The poverty of unattached senior women and the Canadian retirement income system: a matter of blame or contradiction?

Structural and financial inadequacy of Canada's retirement income system, especially with respect to income support benefits (i.e. Old Age Security), are often identified as one major reason unattached senior women experience poverty. While it may be compelling to blame low benefit levels and changing eligibility requirements, particularly because 'crisis' policy discourses have influenced questionable restructuring over time (i.e. the clawback), this paper argues that this is too simplistic of an account of the relationship between these women's poverty and the retirement income system. Other broad social-structural factors are at play in women's lives that have the potential to disentitle their access to income security in old age. Specifically, the mismatch between women's economic situations over the life course and their claims to pension or retirement savings income is presented as an important reason for why many women are still poor despite policy provisions for their retirement.

Key words: poverty, unattached senior women, retirement income, policy discourse

**********

Canada's retirement income system consists of three levels: income supports benefits (Old Age Security), social insurance (Canada Pension Plan) and occupational pension plans and registered retirement savings plans. As a system that expanded with the evolution of the post-war welfare state, it was designed to provide income security to elderly persons who may experience economic risks associated with age related factors such as compulsory retirement and health problems. Despite these intentions, many senior citizens are still poor. According to recent Census data, more than 600,000 Canadians over the age of 65 were living in low income in the year 2000. More unattached senior women live in low income than unattached senior men (approximately 428,300 compared to just over 173,000, respectively) (Statistics Canada, 2003).

The inadequacy of Old Age Security is often identified as one major reason behind the poverty experienced by unattached senior women. This blaming of the first level of Canada's retirement income system is not surprising, especially in view of the 'crisis' policy discourses surrounding population aging and the federal debt that have influenced questionable restructuring of income support benefits over time (i.e. the clawback). However, this is not an accurate portrayal of the relationship between these women's poverty and the retirement income system. It is the argument of this paper that other, broad social-structural factors inhibit women's realization of economic security through the second and especially third level of the system. Across their life course, women overwhelmingly experience occupational segregation, income disparity, and unpaid work responsibilities prior to old age. These significant barriers impede their access to suitable pensions and/or their ability to save for their future economic security through occupational pension plans or retirement savings plans.

In exploring the relationship between the poverty of unattached (widowed, divorced/ separated or ever single) senior women and Canada's retirement income system, the first section of this paper provides a brief description of the structure and provisions of the three levels of the system. Attention is drawn to the persistence of poverty among senior women and how particular discourses, sometimes contradictory, have influenced restructuring of the system in the second section. In the third section, I describe social-structural factors at play in women's lives and show how they have the potential to disentitle them to economic security through the second and third levels of the retirement income system. This discussion demonstrates that the mismatch between women's economic situations over their life course and their claims to retirement income security is an important reason why they are still poor despite policy provisions for their retirement. In the final section of this paper, it is suggested that future restructuring of the retirement income system needs to incorporate a new policy discourse that recognizes this mismatch.

Providing Economic Security for Senior Women (and Men)

Upon disengagement from the labour force, women may seek economic stability through the three levels of the retirement income system. The first level, Old Age Security, was implemented in 1952 and reflected post-war Canada's concern with establishing a 'social safety net' that recognised all individuals' shared vulnerability to forces beyond their control, such as risks associated with unemployment, sickness and old age (Armstrong, 1997, p. 5354). Senior women are eligible for Old Age Security (OAS) if they are over age 65, have Canadian citizenship and have resided in Canada for over 20 years (Rice & Prince, 2000). They qualify for maximum old age pensions providing their net incomes are less than $60,806 (see Table 1). Seniors who have net incomes over this amount have some of their benefits withheld (re: the 'clawback') each month and seniors with net incomes above $98,547 do not receive any benefits (Social Development Canada, SDC, 2005a). The old age pension is subject to income tax but continues to be fully protected against inflation; benefits are reviewed and indexed according to increases in the cost of living as measured by the Consumer Price Index (SDC, 2005b).

Pensioners with zero or very limited income are eligible for another core component of this first level of the system, the Guaranteed Income Supplement (GIS). Implemented in 1967, this benefit is now paid monthly to seniors with net incomes below $35,592 and is non-taxable (SDC, 2005a). In 1975, the Spouse's Allowance was introduced to help low-income persons married to recipients of the Guaranteed Income Supplement. Currently known as the Allowance, it is entitled to persons (age 60-64 years) whose partner has died or who are not entitled to their own pension and are living on the pension of their partner until they become entitled to receive OAS at the age of 65. To qualify for the Allowance, the combined yearly income of a senior couple or the income of the survivor must be below $25,152 (SDC, 2005a) (see Table 1 for an overview of these components of the first level). Although beyond the scope of this paper, readers should note that Canadian senior women and men are also eligible for income security benefits through provincial/territorial governments and benefits (i.e. the Age Credit and Pension Income Credit provided through the income tax system) (Clark, 1998, p. 78; Gee & McDaniel, 1991).

The OAS was intended as a foundation for individuals' retirement incomes. Women and men were expected to add to their pensions through the second and third levels of the retirement income system. The Canada Pension Plan (CPP) was introduced in 1966 to provide basic pension income, survivor and disability insurance benefits to people age 65 and over who were past members of the paid labour force and had contributed to the plans alongside their employers (Armstrong, 1997; NCW, 1999; Rice & Prince, 2000) (for a complete discussion of the different designs and implementation of the Canada and Quebec Pension Plans (C/QPP), see Bryden, 1974). The amount of the CPP is dependent on the number of years worked and the level of past earnings (O'Grady-Leshane, 1993) (see Table 2 for maximum benefit amounts). Outside of Quebec, all members of the paid work force must contribute to the plan, whether they are employees, employers or self-employed. As a pay-as-you go plan, contributions from individuals in the paid labour market are used to pay the pensions of those who are retired (Townson, 1995). These contributions are paid on earnings between $3,500 and $40,500, are tax deductible, and are adjusted for inflation every January (SDC, 2005c).

The CPP provides a retirement pension as early as age 60 for seniors and replaces approximately 25 percent of the contributions paid into it (SDC, 2005c). The period of contributions required for a full CPP benefit is 40 years (O'Connor, Orloff, & Shaver, 1999; Rice & Prince, 2000). For individuals who do not contribute to the plan because of time spent outside of the labour market, the CPP excludes 15 percent of the lowest earnings (roughly seven years) from the calculation of retirement income (Townson, 1995). Time spent outside of the labour force to raise children under the age of seven can also be dropped out of the calculation of CPP (SDC, 2005c).

Aside from accommodating family responsibilities, the CPP offers several other advantages to pensioners, particularly women. The CPP covers all sectors of the economy, it includes part-time and self-employed workers, it is portable upon change of employment, and it can be shared by spouses upon divorce or upon retirement (Townson, 1995; Townson 2000). The CPP provides other supplementary benefits, such as disability benefits to persons unable to work including supplementary child benefits for disability recipients, pensions to surviving spouses, benefits to dependent children of deceased plan members, and lump sum death benefits (Clark, 1998; NCW, 1999: 15-16) (see Table 2 for benefit amounts).

Occupational pension plans and registered retirement savings plans remain as the final and third level. Because they originated in the 1800s, these plans have had a much longer history than the first two levels (Clark, 1998). These plans are sponsored by employers, labour unions, and professional organizations. They are paid into by employees or employees and sponsors to defer wages to provide for their retirement (NCW, 1999). The government assists in seniors' savings for retirement by deducting contributions to these plans from taxable income and not taxing investment income as it is earned. Taxes are paid on these plans when funds are withdrawn or received as pensionable income. The amount of income received by pensioners who participated in employer sponsored pension plans depends on their age of retirement and the plan's benefit formula (SDC 2005e). In comparison, the amount of income a pensioner can expect from their own personal registered retirement savings plan is dependent on how much money they have invested into this plan prior to retirement. These plans are intended to provide pensioners with up to 70 percent of their pre-retirement earnings (NCW, 1999).

Contradictions and Persistent Poverty: System Inadequacies or a Mismatch?

Despite these provisions for their income security in old age, poverty among senior women is persistent. In 1983, 62 percent of unattached women were living in poverty (Baker 1993, p. 294, cited in Baker & Tippin, 1999, p. 93). In 1991, 34.9 percent of women aged 65-69, 48 percent of women aged 70 to 75, and 53 percent of women over 85 who lived alone had incomes below Statistics Canada Low Income Cut Offs (LICOs) (Moore & Rosenberg, 1997, cited in McDonald, 1997). Data from the 2001 Census show that of those living alone in 2000, the low income rate was still higher among women than men (43 percent and 31 percent, respectively) (Statistics Canada, 2003).

Not only are these women more likely to be persistently poor (their incomes are below Statistics Canada's Low-Income Cut Offs for two consecutive years) than women in married or common-law relationships (Lochhead & Scott, 2000), their experiences of poverty differ depending on their marital status. McDonald (1997) discovered that although government transfers were indeed crucial to maintaining the income of some married, widowed, divorced/separated, and ever single women over the Low-Income Cut Offs (LICOs) in 1993, almost one half of all widows over the age of 65 (49%) lived below the poverty line. Widowhood continues to negatively affect women's economic situations. Many widows even experience continuous decline in income years after they are first widowed. Following the impact of widowhood on women aged 65 and over between 1990 and 2001, Li (2004) found that median family income declined 9.8% among widows over approximately five years, compared to the 1.5% decline experienced among women not widowed.

Since unattached senior women depend on government transfers in the form of OAS/GIS and these transfers do not raise them above the poverty line (NCW 1999), this would appear to suggest that the continued poverty among these women is due to an inadequate level of income support benefits. This view is especially compelling because of 'crisis' policy discourses--population aging as paradigm (McDaniel, 1987, p. 334) and deficit as paradigm (McDaniel & Gee 1993, p. 60)--that have increasingly influenced discussions and changes to this level of the system.

In the 1980s, the growing population of seniors (as a result of increased life expectancy and low fertility) and expected rise in pension expenditures, in conjunction with the growing deficit and debt, signalled to the federal government an impending crisis. An already economically stressed system was predicted to become more expensive and less likely to be sustained in the future (Baker & Tippin, 1999). It was in the federal budget speech of April 1989 that the Conservative government announced that seniors with high incomes would receive their old age pension cheques but would be obligated to repay some or all of the amounts they received (NCW, 1999; Cheal & Kampen, 1998) (above $50,000 in individual net income) at a rate of 15 percent in income tax (Pierson & Smith, 1994). Gee and McDaniel (1991) argue that this 'clawback' was largely a response to the crisis discourse surrounding population aging.

The CPP has not escaped the influence of these crisis discourses either, albeit with less controversial but interesting consequences. For example, increases in pension contributions were agreed upon in order to cover the future costs of an aging population (Baker & Tippin, 1999). In 1966, the contribution rate for the CPP was 3.6 percent (1.8% from employees and 1.8% from employers) of contributory earnings. In 1995, this rate was increased to 5.8 percent of earnings (2.9 % from employees and 2.9% from employers), with maximum pensionable earnings set at $35,400, and increased to 7.0 percent in 1999, with maximum pensionable earnings set at $37,400 (Baker & Tippin, 1999). In the year 2003, the contribution rate increased to 9.9 percent (NCW, 1999; SDC 2005f). Other changes made in 1997 negatively affect seniors' economic situations. Freezes made to the maximum death benefit (frozen at $2,500) and the Year's Basic Exemption (frozen at $3,500) means that these CPP provisions and stipulations are no longer increased with inflation.

Blaming inadequate benefits and questionable restructuring of especially the first level of the system for women's poverty, however, is too simplistic of an argument for two inter-related reasons. First, the income guarantee for seniors through OAS has improved even when considering the 'clawback.' For example, in 1952 the maximum OAS for residents living in a large city (over 500,000 population) provided singles $2,956 and couples $5,912 compared to $10,264 and $16,642 in 1995 (constant 1995 dollars) (Battle, 1997, p. 528-529). Second and not surprisingly, the overall poverty among seniors has declined. In 1980, the national poverty rate was measured at 33.6 percent but by 1990 it had declined to approximately 19 percent (NCW, 1992, p. 10-11, cited in Myles and Street, 1995, p. 339) and remained at this level in 1994 and 1995 (Lee, 2000; National Advisory Council on Aging, NACA, 1999). By the year 2000, the poverty rate among seniors declined to 17%, nearly half of the 30% rate in 1980. The decline in poverty is credited to higher CPP premiums and rising OAS and GIS benefits, the latter of which make up 2/3 of the income of low income seniors (Statistics Canada, 2003).

This overall decline in poverty, however, should not mask the fact that when urban and rural differences in measurements of poverty are taken into account, the national poverty rate can be misleading. For example, considering the 19 percent rate of 1995, Lee (2000, p. 30) uses 1996 Census data to show that 25 percent of seniors in cities were poor in this same year--among the 1.4 million elderly in cities, 349,900 lived below the poverty line. The National Council of Welfare (1999, p. 10) confirms that a normal OAS pension, in addition to the maximum GIS, provided single seniors and couples with an income very close to the poverty line for rural areas but below the poverty line of a large city. The combination of maximum CPP benefits with normal OAS benefits does somewhat improve the economic situations of seniors but this too is dependent on whether seniors reside in urban or rural areas and are either single or couple unions. Readers are therefore cautioned to note that depending on what type of measurement of poverty is used, which variables, and whether references are to single unattached seniors or senior couples, a different picture of poverty among seniors can emerge (as noted, the majority of the studies referred to in this paper rely on Statistics Canada's LICOs, Statistics Canada, 1999a).

Moreover, this decline in overall poverty should not mask the need to consider what does account for the persistent poverty among women. Setting aside the convincing and influencing nature of crisis discourses, and given the overall improvement of the economic situations of seniors over time, it appears too easy to credit women's poverty to inadequate old age pensions. Although women are the persistent poor, they are faring better than before the post-war design of policy provisions for their retirement. Other factors must be at play, then, and partially account for the persistence of their poverty. Specifically, what must be taken into account is the mismatch between women's economic situations over the life course and their claims to pension or retirement savings income.

Social-Structural Factors at Play in Women's Work and Family Lives Preceding Retirement

Over the past three decades, women's labour force participation has dramatically increased. In 1953, the labour force participation rate for women was only 23 percent. By 1999, the percentage of women in the labour force had increased to 58 percent; the participation for men was 71 percent (Federal, Provincial, and Territorial Ministers Responsible for the Status of Women, FPM, 2001). Prior to retirement, women are much more likely to be involved in the labour market today than in the past. This involvement, however, is not necessarily a guarantee of their income security in the short-term. Women experience particular social-structural factors in their work and family lives that have implications for their income security in the long-term.

Women have increased participation within the labour force but this has occurred predominantly in non-standard jobs, which are characterised by low wages, lack of security over time, and have little to no benefits (Freiler & Cerny, 1998; NACA, 2000). Women are most often employed as part-time, seasonal, or contract workers, or are self-employed (Baines, 1996; Benoit, 2000; Chaykowski & Powell, 1999) in the service sector or have clerical roles (O'Connor et al., 1999). In contrast, men are over-represented as employees in transport and communications, and manufacturing sectors (O'Connor et al., 2000) and are more likely to occupy jobs that are full-time with supervisory responsibilities (Statistics Canada, 1999b). In 1998, 71 percent of working women worked full-time compared with 90 percent of working men (FPM, 2001). Not surprisingly, Cheal and Kampen (1998) find that women aged 54-63 have about half of the number of years of past work experience of men at this same age and about three fifths of their current weeks of employment.

Women's engagement in work that is unequal to men's in terms of time spent is related to their primary responsibility for domestic labour within the home. Women still engage in more household chores and caregiving than men (Benoit, 2000; Gazso-Windle and McMullin, 2003; O'Grady-Leshane, 1993). Indeed, ideological assumptions about appropriate gender behaviour tend to maintain the division of separate spheres of work for men and women (public or paid labour v.s. unpaid private or domestic labour) in a subtle fashion, even for women who do engage in paid work (Gazso-Windle and McMullin, 2003). Women can experience a high probability of having to juggle their family demands (i.e. housework and child care) with their engagement in paid labour, and thus, face a double day of work in terms of hours spent in domestic labour within the home and hours spent in paid work. Further, many women are expected to interrupt their careers in the work force more often than men to meet their family responsibilities (NACA, 2000). These interruptions are not always for rearing children. An implication of population aging is that many women are now the primary providers of care to aging family members (O'Grady-Leshane, 1993) and have disengaged from the labour market even before the age at which they would retire to provide this care.

This differential participation of women and men in the labour force due to occupational barriers and family responsibilities produces substantial differences in net family incomes. In 2001, the average income of women who worked full-time, full year was $35,258 and was $49,250 for men (Statistics Canada, 2004a). With respect to all earners, including part-time and contract workers, women still had a lower income than men ($24,688 and $38,431 respectively). It is these experiences of occupational segregation, income disparity, and responsibility for family members among women over their life course that are relevant to women's experiences of poverty in old age. As discussed below, these social-structural factors have the potential to inhibit women's realization of economic security through the second and especially third level of the retirement income system.

The Second and Third Levels and Women's Income Insecurity

With respect to the social-structural factors of occupational segregation and income disparity experienced by women, recall that lifetime earnings and number of years of coverage from the CPP affect the amount of old age pension a person receives. Given women's lesser wages and varying degrees of participation in the labour market, it is not surprising that women who engage in paid labour prior to retirement experience unequal CPP coverage compared to men. Data demonstrating gender differences in coverage show that for January 1999, the average monthly retirement pension paid to pensioners (ages 65 to 69) was $533 for men and only $299 for women (the maximum CPP rate was $752); women received 56 percent of the average amount paid to men (NCW, 1999).

Women's ability to obtain adequate income security through the CPP is also unlikely if they engaged in low-wage part-time work that resulted in their incomes being below the Year's Basic Exemptions. Women who spend more than seven years outside of the labour force caring for children or family members will exempt themselves from adequate CPP coverage. Many women do not even receive full CPP pension but rely on survivor benefits that are not equivalent to the full amount (Ross, 2000). Finally, for women who are divorced, their economic security can be challenged by not only the economic consequences of marital dissolution, but also by the fact that provincial governments have the option to over-ride the mandatory credit-splitting of CPP benefits upon marital dissolution. According to the National Council of Welfare (1999), mandatory credit-splitting as an option for economic security for divorced women is far from the norm.

Some women do rely on occupational pension plans and registered retirement savings plans for income security. Relying on data from Statistics Canada Survey of Consumer Finances, the National Council of Welfare (1999) states that of the total number of poor unattached women identified in the Survey (359,000), 29 percent reported income as a result of investing and saving and 15 percent reported income as a result of occupational pension plans. Today, more women in the paid workforce are being covered by employer-sponsored (registered) pension plans in both public and private sectors than in the past (Schembari & Anderson, 2004). Most often, however, women's access to these plans in order to save for their retirement is challenged by their labour force participation and family responsibilities prior to retirement. Understandably, fewer women benefit from private pension plans than men because proportionately fewer women worked all of their lives compared to men, and thus, contributed less to these plans over time compared to men (NACA, 1999).

For women who did engage in paid work, their ability to access occupational pension plans is also not absolute. Occupational pension plans, because they are a private component of the retirement income system, are not readily available to all employees. For example, in 1997, less than half of all Canadian workers (42%) were covered by occupational pension plans (NCW, 1999). By the year 2002, nearly 40% of all paid workers (self-employed, unpaid family workers and unemployed are non-eligible) were covered by employer-sponsored plans (Schembari & Anderson, 2004). The public sector also experiences higher coverage by occupational pension plans than the private sector; women's recent increased access to plans has occurred in the public sector (Schembari & Anderson, 2004). However, since women are more likely to be employed in the private sector in low-wage paying jobs, their lower occupational plan coverage can therefore be attributed to where they work and their lack of financial ability to contribute to these plans. O'Connor et al. (1999) explain that men are more likely to benefit from the third level of the system because they are more likely to be highly paid professionals or managers in organized businesses or unions that are likely to encourage participation in occupational pension plans or enable investment for retirement because they produce disposable income. In addition, women depend on public pensions more than men for their economic security simply because they are more likely to lose private pension credits when they interrupt their employment for family reasons (Baker & Tippin, 1999). Even if women are able to access this level of the retirement income system, they may experience low-income because few occupational plan members have automatic protection from inflation (NCW, 1999). Indeed, while women may receive a pension at age 65, the absence of inflation coverage may mean the pension erodes to the point of insignificance, especially if their life span increases another thirty years.

Women's participation in low paying jobs can further deny their opportunity to save for their retirement through retirement savings plans. Women who engage in low-wage work and experience costs associated with child rearing or the care of a sick family member do not necessarily have the ability to place money aside and benefit from tax savings through registered retirement savings plans. For example, in 1996, only 4 percent of low wage-earners belonged to contributory plans, with an average contribution of $253 for plan members with incomes less than $10,000 (NCW, 1996). Cheal and Kampen's (1998) discovery that fewer women aged 53-64 had substantial retirement income pensions or annuities than men (5.8% and 13.2% respectively) in 1993 is therefore to be expected.

In her study of the factors that affect the economic situations of widows, McDonald (1997) provides excellent evidence of how occupational barriers and family responsibilities can disentitle these women from adequate economic security through the second and third levels of the retirement income system. While the first level of the retirement income system does not protect widows from poverty simply because government transfers in the form of the OAS/GIS/SPA do not raise many widows above the poverty line, the second level is equally ineffective because widows are often self-employed or are the most likely to have worked in bad or non-standard jobs with low wages and few fringe benefits. Prior to the death of their spouse, many widows also retire early out of necessity to caregive for their spouse. The third level has the potential to guarantee women an economic situation that is above the LICOs but few widows have access to these sources of funding simply because of their inconsistent patterns of paid work and family responsibilities (McDonald, 1997) over their life course. In a more recent study, Li (2004) maintains that the decline in family income experienced by widows over time is due to an equal decrease in several income sources. Among all income sources considered (i.e. OAS, pensions, other transfers, earnings, assets, other income), Li found that each contributed to about 20% of the total decline.

O'Connor et al. (1999) maintain that the first two levels of public provision for retirement can be thought of as primarily a welfare state for women and the third level can be thought of as a welfare state for men. This statement captures the mismatch between women's economic situations over the life course and their claims to economic security through the second and third levels of the retirement income system. Although the first level of the retirement income system is not the sole cause of women's poverty, the system is essentially unbalanced simply because its third level is still largely the preserve of high-income seniors (usually men) (Battle, 1997).

Toward Rethinking Retirement Income for Unattached Senior Women: Encouraging a New Discourse

The improvement of unattached senior women's economic situations seems like a daunting task, especially given policy discourses that promote restructuring the retirement income system to be sustainable and affordable in the future and do not make mention of the interrelationships among labour force participation, family responsibilities, and retirement. Population aging as paradigm and deficit as paradigm, as well as interest in targeting benefits to the appropriate segment of the elderly population, continue to influence discussions of restructuring income security benefits for seniors.

In 1995, the Liberal government proposed the combination of the OAS and GIS to form the Seniors Benefit, which was designed to make the system more affordable and sustainable by improving the targeting of benefits to low-income seniors (Armstrong, 1997, p. 65). The benefit was to be based on net family income, rather than individual income, and was non-taxable. It would have paid maximum benefits to seniors with net family incomes under $25,921, but would have been clawed back at a rate of 20 percent over this amount up to $52,000 (McDonald, 2000). The Seniors Benefit would have been completely clawed back from an individual who made over $52,000 in net family income and from a family who made over $78,000 in net family income.

In cutting benefits to middle-income and higher-income seniors by assessing eligibility on the basis of net family income and imposing a large clawback, the Seniors Benefit would target low-income seniors. This benefit, however, would have only paid $120 more per year than the support of the OAS and GIS it was intended to replace (NCW, 1996; McDonald, 2000). And, aside from this very modest increase, the Seniors Benefit would have removed the pension rights of low-income women who were married to men with net family income above the threshold (Baker & Tippin, 1999). Overall, the Seniors Benefit is more accurately seen as a means to resist the impending costs of maintaining an aging population of seniors and as a means to reduce the federal deficit. As McDonald (2000) explains, the federal government would have saved considerable amounts of money through this proposed benefit. Originally anticipated to commence in January 2001, the Seniors Benefit was abandoned in July 1998, and thus, restructuring of the first level of the retirement income system has temporarily stalled.

At the same time that these policy discourses continue to hold sway, two notable Canadian policy publications reflect growing awareness that social-structural factors can negatively impact the economic security of women in their old age. The 2001 Federal, Provincial, and Territorial Ministers Responsible for the Status of Women publication lists the labour market, income and earnings, the balancing of employment with family responsibilities, and unpaid work as the social structural factors that impinge upon a woman's economic situations over their life course (p. 9). Similarly, the National Advisory Council on Aging (1999, p. 54) writes that the economic situation of all future seniors, especially women, can be immediately improved by tackling the polarization of the labour market, the professional ghettoization of women and unequal access to employment, and wage disaparities between women and men for comparable work. By combining these points with the discussion of this paper, I have summarized the major social-structural factors at play in women's work and family lives that implicate their retirement income situations in Table 3.

It is understood that immediate and dramatic corrections of the barriers to women's security in old age are unlikely. This should not mean that there are no possible avenues available to change women's economic experiences in their old age. Indeed, the growing awareness of these social-structural factors suggests a necessary thread of a burgeoning policy discourse on how women's experiences in the short-term impact their economic experiences in the long-term. By recognizing and discussing these social-structural factors in discussions of the experiences of senior women and/or future proposals to restructure the retirement income system, this new discourse has the potential to link with and challenge existing crisis discourses. For example, policy makers need to recognize that the rising number and percentage of the elderly and the growth of non-standard jobs will create heavy pressure on the elderly benefits system in the future (Battle, 1997, p. 541). The less women can achieve income security through the second and third level of the retirement income system because of negative labour market experiences and family responsibilities, the more they will rely on the first level of the retirement income system.

This new discourse also recognizes that the retirement income system is not fully responsible for preventing the poverty among women (Townson, 2000). Other social support systems need to be in place. Community support for the care of elderly family members, affordable, accessible, and quality child care, and adequate pay equity and equal employment opportunity policies can ensure that family responsibilities do not undermine the financial security of women upon retirement. In addition, by inviting further social recognition of the unpaid work done by women in the home, this new discourse prompts serious consideration of one major proposal to ensure women's economic security in old age offered by several feminist policy analysts (i.e. Daly, 2002; Neysmith & Reitsma-Street, 2000; Waring, 1953)--socially recognizing and valuing (and/or financially compensating) women who do exit the labour market to provide care to children and ill or elderly family members.

A most recent development in the province of British Columbia (B.C.) suggests that this idea is not far fetched nor does it need to be gendered. The Human Rights Commission tribunal recently ruled that the provincial government must stop discriminating against disabled people who desire family members as their caregivers rather than government-funded caregivers. It ordered the government to pay more than $100,000 in lost wages, interest and damages to a father who was forced to quit working and access social assistance in order to care for his disabled daughter (The Globe and Mail, July 15, 2004). Now in his 70s, the father provided care for his daughter for seventeen years. Although the B.C. government has responded with a call for a judicial review of this decision of the tribunal, this case may mark the way toward developing policy measures that actually build upon and incorporate this new discourse and in doing so, improve the economic situation of senior women.

Conclusion

Designed with the best of intentions, the retirement income system has not benefited all senior citizens. Unattached senior women consistently experience severe poverty. The influence of crisis discourses that surround discussions of pension policy even appear to sometimes enhance or perpetuate poverty amongst seniors by producing and/or proposing questionable changes to the retirement income system. And yet, it can not be denied that the poverty among senior citizens, including women has declined as a result of targeting benefits to low-income seniors who are in most in need of them. This targeting may even further benefit senior women in the future. The 2005 Budget of the Liberal government proposes boosting the GIS by $2.7 billion over five years. Benefits would climb to $36 for singles and $58 for couples by January, 2007 (Weber, 2005). Blaming the first level of the retirement income system for the continued poverty among unattached senior women denies the existence of other broad, social-structural factors that have the potential to disentitle these women to economic security through the retirement income system. Women do work for less money than men, do engage in more domestic labour than men, and therefore, may not access adequate CPP benefits to 'top up' the income support benefits they receive from the first level and may not be able to save for their retirement through private savings plans.

The intent of this paper was to reveal that these factors must be accounted for in any attempt to understand the poverty among unattached senior women and in policy discussions of restructuring the system to better benefit seniors. Although it can be anticipated that the 'crisis' of population aging and the federal debt will impact future restructuring of the system, this should occur at the same time that a new discourse acknowledges that all three levels of the system interrelate, and more importantly, intertwine with the economic experiences of women over their life course. As Gee and McDaniel (1991, p. 467) argued more than ten years ago: "Income inequalities in later life are largely a function of income inequalities at younger ages." Working toward, investing in, and saving for retirement can not occur if costs associated with population aging hide the costs associated with employment in gendered occupations and primary responsibility for domestic labour, costs most severely experienced by women.

Table 1
Old Age Security Payment Rates (in Canadian funds, January to
March 2005)

                                   Average
                                   Monthly    Maximum     Maximum
Type of                            Benefit    Monthly      Annual
Benefit        Recipient         (Oct. 2004)  Benefit      Income

OAS            all persons         $450.11    $471.76       (a)
GIs            single persons      $388.91    $560.69  $13,464.00
               --spouse of non-    $378.19    $560.69  $32,592.00 (b)
               pensioner
               --spouse of non-    $236.87    $365.21  $17,568.00
               pensioner
               --spouse of non-    $304.27    $365.21  $32,592.00
               allowance
               recipient
Allowance      all persons         $323.74    $836.97  $25,152.00 (c)
Allowance for  all persons         $523.39    $924.04  $18,456.00
the survivor

Source: Adapted from Social Development Canada (SDC) 2005a.

(a) persons with net income above $60,806 must repay part or all of
their OAS, normally through deductions from monthly payments; OAS is
eliminated when net income is $98,547 or above.

(b) The GIS stops being paid at $35,592.

(c) The Allowance stops being paid at $25,152.

Table 2
Canada Pension Plan Payment Rates (in Canadian funds, January to
December 2005)
                                               Average      Maximum
                                               Monthly      Monthly
                                               Benefit      Benefit
Type o f Benefit                             (Oct. 2004)    (2005)

Retirement pension (at age 65)                   $456.92     $828.75
Other benefits:
Disability benefit                               $749.08    $1010.23
Survivors benefit (under age 65)                 $336.68     $462.42
Survivors benefit (65 and over)                  $274.27     $497.25
Children of disabled contributor benefit         $192.68     $195.96
Children of deceased contributor benefit         $192.68     $195.96
Combined survivors & retirement benefit          $622.03     $828.75
  (pension at age 65)
Combined survivors & disability benefit          $889.59    $1010.23
Death benefit (lump sum)                       $2,219.07   $2,500.00

Source: Adapted from Social Development Canada (SDC) 2005d.

Table 3
A New Discourse: Summarizing Social-Structural Factors
of Importance to Women's Lives

* Women and men experience an increasing polarization of good
and bad jobs

* Women and men experience gendered paid work opportunities

* Women have lower incomes from paid earnings than men

* Women spend less time than men in the labour market in terms of
hours worked per week

* Women are still primarily responsible for domestic labour within
the home (including housework, child care, and care of other
family members

* Women's eligibility for the CPP and occupational pension
plans/registered retirement savings plans is dependent on their
wage earnings

References

Armstrong, P. (1997). The welfare state as history. In R. B. Blake, P. E. Bryden & J. F. Strain (Eds.), The welfare state in Canada: Past, present and future (pp. 52-73). Mount Allison University: Irwin Publishing.

Baines, D. (1996). Rebel without a claim: Women's changing bases to claim on the state. Canadian Social Work Review, 13(2), 187-204.

Baker, M., & Tippin, D. (1999). Poverty, social assistance, and the employability of mothers: Restructuring welfare states. Toronto: University of Toronto Press.

Battle, K. (1997). Pension reform in Canada. Canadian Journal on Aging, 16, 519-551.

Benoit, C. M. (2000). Women, work and social rights: Canada in historical and comparative perspective. Toronto: Harcourt Brace & Company, Canada.

Bryden, K. (1974). Old age pensions and policy-making in Canada. Montreal: McGill-Queen's University Press.

Chaykowski, R. P., & Powell, L. M. (1999). Women and the labour market: Recent trends and policy issues. Canadian Public Policy, 25, (1 Supplement), S1-S25.

Cheal, D., & Kampen, K. (1998). Poor and dependent seniors in Canada. Ageing and Society, 18, 147-166.

Clark, C. (1998). Canada's income security programs. Ottawa: Canadian Council on Social Development.

Daly, Mary. (2002). Care as a good for social policy. International Journal of Social Policy 31(2), 251-270.

Federal/Provincial/Territorial Ministers Responsible for the Status of Women. (2001). Women's economic independence and security: A federal/ provincial/territorial strategic framework. Ottawa.

Freiler, C., & Cerny, J. (1998). Benefiting Canada's children: Perspectives on gender and social responsibility. Ottawa: Status of Women Canada.

Gazso-Windle, A., & J. McMullin. (2003). Doing Domestic Labour: Strategising in a Gendered Domain. Canadian Journal of Sociology 28(3), 341-366.

Gee, E., & S. McDaniel. (1991). Pension politics and challenges: Retirement policy implications. Canadian Public Policy, 17(4), 456-472.

Lee, K. (2000). Urban poverty in Canada: A statistical profile. Ottawa: Canadian Council on Social Development.

Li, C. (2004). Widowhood: Consequences on income for senior women. Statistics Canada. Catalogue No. 11-621-MIE-No.015. Retrieved July 24, 2004, from: http://www.statcan.ca/cgi-bin/downpub/listpub.cgi?catno=11-621-MIE2004D15.

Lochhead, C., & Scott, K. (2000). The dynamics of women's poverty in Canada. Ottawa: Status of Women Canada.

McDaniel, S., & Gee, E. (1993). Social policies regarding caregiving to elders: Canadian contradictions. Journal of Aging and Social Policy, 5(1-2), 57-72.

McDaniel, S. (1987). Demographic aging as a guiding paradigm in Canada's welfare state. Canadian Public Policy, 13(3), 330-336.

McDonald, L. (2000). Alarmist economics and women's pensions: A case of 'semanticide.' In E. M. Gee & G. M. Gutman (Ed's.), Overselling population aging: Apocalyptic demography, intergenerational challenges, and social policy (pp. 114-128). Don Mills: Oxford University Press.

McDonald, L. (1997). The invisible poor: Canada's retired widows. Canadian Journal of Aging, 16(3), 553-583.

Myles, J., & D. Street. (1995). Should the economic life course be redesigned? Old age security in a time of transition. Canadian Journal on Aging, 14(2), 334-359.

National Advisory Council on Aging. (1999). Challenges of an aging society: 1999 and beyond. Retrieved January 15, 2004, from: http://www.hc-sc.gc.ca/ seniors-aines/naca/beyond1999/toc_e.htm#index

National Council of Welfare. (1999). A pension primer: A report by the National Council of Welfare. Retrieved December 18, 2004, from: http://www.ncwcnbes.net/htmdocument/principales/onlinepub.htm

National Council of Welfare. (1996). A guide to the proposed seniors benefit. Ottawa: Minister of Supply and Services Canada.

Neysmith, S., & M. Reitsma-Street. (2000). Valuing unpaid work in the third sector: The case of community resources centres. Canadian Public Policy 26(3), 331-346.

O'Connor, J. S., Orloff, A. S., & Shaver, S. (1999). States, markets, families: Gender, liberalism and social policy in Australia, Canada, Great Britain and the United States. Cambridge: Cambridge University Press.

O'Grady-Leshane, R. (1993). Changes in the lives of women and their families: Have old age pensions kept pace? Generations, 17(4), 27-32.

Pierson, P., & Smith, M. (1994). Shifting fortunes of the elderly: The comparative politics of retrenchment. In T. R. Marmor, T. M. Smeeding, & V. L. Greene (Ed's.), Economic security and intergenerational justice: A look at North America (pp. 21-59). Washington: The Urban Institute Press.

Rice, J.J., & Prince, M. J. (2000). Changing politics of Canadian social policy. Toronto: University of Toronto Press.

Ross, D. P. (2000). The Canadian fact book. Ottawa: Canadian Council on Social Development.

Ruggeri, G.C., Howard, R., & Bluck, K. (1994). The incidence of low income among the elderly. Canadian Public Policy, 20(2): 138-151.

Schembari, P. & Anderson, B. (2004). Pension Plans in Canada. Catalogue No. 13F0026MIE2001001. Retrieved September 22, 2004, from: http://www. statcan.ca:8096/bsolc/english/bsolc?catno=13F0026M-2004001

Social Development Canada. (2005a). Old age security payment rates. Retrieved February 22, 2005, from: http://www.sdc.gc.ca/en/isp/oas/oasrates.shtml

Social Development Canada. (2005b). Overview of the old age security program. Retrieved February 4, 2005, from: http://www.sdc.gc.ca/en/isp/oas/ oasoverview.shtml

Social Development Canada. (2005c). Canada pension plan. Retrieved February 27, 2005, from: http://www.sdc.gc.ca/en/isp/common/ris/riscpp.shtml

Social Development Canada. (2005d). Canada pension plan payment rates. Retrieved February 27, 2005, from: http://www.sdc.gc.ca/en/isp/pub/ factsheets/rates.shtml

Social Development Canada. (2005e). Private pensions and savings. Retrieved February 4, 2005, from: http://www.sdc.gc.ca/en/isp/common/ris/ privpension.shtml#ppsl

Social Development Canada. (2005f). 1998 Changes to the Canada pension-plan. Retrieved July 18, 2004, from: http://www.hrdc-rdhc.gc.ca/isp/cpp/ cppchanges.shtml

Statistics Canada. (2004a). Average earnings by sex and work pattern. Retrieved January 14, 2004, from: http://www.statcan.ca/english/Pgdb/labor01a.htm

Statistics Canada. (2004b). The Daily: Wednesday, September 22, 2004: Employer-sponsored pension plans. Retrieved September 22, 2004, from: http://www.statcan.ca/Daily/English/040922/d040922a.htm

Statistics Canada. (2003). 2001 Census: Analysis series, income of Canadian families. Catalogue No. 96F0030XIE2001014. Retrieved February 28, 2005, from: http://www12.statcan.ca/english/census01/products/analytic /companion/inc/pdf/96F0030XIE2001014.pdf

Statistics Canada. (1999a). Low Income Cut Offs. Statistics Canada, Catalogue No. 13-551-XiB. Retrieved December 5, 2004, from: http://www.statcan.ca/english/freepub/13-551-XIB/free.htm

Statistics Canada. (1999b). The Daily: Monday, December 20, 1999: Survey of labour and income dynamics: The wage gap between men and women. Retrieved January 29, 2004, from: http://www.statcan.ca/Daily/English/991220/td991220.htm

The Globe and Mail. (2004, July 15). Family can be paid for caring for disabled, tribunal rules. The Globe and Mail, on-line edition. Retrieved July 24, 2004, from: http://www.theglobeandmail.com.

Towson, M. (2000). Reducing poverty among older women: The potential of retirement income. Ottawa: Status of Women Canada.

Townson, M. (1995). Reforminy, the pension plan: The implications for women. Ottawa: Status of Women Canada.

Waring, Marilyn. (1953). If women counted: A new feminist economics. San Francisco: Harper & Row.

Weber, Terry. (2005, February 23). Goodale charts 'national journey.' The Globe and Mail, on-line edition. Retrieved February 23, 2005, from http://www.theglobeandmail.com/

AMBER GAZSO

University of Alberta

Department of Sociology