Chapter 10 Gendered retirement: The welfare of women and the "new" retirement

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Introduction
A brief history of women's retirement
Modern retirement
What's new about the new retirement
Parallels between the "New Retirement" and women's retirement
Women's work is still women's work
The mismatch between the lives of women and the pension system
Bibliography

Introduction

The dramatic changes associated with economic restructuring and the retrenchment of the welfare state in western industrialized nations have lead to the questioning of the 'education-work-retirement' lockstep of the life course. On one side of the debate it is argued that the threefold model of the life course is "coming undone" to the extent that, "It is impossible to see one's life course." (Guillemard and Rein 1993:496, Guillemard 1997:455) On the other side of the debate several scholars argue that the life course seems to be becoming simply "longer and fuzzier" (Kohli and Rein 1991:22) or blurred (Mutchler et al. 1997). Somewhere in between, Henretta argues that the life course is uniform but becoming more variable in timing (Henretta 1992). Whatever the precise nature of the changes, most agree that there have been modest adjustments in the temporal order of the stages in the life course and changes in the time spent in the various stages (Chappell et al. 2003).

The debate surrounding the shifts in the life course has spilled over into a new debate about the nature of retirement. Retirement, a social institution invented to insure the orderly exit of workers from the labour force, always has been subject to re-definition in accordance with changing social, economic and political forces. Historically, retirement was first introduced to strengthen workers' attachment to the labour force, while today it is used to weaken attachment to the labour force through early retirement programs.

Certainly, retirement is no longer defined by a single chronological age - most people retire between the ages of 55 and 62, well before the institutionalized age of 65 and well before physical decline. Although retirement is viewed as a right as well as a positive social act, many are forced into retirement against their will due to poor health, caregiving responsibilities, unemployment, or the retirement of a spouse. Retirement is also less likely to be circumscribed by pension policy since growing numbers of people use unemployment and disability programs or early retirement packages as bridges into retirement. The devolution of state responsibility for retirement income to the individual also helps to loosen the ties that bind the individual to the state. Furthermore, retirement does not always constitute a one-time event representing an abrupt transition from work to non-work: retirement can be gradual, it can be part-time and it can involve multiple exits from the labour force or, it may never happen.

These variations, taken together, reflect the transformations in the lockstep order of the life course and in the relations between the individual, government and the employer. In short, the blueprint for retirement is being transformed in such a way that the various pathways in and out of retirement are becoming more and more diverse, while ever more responsibility for retirement income is being transferred from the state to the individual. Because of the recentness of these trends, the implications for the well-being of older people are uncertain. They are also in danger of being ignored, especially in view of the strong support for the retrenchment of the welfare state and the existing, but masked, poverty amongst some groups of older persons (McDonald and Robb 2004).

Surprisingly, we know little about women's retirement in Canada. There is little scholarship on Canadian women's retirement addressing the retirement issues faced by women, notwithstanding the abject poverty many older women experience. What is more, for the first time in history we have a generation of women entering retirement who have worked in the labour force for most of their adult lives and who still have managed to provide the greater part of care for the young and, now, for growing numbers of the very old. The common view is that this is a non-issue because today's and tomorrow's women will have worked all their lives and they will have ample pensions in their old age.

The purpose of this paper, then, is to examine women's retirement in the context of these transformations and the implications for their well-being in old age. The basic argument is that the "new" retirement is women's retirement. Women's retirement has always been different from men's and will continue to be different for the foreseeable future, despite arguments to the contrary. The observation applies to the baby boomers and the generation of younger women following in their wake. Women's transition into retirement and their material conditions in retirement are a direct result of gender relations as they play out in work and the family over the life course. These relations are, in turn, tempered by the intersection of race, class and sexual orientation within a particular historical era.

Although Canadian data is favored in the analyses, the retirement transition and the social and economic processes influencing retirement are similar in most western industrialized nations. The Canadian case is particularly informative since Canada has had one of the largest baby booms in the world, it has had the worst recessions in the last two decades and it still supports mandatory retirement. As well, the ethnic diversity of Canada underscores how retirement is experienced differently by different groups.

A brief history of women's retirement

When Simone de Beauvoir observed in 1949 that women "...have no past, no history.", her statement still rings true today when applied to women's retirement. Retirement and its history has been largely a male preserve until very recent times even though the number of women in the paid labour force has steadily increased throughout the twentieth century. In Canada in 1921, 20.5% of women worked (Urquhart and Buckley 1965:63), while in 1999, 55% of women were in the paid labour force (Statistics Canada 2000:99), and this rate reached 61.6% in 2003 (Statistics Canada 2005). In fact, in most developed countries in the world, the average labour force participation rate of women reached 55% by the early 1990s (The World's Women 1995). The limited, if almost nonexistent history of women's retirement is a direct reflection of the situation that, for much of the twentieth century, women were "invisible" workers and hence, "invisible" retirees (Hesse-Biber and Carter 2000:17).

Women's work has been invisible in most historical contexts because traditional economic measures have usually been restricted to paid employment (Hesse-Biber and Carter 2000, Waring 1999, Nyberg 1994, Bose 1987, Beneria and Roldan 1981). Work done outside of the market economy typically was not recorded as work, although there are a few exceptions.1 Marilyn Waring, using data files from numerous countries in support of this observation, included Canada in her examples. As late as 1986 she shows how the Canadian Census contributed to the disappearance of women from the economically productive population. Question 25(a) of the 1986 Canadian Census asks "Last week, how many hours did you work (not including housework, maintenance, or repairs for your own home)?" (Waring 1999:100) It wasn't until 1996 that the Canadian Census included, for the first time, questions about unpaid housework (Arneil 1999).

Omitting the domestic labour of women in census counts is a longstanding practice in Canada. Since the inception of the Canadian Census in 1871, a "gainful occupation" was the conceptual definition used to collect occupational data until 1951 when it was replaced by the labour force concept. A gainful occupation was defined as:

"...one by which the person who pursues it earns money or in which he assists in the production of marketable goods. children, 14 years and over, assisting parents in the work of the farm or in some family business in a 'No Pay' capacity were considered as gainfully occupied, but daughters assisting with household duties in their own homes without wages were not included..." (Urquhart and Buckley 1965:55).

Not surprisingly, "Housewives ... were excluded by this definition." (Urquhart and Buckley 1965:55)

In addition, much of women's paid work has gone unrecognized because remunerative part-time work, like taking in borders or sewing, selling eggs or produce, or doing housework or childcare for others, was considered to be a natural extension of women's service to their families (Prentice et al. 1996).

The rationale behind the data collection practices of most countries stems from the beliefs underpinning the public/private divide, the distinction drawn between the family on the one hand, and markets and states on the other (O'Connor, Orloff and Shaver 1999, Boyd 1997, Sainsbury 1996). The most common view posited by researchers is that during the transition from an agricultural economy to a capitalist industrial economy, economic and technological transformations wrought major social structural changes in society that contributed to the workplace/family divide. Industrialization separated the home and the workplace by moving industry out of the home and into the factory. This process brought about an increased differentiation in the division of labour by gender to the extent that women's work encompassed the private sphere of reproduction and child rearing and men's work involved the public sphere of economic life (Mies 1998, Eichler 1997, Boris 1996, Valenze 1995, Armstrong and Armstrong 1978, Connelly 1978).2

It is not the purpose of this paper to outline the history of men's and women's retirement in Canada (see McDonald 2001) but to underscore that women's retirement was tied to the breadwinner model of the family, a model embedded in our social security system that lingers on today. The following limited examples from the history of the Canadian pension system support this argument.

Using the military as a metaphor, the first private pensions in Canada in 1870 and 1874 were set up to control male behavior through uniforms, hierarchies and strict regulations in exchange for loyalty. Women, of course, were not 'soldiers of labour' and did not qualify. In reference to the Annuities Act of 1908 where both sexes might pay the same amount into their account, the annual returns were lower for women.3 In 1911, Newfoundland, still a British Colony, introduced the first state-run old age pension program, which, consistent with women's place in the private sphere, did not include them at all (Snell 1993a). Even though historian Struthers (1992) heralds the Old Age Pensions Act of 1927 as the first "gender inclusive" social program in Canada, women were not included in the program on an equal basis when compared to men (Struthers 1992:237). Snell (1993b), an historian, provides a number of examples of how the State, through its administration of the Act was careful to maintain the gendered nature of marriage.4 While the Old Age Security Act and the Old Age Assistance Act in 1951 gave women for the first time their own regular source of income independent of their male relatives, 5 the pension was to be augmented by pensions from work and from personal savings, two forms of income not available to the majority of Canadian women at the time. The Canada/Quebec Pension Plan (C/QPP) that resulted from this thinking about personal contributions in 1965 was not overtly biased against women workers but certain provisions that applied equally to women and men were simply disadvantageous for women. Two obvious examples were that all benefits were linked to the level of contributors' incomes and, secondly, a retirement pension varied according to the length of time a person spent in the labour force (McDonald 2001).

When retirement took permanent root with the introduction of social insurance provided by the Old Age Security (OAS) pension and the C/QPP, women were confronted with an anomaly in the public pension system that made their retirement experience somewhat distinctive. The insurance component of the system presumed equal citizenship for women with the accompanying assumption that women could look after themselves in retirement - despite their enduring socioeconomic inequality in the labour force. Having to turn to the Guaranteed Income Supplement (GIS) for social assistance, many women found their retirement income at the bottom of the income ladder. This untenable position of being equal but not equal because of a different relationship to the labour force is the bedrock upon which women's retirement is built and is the source of the many income disparities women experience today. The argument, then, is that we have a very old nineteenth century model for the pension system while women lead a modern twenty-first century lifestyle.

Modern retirement

Today in Canada, retirement is still the norm for most older workers. According to the 2002 General Social Survey, people aged 45 to 59 cited an average planned retirement age of 60.8 years, while 18.4% of Canadians said they did not intend to retire at all (Schellenberg 2004). Although preferences may easily be changed by unforeseen circumstances, the preferences of Canadians are consistent with their retirement behavior.

As indicated in Chart 10.1, 66% of men aged 55 to 64 remained in the labour force in 2004 compared to 86.5% in 1953. For men 65 years of age and over, the labour force participation rate was 11.8% in 2004 compared to 47.5% in 1947, an amazing drop by any measure. Three observations can be made based on the data for males in Chart 10.1. Declines for men have been greater over age 65, although there has been a decline for all ages; early retirement before age 65 seems to be slowing and, during the 1990s, labour force participation seems to have fluctuated somewhat. The fluctuations may represent such factors as government cutbacks that occurred during this time and corporate downsizing (Kieran 2001, Gower 1997). Looked at another way, the median age of retirement for men in 2004 was 61.8, down from the median age of 65.1 in 1976 shown in Chart 10.2.

Chart 10.1 Labour force participation rates for age groups 55 to 64 and 65 and above, by sex, Canda, 1946 to 2004. Opens a new browser window.

Chart 10.1
Labour force participation rates for age groups 55 to 64 and 65 and above, by sex, Canda, 1946 to 2004.


The pattern for women's labour force participation, as seen in Chart 10.1, appears different from men's because the rates represent two trends: the dramatic increase in women's labour force participation starting in the 1960s and the trend to early retirement among men. The increasing midlife participation of women aged 55 to 64 is evident in the chart. In 1954, only 12.9% of women aged 55 to 64 were in the labour force compared to 47.7% in 2004. In contrast, the labour force participation rates for women 65 years of age and over have never exceeded the rate of 6.3 in 1964 and have remained fairly stable since 1946, suggesting that the two trends offset each other for older women. An examination of women's median age of retirement is more telling. As shown in Chart 10.2 the median age of retirement for women dropped from 64.8 in 1976 to 60.8 in 2004 (Statistics Canada 2005).

Chart 10.2 Average and median retirement age by sex, Canada, 1976 to 2004. Opens a new browser window.

Chart 10.2
Average and median retirement age by sex, Canada, 1976 to 2004.

Chart 10.3 clearly indicates women's labour force patterns in relation to those of men. Beginning in the 1970s each successive cohort of women was in the labour force longer. In light of the larger baby boom cohorts, the participation rates rose dramatically as seen in the cross-sectional rates in Chart 10.3. The decline in the participation rates of men, noted above, served to narrow the gap between the rates of men and women and the participation profile of women has begun to resemble the same high and flat shape of men's rates (Sunter 2001). The flat participation rates of women during the 1990s have been attributed to several factors: the full integration of women into the labour force (Beaudry and Lemieux 1999) and/or that the participation rates of successive generations have become more similar leading to slower growth, although, both observations are speculative (Sunter 2001).

Chart 10.3 Labour force participation rates by age and sex, Canada, 1976 and 2004. Opens a new browser window.

Chart 10.3
Labour force participation rates by age and sex, Canada, 1976 and 2004.

In summary, the decline in men's participation, has, over time, become more and more closely associated with age 62. For women, waged labour has become a central element in their lives, tightening the previous weak link between women's paid work and retirement. Most importantly, the patterns indicate that women's retirement will be different than that of men's retirement. Women's late entry into the labour market means that their individual history with retirement is about to begin and this history is likely to be different. At minimum there will be a structural lag on the retirement front where women will be playing pension "catch-up" to men (Moen 1996).

What's new about the new retirement

The life course has been argued to be more asynchronous (O'Rand and Henretta 1999, Han and Moen 1999) or, more vividly argued, is a "...veritable revolution in the age structure of society" (Riley and Riley 1994). Retirement no longer represents an abrupt transition from work to non-work: it can be gradual, it can involve multiple exits, it is multi-layered with other life events and it may never happen. Han and Moen (1999) capture the essence of the new arguments when they conceptualize retirement as "clocking out". Retirement is a multiplex process governed by multiple institutional schedules and by the diverse pacing of individual biographies which intersect with institutional timetables, all of which lead to variability in the age of retirement. Guillemard (2000) specifically argues that the retirement pension system no longer regulates early retirement in Europe, rather, disability insurance creates the bridge between work and retirement. Gardyn (2000) argues that retirement is being reinvented not only to include second careers but also continuing education and volunteerism.

As would be anticipated, changes in the temporal order of the life course has influenced retirement which has been the gateway to the last phase of life and the conduit that links the institutional structures of work and non-work. It has almost become vogue to talk about the "new retirement", although what is "new" about the "new retirement" is not agreed upon by gerontologists. Gathering up the various strands of Canadian research, what seems different about retirement today is that it is evolving into a process without a clear beginning or end, it is much less likely to be chronologically determined, it is not as tightly regulated by the state through public pensions, by firm policies or by one's relationship to the labour force and there are a multitude of pathways leading to retirement that are interwoven with the lives of others. It almost goes without saying that retirement is embedded in a life course. At the heart of the life course perspective is the synchronization of "individual time" and "historical time, and the cumulative impact of earlier life events as shaped by historical forces on subsequent events" (Hareven 1996:31).

The evidence for these changes is growing, although some of the Canadian data cited below are out of date. It is also difficult to make comparisons to earlier times since data was not collected about some of the "new" routes. While the continued lowering of the age of retirement could still be argued to be tied to public and private pensions, what is interesting is that a full 31% of Canadians reported in 1994 that they did not know at what age they would retire. The group most uncertain was women: 36% of women did not know when they would retire compared to 25% of men. The average planned retirement age of employed persons in this survey was 58.5 years for both sexes, well below the age of 60 when pensions first become available in the C/QPP.

The 2002 General Social Survey (GSS) data show a similar pattern. Among Canadians aged 45 or older who said they were not retired, 20% reported no intention to retire, and another 12% did not know when they would retire.6 Thus, as in 1994, just over 30% declined to indicate when they would retire. However, in 2002 the figures for men and women are closer than they were in 1994. Thus, 29.7% of men say they do not intend to retire or do not know when they will retire compared to 31.2% of women (Schellenberg 2004).

Although the evidence is tenuous because of the lack of available data over time, there does seem to be a wider number of routes into retirement than was available in previous times. At least five "new" routes that have come to the fore in recent times would include retirement via early retirement packages, the use of disability benefits, the need to caregive for a family member, involuntary retirement and the reversal of the retirement decision. In the 1992 Survey of Persons Not in the Labour Force, 26% of men and 12% of women indicated that they had retired unexpectedly because they were offered an early retirement package (McDonald, Donahue and Marshall 2000). The corresponding figures, for those who said they had retired, in the 2002 General Social Survey are 19% for men and 9% for women.

Another example, as in the European experience, is the use of disability benefits as a route into retirement. Disability benefits were originally intended to provide income for those who could not earn a living because they had a severe disability. However, in a number of countries they were used as a vehicle to remove older workers from the labour force in response to high levels of unemployment. The uptake was so sharp that in 1990, recipients of disability benefits outnumbered those in receipt of unemployment benefits in 12 of 23 OECD countries (McDonald and Donahue 2000).

In the 2002 General Social Survey, there are data about retirement because of poor health. The prevalence of disability benefits among those citing this reason for retiring is unknown. However, it is notable that among those who said they had retired, just over one quarter (26% for men, and 25% for women) cited health as a reason for their retirement.

In the 1991 survey of Ageing and Independence, 5% of the women who were still employed identified retiring to caregive as a possible reason for their retirement. However, in the General Social Survey five years later, at least twice as many women (13%) reported retiring to caregive (Monette 1996). More recently, the 2002 General Social Survey results indicate that 12% of women and 6% of men had retired in order to take care of a family member. A 1997 national caregiver survey conducted by the National Alliance for Caregiving and the American Association of Retired Persons (AARP) reported that nearly one-quarter of U.S. households contained a family caregiver for someone 50 years of age and over. Nearly two-thirds of family caregivers worked full- or part-time, while about 4% retired early and 6% gave up work entirely to caregive (National Alliance for Caregiving and the AARP1997, 2004). Looking specifically at caregivers who retired to provide care for an aging relative, McDonald, Donahue and Moore (1998) used Canadian data from the national Survey of Ageing and Independence and compared retired caregivers to other retirees. They found that employees who retired to caregive were more likely to be female, were more likely to be on the margins of the economy and were more likely to retire very early. Furthermore, persons who retired to caregive were women with lower personal incomes, were less likely to receive benefits from the public pension system and were less likely to receive private pensions or to have income from investments.

Some Canadians have also reversed their retirement decision. Depending of the data file used, between 13% (General Social Survey 1994) and 17% (Survey of Ageing and Independence 1991) of Canadians reversed their retirement decision and returned to the labour force, mainly to what have been called "bridge jobs" to retirement (Doerringer 1990). In a small, non random study of a telecommunications company in Canada, almost 47% of male and 25% of female early retirees went back to work (Marshall, Clarke and Ballantyne 2001). Among those who said they had retired in the 2002 General Social Survey, 24% of men and 18% of women reported that they had done some paid work after their first retirement.

Most persons in Canada who returned to work, took up part-time work and they tended to be persons who were forced to initially retire because of mandatory retirement provisions or because of early retirement packages (McDonald 1999). In particular, women who were professionals with a high personal income returned to work. Unlike the men, they had planned for their retirement and retired either because they had access to an early retirement package or because there was no available work. The women's pattern of re-employment seemed, however, to be more a form of career progression than a form of retirement (Hayward et al. 1994).

The 2002 General Social Survey also indicates that the rate of going back to work is much higher than average for those who had retired due to mandatory retirement provisions or because of early retirement packages. In this group, 30% did some paid work after their first retirement, compared to 18% for those who did not state such reasons for retiring.

In the United States, it is now estimated that between 30% and 50% of people move into their "final" retirement via partial retirement, or use "bridge jobs" from their career jobs into retirement (Mutchler et al. 1997, Quinn 1999).

Perhaps even more revealing is the number of Canadians who have no intention of ever retiring. Among both men and women aged 45 or more who said they had not retired in the 2002 General Social Survey, 20% reported they had no intention to retire.

Using the longitudinal Health and Retirement Survey in the United States, Ekerdt et al. (2001) showed that the uncertainty about the form and timing of retirement is substantial. In their analyses, 40% of workers aged 51 to 61 could not state how they would exit their job and about 12% did not know the date or age of their retirement. They provide some evidence that the uncertainty was less likely among those who led a more "socially attended life" meaning a more public life influenced by people on the job, at home and by friends. A socially attended life is one under surveillance by others at work and at home. Work in larger, more bureaucratic organizations that recognize, discuss and control the retirement planning of employees makes retirement more visible than work in self-employment. As well, spouses and friends act as surveillants of each other, so that within these relationships people are more likely to mutually discuss and review retirement planning. The findings in this study indicated that uncertainty in the form and timing of retirement was most likely for women, the unmarried and those not enrolled in pension plans - all factors that could lead to a less "socially attended life".

In terms of the influence of other people on one's retirement (linked lives), most research has been done on marriage and the family. The research considers how a partner takes into account their spouse's career and retirement when making their own decisions to retire. In a study of Canadian married couples, McDonald (1996) found that a married woman's age of retirement was influenced by the age difference between herself and her husband, his level of education and occupational prestige, their household income and whether or not he was retired. If the husband was older than his wife, her age of retirement increased, a finding not supported in the American data (Szinovacz and DeViney 2000). The husband's higher levels of education and occupational prestige and income led to lower retirement ages for the wife. The husband's age of retirement was increased if the wife was the same age or older, had high occupational prestige and a larger household (McDonald 1996a). In contrast, Tompa (1999) found that both men and women tend to make joint retirement decisions with their spouses. He found that the larger the C/QPP or other pension income a spouse received, the more likely an individual was to exit the labour force.

Finally, many Canadians will be retired involuntarily. In earlier times, involuntary retirement for some was an expected event in the form of mandatory retirement and is still challenged in the legal system today (Gillin, MacGregor and Klassen 2005, Klassen and Gillin 1999). Most recently, there is a trend in Canada to do away with mandatory retirement. For example, the Ontario Government introduced legislation in June 2005 to end mandatory retirement after extensive consultation across the province. It is therefore not surprising that only about 14% of people are affected by these policies compared to 27% of retirees in 1975. Today involuntary retirement is a hidden issue and is probably higher than what is reported in the current research - about 27% of men and 22% of women retire involuntarily (McDonald, Donahue and Marshall 2000). In the 2002 General Social Survey, 26% of those who said they had retired reported having an involuntary retirement.

Of the one in four Canadians who retired involuntarily, as reported by McDonald et al., the primary reasons are poor health first, followed by mandatory retirement and unemployment (McDonald, Donahue and Marshall 2000). In the 2002 General Social Survey nearly one-half (49%) of those who reported involuntary retirement cited health as a reason for retiring. The involuntary nature of retirement can go undetected by some retirees until after the fact. In a study of caregivers it was found that most did not even realize that they were being forced into premature retirement until after the caregiving was over when they attempted to re-enter the labour force and were unsuccessful (McDonald, Sussman and Donahue 2002).

Although there is preliminary evidence that the very nature of retirement is changing, only time will tell the magnitude and the permanence of these changes. A group of Canadian scholars have stated that, ".Statistics Canada is urged to consider that the terms retirement and retired are no longer valid for understanding a person's paid labour force status." (Payne et al. 1996) However, more than 80% of Canadians currently remain retired, so their suggestion may be somewhat premature. Recent evidence supports the idea that the institutionalization of retirement is still strong, at least in the United States. One study found that retirement anticipation was normative among older workers (Ekerdt, Kosloski and DeViney 2000) and another found that institutionalized retirement criteria have been strongly internalized to anchor people's self-definition as retiree (Szinovacz and DeViney 2000).

Parallels between the "new retirement" and women's retirement

It could be argued that the transformation in retirement that we are witnessing today, is at bottom, women's retirement. In the past, women's retirement was amorphous and fluid and is likely to continue to be so in the future but for different reasons. It also could be argued that, baby boomer women will have an important role in establishing the framework for the "new retirement" (Statistics Canada 2000). The basic principles of this argument rest on the fact that the baby boomers are the first female cohort whose labour force participation will span most of their adult life. As Galarneau (1994:6) describes, in 1971, 54% of the first wave of baby boom women (those born between 1946 and 1955), then aged 16 to 25, were in the labour force. This compared with 39% of all women that year. Ten years later 70% of the second-wave baby boom women (aged 16 to 25) were in the labour force. These women will have the option for the first time of defining their own retirement, at least in terms of a stronger presence in the labour force, their higher levels of educational attainment and their increased occupational prestige. Because of the demographic changes in the Canadian population (older women constituted 57.4% of the population in 1998 and this number will not drop below 56% until after the year 2041), their very presence is likely to have some effect on the retirement.

The question then, is will the baby boomer women's retirement be different from their mother's? The answer is probably yes. The argument is made here that women's retirement, starting in the immediate future, is more likely to look like the "new retirement" for three reasons: the declining importance of marriage; double caretaking and women's work is still women's work. These factors, taken together, will make women's retirement somewhat amorphous, uncertain and irregular.

The decline in the prevalence and stability of marriage

Baby boom women increasingly postponed marriage. This is evident when baby boom women are divided into two waves: the first wave, born between 1946 and 1955 and the second wave, born between 1956 and 1965 (Galarneau 1994). At ages 26 to 35, 20% of second wave women had never married compared with 14% of the first wave and 11% of pre-boom women (those born between 1936 and 1945) at the same age (Galarneau 1994). They also distinguished themselves by having fewer children. Just 13% of second wave women aged 26 to 35 had three or more children compared with 16% for the first wave and 34% of pre-boomers.

These trends set by the boomers will continue. There has been a long-term decline in the proportion of women who are spouses in a husband-wife family because the rate of marriage is down to 20% in 1997 from 24% in 1989. Also in 1997, the average age at first marriage for brides was 28 years, up from 26 years in 1990. What is more, the long-term incidence of divorce in Canada (although most of the rise can be attributed to changes in legislation in 1968 and 1986) has risen and is much higher compared to the 1960s. In 1997 there were 225 divorces per 100,000 of the population. In turn, there is continued growth in the number of women who are lone parents. In 1996, female-headed lone-parent families represented 19% of all families with children and was up from 10% in 1971 (Statistics Canada 2000). As well, recent generations of women are engaging in more common-law unions. More than 40% of women ranging from ages 30 to 39 are expected to choose a common-law union as their first union, while this percentage is expected to reach 53% among 20 to 29 year olds. With common-law relationships there are fewer legal protections and many unions suffer from instability (Statistics Canada 2000).

Taken together, these factors mean that marriage will not be quite as central to women's retirement as it was in the past, an enormous change since the inception of retirement when women's work/retirement was totally embedded in the family. It is well known in the retirement literature in Canada that, for women, marriage is the most important predictor of age of retirement since it determines income in retirement and satisfaction with retirement (McDonald 1996a, McDonald and Robb 2004). For example, marriage has been shown to be a major economic resource that buffers women against poverty in retirement. When marital support disappears, women's secondary poverty becomes all too evident in retirement (Logue 1991). One national study in Canada found that 49% of retired widows, 53% of divorced or separated women and 28% of ever single women lived below the Low Income Cut-offs for Canada compared to 15% of married women (McDonald, Donahue and Moore 1997). The finding that married and single women are more likely to find their income in retirement satisfactory compared to divorced and widowed women comes as no surprise. In a more recent study of unattached women it was found that the separated and the divorced are the poorest of all older unattached women (McDonald and Robb 2004). The whole uncertainty around marital status and the tendency of women "to-go-it-alone" is more than likely to introduce all kinds of unforeseen challenges and uncertainties into the retirement process.

Women will become double caregivers

As Canadian society ages, an emerging issue is the problem of double caregiving - a parent looking after their own children and also caring for their aging parents. Women still have children. Almost two-thirds of women who have ever worked have had a work interruption in paid work of six months or more. Most women on maternity leave (86%) return to work within one year after giving birth. However, one study has shown that after a work interruption, women are not as likely to return to the same job or a full-time job if they had a full-time job prior to their interruption (Fast and Da Pont 1997). In 1998, women employed full-time with children spent 1.7 hours more on unpaid work (about 5 hours per day) and leisure activities than women without children (Statistics Canada 2000). Another study found that, if a labour force adjustment must be made to accommodate family responsibilities, it is usually women who make the adjustment (Townson 2000). Of the men and women with family responsibilities who made adjustments in 1995, two-thirds of women adjusted their work schedules compared to about one third of men. The adjustments included part-time work, irregular schedules, voluntary job absences and voluntary joblessness.

Add to this that there is no historical precedence for the experience of most middle-aged and older women having living parents. In 1940, one in three 50-year-old women had living parents compared to two in three, or double that number, in 1980. In the face of a growing demand for the care of aging parents and relatives, women are also more likely to provide care: 15% of women, compared to 9% of men cared for both children and an aging relative in 1996. The data from one study suggests that employed women provide elder care by carrying out some of their paid work at home and by working in off hours (Marshall 1998). Adding to the instability of women's marital status and all that this implies, the double caregiving aspect of their lives means more time out of the paid labour force, more exits and entrances from the labour force and, as seen in a study of caregivers, many unexpected exits from the labour force with little hope of returning (McDonald, Sussman and Donahue 2002).

Women's work is still women's work

Today, many economic analysts argue that the next generation of women, namely the baby boomers, will have been in paid employment for most of their lives and will have their own private pensions, RRSPs and most will receive C/QPP benefits. While all of this may be true, there are at least two overriding factors that suggest that their pension incomes will still not be equal to men's pension income in old age. To the degree that the private and semiprivate (C/QPP) components of the pension system replicate the inequality in the labour market and as long as women have interrupted work histories due to family responsibilities as noted above, their ability to save and accumulate pension benefits will be affected.

Some of the more obvious patterns of women's labour force participation that will affect their pensions in the future include women's concentration in non-standard and part-time work, their under-representation in unions, their over-representation in the services sector and the continued distribution of their occupations in female employment. The number of women working in non-standard work, including temporary workers, part-time workers, self-employed workers and multiple job holders, has increased from 35% in 1989 to 41% in 1999. Part-time employment continues to be the most common form of non-standard work for women. Since the middle of the 1970s, 7 in 10 women have worked part-time even though a quarter of these women reported that they would prefer full-time work. In 1995, 43% of part-time workers earned less than $7.50 per hour, and less than 20% of part-time jobs were covered by a registered retirement pension plan. Close to 15% of part-time jobs fell below the basic exemption of $3,500 in the C/QPP.

Only 31% of women are covered by unions, a factor that has been directly linked to decreased pension coverage. In terms of the distribution of women's occupations, 70% of employed women in 1999 continued to work in teaching, nursing and related health occupations, clerical or other administrative jobs and sales and services occupations, down from 78% in 1982. It is well known that female occupations traditionally attract lower wages. Women still earn less for every dollar earned by men. The female-to-male earnings ratio in 1997 was 81% for those aged 15 to 24; 76% for women aged 25 to 34; 73% for those 35 to 44; 70 cents on the dollar, for those aged 45 to 54 and 66 cents among those aged 55 and over. The female-to-male average earnings ratio for full-time full-year workers was 72.5 in 1997 and shows signs of dropping since the ratio was 73.1 in 1995 and 73.4 in 1996. When all is said and done, it comes as no surprise that the average income of women in 1997 was $19,874 compared to $30,169 for men (Statistics Canada 2000).

Tomorrow's older women may have their own pensions but their position will be only marginally improved because RPPs, the C/QPP, RRSPs and savings all depend on earnings and length of time in the labour market. Unless these conditions are modified, a change in the size of women's benefits will be small. To add to women's problems, they outlive men so their private pensions are more likely to be subject to inflation in the long run.

The mismatch between the lives of women and the pension system

The outcome of many of the changes highlighted above, portend a retirement that looks a lot like a version of the "new retirement." Women's lives already are, and will continue to be, typified by multiple transitions in and out of the labour force with employment in more than one job that is usually a contingent, nonstandard job, and their work will be segregated in the feminized sector of the economy where they are likely to receive lower wages. They are also likely to follow new routes into retirement especially to caregive and are likely to retire several times.

These irregular work patterns certainly do not match the structure of the current pension system which was developed to meet the needs of the nineteenth century male worker who was a lone breadwinner supporting a large family. The pension system today, with its emphasis on job tenure in a life-long career job, excludes many Canadians who do not match this profile. Specifically, pension policy barely recognizes the burden of institutionalized lower earnings for women or the costs of their unpaid work and ignores the multiple job changes that women have experienced in the last ten years. At its best, the pension system offers social welfare to the poor in the guise of OAS and GIS, which, in the final analysis, does little to relieve their poverty. While the C/QPP has great potential to accommodate women's lives, the size of benefits are small and have become even smaller with the adjustments made in 1998. Any improvements that are made to public or private pensions would have to start with a new vision of the life course in order to reflect the actual lives of Canadian women, not to mention men who are on the margins of the economy (Marshall 2002).

As I have said before

"...why do policy makers persist in making policy according to an orderly life course of education, work and retirement founded on the experience of the 19th century male industrial worker? Why are the changes in the life course being ignored? Why is social policy still being forged on an ad hoc basis? Why do we attempt to build social policy one generation at a time?" (McDonald 1997:395)

Bibliography

Armstrong, Pat and Hugh Armstrong. 1978. The Double Ghetto: Canadian Women and Their Segregated Work. Toronto. McClelland and Stewart.

Arneil, Barbara. 1999. Politics and Feminism. Oxford. Blackwell Publishers Ltd.

Beaudry, Paul and Thomas Lemieux. 1999. Evolution of the Female Labour Force Participation Rate in Canada. CIRANO Project Reports 1999rp to 02. CIRANO.

Beneria, Lourdes and Martha Roldan. 1981. The Crossroads of Class and Gender. Chicago, IL. University of Chicago Press.

Boris, Eileen. 1996. "Sexual divisions, gender constructions." Pp. 19 to37 in Homeworkers in Global Perspective. E. Boris and E. Prugl (eds.). New York, NY. Routledge.

Bose, Christine. 1987. "Devaluing women's work: The undercount of women's employment in 1900 and 1980." Pp. 95 to 115 in Hidden Aspects of Women's Work. C. Bose, R. Feldberg and N. Sokoloff (eds.). New York, NY. Praeger.

Boyd, Susan (ed.). 1997. Challenging the Public/Private Divide: Feminism, Law, and Public Policy. Toronto. University of Toronto Press.

Chappell, Neena, Ellen Gee, Lynn McDonald and Michael Stones. 2003. Aging in Contemporary Canada. Toronto. Prentice Hall.

Clark, R.M. 1960. Economic Security for the Aged in the United States and Canada. A Report Prepared by the Government of Canada, Vol.I, II. Ottawa, Ont. The Queen's Printer and Controller of Stationary.

Cohen, Marjorie. 1988. Women's Work, Markets and Economic Development in Nineteenth-Century Ontario. Toronto. University of Toronto Press.

Connelly, Patricia. 1978. Last Hired, First Fired: Women in the Canadian Workforce. Toronto. Women's Press.

de Beauvoir, Simoe. 1968. The Second Sex. 6th edition. H.M. Parshley (ed.). New York, NY. Alfred A. Knopf.

Doeringer, Peter. 1990. Bridges to Retirement: Older Workers in a Changing Labor Market. Ithaca, NY. ILR Press.

Eichler, Margrit. 1997. Family Shifts: Families, Politics, and Gender Equality. Toronto. Oxford University Press.

Ekerdt, David J., Karl Kosloski and Stanley DeViney. 2000. "The normative anticipation of retirement by older workers." Research on Aging. 22: 3 to 22.

Ekerdt, David J., Jennifer Hackney, Karl Kosloski and Stanley DeViney. 2001. "Eddies in the stream: The prevelance of uncertain plans for retirement." Journal of Gerontology: Social Sciences. 56(B): S162 to S170.

Errington, Jane. 1995. Wives and Mothers, School Mistresses and Scullery Maids: Working Women in Upper Canada 1790 to 1840. Montreal and Kingston. McGill-Queen's University Press.

Fast, Janet and Moreno Da Pont. 1997. "Changes in women's work continuity." Canadian Social Trends. 46, Autumn: 2 to 7. Statistics Canada Catalogue no. 11-008-X1997 0023178.

Galarneau, D. 1994. "Baby boom women." Perspectives on Labour and Income. 6, 4. Statistics Canada Catalogue no. 75-001-XPE.

Gardyn, Rebecca. 2000. "Retirement Redefined." American Demographics. 22, 11: 52.

Gillin, C.T., David MacGregor and Thomas R. Klassen (eds.). 2005. Time's Up: Mandatory Retirement in Canada. Toronto. CAUT/Lorimer Publishing.

Gower, David. 1997. "Measuring the age of retirement." Perspectives on Labour and Income. 99, 99. Statistics Canada Catalogue no. 75-001-XPE.

Guillemard, Anne Marie. 1997. "Rewriting social policy and changes in the life-course organization: A European perspective." Canadian Journal on Aging. 16, 3: 441 to 464.

Guillemard, Anne Marie. 2000. Aging and the Welfare State Crisis. Cranbury, New Jersey. University of Delaware Press.

Guillemard, Anne Marie and Martin Rein. 1993. "Comparative patterns in retirement: Recent trends in developed societies." Annual Review of Sociology. 19: 469 to 503.

Hakim, Catherine. 1996. "Key issues in women's work: Female heterogeneity and the polarization of women's employment." Conflict and Change in Britain Series - A New Audit, 4. London, UK. Athlone Press.

Han, Shin-Kap and Phyllis Moen. 1999. "Work and family over time: A life course approach." The Annals of American Academy of Political and Social Sciences. 562: 98 to110.

Hareven, Tamara K. 1996. Aging and generational relations: life-course and cross-cultural perpectives. Hawthorne, NY. Aldine De Gruyter.

Hayward, Mark D., Eileen M. Crimmins and Linda Wray. 1994. "The relationship between retirement life cycle changes and older men's labor force participation rates." Journal of Gerontology: Social Sciences. 49: S219 to S230.

Henretta, John C. 1992. "Uniformity and diversity: Life course institutionalization and late life work exit." The Sociological Quarterly. 33: 265 to 279.

Hesse-Biber, Sharlene and Gregg Carter. 2000. Working Women in America: Split Dreams. New York, NY. Oxford.

Kieran, Patrick. 2001. "Early retirement trends." Perspectives on Labour and Income. 13, 4. Statistics Canada Catalogue no. 75-001-XPE.

Klassen, Thomas R. and C.T. Gillin. 1999. "The heavy hand of the law: The Canadian Supreme Court and mandatory retirement." Canadian Journal on Aging. 18, 2: 259 to 276.

Kohli, Martin and Martin Rein. 1991. "The changing balance of work and retirement." Pp. 1 to 35 in Time for Retirement: Comparative Studies of Early Exit From the Labor Force. M. Kohli, M. Rein, A.M. Guillemard and H. van Hunsteren (ed.). New York. Cambridge University Press.

Logue, B.J. 1991. "Women at risk: Predictors of financial stress for retired women workers." The Gerontologist. 31, 5: 657 to 665.

Marshall, Katherine. 2002. "Duration of multiple job holding." Perspectives on Labour and Income. 14, 1. Statistics Canada Catalogue no. 75-001-XPE.

Marshall, Katherine. 1998. "Balancing work and family." In Working Arrangements in the 1990s. Ottawa. Statistics Canada.

Marshall, Victor W., Phillipa Clarke and Peri Ballantyne. 2001. "Instability in the retirement transition. Effects on health and well being in a Canadian study." Research on Aging. 23, 4: 379 to 409.

McDonald, Lynn. 2001. Transitions into Retirement. Canadian Labour Congress. Ottawa.

McDonald, Lynn. 1999. The New Retirement. Canadian Association on Gerontology. Ottawa.

McDonald, Lynn. 1997. "The link between social research and social policy options: Reverse retirement as a case in point." Canadian Journal on Aging /Public Policy. 1997 Supplement: 90 to 113.

McDonald, Lynn. 1996a. Transitions into Retirement: A Time for retirement. Toronto, Ont. Centre for Applied Social Research.

McDonald, Lynn. 1996b. The Casualties of Canada's Dualist Pension Policy. In Sociology of Aging. V. Minichiello, N. Chappell, A. Walker and H. Kendig (eds.). The International Sociological Association.

McDonald, Lynn and A. Leslie Robb. 2004. The Economic Legacy of Divorced and Separated Women in Old Age. Hamilton, Ont. SEDAP.

McDonald, Lynn and Peter Donahue. 2000. "Poor health and retirement income: The Canadian case." Ageing and Society. 20: 493 to 522.

McDonald, Lynn, Peter Donahue and Brooke Moore. 1998. The Economic Casualties of Retiring to Caregive. Program for Research on the Independence and Economic Security of the Older Population. Research Paper No. 28. McMaster University.

McDonald, Lynn, Peter Donahue and Brooke Moore. 1997. Widowhood and Retirement: Women on the Margin. Toronto. University of Toronto. Centre for Applied Social Research, Faculty of Social Work.

McDonald, Lynn, Peter Donahue and Victor Marshall. 2000. "The economic consequences of unexpected early retirement." In Independence and Economic Security in Old Age. F.T. Denton, D. Fretz and B.G. Spencer (ed.). Vancouver, BC. University of British Columbia Press.

McDonald, L., T. Sussman and P. Donahue. 2002. The Economic Consequences When the Caregiving is Over. Canadian Association on Gerontology 31st Annual Scientific and Educational Meeting. October 24 to 26. Montreal, Quebec.

Mies, Maria. 1998. Patriarchy and Accumulation on a World Scale: Women in the International Division of Labour. New York, NY. Zed Books.

Moen, Phyllis. 1996. "Gender, age, and the life course." Pp. 171 to 187 in Handbook of Aging and the Social Sciences. 4th edition. R.H. Binstock and L.K. George (ed.). New York, NY. Academic Press.

Monette, Manon. 1996. Canada's Changing Retirement Patterns: Findings from the General Social Survey. Statistics Canada. Ottawa. Ministry of Industry.

Mutchler, Jan E., Jeffrey A. Burr, Amy Pienta and Michael P. Massagli. 1997. "Pathways to labor force exit: Work transitions and work instability." Journal of Gerontology: Social Sciences. 52B: S4-S12.

National Alliance for Caregiving and AARP. 2004. Caregiving in the U.S. Bethesda, MD. MetLife Foundation.

National Alliance for Caregiving and AARP. 1997. Family Caregiving in the U.S.: Findings from a National Survey. Bethesda, MD. MetLife Foundation.

Nyberg, Anita. 1994. "The social construction of married women's labour force participation: The case of Sweden in the twentieth century." Continuity and Change. 9: 145 to 156.

O'Connor, Julie, Ann Orloff and Sheila Shaver. 1999. States, Markets, Families: Gender, Liberalism and Social Policy in Australia, Canada, Great Britain and the United States. Cambridge, MA. Cambridge University Press.

O'Rand, Angela and John C. Henretta. 1999. Age and Inequality: Diverse Pathways Through Later Life. Boulder, CO. Westview Press.

Payne, Barbara J. et al. 1996. "Healthy aging: Insights for research and policy." Canadian Public Policy. 23, 1: 42 to 52.

Prentice, Alison et al. 1996. Canadian Women: A History. 2nd Edition. Toronto. Harcourt Brace and Co., Canada.

Quinn, Joseph. 1999. "Criteria for social security reform." In Prospects for Social Security Reform. Olivia Mitchell, Robert Myers and Howard Young (ed.). Philadelphia, PA. University of Pennsylvania Press.

Riley, J.W. and Jr. Riley, M.W. 1994. "Beyond productive aging: Changing lives and changing structures." Ageing International. June: 15 to 19.

Sainsbury, Diane. 1996. Gender, Equality, and Welfare States. Cambridge, MA. Cambridge University Press.

Schellenberg, G. 2004. The Retirement Plans and Expectations of Non-Retired Canadians Aged 45 to 59. Business and Labour Market Analysis Division. Catalogue no. 11F0019MIF2004223. Statistics Canada. Ottawa

Snell, James G. 1993a. "Maintenance of the Newfoundland old age pension programme, 1911 to 1949." Acadiensis. 23, 1: 86 to 109.

Snell, James G. 1993b. "The gendered construction of elderly marriage, 1900 to 1950." Canadian Journal of Aging. 12, 4: 509 to 523.

Statistics Canada. 2005. Table 282-002 - Labour Force Survey estimates (LFS), by Sex and Detailed Age Group, Annual. (accessed July 27, 2005).

Statistics Canada. 2001. Census of Canada 2001: Visible Minority Groups, Sex, Age Groups for Population, for Canada, Provinces, Territories, Census Metropolitan Areas and Census Agglomerations. Catalogue no. 95F0363X2001004. Ottawa.

Statistics Canada. 2000. Women in Canada : A Gender-Based Statistical Report. Catalogue no. 89-503-XPE. Ottawa.

Statistics Canada.1999. A Portrait of Seniors in Canada. Catalogue no. 89-519-XPE. Ottawa.

Statistics Canada. 1994. General Social Survey. Catalogue no. 11-612-MPE. Ottawa.

Statistics Canada. 1991. Survey of Ageing and Independence. Catalogue no. 91M0002XDB. Ottawa.

Strong-Boag, Veronica J. 1993. The New Day Recalled: Lives of Girls and Women in English Canada 1919 to 1939. Mississauga, Ont. Copp Clark Pitman.

Struthers, James. 1992. "Regulating the elderly: Old age pensions and the formation of a pension bureaucracy in Ontario, 1929 to1945." Journal of the Canadian Historical Association. 3:235 to 255.

Sunter, Deborah. 2001. "Demography and the labour market." Perspectives on Labour and Income.13, 1: 28 to 39. Statistics Canada Catalogue no. 75-001-XPE.

Szinovacz, Maximiliane E. and Stanley DeViney. 2000. "Marital characteristics and retirement decisions." Research on Aging. 22, 5: 470 to 498.

The World's Women 1995. (The) World's Women 1995: Trends and Statistics. New York, NY. United Nations.

Tompa, Emile. 1999. Transitions to Retirement: Determinants of Age of Social Security Take Up. Research Paper No. 6. Social and Economic Dimensions of an Aging Population (SEDAP).

Townson, Monica. 2000. Reducing Poverty among Older Women: The Potential of Retirement Incomes Policies. Ottawa, Ont. Status of Women Canada.

Urquhart, M. and K. Buckley (eds.). 1965. Historical Statistics of Canada. Toronto. The MacMillan Company of Canada Ltd.

Valenze, Deborah. 1995. The First Industrial Woman. Oxford. Oxford University Press.

Waring, Marilyn. 1999. Counting for Nothing: What Men Value and Women are Worth. 2nd Edition. Toronto. University of Toronto Press.


Notes

  1. Using Britain as a case study, Hakim (1996) argues that women's market work is not undercounted due to relative under-reporting and that women's domestic work is not undervalued because it is measured. In Brtitain full-time unpaid domestic work was listed as an occupation in the census classification up until 1871.
  2. Marjorie Griffin Cohen disagrees with this analysis based on her study of women and diary production in Ontario and Quebec in the second half of the nineteenth and the early part of the twentieth centruey. She argues that the division of labour betweeen males and females was well-defined in the pre-industrial stage in Ontario. The characteristics of a staples and export market economy were such that the volatile nature of exports markets, the general scarcity of labour, and the underdeveloped nature of the domestic market made the activitiy of women in the household sector critical to staple development and critical to capital accumulation (Cohen 1988). Her view has been challenged by historian Jane Errington who has argued that often theere was little distinction between production for the household and production fo the market during this time (Errington 1995:83 to 84).
  3. On the maximum total annuity of $1000, a man aged 70 received $135.50 while a women would receive $121.20 (Strong-Boag 1993:188).
  4. In another example in Manitoba in the 1930s it clearly was an affront to the husband as breadwinner if the wife was paid a larger benefit than the husband. The policy basically was, "...Manitoba grants that the wife is eligible, for a higher pension, on account of reduced assets, but cannot see any other alternative, but to keep the wife's pension at the same rate as the husbands." (Quoted in Snell 1993b:514)
  5. While many retired men added their private pensions and their savings from their paid labour to the flat-rate pension, women were solely dependent on the meager flat-rate pension. The stratification that resulted (e.g., in 1957 to 58 a proportion of 128.8 females to 100 males were on social assistance (Clark 1960:265), is a precursor to the situation today with more women depending on Old Age Security (OAS) pension and the Guaranteed Income Supplement (GIS).
  6. Data from the 2002 General Social Survey were computed at Statistics Canada in the Unpaid Work Analysis Division.