Phyllis Moen, Jack Lam and Melanie N.G. Jackson
The (Kenneth) Burke theorem states that “A way of seeing is also a way of not seeing – a focus upon object ‘A’ involves a neglect of object ‘B’” (Burke, 1935, p. 70). Most family scholars focus on “young” families, as young adults move into marriage and parenthood and as working families negotiate paid work while simultaneously raising children. And yet, increased longevity, together with the aging of the Boomer cohort in North America and Europe, meant that growing numbers of contemporary families are “older” families, including older adults whose own aging parents are increasingly frail. While family scholars key in on the early stages of the adult life course (Burke’s “A”) and gerontologists concentrate on aging (and often infirm) individuals (another “A”), there is remarkably little emphasis on “B” – older families.
This chapter takes a gendered life course approach to understanding contemporary families in later adulthood. We draw on a gender perspective together with four interrelated life course themes (Elder, 1974; Moen, 2001; Mortimer and Shanahan, 2003; Moen and Spencer, 2006; Shanahan and Macmillan, 2007): (i) historical timing, policy contexts, and social change; (ii) transitions and trajectories, including the cumulation of advantage and/or disadvantage with age; (iii) linked lives, recognizing that families constitute patterned, interconnected relationships over time that shift with age, age-graded expectations, and exigencies; and (iv) strategic (and often gendered) family adaptive strategies, as older family members respond to changing resources and claims.
We use these interconnected themes to promote understanding of circumstances that (i) are being transformed at the macro-level of society in ways that make contemporary families in later adulthood historically unique, (ii) reflect an ongoing interaction between families and institutions (especially work and retirement), and (iii) are apt to characterize micro-level relationships in the later stages of the family course. Specifically, this chapter employs these four life course themes to capture the nature and experiences of contemporary aging families around three main substantive areas: family composition, families in transaction with the institutions of work and retirement, and intergenerational as well as intragenerational relations, including caregiving, widowhood, and aid to younger generations. While we recognize considerable heterogeneity and inequities by race/ethnicity, nativity, and class in the experience of later family life, given space limitations, we focus here on gender distinctions and potential differences in the contemporary later family course compared to prior cohorts. The focus in this chapter centers largely on the United States, though we also draw on some empirical research on other countries. Clearly, the emerging later family course is an important issue for research throughout the developed as well as the developing world.
Three demographic trends – longer and healthier lifespans, the aging of the large Boomer cohort (born 1946–1964), and delays as well as declines in marriage and fertility – have changed age structures of families, with rising numbers now constituting “older” families. A report by the Organization for Economic Cooperation and Development (OECD) underscores how life expectancy has increased remarkably over the past decades (OECD Family Database, 2011). In 2009, women at age 65 in OECD countries could expect to live another 20.5 years, while men could expect another 17.2 years. This reflects an increase since 1960 of an average of 5.6 years for women and 4.4 years for men. Living longer increases the risk of disability, and a greater proportion of women’s than men’s lives are lived with disability, even though disability at older ages is declining (Robine, Romeiu, and Cambois, 1997; Laditka and Laditka, 2002; Cutler and Wise, 2009).
Relationships take on added significance given increases in longevity, since both couple and intergenerational ties can extend over many years. For example, 18% of children born in 1900 were likely to be orphaned by the time they were 18 years old, whereas 68% of children born in the United States in 2000 were likely to still have four living grandparents by the time they reached 18 (Bengston, 2001). Three- or four-generation families are increasingly prevalent. Thus, the new longevity translates into long marriages and long intergenerational ties, the availability of extended kin such as grandparents and great-grandparents as resources for children, but also more older parents, grandparents, and others requiring care, often for longer periods of time (Bengston, 2001).
Ryder (1965) described social change as occurring through a process of cohort succession, as each cohort moves through the life course. Today’s older families in the United States consist of the large Boomer cohort, born from 1946 through 1964, and the War cohort just preceding them, born from 1942 through 1945, as well as the Silent Generation, born from 1925 through 1941 (Soldo et al., 2007). By 2030, all Boomers will be at least 65 years old, with 71 million adults in this age range (Centers for Disease Control and Prevention, 2010). The aging of the Boomer cohort is transforming what it means to be old, and to be an older family, in contemporary society.
Reuben Hill (1970) and his colleagues laid the groundwork for the study of family development by theorizing the orderly flow of families through life cycles, that is, the patterned sequence as families move through formation (marriage), childbearing, child-raising, child-leaving, and dissolution (widowhood). Hill’s life cycle approach was foundational in that it emphasized the temporal nature of family life. Where families are located in biographical and family time (the ages of children and parents) and in social and institutional time is crucial to understanding their functioning. Hill investigated whether family members felt they were “ahead” of or “behind” normative clockworks as to when key transitions should take place. But identifying the stages of a “typical” family cycle was seen by some (O’Rand, 1996) as prescriptive, that is, a prototype of how families should be. Moreover, today, the “typical” family cycle reflects the experience of an ever smaller portion of contemporary families.
As early as the 1970s, life course scholars began to build upon and extend the life cycle framing of families to capture multiple paths through lives, the dynamics, and diversities of arrangements within and across families. The life course itself can be thought of as a series of role entries, trajectories, and exits that, taken together, constitute both people’s biographies and their family life course (Settersten and Owens, 2002; Elder, Johnson, and Crosnoe, 2003; Kohli, 2007; Elder and Giele, 2009).
Today, there is no “normal” family course. Marriage and children have come to be more of an option than an inevitability. Younger adults (Millennials born 1980–2000 and Gen Xers born 1965–1979) are marrying later or not at all, postponing parenthood, having fewer children or none at all, and moving in and out of jobs and in and out of schooling, while older adults (Boomers born 1946–1964 and members of the Silent Generation born 1925–1942) are also moving in and out of marriage and in and out of retirement. Older families represent the fallout from earlier choices as to whether and when to have children, whether and when to break up existing marriages or partnerships, whether or when to be in or out of the labor market, and whether and when to live close or far from extended family.
Thus, the later family course is characterized by considerable heterogeneity. There are older parents in their 50s, 60s, and 70s whose adult children have moved back home given economic downturns; older adults unexpectedly caring for infirm parents, siblings, and spouses; “old” new parents who have children later in life; couples negotiating two retirements; single mothers with adult children; grandparents caring for grandchildren; and fictive kinships with friends who are more “family” than not. In the context of the recession, the Pew Research Center (Parker, 2012) reports a steady upward trend in the percentage of young adults (ages 25–34, Millennials) living in multigenerational households, from 11% in 1980 to 21.6% in 2010. Most cite economic reasons for moving back. Other forms of shared households, including living with nonrelatives, have also increased (Mykyta and Macartney, 2012). Whereas in the past, the typical older family consisted of marital dyads, growing numbers of families are “stem” or “beanpole” families, consisting of relationships between older parents and adult children or between grandparents and grandchildren (Bengston, 2001; Fischer, 2011).
As concrete repositories and constellations of roles, resources, risks, and relationships, families are where abstract national and global economic, technological, political, and demographic forces play out in real time. Studying family and individual adaptive strategies provides a useful point of entry for rendering visible what are often outdated shared cultural understandings and taken-for-granted rules, roles, and risks within and across the institutions of family, labor markets, business, and social policy development and regulation.
Given the variability in family lifestyles and the growing social acceptance of singlehood and divorce, contemporary older (as well as younger) couple relationships may be more by choice than by custom. Prior family decisions (or nondecisions) also shape family life in later adulthood. Having fewer children reduces the potential pool of caregivers for aging families (Hagestad and Uhlenberg, 2007). In young adulthood, the adaptive strategy of postponing parenthood to advance careers may result in childlessness in older age. Older divorced men estranged from their children may find their adult children feel no filial obligation to care for them (Solomou et al., 1998; Jong Gierveld and Dykstra, 2002; Shapiro, 2003). Divorce and remarriage have long-term intergenerational consequences. College-age children in stepparent families receive less help with college costs than those living with both biological parents (Henretta et al., 2012).
The experiences of older families are shaped by fundamental institutional structures around paid work based on the breadwinner/homemaker family form – the nature of career paths, pensions, social security provisions, unemployment, and disability insurance – along with the clockworks of workdays, workweeks, workyears, and the rewards accruing to those who follow them (Moen, 2012). Key to the family economy is paid work and retirement income, as well as social welfare supports in the form of social security and other protections. The historically constructed division of paid work from unpaid family work and community service has had enormous implications for family functioning and family well-being, as well as for gender inequality (Arber and Ginn, 1995; Moen and Roehling, 2005). In the United States, most contemporary older families consist of members of Boomer, War, and Silent Generation cohorts who in the prime of life took for granted the male “career mystique” that continuous full-time work throughout most of adulthood was the only path to success and life quality (Moen and Roehling, 2005). Accordingly, labor market and social security policies and practices were developed in the mid-twentieth century with seniority leading to job security and benefits tied to whether employees worked full time and full year in “permanent” jobs.
Given that family obligations typically fall on to women, few wives and mothers in the Boomer or prior cohorts have been able to pursue this ideal. Accommodating family life by moving in and out of jobs, in and out of full-time and part-time work, has been costly for women’s wages, job security, pensions, and advancement; these costs become especially apparent in the second half of the life course (Arber and Ginn, 1995; Harrington Meyer and Herd, 2007). Moreover, many older families of minorities, immigrants, and those with little education reap the disadvantages of having been in the secondary labor market. Moving to a global, highly competitive, information economy has meant that career prospects are now less predictable, secure, or stable for Americans, regardless of gender, race, or education, and this insecurity is intruding into the European experience as well. The old “contract” between employers and employees (awarding job and retirement security to workers in return for their commitment) is quickly disappearing (Sweet, Moen, and Meiksins, 2007).
Older families are now living in contexts that upended conventional retirement scripts, enduring gender schema that continue to prioritize women’s family obligations, a growing older workforce that is more educated and healthy than in previous and perhaps future generations, and a volatile global economy eroding traditional job security and retirement protections. Unlike their parents or grandparents, contemporary older families in the United States are increasingly facing two retirements, the husband’s and the wife’s, in policy and economic contexts challenging taken-for-granted blueprints around the timing and nature of “retirement.” Sweet et al. (2007) found that 9 in 10 middle-class couples report some risk in at least one spouse’s job. In only 1 in 10 couples were both husbands and wives confident they would be able to keep their jobs.
Another key source of uncertainty and risk concerns the dismantling of the pension system and retirement protections. Systematic changes in retirement practices (O’Rand, 2003; Hacker, 2006) include employers switching from defined benefits to defined contribution plans. In the United States, defined benefit plans were historically designed so that workers were entitled to receive a given income once they retired. The amount retirees are entitled to in defined contribution plans varies depending on the performance of the fund investment, subject to the volatility of the stock market, as well as the amount employers and employees invest in them. Therefore, defined contribution plans are far more risky than defined benefit plans (Hacker, 2006), contributing to the economic uncertainty of older families. While nonstandard workers – part-timers and contractors – have been shut out of employer pension systems, even full-time employees in the United States are increasingly at risk; 44% of full-time workers in their 50s have neither a defined benefit nor a defined contribution pension plan from their employer (GAO, 2011).
A significant and growing number of Boomer women in the United States are not married; their work and retirement circumstances differ from those of wives, in that they cannot count on other sources of income or resources. Men’s and women’s lives today are more contingent than orderly, regardless of one’s location in the social structure. Ambiguities and uncertainties about the future abound, coloring the sensibilities of older employees and their families as they face the realities of forced early retirements, layoffs, reduced benefits, and greater workloads. Growing numbers of Boomers find themselves unexpectedly “retired” from their primary career jobs in their 50s and early 60s, frequently as a result of corporate restructuring, mergers, and bankruptcies, or poor health. Looking at members of the pre-Boomer cohort in the United States found that most older workers left their career jobs earlier than they had expected as a result of early retirement packages or health difficulties (Han and Moen, 1999, 2001). And there is considerable heterogeneity in the age at which women and men even began planning for retirement.
Rather than being an individual status passage, retirement is a family transition, one that changes the roles, relationships, resources, and identities of all family members. Retirement based on the (male) family breadwinner had implications for homemaking wives and widows, showcasing the interdependencies of family relations. Today, the linking of lives around retirement is evident when two family breadwinners in the same household try to arrange joint retirements or else realign relationships and expectations when one retires and the other spouse does not (Moen, Kim, and Hofmeister, 2001; Kim and Moen, 2002; Moen, Sweet, and Swisher, 2005; Moen et al., 2006). Given many uncertainties, the retirement transition is a time of life without blueprints for couples moving to and through the retirement transition years, as well as for older singles (mostly women) with or without children.
Research on decision making and planning around retirement has focused on individuals’ expectations and preferences. Midlife work experiences, in particular, have been linked with retirement intentions and the ability to realize later-life employment arrangements. In midlife, precarious employment (e.g., with low earnings and without private pension plans or health insurance coverage) predict preferences for later retirement (Raymo et al., 2012), likely because workers do not anticipate having adequate retirement income. Realizing one’s preferences is contingent on a number of prior employment circumstances and is similar for both men and women; those with private pension coverage were less likely to realize preferences for working full time but more likely to realize preferences for not working (Raymo et al., 2010). Men with shorter job tenure and lower incomes expected to work longer but also reported lower life satisfaction when expectations were not met (Clarke, Marshall, and Weir, 2012). For 11 European countries where public retirement benefits have been more adequate, workers in lower social positions were more likely to report early retirement intentions (Wahrendorf, Dragano, and Siegrist, 2012). Expectations for later retirement are reflected in behavior. Using HRS data, Banerjee (2011) reports that fewer Americans are now expecting to retire “early” at ages 62 and 65. In 2006, 7.4% of people over 50 (leading-edge Boomers and Silent Generation cohorts) expected to stop working by 62, but by 2010, this number had dropped to 4.9%. Analyzing social security claims, Johnson (2012) reports that only 26.9% of adults aged 62 and older and eligible to claim early retirement benefits in 2011 did so, the lowest since 1976.
While about a quarter of middle-aged and older respondents believe their standard of living in retirement will be better than the previous generation, nearly half believed that it would be worse (Rix, 2011). In this sample, 67% were working for pay, 16% were not working but looking for work, while 18% were neither working nor looking for work. One concern tied to retirement timing in the United States has to do with the availability of healthcare. Fully 15% of men and women between ages 50 and 64 in 2010 were uninsured (Smolka, Multack, and Figueriredo, 2012). Early retirement (before eligibility for Medicare) is more likely for those having retirement health insurance (Nyce et al., 2011).
Many members of older families continue to work for the income it provides. Job earnings in fact provided 30% of income for adults over 65 in 2008 (Federal Interagency Forum on Aging-Related Statistics, 2010). Other major sources are social security (37%), pensions (19%), and asset income (13%). Others (often the college educated) choose to remain in the labor force because they enjoy working. For retirees who continued to work, 92% report they want to stay active, and 86% report enjoying the work (Helman, Copeland, and VanDerhei, 2011). Of course, fully 90% also provided a financial reason, such as a decrease in savings and investment (72%), needing to make ends meet (59%), or in order to keep health insurance or other benefits (40%). Many older family members (as individuals and as couples) volunteer for community organizations or else take on low-wage jobs for nonprofits, seeking new meaning and social purpose through employment or volunteer work (Freedman, 2007, 2012; Bank, 2010; Moen and Flood, 2012).
What a gendered life course perspective brings to the research and policy agenda is the recognition of the fact that families are in constant transaction with the labor market as “role budgeting centers” (Goode, 1960), allocating family members’ hours, days, and years to both paid work and unpaid domestic care work. In both North America and Europe, one retirement strategy is to take part time, part year, contract, or bridge jobs, including self-employment, in the move from full-time employment to full-time leisure (Cahill, Giandrea, and Quinn, 2006; Ebbinghaus, 2006; Wang et al., 2008; Giandrea, Cahill, and Quinn, 2009; Van Solinge, 2012).
But these are often gender strategies, reproducing gender distinctions and disparities (Moen and Roehling, 2005). Because there are few options regarding the kinds of labor market attachments that provide good incomes and health insurance, couples often prioritize one job per family, typically the husband’s (Pixley, 2008), which leads to gender inequality throughout the later life course (Arber, 2004; Moen and Spencer, 2006; Harrington Meyer and Herd, 2007). The timing of retirement is also a strategic choice. Putney and Bengtson (2003) find work–family conflict associated with greater odds of men and women preferring not to work or to work only part time (vs. working full time) 10 years in the future. Men and women with high work–family conflict when employed report a marked decrease in depressive symptoms upon full or even partial retirement (Coursolle et al., 2010).
If retirement is one challenge of later life, another is caregiving. In European countries, the concern is that state provision of goods and services might “crowd out” family assistance to their aging members; the evidence suggests this is not the case (Künemund, 2008). Given the absence of extensive public supports in the United States, informal/family caregivers (family members, friends, and neighbors) provide the bulk of care for older adults (Family Caregiver Alliance, 2008); approximately 42.1 million family members in the United States in 2009 assisted adults with daily activities (Feinberg et al., 2011). “Primary” caregivers assume the majority of caregiving obligations, including completing everyday tasks and errands. “Secondary” caregivers provide assistance to primary caregivers in a range of domains including emotional, instrumental, and financial support. Tertiary caregivers periodically provide assistance when the primary caregiver requires it, as well as assisting high-functioning care recipients, but are not usually involved in the decision-making aspects of care (Bourgeois et al., 1996; Dilworth-Anderson, Willliams, and Cooper, 1999; Gaugler, Kane, and Kane, 2002; Jackson and Gaugler, 2014). As these definitions suggest, the bulk of the caregiving falls on one individual within a family, the primary caregiver. This has significant implications for both the health and psychological well-being of the caregiver, as well as implications for the quality of care that family member is able to provide.
The closest and most accessible family members tend to become caregivers in what is called the hierarchical compensatory model (Cantor, 1979; Gaugler et al., 2003). Spouses are typically first in line, followed by adult children; then other relatives, friends, and neighbors; and finally formal support services. Besides assistance in activities of daily living (ADLs) tasks, caregivers also help with more cognitively demanding instrumental activities of daily living (IADLs), which can include managing medication, shopping, and housework (Szinovacz and Davey, 2008; Liu and Gallagher-Thompson, 2009; Jackson and Gaugler, 2014; The National Long Term Care Survey). Caregivers are primarily women, typically in their 40s, providing care for relatives who are in their 70s. What is unique to the contemporary caregiving experience is that most family caregivers are also in the workforce.
Caregiving for family members impacts the caregivers’ psychological, emotional, and physical well-being, but this may depend on whether the caregivers perceive their role positively or negatively. Additionally, members of different cohorts perceive caregiving differently in terms of how they view the role as well as their willingness to access services and resources to aid in the care of older relatives. Boomers in Canada insisted they sacrifice far less to be caregivers than had their mothers’ generation, emphasizing other aspects of their lives besides caregiving (Guberman et al., 2012). They were concerned with maintaining their own autonomy, providing less hands-on care and being less willing to sacrifice their work lives in the face of caregiving demands. Nonetheless, they believed that being a caregiver made them less egotistical and more selfless compared to their counterparts not providing care. In contradiction to Guberman et al. (2003), Gans and Silverstein (2006) found US Boomers reported having stronger filial norms than their parents or grandparents reported.
Demographic trends throughout the life course – marrying later or not at all, postponing parenthood, having fewer children or none at all, and divorce – will impact the availability of family members to care for aging Boomers. Boomers are less likely than prior generations to have spouses or children to care for them. The circumstances of Boomers (delayed marriage, reduced fertility) are more similar to an older generation – the parents of the Depression and WWII cohorts (born 1905–1921) – than to the Boomers’ own parents’ generation (born 1922–1940) (Ryan et al., 2012). Depression and WWII cohorts were less likely than the parents of Boomers to be married and to live within 10 miles of one of their children. Boomers will be less likely to have spouses or children to care for them. They have more siblings, but it is unclear whether they will be able to compensate for closer relations.
As the principle of linked lives states, individuals’ lives are interrelated, and the actions of one family member influence the experiences of others. The illness of a parent or spouse can precipitate key turning points as family members move in and out of caregiving, take in an older family member, place an infirm relative in a nursing home, or experience the death of a loved one. These transitions often have significant impacts on the psychological and emotional well-being and social lives of the caregivers. Seltzer and Li (2000) found that participation in leisure activities declined for wives who became caregivers, but daughters who became caregivers increased their social participation. Thus, caregiving may be more disruptive to wives’ family and marital relations, whereas daughters’ primary family relationships with their own husbands and children remain relatively unchanged. Daughters were more likely to exit caregiving by placing their infirm parent in a group residence, whereas wives only exited the role through the death of their husbands. When daughters institutionalized a parent, they reported a significant decrease in burden (while wives tended to continue to care for their husbands at home). Every indication is that caregiving is more demanding and all-consuming for wives than daughters.
When relationship quality is perceived to be high between family members, the transition to caregiving is less emotionally and psychologically disruptive. High relationship quality likely reduces how burdensome the role is perceived. Because burden predicts institutionalizing, higher relationship quality may also reduce the likelihood of institutionalization. Marks et al. (2008) found caregiving sons, but not daughters, showed greater depressive symptoms compared to sons who were not providing care. Lower relationship quality prior to becoming a caregiver, however, resulted in lower psychological well-being for caregiving daughters and sons. In dual-earner couples, caregiving was also associated with psychological distress, but men’s well-being was enhanced when they moved into parental caregiving, possibly taking over caregiving responsibilities from their wives (Chesley and Moen, 2006).
Widowhood is another major transition of later life. Although women are more likely to experience widowhood, widowed or divorced men are more likely to remarry than are widowed or divorced women (Carr and Bodnar-Deren, 2009), providing them with socioemotional support that buffers some of the negative experiences associated with widowhood. Although widows are less likely to regain social support from remarriage, they are more likely to receive support from several sources, such as children, friends, neighbors, and religious groups (Hahn et al., 2011).
Widowed women’s psychological distress was tied to their financial strain, whereas lack of close relationships was linked to widowed men’s psychological distress (Carr, 2004). Men’s self-esteem decreased in the short term if they had been highly dependent on their wives for homemaking tasks.
Not all consequences of widowhood are negative for women. Widowed women have higher self-esteem than married women, perhaps from the confidence of having coped with a significant event (Carr, 2004). Widows also report higher levels of “personal growth” than widowers, which is positively related to the availability of emotional support. If the spouse had been in poor health, widowed men and women reported experiencing significantly more “personal growth” after widowhood. As with caregiving relationships, relationship quality shapes whether widowhood is experienced positively or negatively. Widowed persons expressed greater anger 6 months post loss if they had rated their marriages more positively than their spouses had in contrast to couples with similar marital assessments (Carr and Boerner, 2009). Women who remarry show an increase in psychological and emotional well-being, as also seen in earlier results with men who remarried. Widowed women who have an interest in remarriage are younger and unhappier than widows not interested in remarriage (Moorman, Booth, and Fingerman, 2006). Additionally, remarried widows are less depressed and less worried about finances. This is consistent with research on widowhood conducted internationally, which also shows limited interest by widowed women in remarriage. Dutch data find that only 11% of widowed individuals desire a new partner relationship (Stevens 2002). As in the United States, Dutch widowers were more likely to want a new partner than widows; about 20% of the Dutch sample had some sort of partnership with another person.
Because women are less likely to remarry, they appear to be more likely to seek social support from other relatives. Adult children are most likely to be a source of support for widowed women. Kalmijn (2007) found that widowed mothers in the Netherlands were more apt to receive support and contact from their children than were married mothers, while widowed fathers received the same amount of support but had less contact with their children than married fathers. Widowed parents who remarried received less support from their children than did widowed parents who lived alone – with remarried fathers receiving the least support.
Hill (1966) theorized family stress adaptation in the face of crises and ongoing strains and laid the groundwork for the contemporary life course “cycles of control” model (Elder, 1985; Elder, Johnson, and Crosnoe, 2003). Specifically, when needs or pressures outstrip the available resources with which to manage them, individuals and families seek to regain equilibrium using strategies of adaptation. These strategies include ways of providing care, dealing with the ambivalence of relationships, and helping out between generations.
Family members strategize as to who will provide care to infirm relatives and the kind of care they provide. Ward-Griffin et al.’s (2007) qualitative study of 15 mother–adult daughter dyads in Canada found that mothers and daughters identified four types of relationships: custodial, combative, cooperative, and cohesive. Custodial relationships were defined by the idea of duty, with a high expectation of care but a low expectation of emotional support. Within combative relationships, mothers and daughters competed for control over the care process. Daughters were more controlling of their mother’s behaviors and ADLs, which increased tensions in the relationship. This type of relationship was more likely to lead to neglect and abuse. Cooperative relationships had reciprocity as a key defining characteristic, with both mothers and daughters having a strong family network they could rely on for support. Cohesive relationships were also strength based and positive, with daughters attentive to their mother’s need for independence, which both mothers and daughters found to be rewarding.
When caregiving becomes too burdensome, caregivers identify strategies to reduce the burden, such as retiring from the workforce or institutionalizing the care recipient. Another adaptive strategy is to shift caregivers. Daughters caring for mothers were more likely than those caring for fathers to exchange caregiver responsibilities with another relative (Szinovacz and Davey, 2007). This, along with a study by Ward-Griffin et al. (2007), suggests that there may be specific challenges for caregiving relationships between mothers and daughters. Having more siblings, particularly more sisters, increased the probability of switching caregivers. This provides further evidence for the concept of linked lives, since the decision of one sibling to leave the role of primary caregiver has a direct impact on the lives of the other siblings.
Another strategy for managing caregiving demands is to exit the workforce. This can create a nonnormative, off-time transition for caregivers. Caregiving responsibilities and health/stress factors led Canadian women to retire earlier than expected (Zimmerman et al. 2008). In a US study, husbands caring for their wives were less likely than noncaregiving men to retire, while wives caring for husbands were five times more apt to exit their jobs than other wives (Dentinger and Clarkberg, 2002). Gender expectations may account for this difference.
Other strategies for care include moving infirm relatives to assisted living facilities (ALFs). Kemp (2008) studied 20 couples relocating to ALFs, finding couples transitioned to ALFs together, even though the healthier spouse could have continued to live independently. The transition to ALFs began with an illness diagnosis for one or both spouses. The diagnosis required a transition for the couple, because the caregiving partner was unable to meet the increasing burden of caring for their partner on their own and the couple wanted to be together. Adult children, who were seen as the “lifeline” for the couple, were involved in the decision regarding relocation. In these cases, adult children were the ones who would investigate options, locate facilities, and arrange relocation.
Adaptive strategies can involve providing support for one’s adult children (or vice versa) but also (re)defining the situation, say, feeling more or less ambivalent regarding intergenerational relationships. Typically, there appears to be less ambivalence between older parents and adult children when the children are married and not experiencing significant challenges. In relationships with their adult children, mothers were more ambivalent if they still provided financial support for adult children but less ambivalent toward adult children who had completed college and were married (Pillemer and Suitor, 2002). Mothers voiced more ambivalence toward children who experienced major problems in adulthood (serious illnesses or injury, serious mental or emotional problems, problems with drinking or drugs, and problems with the law) and toward children they felt took more than they gave back (Pillemer et al., 2007).
What is less frequently discussed is the effect that adult children have on the well-being of older parents. Greenfield and Marks (2006) assessed whether parents’ well-being was impacted by adult children having 10 types of problems: chronic disease or disability, frequent minor illnesses, emotional problems, alcohol or substance problems, financial problems, problems at school or work, difficulty finding or keeping a job, marital or partner relationship problems, legal problems, and difficulty getting along with people. Parents with children having more problems reported lower levels of well-being, less positive affect, more negative affect, less self-acceptance, and more family strain. Married parents experienced smaller declines in positive affect in relation to their children’s problems than did single parents. In another study, African American parents were more likely than white parents to have high levels of depression in relation to negative treatment by their (adult) children and their children’s unemployment, as well as higher anger following a child’s illness (Milkie, Bierman, and Schieman, 2008). Children’s divorce, however, decreased anger for African American parents. Pillemer and Suitor (1991) also report a significant relationship between children and parents’ depression and conflict.
In light of the new family demography, most children in North America and Europe have several living grandparents (Hagestad and Uhlenberg, 2007), but a study in Italy shows they have less contact when the grandparents are divorced, widowed, or repartnered, compared to those in stable marriages (Albertini and Saraceno, 2008). In the United States, there were approximately 6.7 million grandchildren under the age of 18 living with a grandparent in 2009, and 2.7 million of these grandparents served as the primary caregivers for their grandchildren, assuming responsibility for their basic needs (United States Census Bureau, 2009). Kohli and Albertini (2008) examined European families’ support (financial transfers, social support) for their adult children’s own family projects. They found differences in support strategies across countries (Sweden, Germany, France, and Spain), but that both social support (including caring for grandchildren) and the combined strategy of social support and financial transfers increased when adult children transitioned into parenthood (see also Silverstein, 2006).
Typically, when the literature refers to the prevalence of grandparenting, it most often refers to the care provided by grandmothers (Arber and Timonen, 2012). Being the primary care provider to a grandchild can be seen as a nonnormative transition, yet other cultural values and contexts may characterize this as an expectable, on-time transition. Landry-Meyer and Newman (2004) interviewed grandparents who had legal custody of their grandchild. This was usually an unanticipated role, but they felt greater clarity regarding the roles once they had legal custody. Participants reported not feeling like a grandparent, however, because they were parenting their grandchildren. The literature also gives less emphasis to grandparents with older grandchildren.
As with children, relationships seem to depend on the younger generation fulfilling expectations. Grandparents and grandchildren both report higher-quality relationships when grandchildren enroll in college than when grandchildren do not enroll in college (Crosnoe and Elder, 2002). This also suggests support for the generational stake hypothesis proposed by Bengston, Schaie, and Burton (1995).
Given the increased longevity and the rising likelihood of chronic illnesses with age, older adults are increasingly likely to become dependent on not only their children but also their grandchildren for care. Approximately 8% of family caregivers of older adults in the United States are grandchildren (Foundation for Accountability and Robert Wood Johnson Foundation, 2001). The increased likelihood of single motherhood, divorce, and widowhood means diminished marital ties that reinforce the beanpole family form of older adults and their descendants. Of course, these relationships differ by class, race/ethnicity, and gender. As Swartz (2009, p. 207) points out, “Intergenerational support, while offering real help, also becomes a largely hidden mechanism by which privilege or disadvantage is transferred through families from generation to generation.”
We began with the Burke theorem, that a focus on “A” excludes “B” and vice versa. Scholars from a range of disciplines have until recently based their theoretical arguments and empirical investigations of paid work, family work, work–family issues, mobility and inequality on conventional taken-for-granted mental maps, and framings of families raising children. These mental maps are useful heuristics, but they direct our attention to some things, and not others. By contrast, a gendered life course approach to aging families puts both work and family in motion and in context, moving the focus beyond families raising children to families of adults living lives in tandem with their aging parents. It also captures the dynamics of family experience in transaction with the labor market and retirement.
The four life course themes organizing this chapter – historical context, transitions and trajectories, linked lives, and adaptive strategies – are both overlapping and interdependent. On the historical context, more needs to be known about these processes in contemporary cohorts regarding families of different composition, race/ethnicity, and class, even as evidence from studies of previous cohorts may well be out of date. Demographic changes, technological advances in communications, immigration, and economic uncertainty, for instance, are challenging assumptions about aging and families based on past cohorts. These social forces necessitate a complex but rich theoretical and research agenda on the contemporary experiences of older families in transaction with larger institutions. For example, as the gendered nature of family, work, and retirement “careers” as they unfold and intersect in later adulthood, they invite examination of families and labor markets as institutionalized processes intersecting and shifting over the life course. These changes have enormous implications for gender – and other – inequalities in resources, status, and future prospects in later adulthood, within and across families and societies.
Transitions and trajectories capture the importance for future research linking life stages, examining the impacts of early experiences (including the biographical pacing of family formation, parenthood, and labor market attachments) on later-life families. The concept of adaptive strategies (Elder, 1974; Hareven, 1982; Moen and Wethington, 1992; Elder, Johnson, and Crosnoe, 2003; Moen et al., 2013) has enabled life course scholars to consider families as decision units, even as they are enabled or constrained by existing (and often outdated) institutional arrangements. This points to the value of investigating ways contemporary older family members are making strategic choices regarding employment, residential mobility, and caregiving, in response to constraints, challenges, and opportunities.
The gendered life course framing of families in the second half of life demonstrates that family members’ adaptive strategies cannot be captured by simply examining individual experiences separately. Recognizing linked lives, they require couple-level and family-level conceptualization and analysis. Historical, cultural, and institutional forces provide a backdrop of constraint, uncertainty, and risk. Work, family, community, and personal ties, as well as demands and resources, shift over the life course, as do their meanings. Constraints and strategic adaptations earlier in the life course play out in and often produce gender, race/ethnic, nativity, and class inequalities in the later adult years. What life course scholars find in both men’s and women’s biographies are more discontinuities than continuities – in family relationships, in paid work and retirement, in personal experience and development, and across cohorts (Rindfuss, Swicegood, and Rosenfeld, 1987; Elder and O’Rand, 1995; Settersten and Mayer, 1997; Han and Moen, 1999, 2001; Pavalko and Smith, 1999).
The authors thank the McKnight Foundation for supporting the research and Jane Peterson for her help in preparing the manuscript for publication.