.  Retirement Pension Benefits  147  The intention that the C P P replace 25 per cent o f earnings up to the A I W at retirement has not changed, however the formula for calculating the amount payable has.  148  Previously retired  C P P members were entitled to monthly payments o f 25 per cent o f an amount based on their average earnings with various adjustments, previous three years Y M P E .  1 5 0  149  up to a maximum based on the average o f the  N o w the maximum is based on the average o f the previous five  years' Y M P E . A s wages tend to increase from year to year, using a greater number o f years to make the calculation w i l l dilute the benefit. This w i l l be especially true for those who are still moving ahead on their career paths at the end o f their working lives as wage increases w i l l reflect promotions as well as cost o f living adjustments. A s women tend to be in the work force for fewer  l44  See supra note 82 and accompanying text.  145  Finance Canada, News Release 97-083, supra note 76.  l46  See supra note 83 and accompanying text.  I47  CPP Act, supra note 70, ss. 44-53, as amended by S.C. 1997, c. 40, ss. 69-71.  148  No change has been made to the annual cost of living adjustment.  Finance Canada News Release 97-083, supra note 76.  l49  Ibid.  l50  -49years than men, it is reasonable to assume that they w i l l be over-represented among those who continue to be upwardly mobile as they near retirement.  Eligibility for Retirement Pension Benefits N o changes have been made to C P P eligibility requirements. Those who work outside the paid labour force are still excluded from receiving benefits in their own right, though they may be eligible for benefits as a dependant o f a C P P member.  Survivor Benefits'  51  Entitlement to and calculation of survivor benefits remain essentially the same, except that when a surviving spouse receives both retirement benefits o f her own and a survivor benefit complex rules have been added that limit the adding up o f benefits.  152  A s before, the combined  retirement/survivor benefit cannot exceed one maximum retirement pension.  Death Benefit  153  The maximum lump sum death benefit has been reduced from a wage indexed amount that was $3,580 in 1997 to $2,500 and frozen at that l e v e l . '  54  B i l l C-2 represents only a stage in the reform o f the C P P , intended to "restore the C P P ' s  CPP Act, supra note 70, ss. 58 and 59, as amended by S.C. 1997, c. 40, s. 76.  l51  Ibid., subs. 58(2) as amended by S.C. 1997, c. 40, s. 76(1).  ]52  " Ibid, s. 57 as amended by S.C. 1997, c. 40, s. 75. 3  Ibid, subs. 57(1.1).  ]54  -50sustainability and make it fairer and more affordable for future generations."  155  Proposals for  changes relating to such issues as partial pensions for those who want to move gradually into retirement, updating survivor benefits, credit splitting during marriage, increasing the limit on pensionable earnings and integration o f Employment Insurance and C P P benefits w i l l be considered over the next couple o f years.  156  It is noteworthy that, notwithstanding Finance  Canada' s statement that these issues "were raised either too late in the review or after consultations were complete" , the issues o f survivors' benefits, credit-splitting and increased coverage were 157  proposed as possible reform measures by the Liberal government in 1982.  158  The issues o f  increased C P P coverage, up-dated survivors' benefits and credit splitting arose during the 'great pension debate' o f the 1980s as w e l l .  159  The pension reform agenda changed, however, in the  1990s. In 1996 the federal government and the nine C P P provinces released a document entitled An Information Paper for Consultations on the Canada Pension Plan,* which focussed on the 60  sustainability o f the plan and completely omitted mention o f the adequacy o f the C P P as part o f Canada's public pension system.  161  The appropriate role for the C P P i n Canada's retirement income system w i l l continue to be  '"Finance Canada News Release 97-083, supra note 76. Ibid.  ,S6  Ibid.  [51  Green Paper, supra note 124 at 32-34.  m  ' "Guess, supra note 3 at 199-201. Department of Finance, An Information Paperfor Consultations on the Canada Pension Plan (9 February  160  1996). Guess, supra note 3 at 286-292. Guess notes, at p. 282, that Canadians understand the word 'sustainable' as "a codeword for cutbacks". l61  -51debated for the next few decades predict Banting and Boadway in their recent study o f retirement income policy reform.  162  A s the middle tier o f the pension system, the C P P shares some  characteristics o f both its neighbours. L i k e Tier I, it is a publicly administered program funded through involuntary contributions (taxation). L i k e Tier III, participation and benefits are tied to contributions, which in turn are tied to workforce participation. Accordingly the C P P provides a middle ground that w i l l be the site where the conflict between those who support an expanded public pension system and proponents o f a privatized, though tax subsidized, pension system is mediated.  (c)  163  Tier I Reform (proposed) In the 1996 federal budget, Finance Minister Paul Martin announced proposed changes to  the first tier o f the retirement income system which were targeted to take effect in 2001. These changes would have completed the transformation o f Tier I from a universal program to one that targets low income elderly.  l62  164  This transformation started i n 1990  165  when a special tax was  Keith G. Banting and Robin Boadway, "Reforming Retirement Income Policy: The Issues" in Keith G.  Banting and Robin Boadway, eds., Reform of Retirement Income Policy: International and Canadian Perspectives  (Kingston: School of Policy Studies, Queen's University, 1997) 1. Ibid.  ia  An Act to amend the Income Tax Act, the Federal-Provincial Fiscal Arrangements and Federal PostSecondary and Health Contributions Act, the Old Age Security Act, the Public Utilities Income Tax Transfer Act, th War Veterans Allowance Act and a related Act, S.C. 1990, c. 39, s. 48. ]M  Arguably the transformation began with the slide of the ratio of OAS to AIW from 20 per cent to 14 per cent where it was allowed to remain while the means tested GIS component of Tier I benefits was increased to the extent that it exceeded the OAS component. (See discussion above at p. 34.) 165  -52imposed on O A S benefits paid to high income recipients.  166  In 1996 a further amendment to the  I T A provided for withholding "an amount equal to the clawback from monthly O A S cheques"  167  which ended "the charade of the federal government sending out cheques to wealthy seniors every month only to recapture the money at tax time."  168  In a news release on July 28,1998 the Finance  Minister announced that the proposed reform would be scrapped, at least for the time being.  169  However the news release also made it clear to Canadians that sustainability continues to be the primary goal o f pension reform and that the objective of the proposed reform, targeted support for needy seniors, remains on the agenda of the Liberal government. Accordingly, an examination o f the proposed reform is as relevant to the discussion o f the current wave o f pension reform as it was before July 28, 1998.  The proposed reform package, called the Seniors Benefit, would replace the O A S / G I S / S P A with a tax free benefit that incorporates current age and pension income tax credits.  170  Key  provisions o f the proposal are: •  individual entitlement would cease for seniors who are married, including those in common law relationships o f more than one year duration. The amount o f the benefit would be based on combined spousal income, but the monthly payments would be delivered in equal amounts separately to each spouse.  In 1995 OAS recipients were taxed at $0.15 for every dollar of net income greater than $53,215, up to $84,484 at which point the entire OAS benefit was repaid. 166  '"Pension Primer, supra note 1 at 4; ITA s. 180.2. Pension Primer, ibid.  m  ™Supra note 130. 1996 Budget Papers, supra note 131.  170  -53the maximum Seniors Benefit would be $120 more per household per year than the maximum combined O A S / G I S .  171  While any increase in benefits would be welcome, an  additional $120 a year does not go very far toward reducing the poverty gap for seniors whose income consists only o f Tier I benefits.  172  universality would end in form as well as substance. Entitlement to the Seniors Benefit would be 'clawed back' at the rate o f 20 per cent on family income above $25,921.  173  Couples with a combined income above $78,000, and singles whose income exceeds $52,000, would not qualify for any portion o f the Seniors Benefit.  This compares  unfavourably with the current clawback, at 15 per cent o f individual income over $53,215.  174  Comparisons between the existing and proposed systems are difficult because  of structural changes, in particular the tax-free and family income based features o f the Seniors Benefit. While the Seniors Benefit would be a tax-free program, the effect o f the benefit is better understood by looking at both the benefit that a particular senior would receive and the income tax that he or she would pay.  175  If the 20 per cent 'clawback' is  regarded as a form of taxation, the effective tax rate on family income over $25,921 could  Senior couples who qualify for the additional $ 120 per year will receive only $60 each; single seniors who qualify for the maximum benefit will receive the full $ 120. 171  In 1995 the poverty gap for such seniors living in Canadian cities ranged from -$2,551 to -$5,555 for singles and -$453 to -$4,800 for couples. (Pension Primer, supra note 1 at 15.) 172  '"Department of Finance, "The Seniors Benefit: Securing the Future - Projected Benefit Levels" in 1996 Budget Papers, supra note 131. I74  ITA, s. 180.2. OAS is completely clawed back for individual seniors with incomes above $84,484.  This point is discussed in John Geddes, "Citizens Revolt", 111-16 Maclean's 12 (April 20, 1998). According to a study reported in this article, when tax consequences are considered (including the clawback as a tax), senior couples with retirement income of over $30,000 from sources other than Tier I benefits will have lower after tax income under the Seniors Benefit program, while those with 'other source' retirement income of $20,000 will be better off by $559 ($279.50 each). l75  -54range from approximately 45 per cent to 70 per cent. This is because income tax would be payable at the taxpayer's normal rate(s) on the income above $25,921, plus Seniors Benefit eligibility would be reduced by 20 per cent, thereby increasing the 'normal' tax rate by 20 per cent. Under the existing system, the amount o f O A S clawed back is deductible in computing a taxpayer's income. •  176  the benefit levels of the Seniors Benefit as well as the eligibility thresholds would be fully indexed to inflation.  •  S P A would continue to operate, and the $120 per year increase in the maximum would apply to this program as well.  Although "[n]ine out o f 10 single senior women [would] be better off [under the Seniors Benefit]" , the family income eligibility requirement represented a step backward for the financial 177  autonomy o f women. Since the lower o f the two incomes that comprise a (heterosexual) family income generally belongs to the female spouse,  178  a family income based system would reduce her  pension income and increase her husband's. Further, since benefits would be higher for single seniors than for married (including common-law) couples, the possibility o f intrusive monitoring of social and sexual relationships for eligibility purposes loomed large.  l76  179  Again, women would  ITA, subs. 180.2(2) and 60(w).  '"Department of Finance, "The Seniors Benefit: Questions and Answers" in 1996 Budget Papers, supra note 131. This is because single senior women are over represented among the poorest of the poor and will accordingly be over represented among those who qualify for the $120 per year benefit increase. In 1993 married women who were employed full time earned, on average, 70 per cent of the earnings of their male counterparts. (Statistics Canada, Women in Canada: A Statistical Report, 3 ed. (Ottawa: Statistics Canada, 1995) at 87.) This statistic does not reflect the fact that many more women than men work part time. In 1995 12 per cent of female workers and 4 per cent of male workers worked full year, part time. (Statistics Canada, The Daily (http//www.StatCan.ca/Daily/English/980317/d980317.htm.) 178  rd  ,79  See Kathleen Jamieson, "Women Will Lose in Pension Reform" (Winter 1997/1998) 14:2 Sparc News 2.  -55have been the most affected, due to their over representation among those who rely most heavily on Tier I retirement income.  W i t h the implementation of the Seniors Benefit, this decade of pension reform would have completely altered the pyramidal shape o f the current system, which prior to reform provided Tier I benefits to all Canadian seniors, Tier II benefits to seniors retired from work force participation, and Tier HI benefits to those with employer sponsored pension plans or who had sufficient disposable income to save for retirement.  The Seniors Benefit would have made C P P the  foundational tier, with Tier HI as the mechanism for providing for consistent pre- and postretirement standard o f living and Tier I as a welfare program that w i l l no doubt revive the "sense of personal failure that accompanied acceptance o f the old age pension under the [1927 A c t ] . "  180  The withdrawal o f the reform proposal is not necessarily something to cheer about, however. The inadequacy o f Tier I benefits and much needed reform o f this aspect o f Tier I is unlikely in the foreseeable future. According to the Finance Minister's press release, the rationale for the decision not to proceed with the implementation o f the Seniors Benefit is that the government has achieved a "balanced budget and a declining debt-to-GDP ratio."  181  The Finance  Minister stressed, however, "that today's fiscal success w i l l need to be matched by continued fiscal  180  Guest, supra note 3 at 137. See also the discussion above at p. 21.  Supra note 130. It is interesting, but not surprising, that no direct mention was made of the political backlash by seniors against this proposed reform, or of the criticism that middle income Canadians would be financially worse off as a result of the reform and the high effective tax rate resulting from the 20 per cent clawback would be a disincentive to individual retirement saving. (See Mary Janigan, "Making the Middle Class Pay", 110-39 Maclean's 42 (September 29, 1997) and Geddes, supra note 175.) m  -56vigilance i n the future".  182  "Sustainability" was mentioned a number o f times in the Finance  Minister's brief announcement, and the government's "commitment to those in need"  183  was  emphasized, foreshadowing future reform o f Tier I that w i l l continue the erosion o f universality already effected through income tax clawbacks o f O A S / G I S benefits and possibly reduce already too low benefit levels. A t this point the form that future reform w i l l take is unknown. Further use o f the tax system as a reform tool is a possibility, or the Seniors Benefit may be revived either i n its original or a revised form. While the specific form o f future Tier I reform may be unknown, the unarticulated but clear message is the government's commitment to individual responsibility for economic well-being in one's retirement years. In his discussion o f the costs associated with Canada's retirement income system, the Finance Minister failed to mention the cost in uncollected tax revenues that support 'private' arrangements for retirement income. In fact, Tier m is not mentioned as part o f Canada's retirement income system at all.  V  WHAT DOES THE TAX SYSTEM HAVE TO DO WITH PENSIONS?  The pension system provides an interesting site for study of the functions of the tax system and its operation as a tool for implementing social policy. While this topic is far too complex to explore in detail in this thesis, some discussion o f the relationship between the pension and tax systems is necessary as background. Some aspects of this relationship have been noted previously, particularly i n relation to the third tier o f the retirement income system, R P P s and R R S P s , the benefits o f which are delivered through the I T A .  Supra note 130.  m  Ibid.  m  -57-  The primary purpose o f the tax system is to raise revenues and the purpose o f raising revenues is to provide services to the public:  All other purposes of the tax system are incidental to the need to pay for public services. It is extremely unlikely that we would have a tax system at all if taxes were not needed to pay for public services. 184  W h i l e its primary function is raising revenue, the tax system has other functions as well including (re)distribution o f economic resources, delivery o f benefits through tax expenditures, regulation of the economy through incentives or disincentives, compensation to society for private use o f public resources and stimulation o f the economy.  185  The retirement income system is closely  connected to the first three o f these functions, and these are interconnected as well.  Once a tax system is in place, public services, such as a pension system, may be paid for either directly by spending revenues raised through the tax system or indirectly by not collecting taxes otherwise payable from taxpayers who spend their own money on the particular service. 'Spending' uncollected tax revenues to deliver a public service is known as a tax expenditure.  186  Although the tax expenditure concept is controversial and the boundaries between what is or is not uncollected tax are contested,  187  the Canadian Department o f Finance recognizes a large number  Ontario, Fair Tax Commission, Report: Fair Taxation in a Changing World (Toronto, University of Toronto Press, 1993) at 35. l84  Ibid. at 35-42.  m  Ibid. at 39. See also Stanley Surrey, Pathways to Tax Reform (Cambridge: Harvard University Press, 1973). The origin of the tax expenditure concept is generally attributed to Surrey. m  NeilBruce, "Pathways to Tax Expenditures: A Survey of Conceptual Issues and Controversies" in Neil Bruce, ed., Tax Expenditures and Government Policy (Kingston: John Deutsch Institute for the Study of Economic Policy, 1988) 21. Tax expenditures were first published in Canada in 1979, then again in 1980, and on a regular l87  -58o f tax expenditure items in its public accounts and publishes a list o f the value o f all tax expenditures each year.  188  The primary relationship between the tax system and the pension system is that tax revenues and expenditures pay for the pension system. Tier I, O A S / G I S / S P A , is funded by direct transfers to elderly Canadians o f income tax revenues; it is a redistribution o f income from taxpayers, in the case o f the O A S to all elderly persons who meet the age and residency requirements, in the case of GIS to needy O A S recipients, and in the case o f S P A to the needy who meet the age and family status requirements. O A S benefits are subject to income tax but the age and personal exemptions effectively-set off any tax payable on incomes consisting o f only O A S and GIS benefits.  The  income tax system is also used to claw back O A S benefits paid to taxpayers who have taxable incomes over $53,215.  189  Tier II, C P P , is funded by an involuntary ear-marked payroll tax imposed on all members of the work force who earn more than 10 per cent o f the A I W and their employers, as well as the income earned on the investment o f those payroll tax revenues.  Employers and employees  contribute equal amounts and the self employed contribute as both employer and employee. Contributors do not bear the full cost o f their contributions as these are tax deductible to employers  annual basis from 1992. The 1998 report entitled Government of Canada: Tax Expenditures was released by the Department of Finance on July 16, 1998 and includes estimates and projections for the period 1993 to 2000. (Correspondence by electronic mail from Leonard Arvelin, Enquiries Officer, Finance Canada, July 21, 1998.) Satya Poddar, "Integration of Tax Expenditures into the Expenditure Management System: The Canadian Experience" in Neil Bruce, ed., Tax Expenditures and Government Policy (Kingston: John Deutsch Institute for the Study of Economic Policy, 1988) 260. l88  See ITA s. 180.2. Following an amendment to the ITA in 1996 (S.C. 1996, c. 21, s. 46) withholding of income tax has been applied to OAS benefits. l89  -59as a business expense and are eligible for tax credits to employees. The effect o f the tax credit is that employees pay only 83 per cent o f the cost o f their contributions. C P P benefits are subject to income taxation when received.  The Tier III portion o f the pension system is both delivered through and paid for (though not completely) by the tax system. The structure o f the tax system shapes the delivery o f Tier HI pension benefits. The I T A sets rules that pension plans and pension savings must comply with to qualify for registration and the accompanying tax benefits. Further, RPPs and R R S P s are funded to a significant extent by tax expenditures.  190  Because the tax subsidy is a deduction rather than  a tax credit, income earners with high marginal tax rates receive greater tax savings for the same contribution than income earners taxed at lower marginal rates. Higher income earners also benefit more than lower income earners from the tax subsidy effected by sheltering the income earned on R R S P and R P P contributions. There are two reasons for this. First, only some taxpayers are positioned to take advantage o f the very favourable tax treatment. The R R S P tax subsidy is available only to those taxpayers who have sufficient disposable income to afford to make the voluntary savings that qualify for the deduction. In the case o f R P P s , only some employers, generally large businesses and the public sector, offer these employee benefits. Second, the amount that may be contributed is a percentage o f income. Thus, higher income earners are both permitted and able to contribute more to their RPPs or their R R S P accounts.  Delivery o f the Tier m portion o f the pension system through tax expenditures has been  At roughly $23 billion in 1994, tax deductions and non-taxation of investment income for RPPs and RRSPs represent Canada's largest tax expenditure. In 1994 the estimated expenditures were $22,780,000. (Tax Expenditures 1997, supra note 116 at 28.) l90  -60severely criticized on the grounds that it is not the privately funded component that it purports to be and that the tax system is not an appropriate or effective vehicle for delivering social programs.  191  Dennis Guest expressed a similar concern:  The use of the tax system as a method of enhancing people's standard of living is as old as the federal system of income tax itself. In sharp contrast to transfers through conventional social security or social welfare programs, whose cost and administration are the subject of regular public monitoring, comment, and study, the benefits provided through the tax system have been hidden from public scrutiny. 192  The fact that the social security program hidden by tax expenditure delivery is one that benefits those who need it the least is particularly objectionable.  A s the historical review of the retirement income system illustrates, social policies, and the tax policies that support them, are reflected in the way i n which governments choose to fund social programs. thrift"  In 1908 the view that poverty in old age "could be avoided by the application o f  shaped the first legislative attempt to deal with the problem o f providing income for  193  retirement. When this program was less than successful, the government o f the day recognized that it had to provide something more, but limited its assistance, from general revenues, to those who met an onerous means test.  194  When the limited scope o f this program proved inadequate, and even  detrimental to the "spirit o f thrift and self reliance"  l9l  195  that restricting benefits to the very needy was  Claire F.L. Young, "Public Taxes, Privatizing Effects, and Gender Inequality" in Susan B. Boyd, ed.,  Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press,  1997) 307 at 308; Austin, supra, note 133. l92  Guest, supra note 3 at 189.  Supra note 8.  m  l94  See discussion of the 1927 Act, above at pp. 12-16.  Supra note 36.  m  -61intended to promote, universality was ushered in with the legislation that implemented the 1951 Program.  196  However, a special tax, known as the 2-2-2 formula, was levied as a concrete reminder  to the public that government responsibility for providing income to senior citizens had a very real price. The next stage of reform o f the 'public' pension system was the addition o f the C P P , again funded by a special tax. In contrast, no tax increases were directly associated with the foregoing o f significant tax revenues either when tax subsidies for 'private' initiatives were legislated or later when they were increased.  While the proposed Seniors Benefit is no longer scheduled for  implementation, a skeletal Tier I program remains a very real possibility, and the rationale is that there are insufficient tax revenues to finance existing Tier I benefits. In conjunction with increased expenditures for 'private' savings, a pared down Tier I underscores a policy that those who look after their own retirement needs w i l l be rewarded for doing so, while those who can't w i l l face an impoverished old age. Most o f the latter w i l l be women.  'See discussion above at pp. 19-22.  -62CHAPTERTWO REALITY AND M Y T H : T H E ECONOMIC SITUATION OF ELDERLY WOMEN  While pension plans have few provisions that discriminate overtly against women, their design results in women receiving disproportionately low benefits.  1  I.  WOMEN AND THE RETIREMENT INCOME SYSTEM  The effect o f the way that the retirement income system is structured is that women may not be able to provide adequately for their retirement years.  This is because, except for the  inadequate basic level o f income provided by Tier I, pension entitlement is directly linked to preretirement participation in the paid workforce. Contributions to the C P P can only be made through employment, R P P membership is available through employment, and contributions to R R S P s are limited to a fixed percentage o f the taxpayer's earned income for the tax year.  2  The rules that  govern contribution limits and consequently benefit levels assume a normal working life, and a normal post-retirement life span, by standards that reflect a male reality. W o m e n generally participate in the work force for fewer years than men, earn lower wages than men, and live longer than men. The failure o f the retirement income system to accommodate the different gender realities is the design flaw that "results in women receiving disproportionately l o w benefits."  3  'Government of Canada, Better Pensions for Canadians: Focus on Women (Ottawa: Department of Supply and Services, 1982) at 3. For the purposes of RRSP contributions, "earned income" includes employment and business income and spousal and child support payments, but not most types of property (ie. investment) income. (Income Tax Act, R.S.C. 1985, c. 1 (5 Supplement) as amended from time to time (hereinafter "ITA"), subs. 146(1).) 2  th  1  Supra note 1.  -63To some extent Canada's retirement income system recognizes that access to pensions through work force participation leaves women inadequately provided for in their retirement years. A secondary means o f accessing the pension system is available, through a spousal relationship. Rather than addressing the underlying reasons for the gender inequity o f access through workforce participation, namely the failure to recognize the unpaid work that women do outside the workforce and the undervaluing o f work that women do in the workforce, secondary access to the pension system through a spousal relationship  assumes that inequity o f access through workforce  participation is gender neutral and is connected to being a spouse rather than being a woman.  (a)  Access Through Workforce Participation While most women do participate in the paid labour force, their contributions to  reproductive work , which is generally done in the family and is unpaid, impairs their ability to 4  achieve income levels necessary for access to maximum participation i n the pension system. Although women's participation in the paid labour force has risen substantially over the past couple of decades, women's wages continue to lag behind those o f men. In 1993 the average pre-tax 5  income for women ($16,500) was 58 per cent o f that for men ($28,600). A l s o in 1993, average 6  In this paper reference to reproductive work is used in the same sense as the term 'social reproduction' is used by Dorothy Chunn, "Feminism, Law and Public Policy: 'Politicizing the Personal'" in Nancy Mandell and Ann Duffy, eds., Canadian Families: Diversity, Conflict and Change (Toronto: Harcourt Brace & Company, Canada, 1994) 177 at 201 note 7: "Social reproduction of the species is linked to the 'motherwork,' 'wifework,' and 'housework' that women perform to maintain the physical and emotional health of individual family members and hence of the society as a whole." Thus 'reproductive work' encompasses much more than work related to biological reproduction and care of dependent children. 4  s  In 1994,45 per cent of all paid workers were women compared with 37 per cent in 1976. Statistics Canada,  Women in Canada: A Statistical Report, 3 ed. (Ottawa: Statistics Canada, 1995) (hereinafter "Women in Canada") rd  at 9. Ibid. Taxable income includes incomefromall sources. It is interesting to note that the gender gap for CPP benefits is roughly the same as the gender gap for income. In 1993 the average CPP retirement benefit paid to women was approximately 58 per cent of that paid to men, and the percentage has fallen by approximately one percentage k  -64annual earnings for women ($ 18,936) were 64 per cent of men's average earnings ($29,599), and 7  the average earnings for women who worked full time for the full year ($28,392) were 72 per cent o f those for men ($39,433).  8  Many of the factors that contribute to the lower average incomes for women can be linked to the responsibilities that women have for unpaid reproductive work, including: 9  the over-representation o f women among the part time employed. In 1994 women held 69 per cent of part time jobs, and 26 per cent o f employed women worked part time. Part 10  time work is not necessarily a choice, however. The number of women working part time who want full time employment has increased in recent years, from 22 per cent in 1989 to 34 per cent in 1994.' Even for those women who 'choose' part time work, the choice may 1  have been thrust upon them:  ...Given the scarcity and high cost of infant and child-care facilities, one parent is usually forced to stay home to look after the child or at least be prepared to interrupt their paid labour to accommodate the all-too-frequent interruptions in child care arrangements. ...If one partner is to drop out of the labour force for the day, week or year, it only makes sense that the lower-income earner, almost  point from 1993 tol996. (Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications, (Ottawa: The Caledon Institute of Social Policy, 1996) (hereinafter "Caledon Institute"), Appendix C graphs entitled "Average Canada Pension Plan Benefit, By Type of Benefit and Sex, March 1996", "Average CPP Retirement Benefit, By Sex, in Constant 1996 Dollars, 1970-1983, 1988-1996", and "Average CPP Retirement Benefit, Women's as Percentage of Men's, 1970-1983, 1988-1996".) Women in Canada, supra note 5.  7  *Ibid. at 95.  1996 census data shows that women perform significantly more unpaid work than men, whether that work was housework, home maintenance, child care or providing care and assistance to the elderly. (Statistics Canada, "1996 Census: Labour force activity, occupation and industry, place of work, mode of transportation to work, unpaid work" in The Daily (http://www.StatCan.ca/Daily/English/980317/d980317.htm at 22-23.) 9  Ibid. at 9. The incidence of men with jobs who worked part time in 1994 was nine per cent.  w  11  Ibid.  -65always the woman, do so. Since she is home, it also makes sense for her to do most of the household chores, especially given that she is trained for the work and probably thinks it's her job anyhow. A new cycle begins. She takes part-time work because it is the work available and because it allows her to do her other job at home, a job that receives little support from men in the form of domestic labour or from the state and employers in the form of child-care facilities, full maternity benefits and protection, decent pay and job opportunities. When her child-care responsibilities diminish, her exclusion from the better full-time jobs is justified on the basis of her interrupted work patterns, her part-time work attributed to preference. 12  •  women are over-represented among those who work i n non-standard jobs such as "own account"  •  self-employment and employment in the home.  13  14  women are more likely than men to drop out o f the paid workforce from time to time because o f family responsibilities.  15  This factor can be inferred from statistical information  which shows that women with children aged five years or younger are less likely to be employed than women whose youngest child is aged six years or older, and that the likelihood o f workforce participation by women increases as the age o f their youngest child increases.  16  This trend is apparent for both lone parent women and women in two parent  families, although workforce participation o f lone parent women is generally lower than for women in two parent families,  17  which further demonstrates the negative relationship  Pat Armstrong, "Women's Work: Women's Wages" in Greta Hoffman Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside Limited, 1987) 354 at 371. l2  "Own account" workers are those "who work by, and for, themselves". Analysis of 1996 census data shows an increasing trend toward "own account" work by women. Over the five year period from 1991 to 1996 "own account" work increased by 62 per cent for women and 27 per cent for men. (The Daily, supra note 9 at 3.) l3  7.3 per cent of women workforce participants and 5.3 per cent of men workforce participants worked at home in 1996. Work at home is more likely to be part-time for women than for men. (Ibid, at 13-15.) H  "Ibid. See also Monica Townson, "The Special Importance of the Canada Pension Plan for Women" in Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications, (Ottawa:  Caledon Institute of Social Policy, 1996) 21 at 22. Women in Canada, supra note 5 at 64-65.  >(,  "Ibid.  -66between family responsibilities, 100 per cent o f which fall on the lone parent woman, and workforce participation. •  on average, women who work full time work fewer hours per week (40.8 hours i n a 1996 census reference week) than men who work full time (45.7 hours in a 1996 census reference week).  •  18  regardless of the degree of workforce participation, women bear the primary responsibility for unpaid reproductive work. In 1992 women employed full time with a spouse and at least one child under age five spent an average o f 5.3 hours per day on household activities while their male counterparts averaged two hours less.  •  19  women are more likely than men to retire from the workforce before age 65 "to undertake family-related activities such as caring for a sick relative or a spouse" or to enhance the 20  enjoyment o f retirement for her spouse. ' 2  In short, women participate in the paid workforce to a lesser extent than men, and men participate in unpaid reproductive work to a lesser extent than women.  C P P (Tier II) benefits are based on average lifetime earnings. Since men earn more during their 'standard' working lives, they get more benefits on retirement. C P P does not entirely ignore  *The Daily, supra note 9 at 3.  l  ^ Women in Canada, supra note 5 at 9. Statistics Canada, Canada's Retirement Income Programs: A Statistical Overview (Ottawa: Statistics  20  Canada, 1996) (hereinafter "Statistical Overview") at 95. Ibid. As mentioned in the discussion of SPA (see above. Chapter One at pp. 35-36), which specifically provides relief for the economic disadvantage to men of the 'normal' age gap between spouses, men tend to have spouses who are younger than they are. 21  -67women's workforce realities, however: •  C P P covers all workforce participants, regardless o f whether they are full, part-time, temporary, contract or self employed workers, and regardless o f the sector in which they are employed;  •  22  C P P ' s child care dropout provision recognizes to some extent women's responsibility for reproductive work;  23  •  C P P is portable, an advantage to women, who change jobs more frequently than men;  •  C P P allows for flexibility in retirement age, allowing benefits to begin at any time between the ages o f 60 and 70;  •  25  C P P is fully indexed for inflation, an important feature for those who live long past retirement age;  •  24  26  spouses may split C P P credits earned during the marriage or common law relationship on divorce, separation or death.  O n retirement, spouses may choose to share their C P P  incomes, increasing the retirement income o f the lower earning spouse (wife).  27  Through these provisions the C P P does accommodate gender differences to a some extent, but the  22  Caledon Institute, supra note 6 at 25-26.  Hbid.  2  *Ibid.  2  21  Ibid. See also above, Chapter One at pp. 31-33.  Ibid. Note that women tend to live longer than men, and can expect to live approximately 20 years beyond the age of 65. (See below at pp. 79-81.) 26  Ibid. See also Monica Townson, Protecting Public Pensions: Myths vs. Reality (Ottawa: Canadian Centr for Policy Alternatives, 1997) at 47-48 and National Council of Welfare (Canada), A Pension Primer (Ottawa: Supply and Services, 1996) at 28-29. 21  -68program still favours men. In 1996, for example, the average C P P monthly benefit payable was $517 for men and $289 for women.  28  The problems noted at Tier II are magnified at Tier III. N o concessions are made for reproductive work, and employees in non-standard jobs, including part-time, in-home and 'self account' employment are most common in sectors o f the economy where R P P coverage is l o w .  29  In addition, part-time employees obviously w i l l earn less than full-time employees and w i l l therefore have lower R R S P contribution limits, as well as fewer financial resources with which to make contributions. Periods of withdrawal from the workforce and early retirement, also 'standard' features o f a woman's working life, mean that women generally w i l l have fewer years in which R R S P and R P P  3 0  contributions can be made, which in turn w i l l result in reduced benefits on  retirement.  In addition to the quantitative aspect o f limited access to the pension system for women (they spend less time in the workforce than men and therefore have lower annual and lifetime earnings), there is a qualitative aspect that is even more insidious. Women's participation i n the work force is generally accorded secondary importance and is therefore undervalued and  Pension Primer, ibid, at 26.  2S  Whether full or part time, most women in the workforce hold jobs in the retail sector, and in community, business and personal services. In these sectors employers tend to be small businesses that do not provide much in the way of benefits to employees. Public sector employees and unionized employees are the most likely to have RPPs. Women workers are less likely than men workers to be public sector or unionized employees. (Monica Townson, 29  Women's Financial Futures: Mid-Life Prospects for a Secure Retirement (Ottawa: Canadian Advisory Council on  the Status of Women, 1995) at 37-39.) If she is among the fortunate few women who are RPP members. In 1993 41.9 per cent of women paid workers (the self employed are not included) were RPP members. (Statistical Overview, supra note 20 at 50-53.) 30  -69accordingly underpaid.  31  The rationale for the secondary status accorded women's workforce  participation is the assumption that men work to provide for their families' needs while women choose to work not because of need but for some secondary purpose. Women's realities belie this assumption. Historically and currently most women enter the workforce because o f economic need.  32  The importance o f women's earnings to their families' economic well being is underscored  by the fact that i n 1992 four per cent o f dual earner families had family incomes below the poverty line and that figure would increase to 16 per cent i f the wives' incomes were excluded.  33  Further,  in 1991 16 per cent o f all families with children i n Canada were lone parent families headed by a woman.  34  Traditionally women and men have been segregated into different occupations, with women over-represented i n occupations i n which care-giving is an integral component such as clerical, sales and service work, and men over-represented i n industrial occupations such as mining, forestry, agriculture, fishing, manufacturing and transportation.  35  A s well as an occupational  'See Jane Lewis, "Towards a Framework for Analysing the Position of Women under Income Security Policies" in E. Diane Pask, Kathleen E. Mahoney and Catherine Brown, eds., Women, the Law and the Economy (Toronto: Butterworths, 1984) 145 at 145 and Maureen Maloney, "What is the Appropriate Tax Unit for the 1990s and Beyond?" in Allan Maslove, ed., Issues in the Taxation of Individuals (Toronto: University of Toronto Press, 1994) 116 at 125-129. These articles discuss the assumption that women enter the work force not because of need but for other reasons secondary to men's workforce participation, and show that the reality of women's lives belies this assumption. See also the discussion below at pp. 90-92 which connects this assumption to familial ideology and the 'family wage'. See also Pat Armstrong, "Restructuring Public and Private: Women's Paid and Unpaid Work" in 3  Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law and Public Policy (Toronto: University  of Toronto Press, 1997)37. 32  See Armstrong, supra note 12 and ibid. Women in Canada, supra note 5 at 10.  K  lbid. at 18.  M  Armstrong, supra note 31 at 45. See also Karen Messing, "Do Men and Women Have Different Jobs Because of Their Biological Differences?" in Greta Hoffman Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside Limited, 1987) 341 and Pat Armstrong, "Women's Work: 35  -70-division, there is a wage division; jobs that employ primarily women pay less than jobs that employ primarily men.  36  While some important advances toward equality in the workforce have been made  over the past 50 years,  37  segregation o f work and wages has not yet been relegated to history:  Of the 200 jobs selected in the 1991 Census summary, 35 are at least 70 per cent female and 81 are at least 70 per cent male. More than two-thirds of the women, 68 per cent of them, are employed in these 35 female-dominated jobs... The gendered segregation is not limited to the kinds of work women do, however. It is also found in terms of wages. The salaries in the female dominated jobs range from $13,307 to $37,694, while those for male-dominated occupations go from$16,135 to $111,261. ... The wage segregation is particularly evident if we look at the ten lowest and ten highest paid occupations. Women accounted for only 20 per cent of those in the ten highest paid occupations, but made up over 70 per cent of those in the ten lowest paid occupations. 38  The 1996 census data show continued consistency in job segregation. O f the ten most frequent jobs for women only one, retail salespersons, is among the ten most frequent jobs for men. It is also 39  40  Women's Wages", supra note 12. "Ibid.  "Improvements include increased labour force participation by women, increased numbers of women belonging to unions (from below 25 per cent in 1970 to almost 33 per cent in 1992), improvement in women's average wages in relation to men's wages as well as absolutely, successful demands by women for equal pay for equal work, increased minimum wage, implementation of minimum employment standards, enactment of human rights legislation which prohibits discrimination based on gender and marital status, improved job security and benefits for pregnant women, and state required maternity leave. See Armstrong, supra note 31 at 47-48. Ibid. at 46.  n  Retail sales persons, secretaries, cashiers, registered nurses, accounting clerks, elementary teachers, food servers, general office clerks, babysitters and receptionists. These jobs accounted for 32 per cent of all jobs held by women. (The Daily, supra note 9.) 39  "Truck drivers, retail sales persons, janitors, retail trade managers, farmers, sales representatives (wholesale trade), motor vehicle mechanics, carpenters and construction trade helpers. The increased participation of men in retail sales reflects the fact that sales and service has become the largest sector of the Canadian economy and its growth is continuing while there is negative growth in the traditionally male dominated sectors of the economy. Urbanization is also a factor. "Retail salespersons and sales clerks was the leading job group among all workers in Canada's four  -71noteworthy that the ten most frequent jobs for women continue to be care-giving occupations.  In general, women cannot  41  and do not participate in the paid labour force to the same  extent as men because they bear children and have been assigned the responsibility for caregiving within both the home and the workplace. "Secondary status and income go with these roles."  42  Economic barriers for women in general are further "complicate[d] and deepen[ed]" for visible 43  minority, immigrant, First Nation and disabled women.  44  In addition to gender discrimination,  these women experience discrimination in the workforce based on race, ethnicity and disability.  45  largest census metropolitan areas in 1996." Ibid. At least until changes to the workplace that will alleviate the barriers to women's fuller participation are implemented, such as accessible, affordable and safe child care, as well as fairer maternity leave provisions. (See Nitya Iyer, "Some Mothers Are Better Than Others: A Re-examination of Maternity Benefits" in Susan B. Boyd, ed., 41  Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press,  1997) 168 and Katherine Teghtsoonian, "Who Pays for Caring for Children? Public Policy and the Devaluation of Women's Work" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy  (Toronto: University of Toronto Press, 1997) 113.) 42  Shelagh Day and Gwen Brodsky, Women and the Equality Deficit: Restructuring Canada's Social  Programs (Ottawa: Research Directorate, Status of Women Canada, 1998) at 7. Ibid.  Al  •"While this thesis focuses on gender inequality in the retirement income system, it is important to note that race, ethnicity and disability intersect with gender and consequently visible minority, First Nation, immigrant and disabled women experience barriers to their workforce participation in addition to the barriers imposed by gender alone. The failure of this thesis to explore these intersections in detail is not intended to minimize the plight of these women. Although their experiences are different and their economic disadvantages compounded, like women in general, visible minority, immigrant, First Nation and disabled women will carry their economic disadvantages into their retirement years. Since their economic disadvantages are greater in their pre-retirement years, their likelihood of poverty in retirement is also greater. Nitya Iyer explores the experiences of visible minority, First Nation and disabled women that 'complicate and deepen' gendered barriers to workforce participation in the context of access to maternity benefits. Her insights are informative in the context of understanding that the experience of gendered access, to the retirement income system as well as to maternity benefits, cannot be presumed to be homogeneous. (See Iyer, supra note 41. See also the discussion of the experiences of immigrant women in the garment industry by Amanda Araba Ocran, "Across the Home/Work Divide: Homework in Garment Manufacture and the Failure of Employment Regulation" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public  Policy (Toronto: University of Toronto Press, 1997) 144.) See Women in Canada, supra note 5 at 117-166. The statistical information reported in this document shows that visible minority, First Nation and disabled women earn less than their male counterparts and, as well, earn less than non disabled women who are Caucasian in race and white in colour. The experience of visible minority 45  -72The effect o f their double (perhaps even triple) disadvantage is that average incomes o f visible minority (including immigrant visible minority women), First Nation and disabled women are lower than incomes o f other women.  46  A s well, visible minority, immigrant, First Nation and  disabled women are over-represented among those whose incomes fall below the poverty l i n e .  47  Thus, a retirement income system that defines the extent o f access to its benefits by reference to the  value attributed  to a person's workforce participation effectively creates  systemic  discrimination against women, and for many women the experience o f that discrimination is only 48  partly based on gender.  (b)  Access T h r o u g h Spousal Relationship Access to pension benefits through a spousal relationship operates at all three tiers o f the  retirement income system. Even Tier I, which is not tied directly to work force participation,  women illustrates the causal connection between racism and inferior employment opportunities. Although visible minority women are more likely than other Canadian women to have a university degree (p. 136), they are less likely to be employed in professional or managerial jobs (p. 137). Though raced as well, the workforce experiences of First Nation women differs from those of other women. First Nation women participate in the workforce proportionally less than other women (pp. 151-152). One factor that contributes to lower workforce participation is that First Nation women have higher fertility rates than other women (p. 150). Unemployment is higher for immigrant (10.7 percent), visible minority (13.4 per cent), First Nation (17.7 per cent)and disabled (16.0 per cent) women as well, compared with other women workforce participants (9.9 per cent) (p. 9). * Ibid. at pp. 138, 152 and 166. 6  In 1990 the incidence of low income for immigrant women (white and visible minority) was 21 per cent; for visible minority women (including immigrant visible minority women) it was 28 per cent; for First Nation women it was 33 per cent. For non First Nation women it was 17 per cent and for non immigrant/visible minority women it was 16 per cent. (Ibid, at 123, 138 and 153.) 47  Caledon Institute, supra note 6, Appendix C - graphs entitled "Average CPP Retirement Benefit, by Sex, in Constant 1996 Dollars, 1988-1996"; "Taxfilers Contributing to RRSPs, by Age and Sex,1992 "; "Average Contribution to RRSPs by Age and Sex, 1992" and "Percentage of Seniors with Employer-Sponsored Pension Plan Benefits, By Sex, 1981 - 1992". In 1996 women's average CPP retirement benefit was approximately 58 per cent of men's. In 1992 at all age levels proportionately fewer women than men taxfilers contributed to RRSPs and women's contributions were, on average, less than those of men. Also in 1992, 57.75 per cent of men seniors and 30.49 per cent of women seniors received RPP benefits. 48  -73permits early access to Tier I benefits, through S P A , by a spouse o f an O A S recipient. A s for Tiers II and HI, surviving spouses are entitled to survivor benefits on the death o f their C P P contributor spouses and may be so entitled on the death o f their R P P member spouses. In addition, taxpayers with R R S P contribution room may choose to contribute some or all o f their R R S P savings to a spousal R R S P rather than to R R S P s in their own name.  Tier I: SPA The S P A component o f Tier I simply extends benefits equivalent to the O A S and GIS to needy people between the ages o f 60 and 65. However, only those who are either i n a spousal relationship with a GIS recipient or are widows or widowers are eligible. Need is determined by reference to the combined incomes o f the spouses and accordingly spousal status determines the amount o f S P A entitlement as well as eligibility. The effect o f S P A is that the family income o f a pensioner whose spouse is younger by five years or less w i l l be the same as it would be i f both spouses were pensioners.  49  Tier II: CPP Survivor Benefits Few women qualify for the maximum C P P retirement benefit because the C P P does not recognize unpaid (reproductive) work which takes many women out o f the paid work force for at least a portion o f the time period that represents a normal working life for a male worker. In addition, women tend to be employed in lower paying jobs and earn less than the A I W .  5 0  However,  John Burbidge, Social Security in Canada: An Economic Appraisal (Toronto: Canadian Tax Foundation, 1987) at 12-13. 49  50  Armstrong, supra notes 12 and 31.  -74a women who is or was in spousal relationship with a man who is a C P P member is entitled to a retirement benefit, referred to as a survivor benefit, on the death o f her spouse.  51  The maximum  survivor benefit is 60 per cent o f the deceased spouse's retirement benefit provided that the combined survivor benefit and the surviving spouse's own C P P benefit cannot exceed the maximum C P P retirement benefit payable to an individual. The age o f the surviving spouse, whether she has dependent children and whether she is or becomes disabled are additional factors that affect the amount o f her survivor benefit. The full benefit o f 60 per cent o f the deceased's spouse's benefit is payable only to a surviving spouse aged 65 or above.  Tier III: RPP Survivor Benefits Women face the same kinds o f obstacles with RPPs as they do with the C P P - shorter 'working' life (due to time away from paid employment to do unpaid reproductive work) and lower wages (due to undervalued 'women's' work). However, a women who is in spousal relationship with a man who is a R P P member may be entitled to a pension benefit on the death o f her spouse. In the late 1980s legislation governing RPPs was reformed to make it mandatory for R P P s to provide a pension o f no less than 60 per cent o f the contributor's benefit to the surviving spouse of a deceased contributor who dies after retirement.  52  If the contributor dies before retirement, "the  death benefit is either a lump sum to the beneficiary or an annuity to the spouse equal to the value o f the member's pension credits at the time o f death."  53  These surviving spouse benefits operate  'Registration restrictions set out in the ITA limit entitlement to survivor benefits to a spouse, former spouse or dependant of a deceased RPP member. (ITA, supra note 2, Reg. 8502(c).) 5  "Robert L. Brown, Economic Security in an Aging Population (Toronto: Butterworths, 1991) at 63-64; Monica Townson, supra note 29 at 48-49. "Brown, ibid, at 64.  -75on existing and new plans only from the date that the relevant legislation came into force, and a written waiver signed by both spouses w i l l override these provisions.  54  Tier III: Spousal RRSPs Because R R S P contribution limits are tied to income level, women generally qualify for lower contribution levels than men. However, a taxpayer may chose to contribute to a spousal R R S P rather than an R R S P i n her or his own name, although the taxpayer's contribution limit remains the same regardless o f where the contributions are directed.  55  This provision operates as  an incentive to the higher income earning spouse (husband) to provide for retirement needs o f the dependent non or low income earning spouse (wife). The advantage o f spousal R R S P s is that retirement income from this source can be split so that it w i l l be taxed in the hands o f the lower income spouse, effecting lower overall tax liability for the spousal unit. Since men generally earn more than women for the reasons previously discussed, spousal R R S P s are generally in the name of the female spouse.  (c)  Rationale for Spousal Provisions  Spouses are Women A l l three tiers of the retirement income system have their own definitions o f 'spouse'. F o r the purposes o f the S P A , "spouse" is defined by the Old Age Security Act , s. 2: 56  'Brown, ibid, and Townson, supra note 29. 'ITA, supra note 2, s. 146. 'R.S.C. 1985, c. 0-9.  -762. In this Act, "Spouse", in relation to any person, includes a person of the opposite sex who is living with that person, having lived with that person for at least one year, if the two persons have publicly represented themselves as husband and wife;  W h i l e a person could have both a married and a common law spouse under this definition, both would not qualify for the S P A benefit.  57  The Canada Pension Plan™ defines "spouse" as:  2. (1) In this Act, "spouse", in relation to a contributor, means, (a) except in or in relation to section 55, (i) if there is no person described in subparagraph (ii), a person who is married to the contributor at the relevant time, or (ii) a person of the opposite sex who is cohabiting with the contributor in a conjugal relationship at the relevant time, having so cohabited with the contributor for a continuous period of at least one year,  Under this definition a married spouse who is separated from her C P P contributor or recipient husband may be eligible for the surviving spouse's benefit as long as her husband did not also have a common law spouse.  59  Access to Tier m spousal provisions, including R P P survivor benefits and  spousal R R S P s , is determined by the definition o f "spouse" in the I T A :  Subs. 252(4) In this Act, (a) words referring to a spouse at any time of a taxpayer include the person of the opposite sex who cohabits at that time with the taxpayer in a conjugal relationship and  Ibid., subsection 19(1) provides that "...a spouse's allowance may be paid to the spouse of a pensioner if the spouse (a) is not separated from the pensioner..." 51  58  R.S.C. 1985, c. C-8.  The eligibility of a surviving spouse under age 65 for CPP survival benefits depends on whether she has reached the age of 35 years, has dependent children or is disabled. (Ibid., para. 44(1 )(d).) 59  -77(i) has so cohabited with the taxpayer throughout a 12-month period ending before that time, or (ii) is a parent of a child of whom the taxpayer is a parent...  It is only since 1994 that there has been consistency throughout the retirement income system in recognizing common law heterosexual relationships o f at least one year duration as spousal relationships.  60  In A p r i l , 1998, for the limited purpose o f its application "to the registration o f  pension plans or amendments to registered pension plans", the Ontario Court o f Appeal extended the definition o f 'spouse' in the I T A to include same sex persons who cohabit in a conjugal relationship.  61  While the legislation sets limits on who is a spouse, meaning is also derived from context. When S P A was enacted in 1975, 90 per cent o f the persons eligible for these benefits were women.  62  89 per cent of C P P survivor benefits go to women and it is likely that a similar pattern 63  applies to R P P survivor benefits. Although "[o]nly 12 % o f women who accumulated R R S P savings i n 1989 received all or part o f their deposits from husbands",  64  in 1991 essentially all  contributions to spousal R R S P s were made by husbands to their wives' R R S P accounts.  65  In the  The definition of "spouse" under the ITA was amended to include common law relationships in 1994. (S.C. 1994, c. 7, s. 140.) The SPA benefit was restricted to legally married couples until 1975 when the definition of "spouse" was amended to include common law heterosexual relationships. (S.C. 1974-75-76, c. 58, s. 1.) CPP benefits were extended to common law couples by S.C. 1986, c. 38, s. 1(3) which added the definition of 'spouse' to the 60  Canada Pension Plan.  "Rosenberg v. Canada (Attorney General) (1995), 25 O.R. (3d) 612 (G.D.), reversed [1998] O.J. No. 1627 (Docket No. C22807, April 23, 1998). See discussion in Chapter Three, below. 62  Dennis Guest, The Emergence ofSocial Security In Canada 3 ed. (Vancouver, UBC Press, 1997) at 187.  63  rd  Caledon Institute, supra note 6 at 13.  ^Townson, supra note 29 at 57. "Hubert Frenken, "Women and RRSPs" (1991) 3:4 Perspectives on Labour and Income 8.  -78context o f Canada's retirement income system, then, spouses are generally women, and spousal provisions represent legislative initiatives to provide some relief for the unfairness of the workforce related access to the pension system.  66  B u t N o t A l l W o m e n are Spouses The flaw in providing relief for pension system unfairness through spousal provisions is that, while a spouse is generally a woman in the context o f the retirement income system, not all women are spouses.  This was noted during the debate on the bill amending the Old Age Security  Act to include S P A :  Stanley Knowles, the veteran New Democratic Party member of Parliament, made the most pertinent comment. He pointed out that the Spouse's Allowance Bill treated women as dependants rather than as persons: 'They get a pension on the basis of being able to answer the question: Have you got a man? Dead or alive, have you got a man?' 67  Women who do not 'have a man, dead or alive', that is women who never married, are divorced or separated, or who are in lesbian relationships do not qualify for S P A or C P P benefits that are tied to a spousal relationship, or for spousal R R S P s , although since A p r i l 1998 they may qualify for R P P spousal benefits. Also excluded from S P A , C P P and R R S P spousal benefits are dependent partners i n gay male relationships - they "have a man" but their relationships do not meet the heterosexual requirement.  68  While Tier I eligibility is not directly related to workforce participation, the age of eligibility for OAS and GIS is tied to the normal age of retirement from the work force, age 65. 66  67  Guest, supra note 62 at 188.  The exclusion of gay and lesbian couples from spousal benefits has been the subject of litigation based on the argument that the exclusionary provision of spousal benefits is an example of heterosexual privilege. This issue 68  and the decisions in Egan v. Canada, [1995] 2 S.C.R. 513 and Rosenberg v. Canada (Attorney General), supra note  61 are discussed below in Chapter Three.  -79(d)  T h e Longevity F a c t o r The unfairness for women inherent within the retirement income system is compounded  by the reality that women tend to live longer than men.  69  This means that women are much more  likely than men to outlive their spouses and to spend their final years alone. ' A s noted previously 70  7  in the discussion o f S P A , men tend to marry women who are younger than themselves. Therefore the length o f time that an elderly widow can expect to live alone w i l l likely be greater than the discrepancy between the life expectancies of men and women.  72  When one considers that close to  one-half o f unattached individuals (52.9 per cent for women and 31.8 per cent for men) aged 65 and over had incomes below the poverty line in 1994, compared with 7.1 per cent for families 73  headed by persons aged 65 and older, the impact o f women's longevity on their financial well74  being is both obvious and startling. A s Louise Dulude has put it:  The life expectancy for a woman born in 1991 is 80.9 years, while for a man it is 74.6 years. A woman aged 65 in 1991 could expect to life an additional 19.9 years, to age 84.9; a man aged 65 years in 1991 could expect to live an additional 15.7 years, to age 80.7. (Women in Canada, supra note 5 at 41.) 69  In 1995 women comprised 57.8 per cent of the Canadian population aged 65 and over, a slight drop from 1991 when women comprised 58 per cent of the population in the over 65 age group. (Statistics Canada, A Portrait ofSeniors in Canada, 2 ed. (Ottawa: Statistics Canada, 1997) (hereinafter "Portrait ofSeniors") at 18.) Within age groups under the age of 65, in 1991 the percentage of women ranged from 48.8 per cent to 50.7 per cent. When the over 65 age group is further segmented, the ratio of women to men increases with age. In 1991 women comprised 55.1 per cent of the population aged 65-74, 60.4 per cent of the population aged 75-84 and 69.5 per cent of the population aged 85 and over. (Women in Canada, ibid, at 14.) While a similar breakdown in relation to more recent statistics has not yet been published, there is no reason to expect that the pattern of the proportion of women to men increasing with age within the over 65 age group has changed significantly since 1991. 70  nd  "Statistics Canada analysis of 1996 census data shows that almost half of women living alone in 1996 were aged 65 and over, and that three-quarters of all people living alone and aged 65 and older were women. (Statistics Canada, The Daily, No. 11-001-XPE (Tuesday, June 9, 1998) at 3.) Analysis of 1991 census data shows that 79.5 per cent of men and 53.9 per cent of women aged 65-74 lived with a spouse or common law partner. For people aged 75-84 thefiguresare 71.5 per cent for men and 29.6 per cent for women, and at age 85 and older the figures are 53 per cent for men and 10.7 per cent for women. (Portrait of 72  Seniors, supra note 70 at 36.)  Caledon Institute, supra note 6, Appendix C - graph entitled "Poverty Rates, Elderly Unattached Individuals, By Sex, 1980-1994". 73  Ibid., graph entitled "Poverty Rates, Families with Elderly Heads and All Families, 1969-1994".  14  -80The most tragic fact of life for old women is that the blessing of their superior longevity too often turns out to be a curse. Instead of more twilight years of fulfilment and serenity, it brings to many of them a decade or more of loneliness, ill health and poverty. 75  Women's superior longevity has implications that connect to gendered poverty at all three tiers o f the retirement income system. Statistics demonstrate that Tier I benefits are too low for unattached seniors; those who must rely on these benefits only are condemned to living on an income that falls seriously below the poverty line.  76  Their longevity means that more women than  men are unattached seniors. The survivors benefit for C P P (Tier II) is 60 per cent o f a deceased spouse's retirement benefit. Because o f their longevity, more women than men w i l l have a C P P benefit that consists o f the full retirement benefit o f the lower income earning spouse (usually the wife) plus 60 per cent o f the retirement benefit o f the higher income earning spouse. That is, on the death o f a spouse a surviving husband is more likely to be financially better off than a surviving wife.  77  The same reasoning applies in relation to R P P defined benefit plan survivor benefits,  although the gendered effect is compounded by the fact that only a portion o f the workforce are members o f defined benefit R P P s and the majority o f these are men. In addition, unlike C P P many defined benefit R P P s are not indexed to the cost o f living, a vital concern for women who can expect to live for approximately 20 years beyond age 65, and probably more than that beyond the age o f her husband's retirement.  78  A s for R R S P s and defined contribution (money purchase) R P P s ,  "Louise Dulude, "Getting Old: Men in Couples and Women Alone" in Greta Hoffman Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside Limited, 1987) 323 at 324. Caledon Institute, supra note 6. See also graph entitled "Maximum Old Age Security/Guaranteed Income Supplement versus Poverty Line for Different Areas, Single Senior, 1995". 76  This makes the issue of whether 60 per cent is an adequate survivor benefit much more crucial for women than for men. 77  ™Supra note 69.  -81maximum contribution limits are based on amounts projected as sufficient to purchase a life annuity from an insurance company to replace income on retirement for the remainder o f the annuitant's life.  Since a woman is statistically likely to live longer than a man, "insurance  companies charge women more for the same pension benefits, or else give women lower pensions for the same amount."  79  Therefore it makes sense that a woman's maximum contribution level  should be higher than a man's so that she can purchase an annuity that w i l l provide, until her death, the same income as a man in her financial circumstances. Alternatively, the use o f gender based annuity tables to set the purchase price of, or benefits from, a life annuity should be prohibited by law.  (e)  80  N e u t r a l is N o t E n o u g h Although neither o f the means o f access to pension benefits, through workforce  participation or a spousal relationship, overtly favours one gender over the other, in their effects both are implicated in the gendered poverty o f the elderly.  81  The cause of [the] huge gap in the living standards of elderly men and women is that Canada's pension system was designed by men to benefit men. This was not done to spite women, but out of the traditional belief that the world is composed of only two categories of people: full-time participants in the labour market (husbands and fathers), and the people they support (women and children). If you provide pensions to the first group, it was felt, the second group would automatically be taken care of. The problem is that in the real world, people do not fit in such convenient  Louise Dulude, Pension Reform With Women in Mind (Ottawa: Canadian Advisory Council on the Status of Women, 1981) at 12. 79  80  See Peter Bartlett, "Sexual Distinctions in Pensions" (1986) 24 Osgoode Hall L. J. 833 at 849-865.  "'Caledon Institute, supra note 6 shows that women who are not in spousal relationships (ie. unattached) are the poorest of the elderly.  -82categories. About 10 per cent of women never marry, more than half of those who do marry become widowed at some point, and the rate of marriage dissolution through divorce and separation is growing by leaps and bounds. As a result, the National Council of Welfare estimated, fully three-quarters of all women end up having to support themselves and, increasingly, their children as well. Since very few of these "unprotected" women had life-long full-time jobs giving them access to good pension plans and allowing them to accumulate substantial savings, it is inevitable that most of them will end up poor when they are old. 82  While Dulude attributes the design flaw to men's traditional belief in a simple dichotomized world of breadwinners and dependants, Chunn argues that women must accept complicity in accepting and promoting this view as w e l l .  83  Liberal feminists, who dominated the feminist movement in the  1960's through the m i d 1970s, fought for, and achieved, gender neutrality in legislation as recognition o f gender equality.  Access to the retirement income system, whether through  workforce participation or a spousal relationship, was acceptable to liberal feminists as long as the rules governing such access were expressed in gender neutral terms.  Gender neutral rules have failed to achieve fairer access to the retirement income system for women.  84  The persistent over-representation o f elderly women among those whose incomes  Louise Dulude, supra note 75 at 330 -331. The trend is that increasing numbers of women live alone. In 1991, 12.3 percent of women over the age of 15 lived alone, up from 6.6 percent in 1971 and 10.6 per cent in 1981. 38.1 per cent of women aged 65 and over lived alone in 1991. (Women in Canada, supra note 5 at 21.) The trend appears to have continued in the 1996 census. Although gender breakdowns are not available as yet, the number of one person households increased by 15 per cent from 1991 to 1996. "[0]ne person households accounted for 24% of all households in 1996, up from 20% in 1981." 36 per cent of persons living alone in 1996 were aged 65 and over, and approximately 75 per cent of these were women. (The Daily, supra note 9.) 82  s3  Chunn, supra note 4.  The positions taken by the SCC justices in Egan v. Canada, supra note 68 (discussed in Chapter Three, below) notwithstanding, many who have written about this issue agree that the spousal provisions are intended to provide some relief for women who are systemically limited by their ability to access retirement income through work force participation and by the failure of the system to recognize unpaid reproductive work. See Claire F.L. Young, "Public Taxes, Privatizing Effects, and Gender Inequality" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 307; Bartlett, supra note 80; Neil Brooks, "The Irrelevance of Conjugal Relationships in Assessing Tax Liability" in John G. Head and Richard Krever, eds., Tax Units and the Tax Rate Scale, (Sydney: The Australian Tax Research Foundation, 1996) 35. 84  -83are below the poverty line, and particularly the over-representation o f single elderly women among the poorest o f the poor, underscores this failure. The durability o f gender inequality in the face o f gender neutral reform initiatives by liberal feminists became, and continues to be, a focal point for feminists o f the 1980s and 1990s. A m o n g these are socialist feminists, who perceive the law as informed by an existing social superstructure and reason that changes to the law alone cannot alter inequalities that are integral to the superstructure.  85  A s Susan B o y d observed, these feminists are  interested i n investigating "the relationship between law and ideology... to explore and explain the elastic capacity o f the legal system to absorb reforms initiated by the women's movement and turn them into patriarchal ends." The ideology that many feminists consider to be central to explaining 86  the perpetuation o f women's inequality is familial ideology, which reinforces patriarchy and women's subordination.  87  What follows is an exploration o f the relationship between familial  ideology and the spousal provisions o f the retirement income system and its implications for improving the economic circumstances o f elderly women.  II  WOMEN: REAL OR IDEAL?  (a)  W h a t is ideology? Before employing the concept of'ideology' to provide theoretical insight into the gendered  effects o f the spousal provisions in legislation governing the retirement income system, it is helpful,  85  Chunn, supra note 4.  Susan Boyd, "Child Custody, Ideologies, and Employment" (1989) 3 Canadian Journal of Women and the Law 111 at 114. 86  87  Chunn, supra note 4 at 183.  -84i f not necessary, to clarify the "difficult, slippery and ambiguous"  88  concept itself.  Ideology is a concept that was imported into legal theory from social theory, generally associated with Marxist social theory.  90  89  and is  The concept eludes a unified definition; Marxist  literary critic Terry Eagleton identifies no less than sixteen definitions o f ' ideology' i n current use,  91  some o f which are incompatible with others and all o f which, in Eagleton's analysis, are problematic to some extent. Eagleton prefers a functional description o f the concept:  [Ideology] is neither a set of diffuse discourses nor a seamless whole; if its impulse is to identify and homogenize, it is nevertheless scarred and disarticulated by its relational character; by the conflicting interests among which it must ceaselessly negotiate. It is not itself as some historicist Marxism would seem to suggest, the founding principle of social unity, but rather strives in the teeth of political resistence to reconstitute that unity at an imaginary level. As such it can never be ' other-worldliness' or idly disconnected thought; on the contrary, it must figure as an organizing social force which actively constructs human subjects at the roots of their lived experience and seeks to equip them with forms of value and belief relevant to their specific social tasks and to the general reproduction of the social order....Ideology contributes to the constitution of social interests, rather than passively reflecting pre-given positions; but it does not, for all that, legislate such positions into existence by its own discursive omnipotence. 92  While Eagleton's insights into how ideologies operate give meaning to the concept, Shelley Gavigan, drawing on the work o f sociologist Roger Cotterrell, adds a further dimension to this meaning by identifying three characteristics o f 'ideological thought':  Alan Hunt, "The Ideology of Law: Advances and Problems in Recent Applications of the Concept of Ideology to the Analysis of Law" (1985) 19 Law & Society Review 11 at 31. 88  Shelley A . M . Gavigan, "Law, Gender and Ideology" in Anne Bayefsky, ed., Legal Theory Meets Legal Practice (Edmonton: Academic Printing and Publishing, 1987) 283 at 285. 89  90  9l  Hunt, supra note 88 at 12.  Terry Eagleton, Ideology: An Introduction (London: Verso, 1991) at 1-2. Ibid. at 222-223.  92  -851. It appears to be 'common sense', obvious and natural and hence not requiring specific justification. It provides a basic structure of perceptions and beliefs in relation to which experience is interpreted. 2. This structure of beliefs and perceptions tends to assert its own completeness and timeliness. 3. This claim to completeness and self-sufficiency is maintained by emotional commitments which may justify selective consideration of empirical evidence.  93  Thus ideologies seek to reproduce the existing social order and operate to stop, or at least dilute the force of, criticism by those who question the validity o f the social and economic status quo. The power o f ideologies is in their resistence to inquiry - there is simply no need to inquire into, to justify, the 'obvious and natural'.  94  This 'common sense' aspect o f ideology is examined by  Eagleton as well:  Successful ideologies are often thought to render their beliefs natural and self evident - to identify them with the 'common sense' of a society so that nobody could imagine how they might ever be different. ...On this view a ruling ideology does not so much combat alternative ideas as thrust them beyond the very bounds of the thinkable. 95  The problem is that the 'obvious and natural' may not reflect the real lives o f all social interests that are subsumed within a particular ideology. This is not to suggest that the 'obvious and natural' is a myth constructed by the powerful to fool the masses into unquestioningly accepting their subservience. A s Gavigan points out:  It is important to be clear that ideology must mean more than false consciousness, false image, or myth. Ideologies must have some legitimizing component which likewise does not mean that people are duped but rather that something 'real' is  93  Gavigan, supra note 89 at 121-2.  This is obviously an oversimplification of a complex social theory. See Hunt, Gavigan and Eagleton, supra notes 88, 89 and 91 for reasoned analyses of the effectivity of ideologies. 94  95  Eagleton, supra note 91 at 58.  -86at times gained or reflected in the ideology. For instance, while research on family violence has illustrated that 'the home', farfrombeing a haven for women may be more aptly considered a danger zone, the strength of familial ideology is such that even feminists are hard-pressed to be critical of the family. 96  Gavigan's family violence example raises the interesting question o f why people accept as true ideas that they experience as contradictions. W h y is the idea o f the family as a safe haven unquestionable when the experience of domestic violence contradicts that 'truth'? Eagleton argues that, while contradictions may be experienced, a feature of a dominant ideology is that it operates 97  to inculcate beliefs that any injustices associated with the contradiction are either  ...en route to being amended, or that they are counterbalanced by greater benefits, or that they are inevitable, or that they are not really injustices at all. [A dominant ideology] can do this either by falsifying social reality, suppressing and excluding certain unwelcome features of it, or suggesting that these features cannot be avoided. 98  Ideologies, then, resist inquiry even by those who recognize that they may not reflect the 'obvious and natural'. However, for those whose interests are impaired by the effects o f particular ideologies, such inquiries are crucial to moving towards eliminating the impairment. A s Eagleton aptly put it, "[t]he study of ideology is among other things an inquiry into the ways in which people may come to invest in their own unhappiness."  99  To return again to Gavigan's example, inquiry  into familial ideology is crucial to understanding domestic violence, and failure to move toward understanding is tantamount to accepting, or investing in, the injustice.  Gavigan, supra note 89 at 293. Hunt, supra note 88 and Eagleton, supra note 91 also reject the notion that ideologies are based on 'false consciousness', though they may involve falsity. 96  A dominant ideology is an ideology that operates as a force to legitimate the power of a dominant social group or class. (Eagleton, supra note 91 at 5.) 97  *Ibid. at 27.  9  "Ibid, at xiii.  -87Building on advances by neo-Marxist social theorists i n reformulating the traditional Marxist concept o f ideology, A l a n Hunt proposed that the concept o f ideology be used in legal analysis "to explore the connection between ideas, attitudes, and beliefs, on the one hand, and economic and political interests, on the other." irreplaceable tool o f analysis" reproduction o f hegemony,"  102  100  Hunt regards ideology as "an indispensable and  for the study o f the relationship o f law "to the production and  101  and "the way legal ideology is articulated in conjunction with  economic, political, and cultural ideologies."  103  The presence over the past decade o f ideological  analyses i n numerous works by feminist legal scholars interested in "understanding the contribution of the law to the social position o f w o m e n "  104  underscores the value o f ideology as an tool in legal  analysis.  l00  Hunt, supra note 88 at 13.  Ibid. at 31.  m  Ibid.  m  Ibid.  m  '""Gavigan, supra note 89 at 285. See also Shelley Gavigan, "Paradise Lost, Paradox Revisited: The Implications of Familial Ideology for Feminist, Lesbian and Gay Engagement to Law" (1993) 31 Osgoode Hall L.J. 589; Boyd, supra note 86; Susan Boyd, "(Re)Placing the State: Family, Law and Oppression" (1994) 9 Canadian Journal of Law and Society 39; Susan Boyd, "Can Law Challenge the Public/Private Divide? Women, Work, and Family" (1996) Windsor Yearbook of Access to Justice 161; Susan Boyd, "Challenging the Public/Private Divide: An Overview" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy  (Toronto: University of Toronto Press, 1997) 3; Carol Smart, "The Women of Legal Discourse" (1992) 1 Social and Legal Studies 29; Marlee Kline, "Complicating the Ideology of Motherhood: Child Welfare and First Nation Women" (1993) 18 Queen's Law Journal 306; Lisa Philipps and Margot Young "Sex, Tax and the Charter: A Review of Thibideau v. Canada" (1995) 2 Review of Constitutional Studies 221; Janine Brodie, Politics on the Margins: Restructuring and the Canadian Women's Movement (Halifax: Fernwood Publishing, 1995); Jane Ursel, Private Lives, Public Policy: 100 Years of State Intervention in the Family (Toronto: Women's Press, 1992).  -88(b)  T h e Ideology of the F a m i l y Several ideologies have been identified as relevant to feminist legal analysis.  105  However,  the relationship between familial ideology and the oppression of and discrimination against women is a central concern in feminist legal research directed toward understanding women's inequality.  106  Gavigan summarizes the importance o f familial ideology as follows:  In my view, the importance of analysing law as ideology is best illustrated by a consideration of 'the family', writ large as it often is. 'The family' is presented both in law and in popular culture as the basic unit in society, a sacred, timeless and so natural institution that its definition is self evident. Its privacy is sought to be protected and its sanctity proclaimed. That it is the fittest place to raise children is again so self-evident as to not merit question, and the hold of the family is strong despite the knowledge that large numbers of individuals live in households which bear no resemblance to the ideal family. (The ideal family, despite the gender-neutral references to 'spouse' and 'parent' in legislation, is still taken to mean a social relationship, sanctified by law and preferably by religion, comprising a male adult, female adult, and their biological or adopted children.) 107  Within the idealized family ("heterosexual, privatized, nuclear, largely middle class and white" ) 108  women and men are assigned gender specific roles - "the wife/mother is a full-time homemaker and the husband/father is the sole breadwinner." woman's economic dependence on a man.  109  Thus one feature o f familial ideology is  While only a minority o f families may actually  For example the ideology of motherhood, public/private ideology, the ideology of equality, restructuring  105  ideology. Gavigan, supra note 89 at 290-294. See also Gavigan, supra note 104; Frances E. Olsen, "The Family and the Market: A Study of Ideology and Legal Reform" (1983) 96 Harvard L.R. 1497; Chunn, supra note 4, Boyd, supra notes 86 and 104; Marlee Kline, "Blue Meanies in Alberta: Tory Tactics and Privatization of Child Welfare" 106  in Susan B. Boyd, ed.,  Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto:  University of Toronto Press, 1997) 330; Claire Young, supra note 84. l07  Gavigan, supra note 89 at 293.  ,08  Boyd, "(Re)Placing the State...", supra note 104 at 60.  109  Chunn, supra note 4 at 183.  -89resemble the idealized family,  110  that family forms the content o f dominant family ideology. ' This 11  dominant ideology, with its socially constructed gender roles, reinforces and perpetuates patriarchal relations and the gender inequalities that are integral to those relations.  112  A second, though related, aspect of dominant familial ideology is reification o f the family as a safe haven from the outside world:  ...home life was celebrated as the reward for which men should be willing to suffer in the earthly world of work. The family and home were seen as safe repositories for the virtues and emotions that people believed were being banished from the world of commerce and industry." 3  This view o f home and family "encouraged women to be generous and nurturing but discouraged them from being strong and self-reliant; it insulated women from the world's corruptions but denied them the world's stimulation."  114  'Idealized family' members are assumed to be altruistic  in their relations with each other - individual interests are subordinated to the good o f the whole family. Sharing and sacrifice among family members is considered to be the norm.  Although dominant familial ideology is contradicted by life experience (ie. many, i f not  Ibid.  m  "'Gavigan, supra note 104 at 597. Ibid.; Boyd, supra note 104 at 114-115; Chunn, supra note 4 at 183.  u2  " 01sen, supra note 106 at 1499. 3  Ibid. at 1500.  m  -90most, families do not fit the n o r m " and homes are frequently unsafe for women and children' ), 5  as Eagleton and Gavigan have pointed out'  16  17  this does not mean that the ideological representation  of the ideal family as 'obvious and natural' is based upon a collective false consciousness. However, it does mean that the ideology operates to mask the differences between 'ideal' and 'experienced' families.  118  What, then, are the injustices for women associated with the differences  between 'ideal' and 'experienced' families that are effected (and masked) by dominant familial ideology?  Familial Ideology and Dependency The Family Wage Familial ideology assumes that within families there is a direct correlation between gender and economic independence - men are independent and women are dependent on m e n . "  9  This  follows naturally from the assigned gender roles o f husband as breadwinner and wife as homemaker, which in turn assigns gender locations - men go out into the workforce and women stay at home. Associated with this gendered division o f labour are the family wage concept, that "...a working man should earn enough to support his f a m i l y " ' , and the notion that any wage 20  '"Mary Jane Mossman and Morag MacLean, "Family Law and Social Welfare: Toward a New Equality" (1986) 5 Canadian Journal of Family Law 78 at 104. " Gavigan, supra note 89 and supra note 104 at 601-609. 6  " See discussion above, at p. 84-86 7  ' '"Eagleton, supra note 91 at 5. ' The dependency of women on men within the idealized family is part of the patriarchal nature of that social unit and its sexual division of labour. Accordingly this is a central issue in feminist critique of the family. This simplified overview of the issue is drawnfrommy understanding of the writings of Chunn, Boyd, Gavigan, Olsen and Mossman and MacLean, cited supra at notes 4, 86, 89, 104, 106 and 115. l9  120  Gavigan, supra note 104 at 599.  -91eamed by a wife was secondary. A s Jane Lewis points out, both men and women endorsed the family wage concept because o f its promise o f an easier life style for working class families, and "[t]he argument that men had families to support... played a crucial part in union wage bargaining strategies..." . 121  The economic effect for women o f their assumed 'obvious and natural'  dependence on a man, then, was that political energies were directed at reinforcing and reproducing that dependency relationship through the struggle for breadwinners (men) to be paid a family wage.  Gavigan suggests that women paid a price in the quest for the family wage - their work force participation was ghettoised and poorly paid.  122  Although there have been improvements to  workforce accessibility and wages for women i n recent years,  123  segregation o f the labour force,  the effect o f the compromised distribution o f work and wages, "has remained remarkably stable throughout the postwar period."  124  In spite of the price women paid for it, the promise o f an easier  life for families through the family wage was not realized. In terms o f providing an adequate family income "for many, i f not most, working class people" , the work force participation o f 125  women in families is not, i f it ever was, secondary. Many women's families depended on their wages, and their real lives were (and are) lived straddling the workforce/family divide doing double duty at unpaid work in the home and underpaid work i n the labour market.  12l  Lewis, supra note 31 at 146.  Gavigan, supra note 104 at 599 notes 34-35. See also Lewis, supra note 31, Armstrong, supra note 12 at 369-370 and Messing, supra note 35. See also the discussion above at pp. 69-71. 122  Supra note 37.  m  124  l25  Armstrong, supra note 31 at 47.  Gavigan, supra note 104 at 599.  -92A s the work and wage segregation o f women illustrates, the 'natural and obvious' dependency o f women on men inscribed in familial ideology is not limited to women who are in familial relationships. Accordingly women who must enter the workforce to support families, whether as sole supporters or so-called 'secondary' earners, are unjustly affected by the gendered work and wage differentials which familial ideology reinforces. Speaking in general terms, they must support their families on lower wages that their male counterparts.  Women who support  themselves are similarly unjustly affected in comparison with their male counterparts.  Thus  familial ideology informs processes that penalize economically women who fail to fulfill their prescribed roles in the idealized family.  126  Dependency and Family L a w The assumed dependency o f women is particularly problematic in light o f the significant increase in the divorce rate over the past few decades, with a corresponding growth in the number o f lone parent families with dependent children headed by women.  127  O n divorce (ie. when they  step outside their ideologically assigned role) women are assumed to be as economically selfsufficient as their husbands, or capable of becoming so within a relatively short period o f time.  128  Women whose real livesfitthis idealized family role are economically penalized as well, to the extent that economic resources are not shared equally within their families. See infra note 144. (Like all generalizations, there are, of course, exceptions. In the case of familial ideology affecting the level of remuneration that women in the work force receive, the exceptions (ie. women and men are remunerated equally) are more likely to be found where women have infiltrated traditionally male occupations and follow a male work pattern.) l26  As Mossman and Maclean, supra note 115 at 87 have noted, following a marriage breakdown women are more likely than men to have custody of children. While the proportion of lone parent families headed by women has remained constant at around 80 per cent from the 1960s to the 1991 census, the number of lone parent families has increased over the years; in 1971 5 per cent of women were lone parents and by 1991 the proportion climbed to 7 per l27  cent. (Women in Canada, supra note 5 at 1 and 18.) l28  Mossman and MacLean describe a divorced wife's entitlement to support as "'rehabilitative' support".  Ibid, at 94.  -93This assumption operates unfairly for women with (custody of) dependent children as well as for those whose children are no longer dependent.  [e]ven where a divorce occurs after children have grown up, a wife who has worked for many years as homemaker and child care provider faces greater obstacles to becoming self-sufficient, having lived a lifetime of economic dependence with an expectation that her full-time homemaking would lead to economic security in old age. The obligation of self-support appears especially harsh for such women because their failure to participate in the paid labour force often results in a denial of access to pension entitlement. This situation further contributes to the problem of poverty among elderly women in Canada who rely for security on the basic federal pension. 129  Given the economic disadvantages  130  that women in general, and particularly those with  responsibility for caring for and supporting their children, face in the work force, the effect o f assuming a woman's independence immediately on the breakdown o f a relationship that valorizes her economic dependence and reinforces her responsibility for unpaid work in the home is that "women bear the costs o f marriage breakdown" . The assumed dependency o f women within the 131  idealized family is thus implicated in the 'feminization o f poverty' when the marital relationship comes to an end.  Even for those women whose realities conform to the idealized family and therefore resist the characterization o f their relationships with their husbands as oppressive, the tenuousness o f  Ibid. at 94-95.  m  Lower wages and work segregation are only part of the picture; lack of access to child care and the gendered responsibility for unpaid work in the home contribute further to economic disadvantage for women in the workplace. l30  13  'Mossman and Maclean, supra note 115 at 103.  -94their economic (and social ) status illustrates their dependency: 132  In the context of family law there is ... increasing masking of gender role differences within the family, and economic consequences which such differences can produce, especially when a relationship breaks up: '...the structure of marriage legally, economically and normatively, subordinates women to their husbands. While a good relationship can overcome that structural subordination, the minute the marriage breaks down or ends (through death, divorce or desertion) the unmediated structure of control reasserts itself. Millions of women leading comfortable lives discover to their dismay that they are just a divorce or death away from poverty.' 133  Contradictory Effects: Family and Welfare L a w Women who depart from their assigned role as dependent spouse, whether voluntarily or involuntarily, may find themselves facing another type o f dependency relationship. If she is unable to overcome the obstacles to her economic independence discussed in the previous section, she may have no choice but to turn to the state for economic support. Welfare law defines the dependency relationship between an individual and the state. It starts with the presumption that a woman is dependent on a man and w i l l deny benefits to women who cannot demonstrate the absence o f such support.  134  B y assuming that a woman is, or ought to be, dependent on a man and not the state, the  state may avoid a costly dependency relationship. O n the other hand, spousal support law defines  Related to the dependency role ascribed to women through familial ideology is the observation that gender and class intersect, and that women obtain their class status from their relationships with men. (Gavigan, supra note 104 at 594.) l32  l33  Boyd, supra note 86 at 123, quoting from Meg Luxton and Harriet Rosenberg, Through the Kitchen  Window: The Politics of Home and Family (Toronto: Garamond Press, 1986) at 11.  Mossman and Maclean, supra note 115 at 95-100. The authors' discussionof "male breadwinner" welfare benefits (unemployment insurance, workers' compensation and the like which are available as of right) for which men generally qualify and "male pauper" welfare benefits (social welfare benefits which are usually available to unemployable persons who meet a means test) for which women generally qualify is particularly interesting. (See also Judy Fudge, "The Public/Private Distinction: The Possibilities of and the Limits to the Use of Charter Litigation to Further Feminist Struggles" (1987) 25 Osgoode Hall L.J. 485 at 522-523.) 134  -95the dependency relationship between a husband and wife at the precise time that the presumed idealized family is breaking down. B y assuming the pure  support(husband)/dependence(wife)  dichotomy within familial ideology, equal and independent status when the family unit ceases to exist may make sense. However, this reasoning loses its validity when the economic disadvantages for women in the workforce and the failure to value home (reproductive) labour are recognized.  In spite o f their opposite presumptions about dependency o f women, both welfare and family law reinforce and reproduce patriarchal relations o f the private male supporter and female dependant. Because o f workplace economics which favour the presumed supporter, and the caregiving roles that are 'natural' to women and not men, women's access to full participation in the workforce, and thus to male equivalent economic independence, is restricted. Yet a woman must earn at least a subsistence level income or she w i l l be relegated to dependency status, either on a m a n  135  or on the state. However, as B o y d notes, "[t]he state has rarely been eager to give  benefits to mothers who depart from the patriarchal familial mode o f l i v i n g . "  136  Thus women are  encouraged by the state to look to a relationship with a man for their economic well being. Rather than recognizing and redressing the unfair distribution of workforce access, wages and caregiving responsibilities, the state fills i n and provides welfare (subsistence level) benefits to women who need support but have no man on whom they can depend.  She is required to seek support based on an existing or former relationship. See Mossman and MacLean, supra note 115 at 95-100. l35  Boyd, "(Re)Placing the State...", supra note 104 at 61. See also the discussion under the heading "Spouse in the House or Louse on the Lam" in Gavigan, supra note 104 at 616-632; and Mossman and MacLean, supra note 115 at 89-103. 136  -96T h e F a m i l y - P r i v a t e Safe H a v e n ? Familial ideology portrays home and family as a private safe haven from the outside world of work.  137  In contrast with the outside world of work (the market) where self interest is assumed  and extolled as the force that w i l l lead to a better society, the family is assumed to be a site o f "sharing and sacrifice among family members."  138  Associated with the assumptions o f family as  private and safe is the idea that the state should not interfere with the family; it should take a neutral stance.  139  However, as discussed above, the family in its idealized form assigns roles to its  members which define power relations within the family that are assumed to be 'obvious and natural'. Accordingly, noninterference in the family amounts to state endorsement o f the status quo within the family.  The status quo is treated as something natural and not as the responsibility of the state. Actual inequality and domination in the family - as in the free market - are represented as private matters that the state did not bring about, although it could undertake to change them. 140  Thus noninterference is not neutral.  By insisting that the family should not be subject to state regulation, men have been able to retain their excessive power. Furthermore, men in fact use the coercive power of the state to reinforce and consolidate their authority over wives and children. 141  L i v e d experiences o f child and wife abuse discredit the assumption o f safety within the  137  01sen, supra note 106.  Ibid. at 1505.  m  Ibid.  m  Ibid. at 1506.  [40  Ibid. at 1510.  w  -97family; rather the home may be a "danger zone" , not a haven, for many women and children. 142  143  A s well feminist research on the subordinate status o f women in the family referred to previously portrays equal sharing within the family as a mythical construct that is not grounded in reality.  144  Familial ideology thus helps to mask women's inequality within the family and acts as a barrier to state 'interference' in redressing these inequalities.  145  Marginalization of Non-conforming Families Dominant familial ideology constructs a particular model o f 'family' and holds this out as normative. However, many types o f families that do not meet the idealized form exist in Canadian society, including lone parent families, "cohabiting spouses which occur as initial families or as second families following divorce or separation (which may or may not include children), and gay and lesbian families."  146  Race and class inform the construction o f family as well; the idealized  family is white and middle class.  147  A s the idealized family is neither race, class or gender neutral,  laws and social policies that assume the idealized family form have differential impacts on families that vary from what is set out as the 'norm'. For example, lesbians and gay men do not qualify for  l42  Gavigan, supra note 104 at 293.  For a discussion of feminist scholarship on violence within the family see Jennifer Koshan, "Public/Private Dichotomy, Violence, and Aboriginal Women" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 87 at 89-94, and Carol Smart, Feminism and the Power ofLaw (London: Routledge, 1989). l43  Gavigan, supra note 104; Fudge, supra note 134; Boyd, supra notes 86 and 104; Mossman and MacLean, supra note 115. The fallacy of husbands and wives sharing economic resources is discussed in Maureen A. Maloney, supra note 31 at 125-129. l44  Not all feminists regard the family as a primary site of women's oppression. For example, as discussed below at pp. 99-100 some black feminists consider the family to be a refuge from the oppression of racism. l45  l46  Mossman and Maclean, supra note 115 at 105.  l47  Gavigan, supra note 104 at 591; Boyd, "(Re)Placing the State  supra note 104 at 60.  -98many family based state benefits.  148  Others must establish that they are not part o f a 'traditional'  family to qualify for state benefits. Lone parent mothers, 83 per cent o f whom had incomes below the poverty line in 1995,  149  often find that they have no choice but to seek state assistance in the  form o f social welfare benefits.  150  These women pay a high price for failing to comply with the set  family ' n o r m ' . A s only the 'deserving' poor qualify for state assistance, their privacy is subject to state scrutiny.  151  First Nation women are particularly vulnerable to state intrusion into their privacy  as they are twice as likely as other women to be heads o f lone parent families and therefore more likely to have to rely on state assistance.  152  These intrusions can have severe consequences. Susan  B o y d succinctly summarizes how familial ideology is implicated in the construction o f First Nation mothers as unfit, and the destructive effect that this construction has on First Nation families:  In the case of child welfare law, First Nations mothers who depart from the norm of mothering in a nuclear family setting are viewed negatively through a racist and sexist lens by decision-makers and are more likely to lose their children, thereby reinforcing the ideological standard-bearer of white motherhood in a nuclear family setting. 153  Within the retirement income system, for example, only heterosexuals can qualify for SPA, CPP survivor benefits and spousal RRSPs. See the discussion of the rational for spousal provisions above at pp. 75-78 and the discussion of Egan v. Canada in Chapter Three, below. l4R  National Council of Welfare (Canada), Poverty Profile 1995 (Ottawa: Supply and Services, 1997) at 35. Not only are lone parent mothers likely to be poor, the depth of their poverty is severe. The average income of lone parent mothers in 1995 was 64.2 per cent of poverty line income. Expressed in dollars, their average income of $10,126 was $8,851 below the poverty line. (Poverty Profde at pp. 51-52.) l49  ""Mossman and MacLean, supra note 115 at 95-100. Ibid. See also Gavigan, supra note 104 at 616-623.  m  152  Women in Canada, supra note 5 at 150.  " Boyd, "(Re)Placing the S t a t e s u p r a note 104 at 61-62. For a detailed analysis of this issue see Marlee Kline, supra notes 104 and 106. Kline's analysis of child welfare law and First Nation women demonstrates that the ideology of motherhood is implicated in the construction of First Nation mothers as 'bad' mothers because their mothering patterns diverge from those attributed to the 'normative' mother in the idealized family: 3  ...the dominant ideology of motherhood individuates mothers and mothering. Expectations such as those of primary care and a "proper" home environment  -99Dorothy Chunn observed the connection between familial ideology and state surveillance as well, in her discussion o f state regulation o f sexuality, sexually transmitted disease and reproduction:  Generally speaking, the more closely men, women, and children adhere to the norms governing their respective roles in the white, middle class, nuclear family, the more freedom from state surveillance they will enjoy. 154  A s the foregoing examples illustrate, the fact that the experiences o f 'non-traditional families' diverge from the set 'norm' does not mean that there is homogeneity among these families or in the way that they experience 'difference'. Although the specifics o f their experiences vary, 'nontraditional families' do share the experience of state assessment of their difference from the 'norm' to determine their (dis)qualification for benefits or to rationalize state interference.  While non-traditional families may be marginalized by laws and policies that assume the idealized family form, feminists writing about non-traditional experiences do not necessarily agree that the family is linked to women's oppression. In response to (white) feminists reification o f the  establish a conceptual framework within which individual First Nation women are blamed for the difficulties they experience in child-raising.... ...the dominant ideology of motherhood operates to impose dominant cultural values and practices in relation to child-raising on First Nations, and concomitantly to devalue, marginalize, and sometimes even deny different values and practices of First Nations, including, for example, the active involvement of extended family members in child raising. The combined effect in First Nation child welfare cases of the individuation/obfuscation and imposition/devaluation processes rooted in the ideology of motherhood is to leave First Nation women particularly vulnerable to being constructed by courts as 'bad mothers' in child protection proceedings, and to having their children taken away as a result of this construction, (emphasis added) Kline, supra note 104 at 339-340. ""Dorothy E. Chunn, "A Little Sex Can Be a Dangerous Thing: Regulating Sexuality, Venereal Disease, and Reproduction in British Columbia, 1914-1945" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 62 at 63.  -100family as the primary site o f women's oppression, black feminists countered with their reification of the black family as "a bastion against a racist society."  155  These seemingly contradictory views  of the family do not negate feminist critique o f the idealized family. Rather, the fact that black women experience 'family' differently from white women underscores the point that familial ideology plays out in specific ways and impacts upon particular experiences o f gender, class and race in discrete ways. A s Gavigan points out, laws and policies that assume the idealized family as the model for all families is especially problematic for women whose experiences are removed from the roles that the idealized family assigns to them.  156  The availability of economic benefits to families that fit or approximate the idealized form has spurred some lesbians and gay men to accept the idealized family form in principle but argue that its boundaries should be stretched so that their family forms fit within, rather than outside, the margins. The legal argument posed, based on section 15 of the Charter, is that linking availability o f benefits to family status (ie. a spousal relationship) amounts to discrimination on the basis o f sexual orientation.  157  Within lesbian and gay communities the long term value o f fighting for  family status and its associated benefits is the subject o f debate.  158  Some argue that the "historic  '"Gavigan, supra note 104 at 602. >Ibid. at 601-609.  ]5(  l37  See, for example, Egan v. Canada, supra note 68; and Rosenberg v. Canada (Attorney General), supra  note 61. In Egan, a gay couple (Egan and Nesbitt) who had resided together since 1948, argued (unsuccessfully) that the availability of the Spousal Allowance under the Old Age Security Act to a 'spouse' (defined to include only heterosexual relationships) offended section 15 of the Charter. A similar definition of'spouse' under the Income Tax Act was challenged (unsuccessfully at the trial level but successfully before the Ontario Court of Appeal) in Rosenberg by a lesbian and her employer who wanted to extend survival benefits under the employer's Registered Pension Plan to partners of gay and lesbian employees. (See discussion in Chapter Three, below.) l58  Brenda Cossman, "Family Inside/Out" (1994) 44 University of Toronto Law Journal 1.  -101-  and continuing e x c l u s i o n "  159  of lesbians and gay men from social and economic benefits available  to persons i n heterosexual relationships "is discriminatory, and violates their equality rights guaranteed by human rights legislation and the Constitution." oppressive and exclusionary institution"  161  160  Others argue that the family is "an  and that lesbians and gay men "should not be trying to  get inside the family, but should be trying to decentre, deconstruct, and/or otherwise destroy i t . "  162  Gavigan makes the compelling argument that seeking equality by accepting much o f familial ideology but challenging its heterosexual privilege  163  may achieve some positive results i n the short  term, but is too narrow an approach to achieve anything but a compromised equality for all whose experiences are marginalized by the idealized family n o r m .  164  ...['Heterosexual privilege'] implies symmetry when asymmetry is surely a more accurate characterization of spouses in the familial context, notwithstanding the contrived formal equality, gender neutrality, and contractual language of family law legislation.... The focus on the specific contexts of lesbian and gay relationships has paradoxically decontextualized heterosexual relationships. More precisely, it has decontextualized and declawed heterosexism. The analysis must be extended to explain core familial phenomena in our country such as: wife assault and child abuse; the presumed dependency of a woman in need of either social assistance or a job upon a man; the enforced dependency, or poverty, of many sponsored immigrant women; and the terrifying isolation of the battered woman whose first language is not one of the official languages. The dearth of quality and  /6id.atl.  ,59  Ibid. Canada (A.G.) v. Mossop, [1993] 1 S.C.R. 554 provides an example of a legalfightby a gay man to extend family status under human rights legislation to him and his partner. Mossop, a gay man, argued (unsuccessfully) that his employer's denial of bereavement leave to attend his partner's father's funeral was discrimination on the basis of family status. Such leave was available to immediately family, which was defined as including a spouse and common-law spouse, under the collective agreement that governed Mossop's employment. m  'Cossman, supra note 158 at 1.  l6  Ibid  l62  Gavigan, supra note 104 at 609-616 notes that society accords certain benefits to social units that are acknowledged as families and uses the term "heterosexual privilege" to describe the imperative that a family can only be heterosexual. l63  l64  Gavigan, supra note 104 at 609-616.  -102accessible child care and safe, affordable housing; the modicum of economic and physical security for women who leave unhappy or violent relationships; and the different class positions reflected and reinforced in different legislation, must also be explained. The concept of heterosexual privilege does not even begin to do this. It simply cannot cope with the enormity of the family as an ideological construct and, as a result, is neither a helpful analytical tool nor an accurate descriptive tool. (emphasis added) 165  In a thoughtful analysis o f the debate, Brenda Cossman argues that the struggle for equality for lesbians and gay men requires the contradictory approach of adopting o f both strategies - struggling to come within the family and struggling to decentre and deconstruct i t .  166  However, both Cossman  and Gavigan recognize the problematic, obstructive nature o f familial ideology i n the pursuit o f equality for those whose lived experiences vary from the idealized family norm.  The heterosexual privilege argument relies on the complex and sometimes contradictory nature o f a dominant ideology. L i k e all dominant ideologies, familial ideology is not static. Rather, as Eagleton notes, it "is scarred and disarticulated by its relational character".  167  A  dominant ideology seeks to relate the existing social order to the lived experiences o f the human beings that comprise society.  T o survive, a dominant ideology may have to absorb certain  contradictions between the social order that it seeks to maintain and lived human experiences. This is the process through which it becomes a "scarred and disarticulated" version o f its original form. The relatively recent extension o f certain social benefits, such as spousal pension benefits, to common law family relationships, exemplifies the "relational character" o f familial ideology to the social context (collective lived experience) in which it resides. While familial ideology not so  Ibid. at 613-614.  m  l66  Cossman, supra note 158 at 36.  l67  %?ranote91.  -103very long ago made marriage a requisite qualification for family status, the real life experience in our society was that a marriage ceremony did not define what was or was not a family. Those who seek equal access to social benefits for lesbians and gay men see the "relational character" o f familial ideology as potentially elastic enough to let them in. This strategy, however, is short sighted from an equality seeking perspective. It seeks inclusion o f lesbians and gay men among the privileged and thus legitimizes the underlying systemic discrimination that family status based benefits impose on others who remain excluded.  (c)  Familial Ideology and (In)Equalitv Underlying the preceding discussion o f familial ideology is the relationship between  familial ideology and gender (in)equality. To summarize, familial ideology assigns a subordinate and dependent role to women. A s indicated by the discussion o f the family wage, social policies that assume the idealized family as the norm reinforce and reproduce women's dependent and unequal status. The undervalued (unvalued) caregiving role assigned to women but not to men serves to compound economic inequality - more women than men are sole custodial parents so greater responsibilities, social and domestic as well as economic, often accompany limited access to workforce participation. The impoverishment o f women on family breakdown contradicts the family law assumption o f economic equality on divorce or separation and illuminates the failure o f gender neutrality to achieve equality in the absence o f interrogating the nature o f the relationships that comprise the status quo. That some women are unexpectedly impoverished by the death o f a spouse challenges the assumption that economic resources are distributed equally within the family. The policy o f noninterference by the state i n the family likewise preserves unequal power structures to the disadvantage o f women. The marginalizing effect o f familial  -104ideology on non-normative families further implicates it in reproducing and reinforcing existing inequalities.  168  The Charter's promise o f gender equality has failed to dislocate inequalities that are obscured by the assumption that the idealized family is the 'obvious and natural' family form. In a review o f decided cases on Charter equality rights, Judy Fudge observed that "...familial ideology has presented a barrier to members o f non-traditional family units who seek to rely on the equality guarantees to obtain state benefits which are available to traditional families."  169  She notes that  although the Charter has been effective in achieving formal equality, the impact o f these achievements has sometimes been retrogressive for w o m e n .  170  Formal equality assumes gender  neutrality and does not look beneath the surface to uncover the unequal power relations that inform existing gender relations.  ...in general, the cases illustrate what amounts to either a profound unwillingness or inability on the part of the courts to regard the different treatment of men and women in terms of the legal regulation of the family as in any way related to the subordination of women either within or without the family. 171  Although some feminists regard formal equality as a necessary first step toward achieving  The foregoing is obviously a simplified summary of the very complex relationship between familial ideology and women's inequality. l68  Supra, note 134 at 523. See also the discussions of Egan v. Canada and Rosenberg v. Canada (Attorney  m  General) in Chapter Three, below. Fudge, ibid, at 529-530 suggests that when the equality provisions of the Charter came into force "of the few remaining explicit uses of gender-based classification in the statute books most benefit women" and that the significant number of successful Charter challenges by men resulted from the courts taking a formal approach to equality. On the other hand, the fewer, and generally less successful, Charter challenges brought by women were founded on claims of systemic inequality. As Chunn, supra note 4 at 196 put it, "women lose and men gain overall when the courts adhere strictly to the principle of formal legal equality." l70  l7l  Fudge, ibid, at 524.  -105substantive gender equality, they recognize that it is not sufficient. A recurring theme in recent feminist writing is that a contextualized approach, which requires that "recognition be given to women's private experience o f subordination" , must be taken i f substantive equality for women 172  is to be achieved.  173  W h i l e familial ideology may be implicated in decontextualizing gender inequality and impeding the achievement o f substantive equality, it doesn't account for all that happens in the reproduction and reinforcement o f existing power relationships. A n d the value o f familial ideology is not as a source o f blame, a place to point the finger, but as a tool for understanding the relationship between law and women's oppression. Examining how familial ideology operates in the context o f gender neutral legislative language exposes the fallacy that neutral is equal. A s Susan B o y d explained:  Examining the ideological elements of legal discourse enhanced my understanding of the indirect relationship which law often has to relations of oppression and power. That is, law's role in women's oppression is not always direct, especially in the current period, but rather indirect in its construction of women in problematic ways, often along lines of race, class, and sexual orientation. This often occurs through the historically specific ways in which regulation of "the family" operates to reinforce women's roles. This regulation often occurs through ideological rather than directly coercive state practices, and often but not always occurs through law. 174  (d)  The Family and the Public/Private Divide The public/private divide refers to the separation of various aspects o f life into distinct and  Ibid. at 496.  m  Ibid.; Chunn, supra note 4 at 194-199; Kline, supra note 106; Mossman and MacLean, supra note 115; Gavigan, supra note 104; Boyd, supra notes 86 and 104. m  174  Boyd, "(Re)Placing the State...", supra note 104 at 46.  -106-  opposing spheres o f activities and responsibilities within western societies including Canada. These spheres co-exist and define each other in the context of three relationships: the state-market relationship in which the state is public and the market private, the market-family relationship in which the market is public and the family is private, and the state-family relationship in which the state is public and the family is private.  175  Critique of the construction o f the family as a separate  sphere from both the market and the state, and the construction o f women as inhabitants o f that sphere, has generated a substantial body of literature within feminist legal scholarship. W h i l e this subject w i l l be developed further in a later chapter, some familiarity with the relationship between familial ideology and the public/private divide at this point w i l l be helpful to understanding the public/private divide within the retirement income system itself, gendered poverty o f the elderly.  176  and its connection to the  177  Familial ideology informs the divide between family (private) and market (public). The idealized family, consisting as it does o f a bread-winning husband (whose family wage from his work i n the market sphere is sufficient to support his family), a caregiving and nurturing wife (whose unpaid work in the family provides the care and nurturing that her family needs) and their children, is a place where altruism rules and responsibilities and resources are shared for the common good of the all family members. Thus family and market are designated as separate and  '"Young, supra note 84 at 307. The retirement income system has its own public/private divide. Tiers I and II (OAS/GIS/SPA and CPP) are designated as the 'public' parts of the system and Tier III (tax subsidized RPPs and RRSPs) as the 'private'. l76  The summary that follows draws from my reading of the writings of Olsen, supra note 106; Susan Boyd, supra notes 86 and 104; Shelley Gavigan, supra notes 89 and 104; Judy Fudge, supra note 134; Dorothy Chunn, supra note 4, Jane Ursel, supra note 104, Janine Brodie, supra note 104; Faith Robertson Elliot, "The Family: Private Arena or Adjunct of the State?" (1989) 16 Journal of Law and Society 443 and Nikolas Rose, "Beyond the Public/Private Division: Law, Power and the Family" (1987) 14 Journal of Law and Society 61. 177  -107gendered spheres.  In relation to the state, both family and market are 'private'.  The state is an active  participant i n the division between itself (public) and the family and market (private); to the extent that it can characterize an activity or responsibility as private (family or market responsibilities), the state can shape its own sphere of operation.  178  The responsibilities that the state assigns to the  family are based on assumptions derived from familial ideology. It follows from the gendered nature o f the idealized family (female, caring and nurturing) that the natural place to assign responsibilities requiring those traits (care o f children, the sick, disabled and elderly) is to the family where they w i l l be taken care o f through women's unpaid or under paid labour. There is no need for the state to regulate responsibilities assigned to the family as the idealized family regulates its own internal affairs for the common good. The dividing line between state (public) and family (private) is complicated where the assumption that all families share the characteristics o f the idealized family breaks down. A s discussed above, the state does intervene in these circumstances, taking 'public' responsibility for certain family matters that fail to conform to the idealized family 'norm'. The state/market divide is also constructed by the state; the state defines itself in relation to the market (private) in the way that it takes or declines responsibility for market related matters.  Feminists have deconstructed the idealized family and exposed the role o f the state in its  'Fudge, supra note 134 at 488-489; Boyd, "Can Law Challenge ..." supra note 104 at 164.  -1 r e construction.  They have demonstrated the relationship between familial ideology and the  179  public/private divide and its gendered effects. Further, feminist scholarship, which is informed by the real (as opposed to idealized) lives o f women, recognizes that the 'public' and 'private' are not autonomous and that no sharp line can be drawn between them.  In actual practice, the dividing line between the two spheres is wiggly, slippery, and impossible to draw, especially when differences among groups of women along such lines as class and race are taken into account. For example, some paid work is done in the home, notably by women, and some women have worked in both market and family, often at the same time. The indetenninancy of the dividing line is, however, due to the fact that it is not a real line, but rather an ideological construct that is created and produced through state action and that often shifts. 180  This notion o f the public/private divide as a construct that has a shapeless, shifting quality fits earlier observations that the designation of Tier IH as a separate and 'private' part o f the retirement income system is problematic. ' 18  B o y d ' s insight that gender, race and class complicate the  public/private distinction assists in understanding that equalities can be hidden behind such designations as  'public' and 'private', and that an informed critique o f (in)equality i n the  retirement income system requires looking beyond such designations.  There is a second reason for being cognizant of the role that the public/private divide plays in the retirement income system. B o y d notes that "the way that society is organized materially along the lines o f public and private has not been fundamentally challenged by litigation or  See particularly the discussion of state involvement in constituting the family as a place where women are segregated as a means of ensuring and controlling social reproduction in Fudge, supra note 134, and in Jane Ursel, supra note 104. 179  l80  m  B o y d , "Can Law Challenge ..." supra note 104 at 164.  S e e Chapter One above, at pp.26-27.  -109legislative reform."  182  This comment underscores the observations o f Judy Fudge that the gender  inequality that is embedded in the idealized family and thus in the public/private divide defines the potential (in)effectiveness o f Charter litigation in advancing gender equality:  Subjects appear free and equal before the law, yet this formal legal equality and freedom is embedded in a social context of overreaching inequality. In other words, the formal legal equality before the state masks the systemic inequality in the private sphere. And this is exactly the significance of the public/private distinction for liberal constitutional adjudication; the Canadian Charter reifies the split and identifies the public sphere as the source of oppression while systemically obfuscating the extent to which the public sphere constitutes the private. 183  Thus the public/private divide has implications for strategies aimed at redressing (in)equality in the retirement income system as well.  The relationship between familial ideology and the public/private divide illustrates that complex and interconnecting factors work together to reinforce and reproduce unequal distribution o f power and resources along gender lines. This relationship informs the examination in Chapter Three o f the role o f family ideology in the delivery of pension benefits and the consequent systemic discrimination against women, the discussion i n Chapter Four o f the responsiveness o f pension reform to gender inequalities in a political climate that has shifted the boundary between public and private, and the exploration in Chapter Five o f strategies to redress these inequalities.  !  Boyd, "Can Law Challenge ..." supra note 104 at 182.  'Fudge, supra note 134 at 534-535.  -110CHAPTER THREE FAMILIAL IDEOLOGY AND (IN)EQUALITY IN THE RETIREMENT INCOME SYSTEM  The investigation of the relationship between law and ideology has been adopted by feminists to explore and explain the elastic capacity of the legal system to absorb reforms initiated by the women's movement and turn them into patriarchal ends. While in different periods of history, the legal system has oppressed the lives of women in fairly obvious and instrumental ways, the striking feature of the late 20 century is the perpetuation of patriarchal relations despite the removal of these obstacles.' th  Gendered poverty among the elderly is a statistical fact.  Since pensions are the major  source o f income for persons aged 65 and over, the retirement income system is implicated as a 2  site o f gender inequality. While theoretically neither o f the two modes o f access to retirement income above subsistence level, through workforce participation and spousal status, is subject to gender restrictions, the fact of gendered poverty among the elderly implies that access to retirement income is indeed gendered.  The discussion in Chapter T w o connected gender assignment to  workforce participation (assume a breadwinning husband) and spousal status (assume a dependant wife) with familial ideology. The discussion in this chapter examines the role o f familial ideology in a retirement income system that contributes to the tenuous economic security of elderly women.  The benefits o f the retirement income system are not limited to entitlement to retirement pension income. In addition to retirement pension from O A S / G I S / S P A , C P P , R P P s and annuities  'Susan Boyd, "Child Custody, Ideologies, and Employment" (1989) 3 Canadian Journal of Women and the Law 111 at 114. Revenue Canada, Tax Statistics on Individuals: 1995 Tax Year (Ottawa: Revenue Canada, Statistics Division, 1997) at 131-132. 2  -Illpurchased with R R S P s , tax benefits that enhance the ability o f some to accumulate pension income entitlement and to reduce tax liability during retirement are an integral part o f the retirement income system.  3  Entitlement to both types o f benefits (retirement pension and tax benefits)  delivered through the retirement income system is available either directly, as o f right, or indirectly as the spouse o f a person with entitlement as o f right. Insight into gendered inequality in the retirement income system is assisted by analysis o f the ideological assumptions about family that inform the nature and structure o f the system, whether access is as o f right or as a spouse, and whether the benefit is delivered during retirement as a pension benefit or prior to retirement as a tax benefit that enhances the ability to accumulate retirement savings.  I  DELIVERY OF RETIREMENT PENSION BENEFITS  (a)  A s of R i g h t A t Tiers II and III the level o f retirement pension benefits to which retired persons are  entitled as o f right is dependent upon the contributions that they make during their years o f participation i n the work force. The contributions that they must make (for C P P ) or may make (for R P P s and R R S P s ) are i n turn determined by their annual incomes. The rules that tie level o f pension entitlement to contributions and contribution limits to income favour those who earn relatively high incomes (at and above the A I W ) and participate i n the work force for the greatest number o f years.  A s discussed previously, women face greater barriers to full workforce  participation than men do, and wages for women who work are, generally speaking, lower than  See discussion above at pp. 56-61.  3  -112men's wages. Just as pension entitlements accumulate over a working life, so the effects o f lower 4  participation rates and wages are cumulative, and women carry with them into retirement the economic disadvantages o f their impeded workforce participation.  A t Tier I O A S benefits are available as of right to all residents of Canada aged 65 years and older provided they have resided in Canada for ten years after age 18, and these pensioners may also qualify for needs tested GIS benefits. O A S / G I S is particularly important for women because it provides a source o f income on retirement that is not dependent on workforce participation. The 5  problem with O A S / G I S is the low level o f income that it provides. The gendered inequality connected to O A S / G I S benefits is that women are over-represented among those who qualify for the needs tested GIS and whose retirement incomes consist primarily or entirely o f below poverty level Tier I benefits. One way o f ensuring that elderly women have equal access to "incomes high 6  enough to allow them to live in dignity no matter what their circumstances were i n their working years" is to increase O A S / G I S benefits to a level that does not fall below the poverty line. In fact, 7  the National Council o f Welfare recommended a $100 per month increase i n Tier I benefits to unattached elderly Canadians and $50 per month for elderly couples "as a step toward getting all  See above, Chapter Two at pp. 63-72.  4  Many women consider this income to represent society's recognition for their unpaid reproductive work. (Monica Townson, Protecting Public Pensions: Myths vs. Reality (Ottawa: Canadian Centre for Policy Alternatives, 1997) at 9.) 5  6  In 1993 women comprised 65 per cent of GIS recipients. (Statistics Canada, Canada's Retirement Income  Programs: A Statistical Overview (Ottawa: Statistics Canada, 1996) (hereinafter "Statistical Overview") at 111.) In  1994 OAS accounted for 39.8 per cent of the incomes of unattached senior women compared with 25.1 per cent for unattached senior men. (Statistics Canada, A Portrait ofSeniors in Canada, 2 ed. (Ottawa: Statistics Canada, 1997) nd  (hereinafter "Portrait of Seniors") at 109.)  'Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Welfare Canada, 1982) (hereinafter "Green Paper") at 1.  -113seniors above the poverty line."  8  The discussion in Chapter T w o linked women's responsibilities for unpaid reproductive work to their lower workforce participation and lower wages. Familial ideology informs this linkage. Familial ideology designates the family as the "fittest place to raise children" and assigns 9  specific gender roles within the idealized family to the wife/mother (full-time homemaker) and to the husband/father (sole breadwinner). The economic dependence of women (and children) on men  is integral to familial ideology, though in the idealized family this dependency is not  problematic because altruism prevails and economic resources are understood as shared within the family for its collective good. W o m e n in the idealized family are not supposed to enter the workforce for two reasons - they have other responsibilities which are to be carried out within the home, and they have no economic needs as these w i l l be met by the men who support them. The gendered division o f labour and its associated dependencies fuelled the struggle between capital and labour for men to be paid a 'family wage'.  10  Since the same argument held little force in  relation to women's wages," women workforce participants, who tended to be segregated into service oriented occupations that employed primarily women, were poorly paid in comparison with  "National Council of Welfare (Canada), Women and Poverty Revisited: A Report (Ottawa: Supply and Services, 1990) at 108. In the face of this recommendation, the proposed increases of $120 per year for unattached seniors and $60 for couples in the Seniors Benefit seems insulting, particularly when used in support of political rhetoric such as "The vast majority of seniors will be as well or better off. Fully 75 per cent of single seniors and couples will receive the same or higher benefits. Nine out of 10 senior women will be better off under the new system." (Department of Finance, "The Seniors Benefit: Summary" mBudget Papers (Ottawa, Department of Finance Canada, 1996).) 'Shelley A . M . Gavigan, "Law, Gender and Ideology" in Anne Bayefsky, ed., Legal Theory Meets Legal Practice (Edmonton: Academic Printing and Publishing, 1988) 283 at 293. ie. a wage that would be adequate to provide for a working man and his family. See discussion atpp.90-92.  w  "Women, after all, should not and need not participate in the paid the labour force.  -114men. Segregation o f women i n the labour force continues today, and, i n spite o f increased labour 12  force participation, women's wages continue to lag behind those o f men. 13  14  Thus the role assigned  to women within familial ideology has limited their access to, and ability to benefit economically from, full participation i n the workforce, whether the measure o f participation is i n terms o f time or wages. Since the opportunity to make maximum contributions to C P P , R P P s and R R S P s is determined by both measures o f workforce participation (time and wages), women's access to retirement pension income as o f right is also limited, and women carry the second class economic status accorded to them i n the workforce into their retirement years.  There are two aspects, then, to the insight that familial ideology offers into the gendered nature o f entitlement to retirement pension benefits through workforce participation. First, women's assigned responsibility for unpaid reproductive work limits the time that they are able to contribute to workforce participation, and second, the assumption that men w i l l provide for women limits the wages that women receive i n the paid workforce. Not only do these gendered effects operate independently to limit retirement pension benefit entitlement, but as Pat Armstrong has  l2  See discussion above at pp. 69-72.  In 1994 45 per cent of all paid workers were women, compared with 37 per cent in 1976. The percentage of women employed increased from 41.7 per cent in 1976 to 51.9 per cent in 1994, while the percentage of men employed decreased over the same period of time, from 72.8 per cent in 1976 to 65.4 per cent in 1994. (Statistics Canada, Women in Canada: A Statistical Report, 3 ed. (Ottawa: Statistics Canada, 1995) (hereinafter "Women in l3  rd  Canada") at 71.)  '"See above, Chapter Two at pp. 63-64 and 68-72. Pat Armstrong links the increased workforce participation by women over the past few decades to economic necessity, not the secondary purposes assumed within familial ideology. (See Pat Armstrong, "Restructuring Public and Private: Women's Paid and Unpaid Work" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, andPublic Policy (Toronto: University of Toronto  Press, 1997) 37 at 42-44.) Accordingly the rationale for lower wages for women no longer has any basis in reality, if it ever did. (Recall the discussion of this point above in Chapter Two at pp. 68-69.)  -115observed,  they may also operate together to produce a workforce participation cycle that  15  reinforces and reproduces the gender roles assigned by familial ideology.  16  A s mentioned previously, the C P P child-rearing dropout provision enacted i n 1983 recognizes to some extent that women's workforce participation, as measured by time in the workforce, is limited by their reproductive responsibilities.  17  This is an example of the "relational  character" of familial ideology, its capacity to maintain its legitimacy by changing to incorporate 18  (relate to) the social context in which it resides. Child-rearing dropout provisions absorb the late twentieth century social reality that contradicts the role assigned to women by familial ideology, namely that increasingly women are, and need to be, in the workforce. A t the same time these 19  provisions  preserve the  idealized  20  family's  gender  roles.  Generally  it  will  be  the  woman/wife/mother who drops out o f the workforce because she w i l l be the one with the lower wages. While the child-rearing dropout provisions are a step toward the goal o f gender equality  "Pat Armstrong, "Women's Work: Women's Wages" in Greta Hoffman Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside Limited, 1987) 354 at 371. See also the discussion above at pp. 64-65. Ibid. The ideological expectation is that, on the birth of a child one parent will interrupt his or her workforce participation so that the child can be raised within the home. This result is reinforced by lack of state support for daycare. Familial ideology informs the choice of which parent's career will be interrupted, through the gender roles it assigns and the economic effects of that role assignment. The lower paid spouse (wife) will interrupt her career and, since she is at home, her responsibilities for unpaid household work will increase. When she is ready to return to the paid workforce, her opportunities for a decent job at decent wages will have decreased. [6  "The child-rearing drop out provision allows a parent (gender neutral for mother) who stays home with her child(ren) under the age of seven to exclude those years from the calculation of lifetime earnings on which CPP entitlement on retirement is based. See supra Chapter One note 87 and the discussion above in Chapter Two at pp. 66-67. "Terry Eagleton, Ideology: An Introduction (London: Verso, 1991) at 222-223. See also the discussions above at pp. 84 and 102-103. 'Supra note 13.  19  Supra note 14.  20  -116in the delivery o f C P P retirement benefits, more is required. Effective reform w i l l have to address both aspects o f limited workforce participation by women, time and wages, as well as the ideological assumptions about women's familial roles that underlie them.  A s flawed as the C P P child-rearing dropout provisions may be, at least they make some concession to the reality that women's working lives do not follow a male standard. N o such concessions have yet been made to improve women's access in their own right to Tier III pension benefits.  (b)  A s a Spouse The access accorded to retirement pension benefits by a spousal relationship suggests a link  between familial ideology and the retirement income system. A s Stanley Knowles observed, access is given to women who are dependent on men and denied to those who are not.  21  The inequities  for women associated with the assumption of their dependency that have been discussed i n Chapter Two in the context o f familial ideology are relevant to the discussion i n this chapter and w i l l be 22  referred to, though not repeated. However, the following is worth emphasizing:  [dependency provisions ... undermine the important contribution that women working in the home make to the economy. Equally important, they raise the costs of entering the workforce for women, thereby distorting their choices and undermining their autonomy. 23  Spousal (dependency) provisions are found at all three tiers o f the pension system and, although  2l  See discussion above at p. 78.  22  See above, pp. 90-95.  Maureen Maloney, "What is the Appropriate Tax Unit for the 1990s and Beyond?" in Allan Maslove, ed., Issues in the Taxation ofIndividuals (Toronto: University of Toronto Press, 1994) 116 at 146. 23  -  i  n  -  different legislation defines 'spouse' for each tier, generally throughout the system a person qualifies as a spouse o f a pensioner i f that person is o f the opposite sex and is either married to the pensioner or has lived with the pensioner in a common law relationship for at least one year.  24  S P A (Tier I) S P A recognizes that the normal age o f retirement is 65 years, and that the breadwinner spouse may retire from the workforce before the dependent spouse with insufficient retirement income to support both o f them. In providing early access to Tier I benefits to a pre-retirement aged spouse of an O A S / G I S recipient, S P A assumes the wife/homemaker and husband/breadwinner model reified i n familial ideology.  25  Family pooling o f income is also assumed; the maximum 26  S P A is essentially the equivalent of half o f the O A S / G I S maximum for a couple (ie. married or in a common law heterosexual relationship). M a x i m u m benefits payable under O A S / G I S to a single pensioner and maximum S P A benefits for a widowed person are more than one-half o f the O A S / G I S maximum for a couple or the maximum S P A .  W h i l e S P A assumes a male support and female dependency model, qualification is triggered by the failure o f the breadwinner to provide for his family on retirement. Thus S P A is  See discussion of the definitions of 'spouse' in the Income Tax Act, R.S.C. 1985, c. 1 (5 Supplement) (hereinafter "ITA"), subs: 252(4), the Old Age Security Act, R.S.C. 1985, c. 0-9, s. 2 and the Canada Pension Plan, R.S.C. 1985, c. C-8, subs. 2(1), above at pp. 75-78. See also the discussion of Rosenberg v. Canada (Attorney General) (1995), 25 O.R. (3d) 612 (G.D.), reversed [1998] O.J. No. 1627 (Docket No. C22807, April 23, 1998), below at pp. 129-130. 24  th  As discussed earlier, the younger spouse is usually the wife. Since familial ideology assumes that the wife is dependent on the older OAS/GIS recipient spouse (husband), SPA was enacted as a mechanism for supplementing the breadwinner's OAS/GIS pension amount. See discussion above at pp. 35-36, 73 and 78. 25  Although at this subsistence income level, this is perhaps less an assumption of pooling than recognition of economies of scale that justify a lower 'welfare' payment to persons who share living space. 26  -118a welfare program that exemplifies the relationship between welfare law and familial ideology discussed earlier in Chapter T w o .  27  Welfare law assumes that a woman is, or ought to be,  dependent on a man and not on the state. However, when the man qualifies for the needs tested GIS, the state recognizes that he has insufficient income to support his not yet pension aged  28  wife  and extends S P A benefits to her subject to demonstration o f need based on family income.  The overriding requirement o f a patriarchal familial relationship is evident in the improvements in S P A over the years as well. Initially S P A benefits were available only within marital relationships; they are now available within common law relationships and to widows and widowers whose spouses would, i f not deceased, be 65 years or older.  29  S P A is discontinued on  separation or divorce and, in the case o f recipient widows and widowers, on remarriage.  30  Needy  though they may be, S P A benefits are not available to women (mostly) who never married, are divorced or separated, or to the younger partner in nontraditional relationships (eg. lesbian and gay  "See above at pp. 94-95. Put simply, welfare law requires that a woman look first to a spousal relationship (ie. a relationship to which familial ideology assigns a breadwinner role) for economic support before welfare (subsistence level) benefits are available from the state. The state steps in only where and when a breadwinner fails or cannot be found. On this point it is interesting to contrast the decisions in Egan v. Canada, [ 1995] 2 S.C.R. 513, discussed below at pp. 119-124, and Rosenberg, supra note 24. In Egan SPA benefits were denied to persons in same sex relationships while the Rosenberg decision permitted the extension of RPP survivor benefits to the surviving partner in a same sex relationship. SPA is a government program paid for out of general revenues and available to certain needy persons who are close to retirement age. Need is not relevant to RPP benefits which, though publicly funded through tax subsidies, are considered private arrangements between employers and their employees. The age range to which SPA applies (60 to 64) assumes that a dependent spouse will be no more than five years younger than her provider spouse. 28  29  National Council of Welfare (Canada), A Pension Primer (Ottawa: Supply and Services, 1996) (hereinafter  "Pension Primer") at 12. Ibid.  30  -119relationships).  31  These are persons who either live alone, with non-family members or in families  that diverge from the idealized family headed by a husband/breadwinner who provides for his dependent wife/homemaker; they do not meet the "patriarchal familial mode o f l i v i n g "  32  test  implicit in the definition o f a 'spouse', and only a 'spouse' qualifies for S P A . T o achieve gender 33  equality spousal status should not be a determinative factor for eligibility for S P A benefits. Rather, all persons who meet the need and age requirements should be eligible. This would eliminate the bi-level system that S P A created where those o f S P A qualifying age who are not spouses and are not in the paid labour force are left to provincial welfare systems which generally provide lower income assistance than even the inadequate S P A does.  34  S P A has been challenged as representing an unconstitutional heterosexual privilege. Egan v. Canada  36  35  hi  a male homosexual couple applied for a declaration that the definition o f  'spouse' in the Old Age Security Act offended section 15 o f the Charter because its effect "is to  See discussion in Chapter One above, at pp. 35-36 and Chapter Two above, at pp. 73 and 75-78. While access to SPA is phrased in gender neutral terms, far more women than men meet the eligibility requirement of being the younger spouse. Looked at through the familial ideology prism, SPA is as much a program for men as for women it relieves retired men of limited financial means from the obligation, imposed by familial ideology, of having to support a spouse. 3l  Susan Boyd, "(Re)Placing the State: Family, Law and Oppression" (1994) Canadian Journal of Law and Society 39 at 61. 32  According to the definition of 'spouse' in the Old Age Security Act, supra note 24, s. 2 (reproduced above in Chapter Two at pp. 75-76), spouses must be persons of the opposite sex in a relationship that either is a husband/wife relationship or is presented publically as a husband/wife relationship. In the latter case, the couple must be living together and have lived together for at least one year. 33  Joel Bakan's comment that social assistance "provides little more than means for survival allowing people to live in poverty rather than die from it" is equally descriptive of OAS/GIS/SPA. (Joel Bakan, Just Words: Constitutional Rights and Social Wrongs (Toronto: University of Toronto Press, 1997 at 139.) 34  35  The concept of heterosexual privilege was discussed above in Chapter Two at pp. 101-103.  Supra note 27.  36  -120restrict the allowances to spouses in a heterosexual union, ie. those who are legally married or who live in a common law relationship." Four o f the nine member panel who heard the case, LaForest 37  J., Lamer C.J.C., Gonthier J. and Major J. (the "LaForest group"), found that the purpose o f providing the S P A benefit only to heterosexual couples:  ...is both obvious and deeply rooted in our fundamental values and traditions, values and traditions that could not have been lost on the framers of the Charter. Simply stated, what Parliament clearly had in mind was to accord support to married couples who were aged and elderly, and this for the advancement of public policy central to society. 38  The LaForest group equates marriage with procreation, implying that the S P A benefit is recognition for the contribution to reproductive work, by those who are married:  ...Suffice it to say that marriage has from time immemorial been firmly grounded in our legal tradition, one that is itself a reflection of long-standing philosophical and religious traditions. But its ultimate raison d 'etre transcends all of these and is firmly anchored in the biological and social realities that heterosexual couples have the unique ability to procreate, that most children are the product of these relationships, and that they are generally cared for and nurtured by those who live in that relationship. In this sense marriage is by nature heterosexual... 39  and by those who are in common law (heterosexual) relationships:  But many of the underlying concerns that justify Parliament's support and protection of legal marriage extend to couples who are not legally married. Many of these couples live together indefinitely, bring forth children and care for them in response to familial instincts rooted in the human psyche. These couples have need for support just as legally married couples do in performing this critical task, which is to benefit all society. Language has long captured the essence of this relationship by the expression 'common law marriage'. 40  "Ibid, at 618. Ibid..at 625.  n  Ibid.  i9  Ibid. at 626.  A0  -121But, despite the LaForest group's suggestion that support for reproductive work lies at the heart o f the S P A benefits, many women who do not qualify for S P A have had and raised children, perhaps even, for the separated and divorced, within a 'traditional' family. The test for S P A , though, is not connected to any real contribution to reproductive work; the only test is whether the applicant (wife) is, at age 60 to 64, the spouse o f a man who has failed to properly provide for their retirement.  In contrast with the sexist and heterosexist reasoning o f the LaForest group, another four members o f the panel, Cory J., Iacobucci J., M c L a c h l i n J. and L'Heureux-Dube J. (the "Cory group") agreed that the purpose o f the impugned legislation is:  ...to ensure that, when one partner in a couple retires, that couple will continue to receive income equivalent to the amount that would be earned were both members of the couple to be retired, provided, of course, that the non-retired spouse be between the ages of 60 and 64. To this end, the Act is geared towards the mitigation of poverty among "elderly households". I reach this conclusion after reviewing the design of the legislation, as well as the legislative debates and policy statements accompanying the introduction of the spousal allowance. 41  S P A is characterized as simply an anti-poverty program for elderly couples. This group rejected the argument that S P A was intended to reduce poverty among elderly women i n spousal relationships though it recognized that 87 per cent o f recipients are women, and considered that 42  S P A has no social significance beyond a measure that provides '"relief in situations where two persons would otherwise have to live on the pension o f one'"  43  during a transition period when the  younger person is between the ages o f 60 and 65. Having characterized the purpose o f the  "Ibid, at 678.L'Heureux-Dube's concurring remarks on this point are at 650. Ibid. at 679.  n  *Ibid.  -122legislation in this manner, the Cory group found that the legislation offended section 15 o f the Charter and that it could not be justified under section 1.  One panel member, Sopinka J. saw the purpose o f the legislation as "alleviation o f poverty of elderly spouses" , a seemingly minor difference from the Cory group whose members focussed 44  on 'couples' ratherthan 'spouses' in their characterization of the purpose. However, the difference was sufficiently significant for Sopinka J. to justify the otherwise unconstitutional provision under section 1 o f the  Charter:  45  Given the fact that equating same-sex couples with heterosexual spouses, either married or common law, is still generally regarded as a novel concept, I am not prepared to say that by its inaction to date, the government has disentitled itself to rely on s. 1 of the Charter.'*'  Sopinka J.'s judgment was both pivotal and problematic. H i s concurrence with the Cory group that the definition o f 'spouse' in section 2 o f the Old Age Security Act did infringe section 15 o f the Charter  made Cory's the majority decision on that issue. O n the section 1 issue, however, his  finding that the otherwise unconstitutional provision was justified gave the LaForest group's decision majority status on the ultimate issue that denied Egan and Nesbitt the remedy they sought. Although the LaForest group stated its agreement with Sopinka on the section 1 issue, it was unnecessary for that group to continue to a section 1 analysis having found no infringement o f section 15. Without a clear majority on either o f the two stated issues, what the S C C really decided  "Ibid, at 654.  "'Section 1 provides that the Charter "guarantees the rights and freedoms set out in [the Charter] subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society." Supra note 27 at 656.  A6  -123was that otherwise unconstitutional discrimination could be "demonstrably justified in a free and democratic society" , and Sopinka J.'s judgment provided the key to that decision. Through its 47  alignment with both the Cory and the LaForest groups on different issues, Sopinka's lone judgment became the majority decision on an issue that transcends the question that the S C C was called on to resolve - whether a legislative provision that denies S P A benefits to cohabiting lesbians and gay men is constitutional.  Familial ideology i n its purest form is clearly alive and well in the LaForest group's 'reasoning' that S P A was intended as support for elderly couples in marriage or marriage-like heterosexual relationships, in recognition o f their contribution to society i n producing and caring for children. It is there in the Cory group reasons as well, but in a form that is "scarred and disarticulated" by its relationship to current social realities. While this group would extend the 48  benefit to lesbians and gay men who have cohabited in a conjugal relationship for at least 12 months, their rationale for the extension is that the economic relationships within these couples can be equated with the provider/dependant relationship that is central to familial ideology and assumed to exist i n heterosexual married and common law couples. Because public perception o f same sex relationships has changed considerably over the past few decades, heterosexuality as a defining 49  qualification o f 'family' has become contradictory and the Cory group allows familial ideology to  Supra note 45.  41  48  Terry Eagleton, supra note 18 at 222.  Didi Herman makes the point that "the lesbian and gay rights movement has been one of the most successful social movements in this century" but notes as well that this acknowledgement "does not mean that homophobia and anti-gay discrimination are not a part of everyday life." (Didi Herman, "(Il)legitimate Minorities: The American Christian Right's Anti-Gay-Rights Discourse" (1996) 23 Journal of Law and Society 346 at 358. See also Didi Herman, "Beyond the Rights Debate" (1992) 2 Social and Legal Studies 25.) 49  -124absorb that contradiction by stretching itself to allow lesbians and gay men i n .  50  However, as with  heterosexual relationships, only certain lesbian and gay men relationships qualify for 'spousal' status and the access to benefits that such qualification allows. These are the relationships that most closely resemble the idealized family, thus allowing the myths o f a dependent and provider 'spouse', o f shared economic resources, and o f the family as a private safe haven to remain unchallenged. A l s o unchallenged is the validity o f extending benefits to 'spouses' while excluding all unattached persons and persons in other types o f relationships from access to those benefits. While the extension o f spousal status to lesbians and gay men whose relationships, except for sexual orientation only, have the same conjugal and length o f cohabitation characteristics as heterosexual common law 'spouses' expands the traditional notion o f 'family', the Cory group's reasoning reinforces, rather than shifts or dislodges, the ideological nature o f the family. Sopinka J. acknowledges the "scarred and disarticulated" form of'family' that the Cory group accepts, but is not sure that the scars are permanent. He considers that the 'novel' idea o f according spousal status to homosexual couples falls within the domain o f political, not judicial, decision-making. Accordingly Sopinka J., in reasoning that can itself be described as ' n o v e l ' , finds that the government is justified, at least for the time being, in discriminating against lesbians and gay men by denying S P A benefits to them. In the end, then, the expansion o f familial ideology to let certain lesbian and gay men i n was held i n check as the S C C deferred to the government's jurisdiction over 'novel' matters.  'Recall the discussion of the idealized family as heterosexual privilege in Chapter Two, above at pp. 100-103.  -125S u r v i v o r benefits C P P - (Tier ID C P P is a publicly operated and funded, albeit out of contributions determined by workforce participation, pension program. However, C P P has attributes o f the private sphere as well. It provides pension coverage to virtually everyone in the paid labour force but not to those who labour only in the family, reinforcing the assumption within familial ideology that the private market (ie. workforce participants) provides for the private family through altruistic redistribution within the family. C P P does make some concessions to women's responsibilities for reproductive work through child-rearing drop out provisions, breadwinner/homemaker model.  51  but otherwise it adopts the  familiar  The amount o f pension entitlement is related to lifetime  'breadwinnings', up to a maximum entitlement  52  and it is assumed that the C P P member w i l l  share his pension with his dependent spouse during his lifetime. O n the C P P member's death, 53  C P P payments continue at a reduced rate (60 per cent, to a defined maximum ) and are payable 54  to the deceased C P P member's surviving spouse  55  until her death. The economic realities o f  women's lives dictate that they are less likely than men to qualify for full C P P retirement benefits 56  5l  See discussion above at pp. 115-116.  "See discussion above at pp. 30-31 "Spouses may apply to have their CPP pension payments split, with separate and equal cheques going to each spouse. While this is a step forward in recognizing economic autonomy, it is a small one. One would predict that spouses would agree to make such an application only in relationships where sharing is already ongoing and the splitting effects a tax advantage. Splitting of CPP pension credits applies on marriage breakdown as well, though this may be over-ridden by provincial pension legislation. (Monica Townson, Women's Financial Futures: Mid-Life Prospects for a Secure Retirement (Ottawa: Canadian Advisory Council on the Status of Women, 1995) at 34-35.)  54  See discussion above at pp. 73-74.  55  56  A heterosexual married or common law relationship is required. (Supra note 24.)  See discussion above at pp. 63-72.  -126and biological realities predict that their own C P P benefits, i f any, w i l l eventually be supplemented by C P P survivor benefits.  57  To some extent C P P survivor benefits adjust for gender inequality in access to pension entitlement through work force participation, but they do so for only some women - widows.  58  These benefits recognize that women who are 'spouses' may not have been able to provide for their own retirement income needs and extend to these women a portion o f their deceased spouses' retirement income benefits to continue until they, too, are deceased. The separated, divorced, never married, and lesbians and gay men are again excluded. Even for those who qualify, the amount o f the 'gender inequality adjustment' available to the widow o f a C P P recipient is reminiscent o f social welfare rather than entitlement.  59  The maximum survivor benefit could reduce the standard  of living o f a surviving spouse from around the poverty line (assuming family income o f O A S for both spouses and maximum C P P for the deceased spouse prior to death) to well below it.  60  In the  absence o f a supporting spouse, the state reluctantly steps in with subsistence level support, but only for those who fit the "patriarchal familial mode o f l i v i n g " . 61  C P P survivor benefits assume that the degree o f independence, or alternatively the chance  "In a 1996 publication the National Council of Welfare reported that 89 per cent of survivor pensions were paid to women. (Pension Primer, supra note 29 at 20.) '"Originally the benefit was conditional; until 1987 it was discontinued on remarriage. (Ibid, at 21.) For an interesting discussion of gender differences in social welfare benefits, which argues that the benefits that men generally qualify for (eg. employment insurance, workers' compensation) are deserved entitlements while those generally available to women are undeserved charity, see Mary Jane Mossman and Morag MacLean, "Family Law and Social Welfare: Toward aNew Equality" (1986) 5 Canadian Journal of Family Law 78 at 95-100. 59  Pension Primer, supra note 29 at 28.  m  6i  Boyd, supra note 32.  -127o f finding another man, is inversely related to age. Only widows aged 65 years and older can qualify for the full benefit.  62  The maximum benefit for widows between the ages o f 45 and 64 is  a flat rate plus 37.5 per cent of the deceased spouses benefit, and there is a pro-rated reduction o f that amount for widows between the ages o f 35 and 44.  63  Widows under the age o f 35 are not  entitled to survivor benefits unless they have dependent children or are (or become) disabled.  64  L i k e the inequitably low level o f even the maximum survivor benefit, these age restrictions signify 'welfare'; they suggest that young, able bodied and unencumbered widows can and should find another man to provide for them.  65  The state thus reinforces familial ideology by accepting the role  o f provider only for the old, the disabled and the encumbered, whose 'breadwinners' have predeceased them and whose opportunities for establishing a replacement dependency relationship are minimal.  RPPs - (Tier III) Because o f the low retirement income from C P P survivor benefits, widows (who for reasons previously discussed are likely to have low or no workforce related retirement income) must look to Tier III (RPPs and RRSPs) to keep them from a significant drop i n lifestyle on the death o f a spouse. L i k e the C P P , access to RPPs for many women is limited to survivor benefits.  While the C P P covers virtually all members o f the paid labour force, less than half o f those  'Pension Primer, supra note 29 at 20. 'Ibid. 'Ibid.  See discussion above, in Chapter Two at pp. 94-95.  5  -128workers are covered by R P P s .  66  A gender breakdown shows R P P coverage o f 46.8 per cent of men  paid workers and 41.9 per cent of women paid workers in 1993,  67  It is probably fair to assume that  in some families both spouses have R P P coverage, and that some o f those covered are unattached. This means that the total workforce coverage (44.6 per cent in 1993) does not reflect the percentage o f 'breadwinners' with post retirement income from RPPs; that percentage is undoubtedly lower. Further, it has only been since the late 1980s that RPPs have been required to include survivor benefits, and this requirement was not made retroactive.  68  Accordingly, while R P P survivor  benefits allow access to retirement income through a spousal relationship, such access is potentially available to only a small percentage o f spouses.  Others are reduced to the subsistence level  incomes provided by O A S and C P P survivor benefits. For widows the goal o f maintaining a preretirement lifestyle more often than not dies with their spouses. Since tax subsidies (expenditures of public funds) fund RPPs to a considerable extent, this amounts to state supported (re)distribution o f income in a manner "that institutionalizes women's poverty in old age."  69  In 1993 coverage was only 44.6 per cent of the work force. It is unlikely that the percentage of coverage has increased since 1993; in the ten year period from 1983 to 1993, CPP work force coverage dropped from 45.4 to 44.6 per cent. (Pension Primer, supra note 29 at 31 and 37.) See also the discussion supra, Chapter Two note 29 and accompanying text. 66  "Coverage for women increased from 35.9 to 41.9 percent from 1983 to 1993 while it dropped for men from 52.4 to 46.8 per cent. This change is attributed to increasing participation in the work force by women, who tend to be well represented in public sector employment (as teachers, nurses, social workers, etc.) where RPP coverage was 85 per cent in 1993, while private sector coverage was only 31 per cent. (Pension Primer, ibid.) Robert L. Brown, Economic Security in an Aging Population (Toronto: Butterworths, 1991) at 63-64; Townson, supra note 53 at 48-49. This requirement may be waived by written direction from both spouses. This is less than ideal given the patriarchal nature of the family and associated unequal power relations between husbands and wives. 68  69  Faye Woodman, "The Charter and the Taxation of Women" (1990) 22 Ottawa Law Review 625 at 688.  -129A s with C P P survivor benefits, the separated, divorced, and never married are excluded. 70  Until the Rosenberg  decision, lesbians and gay men were also denied access to R P P survivor  benefits. These exclusions reflect the ideological assumption that the only acceptable economic dependency relationship is a spousal relationship, consisting as it does o f a couple with assigned roles - breadwinner (husband) and a homemaker (wife).  The spousal requirement for access to R P P survivor benefits has been challenged under the Charter as discrimination on the basis o f sexual orientation. In Rosenberg General) ^  v. Canada  (Attorney-  two lesbians and their employer, C U P E , challenged the constitutionality o f the  1  definition o f 'spouse' in the I T A as it related to registration o f RPPs on the grounds that its effect was to discriminate against lesbians and gay men.  72  surviving spouses o f gay and lesbian employees.  C U P E wanted to extend survivor benefits to To achieve this goal C U P E amended the  definition o f 'spouse' in its pension plan to include lesbians and gay men. Revenue Canada advised C U P E that it would not accept the amended definition. Therefore the plan would be deregistered under the I T A and, as a result, would not qualify for the tax subsidies available to registered plans. The trial judge considered that she was bound by the decision o f the Supreme Court o f Canada in Egan v. Canada  12,  and concluded that, although the definition o f 'spouse' i n  A CPP member may designate a former spouse as a beneficiary of his survivor benefits, but if he has a current spouse that person must be designated (subject to the waiver referred to in note 67). 70  Supra note 24.  1]  Although the litigants did not explicitly state the issue in those terms, their challenge that the restriction of RPP survivor benefits to heterosexual relationships was unconstitutional is a challenge to that restriction as a heterosexual privilege. See the discussion of the problematic nature of such challenges in Chapter Two above, at pp. 100-103. 72  Supra note 27.  n  -130the I T A amounted to discrimination under section 15 of the Charter, the offending provision was justified under section 1. In a unanimous decision delivered A p r i l 23, 1998, the Ontario Court o f Appeal overturned the trial judge's decision, holding that the exclusion o f lesbian and gay male relationships from the definition of'spouse' in the I T A is discriminatory and cannot be justified under section 1 o f the Charter. The remedy granted was the inclusion o f same sex partners in the definition o f 'spouse' in the I T A as it applies to pension plan registration. B y allowing the appeal 74  period to expire, the federal government has accepted the decision of the Ontario Court of Appeal. Employer sponsored pension plans may (not must) now extend survivor benefits to same sex spouses without losing their qualification for income tax benefits.  75  W h i l e a favourable decision for Rosenberg and C U P E is a step forward for economic security of lesbians and gay men in stable (more than one year duration) conjugal relationships, this decision reflects acceptance of the idealized family form (comprised o f a provider and a dependant spouse) in principle, albeit with somewhat larger boundaries. To paraphrase Gavigan, the Ontario Court o f Appeal examined same sex relationships through the prism o f familial ideology and judged that sexual orientation is not a defining criterion o f what constitutes a family, at least for the limited purpose o f pension plan registration.  76  But, as Gavigan argues, this type o f reasoning  Writing for the unanimous court, Abella J.A. stated, "The appropriate remedy under s. 52 of the Constitution Act, 1982, is, in my view, to read the words "or same sex" into the definition of 'spouse' in s. 252(4) of the Income Tax Act for the purpose of the registration of pension plans and amendments to registered pension plans." (emphasis added) {Supra note 24 at para. 53.) 74  "Following the decision in Rosenberg, Nova Scotia and British Columbia announced changes to their public employee pension plans that will extend the same benefits to lesbians and gay men as are provided to heterosexual employees. See Claire F.L. Young, "Spousal Status, Pension Benefits and Tax: Rosenberg v. Canada" (1998) 6 Canadian Employment and Labour Law Journal 1. Shelley A . M . Gavigan, "Paridise Lost, Paradox Revisited: The Implications of Familial Ideology for Feminist, Lesbian and Gay Engagement to Law" (1993) 31 Osgoode Hall L.J. 589 at 605. ("...[T]he ideology of the partriarchal nuclear family provides a prism through which relationships are examined, and the measure against which 76  -131is too limited to go very far i n achieving substantive equality for women and others whose real lives diverge from the roles assigned to them by the idealized family ' n o r m ' .  77  The separated,  divorced, never married and those who are not in a long term conjugal relationship are still 78  denied access to survivor benefits, even though these persons may be in financial need and someone may desire to name them as a pension plan surviving beneficiary. They are not spouses and accordingly no one has been, or can be, assigned the role o f supporting them.  Spousal R R S P s (Tier III)  79  The tax system encourages a higher income earning spouse to contribute some or all of his allowed R R S P contribution amount to an R R S P account in his spouse's name, rather than his own. The incentive is the income splitting that arises on withdrawal as retirement income; the retirement income w i l l be taxed i n the lower income earner's (wife's) hands, at a lower rate. While the entire tax subsidy goes to the husband and there is no direct tax advantage to the wife , the idea is that 80  81  she w i l l have control over the R R S P contribution to her account and thereby have her own source o f retirement income.  they are judged.") 77  See the discussion above in Chapter Two at pp. 100-103.  78  including lesbians and gay men.  Spousal RRSPs both provide a source of pension income and deliver a tax benefit. The brief discussion here focuses on spousal RRSPs as a source of retirement income that can be accessed only through a spousal relationship. Spousal RRSPs are discussed further, as tax benefits, below at pp. 135-138. 79  80  As discussed below at p. 134-136.  'In fact she may have a higher tax liability as a result. However, again there is an assumption that she will benefit from her husband's sharing of his tax saving. 8  -132Spousal R R S P s provide access to retirement income derived from retirement savings made by the higher income earning spouse's (husband's) income i n the name o f the lower income earning spouse (wife). A father cannot contribute to an R R S P account in his daughter's name, nor can a friend contribute to a friend's R R S P account. Thus access requires and rewards only the type o f dependency that is a familiar feature o f familial ideology:  Together with the provision of survivor benefits, this rule highlights a general problem with the tax subsidization of pension plans and RRSPs. State-subsidized benefits are being provided to individuals solely on the basis that they are in a particular defined relationship with another person. Single persons and those whose relationships are not recognized by the tax system are discriminated against. 82  A s noted previously i n the discussion o f the Rosenberg decision, the tax system employs familial ideology as the yardstick against which relationships that it w i l l recognize are measured. Although Rosenberg ruled that same sex couples cannot be excluded from the definition o f 'spouse' i n the I T A as it applies to registration o f pension plans, access to the benefits o f spousal R R S P s are not extended to lesbians and gay men by that decision.  In sum, pension benefits beyond subsistence level are delivered primarily through workforce participation (CPP, RPPs and RRSPs) and secondarily through a spousal relationship ( S P A , survivor benefits and spousal R R S P s ) .  82  83  This might be a sensible delivery model i f all  Claire F.L. Young, "Public Taxes, Privatizing Effects, and Gender Inequality" in Susan B. Boyd, ed.,  Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press,  1997) 307 at 321-322. To clarify, I use "primarily" and "secondarily" here in parallel with direct and indirect access to the retirement income system. The primary mode of delivery of benefits is through entitlement that is acquired directly by the beneficiary, and the means of acquisition is through workforce participation. The secondary mode of delivery of benefits is through entitlement that is acquired indirectly, by being in a particular type of relationship with a person who has acquired entitlement directly through workforce participation. Survival benefits, for example, are secondary to benefits through workforce participation in that they are payable only after the 'primary' beneficiary dies and his benefits cease to be payable. Spousal RRSPs are also secondary in the sense that the person entitled to contribute may 83  -133families had the attributes o f the idealized family and everyone was part o f such a family. However, the lives o f real women and men diverge in many ways and in varying degrees from their idealized forms. Familial ideology operates in this context to make sense o f the contradictions between real and ideal by substituting assumptions about families and economic relations for realities i n the design of the retirement income system and the delivery o f pension income benefits. Distortions are thereby concealed, and inequalities masked.  II  DELIVERY OF TAX BENEFITS  (a)  A s of R i g h t  8 4  Taxpayers are entitled, as of right, to deduct from their taxable incomes contributions, up to a limit that is related to their income levels, made by them to their own R R S P s or to an R R S P in their spouses' names. Their contributions to employer sponsored pension plans (RPPs) are also tax deductible and they receive tax credits for their C P P contributions.  A further benefit is  delivered through the non taxation o f income earned on R P P and R R S P investments.  While there is nothing that overtly suggests gender inequality i n the policy o f assigning responsibility for pension income to the private sphere o f the market, or i n the tax incentives that encourage this policy, as Claire Y o u n g points out, "[w]omen are disproportionately represented in  choose to share his entitlement with his spouse. In all cases (SPA, survival benefits and spousal RRSPs) the spousal benefit is secondary to a prior entitlement. The discussion of access to pension benefits as of right earlier in this chapter, above at pp. 111 -116, applies equally here and will not be repeated. Barriers to full workforce participation are an impediment to equal opportunity for women to acquire RPP entitlement and to accumulate RRSP savings in their own right. 84  -134the group that is unable to take advantage o f the [RPP] tax benefits."  85  In addition, because o f the  economic disadvantages that women face in the work force, they generally earn significantly less than men and therefore have both lower R R S P contribution limits and less discretionary income to make R R S P contributions. Further, because o f their responsibilities for unpaid reproductive work, many women do not participate in the paid work force and accordingly have no direct access to R P P or R R S P tax advantages.  (b)  A s a Spouse The tax system responds to the market's failure to provide adequate retirement income for  women by including special rules that permit RPPs to continue paying a portion of the pensioner's pension to his surviving spouse and that allow R R S P contributions to be made i n the name o f a 86  spouse. That is, the tax system first encourages the market to provide adequate retirement income, then looks to the idealized breadwinner/dependant family to " f i l l the gap[s]" and rewards him for 87  doing so. A l l of the tax savings - the tax deductions for contributions, the sheltering from taxation of the income earned on the contributions and the reduced tax on the pension income when it is paid - accrue to the breadwinner (husband) from whose pension savings entitlement the R P P survivor benefit or the spousal R R S P derived.  Young, supra note 82 at 320. See also the discussion of access to RPP survivor benefits earlier in this chapter, at pp. 127-131. 85  ITA, supra note 24, Part L X X X V . The ITA rules permit, but do not require, a RPP to provide survivor benefits. Employer sponsored pensions are governed by provincial or federal pension legislation, depending on which of these bodies has jurisdiction over the employment. This legislation mandates that pension plans must provide survivor benefits to the employee's spouse unless both the employee and his spouse waive the requirement in writing. (See Brown, supra note 68 at 63-64.) Inequalities associated with access to RPP benefits as a surviving spouse are discussed above, at pp. 127-131. 86  87  Young, supra note 82 at 320.  -135Spousal R R S P s O f the Tier III related tax savings referred to above, only the reduced tax on pension income derived from contributions to a spousal R R S P is, strictly speaking, a tax benefit to which access is restricted to spouses.  88  The policy choice of delivering a pension related tax benefit to the higher  income earning spouse (husband) does not look to the real lives o f real women and their relationships with their spouses but accepts the altruistic husband model and assumes that the benefit to him, lower tax liability on his retirement income, is also a benefit to his spouse, and that the detriment to her, higher tax liability on retirement, w i l l be shared. Dependency o f women on men is thus assumed, reinforced and reproduced. This is not achieved through explicit gender bias, but more subtly by designing tax subsidies for retirement savings "around... assumptions ... about the nature o f economic relations within the family." The spouse without access to R R S P or R P P 89  tax benefits as o f right still has no direct access to those benefits through a spousal R R S P ;  the  supporting spouse gets the tax benefit, not the spouse who needs it:  ...the subsidy is delivered to the economically dominant person in the relationship and not the 'dependent' person who needs it. This manner of delivery assumes that the income will be pooled and redistributed equitably in the relationship, but,... such pooling does not occur in a majority of relationships. 90  The misguided nature o f this policy choice is summarized by Phillips and Y o u n g :  While contributions to a spousal RRSP are tax deductible to the contributor, and the income on the invested contributions is tax free, spousal RRSP contributions can only be made with the contributor's RRSP contribution room. If he had no spouse, presumably he would make the contribution to his own RRSP. 88  Lisa Phillips and Margot Young, "Sex, Tax and the Charter" (1995) 2 Review of Constitutional Studies 221 at 279. 89  Young, supra note 82 at 322. Tax scholar Neil Brooks also notes that "The claim that there is widespread pooling of resources [between spousesl seems particularly naive." (Neil Brooks, "The Irrelevance of Conjugal Relationships in Assessing Tax Liability" in John G. Head and Richard Krever, eds., Tax Units and the Tax Rate Scale (Sydney: The Australian Tax Research Foundation, 1996) 35 at 63.) See also Maureen Donnelly, "The Disparate Impact of Pension Reform on Women" (1993) 6 Canadian Journal of Women and the Law" 419 at 446-447. 90  -136Though they may be benevolently motivated, taxation policies which react to women's lack of economic power by simply leaving space for support to be provided to them privately, by men, may do nothing more than legitimate women's and children's continuing poverty. 91  The assumption o f altruistic equitable distribution by the recipient o f the tax benefit breaks down completely for women who do not have a man on whom they are dependent. Gendered economic disadvantages are not restricted to attached women. Unattached women, and particularly unattached women with dependent children, are vulnerable to the gendered obstacles to full workforce participation which flow from assumptions about the 'obvious and natural' dependency o f women. These women are, i n effect, punished by the tax system for their failure to adopt the dependency role assigned to them by familial ideology. Their ability to take advantage o f tax subsidies for retirement savings is limited  92  but no man is altruistically impelled (or coercively  compelled) to re-distribute his excessive tax (welfare) benefits to these more needy persons.  While the husband gets the tax benefit associated with a spousal R R S P , the wife gets her own source o f her retirement income.  Or does she?  Maureen Maloney points that out a  (re)distribution for tax planning purposes may not be a surrender o f de facto control by the contributor: ...it is unlikely that many wives will obtain decision-making power over assets transferred to them legally, and I suspect this is especially true when the transfers are made as a tax-planning device. 93  This provision thus works i n practice more as compensation to a husband for supporting his wife  Supra note 89.  9l  92  See earlier discussion, above at pp. 63-72 and 74-75.  93  Maloney, supra note 23 at 137.  -137than an incentive to share R R S P savings between spouses.  94  Moreover, i f there is sharing between  spouses, as familial ideology assumes, this tax splitting incentive to share is unnecessary. Although the idea o f providing an incentive to share economic resources within the family seems contradictory to familial ideology (which assumes altruism and sharing), the nature o f the incentive (giving a benefit to the provider spouse on the assumption that he w i l l share with his dependent spouse, rather than giving the benefit directly to the lower income spouse) is clearly grounded in familial ideology. The economic benefit o f tax splitting may not work to induce sharing either in fact or in theory, but it is available only to those who qualify as 'spouses' and not to others, thereby reinforcing the provider/dependency relationship between women and men. Indeed, maintaining actual economic control in the hands o f the contributor (husband) is completely consistent with familial ideology; the assumption o f his altruism remains intact.  Although they assume, reinforce and reproduce the economic power relations within the idealized family, two other aspects o f spousal R R S P s are suspect from the perspective o f their use as providing an independent source o f retirement income for a spouse.  First, spousal R R S P  contributions need not remain invested until retirement and may be withdrawn after three years. "The advantages o f income-splitting can thus be accessed regardless o f whether the funds accumulate, or were ever intended to accumulate, for use in the wife's retirement."  95  Accordingly,  the contributor's (husband's) de facto control over the family's economic decision making is postponed for a short time only. Second, i n the absence of an over-riding provision i n the I T A , the husband may be able to invoke the common law doctrine o f resulting trust to revoke the transfer  94  Maureen Donnelly makes a similar suggestion in respect of the spousal credit. Supra note 90 at 428.  Ibid. at 446.  9i  -138and defeat an assertion of ownership o f the spousal R R S P by his wife.  96  The contradictory way in  which familial ideology informs this aspect o f the tax system, by providing R R S P access (albeit limited) to the lower income spouse while placing minimal restrictions on the control o f that access to the economically advantaged spouse, may be a factor that contributes to the low use o f spousal R R S P s . Indeed, history has shown that tax incentives involving transfer o f property by husbands to their wives are unlikely to be well used.  97  Spousal R R S P s provide access to retirement income benefits through dependency. They privatize a dependent spouse's (wife's) retirement income through a provider spouse's (husband's) altruistic (re)distribution of his earned income, thereby reinforcing the assumption that reproductive work has no economic value. They are completely voluntary arrangements, 'gifts' with strings attached, that leave otherwise payable tax dollars in the pockets o f men while doing little to ensure that their spouses are provided for in their retirement years. A s ineffective as spousal R R S P s may be, though, they do offer an opportunity for economic advantage to women i n spousal relationships that is denied to women who are not spouses.  (c)  T h e P r o b l e m w i t h T a x Benefits Thus, the modes of delivery o f pension related tax benefits, as o f right or through a spousal  relationship, parallel the modes of delivery of pension benefits. Although the benefits themselves  Ibid:  96  "Only 12 per cent of women who accumulated RRSP savings in 1989 received all or part of their deposits from their husbands." (Monica Townson, supra note 53 at 57.) See also Hubert Frenken, "Women and RRSPs" (1991) 3:4 Perspectives on Labour and Income 8. Recent statistics on spousal RRSP contributions are not available. Donnelly suggests "The reason must be either that [Canadian men] are not rational and self-interested or, what seems more likely, that counter-veiling perceptions of self-interest deter them from transferring wealth to their wives." (Ibid.) 97  -139are different, both modes o f accessing these benefits are based on assumptions about men and 98  women and their economic relationships that are grounded in familial ideology.  A s these  assumptions do not represent reality for all (or even most) women, the result is that women are disadvantaged by both modes of access. Their abilities to accumulate their own sources of pension income is impeded and they are forced to rely on dependency relationships with men for their economic well being. Only certain dependency relationships are acceptable, however, and family ideology determines the norms against which acceptability is measured.  A s unfair for women as the limited Tier III tax benefits maybe, the policy choice reflected in choosing the tax system to deliver pension related benefits is even more problematic. Tax deductions for R P P and R R S P contributions and tax sheltering o f income earned on those contributions reflect a public policy o f characterizing income security on retirement as a private responsibility. The assumption is that the (private) market w i l l provide adequate income for a breadwinner and that he w i l l provide for his family.  99  retirement  Through deductions for  contributions to R P P s and tax sheltering o f income on those savings, the tax system encourages employers and employees to make retirement benefits part o f the remuneration for employment. Similar tax incentives are extended to employees whose employers do not provide retirement pensions - they may provide for their own retirements by contributing a portion o f their incomes  Pension benefits are retirement income, while pension related tax benefits subsidize contributions to pension plans and pension savings thereby enhancing the accumulation of financial resources from which income will be derived on retirement. 98  "As explained above in Chapter One, Tier III of the retirement income system is intended to top up the subsistence level pension benefits from the lower two tiers and provide sufficient retirement income to maintain preretirement lifestyles. It is in this context that "adequate" is used in this discussion. (I refer to Tiers I and II together as providing subsistence level income because the maximum amounts available under these combined pension programs merely hovers around the poverty line.)  -140to an R R S P . Tax benefits become a retirement income producing social program.  B y funding Tier JJI through tax expenditures (public monies from taxes otherwise payable) the government is, i n effect, paying for a social security program that is not publicly accountable. W h i l e the amount that the government 'spends' on Tier IU is made public each year in the published tax expenditure accounts,  100  there is no public accountability with respect to ensuring the  effectiveness o f the expenditure i n achieving the goals set by the government for the retirement income system. Rather, it is assumed that leaving additional monies in the pockets o f a person who saves for retirement w i l l provide for that person's retirement income needs as well as the retirement income needs o f those who are dependent on him. Indeed, the savings need not be for retirement; R R S P investments may be 'cashed i n ' at any time though this w i l l trigger income tax liability.  The problematic nature o f delivering social security benefits indirectly through the tax system where they are "hidden from public scrutiny"  101  rather than directly through programs  developed for those particular purposes has been discussed previously.  102  The standard model for  social programs is that they are directed at particular social needs and monitored to ensure that  l00  See supra, Chapter One notes 187-188 and accompanying text.  ""Dennis Guest, The Emergence ojSocial Security In Canada, 3 ed. (Vancouver, UBC Press, 1997) at 189. Again, it is not the amount of the expenditure which is hidden, but the connection that should exist between the expenditure and its goal which, because it is not monitored, is "hidden from public scrutiny". The process of pension reform over the past decade has graphically demonstrated Guest's point. Reform initiatives directed at the parts of the retirement income system that delivers pension benefits through programs designed for those purposes (OAS/GIS/SPA and CPP) have triggered much public debate while reform of tax system delivered pension related benefits was achieved without public controversy. While the sustainability of the so called 'public' parts of the retirement income system has been questioned by a government that seeks to cut 'public' spending, the 'private', tax system delivered part of the retirement income system not only escaped such scrutiny but these expenditures were increased. rd  102  See above, Chapter One at pp. 59-60.  -141goals aimed at addressing those needs are being met effectively. The retirement income system is a social program intended to achieve two broad goals.  103  Tiers I and II are directed at the first  goal, "... to ensure that elderly people have incomes high enough to live i n dignity no matter what their circumstances during their working years"  104  while the second goal, "to maintain a reasonable  relationship between income before and after retirement so that old age does not bring a drastic reduction i n standard o f l i v i n g "  105  is left to Tier H J .  106  A s discussed in Chapter T w o women are  more vulnerable than men to serious reductions in lifestyle on, or at some point during, retirement and to live their final years in poverty.  107  Logically, then, Tier III w i l l not effectively achieve its  goal unless women are over-represented among those to whom the tax expenditures directed at income security on retirement are distributed.  108  However, as women are likely to have less income  than men, they w i l l also receive a smaller share o f the tax savings that enhance retirement savings.  109  The fact that Tier HI is less accessible for women than for men underscores the point  made by Claire Young that " i n many circumstances the tax system is a very blunt instrument by  Pension Primer, supra note 29 at 1.  103  Ibid.  m  Ibid.  m  ""Theoretically Tier II should be directed toward the second goal as it is tied to earned income and is, in that sense, income replacement. However, Tier I income levels are so low that the first goal is not achieved unless Tier II benefits are also available. Even with full Tier II benefits, retirement incomes hover around the poverty line. ,07  See above pp. 62-81.  The way in which the tax system heavily subsidizes retirement savings and thus increases the funding for this source of retirement income was discussed earlier. See above, Chapter One at pp. 22-27 and 44-46. l0  ln 1993 the average RRSP contribution for persons aged 25 to 64 whose taxable incomes were over $50,000 was $5,115. For those earning (a) under $10,000, (b) $10,000-20,000, (c) $20,000-30,000, (d) $30,00040,000, and (d) $40,000-50,000 average contributions were (a) $1,115, (b) $1,661, (c) $2,058, (d) $2,621, and (e) $3,208 respectively. (Pension Primer, supra note 29 at 43.) Since the rate of tax savings increases as a taxpayer's marginal tax rate increases, the cost of each dollar saved for retirement is less for high income earners than for low income earners. I09  -142which to deliver sophisticated social programs.""  0  The use o f tax expenditures (public funding i n the form o f uncollected tax revenues) as an instrument to deliver an important part o f Canada's retirement income program is not only blunt, it is aimed i n the wrong direction. Y o u n g notes the "upside down effect"" that results from 1  delivering Tier HI benefits as tax deductions. A s a tax deduction is "tied to the rate at which a taxpayer pays tax" , public funds are used to deliver a better pension program to the economically 112  advantaged (higher income earners, who pay tax at a higher rate) than to the disadvantaged. Faye Woodman neatly summarizes the problem:  The tax concessions to deferred income plans are welfare type tax expenditures which ration benefits on the basis of income. Thus, the more one has, the more one receives. Those who have less, receive less. 113  W h i l e delivering more pension related benefits to those with less need is counterintuitive, familial ideology again obscures the contradictions between the real and the ideal and validates the delivery of more to men (providers) and less to women (dependents). From this perspective women and men benefit equally from the delivery o f Tier HI benefits, and i f women receive their equal shares indirectly through their spouses, that is only 'natural'. However, not all women are spouses and not all men are ideal, and it is women who are disadvantaged by this reality.  The delivery o f tax benefits to encourage 'private' retirement arrangements effectively  ""Supra note 82 at 308. Ibid.  m  Ibid.  U2  Woodman, supra note 69 at 687.  1l3  -143privatizes "economic responsibility for dependent persons.""  4  It valorizes private initiative, the  effect o f which is to construct those who rely on the ' p u b l i c " parts o f the pension system 15  (women) as dependent on government. A t the same time it obscures "government support [tax expenditures...] of high-income individuals and corporate entities that generally are represented as operating without public assistance.""  6  The exclusion o f any reference to Tier III as part o f  Canada's retirement income system in the Finance Minister's July 28,1998 news release illustrates this point."  7  Recent reform has not only ignored the flaws in the tax incentive programs but has exacerbated these by increasing the role o f tax incentives in responding to the social need o f providing economic security on retirement. It is important to note here that using the tax system to  "transfer more and more o f the risk and responsibility for providing for retirement on to  individuals"" is not the same as shifting the costs to individuals. Lost tax revenues provide 8  financial assistance to those who make the kinds of'private' retirement income arrangements that the tax system encourages.  Even i f the increased tax subsidies effected by reform could be  successful incentives to private retirement arrangements, the shifted risk may not reduce the need  ' Young, supra note 82 at 321. l4  '"Although Tier III is heavily subsidized by public resources in the fonn of tax expenditures, the term 'public' here follows common usage and denotes Tier I (OAS/GIS/SPA) and Tier II (CPP). The features of Tier III from which the public/private distinction derives in this context are discussed above in Chapter One at pp. 26-27. ' Marlee Kline, "Blue Meanies in Alberta: Tory Tactics and Privatization of Child Welfare" in Susan B. i6  Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto  Press, 1997) 330 at 338. "'Department of Finance Canada, News Release 98-071, "Finance Minister's Statement on the Seniors Benefit" (28 July 1998). See also above, Chapter One at pp. 55-56. ' "Townson, supra note 5 at 6.  -144for social security programs.  If the market fails, retirement savings cannot be relied upon to  provide adequate retirement income. Thus this shift could cost more and provide less retirement income security than might be achieved by collecting the forgiven taxes and directing these monies to effect improvements to the social security parts o f the system. Further, the shift to a more privatized retirement system is a regressive step for women (and others) who were already disadvantaged by the retirement income system as it existed prior to reform.  -145CHAPTERFOUR RESTRUCTURING AND (IN)EQUALITY IN THE RETIREMENT INCOME SYSTEM  [I]t is clear that the main thrust of government policy on retirement incomes is to reduce the role of the state; to move away from collective responsibility for and to seniors; and increasingly to expect Canadians to go it alone as far as providing for a secure retirement is concerned. 1  Economic inequality between men and women is embedded within Canada's retirement income system, not necessarily because the system intends to advantage men over women, but because the system accepts the social and economic construction o f women and men implicit in familial ideology. T o recap, women are assigned the caretaking work that society requires and are expected to perform this work for no  2  or l o w  3  economic benefit.  M e n are assigned the  'breadwinning' role; they are expected to participate full time in the labour market and to share the economic benefits o f their labour within their families. Economic autonomy for women is incompatible with familial ideology, which casts women as economically subordinate to, and dependent upon, men. A s discussed in the preceding chapters, the retirement income system carries this economic model forward into retirement years - except for the less than adequate O A S benefits, ability to accumulate and access retirement income benefits is determined by workforce participation and spousal status. In short, Canada's retirement income system imposes economic inequality on women because it is based on a 'common sense' representation o f women that  'Monica Townson, Protecting Public Pensions: Myths vs. Reality (Ottawa: Canadian Centre for Policy Alternatives, 1997) at 1. ie. unpaid reproductive work within the family. (Recall that 'reproductive work' as used in this thesis means much more than the work related to biological reproduction. See supra Chapter 2 note 4.) 2  ie. poorly paid 'women's work' within the labour market.  3  -146distorts women's realities, and familial ideology is the "organizing social force" that legitimizes 4  the distortion.  A s pointed out i n Chapter One, inequalities for women within the retirement income system were acknowledged and targeted for reform almost two decades ago, when the Green Paper Pensions  was released by the Minister o f National Health and Welfare in 1982.  5  on  W h i l e some  reforms were implemented in response to women's equality concerns during the 1980s, these can 6  be characterized as piecemeal measures that did not seriously call into question the economic model, informed by familial ideology, on which the retirement income system was superimposed. While reform o f the system continued, over the past decade these changes have not been driven by women's equality issues. This chapter explores the relationship between the contemporary political climate and the failure o f recent pension reform to address the economic inequality for women that was acknowledged in 1982 to be a serious problem "throughout much o f the retirement income system."  4  7  Terry Eagleton, Ideology: An Introduction (London: Verso, 1991) at 222.  Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Welfare Canada, 1982) (hereinafter "Green Paper"). See also above, Chapter One pp. 39-43. Tier I: In 1985 SPA was extended to any widow or widower who met the age and residency requirements. (See discussion above in Chapter One at pp. 35-36.) Tier II: In 1986 the provision by which CPP survivor benefits ceased on remarriage was repealed, and CPP survivor benefits were extended to common law spouses. (See discussion above in Chapter One at pp. 32-33.) Child-rearing dropout provisions were added to the CPP in 1983. (Supra, Chapter One note 87.) Tier III: New rules for RPPs were implemented through amendments to provincial and federal legislation governing these 'private' pension schemes, including requirements that the plan vest after two years rather than ten (to accommodate the interruptions in workforce participation experienced by many women), that part time employees be allowed to join any plan provided for full time workers, and that plans provide mandatory benefits for surviving spouses unless both spouses agree in writing to waive the requirement. (Monica Townson, Partnerships in Pensions: Who's Responsible Anyway (Ottawa: The Canadian Centre for Policy Alternatives, 1990) at 10.) RPPs remain seriously flawed as far as women are concerned - few women workers are employed where coverage is available, and inflation protection is not mandatory. (Monica Townson, Women's Financial Futures: Mid-Life Prospects for a Secure Retirement (Ottawa: Canadian Advisory Council on the Status of Women, 1995) at 63.) 1  Green Paper, supra note 5 at 2.  -147I  T H E RETIREMENT INCOME SYSTEM - REFORMED OR RESTRUCTURED?  (a)  The Reformed System The shape o f Canada's retirement income system has altered over the past decade; the role  o f the income tax system has been expanded. This expansion has taken two forms, the effects o f which have been to reduce the broad population base that Tier I serves and to increase the level o f tax benefits to the narrower population base that Tier JJI serves. First, the tax system has been used to erode the once universal character of O A S .  8  In 1990 legislation was implemented that imposed  a special 15 per cent tax on O A S benefits paid to persons with incomes exceeding $53,215. In 1996 the ruse o f maintaining universality by paying O A S to all seniors and then taxing it back from some was eliminated when the Income Tax Act was further amended to provide for the 9  withholding from O A S cheques o f an amount equal to the clawback for the previous tax year.  10  Second, pension related tax benefits available to high income workforce participants" were increased.  12  U n t i l tax reform in 1991, the R R S P contribution limit topped out at taxable incomes  An Act to amend the Income Tax Act, the Federal-Provincial Fiscal Arrangements and Federal PostSecondary and Health Contributions Act, the Old Age Security Act, the Public Utilities Income Tax Transfer Act, the %  War Veterans Allowance Act and a related Act, S.C. 1990, c. 39. See also the discussion above, in Chapter One at pp. 51-52. "R.S.C. 1985, c. 1 (5 Supplement) (hereinafter "ITA"). th  ™An Act to amend the Income Tax Act, S.C. 1996, c. 21, s. 46 (Chances are that the population base served by the OAS will be further eroded in the not too distant future. See discussion of the proposed Seniors Benefit above, Chapter One at pp. 52-56.) "Tier III income tax benefits are workforce related in that RRSP contributions are deductible from earned income which includes business and employment income, but not income earned from property, and RPPs are available as part of employee remuneration packages. The relationship is not perfect, though, as spousal and child support payments, which are not directly workforce related, are also included in earned income for RRSP deduction purposes. l2  See discussion above, Chapter One at pp. 44-46.  -148o f $37,500.  13  B y 2004 the contribution limit w i l l top out only for taxpayers with taxable incomes  o f $86,111 and above. A s a result of tax reform, economically advantaged taxpayers who have the financial resources to make maximum R R S P contributions can expect to enjoy publicly funded or subsidized retirement incomes o f at least twice the A I W .  1 4  In contrast, low income earners, who  have to rely on O A S and C P P , can count on maximum income replacement on retirement o f only 40 per cent o f the A I W .  W h i l e pension related tax benefits have increased i n recent pension reform, retirement benefits have struggled to hold their own. C P P reform implemented i n 1998" was directed at the financing aspects o f the C P P , not to improving its benefits. Indeed the few changes that were made on the benefit side reduced rather than enhanced the benefits.  16  More drastic benefit reductions  were on the reform agenda but in the face of public opposition these were not implemented.  17  The  It makes sense, though, that those with higher incomes were more likely to make the maximum contribution as they had higher disposable incomes. l3  The maximum RRSP limit is set at a level calculated to replace 70 per cent of 2.5 times the AIW from RRSP investment income (70% x 2.5 AIW = 175% of the AIW). CPP will replace 25 per cent of the ArW. Iftoday's rules continue, OAS will be partially clawed back for those with retirement incomes at twice the AIW (roughly $69,000 in 2004 dollars) and will be completely clawed back for those with retirement incomes of $84,484. (See discussion above, Chapter One at pp. 51 -52 (OAS clawback) and 44-46 (effect of increased RRSP deduction limits).). M  An Act to establish the Canada Pension Plan Investment Board and to amend the Canada Pension Plan Act and the Old Age Security Act and to make consequential amendments to other Acts, S.C. 1997, c. 40. (See l5  discussion above, Chapter One at pp. 46-51.) For example the amount payable as retirement pension benefits is now calculated on the average of the retirees previous five years of earnings while prior to reform it was based on the previous three years of earnings. See l6  ibid.  The reform agenda was set by Department of Finance, An Information Paper for Consultations on the Canada Pension Plan (9 February 1996) (hereinafter " CPP Information Paper") and included such benefit reducing measures as a proposal to reduce the maximum pension payable from 25 per cent of the AIW to 22.5 per cent. 17  Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications, (Ottawa: The  Caledon Institute of Social Policy, 1996) (hereinafter "Caledon Institute") discusses the gender implications of the proposed reform package and represents an example of public resistence that undoubtedly influenced the decision not to include the more intrusive benefit reduction proposals in the reform package as it was implemented in January 1998.  -149erosion o f O A S universality has been discussed above. Additionally, proposals for reforming O A S / G I S were part o f the federal government's agenda in the 1996 federal budget, with implementation set for 2001. A s with C P P reform, improved benefits received little more than lip service i n the Seniors Benefit package that would replace O A S / G I S . While the poorest seniors would receive an additional $120 per year, the imposition o f family based (rather than individual based) entitlement and clawback at a higher percentage starting at a lower income level meant that most seniors would look forward to reduced retirement incomes under the Seniors Benefit proposal.  18  Negative public response to the proposed reduction o f Tier I benefits eventually led to  the withdrawal o f the reform package on July 28, 1998.  19  During the debate over C P P and O A S / G I S reform Canadians were told by the Liberal government that we are lucky to have the existing system, as purported increasing costs o f these programs jeopardize their sustainability. Hidden as it is within the income tax system, no such 'sustainability' debate surrounded Tier III reform. The clear message is that individual initiative in planning for retirement w i l l be rewarded by the government while collective (state) responsibility for economic well being i n old age, reflected i n the 'public' parts o f the retirement income system, cannot be permitted to grow and must be contained. The pyramidal structure o f the three tiered system has thus been inverted. More public funds are going into the top tier, to which the fewest and most economically advantaged Canadians have access, while the level o f public funding to the bottom tiers has been targeted for no growth and is likely to be further  See discussion above, Chapter One at pp. 51-55.  18  '"Department of Finance Canada, News Release 98-071, "Finance Minister's Statement on the Seniors Benefit" (28 July 1998). See discussion above, Chapter One at pp.55-56.  -150reduced in the future. Distribution of benefits, too, is inversely related to economic well being. The relatively small portion o f the population favoured with economic advantages receive more benefits from the retirement income system than they did prior to reform while the economically disadvantaged, a classification comprising more women than men, have only just held onto their inadequate pre-reform level o f benefits.  (b)  R e f o r m and the 1982 R e f o r m A g e n d a In 1982 the Liberal government set out these principles for pension reform:  • Elderly Canadians should be guaranteed a reasonable minimum income. • The opportunities and arrangements available to all Canadians to provide for their retirement should be fair. • Canadians should be able to avoid serious disruption of their pre-retirement living standards on retirement. 20  The language o f these principles is gender neutral and therefore on their face they encompass the goal o f eliminating the acknowledged gender inequalities within the retirement income system. H o w does the 1990s reform measure up to the principles articulated in the Green  Paper ? 1  Guaranteed reasonable minimum income The Green Paper acknowledged that the incomes o f some single seniors were too low and stated that remedying this situation was "one o f the priorities among the pension reform initiatives that the Government o f Canada w i l l pursue."  Green Paper, supra note 5 at 11.  20  Ibid. at 25.  2]  21  These needy seniors, the vast majority o f whom are  -151women, were targeted to receive an additional $120 per year on the implementation o f the Seniors Benefit in 2001. This increase would be a mere drop in the bucket toward the goal o f a 'reasonable minimum income' as the poverty gap  22  for single seniors in 1995 ranged from over $5,500 in  densely populated urban areas to $1400 in rural areas.  23  In any event, the Seniors Benefit is no  longer scheduled to proceed. In spite of the fact that a recovering economy is the reason given for scrapping the Seniors Benefit, to date no government initiative has been taken to improve the economic situation o f Canada's poorest seniors (unattached women) outside o f that proposed reform package.  Fair opportunities and arrangements available to all Canadians A s discussed in Chapters Two and Three, except for the inadequate Tier I benefits, "the opportunities and arrangements... to provide for... retirement" are available to Canadians through 24  workforce participation and spousal relationships. The level o f workforce participation required for entitlement to full C P P and R P P benefits, and for the accumulation o f sufficient R R S P savings to generate retirement income equivalent to full R P P benefits, is measured by a male standard in terms o f both wages and time spent in the workforce.  25  While virtually all women workforce  participants face this 'male standard' barrier to full access to benefits delivered through Tiers II and  In this context 'poverty gap' means the difference between maximum OAS/GIS and the poverty line (Statistics Canada's low income cut-offs). 22  Caledon Institute, supra note 17, Appendix "C", graphs entitled "Maximum Old Age Security/Guaranteed Income Supplement Versus Poverty Line for Different Areas, Single Senior, 1995" and "Poverty Gap for Single Seniors and Elderly Couples with Maximum OAS/GIS, by Community Size, 1995". These graphs also show that the additional $120 per year promised to needy senior couples by the Seniors Benefit will also miss the mark for those living in communities with populations greater than 30,000. 23  Supra note 20.  u  "See discussion in Chapter Two above, at pp. 63-72.  -152ni, the only barrier that the retirement income system recognizes is the 'spousal' barrier, which assumes that less than full workforce participation by women is causally linked to spousal status and that providing access through spousal status w i l l effect gender equality. The reality is, though, that the 'spousal' barrier is not the only barrier to full access to the benefits o f the retirement income system that women face, and not all women who face barriers are spouses.  26  Accordingly,  neither mode o f access to the retirement income system can be considered fair for Canadian women.  Gendered access problems existed before the 1990s pension reform and they still exist today. Indeed, since Tier I and II benefits have not been improved, the expansion o f the tax subsidized part of the retirement income system has effectively increased the gap between women's economic realities and the goal o f economic equality for women i n the retirement income system. A s discussed previously fewer women than men are R P P members, and women generally make 27  lower R R S P contributions than men.  28  Women's average wages are much lower than m e n ' s and 29  accordingly fewer women than men w i l l have had their R R S P limits extended by the 1991 Tier III  Perhaps some who are spouses would prefer not to be but their choices are constrained by their economic inequality. As Claire Young pointed out, access to tax subsidies through spousal relationships "imposes significant limitations on [women] by restricting their choices about how and with whom they live their lives." (Claire F.L. Young,, "Public Taxes, Privatizing Effects, and Gender Inequality" in Susan B. Boyd, ed., Challenging the 26  Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 307 at 323.) 27  See above, Chapter Two at p. 68 and Chapter Three at pp. 111-116 and 133-134.  The primary reason that women make lower RRSP contributions than men is that women are under represented among those who are sufficiently economically advantaged to make full use of their RRSP limits. In addition, because they generally earn less annual income than men, their RRSP contribution limits are generally lower than men's. (Recall that RRSP limits are calculated as 18 per cent of annual earned income provided that contributions cannot exceed a statutorily set level, which for 1998 is $13,500. See supra Chapter One note 131.) Therefore, even if they could afford to make full use of their contribution limits, women would accumulate less RRSP savings, and benefit less from related tax subsidies, than men. 28  'See discussion above, Chapter Two at pp. 63-64.  -153reform. In fact, most women, because they earn less than $41,667 per year, w i l l have had their contribution limits reduced by the reform.  30  The purpose o f Tier III reform was to make the "opportunities and arrangements" within Tier HI fairer, by equalizing R R S P and R P P tax benefits and by allowing contribution limits to be carried forward rather than limited to one taxation year.  31  While fairness to Canadians who have  access to Tier III (mostly men) may have been improved, fairness to all Canadians within the retirement income system as a whole has not. Allocating additional financial support to Tier III, instead o f to Tier I which requires additional funding to ensure that elderly Canadians are "guaranteed a reasonable minimum income", or to Tier II which presently replaces much too l o w a percentage o f the A I W to ensure continuation o f pre- and post- retirement lifestyle for average income earners and which makes some accommodation for women's culturally assigned responsibilities for reproductive work, advantages the already advantaged and compounds unfairnesses that previously existed in the retirement income system.  A v o i d serious disruption o f pre-retirement living standards Tiers II and III were intended to be the parts o f the retirement income system that gave Canadians the opportunity to maintain their pre-retirement living standards on retirement. C P P was deliberately set at a level that would give working Canadians a retirement income o f 45 per cent o f their pre-retirement incomes, provided that their pre-retirement incomes did not exceed the  The 1991 ITA amendments raised the contribution limit but reduced the percentage of annual income that could be contributed from 20% to 18%. See supra, Chapter One note 132. 30  3i  See discussion above, Chapter One at pp. 44-46.  -154AIW.  3 2  Legislators recognized that continuing a pre-retirement lifestyle i n retirement required 55  to 80 per cent o f pre-retirement income and left the topping up by 10 to 35 per cent to Tier LTl.  33  Because O A S benefits lost ground in relation to the A I W during the 1970s, in effect C P P lost some of its lifestyle replacement effectiveness and instead topped up O A S to give workforce participants "a reasonable minimum income" . Tier I l l ' s lifestyle replacement role for average income earners 34  was accordingly increased.  Recent reform o f C P P did nothing to improve the ability o f Canadians "to avoid serious disruption o f their pre-retirement standards on retirement." Rather, C P P retirement income benefits w i l l be lower for many, i f not most, Canadians because the formula for setting the maximum benefit has changed from an average of the Y M P E over the previous three to a five year average.  35  Nor did the '90's reform restore O A S ' s position relative to C P P . However, Tier III reform improved the ability o f economically advantaged persons to maintain pre-retirement lifestyles associated with incomes up to more than 2.5 times the A I W .  3 6  Thus, recent reform o f the  retirement income system fails to respect the third principle advocated by the Green Book, and for the same reason as it failed to respect the second principle - its attention was focussed on equalizing the benefits within the top tier of the system only. The impact on the retirement income system as  See discussion above, Chapter One at pp. 28-29. Recall that at the time CPP was enacted OAS benefits were 20% of the AIW. By 1980 OAS benefits had dropped to 14 per cent of the AIW. 32  "Ibid.  Full OAS and CPP benefits provide an income that hovers around the poverty line. See discussion above, Chapter One at pp. 37-38. 34  35  See discussion above, Chapter One at p. 48.  See discussion above, at pp. 147-150. I say "more than 2.5 times the AIW" because the potential Tier III income replacement is 70 per cent of pre-retirement income and these retirees will also, in all likelihood, receive full CPP. 36  -155a whole was ignored. Only a few advantaged Canadians benefit from improvements to the top tier, whereas improvement to the O A S would ripple upward so that the lifestyle replacement benefit o f the C P P would be improved for a greater number o f Canadians. Alternatively the additional public financial support to Tier m could have been directed toward a form o f C P P reform that would increase the post-retirement income replacement potential o f C P P to more than 25 per cent o f the A I W , an improvement that all workforce participants would benefit from and one that would be fairer for women because o f the C P P ' s concessions to their reproductive work.  (c)  R e f o r m and the Restructuring A g e n d a H o w did pension reform stray so far from the principles that were supposed "to provide the  basis for reform" ? 37  W h y did reform i n the 1990s fail so completely to address what was  acknowledged to be among "the most serious shortcomings"  38  i n Canada's retirement income  system - equality and economic security for women i n their retirement years? Janine Brodie's analysis o f this period o f the contemporary political climate in Canada and its impact on the Canadian women's movement provides some insight into these questions.  39  A t the time that the Green Paper was released in 1982, Canada was nearing the end o f what  ' Green Paper, supra note 5.  3 1  Ibid. at 2.  M  Janine Brodie, Politics on the Margins: Restructuring and the Canadian Women's Movement (Halifax:  39  Fernwood Publishing, 1995). See also the collection of essays in Isabella Bakker, ed., Rethinking Restructuring: Gender and Change in Canada (Toronto: University of Toronto Press, 1996) (hereinafter "Rethinking Restructuring"). The discussion that follows is necessarily a simplification of the politics of restructuring. Its purpose is to add to an understanding of how pension reform failed to address the issues that were central to the reform initiative by setting the political context in which recent reform occurred. The sources from which I have drawn provide a comprehensive and provocative analysis of the topic from a feminist perspective.  -156Brodie refers to as its 'Keynesian welfare state' era.  40  During this era "[t]he pre-war idea o f a  means-tested welfare was replaced with the concept o f social security for all its citizens as a universal right."  This political philosophy is reflected in the opening statement o f the Green  41  Paper.  One of the principal goals of any society is to ensure that its citizens are able to enjoy fulfilling and constructive lives without undue fear for their basic security. 42  and in the Green Paper's elderly Canadians.  43  stated principles for pension reform directed at achieving that goal for  B y the time pension reform began to be implemented, i n the late 1980s and  into the 1990s, Canada was entering what Brodie refers to as the third period o f major social transformation in Canada, from Keynesian economics to neoliberalism as the state's governing orthodoxy.  44  The impetus for the transformation to neoliberalism "as the new wisdom o f governing" is globalization, the recent trend toward international production o f goods and services.  Technological and managerial changes are taking place that allowfirmsto divide the different aspects of their operations globally in order to take advantage of the lowest-cost raw materials, the best research and development, the highest-quality  40  Brodie, ibid, at 39-47.  "Ibid, at 39. Green Paper, supra, note 5 at 1.  42  These are the principles set out above at p. 150 and discussed at pp. 150-155.  43  ^Brodie, supra note 39 at 16-18. Ibid. at 10.  4i  45  -157assembly, and the most effective marketing.'  Other changes within Canadian society occurred more or less coincidently with increased globalized production, including economic recession, increased unemployment, expanding divorce rate, growing numbers o f lone parent families, an aging population, a growing national debt.  47  While there probably is not a causal relationship between globalization and all o f Canada's economic woes,  48  proponents o f a market driven economy seized the opportunity to connect  globalization to government spending on social programs and to promote the view that in a competitive international economy the country has no choice but to divest itself o f the costly welfare state.  49  .. .Canadians are being bombarded at every turn with the message that things have to change, that we are uncompetitive in an increasingly global economy, and that we can no longer afford the security and services that were once guaranteed to all Canadians by the postwar welfare state. 50  46  Isabella Bakker, "Introduction: The Gendered Foundations of Restructuring in Canada" in Rethinking  Restructuring, supra note 39 at 3.  Ibid. at 5-6 and Brodie supra note 39 at 14-15.  41  The view that Canada's economic policies over the past decade, supposedly in response to the trend toward international production of goods and services, have been the cause and not the effect of Canada's economic problems, is argued by Duncan Cameron and Ed Finn, "10 Deficit Myths: The Truth about Government Debts and Why They Don't Justify Cutbacks" (Ottawa: Canadian Centre for Policy Alternatives, 1996) and Lisa C. Phillips, "The Rise of Balanced Budget Laws in Canada: Legislating Fiscal (Ir)Responsibility" (1996) 34 Osgoode Hall Law Journal 681. 48  'Ibid. Note Bakker's observation that welfare state spending appears to have played less of a role in contributing to the deficit than decisions made on the revenue side: "Revenue decisions such as a series of tax expenditures or tax breaks for individuals and corporations were reported to have cost the Canadian federal treasury at least $30 billion by the late 1980s. By using tax expenditures, governments could give the appearance of being fiscally responsible while still offering fiscal advantages to specific individuals." (Supra note 46 at 5.) In the current pension context, Claire Young notes that "For 1998, it is estimated that the value of the deduction and deferral [for RPP contributions] will be over $18 billion making it the single largest tax expenditure in that year." (Claire F.L. Young, "Spousal Status, Pension Benefits and Tax: Rosenberg v. Canada" (1998) 6 Canadian Employment and Labour Law Journal 1.) 45  50  Brodie supra note 39 at 9.  -158In Western democracies, including Canada, there is little question that, over the past decade or so, the Keynesian welfare state has crumbled and neoliberalism has moved in to fill the void.  Neoliberalism demands that governments respond to globalization by:  • maximiz[ing] exports • reduc[ing] social spending • curtailing] state economic regulation • enabling] market forces to restructure national economies as part of transnational or regional trading blocs. 52  Keynesian welfare state economics, which "asserted the primacy o f the state over the 'invisible hand' o f the market and engendered widespread public expectations that governments were responsible for meeting the needs o f their citizens" , clearly does not fit the neoliberal agenda 53  which seeks to dilute democratic controls over economic life.  Obviously such a dramatic  transformation does not happen either quickly or painlessly; rather it occurs over a period o f 'restructuring':  Restructuring is a key word which refers to a prolonged and conflict-ridden political process during which old assumptions and shared understandings are challenged and are eventually either rejected or transformed while social forces struggle to achieve a new consensus - a new vision of the future to fill the vacuum created by the erosion of the old. The concept of restructuring represents the simultaneous 'combination of falling apart and building up again' of an entire political-cultural order. 54  'Brodie considers that the changing governing order began to emerge following the elections of Margaret Thatcher as Prime Minister of Britain in 1979 and Ronald Regan as President of the United States in 1980. Ibid, at 16. 5  Ibid.  52  "Ibid, at 15. Ibid.  54  -159Recent pension reform has taken place since the late 1980s, a period when restructuring dominated Canada's political agenda.  55  Associated with the political restructuring from Keynesian to neoliberal economic policies is shrinkage o f the public portion o f the public-private continuum and transformation o f the social 56  safety net "to make it fit with the market-based, self-reliant, and privatizing ideals o f the new order."  Recent changes to the retirement income system reflect these characteristics o f  57  restructuring. A s M o n i c a Towson observed,  ...government policy on pensions is a clear fit with trends in most other areas of current economic and social policy. The emphasis is to be on smaller government, privatization, individual initiative, and market driven solutions. 58  Thus, where the reformed pension system misses the mark with respect to the principles prescribed in the Green Paper, it scores on the restructuring agenda. The policy o f sustainability at best for Tiers I and II benefits while expanding Tier l U benefits makes no sense when measured against the reform agenda set in 1982. O n the other hand, a policy o f rewarding the self-reliant who make their own retirement income arrangements in the private market while containing the level of assistance that the state w i l l provide directly to those who have not taken care o f their own needs is clearly in line with the demands that neoliberalism has placed on governments. It appears that the pension system has not been reformed; it has been restructured.  Tbid. at 17.  'Ibid, at 49-63. Ibid. at 57. These two characteristics of restructuring are discussed in greater detail below, at pp. 160-171.  7  *Townson, supra note 1.  -160II  (IN)EQUALITY AND THE RESTRUCTURED RETIREMENT INCOME SYSTEM  The gendered effects of restructuring has been the subject o f much recent Canadian feminist scholarship.  59  The discussion o f the gendered effects o f restructuring on the retirement income  system that follows draws from these sources.  60  The focus o f this part o f the thesis is the  relationship o f two key features o f restructuring, the shifted public/private divide and the related transformation of the social safety net, to continuing gender inequality i n the restructured retirement income system.  (a)  The Public/Private Divide A s discussed previously, many aspects of western democratic societies are organized around  'public' and 'private' activities.  61  While feminist scholars have exposed the separate and  autonomous 'public' and 'private' as constructed around ideas of social and economic relations that often do not represent reality, the designation o f an activity as 'public' or 'private' has very real, and often gendered, effects. B y characterizing an activity or responsibility as private or public, the state can change the shape o f the boundary between public and private and enlarge or diminish its  See Brodie, supra note 39; Janine Brodie, Politics on the Boundaries: Restructuring and the Canadian Women's Movement (Toronto: Robarts Centre for Canadian Studies, York University, 1994); Shelagh Day and Gwen Brodsky, Women and the Equality Deficit: Restructuring Canada's Social Programs (Ottawa: Research Directorate, Status of Women Canada, 1998); Martha Jackman, "Women and the Canada Health and Social Transfer: Ensuring Gender Equality in Federal Welfare Reform" (1995) 8 C.J.W.L. 371; and the work of the many feminist scholars who contributed to Rethinking Restructuring, supra note 39, to Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997), and to Janine Brodie, ed., Women and Canadian Public Policy, (Toronto: Harcourt Brace, 1996). 59  As the effects of restructuring on women and members of other disadvantaged groups more generally is complex, a comprehensive analysis of their many and often contradictory features is beyond the scope of this thesis. Those seeking a better understanding of this issue should refer to the sources cited in the previous note. 60  6l  See discussion above, Chapter Two at pp. 105-109.  -161own sphere o f operation.  62  Canada has experienced three periods o f restructuring,  63  each characterized by a shifted  boundary between public and private that altered the relationships between the state (public) and the market (private), and between the state (public) and the family (private).  64  The laissez-faire  state, characterized by a rigid dichotomy between the public sphere (the state) and the private sphere (the market and family), emerged from the first transformation and lasted from around the 1840s to the depression o f the 1930s.  65  The second transformation, to the Keynesian welfare state,  moved the public/private boundary to allow the state to regulate the economy " i n order to maximize the collective welfare" . The provision o f basic needs was no longer completely within 66  the domain o f the private family; the state took responsibility for providing these "especially when private mechanisms failed" . 67  The neoliberal state, which  seeks to shrink the public and  "increas[e] the autonomy o f market forces and o f the family"  68  is emerging from the third  restructuring period. Thus history confirms that the public/private divide is a construct, not a  See Susan Boyd, "Can Law Challenge the Public/Private Divide? Women, Work and Family" (1995) Windsor Yearbook of Access to Justice 161, and Judy Fudge, "The Public/Private Distinction: The Possibilities of and Limits to the Use of Charter Litigation to Further Feminist Struggles" (1987) 25 Osgoode Hall Law Journal 485. See also the discussion of the relationship between the public/private divide and the way that states "state" themselves in Brodie, supra note 39 at 27-28. 62  Brodie, ibid, note 39 at 33-34. The third restructuring period is currently ongoing.  63  Ibid.  M  Ibid. at 34-38.  6s  Ibid. at 39.  6h  "Ibid. Hbid. at 49.  -162distinct line.  69  It is a dynamic boundary that has shifted over time " i n relation to the role o f the  state at particular moments, in particular contexts and in relation to particular issues."  70  A s discussed previously, there are close connections between familial ideology (the dominant family form) and the public/private divide, and between familial ideology and gender 71  relations.  72  The effect o f these inter-connections is that a shift in the location o f the boundary  between public and private is not, and has not been, gender neutral.  73  Armstrong encapsulates the  gendered effect o f shifting from the boundary set by the Keynesian welfare state to the boundary that neoliberalism seeks:  As many feminists have pointed out, the state has never been a neutral arbitor [sic] or primarily a defender of women. But the state can, and does, set rules or provide services that can protect the weakest. Private firms and individual households offer much more arbitrary and unreliable protections, especially for the weakest. ...[T]he shifts in public and private will be played out in women's lives. 74  The restructured 'private' There are two aspects to the 'private' sphere - the family and the market. Restructuring  69  See discussion above, at pp. 106-108.  70  Susan Boyd, "Challenging the Public/Private Divide: An Overview" in Susan B. Boyd, ed., Challenging  the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 3 at 4. 7l  See discussion above, Chapter Two at pp. 105-109.  72  See discussion above, Chapter Two at pp. 103-105.  See Brodie, supra note 39 at 33-48 for a discussion of the gendered effects of the transformations to the laissez-faire state and to the Keynesian welfare state. 73  74  Pat Armstrong, "Restructuring Public and Private: Women's Paid and Unpaid Work" in Susan B. Boyd,  ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press,  1997) 37 at 57.  -163-  seeks to expand both, and to limit the state's role in matters relegated to them.  75  Regulation o f the  economy is being relegated to the private market, with redistribution of market earnings left to the family. Increasingly responsibility for social programs is being relegated to the community where much o f the work is done by society's caregivers (women) either i n the market (for low pay) or within the family (for no pay).  76  Market regulation o f the economy The Keynesian welfare state took an active role i n managing the economy, developing an elaborate bureaucracy and complex legislation consistent with its agenda o f "promoting domestic welfare and protecting the national economy from unstable international forces."  77  A key  imperative of neoliberalism is a self-regulating, market driven economy with little regulation and minimal protection for national economies from international market forces. Downloading control over the economy to the private market, accompanied by promotion o f international competition through trade agreements, has resulted i n the growth o f low paying, part time and short-term 78  This is discussion obviously presents a simplistic characterization of the restructured relationships between state and market, state and family and market and family. The complexities of these relationships cannot be encapsulated without going into more detail than can be contained within the parameters of this part of the thesis. For a comprehensive understanding of the restructured public and private, see the resources listed supra note 59. 75  Marlee Kline, "Blue Meanies in Alberta: Tory Tactics and Privatization of Child Welfare" in Susan B. Boyd,ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 330 illustrates this aspect of restructuring in the province of Alberta. As well, the gendered impact of (re)assigning social programs to the 'private' sphere is discussed by Armstrong, supra note 74. 76  Brodie, supra note 39 at 16. Brodie notes that political intervention into the economy by the Keynesian welfare state "presumed a stable working and middle class" and the form of family that is reified in familial ideology. (Supra note 39 at 15.) See also Janine Brodie, "Restructuring and the New Citizenship" in Isabella Bakker, ed., Rethinking Restructuring: Gender and Change in Canada (Toronto: University of Toronto Press, 1996) 126 at 128130. 77  See Marjorie Griffin Cohen, "New International Trade Agreements: Their Reactionary Role in Creating Markets and Retarding Social Welfare" in Rethinking Restructuring, supra note 39 at 187 and Christina Gabriel and Laura MacDonald, "NATFA and Economic Restructuring: Some Gender and Race Implications" in Rethinking Restructuring, supra note 39 at 165. 78  -164contract work, some o f which is performed in the unregulated environment o f the home.  79  A s well  paying, full time, secure employment increasingly gives way to these new job forms, the necessity for two wage earners to support a family also increases. Unfortunately the improvement o f the conditions that would relieve women o f some o f the reproductive responsibilities that hamper their full workforce participation has not been on the restructuring agenda.  80  Brodie identifies five ways in which the neoliberal imperatives o f shrinking the public sector and allowing market forces to restructure national economies i n line with international trade have impacted on women: •  81  poverty has become increasingly gendered and is "particularly acute among female-headed households and among elderly women";  •  82  "women have acted as 'shock-absorbers' during adjustment, both by curtailing their own consumption and by increasing their work to compensate for household income loss";  •  83  privatization o f social programs and cuts in state funding for social welfare "mean that  Armstrong, supra note 74; Pat Armstrong, "Feminization of the Labour Force: Harmonizing Down in a Global Economy " in Rethinking Restructuring, supra note 39 at 29; Jane Jensen, "Part-Time Employment and Women: A Range of Strategies" in Rethinking Restructuring, supra note 39 at 92; Amanda Ababa Ocran, "Across the HomeAVork Divide: Homework in the Garment Industry and the Failure of Employment Regulation" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 144. 79  See Katherine Teghtsoonian, "Who Pays for Caring for Children? Public Policy and the Devaluation of Women's Work" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 113 and Nitya Iyer, "Some Mothers Are Better Than Others: A Reexamination of Maternity Benefits" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press, 1997) 168. 80  Brodie, supra note 39 at 19-20 While the list is derived from a study by the U S Agency for International Development, a report by the National Action Committee on the Status of Women entitled "Review of the Situation of Women in Canada" shows the same trends developing in Canada. 8l  * Ibid. at 19. 2  Ibid.  K  -165social services are shifted from the paid work o f women in the public sector to the unpaid work o f women in the domestic sphere"; •  84  "gains made toward the goal o f gender equality during the 1970s are being rapidly eroded due to shifts i n the labour market which are producing few good jobs for women (or men, for that matter), and because o f reductions i n childcare, education, and retraining programs";  •  85  and  because o f public expenditure constraints, progress toward improved working conditions for women i n the public sector, such as pay equity, "has come to a standstill".  The gendered effects o f an expanding market are relevant to Canada's  86  restructured  retirement income system. First, poverty among the elderly is increasingly 'feminized'. A m o n g the unattached elderly who have incomes below the poverty line, the gender gap increased from 10.9 per cent in 1980 to 22.1 per cent in 1995.  87  Second, the sharp increase in the number o f  women participants in the work force (from 42 per cent in 1974 to 52 per cent in 1994 ), and 88  particularly women with children (from 50 per cent in 1981 to 63 per cent in 1994 ), suggest that 89  women are "increasing their work load to compensate for household income loss."  90  However, as  Ibid. at 19-20.  x  Hbid. at 20. Hbid.  'National Council of Welfare (Canada), Poverty Profile 1995 (Ottawa: Supply and Services, 1997) at 19. "Statistics Canada, Women in Canada: A Statistical Report, 3 ed. (Ottawa: Statistics Canada, 1995) at 9. rd  'Ibid.  °Brodie, supra note 39 at 19.  -166more women are entering the work force at the same time that ' g o o d ' jobs are becoming scarcer, 91  this increased work load is unlikely to translate into improved opportunities to accumulate entitlement to lifestyle replacement retirement benefits. Thus the shock that women absorb for their families during their working lives w i l l continue to be felt by them on retirement. Third, the shifting o f social services work from women who are employed in the public sector to women i n the home means that fewer women w i l l be employed in jobs that provide RPPs and the ability o f more women to participate fully in the paid workforce, and thereby earn retirement pension entitlement, w i l l be curtailed.  92  Fourth, the erosion o f gender equality gains in the workforce can  only increase the difficulty that women have i n accessing the third tier o f the retirement income system. That, in turn, w i l l increase the gender gap in retirement income and force increased reliance by women on access through dependency on a spouse. Finally, pay equity and similar improvements for women are essential to effect equality o f access to pension and pension related tax benefits through workforce participation.  'Re'-privatizing social welfare One of the imperatives of the current restructuring is that people should look after their own needs without making demands on the state.  93  Because o f inequalities that are steeped i n their  responsibilities for unpaid care giving work and assumptions o f their dependency, women are unable to look after their own needs to the extent that men are. While women have always been expected to look to their altruistic families (spouses) to provide for them, social programs  Ie. secure full time jobs with decent wages and benefits.  9{  92  This effect is discussed further below, in the discussion of'"Re'-privatizing social welfare".  93  See discussion of the 'new citizenship' below at pp. 171-175.  -167implemented by the Keynesian welfare state provided support to women when recourse to family support failed.  94  The leaner, meaner restructured state seeks to '"re'-privatize" social programs 95  to the market and the family, placing "a renewed emphasis on the heterosexual patriarchal family as the fundamental building block i n restructuring."  96  Since it is the economically disadvantaged  members o f society (among whom women are over represented) who are the intended recipients of social services, entitlement removes barriers to access. 'Re'-privatizing removes entitlement and access is constrained by the power relationships within the family.  Downloading social programs to the market has gendered effects. Services provided by the market are inaccessible to those without financial resources, and many women have to rely on economic redistribution within the family, which is subject to gendered power relations and generally not within their control.  97  Further, the market w i l l continue to provide only programs that  are profitable. Where the market does not take over, or continue, programs that have been cut by  As discussed in Chapter Three, social programs, including the retirement income system, developed during the Keynesian welfare state period were highly gendered. I do not mean to valorize these programs here, only to draw the distinction that the state filled in where the family failed in the Keynesian welfare state while 're'-privatizing social programs removes entitlement to state support. 94  Brodie highlights the message hidden in the 're' rhetoric of neoliberalism - that the services delivered through social welfare programs "are being returned to some place where they naturally belong." (Supra note 39 at 53.) 95  96  Boyd, supra note 70 at 19.  The gendered nature of the distribution of income, that is that men earn more than women, has been discussed above. (See Chapter Two at pp 63-72.) The distribution of wealth is gendered as well, according to Lisa Phillips, "Tax Policy and the Gendered Distribution of Wealth" in Rethinking Restructuring, supra note 39 at 141. The fallacy of the assumption of altruistic (re)distribution of economic resources within the family, and its gendered effects, have been discussed earlier in this thesis. (See above, Chapter Two at pp. 95-97 and Chapter Three at pp. 135138.) In "Sex, Tax and the Charter" (1995) 2 Review of Constitutional Studies 221 at 279, Lisa Phillips and Margot Young connect this fallacy to inequality for women within the ITA. (See supra Chapter Three notes 89 and 91 and accompanying text.) That same reasoning is applied here in a more general context. That is if women's realities reveal that sharing of tax benefits ought not to be assumed, then the sharing of income ought not to be assumed either, and if income is not shared, the person who controls it also controls how it is spent. 97  -168the state, responsibility falls to the family and the unpaid work o f women.  98  W i t h the additional  work imposed on them, the ability o f women to provide for themselves through workforce participation is further inhibited. In addition, more women than men were employed by the state in jobs relating to social service delivery. When such programs are cut, these women are forced to leave reasonably remunerated employment in a sector where R P P membership is high and to either find similar work i n the market where full time work with decent pay and benefits is becoming increasingly scarce " o r to face unemployment.  The restructured retirement income system has downloaded responsibility for providing for continuation o f pre-retirement lifestyle onto the market and family by expanding Tier III while containing Tiers I and II. It is a mistake to think o f Tier III as private, however, as it has both public and private aspects.  100  The economically advantaged (RPP members and those with the  economic resources to make full use o f their R R S P contribution limits - mostly men) reap the public benefit o f significant tax savings.  101  The public benefit available to the economically  disadvantaged (mostly women) is limited at best, and these persons are more likely to experience the full negative impact o f Tier I l l ' s private aspects. They w i l l be dependent on altruism within the family for their economic well-being. Although restructuring did not create this effect, it has compounded it in two ways. First, the impact o f restructuring has eliminated many o f the well  9S  Day and Brodsky, supra note 59 at 10-11.  "See Armstrong, supra notes 74 and 79. 100  See discussion above, Chapter One at p. 26-27.  Tax subsidies to encourage 'private' activities exemplify a distinction between the laissez-faire state, which endeavoured to keep the government and the market completely separate, and the neoliberal state which "requires political intervention" to perform the "job of convincing us that market rationality (itself a questionable term) should take precedence over political negotiation and the collective will of the public." (Brodie, supra note 39 at 51.) 101  -169paying jobs with good benefits that women held, thereby reducing their opportunities to provide adequate retirement income entitlement for themselves. Second, the way i n which the retirement income system has changed its shape through restructuring puts more o f the system's benefits i n the hands o f the person within the family who has the most economic power and relies on his altruism for fair distribution o f these benefits.  The restructured 'public' If economic regulation and social welfare are relegated to the 'private' sphere, then what is left i n the 'public'? In contrast with the laissez-faire state form, for neoliberalism a smaller 'public' does not necessarily mean a smaller government. The state's role is to promote the demands o f neoliberalism as the only choice o f state form i n a globalized economy and to "enforc[e] the market model in virtually all aspects of everyday l i f e . "  102  The restructured retirement  income system provides an example of how the state carries out this role and exposes the malleable nature o f the public/private boundary.  In spite o f the significant state (public) financial support, Tier in is designated the 'private' part o f the retirement income system.  103  Recently this 'private' part o f the retirement income  system has been strengthened at the expense o f the 'public' part. That strength exchange is obscured for many by the different types of benefits delivered (retirement pension and tax benefits). This obscurity could  l02  be clarified for Canadians i f there was political w i l l to do so.  An  Brodie, supra note 39 at 51. See also Bakker, supra note 46 at 3-4.  This separation of the retirement income system into 'public' and 'private' parts is seen in government publications relating to recent reform of the system. (See, for example, Government of Canada, Securing Canada's Retirement Income System (Ottawa: Supply and Services, 1997).) The distinction is also made by those who critique the system, possibly for consistency. (See, for example, Townson, supra note 1.) l03  -170understanding o f the role of Tier HI that is informed by tax expenditure analysis exposes the fallacy o f separating Tier HI from the rest o f the retirement income system and labelling it as 'private'. Tax expenditure analysis considers tax subsidies, that is taxes that would be collected but for tax breaks used by the government to reward certain kinds o f behaviour, to be government revenues. Although they remain in private pockets, uncollected tax revenues are as much public funds as are collected taxes. Indeed, to the extent that the Canadian government publishes tax expenditure accounts, it recognizes these as public funds. However, at least as far as the retirement income system is concerned, the government describes the programs that are paid for by uncollected tax revenues as 'private'. This public/private distinction masks the connection that exists between the tax and pension systems and their benefits.  104  While the government could remove this mask and  consider the whole retirement income system as a 'public' program, it is important to the restructuring agenda that Tier III, the part o f the system that responds to private initiative in the market (albeit with 'public' monies), be promoted as 'private' rather than 'public'. B y separating the 'public' from the 'private', different treatment can be imposed - Tier III, the tier that rewards private initiative, can be expanded, while Tiers I and II, the tiers that provide economic security based on collective entitlement, can be contained. In this context, the politically constructed public/private dichotomy provides a political rationale for enhancing benefits to the economically advantaged while reducing or containing those available to the economically disadvantaged. Without the divide the grossly inequitable nature o f the different treatment, which provides public  '""Government rhetoric surrounding the Seniors Benefit illustrates this point. For example, the Department of Finance portrayed the Seniors Benefit as completely tax free and therefore completely disconnected from the tax system. (See Department of Finance, "The Seniors Benefit: Summary" in Budget Papers (Ottawa, Department of Finance Canada, 1996).) However, critics pointed out that the financial impact of the Seniors Benefit could only be properly understood by looking at both the benefit that a particular senior would receive and the income tax that he or she would have to pay. (See John Geddes, "Citizens Revolt", 111-16 Maclean's 12 (April 20, 1998). See also the discussion in Chapter One, above at pp. 53-56.)  -171fmancial support for Canadians earning up to 2.5 times the A I W to maintain their lifestyles on retirement while those earning the A I W and below can look forward to retirement incomes ranging from at to well below the poverty line, would be exposed.  (b)  105  Transformed Social Safety Net Neoliberalism valorizes individual initiative and self-reliance and restructuring seeks to  transform the social safety net accordingly. Thus social programs have not been cut primarily as a cost saving measure, but because the Keynesian idea that everyone is entitled to a degree o f protection from market forces is incompatible with the neoliberal view that individuals should take care o f their own needs and not make demands on the state.  106  Although transforming the welfare  state is related to an expanded 'private' and shrunken 'public', there is more to the transformation than the state trimming itself down to a smaller size.  The new citizenship The valorization o f individual initiative and self reliance has changed what it means to be a Canadian citizen.  107  The postwar welfare state acknowledged that such social ills as poverty often  reflect social, not individual, fault and that collective responsibility is the appropriate response. The neoliberal state rejects this notion o f citizenship and seeks to replace it with a citizenship that corresponds with market values.  105  l06  See discussion above at pp. 147-150.  Brodie, supra note 39 at 54-56.  Ibid. at 56-63. See also Bakker, supra note 46. As well, this theme may be observed in Day and Brodsky, supra note 59, and Jackman, supra note 59. i01  -172The neo-liberal economic agenda... dictates a particular approach to government. The private sphere - the home, the market - is considered worthy of enlargement and sanctification, and the public sphere, including the institution of government itself, is considered dangerous and best kept small. When governments follow this agenda, they treat their capacity to impose limits on the market, in the name of collective values, as suspect, and permit the unqualified assertion of marketoriented values, such as self-reliance and competition. The citizenry becomes individualized. The emphasis is not on understanding and addressing the "social and structural foundations of dependency" but on 'individual solutions to what are perceived to be individually determined problems.' 108  Social problems such as unemployment and poverty have been redefined as deriving from the individual and it is the individual's obligation as a citizen to solve his or her problems without putting demands on the state. Brodie points out the gendered effects o f this new citizenship:  ..the clear message is that it is up to every 'good individual' to become more flexible and self-reliant and to make fewer demands on the state. But, a deeply gendered and unequal social order, by definition, means women can only be gendered individuals. As much as this [neoliberal] rhetoric tries to cast women as individuals detached from a deeply gendered social order, then, it must necessarily recast them as 'bad individuals' - the ones who are different, dependent, and blameworthy for not successfully leaping into independence. Obviously it is much harder to leap when an individual is burdened by unpaid caring work and gender-based discrimination. 109  B y its very nature, individualized citizenship leaves no opening for claims based on group related systemic inequalities, whether the claims are based on gender, race, sexual orientation or other culturally reinforced disadvantage. In short, individualized citizenship imposes limitations on the meaning o f equality.  The 'ordinarv'/'speciar dichotomy Brodie assigns the terms "ordinary Canadian" and "special" to the to the 'good' (ie.  'Day and Brodsky, supra note 59 at 9. 'Brodie, supra note 39 at 57.  -173"flexible and self-reliant" ) and ' b a d ' (ie. "different, 110  dependent, and blameworthy"" ) 1  individuals." The "ordinary Canadian" is someone who claims no special interest and makes no 2  demands on the state - he (sic) is defined by what he is not:  The ordinary Canadian is disinterested, neither seeking special status nor treatment from the state. He is neither raced, nor sexed, nor classed; he transcends difference. So who is he? A close reading of the current conception of the ordinary Canadian reveals that he can only be a white, heterosexual, middle-class, English-speaking male because, in contrast to him, everyone else is 'special' in some way. 113  Neoliberalism seeks to reify the 'ordinary Canadian' and to identify all others (ie. those who make demands on the state) as self interested members of special interest groups who seek from the state something they did not earn and therefore do not deserve. B y definition, then, the interests o f 'special' Canadians must be adverse to those o f the 'ordinary' Canadian. This ordinary/special dichotomy imposed by neoliberalism "serves to delegitimize and silence all those who declare themselves to be different, marginalized, and structurally disadvantaged.""  4  A s the 'special' are  cast outside the community o f 'ordinary' Canadians, the structural inequalities that advantage the 'ordinary' and impede the 'special' from transcending the experiences o f their gender, class and race become obscured and claims by the disadvantaged for equality become marginalized."  5  See ibid. These individuals make 'private' retirement income arrangements and do not need 'public' pension programs. ll0  ' 'Ibid. These individuals have not taken care of their retirement income programs and look to the government to provide for them, 1  " Brodie, supra note 39 at 71-73. 2  Ibid. at 73.  m  Ibid.  m  "Although this thesis has focussed on gender inequalities within the retirement income system, at this point it may be helpful to recall that class, race, citizenship status and disability intersect with gender to "complicate and deepen" economic barriers faced by women in general. See supra, Chapter Two notes 44-47 and accompanying text. 1  -174The 'ordinary'/'special' dichotomy is evident in the restructured retirement income system. Enhanced Tier HJ benefits effected by amendments to the I T A reduced the inequalities within that tier  116  and improved the system for those who are not afflicted with gender, race, class or other  structural disadvantage (ie. 'ordinary' Canadians). In contrast, no improvements were made to Tiers I and II, the parts o f the system that involve the state, rather than the individual, directly in providing retirement income. The 'ordinary' Canadian, who provides for himself and makes no demands on the state, does not need improvement to Tiers I and II. Directing improvement only within Tier III obscures the raced, sexed and classed nature o f access to that part o f the system and illustrates the way in which programs directed at the 'ordinary' Canadian marginalize the claims o f equality seeking groups.  Targeting Closely connected with individualized citizenship and the 'ordinary Canadian' is the concept o f targeting. In a 1994 publication the federal government stated that social programs "must be targeted at those who demonstrate a willingness and commitment to self-help."  117  This  means that social welfare funding w i l l be directed from income maintenance programs to programs aimed at helping people to overcome their shortcomings so that they can (re)enter the workforce and become self-sufficient 'ordinary' Canadians. Thus social problems are individualized, and those who suffer from them are "pathologize[d]" and targeted for state imposed "rehabilitation m  " See discussion above, at pp. 153-155. 6  " Human Resources Development Canada, Improving Social Security in Canada: A Discussion Paper 7  (Ottawa: Minister of Supply and Services Canada, 1994) at 25. " Brodie, supra note 77 at 74. 8  -175to make them normal."  119  While some income maintenance programs w i l l still be necessary, these  w i l l be targeted at groups who are identified as incapable of responding to rehabilitative treatment. Although targeting responds to difference, it does not respond to the structural disadvantages that are linked to difference. Rather, targeting is "used as a way to divide citizens between those who are deserving and undeserving, between the normal and those with special needs."  120  In the retirement income context targeting works in two directions. First,  increased  financial support is targeted at those who save for their own retirements (enhanced Tier m benefits). These people have shown a "willingness and commitment to self-help" . Second 121  erosion o f the universality o f Tier I benefits is a step i n the direction o f targeting this subsistence level income maintenance program to the pathologically needy. The proposed, but now scrapped, Seniors Benefit would have furthered this transformation:  The Seniors Benefit will help make the public pension system more affordable and sustainable. It will target help to those who need it most. Over the long term, better targeting will slow the rate of growth in public pensions that are paid out of the government's general revenues. 122  Both types o f targeting ignore the systemic inequalities that may put the rewards held out by Tier in out o f reach, or limit opportunities to access retirement income beyond that provided by Tier I.  Ibid.  m  l20  l2  Bakker, supra note 46 at 18.  '%?ranote 117.  '"Department of Finance, "The Seniors Benefit: Questions and Answers" in Budget Papers (Ottawa: Department of Finance, 1996).  -176-  The new senior citizenship Just as restructuring has changed the nature o f what it means to be a Canadian citizen, so the reformed (restructured) retirement income system has changed what it w i l l mean to be a senior citizen in Canada.  First, the enhanced tax subsidies that reward individuals for taking  responsibility for providing their own retirement incomes, and which provide benefits in direct correlation with economic advantage rather than need, w i l l widen the gap between the advantaged and disadvantaged on retirement. A s this gap is already gendered, retirement incomes of men and women w i l l become increasingly disparate, and the gender gap among the elderly poor w i l l likewise increase. Second, the restructured system deepens the gender inequalities linked to the provisions that reinforce the dependency of women on men. B y increasing the role o f the market and family in providing for retirement income, women's dependency on redistribution of economic resources within the family w i l l increase accordingly. This is so even though women's workforce participation is, o f necessity, increasing. The increase i n the proportion o f 'bad' jobs available to women  123  as well as the elimination o f many o f the 'good' public sector jobs that traditionally  employed women, mean that increased workforce participation w i l l not necessarily provide more or better opportunities for women to provide for their own retirement incomes. Downloading social programs onto the unpaid work o f women within the family w i l l also reduce workforce participation o f many women who might otherwise seek the economic independence o f employment. Finally, women seeking redress for enhanced structural disadvantages, whether through reform of Tiers I and II or otherwise are unlikely to be heard. The fact that 'sustainability' has been set as the restructured goal for these programs means that means that the government  And to men as well, as the labour market trend is toward increases in part time and short term jobs while full time, long tenn employment opportunities are on the decline. See supra note 79 and accompanying text. 123  -177accepts the status quo as the limit o f its commitment to the 'public' parts o f the retirement income system and implies that demands for gender equality enhancing improvements would be considered 'special' treatment.  Such demands are antithetical to the neoliberal state's interest in supporting  a market driven economy because they require the government to intervene in the market and to alter the way that it (the market) distributes economic resources. Neoliberal orthodoxy does not accept that 'ordinary' Canadians would ask their government for this kind o f 'special' treatment. Claims o f structural disadvantage and demands to reduce or eliminate these must therefore, by neoliberal reasoning, necessarily be made from outside the community o f 'ordinary' Canadians. The legitimacy o f these claims and demands is thus undermined and they w i l l not easily find their way to the top o f the state agenda. In short, while the economic aspects o f senior citizenship were gendered prior to recent reform (restructure) o f the retirement income system, reform (restructure) has deepened these gendered effects and legitimized the systemic inequalities from which these effects derive.  (c)  Restructuring - The 'No Choice' Alternative? One o f the features o f restructuring is the refrain that in the competitive global economy  Canada can no longer afford its costly social programs and has no choice but to reduce social spending.  124  The same type o f restructuring rhetoric is used by the government to deliver the  message that its reform (restructure) o f the retirement income system is essential:  Ensuring that Canada's seniors have an adequate retirement income is one of the most important social policy initiatives ever undertaken in this country. And while our system of public and private pensions is widely seen as one of the best in the world, its future sustainability is being threatened by major demographic and economic shifts that have occurred since these programs were set up almost  'Brodie, supra note 39 at 49-51 See also the discussion above, at pp. 157-159.  -17850 years ago.  The government has worked very hard over the past three-and-a-half years to bring about the changes necessary to preserve this system. ... I25  Monica Townson  126  taken on pension reform.  127  •  discredits the necessity for the position that the present government has The following is a summary o f the key points o f her critique:  in response to the concern expressed in the Finance Minister's 1995 budget that the public pension is at risk because o f rising costs, Townson notes that Canada spends a lower portion o f its Gross Domestic Product ( G D P ) on public pensions that any other O E C D  1 2 8  country. Townson refers to W o r l d Bank statistics for 1991 that show average cost o f public pensions i n O E C D countries at 9.2 per cent o f G D P while in Canada the cost in 1993 was  '"Government of Canada, supra note 103 at p. 2. See also Department of Finance, "Protecting Canada's Elderly" in Budget Speech 1995, http://www.fin.gc.ca/budget95/speeche/speech7e.httnl. Ms. Townson is a member of the Canada Pension Plan Advisory Board and a Board member of the Canadian Centre for Policy Alternatives; she was a member of the Pension Commission of Ontario from 1986 to 1997. As indicated in the bibliography to this thesis, she has written extensively about women and the Canadian retirement income system and the impact of pension reform on women. 126  l27  Townson, supra note 1; Monica Townson, Our Aging Society: Preserving Retirement Incomes into the  2F' Century (Ottawa: Canadian Centre for Policy Alternatives, 1995) and Social Contract for Seniors in Canada: Preparing for the 21 ' Century (Ottawa: Minister of Supply and Services, 1994). (See also Monica Townson, s  Partnership in Pensions: Who's Responsible Anyway, supra note 6; The Economic Situation of Canada's Seniors  (Ottawa: National Advisory Council on Aging, 1991); Tax Facts: What Every Woman Should Know (Ottawa: Canadian Advisory Council on the Status of Women, 1993); Women's Financial Futures: Mid-Life Prospects for a  Secure Retirement, supra note 6; Independent Means: A Canadian Woman's Guide to Pensions and a Secur  Financial Future (Toronto: Macmillan Canada, 1997); and "The Special Importance of the Canada Pension Plan for Women" in Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications,  (Ottawa: The Caledon Institute of Social Policy, 1996).) 0 E C D is an acronym for "Organization for Economic Cooperation and Development", founded in 1961. Canada is among the founding members, along with the United States, the United Kingdom, Ireland, Austria, Belgium, Denmark, France, Germany, Greece, Iceland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and Turkey. Japan, Australia, New Zealand and Finland joined during the 1960s-1970s. Since Mexico, the Czech Republic, Hungary, Poland and Korea joined between 1994 and 1996 their public pension costs would not have been included in the World Bank's 1991 average, (http://www.oecd.org/about/member-countries.html.) I28  -1795.3 per cent o f G D P with projections for 2030 at an 'unaffordable' rate o f 8 per cent. •  129  Townson argues that government projections of the long term costs of public pensions are overstated.  130  While O A S / G I S costs stated by the government are net of amounts that w i l l  be recouped in taxes and clawbacks, projections ignore the fact that the O A S is fully indexed to the Consumer Price Index (CPI) while income tax and clawbacks "are indexed only to the extent that the annual increases i n the C P I exceed 3 % . "  131  The result o f this  omission is overestimation of the likely cost of the public pension system when retired baby boomers put peak pressure on the system and a corresponding underestimation o f future tax revenues. Townson explains:  ..failure to fully index the tax system allows the government to take proportionately more of all forms of income each year. The phenomenon has been described by the National Council of Welfare as the biggest hidden tax grab in the system. People whose wages just keep pace with inflation, for example, will actually pay a higher percentage of their income in taxes, even though the purchasing power of their wages has not increased. This also means that, while the total cost of the OAS program will increase as the population ages, the government will take proportionately more of these benefits in taxes and clawbacks because of the different indexing regimes. 132  projections based on demographics without allowing for future economic growth overestimate the proportion o f Canada's future G D P that w i l l be required to support the public pension system.  133  important factors were omitted from the calculation o f dependency ratios relied on to  l29  Townson, supra note 1 at 2.  Ibid. at 13-14.  m  Ibid, at 13.  m  '"Townson, Social Contractsupra  note 127 at 16-17.  -180demonstrate that the aging population w i l l be too large an economic burden for the future workforce to support. The decreasing population o f young people w i l l undoubtedly lower the demand on public resources for education, health care and other youth related needs.  134  In addition, as Dennis Guest points out, the government's pessimistic projection o f one pensioner to every three workers makes  no mention of events that might alter this dire forecast, such as the implementation of a national day care program that could release more women into the full-time workforce; a policy of full employment to provide jobs for the thousands of unemployed and underemployed; the possibility that immigration policy could introduce younger workers into the economy; or even changes in the birthrate. 135  cost saving measures w i l l not reduce the needs o f Canada's seniors. There is no evidence that the government has analysed the future needs o f Canada's aging population, or studied the effects of the impact of pension reform on those needs. Absent such studies to provide a reference point, assertions o f unfairness and unsustainability lack meaning.  136  D a y and  Brodsky add to this point, noting that cost cutting in public pensions to the extent that economic needs o f seniors cannot be met may have the effect o f increasing government spending elsewhere:  ...it is an inadequate analysis of cost and benefits that chooses to focus only on the costs associated with social spending, without taking into account the costs of not engaging in adequate social spending. Social spending is not just an expense; it is a necessary social investment, the lack of which also has costs. 137  Ibid. at 11-14.  w  135  Dennis Guest, The Emergence ofSocial Security In Canada, 3 ed. (Vancouver, UBC Press, 1997) at 284-  136  Tbwnson, supra note 1 at 2.  rd  285.  Day and Brodsky, supra note 59 at 10. See also Joel Bakan, Just Words: Constitutional Rights and Social Wrongs (Toronto: University of Toronto Press, 1997) at 150-151. 137  -181Examples spring easily to the mind - increased health costs for seniors due to illness related to inadequate income to provide adequate food and shelter, increased nursing home admissions, increased social costs arising from pressures within families as a result o f bearing the responsibility and costs of providing direct care to aging family members. •  perhaps the most telling sign that the sustainability o f the pension system is not "threatened by major demographic and economic shifts" necessary to preserve [the] system"  139  138  that limit pension reform to "changes  is the dramatic increase i n public funding support for  R R S P contributions through tax subsidies.  140  This indicates first, that improvements to the  pension system are not unaffordable, and second, that the shape o f reform is not driven by economic necessity but by an ideological shift from "collective responsibility for and to seniors"  141  to "a market-based solution."  142  A s for the argument that this reform w i l l reduce  post retirement poverty, Townson points out that there is no requirement that R R S P funds be used for retirement,  143  contributions to R R S P s .  144  and that most Canadians cannot afford to make regular  Supra note 103.  m  Ibid.  m  140  Townson, supra note 1 at 52, and Townson, Social Contractand  Our Aging Society..., supra note 127  at 26 and 32-33 respectively. See also the discussions in Chapter One, above at pp. 44-46 and in Chapter Three, above at pp. 138-144. It is interesting to note that when the reformed RRSP contribution limit of $ 15,500 is phased in, those earning $86,000 a year or more will be eligible for an annual tax subsidy of $7,500 on their contributions alone, not including tax subsidies arising from non-taxation of interest earned on previous years' contributions (Our Aging Society at 32). Note also the observation of Bakker, supra note 46 at 5, that tax expenditures have played a greater role in increasing Canada's deficit than spending programs. M  'Townson, Social Contract..., ibid, at 19. lbid.  I42  Townson, Our Aging Society..., supra note 127 at 33. See also the discussion in Chapter Three above, at pp. 137-138. l43  'Townson, supra note 1 at 52.  -182In sum, the 'only choice' argument that resulted in the current evolution o f the retirement income system is flawed at best. The focus o f this argument on the costs o f Tiers I and II has taken a very narrow view in projecting future costs and in failing to consider cost consequences o f cost cutting. Further, the 'only choice' argument fails to consider the retirement income system as a whole in its costs analysis. A mythical division o f the system into public and private parts leaves the public costs o f Tier HI outside the framework for the discussion o f reform o f the 'public' pension system, and masks the fact that public monies expended on retirement income programs for the economically advantaged have increased significantly at the same time that programs that the less advantaged rely upon, the benefits o f which are also extended to the advantaged, are targeted as unaffordable and unsustainable. These flaws seriously call into question the validity of 'only choice' argument. A reformed (restructured) retirement income system that is based on flawed premises and has the effect o f compounding previously existing economic inequalities for women cannot be, and ought not to be accepted as, the only available choice.  -183C H A P T E R FIVE EQUALITY FOR WOMEN IN THE RETIREMENT INCOME SYSTEM: WHERE TO FROM HERE?  Social equality is an absence of major disparities in people's resources, political and social power, well-being and of exploitation and oppression. ... Social inequality is pervasive in Canadian society... Social inequality is not inevitable. That is why the struggle against it is worthwhile. Transformation of the causes of social inequality is possible: the question is how to achieve it.'  In spite o f the promise that it held, recent pension reform has not been responsive to gender (in)equality within the retirement income system.  In 1982 inequitable treatment o f women  "throughout much of the retirement income system" and insufficient income for unattached elderly 2  women were enumerated among "the most serious shortcomings" o f the retirement income system. 3  A s argued i n Chapter Four, during the past decade of reform o f the retirement income system, the "shortcomings" relating to gender inequality slipped from the reform agenda. T w o clear trends can be observed i n pension reform. First, a trend toward privatization is seen in the increased tax subsidization for retirement savings and employer sponsored pension plans. A s discussed earlier, the increased tax expenditures i n this area systemically advantage men who are over represented among those whose economic circumstances place them in a position where the opportunity to avail themselves o f these benefits is real, not merely theoretical. The second, and related, trend is away from universality of, and toward cutbacks to, state delivered pension programs.  Also,  'Joel Bakan, Just Words: Constitutional Rights and Social Wrongs (Toronto: University of Toronto Press, 1997) at 47. Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Wei fare Canada, 1982) (hereinafter "Green Paper") at 2. Hbid.  -184because o f their economic circumstances during their pre-retirement years, elderly women depend 4  on state delivered 'public' pension programs for their financial autonomy to a much greater extent than men. Paradoxically then, although gender (in)equality was ignored in the reform process, the effects o f reform are deeply gendered and compound the systemic economic inequalities for women that existed prior to reform.  There is no doubt that the reform process has been a setback for women's economic equality in retirement.  For those who continue to view the struggle for economic equality in  retirement as worthwhile, two key questions follow from this setback. First, is the government's assertion that the reformed (restructured) retirement income system is the only viable option in the current economic climate valid? A n d second, what alternatives advance, rather than impede, progress toward gender equality?  Chapter Four concluded with arguments that undermine the  'only viable option' rationale. The discussion in this chapter explores reform alternatives and discusses strategies for advancing beyond reform (restructuring) toward gender equality.  I  ALTERNATIVES TO THE REFORMED (RESTRUCTURED) RETIREMENT INCOME SYSTEM  One o f the themes developed in this thesis is that gendered inequality in retirement years is a continuation o f pre-retirement gendered inequalities within the home and the workforce. Accordingly, reform initiatives directed at pay equity, child care, maternity leave and other pre-  "Responsibility for unpaid reproductive work and consequent lower workforce participation than men in terms of both time and remuneration. See discussion in Chapter Two above, at pp. 62-72.  -185retirement gender equality issues are also retirement income reform initiatives. W h i l e discussion 5  of strategies directed at equality enhancing alternatives to recent pension reform could, and probably should, explore these kinds of initiatives, the focus of the discussion in this thesis w i l l be limited to reform possibilities within the retirement income system itself.  (a)  A Progressive R e f o r m A g e n d a Based as it is on assumptions about men and women and their economic relations, Canada's  retirement income system fails many women because it does not recognize differences among women and does not respect differences between men and women. A s a result women's access 6  7  to workforce related benefits is limited and some women are excluded entirely from spousal benefits. Economic equality for women (and other economically disadvantaged groups) during their senior years requires "[a] pension system that is more universal and less exclusionary."  8  Exclusion increases within the current retirement income system as one moves from the foundational Tier I level through Tier H and into Tier III. It makes sense, then, that "[constructing a progressive agenda for pension reform has to be built on strengthening the role of public pension plans" which "are far more effective at providing retirement incomes for everyone" 9  10  than the  A corollary to this point is that reform of the retirement income system by itself is unlikely to achieve complete elimination of economic inequality for elderly women. 5  6  For example, not all women are spouses residing within 'ideal' families.  7  For example, women's workforce participation is limited by their reproductive responsibilities.  "Claire F.L. Young, "Public Taxes, Privatizing Effects, and Gender Inequality" in Susan B. Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University of Toronto Press,  1997) 307 at 322. 'Monica Townson, Protecting Public Pensions: Myths vs. Reality (Ottawa: Canadian Centre for Policy Alternatives, 1997) at 53. See also Young, ibid.; Dennis Guest, The Emergence ofSocial Security In Canada, 3 ed. (Vancouver, UBC Press, 1997); John My les and Les Teichroew, "The Politics of Dualism in Canada" in John Myles and Jill Quadagno, eds., States, Labor Markets, and the Future of Old-Age Policy (Philadelphia: Temple University rd  -186heavily tax subsidized 'private' retirement income arrangements.  Strengthening the 'Public' Tiers The low level o f benefits that it delivers implicates O A S / G I S / S P A in the gendered poverty o f the elderly. Tier I ought to be designated as the site for achieving the first goal o f the retirement income system, providing sufficient income for the elderly "to live in dignity" regardless o f their 11  pre-retirement economic circumstances. Strengthening O A S / G I S / S P A is a matter o f increasing the level o f benefits for all pensioners, but particularly for the unattached.  Because it is a mandatory scheme that provides coverage to all workforce participants, not just to those who are fortunate enough to be R P P members or who can afford to contribute to R R S P s , C P P is well situated to be the primary site for achieving the second goal o f the retirement income system, continuation o f pre-retirement standard o f l i v i n g .  12  However, in its present form  the C P P is capable o f providing only minimal earnings replacement (25 per cent o f the A I W ) even for average income earners. While it provides a source o f retirement income for all workforce participants, C P P is an especially important part of the retirement income system for women, whose economic realities seriously limit their access to Tier III benefits.  13  Although its benefits  Press, 1991) 84; Maureen Donnelly, "The Disparate Impact of Pension Reforms on Women" (1993) 6 C.J.W.L. 419; and Barbara Austin, "Policies, Preferences and Perversions in the Tax-Assisted Retirement Savings System" (1996) 41 McGill L.J. 571. '"Townson, ibid. ' Green Paper, supra note 2 at 1.  1  Ibid.  n  l3  See discussion in Chapter Two above, at pp. 63-72.  -187are linked to workforce participation, the fact that the C P P makes some accommodation i n recognition o f the differences between women's and men's workforce realities marks the C P P as a promising site for improving gender equality i n the retirement income system.  Various  suggestions for improving gender equality in the C P P have been implemented i n the past and 14  others, including homemaker pensions, mandatory credit sharing and increasing the C P P income replacement limit, have been proposed.  15  Strengthening the C P P , then, would extend the  opportunity to continue pre-retirement lifestyles to those with limited access to Tier in and would enhance opportunities for implementing reform initiatives that recognize workforce participation differences between women and men and value women's reproductive work.  R e d u c e d Role for the ' P r i v a t e ' T i e r Logically, an increased role for the 'public' parts o f the retirement income system means  l4  See discussion in Chapter Two above, at pp. 64-68.  "Following the release of the Green Paper (supra note 2), various specific recommendations were made to improve employer sponsored pension plans and the CPP for women, including making employer sponsored pension plans mandatory, increasing the income replacement level of CPP to 50 per cent of the AIW, and implementing homemaker pensions through the CPP. (Susannah Worth Rowley, "Women, Pensions and Equality" (1986) 10 Dalhousie L.J. 283 at 311-323 and 339-340.) As Rowley notes, some of the suggested reforms were discussed in the Green Paper itself.) Although the debate over the pros and cons of these specific proposals will undoubtedly be revived when gender equality is again put on the pension reform agenda, for the purposes of this thesis it is unnecessary to engage in a discussion of the details of these proposals. (The following discuss these specific proposals to varying degrees: Yvonne C. Chenier, "Pension Reform to Benefit Women: How Far Have We Come?" in E. Diane Pask, Kathleen E. Mahoney and Catherine Brown, eds., Women, The Law and the Economy (Toronto: Butterworths, 1985) at 121; Government of Canada, Better Pensions for Canadians: Focus on Women (Ottawa:  Department of Supply and Services, 1982); Dennis Guest, supra note 9; Jane Lewis, "Towards a Framework for Analysing the Position of Women under Income Security Policies" in E. Diane Pask, Kathleen E. Mahoney and Catherine Brown, eds., Women, the Law and the Economy (Toronto: Butterworths, 1984) 145; National Council of Welfare (Canada), Elderly Women in Canada: Their Economic Plight (Ottawa: Supply and Services, 1982); National Council of Welfare (Canada), Women and Poverty Revisited: A Report (Ottawa: Supply and Services, 1990); E. Diane Pask, "Pensions and the Elderly: Selected Legislative Issues Concerning Income Splitting" in M.E. Hughes and E.D. Pask, eds., National Themes in Family Law (Toronto, Carswell, 1988) 117; Andrea J. Seale, "Can the Canada Pension Plan Survive the Charter? Section 15(1) and Sex (In)Equality" (1985) 10 Queen's L.J. 441; Caledon Institute of Social Policy, Round Table on Canada Pension Plan Reform: Gender Implications, (Ottawa: The Caledon Institute  of Social Policy, 1996); and the various Townson sources, cited supra Chapter Four note 127.  -188a reduced role for the 'private' part. Since the dual objectives o f a retirement income system  16  could be achieved for most Canadians through improvements to Tiers I and II, perhaps Tier HI should not even play a role.  If, for example, O A S benefits were increased to 20 per cent o f the  A I W and C P P benefits were increased to 50 per cent o f the A I W , persons earning up to the A I W would not need Tier HI benefits. Increasing the C P P contributions and benefits to one and one-half times the A I W would further reduce the number o f persons who would rely on Tier III to ensure lifestyle maintenance on retirement. This raises the question o f the level to which public monies should be used to ensure a consistent lifestyle. Surely the current level o f public support that the economically advantaged can access through Tier III, up to two and a half times the A I W , is too much. The point here is that a prerequisite to defining the specifics o f an equality enhancing reform agenda is "a comprehensive and coherent retirement incomes strategy."  17  This strategy  should be based on the needs o f Canada's aging population and how they can best be met, not on how the risk of ensuring adequate income can be shifted onto individuals rather than the collective. Cost is an important factor, o f course, but cost o f the system as a whole must be considered, not just its 'public' parts, and costs o f not meeting the needs must enter into the calculation as well.  Suggestions for Setting an Equality Enhancing Agenda While is relatively easy to move from the understanding o f the bases o f gender inequality in the retirement income system developed in this thesis to setting a general progressive reform agenda, the specifics are more elusive. Recognizing that it is beyond the scope o f this thesis to  /e. to provide a level of income that allows all elderly persons to live in dignity and to prevent a drastic change in lifestyle on retirement. (See supra notes 11 and 12 and accompanying text.) 16  "Townson, supra note 9 at 53.  -189suggest a comprehensive gender equality enhancing strategy for pension reform, the observations 18  and suggestions that follow are informed by the exploration in this thesis o f the issue o f gendered poverty among the elderly: •  an expanded role for Tiers I and II is clearly desirable in meeting both o f the goals o f a retirement income system, up to a set income replacement level (for example 70 per cent o f the A I W ) ; replacement o f universality with targeting is regressive and undesirable.  A s well it  diminishes the income replacement role o f the system as a whole; the level o f income from Tier I should be increased to the poverty line. T o achieve this the benefits for unattached seniors w i l l have to be increased by a greater amount than for senior couples; •  the level o f Tier II benefits should stack onto Tier I to achieve the set income replacement level;  •  spousal status should be eliminated as an eligibility criterion for S P A and C P P survivor benefits. Rather, the needs that these provisions were intended to address should be studied and a non discriminatory mechanism for meeting those needs should be implemented;  •  19  improvements to the C P P that accommodate the access disadvantages that women suffer  ' And beyond the expertise of its writer as well. Indeed it would be arrogant to suggest that this thesis could, by itself, serve as a guide those who are out there in the real world actively engaged in the struggle for economic equality for women. To lift the words used by Didi Herman in a different, though related, context: "Each analysis, like pieces of an analytic jigsaw, contributes a partial and necessary explanatory element. At the same time, no one approach by itself completes the picture." ("The Good, the Bad and the Smugly infra note 36 at 604.) 8  Specifically, SPA should be available based on need, not spousal status, and a gender equal CPP would set eligibility criteria for the survivor benefit which reflect the reality that there are numerous types of dependency relationships in addition to spousal relationships. See, for example, Deborah Jones, "What About No-Sex Partners", The Globe and Mail (21 March 1998) D3 where an interviewee stated that she would like her RPP survivor benefit to go to her lower income earning sister who depends on the interviewee for support. 19  -190because o f unpaid reproductive work and underpaid work in the labour force should be implemented; •  public funding o f Tier JJI should be reduced, i f not eliminated altogether, and redirected toward improving Tiers I and II. If public funding (tax subsidies) for Tier HI is continued, though reduced, it should be i n the form o f tax credits, not tax deductions.  Having set a tentative reform agenda, the question o f how to move toward achieving these goals remains.  II  BEYOND RESTRUCTURING - STRATEGIES FOR EQUALITY SEEKING REFORM  This section o f the thesis explores the possible courses o f action for moving from recent regressive reform to equality. T w o general types o f responses are available - legal action and political action.  Although these are discussed below under separate headings, it would be  misleading to suggest that a dichotomous relationship exists between these responses. Legal action is political, particularly when it is used to advance an agenda that challenges the status quo.  (a)  Legal Action Gender equality is constitutionally protected by the Charter.  20  Courts may scrutinize  ^Canadian Charter ofRights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c l 1. Subsection 15(1) of the Charter states: 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.  -191actions o f the provincial or federal governments, and i f the impugned action does not comply with the Charter the court w i l l conclude that the government's action was illegal unless the government convinces the court that the action was justifiable in a free and democratic society. The retirement 21  income system is a statutory creation and is therefore subject to the Charter.  In addition, Canada  is a signatory to several international human rights agreements in which commitments are made to gender equality and arguably is legally bound to act according to these commitments. Both o f these offer possibilities for making progress toward gender equality in the retirement income system.  The Charter One o f the ironies o f the regressive turn that pension reform has taken when considered from a feminist perspective is that it occurred in the Charter era, when one would have expected gender equality issues to remain prominent on any reform agenda. The failure o f the Charter  to  live up to the expectation that it would eliminate social inequality has provided legal academics with a rich topic for scholarly study. This discussion is informed by their work.  The effective date o f section 15 o f the Charter  was delayed for three years to give  governments time to bring legislation in line with the equality guarantees. In an article that predated the effective date, Andrea Seale concluded that the imminent coming into force o f section 15, on A p r i l 17, 1985, required that gender inequalities i n the retirement income system be  Ibid., s.l. In addition, section 33 of the Charter allows governments to immunize specific actions from Charter scrutiny by declaring in legislation that the legislative provisions that govern those actions may operate notwithstanding the equality provisions in the Charter. This 'notwithstanding' provision expires after five years but can be re-enacted. 2>  -192eliminated immediately. In 1986 Susannah Worth Rowley examined Canada's retirement income 22  system and concluded that even i f the proposals for reform that were then being debated i n the aftermath o f the Green Paper (primarily reform to Tier III) were implemented the system could not withstand a section 15 analysis.  23  Rowley suggested that legal action using strategies employed in  the Supreme Court o f the United States i n Brown v. Board of Education , namely the use o f 24  sociological evidence to give a contextualized meaning to 'equality', might be effective in implementing equality enhancing improvements to the retirement income system.  25  Two years after the equality provision o f the Charter came into force, Judy Fudge noted a trend in judicial decision making that led her to conclude that "feminists ought to approach the arena o f Charter litigation with great trepidation, i f at a l l . "  26  Her analysis o f early Charter  decisions connects familial ideology to women's subordinate status and consequent inequality. Charter decisions, she argued, show familial ideology at work in the judicial interpretation o f "the abstract and general language" o f section 15 and other Charter rights. While feminists may interpret these rights in a progressive way, "they are equally amenable to formal and narrow  22  Seale, supra note 15.  23  Rowley, supra note 15. (1954) 347 U.S. 483.  24  It is interesting that advancing a contextualized meaning of 'equality' still dominates discussions about the use of the Charter in equality seeking litigation. Day and Brodsky emphasize the importance of expanding the concept of 'equality' to include economic equality. (Shelagh Day and Gwen Brodsky, Women and the Equality Deficit: Restructuring Canada's Social Programs (Ottawa: Research Directorate, Status of Women Canada, 1998). See the discussion in this chapter, below at pp. 203-207. 25  Judy Fudge, "The Public/Private Distinction: The Possibilities of and Limits to the Use of Charter Litigation to Further Feminist Struggles" (1987) 25 Osgoode Hall Law Journal 485 at 554. (See also Judy Fudge, "The Privatization of the Costs of Social Reproduction: Some Recent Charter Cases" (1989) 3 Canadian Journal of Women and the Law" 246.) 26  -193interpretations which are antithetical to feminist struggles."  27  Fudge was particularly critical o f an  early decision o f the Supreme Court o f Canada which shielded 'private' (non governmental) activity from Charter scrutiny.  28  Establishing a public/private divide for Charter litigation  effectively limits the scope o f gender equality to formal equality, argues Fudge, because furthering feminist struggles through litigation requires a contextualized approach to gender equality, and the context o f women's inequality is contained within the private sphere o f the market, where women work for l o w wages, and the home, where women are subordinate to their husbands and work for no wages.  29  In her early Charter commentaries, Fudge did not go so far as to exclude Charter litigation altogether from the struggle for women's equality, but relegated it to a limited role.  ...it is crucial to understand that Charter litigation, although at times a necessary arena of struggle, is not itself sufficient (even when successful) to achieve substantive equality. However, it is equally important to understand that liberal rights and freedoms are contradictory and insufficient rather than illusory, and the struggle for social transformation may often involve pushing bourgeois forms of legality to their limit. Feminists must be vigilant against allowing litigation to dominate the political and social struggle to obtain substantive equality for women, because the tendency of litigation strategies is to transform politics into a series of narrow issues packaged as private and individual cases. ...it is important to distinguish between litigation as a tactic which is part of an overall political strategy for social justice, and litigation as the strategy for attaining social justice. (emphasis in original) 30  27  "The Public/Private Distinction...", ibid.  28  Retail, Wholesale and Department Store Union, Local 580 v. Dolphin Delivery, [1986] 2 S.C.R. 577.  Fudge's argument remains valid today, even though some gains have been made in extending the reach (albeit indirectly) of the Charter to 'private' activities that are subject to or governed by legislation. For example in Vriend v. Alberta (1998), 156 D.L.R. (4 ) 385 the Supreme Court of Canada ruled that Alberta human rights legislation contravened the Charter because it did not include sexual orientation as an enumerated ground of discrimination. 29  ,h  30  Fudge, "The Public/Private Distinction  supra note 26 at 548.  -194The theme o f the ineffectiveness o f the Charter to transform causes o f social inequality appears again i n Fudge's later work. In a co-authored article Fudge and Harry Glasbeek argue that enumerated grounds o f discrimination such as those found i n the section 15 o f the Charter and i n human rights legislation cannot protect people from the "economic subordination which may be expressed through these legally recognized differences."  31  The enumerated distinctions, they argue,  are promoted by the liberal state to "deny, or at the very least obfuscate, the possibility o f interconnectedness o f the sources of their various oppressions." "always generates economic inequality."  33  32  The source is capitalism, which  Fudge and Glasbeek advocate that rights groups  abandon legal action which, by its very structure, isolates rights claims, and acknowledge the common sources o f their inequality, patriarchy and the capitalist class structure. Rights groups, they argue, should shift their focus from the symptoms o f their oppression to its underlying common cause. transformation"  34  They should turn their fragmented  rights claims into " a politics o f  directed at "confront[ing] the larger structural changes within which concrete  political struggles are taking place."  35  Charter litigation requires and reinforces fragmentation and  has no role to play i n the struggle.  D i d i Herman takes issue with the Fudge and Glasbeek denial o f the Charter as a tool that  Judy Fudge and Harry Glasbeek, "The Politics of Rights: A Politics with Little Class" (1992) 1 Social and Legal Studies 45 at 61. 3l  "Ibid. "Ibid, at 63. Ibid. at 66.  i4  Ibid.  35  -195may be useful in furthering women's equality in Canada.  36  Herman points to the experiences o f  lesbian and gay movements i n using the Charter as indicators that Charter litigation can be a useful strategy in moving toward equality, though she acknowledges that Charter litigation on its own is unlikely to effect equality.  37  Although formal equality is less than substantive equality, she argues  that the achievement of formal equality through Charter litigation may be a strategically necessary first step along the way. In addition, Charter litigation and the media attention it gets can stimulate people to think about the issues that are being litigated. A s one gay litigant, whose litigation was ultimately unsuccessful in terms o f the result, put it in an interview with Herman, "... no one questions, as they might have before the Charter, whether there is such a thing as 'rights'... 'rights' gives you a language."  38  Herman disagrees with Fudge and Glasbeek that rights activists are i n  danger of naive over-reliance on the Charter as a mechanism for redressing social inequalities. A n excerpt from an interview with activist Gwen Brodsky summarizes what appears to be Herman's view o f Charter litigation in advancing an equality agenda:  Where the Charter is concerned, I prefer to think of myself as a thoughtful realist. I believe that the Charter can have a positive role in struggles for progressive social change, but that in order for its potential to be achieved Charter litigation needs to be: rooted in community based politics, combined with other political struggles, and regarded critically by its users. The Charter has the power to effect change, whether that change be regressive or progressive; that is why I have devoted energy to arguing in favour of progressive interpretations, public funding for access to Charter rights by members of disadvantaged groups, and increased accountability on the part of governments in Charter cases. I also  Didi Herman, "Beyond the Rights Debate" (1992) 2 Social and Legal Studies 25. (See also Didi Herman, "The Good, the Bad, and the Smugly: Perspectives on the Canadian Charter of Rights and Freedoms" (1994) 14 Oxford Journal of Legal Studies 589.) 36  In the sense that it recognizes a limited role for Charter litigation in the pursuit of (in)equality claims, Herman's critique of Fudge and Glasbeek is reminiscent of Fudge's earlier commentary on the Charter. Fudge's later revised her view, arguing that Charter litigation should be abandoned as a strategy for effecting equality. 37  38  Herman, "Beyond the Rights Debate...", supra note 36 at 34.  -196believe in the value of allowing the voices of marginalised groups to be heard.  39  Herman acknowledges that she subscribes to a pragmatic approach to use o f the Charter as a tool for forwarding an equality agenda, but the commentary o f those who either debunk or promote the Charter inform her pragmatism.  40  She includes Fudge and Glasbeek among the  "economistic critics" who debunk the Charter and, while she does not subscribe to their view that 41  the Charter is necessarily an ineffective tool for social transformation, she acknowledges that "an analysis o f capital and class is as relevant as it ever was" to the struggle for social equality. Herman accepts the Fudge and Glasbeek observation that the Charter has not been successful i n undermining the existing economic order, but not their conclusion "that such litigation is 'bad'."  42  However, she warns those who consider using the Charter to be wary that "the courts.. .exist within a state structure and operate with all o f the constraints and functions appropriate to their role namely facilitating the reproduction, rather than the subversion, o f the status quo."  43  Joel Bakan offers a comprehensive and accessible critique o f the role o f Charter litigation in forwarding a social equality agenda i n his book Just Words. H i s message is that the Charter 44  is just a document (just words) and, while there is potential for those words to be interpreted as serving the end o f social justice (to be words o f justice), meaning does not come from the  "Ibid, at 35.  ""Herman, "The Good, the Bad and the Smugly Ibid, at 601.  41  Ibid. at 603.  42  Ibid.  4,  Supra note 1.  44  supra note 36.  -197document but from the context in which the Charter operates.  All political institutions, including the Charter and rights, are necessarily constrained in their operation by the wider social system that they are established to govern. That is why it is necessary to be sceptical of both Charter optimism and pessimism when they are based on allegedly essential features of the Charter or rights. The emancipatory and egalitarian potential of the Charter depends on the social and historical circumstances surrounding its use. (emphasis added) 43  Bakan ascribes progressive meanings to 'equality', 'freedom' and 'democracy'  46  (the Charter's  'just' words) and these progressive meanings inform his critique o f Charter effectiveness in advancing the struggle for social equality.  He observes that the ideals represented in these  progressively defined words are "only partly realized in Canada today" , and the Charter cannot 47  "protect and advance a progressive conception o f social justice" that is not yet realized within the 48  broader social context.  The Charter ... cannot compensate for the systemic undermining of ideals of social justice by the routine operation of society's structures and institutions.... Constitutional rights strategies share with all other forms of political action some potential and some dangers, yet they cannot substantially redress social injustice because it is produced by a multitude of factors beyond their reach. 49  Bakan is sceptical o f the effectiveness o f section 15 o f the Charter as a tool to be deployed in the struggle for social justice, and in particular as a tool for redressing "the economic dimensions  Ibid. at 9.  45  Ibid. at 9-10.  46  Ibid, at 11.  41  Ibid.  4%  lbid.  m  -198o f social inequality" , which "are rooted in intersecting relations o f class, gender and race, not i n 50  particular acts o f government or private actors."  51  Like Herman, Bakan questions the trust placed  in the judiciary as interpreters o f the Charter's 'just' words and o f the scope o f the Charter's application. He points out, though, that the judiciary is not the only constraint on the Charter's potential for progressing toward economic equality:  ...even if anti-poverty advocates succeed in persuading courts to accept their claims to positive rights, which is at least a theoretical possibility, this would have little effect on the nature and scope of poverty in Canada. Though equality rights might be capable of accommodating claims for more and better social welfare programs, it cannot recognize, nor be used to remedy, the conditions that cause poverty and make those programs necessary.  The Charter contemplates only particular instances of state regulation and spending; at best it can require the state to provide remedies for poverty's ill effects. In the meantime, however, national and international political economic developments, far beyond the scope of the Charter, increasingly intensify social inequality and injustice. 52  Bakan leaves the ultimate decision o f whether or not to employ Charter litigation in the struggle for economic equality to those who are closer to front line activism:  ...academic expertise does not qualify people such as me to tell actors in social movements what they should or should not be doing.... I am content to suggest that people who work with rights should add some social theory to their strategic  Ibid. at 51. The basis for Bakan's scepticism is his observation that judges' interpretations of section 15 have been constrained by "the traditional conception of rights as protecting individuals from public (state) interference in their private affairs but not requiring positive assistance by the state". He argues that even if courts adopted a progressive conceptualization of equality and imposed positive obligations on the state, the atomistic nature of rights (that rights belong to discrete entities in relation to corresponding duties or obligations of another) would preclude rights claims from "getfting] at the causes of inequality and other social ills; they deal only with symptoms, leaving underlying social structures untouched." At this point it appropriate to urge those who are interested in use of the Charter's in the struggle for social justice to read Bakan's book in its entirety. While hopefully the parts of his argument that relate to the themes of this thesis are accurately presented, the points extracted and summarized in this thesis do not capture the totality of Bakan's message. S0  "Ibid.  Ibid. at 54-55.  l  -199thinking, if they have not already done so. They are better positioned than I am to determine where this will lead them. 53  Charter litigation, based on the equality rights provisions i n section 15, might be considered as a possible strategy for advancing the progressive agenda for reform o f the retirement income system.  The flaws i n the system from a gender equality perspective are the spousal  provisions, the almost total lack o f recognition for reproductive work, and the skewed distribution of benefits that favours the economically advantaged. Taking into consideration the critiques by Fudge and Glasbeek, Herman, and Bakan, Charter litigation aimed at redressing the gender inequality aspects o f these flaws is problematic. The spousal provisions provide needed financial support for the women who have access to them, and their financial security might be diminished by a claim that these provisions unfairly discriminate against women. The source o f the gender inequality i n the spousal provisions is the power relations between women and men that are part of the underlying social structure on which the Charter sits, and which all four o f the scholars referred to above agree is beyond the power o f the Charter to transform. Because some good is achieved by the spousal provisions, this is exactly the kind o f situation where the power o f the Charter, limited as it is to "deal[ing] only with discrete symptoms, leaving underlying social structures untouched"  54  could have detrimental effects.  Further, the three identified flaws  55  (spousal provisions, the lack o f recognition for reproductive work and the skewed distribution o f benefits) reflect "economic dimensions o f social inequality" , which all four scholars agree is the 56  "Ibid, at 144. "Ibid. 55  See discussion below at pp. 200-202.  56  Bakan, supra note 1 at 51. See also supra note 50 and accompanying text.  -200type o f inequality that Charter litigation has been least effective in redressing.  W h i l e legal scholars may agree that Charter litigation is unlikely to be effective in redressing the types o f gender inequalities that permeate the retirement income system, activists are unlikely to rule out Charter challenge on the basis of academic argument. A s noted by Herman, activists are informed b y academic analyses o f the Charter's limitations but tend to take a pragmatic approach to Charter litigation, as well as any other potentially effective course o f legal action. The pragmatic question is - can this particular equality challenge negotiate around the Charter's recognized limitations i n a manner that advances, rather than impedes, the goals o f the (progressive) reform agenda?  37  The answer to this question w i l l necessarily depend on the specific  circumstances o f a particular case. The activist works closer to the front line and is therefore well positioned to judge whether a particular set of facts might make litigation worthwhile despite well founded scepticism. Some of the factors to be considered in making that judgment in the context o f advancing progressive reform o f the retirement income system are discussed below.  Spousal Provisions O f the items on the 'equality enhancing agenda', the spousal provisions are the most obvious choice for a challenge under section 15 o f the Charter In fact, two o f these provisions,  "Implicit in this question is recognition that not all section 15 challenges brought by women advance a progressive equality enhancing agenda. Symes v. Canada, [1993] S.C.R. 695 is a case in point. Beth Symes, a lawyer, argued that Revenue Canada's disallowance of her child care expenses as a business expense amounted to discrimination based on gender. Had Symes been successful, a benefit (tax subsidy) would have been made available to economically advantaged professional and business women, but not to the majority of women in the workforce who are employees for low wages and therefore cannot take a business deduction even though they need, and spend a greater portion of their incomes on, child care. (See Claire F.L. Young, "Child Care - A Taxing Issue" (1994) 39 McGill Law Journal 540 and Audrey Macklin, "Symes v. M.N.R.: Where Sex Meets Class" (1992) 5 Canadian Journal of Women and the Law 498.)  -201S P A and R P P survivor benefits, have been the subject o f recent Charter litigation.  58  However, the  approach taken i n Egan and Rosenberg would not satisfy the goal for challenging spousal provisions set in the 'equality enhancing agenda' - to eliminate spousal status as a criterion for these benefits. One o f the Charter litigation limitations that a plaintiff bringing such a challenge w i l l have to negotiate around is that it is only discrimination based on "an enumerated or analogous ground" that offends section 15 o f the Charter. A plaintiff who is a woman (but not a spouse) would have difficulty basing her challenge on the enumerated ground o f "sex" (gender) because women are over-represented among those who benefit from the spousal provisions. Spousal status is not an enumerated ground i n section 15 and asserting that it is an analogous ground adds an additional hurdle to the litigation. It might be possible, however, for an impoverished unattached elderly woman to challenge the spousal provisions. In Dartmouth/Halifax County Regional Housing Authority v. Sparks the N o v a Scotia Court o f Appeal determined that public housing 59  tenants are an analogous group to whom section 15 o f the Charter applies because they share the characteristic o f poverty.  ...As a general proposition, persons who qualify for public housing are the economically disadvantaged and are so disadvantaged because of their age and correspondingly low incomes (seniors) or families with low incomes, a majority of whom are disadvantaged because they are single female parents on social assistance, many of whom are black. The public housing tenants group as a whole is historically disadvantaged as a result of the combined effect of several personal characteristics listed in section 15(1). As a result, they are a group analogous to those groups specifically referred to by the characteristics set out in s. 15(1) of the Charter being characteristics that are most commonly the subject  58  See the discussion of Egan v. Canada, [1995] 2 S.C.R. 513 and Rosenberg v. Canada (Attorney General)  (1995), 25 O.R. (3d) 612 (G.D.), reversed [1998] O.J. No. 1627 (Docket No. C22807, April 23, 1998) above, in Chapter Three at pp. 119-123 and 129-131. Both challenges were made on the ground of sexual orientation. Neither challenged the underlying power structure; their claims were that their relationships were 'like' heterosexual spousal relationships. (1993) D.L.R. (4th) 224 (N.S.C.A.).  -202of discrimination.'  Based on the reasoning employed by the N o v a Scotia Court o f Appeal and the compelling statistical evidence o f their poverty, unattached elderly women would qualify as a "group analogous to those groups specifically referred to by the characteristics set out in s. 15(1)" . A member o f 61  this group could advance a claim that the spousal provisions (or any one o f them) discriminate against her. Even i f successful, though, the remedy available might impede, rather than advance, the reform goal. Obviously the plaintiff would seek a remedy that extends the benefit to those who are not spouses, such as modification o f the qualifying criteria in a way that, disregarding the assumptions within familial ideology, reflects the need that the provision was intended to meet. However, a remedy that requires the government to increase its spending on a particular program is rarely granted.  62  If the court w i l l not grant the remedy that the plaintiff seeks, it must grant some  other remedy because it cannot allow an unconstitutional provision to continue to operate. A probable outcome is that the unconstitutional provision would be struck, and the benefit(s) would no longer be available even to the women (mostly) who once qualified.  Finally, even in  circumstances that satisfy the 'pragmatic question' posed above, there are practical obstacles to  Ibid. at 234. See also R. v. Rehberg (1994) 111 D.L.R. (4 ) 336 (N.S.S.C.) which applied the Sparks decision and determined, at p. 364, that "poverty is analogous to the listed grounds in s. 15(1)." See also the discussion of this issue in Day and Brodsky, supra note 25 at 123-126. m  6]  th  Dartmouth/Halifax County Regional Housing Authority v. Sparks, ibid.  In her analysis of the failure of Egan's section 15 claim in relation to SPA and the success of Rosenberg's section 15 claim in relation to RPP survivor benefits, Claire Young suggests that the cost to the public purse could explain the different outcomes. Egan was claiming a right to SPA benefits; Rosenberg was claiming therightnot to be excluded from access to her partner's RPP survivor benefits that the employer would extend to gay and lesbian couples but the ITA would not. (See Claire F.L. Young, "Spousal Status, Pension Benefits and Tax: Rosenberg v. Canada", (1998) 6 Canadian Employment and Labour Law Journal 1.) Note however, that the Supreme Court of Canada recently rendered a decision that has the effect of requiring government spending to comply with section 15 62  of the Charter. In Elridge v. British Columbia (Attorney General) (1995), 151 D.L.R. (4 ) 577 the British Columbia ,h  government's withdrawal of medical interpreting service to deaf people was found to discriminate on the ground of disability.  -203litigating this issue as well.  Impoverished elderly women are unlikely to have the necessary  resources (economic, emotional, physical, social, political and otherwise) to initiate this type o f litigation.  Notwithstanding the considerations discussed in the previous paragraph, however, it goes against the pragmatic grain to take a completely pessimistic stance on Charter litigation. The right set o f circumstances for a successful challenge to the spousal provisions might come along, or 63  perhaps a long term strategy might include risking a failed Charter challenge for the political value o f the publicity that the litigation, regardless o f outcome, might generate.  64  Economic equality The other items on the 'equality enhancing agenda' reflect structural flaws in the retirement income system and are difficult to fit into a litigation framework. There are no specific legislative provisions to impugn, rather the structure o f the system is the source o f the inequality. However, feminist activists Gwen Brodsky and Shelagh Day argue that Charter litigation directed at  While it is impossible to define exactly what these circumstances might be, at a rninimuin they would have to offer a high degree of probability of overcoming the obstacles described in above. To return to the example of the impoverished unattached woman who might challenge the SPA provisions, the right set of circumstances might be present if she had the necessary non financial resources to take on lengthy litigation, access to the necessary legal resources or financial backing to make that access possible, and Charter remedies were further extended (possibly in a future decision that applies Elridge (supra note 62), or perhaps through litigation that expands the Charter notion of equality to include economic equality as discussed below). Political activism regarding the issue to be litigated may create (or add to the creation of) the right circumstances. As Day and Brodsky note: "The most important legal victories of the past decade - in cases such as Brooks, Janzen and Morgentaler - were achieved because they were preceded and accompanied by an enormous swell of activism, which created favourable conditions for litigation." (Supra note 25 at 136.) 63  Day and Brodsky, ibid, at 137, note that a Charter challenge "can be a way of gaining legitimacy and public sympathy for an issue to which elected officials have not been responsive." They discuss M.N.R. v. Thibideau, [ 1995] 1 C.T.C. 382 (S.C.C.) as an example of unsuccessful Charter litigation that had a positive political result. See also supra note 38 and accompanying text. 64  -204achieving economic equality for women is not necessarily futile.  65  They see Charter litigation as  a (but not the) strategy for expanding the meaning o f equality to include economic equality for women. This expanded meaning is critical to any progressive notion o f gender equality:  It is crucial that equality rights be expanded to recognize that, in relation to women's equality aspirations, economic autonomy interests are no less fundamental than liberty interests, and that for women, these interests are actually inseparable from one another. 66  They point out that "Canadian courts have not been asked to consider a Charter case that squarely raises the issue o f women's right to an adequate standard o f l i v i n g "  67  and direct their energies at  analysing how the Charter can be effectively used i n litigating this issue rather than why it might not be effective. In this pursuit they look to Canada's international commitments.  68  W h i l e the use  o f international human rights agreements in advancing economic equality claims is discussed in detail i n the next section, the point to be made here is that such an expanded notion o f equality would significantly expand the scope o f Charter litigation as a tool for redressing the gendered and classed (as well as raced) economic inequalities in the retirement income system. Although it still might be that discrete parts of the system would be easier to challenge than the system's structure, constitutional protection for economic equality would open up the possibility o f impugning as unconstitutional the far too low level o f Tier I benefits for elderly unattached women. Given the judicial acknowledgement that poverty is an analogous ground under section 15 o f the Charter and  Supra note 25.  65  Ibid. at 142.  66  67  'Ibid, at 116.  See also Martha Jackman, "Women and the Canada Health and Social Transfer: Ensuring Gender Equality in Federal Welfare Reform" (1995) 8 C.J.W.L. 371. 68  -205that poverty is linked to senior citizenship, perhaps it would be an effective strategy to make the 69  retirement income system the site where the argument that equality includes economic equality is made.  International Human Rights Agreements Day and Brodsky argue that international human rights agreements to which Canada is committed require that a contextualized and progressive interpretation be ascribed to the concept o f 'equality', and that the Charter must be interpreted i n a manner consistent with Canada's international human rights obligations because "they 'reflect the values and principles that underlie the Charter itself. " '  7 0  Section 15 o f the Charter must therefore be interpreted within the framework  o f the following propositions that are central to the International Covenant on Economic, Social and Cultural Rights ( " I C E S C R " ) , the Convention on the Elimination of All Forms of Discrimination Against Women ( " C E D A W " ) , as well as the United Nations Report on the Fourth World Conference on Women held in Beijing in 1995:  • Equality has a bottom,... equality includes, as part of its meaning, the social and economic rights of the ICESCR, including an adequate standard of living; • Equality requires the elimination of economic disparities between men and women as groups; • Governments have positive obligations to create conditions of social and economic equality for women; • Those obligations do not permit governments to delay in taking the appropriate measures to meet them or to move backwards; and • Economic policies violate women's rights to equality if they permit or foster poverty among women, or if they perpetuate, and do not repair, the status quo of  'Supra notes 59-61 and accompanying text. 'Day and Brodsky, supra note 25 at 119, quoting Dickson J. in R.vXeegstra, [1990] 3 S.C.R. 697.  -206women's economic inequality.  71  Day and Brodsky propose that 'equality' so contextualized can be the basis for a Charter challenge to federal legislation that has had the effect o f undermining social programs on which women depend for their economic security.  72  The restructuring o f the retirement income system has the same effect - the privatizing effect o f expanding Tier III benefits while containing the benefits o f Tiers I and II undermines women's opportunities for economic security during their senior years. Although grounds for scepticism based on the Charter's record o f setting issues o f social and economic policy outside its scope are undeniable,  73  it is difficult to resist the appeal o f a an original and creative legal  argument that might forward the struggle for women's economic equality. This is especially so in the present political climate which has proven to be ripe for regressive measures respecting  '"Ibid, at 112.  Day and Brodsky's discussion of reform strategies is in relation to the repeal of the Canada Assistance Plan through which the federal government could impose national uniformity on provincial social programs by making federal funding conditional on meeting national standards. In 1995 the Budget Implementation Act altered this thirty year old arrangement by implementing block funding to provinces for social assistance, health and post-secondary education spending. Conditions attached to social assistance spending were removed, as well as the separation of funds for each of the three areas of spending. This reform (restructuring) measure is regarded by feminists as a regressive step for women who rely on social assistance programs and who are at risk that the provinces (any or all of them) will divert the federal funds that formerly were directed to these programs to health or post-secondary education and that regressive standards for social assistance programs will be imposed in some or all provinces. The equality issues raised by the repeal of the Canada Assistance Plan are similar to those raised by the restructure of the retirement income system. Both restructuring measures compound the pre-existing economic inequality of women. 72  73  This issue is discussed by Day and Brodsky, supra note 25 at 83-99, and includes analysis of relevant  Charter decisions including M.N.R. v. Thibideau, supra note 64, and Symes v. Canada, supra note 57. These two  decisions involved challenges of gender inequality by women to provisions of the Income Tax Act. Though neither was successful, the ultimate issue was subsequently resolved in Thibideau's favour by legislation, and Symes' success would have been problematic for women's equality in general. Both decisions are instructive for future litigation on the issue of economic equality for women, a point that Day and Brodsky emphasize, as well as others including Lisa Philipps and Margot Young, "Sex, Tax and the Charter" (1995) 2 Review of Constitutional Studies 221; Claire F.L.Young, "(In)visible Inequalities: Women, Tax and Poverty" (1995) 27 Ottawa L. Rev. 99 and "Public Taxes, Privatizing Effects, and Gender Inequality", supra note 8.  -207women's economic security. However, Day and Brodsky are mindful o f the limits o f Charter litigation and consider that it must be strategic, "one element o f a long term struggle to overcome ideologies o f dominance, and to establish understandings that can actually speak to the material conditions o f women's inequality." They are sufficiently sceptical to acknowledge that litigation cannot be effective on its own, but argue that it should not be discarded either, as "feminists should use all the spaces that are available for feminist advocacy, and should also push for spaces to be created and expanded."  74  Day and Brodsky do not limit the use o f Canada's international human rights agreements to interpretive aids for Charter litigation. Although there are practical hurdles to be dealt with in using international agreements and fora, Day and Brodsky see these as areas for exploration followed by activism, spaces available for feminist advocacy that should be expanded.  75  W h i l e the  motivation o f opening all available spaces for promoting feminist issues is valid, it makes sense that results are more likely to come from within Canada than outside. While the possibility o f successful Charter litigation based on economic equality may be remote, i f it materializes the result w i l l be an equality enhancing remedy enforceable against the government. The effectiveness o f a remedy granted by an international forum is much less certain. If political embarrassment is the desired result o f international litigation, perhaps it could be achieved through more effective means.  (b)  Political A c t i o n The limitations associated with Charter litigation, particularly i n relation to altering the  'Day and Brodsky, supra note 25 at 112. 'Ibid, at 143-151.  -208economic aspects o f social inequality, designate political (non legal) action as the dominate strategy for effecting the goals o f the equality enhancing agenda outlined above.  It follows from the  previous discussion, however, that legal and political action are not either/or choices. If there is a choice, it is political action without legal action or political action that incorporates strategic litigation within its overall strategic plan. Although legal action may be capable o f dealing only with the symptoms o f inequality, resolving legal issues relating to symptoms may be an important strategic step toward the ultimate goal. Further, unsuccessful litigation may effect a positive political result by raising public interest and awareness.  While he is sceptical that the Charter can effectively advance the struggle for economic equality, Bakan concludes that positive social transformation is attainable through other forms o f political action:  The struggle for social justice is much larger than constitutional rights; it is waged through political parties and movements, demonstrations, protests, boycotts, strikes, civil disobedience, grassroots activism, and social commentary and art. In Canada we must use all of these measures to combat the vast disparities of welfare and power among us and to ensure that all people have an effective voice in determining the conditions of their lives. 76  Day and Brodsky would add constitutional litigation, reinforced by international human rights agreements, and possibly action before international fora to their arsenal, for use in the strategically appropriate circumstances. Herman would probably not reject Charter litigation out o f hand either, though she would be wary o f its use in pursuing a claim for economic equality. W h i l e all would place political action above legal action, or more accurately would subsume legal action within political action, Fudge and Glasbeek would be alone in their absolute rejection o f litigation. There  'Supra note 1 at 152.  -209is unanimity among all o f these scholars, however, on the regressive effects o f restructuring and the neoliberal agenda on social (in)equality and the need for political resistence to combat these effects. There is no magic formula, then, no step by step guide, for implementing an agenda that promotes equality for women in the retirement income system. T w o points seem clear, however. First, redressing the inequalities for elderly women within the retirement income system requires a progressive reform strategy that is informed by an understanding o f the linkages between gendered economic inequality on the one hand and familial ideology, the public/private divide, the Keynesian welfare state, neoliberalism and restructuring on the other, and the interconnections between these. Second, progressive reform o f the retirement income system requires gendered political action in a degendered political climate. The effect of restructuring has been that the step forward for women that pension reform promised was transformed into two steps back. Brodie suggests that women w i l l be unable to regain the political power that they have lost over the last decade without taking the necessary first step o f contesting neoliberal orthodoxy.  77  This w i l l not  be an easy goal to achieve because, as Janine Brodie has pointed out, one o f the characteristics o f restructuring is that issues are degendered and the voices o f women (and others) who seek state activism on their behalf have been pushed to the margins o f society.  78  To regain their pre-  restructuring position women must take whatever political action is available to them to destroy the consensus, reinforced by government rhetoric, that social programs must be cut because they are unaffordable, and to expose this rhetoric as a mask that hides the political desire to increase the  Janine Brodie, Politics on the Margins: Restructuring and the Canadian Women's Movement (Halifax:  77  Fernwood Publishing, 1995) at 47. See also Brodie's discussion of a feminist response to restructuring at 74-83. '"Ibid, at 80-83.  -210power o f the market (capital) through government abdication o f democratic control over economic life. But the struggle cannot stop there. Gendered poverty was a serious problem within the retirement income system prior to restructuring.  Although restructuring has exasperated the  problem, a return to the pre-restructuring (Keynesian welfare state) political agenda w i l l merely be a return to the problems identified in the Green Paper. However, an activist state is more conducive to the struggle for progressive reform than one that subordinates its role to market forces.  -211CONCLUSION  M y attention was first directed toward the issue o f poverty among the elderly two years ago when a woman in her m i d eighties came to me for legal advice on the admission o f her husband of more than sixty years into a nursing home. While I was able to help her through the bureaucratic red tape that this change entailed, I could not give her the assurance she sought, that her financial well-being would not be affected. Her husband had worked as a farm labourer and retired without an employer sponsored pension. Their post retirement income consisted only o f O A S and GIS. They had always lived frugally and had accumulated savings o f $20,000 over the sixty years o f their marriage. In addition they had a home valued at less than $60,000. To offset the cost o f nursing home care, the husband's O A S and his portion o f the GIS followed him to the nursing home, leaving her with half of their already meagre income to maintain the home and provide for her needs. The savings too, except for an amount deemed sufficient for burial expenses, was subject to expropriation to offset nursing home expenses. While she could remain in the home, it looked very doubtful that she could continue to afford to maintain it on her much reduced income, ' and i f she sold it the proceeds would be subject to the same expropriation rules that applied to the savings. In short, on top o f the emotional stress that accompanied her reluctant admission that she could no longer cope with the care that her husband required, this woman had to face a sudden plunge from a frugal lifestyle to the depths o f poverty.  Initially I considered my client's situation as a degendered issue. After all, i f she had been the one admitted to the nursing home, her husband would be in the same financial situation that she now faced. However, I had only scratched the surface in exploring the issue when I realized that  -212elderly women are much more vulnerable to poverty than elderly men, and that poverty is very much a gendered issue at all age levels. Since pensions are the major source o f income for those over age 65,1 sought in this thesis to understand the relationship between the tenuous financial well-being o f elderly women and the retirement income system.  U s i n g familial ideology as an analytical tool, I examined the legislative provisions which permit access to the retirement income system and discovered that access is either through workforce participation or spousal relationships. I came to understand that these provisions, grounded as they are in assumptions about the 'natural and obvious' roles o f men and women, reinforce and perpetuate gender inequality. Though the relevant legislative language is gender neutral, access through workforce participation ignores contributions made by women to reproductive work and assumes that the male pattern o f workforce participation, in terms o f both time and wages, applies to all workforce participants. A s for access through a spousal relationship, the gender neutral term 'spouse' is informed by dominant familial ideology and, i n the context o f the spousal provisions relating to the retirement income system, it means dependent spouse, which in turn means 'woman'. Accordingly, gender neutral language does not denote gender equality. The informing ideology survives, and maintains inequalities masked by the neutral language.  I have come to understand that familial ideology is also reflected i n the limitations women face in their access to, and ability to benefit economically from, full participation in the workforce. It is perhaps more clearly reflected within the spousal provisions relating to the retirement income system, which legitimize the privatization o f the economic well-being o f women. The benefits o f these provisions are extended to women only where a man cannot provide for her - specifically,  -213when the supporting spouse has only the subsistence level income provided by O A S / G I S and on his death (survivor benefits). The benefit o f tax incentives for retirement savings are defacto given to men; it is assumed that their spouses w i l l be taken care o f by voluntary sharing within the family. In addition, because a spousal relationship triggers all o f these benefits, many who also may need these are excluded.  I agree with Shelley Gavigan's opinion that  "...one's access or entitlement to social benefits, to health care, dental care, longterm disability benefits, and bereavement leave is, in sum, one's dignity and personal and economic security. It should not, and need not, depend upon being situated in or relegated to a familial relationship." 1  H o w , then, can gender equality to such access and entitlement be achieved?  Following the  publication o f the Green Paper which held out hope for redressing the inequalities for women in 2  the retirement income system, several feminist scholars suggested concrete changes that, i f implemented, promised to advance equality for women within the retirement income system.  3  However, as pension reform was implemented i n the 1990s, gender (in)equality issues disappeared from the reform agenda. While, as Susan B o y d has pointed out, there is no easy way to overcome the effects of ideologies, 1 have come to understand that the omission o f gender equality enhancing 4  initiatives from the reform o f the retirement income system cannot be accounted for by the  'Shelley A.M., Gavigan, "Paridise Lost, Paradox Revisited: The Implications of Familial Ideology for Feminist, Lesbian and Gay Engagement to Law" (1993) 31 Osgoode Hall L.J. 589 at 624. Government of Canada, Green Paper on Pensions (Ottawa: Departments of Finance and Health and Welfare Canada, 1982). See supra Chapter Five note 15.  3  "Susan Boyd, "Can Law Challenge the Public/Private Divide? Women, Work and Family" (1995) Windsor Yearbook of Access to Justice 161 at 182-185.  -214resilience o f familial ideology alone. The direction that reform has taken follows the agenda prescribed by the emergence o f neoliberalism as Canada's dominant governing orthodoxy, and the effect is a deeper entrenchment o f gender inequalities in the retirement income system.  A s Day and Brodsky point out, the "tendency for discussions about poverty and social programs to become divorced from a critique o f underlying unequal power relations and o f social institutions" presents an obstacle to understanding "the gendered dimensions of poverty." In this 5  6  thesis my object has been to overcome that obstacle and thereby achieve an understanding o f postretirement gendered poverty that can effectively inform strategies for moving toward economic autonomy for elderly women. A t best this understanding is only a beginning for the complex task o f improving a retirement income system that, i n its present form, imposes the cruel penalty o f poverty on elderly women during their final years.  Shelagh Day and Gwen Brodsky, Women and the Equality Deficit: Restructuring Canada's Social Programs  5  (Ottawa: Research Directorate, Status ofWomen Canada, 1998) at 5.  -215BIBLIOGRAPHY A l m e y , Marcia, Finding Data on Women: A Guide to Major Resources at Statistics Canada (Ottawa: Research Directorate, Status o f Women Canada, 1998). Armstrong, Pat, "Women's Work: Women's Wages" i n Greta Hoffman Nemiroff, ed., Women and Men: Interdisciplinary Readings on Gender (Montreal: Fitzhenry and Whiteside Limited, 1987)354. Armstrong, Pat, "The Feminization o f the Labour Force: Harmonizing D o w n i n a Global Economy" i n Isabella Bakker, ed., Rethinking Restructuring: Gender and Change in Canada (Toronto: University o f Toronto Press, 1996) 29. Armstrong, Pat, "Restructuring Public and Private: Women's Paid and Unpaid W o r k " in Susan B . Boyd, ed., Challenging the Public/Private Divide: Feminism, Law, and Public Policy (Toronto: University o f Toronto Press, 1997) 37. Austin, Barbara, "Policies, Preferences and Perversions i n the Tax-Assisted Retirement Savings System" (1996) 41 M c G i l l L . J . 571. Bakan, Joel, Just Words: Constitutional Rights and Social Wrongs (Toronto: University o f Toronto Press, 1997). Baker, Maureen, Aging in Canadian Society: A Survey (Toronto: M c G r a w - H i l l Ryerson, 1987). Baker, Maureen, ed., Families: Changing Trends in Canada (Toronto: M c G r a w - H i l l Ryerson, 1996). Bakker, Isabella, ed., Rethinking Restructuring: Gender and Change in Canada (Toronto: University o f Toronto Press, 1996). Bakker, Isabella, "Introduction: The Gendered Foundations o f Restructuring in Canada" in Isabella Bakker, ed., Rethinking Restructuring: Gender and Change in Canada (Toronto: University o f Toronto P