Chapter 9 Gender and family - Major dimensions of retirement research

Retirement is a concept derived from labour force participation and therefore does not capture activity outside of the paid labour force. Historically, women were essentially invisible workers and therefore invisible retirees because they did not work for very long in the paid labour force. During the industrialization of Canada, work and retirement became firmly linked with the public sphere for men while women's retirement was buried in the recesses of the family. Women's work encompassed the private sphere of reproduction and child rearing and men's work involved the public sphere of economic life.

Single women who worked in the paid labour force "retired" into unpaid labour in marriage and married women "retired" from the unpaid labour of marriage into widowhood and/or back into the paid labour force, while some women worked all their lives on the margins of the economy for women's wages as unacknowledged breadwinners. Thus, women's retirement was originally anchored in the family wherein women's welfare in old age was dependent on the nineteenth century male model of retirement - that of an economically dependent wife supported by a breadwinner with a pension.

It comes as no surprise then, that the study of women's retirement took a secondary place to men's retirement. Because of the strong attachment of men to the labour force, retirement was solely a male preserve. At the outset, women were usually slipped into the retirement research as part of their husbands' retirement. At best, women were considered retired if their spouse retired and the retirement decision usually involved an evaluation of the current and future largesse of the husband's pension.

Beginning in the 1960s, the growing participation of women in the labour force helped to shift the focus of the research to an interest in the attributes of retired women that lasted through the 1980s. Perspective on women's retirement at this time hinged on the impact of family and spouse on women's retirement and compared retired women and homemakers since the main sphere for women was the home and family and retired women were seen as an anomaly.

By the mid 1990s, it is likely that the impressive labour force participation rates of baby boom women, their continuous labour force attachment and inevitable retirement, triggered a move to directly apply the male model of retirement to women. Researchers studied separate subsamples of men and women with a new focus on women's employment history, occupation, industry, job tenure, wages etc. However, the harder researchers tried to shoe-horn women into a male retirement model based on paid labour force participation, the less successful they were. This individualist approach based on economic participation ignored context and did very little to explain much of women's retirement.

The thinking at the time was that women's and men's retirement were beginning to converge and that women were headed toward a comfortable retirement - a far cry from the poverty they experienced in earlier decades. Related to this view was the startling realization that the work and retirement patterns of baby boom women would influence the number of children they had and, as a result, the size and composition of the future labour force; their enhanced work experience and higher levels of education would be a valuable source of labour that could influence labour supply practices and policies and, because of their lengthy work history, they were likely to make demands on the pension system never made before. Gender started to matter insofar as it was recognized that women might have some impact on retirement, at least in male terms.

While these observations were correct, the model behind it was flawed. The male model of retirement with its emphasis on individual, rationale choice, founded on a one job, continuous work history with pension trade-offs made no sense for women because it was lop-sided. The model ignored gender differences in work and the fact that the majority of men and women live or have lived in some type of family arrangement. The model also did not recognize that to understand retirement, a life course perspective is required. A new model that included gender and family was needed that captured the interdependence between men and women and their families over time.

Many researchers today would agree that retirement is a complex transition that reflects linked lives, historical and political events across the life course and changes in family structures as people grow older. Most would also recognize that a broader framework with new theoretical underpinnings is long over due. At minimum, a broader approach to gender relationships that addresses the relationships between men and women and the totality of these relationships as manifested in societal institutions such as the family, law, welfare states and employment regimes is necessary.

It is to these matters that researchers turn in the second section of this book. The four chapters address the issues of gender and family in retirement and initiate arguments that underscore the problems with a male model of retirement.

Lynn McDonald briefly sets the stage for a discussion of gender and family by tracing the history of the male model of retirement in Canada and how it has never applied to women, especially in light of changes in the labour force patterns of men and women over time. The data portends the possibility of a "new retirement" wherein retirement no longer represents an abrupt transition from work to non-work. Retirement today is evolving into a process without a clear beginning or end, it is much less likely to be chronologically determined, and it is not as tightly regulated by the state through public pensions, by firm policies or by one's relationship to the labour force. Furthermore, there are a multitude of pathways leading to retirement that are interwoven with the lives of others. It is argued that this transformation in retirement is actually women's retirement.

Baby boom women, who will be the first female cohort to participate fully in the labour force over most of their adult life, will help define the parameters of this new retirement for three reasons. Taken together, the declining prevalence of marriage, the need for double caretaking and the fact women's work is still women's work will make women's retirement somewhat amorphous, uncertain and irregular. Being in the majority, baby boom women are likely to question the inadequacies of policies based on a male model of retirement and be pioneers in creating new forms of retirement.

Maximiliane Szinovacz picks up the family thread and the need for a new theoretical perspective by providing a model of family-retirement linkages which is grounded in a life-course perspective. The model emphasizes the importance of different contexts such as work, marriage and family, peoples' experiences in various roles and how these contexts are interdependent. At the same time, retirement transitions are affected by family members through their linked lives and by changes in other life spheres according to when they occur and in what sequence. Research about how families affect retirement is considered from the perspective of marital and family contexts with the proviso that the impacts often vary by gender and race or ethnicity. She goes on to review the less voluminous research on how retirement affects families in terms of marital and family relationships. The links between retirement policies and families are identified as the shifting of economic responsibility for older adults from the family to the state, by providing financial security for disabled workers and survivors and by incorporating regulations that favour specific types of family relationships. Family behaviours have, in turn, consequences for old age social security policies.

Having provided examples of the multiple interdependencies in the retirement transition, the author identifies what research is needed next about families and retirement. In this capacity she expands the idea of the family beyond the couple to include the widowed and divorced, she suggests further explorations of the accumulation of life events (illness, death) around the retirement decision and the interplay between old age security polices and their regulations and the needs of families.

Grant Schellenberg, Martin Turcotte and Bali Ram provide evidence that the interdependence of men and women in dual-earner couples is fast changing, heralding new relationships that affect the retirement decision. They ask, "What are the correlates of the intentions and experiences of joint retirement in Canada for dual-earner couples approaching retirement (aged 45 to 54)?" They suspect that women's increased labour force participation, the gains they have made in their contribution to household income, their growth in pension coverage and increased involvement in managerial and professional occupations will influence how couples approach joint retirement. Using multiple data sources from Statistics Canada, they make their case about these changes in the characteristics of women employees and show how the changing labour market characteristics of these women have altered the characteristics of couples approaching retirement. The outcome of these changes is that in Canada we have a cohort of well-educated dual-earner couples with high incomes on the verge of the retirement transition. Given these improvements, the researchers observe that dual-career couples will now have to make two retirement decisions instead of one. Indeed, almost one half of married individuals now approaching retirement say they intend to retire at the same time as their spouse, although only about one-third of married couples manage to achieve this coordination. Using the General Social Survey from 2002 they find that the factors that could be associated with the economic independence of women are not correlated with expectations of joint retirement and that more economically independent women may view the timing of their retirement as an event independent of their spouse.

In the last paper in this section, Nathalie Deschênes and Leroy Stone recognize the significance of two dimensions of the retirement transition frequently overlooked: the unavoidable fact that the retirement transition is a process and the importance of measuring this retirement process. Using the Survey of Labour and Income Dynamics, the researchers explore the differences in the retirement transitions of women and men focussing on the different processes involved in reaching the status of retirement from the initiation of retirement until formal retirement or to the end of the observation period. They separate the sample into completed and incomplete retirement trajectories that are voluntary or involuntary. Three hypotheses are proffered. It is predicted that women retire at a more rapid pace than men; of the involuntary retirees, men are more likely to still be in the labour force at the end of the observation period than are women and, finally, family and household factors are predicted to be more influential in achieving retirement for women than for men. While the first two hypothesis are supported - women have greater odds of being retired within the first year and involuntary male retirees are still in the labour force at the end of the observation period compared to involuntary women - it is surprising that the last hypothesis is not supported. There is no question that gender matters in the retirement transition since the male/female trajectories are different, however, family characteristics like marriage, having a child or parent in the home or a person who is dependent, does not figure in the actual transition process which can take up to four years. Although there may be several explanations for this finding, what is clear is that we know very little about the actual process of the transition into retirement for men or women.

Together these papers signal that change is afoot in the study of retirement. If we pull all the threads together from the previous papers, at least four observations can made; gender is a crucial aspect of the study of retirement simply because the majority of women work; people who retire live in families where the interdependence of people and events influence retirement and vice versa; the study of retirement is in the state of transformation where dated ideas about retirement coexist with new ideas; and scholars in retirement research need to move forward to develop more complex models of retirement that cover the exigencies of ordinary people, both male and female and their interdependence.

In the first case McDonald sketches the conditions of work and retirement of baby boom women, while Szinovacz's review of the literature on families and retirement show significant differences by gender. Schellenberg, Turcotte and Ram follow up with a demonstration of how the advanced economic position of women in dual-earner couples affects the joint retirement decision of couples. Deschênes and Stone provide evidence that the actual transition into retirement differs according to gender. Although these papers provide only a brief introduction to the "new retirement" literature they are compelling about the value of studying gender as vital to understanding retirement transitions.

To think that the first text devoted to families and retirement came out as late as 1992 is curious because it highlights the tenacity of researchers to cling to the idea that retirement is only about economic security and individual processes of adaptation. Szinovacz, a true pioneer in the study of the family and retirement, unmistakably shows the effect of the family on retirement and the effect of retirement on the family. The marital relationship as a form of family structure is to be ignored at one's peril in the coordination of joint retirement. Failure to recognize that husbands and wives wish to coordinate their retirement decisions can distort the impact of any policy initiatives. Family can also take on other forms such a divorce, single-parent families, or common-law relationships - several of the real possibilities for baby boom women that will have lifetime implications for their retirement transitions. That the short window for exit from work to retirement is immune to family influences for men and women suggests that other factors must be at work that have yet to be considered.

The coexistence of the male model of retirement with newer forms of retirement is apparent in each and every paper. After arguing that baby boom women's retirement will be fluid and amorphous, McDonald illustrates how pension policy is built on a male model of retirement that does not accommodate interdependence, care giving or intermittent and contingent work histories. In the same vein, Szinovacz makes the point that it is almost impossible for many couples to jointly retire because social security rules militate against it. Any age difference between spouses means a reduction in benefits if either spouse retires before reaching full benefit eligibility. This same problem is echoed by Schellenberg, Turcotte and Ram in their study of dual-earner couples, while Deschênes and Stone indicate that their model of the transition to retirement may not capture the interdependence of women's lives, a situation less applicable to men. The co-existence of the old (individualist) and the new (interdependence) in theory, research and policy is likely part of the dilemma that holds back the advancement of new thinking about retirement.

It would seem, therefore, that the time has come for scholars to go back to the drawing board and address the lack of theoretical progress about gender, families and retirement. At minimum, a life course perspective with revised definitions of retirement that include gender/family/work linkages would be valuable in reflecting the experiences of people and would lead to policies that would have a better chance for success. As the situation stands now, current social security and pension rules do not reflect women's interwoven life styles of work and care giving, as well as the interdependence of family members.