Sociology

50th Anniversary Feature Article The Gerontologist © The Author 2010. Published by Oxford University Press on behalf of The Gerontological Society of America. Vol. 50, No. 6, 720–734 All rights reserved. For permissions, please e-mail: [email protected]. doi:10.1093/geront/gnq085

The Gerontologist720

As a concept in gerontology, gender appears as lists of traits learned through socialization when theo- rized at all. I argue for a framework that theorizes the intersections of relations of gender inequality with those of age. This framework holds that men and women gain resources and bear responsibilities, in relation to one another, by virtue of mundane categorization into naturalized, stratified groups. Current research shows that this approach allows explanation of gender differences, which appear in many reports but which usually go untheorized, as responses to social inequality. I illustrate applications to research and practice in relation to three areas of old age experiences: financial security, spousal care work, and health. Throughout, I discuss implications of focusing on inequality to enhance our abilities to engage in effec- tive research, practice, and policy for older people, women and men alike. For instance, an understand- ing of the gender division of labor and workplace discrimination makes clear that financial status in later life cannot be reduced to individual choices concerning paid labor or retirement planning. And understanding that people orient their behaviors to gender ideals allows us to see that men and women perform spousal care in similar and different ways that require varied responses from practitioners; it also reveals contexts in which men engage in posi-

tive health behaviors. Finally, I argue that gerontolo- gists interested in facilitating favorable outcomes for old people should consider research and practice that would disrupt, not reinforce, the bases of gender inequalities in later life.

Key Words: Intersections, Social inequalities, Economic/ financial security, Caregiving/care work, Health

It is no longer unusual to see gender mentioned in gerontological research or practice. Even when they do not focus on it, most gerontologists dem- onstrate awareness of gender’s potential impact and thus include it as a variable in their analyses or give at least fleeting thought to gender differences as they fashion policies. To move beyond this and theorize gender gives us a framework within which we can place and comprehend gender differences— not only why they occur but also why and how they matter. Such an understanding keeps us from formulating well-intentioned policies or programs unlikely to be effective—for instance, trying to take what we learn about men’s success at coping with care work and telling women simply to do likewise. A theoretical understanding of gender can also move us closer toward research and inter- ventions that are more equitable—for example,

Gender Relations and Applied Research on Aging

Toni Calasanti, PhD*,1

1Department of Sociology, Virginia Tech, Blacksburg.

*Address correspondence to Toni Calasanti, PhD, Department of Sociology, Virginia Tech, Blacksburg, VA 24061. E-mail: [email protected]

Received August 3, 2010; Accepted September 14, 2010 Decision Editor: William J. McAuley, PhD

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 721

devising strategies to persuade men to engage in better health behaviors that do not simultaneously reinforce their positions of dominance over women. I urge recognition that men as well as women have gender, a set of ideals created in rela- tions of inequality and in concrete institutions, motivating behaviors that shape experiences of old age. Scholars should study masculinities as well as femininities, understood in these relational, contex- tual terms, in all areas of aging research—not only just those in which we think men might encounter difficulties (such as care work or health outcomes) but also in the areas in which we think men might do better than women (e.g., financial security).

My goal in this article, then, is to outline a frame- work for exploring gender that allows us better to conceptualize research, policies, and practices that are both effective and equitable and thus improve the lives of old people—women and men alike. Through- out, I distinguish between this approach and others that treat gender as a background variable, indi- vidual attribute, or set of stable characteristics.

Understanding Gender Relations

Scholars of gender emphasize that gender is relational, that is, men and women gain identities and power in relation to one another. Gender rela- tions are dynamic constructed power relations embedded in social processes and institutionalized in social arenas to which people orient their behav- iors to ideals of manhood and womanhood with important consequences for life chances (Glenn, 1999; West & Fenstermaker, 1995). Societies organize on the basis of gender such that popular ideals of manhood and womanhood result from and affirm gendered divisions of labor, authority, and status. Gender identities are internalized natu- ralized ideals of behavior, formed and reshaped in interaction with others. “Doing gender” by appear- ing to conform to the dictates of biological givens affirms inequalities in access to social and material resources such that people feel compelled to sustain them even at their own expenses (West & Fenstermaker, 1995). In the process of enacting gender identities, people also behave in ways that largely serve to privilege men—give them unearned advantages—while they tend to disadvantage women, even as people resist and reformulate seemingly “natural” gender differences and gender meanings. A key point, which is often overlooked, is that while people orient their behaviors to gen- der ideals, what these actions are can vary by con-

text. That is, even if a group of traits are thought to be feminine or masculine, the ways in which women or men behave depends upon the situation. For instance, women may be forceful or aggressive in situations where they feel their child is being threatened, behavior that seems to violate abstract ideals of femininity as delicate or nonassertive.

Gender relations generally go unchallenged because they are embedded in taken-for-granted rou- tines of such social institutions as paid work and family life. Focus on the institutional embeddedness of seemingly natural behavior directs our attention from individuals and forces of nature toward the force of organizational life. That is, although one can always view the enactment of gender as personal behavior, gender relations do not depend on any one person’s intentions or actions any more than they depend exclusively on hormones and genes. To begin to take apart the disadvantages or privileges that accrue to gender relations, then, requires that we look at the operation of institutions.

Finally, gender relations are dynamic. Ideals and enactments of gender identities change over the life course, both in response to historical and life-stage changes but also to age-based changes— not only just physically or as bodies change but also as our age statuses change. Changes to how one dresses, for example, signal shifts in gender as, for instance, many women move from empha- sizing (hetero)sexual attractiveness to signifying grandmotherhood. Certainly, an older woman can still be sexual. But highlighting such sexuality in dress, by wearing low-cut shirts and short skirts, can draw derision and pressure to dress in more age-appropriate manners. This dynamism merits attention in our research and practice.

The previous example points to an additional form of inequality that we must consider as we explore gender in old age: ageism itself. Old age is a devalued status, so much so that people will try to avoid it at all costs (e.g., Hurd, 1999). Understanding how gender influences later life thus includes account- ing for the ways that people do gender over their life- times, and the struggles to enact gender in later life as bodies change in a context of age inequalities.

This picture gains even greater specificity (and poses greater challenges to researchers) if we con- sider how gender relations intersect in a larger sys- tem of inequalities between groups distinguished by categorical identity. Just as gender shapes aging, so do other hierarchies influence both. For instance, multiple masculinities, shaped by inequalities, coexist but usually orbit around a hegemonic set

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist722

of ideals that reflect and serve the interests of elite men (Connell, 1995). The disempowerment of the majority of men often results from discrimination based on their positions in other systems of inequality; and many men living up to ideals of oppositional masculinity lead lives foreshortened by violence and lack of health care, both perpe- trated by others and acceded to in the name of manhood. Thus, while I focus on gender in this article, I recognize that old men and women do not exist apart from their racial and ethnic, sexual, and class-based locations, and I will point to these intersections where possible in my discussion.

Below, I apply this understanding of gender to paid labor and financial status in old age; to such unpaid labor as informal care work; and to health behaviors. Within my brief treatment of each topic, I aim to show the importance of understanding gender relations for research and practices that are both effective and equitable.

Economic in/Security

That women face greater financial insecurity in old age is evident in their higher rates of poverty relative to men; in 2008, old women’s poverty rates were almost double those of old men (11.9% vs. 6.7%). This gender gap is also reflected in incomes; in 2008, the median income of men aged 65 years and older was $25,503 and women’s was only $14,559 (Administration on Aging, 2010).

Describing such gender-based financial differ- ences is important but, by itself, tells little about the causes of or solutions to these inequities. Distin- guishing individual units of analysis by gender with- out a theory of gender relations might lead to a focus on individuals and the labor market alone and urge women to spend more years working for pay, choose better jobs, and engage in more financial planning and pension investment. Although I will not speak to each of these in depth, I argue that such suggestions would be neither effective nor equitable as they ignore the underlying power relations that create women’s poorer financial situation. I outline a gender relations approach subsequently, which situates labor market behaviors and outcomes within the larger gender division of labor.

Societies organize on the basis of gender, age, and race by assigning tasks, roles, and naturalized categorical identities (West & Fenstermaker, 1995). Work by such scholars as Glenn (1992) and Jones (1985) has shown that most such divisions of labor, authority, and status reproduce 19th century indus-

trial order in which a relatively small group of land- owning White men performed paid managerial work and owned family members and servants, whose labor supported them. A relatively small group of White women occupied exalted but pow- erless positions as spousal property, providing legit- imate sexual access and heirs. Most Whites labored in paid service or the production of physical goods. Most non-Whites slaved at more menial versions of the same supporting tasks: labor-intensive farm work, construction, and domesticity. As slaves gained emancipation, they mostly retained occupa- tional niches in manual labor and menial service, although some racial minorities have expanded into civil service and small-businesses ownership in ser- vice sectors. As women gained citizenship and inte- grated many paid occupations, they have created alternatives to marriage but retained most responsi- bility for unpaid domestic work and mostly entered service occupations associated with those tasks (nursing, teaching, cleaning, and other helping pro- fessions). These inequitable divisions of labor, authority, and status shape experiences of old age, including financial security.

Women’s responsibility for domestic work can reduce income by limiting their time in paid labor, thereby lowering their earnings and retirement income or reducing their income to a spousal ben- efit from Social Security. The persistent assump- tion of heterosexuality and marriage further constrains women, as those who are unmarried receive still lower earnings, on average, and do not have the possibility of a higher survivor’s benefit (Calasanti & Slevin, 2001; Herd, 2009). With income thus diminished by the demands of domes- tic support, women enter retirement with fewer economic resources than men.

The history of race and ethnicity outlined pre- viously shapes the life courses of men and women such that gender relations play out differently for various groups. For instance, a more detailed look at poverty rates in old age documents the follow- ing, in ascending order: White men, 5%; White women, 9%; Asian women and Asian men (both), 12%; Black men, 17%; Hispanic men, 18%; Hispanic women, 21%; and Black women, 27% (Carr, 2010). Regardless of gender, Whites fare bet- ter than any other racial and ethnic group member. This does not reflect the higher status of White women, however, but rather their greater access to family alignments with White men. Their better financial status in old age reflects their present or pre- vious positions as wives of relatively privileged men.

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 723

To clarify how gender relations influence finan- cial status in old age, I begin with Social Security, the most important source of financial security for those aged 65 years and older—almost two thirds (64%) rely on it for more than half of their income (Social Security Administration, 2009). Herd (2006, 2009) argues that the current Social Security benefit structure, which rewards consistent long work his- tories based on individual contributions or on long- term marriages through dependent and spousal and survivor benefits, leaves many women vulnerable to poverty. If they are married and claim Social Secu- rity income on that basis, their benefit levels depends upon the earnings of their spouses. If they work for pay, the gender division of responsibility for house- hold labor tends to depress their earnings.

This constraint on women’s earnings appears in research on divisions of family responsibilities. In 1980, for instance, the presence of children in households accounted for 35% of the wage gap between men and women who lived in them; by 1991, this had increased to 51% (Harrington Meyer & Herd, 2007). Although men do more domestic labor than they did prior to women’s increased movement into the paid labor force in the 1960s, women still spend almost twice as much time performing it (Sayer, 2005). Furthermore, women remain far more likely than men either to withdraw from the labor force or to reduce their participation to care for children (Herd, 2009). For instance, Sayer (2005) found that, among mar- ried couples, women’s time in paid work is 80% than that of men’s; among mothers, this drops to 60%. Becker and Moen (1999) discovered that, among middle-class dual earner couples, women were far more likely than men to engage in “scal- ing back” strategies; two thirds of those who lim- ited their work time or changed from careers to jobs were women. And Sanchez and Thomson’s (1997) longitudinal analysis revealed that if a woman becomes a mother for the first time, her paid-work time is reduced by a full workday per week. Having two or more children decreases her time in employment by an additional 12 hr. By contrast, fathers’ employment does not decline, and birth of a second child leads to a slight increase in their paid-work time. Earnings themselves are also affected. In 2009, women working full-time earned 78.6% of what their male counterparts earned; mothers’ earnings to fathers stood at 70%. This gap has changed little from 2001, when these figures were 73.6% and 66%, respectively (Bureau of Labor Statistics, 2002, tables 8 and 15; Bureau of

Labor Statistics, 2010a, tables 9 and 12 ). Finally, Cha’s (2010) analysis of national longitudinal data showed that among dual-career couples, having husbands who work long hours makes it more likely that wives will quit their jobs, especially if they have children; by contrast, husbands’ (and fathers’) employment was not affected by their wives long work hours. Thus, women’s Social Security benefits are deflated by their greater shares of responsibility for unpaid labor in their homes. Labor-force with- drawal (even if temporary) increases the likelihood that women will have years of zero earnings among the 35 years used to calculate benefit levels while reduced hours in paid work depresses earnings, especially in professional jobs (Cha, 2010).

Employers also devalue women’s paid work, diminishing their potential retirement income. The female-to-male earnings ratio for full-time workers has changed little in the new millennium, fluctuat- ing between 0.75 and 0.78 since 2000 (DeNavas- Wait, Proctor, & Smith, 2009). Women are segregated into lower paying fields that are often aligned with such presumably inherent traits as nurturing. Work taken to express women’s natu- ralized qualities, even when paid, reaps relatively few monetary rewards (Bielby, 2000).

Histories of specific jobs have shown that what determines scales of pay is the gender of those who mainly fill the posts rather than established under- standings of the work, which alter as women move into niches previously dominated by men (Milkman, 1982). The degree to which employers regard various paid jobs as appropriate for women can change dramatically for such reasons as dwindling male labor supply due to war, technological shifts that automate jobs and make them less attractive to men, labor conflict, or bureaucratization that changes job ladders. Regardless of reasons for such regendering of jobs, “the specific idiom of sex typ- ing . . . is flexibly applied to whatever jobs women and men happen to be doing” (Milkman, 1982, p. 341). For instance, during World War II, jobs previously seen to involve “the very quintessence of masculinity” were recast for women such that cutting patterns for aircraft parts was equated with dress pattern cutting; similarly, women’s kitchen skills were touted as helping her to “cook gears” or operate a drill press as if it were a juicer (Milkman, 1982, p. 341). Light (1999) shows that managers can convince themselves that even the most complex of tasks express women’s natural tendencies toward doing work that is “repetitive,” “domestic,” and “soft,” (pp 460–461, 469). And

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist724

when employers have redefined jobs to express the naturalized traits of women rather than those of men, they have adjusted pay scales accordingly (down).

Thus, gender gaps in pay result from more than women’s selections into occupations in which pay scales have historically been low. The gender gap remains largest among the most highly paid work- ers, in fact. A woman in the 95th earnings percen- tile for women makes 28% less than her similarly situated male counterpart (Leonhardt, 2006). Roth (2003) found such discrepancy among men and women who graduated at the same time from five elite graduate programs in finance and who found work in one of the top nine Wall Street firms. Four to seven years later, the women earned, on aver- age, only 60% of what men took home ($223,368 less per year). Examining this disparity, Roth con- trolled for presence of children, background char- acteristics, number of hours worked per week, sex segregation, and organizational measures relating to position in the firm. Gender still exerted a strong independent effect on earnings.

As a result and despite recent increases in women’s labor force participation, female retirees are about as likely to draw benefits based on their spouses’ earnings today as they were in 1960 (Herd, 2009). In 2008, 58% of women receive either spouse or survivor benefits; half of those women could have opted for benefits based on their own work histories. But the average monthly benefit for female workers was $988 in 2008 com- pared with a widow’s benefit of $1,292 (White women’s averages are higher; Social Security Administration, 2010a, table 5.A7). In fact, Harrington Meyer, Wolf, and Himes (2006) argued that marital status is more important than work history for older women’s economic security. However, they found that declining marriage rates and shorter marriages among those women born in the 1960s means that about 80% of White and Hispanic women will be eligible for either spousal or survivor benefits, whereas only about half of Black women will be so entitled. Racial and ethnic minority women’s inability to draw upon spousal or survivor’s benefits portends even greater finan- cial vulnerability (Herd, 2009). Even though Black women have historically high labor-force partici- pation rates, low pay associated with racism and their concentration in manual labor and menial service precludes them from receiving high Social Security benefits based on their paid work.

Earnings and pensions—the next sources of income in later life (Social Security Administra-

tion, 2010b)—are similarly gendered. The gender gap in earnings among older workers is even more pronounced than that among younger cohorts; in 2010, among full-time workers aged 65 years and older, women earned only 69% of men’s earnings. Figures by race and gender show even greater disparities. Looking at workers aged 55 years and older and using White men’s median weekly earn- ings as the standard, we find that Hispanic women (of any race) earned only 53% of White men’s earnings; Black women’s wages stood at 64%; White women’s, 72%,; and Asian women’s, 79% (Bureau of Labor Statistics, 2010b).

Pensions are just as unequally divided. The shift to defined contribution pensions has been touted as a positive development for women in that they offer greater access to pensions and avoid some of the vesting issues of the defined benefit plans (Calasanti & Slevin, 2001). In this sense, they seem gender neutral. And indeed, in 2008, full-time employed women and men participated in employer-spon- sored pensions plans at equivalent rates (51%) and across age groups (Johnson, 2008; Purcell, 2009). Pension access varies by race and ethnicity, however (Purcell, 2009); thus, we find that among workers in their early 50s, 80% of White women reported pension coverage in their current jobs in 2004 com- pared with 73% of Black women and only 50% of Hispanic women (Johnson, 2008).

Beyond participation, however, gender rela- tions are embedded in the defined contribution pensions’ assumptions concerning work and fam- ily that relate to (White, middle-class) men’s more privileged positions. These plans require that indi- viduals set aside funds and assume investment risks. Employers are not mandated to contribute to these plans, and employee participation levels are tied to earnings, with the expected result that those with less discretionary income participate less (Purcell, 2009). Women and men contribute similar percents of their salaries to 401(k) plans, but women’s (continued) lower earnings mean that they contribute less in absolute dollars (Harrington Meyer & Herd, 2007). This has a cumulative effect such that, among workers aged 60–64 years, women contribute 32% less than men to their pensions per year ($3,485 vs. $5,111; Johnson, 2008). Thus, despite their similar partici- pation rates, women draw only about half as much in benefits as do men (Shaw & Hill, 2002), a proportion that has shown little change from the past and one that is not projected to change in the near future.

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 725

Gender relations influence pensions in other ways that relate to the division of labor. Women’s investment inexperience and more conservative strategies (Calasanti & Slevin, 2001) may relate to men’s traditional position as breadwinners, but gendered positions within work and family spheres also have other impacts. When workers leave jobs that distribute pensions in lump sums, men are more likely to reinvest them in other retirement accounts, homes, stocks, or bonds. By contrast, women earn lower wages and are more likely to draw on these lump sums to pay for medical care, bills, loans, or for children’s education. As a result, women are also only half as likely as men to receive pensions upon retirement (Shaw & Hill, 2002).

This discussion makes clear that old women’s financial situations do not simply result from lack of employment, choices of the wrong jobs, or fail- ures to engage in financial planning. Research or programs that focus on such recommendations might be somewhat effective (e.g., women who are relatively well off will probably manage better with some planning than without; women with longer employment histories will fare better, on average, than those without). But research and policies that target the gendered nature of social institutions hold greater promise both for effec- tiveness and for equity—subsidized high-quality childcare, for instance. Changes to Social Security, for instance, could alter the traditional gender rela- tions embedded in the legislation. Herd (2005, 2006) examined some of these proposals and found that universal minimum benefits (eligibility is based on being a resident of the United States) would be most advantageous for poorer women (Herd, 2005). Furthermore, benefits linked directly to parenthood rather than to marriage (based on care credits, a common approach in other coun- tries) would benefit Black and poor women better depending on how these are including in calcula- tions (Herd, 2006). Together, these proposals would address some of the gender and class inequalities within the Social Security system.

Strategies to eliminate gender disparities will receive little notice in the present political eco- nomic climate, however. The greater push is to cut back on Social Security spending, and a variety of proposals for “reforming” Social Security have been aired, each with ramifications for gender inequality. Proposed changes in benefit calculation formulae, such as increasing the number of years used to compute benefits to 38 or 40, move in the opposite direction of gender-parity proposals out-

lined previously; women are far more likely than men to have zero earnings years due to child- rearing responsibilities. Women, and particularly working class and women of color, would be most penalized by any across-the-board benefit cuts; they would also be most disadvantaged by calls to increase the age for collection of full benefits to 70, as they are most likely to suffer chronic health issues. Women also face ageism in employment at younger ages than do men, such that they face greater chances both of losing jobs and difficulty finding new employment. Especially, problematic are plans to privatize portions of Social Security. As with defined contribution plans, such proposals appear gender and race neutral; indeed, by promis- ing individual control, they do not address the inequalities that are embedded in the very nature of such investment. As the previous discussion makes clear, women, people of color, and working class members would begin from a disadvantaged position (Calasanti & Slevin, 2001). Such groups are most likely to gain from keeping risks collec- tive, spreading the costs of failed investments over large populations, and keeping incidence of house- hold destitution low.

The Congressional Budget Office projected the returns on a privatization scheme, assuming aver- age rates of return on investments in stocks and bonds, and found that a low-income retiree in 2035 would receive annual Social Security benefits (including the annuity from a private account) of $9,100—down from the $9,500 forecast under the present program. A median-waged retiree would see a larger decline, from $17,700 to $13,600. (Lowenstein, 2005)

The point of this discussion, for those conduct- ing gerontological research of whatever kind— epidemiological, quantitative, personal interview, site-based evaluations, etc.—is that systematic comparisons of groups categorized by age, gen- der, class, race, etc., in light of structural vari- ables theorized to affect disparate experiences, is most likely to provide findings that shed light on the outcomes of interest, such as financial status or health behaviors in old age. We understand variations in those outcomes to result in part from the unequal divisions of resources and responsibilities that ensue from categorization into (largely naturalized) social groups. I turn next to a consideration of how gender relations shape unpaid labor in later life, specifically spou- sal care work, which turns out to be no less struc- tured by gender inequality.

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist726

Spousal Care Work

Although women perform the majority of unpaid care work for all age groups across the life course, in later life, spouses are the preferred caregivers. Not only do spouses perform 25% of informal care but also husbands provide the same amount and types of care as do wives (Arber & Ginn, 1995; Shirey & Summer, 2000 ). Because spousal caregiv- ing also controls for type of relationship between primary care worker and receiver, it provides an interesting opportunity to examine how gender relations shape care work in later life. Because research on spousal care has generally been conducted on White, middle-class heterosexual couples, my discussion is limited to how gender shapes spousal care for this group.

Scholars often point to men’s reports of rela- tively low levels of caregiver stress (e.g., Baker & Robertson, 2008), but what this means is unclear. Certainly, men experience stress, and why they report so little has been discussed in both caregiver and health literatures (e.g., Calasanti & King, 2007; Courtenay, 2000). However, debates con- cerning men’s and women’s stresses, caregiving styles, and the like tend to ignore the force of gen- der relations. Researchers tend to treat gender as a background, control variable, or treat manhood and womanhood as stable attributes of individuals rather than as positions in relations of structural inequality (Thompson, 2000). In line with the last approach, I argue that men and women apply situ- ational ideals of behavior based upon their respec- tive structural locations, thereby creating caregiving meanings and experiences that vary by gender and bear explanation in terms of the divisions of labor, authority, and esteem reviewed above. I do not assume that one approach is inherently better than the other. Instead, I explore how gender repertoires—sets of skills and resources learned over the life course that affirm gender identities formed in a context of inequality—influence how they approach care work, the stresses they encounter and how they cope with these, as well as the aspects that they find rewarding in order to demonstrate the links between gender and experiences of care work that research, policies, and interventions can address effectively and equitably.

Of course, many similarities accrue to husbands and wives who give care, beyond their resem- blances in their tasks performed and time spent. Both express comparable motivations for their care work: duty, the marital bargain, and senses

of reciprocity and commitment (Calasanti, 2006; Hayes, Boylstein, & Zimmerman, 2009; Ribeiro & Paul, 2008; Russell, 2007). By the same token, they also approach their care work differently in some respects.

Based on their experiences in the public realm of paid work, men tend to approach care work as tasks to master and problems to solve and often view their work as organizational in nature (Russell, 2007). This orientation can provide greater feelings of control and self-efficacy; it may also help men to separate their emotions from tasks at hand and take respite when needed (Calasanti & King, 2007; Russell, 2007; Thompson, 2000). As we find among men employed in female-dominated occupa- tions, such as nursing, men bring their identities to bear on the work and describe it in masculine terms (Williams, 1995). Similarly, husbands describe care work as challenges that men should face, learning it like a trade, applying strength and problem solving, and likening it to military service (Calasanti, 2006; Ribeiro, Paul, & Nogueira, 2007).

Their approach does not mean that men do not care affectively; they express concern and emotional commitment to their spouses (Calasanti & King, 2007; Russell, 2007; Thompson, 2000). But care- taking is not a part of the labor for which men hold themselves primarily responsible. They do not iden- tify themselves as natural caregivers or nurturers (e.g., Ribeiro et al., 2007). Men thus tend to evalu- ate their caregiving based more on tasks performed successfully than on their notions of themselves as providing emotional support or making their wives happy when conflicts arise (a common choice that arises in caring for spouses with dementia; Calas- anti & King, 2007). Understanding their approach helps explain why husbands who use problem- focused coping strategies report positive caregiving experiences (Baker & Robertson, 2008).

Women also uphold gender divisions of labor across the life course, which designate them primar- ily responsible for such domestic labor as care work. As a result of their previous experiences as caregivers and the emphasis on women’s nurturing, they expect that they will care effortlessly for spouses physically and emotionally (Calasanti, 2006). They thus approach their care work for spouses with greater concern for the care receiver as a whole and with preserving the emotional dynam- ics of the relationship, including mutuality (Hayes et al., 2009; Lyons, Stewart, Archbold, & Carter, 2009; Rose & Bruce, 1995). The generally high expectations—by themselves and by others—about

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 727

their ability to handle all the problems that come with caregiving (especially for cognitively impaired spouses) creates greater potential for stress than husbands may experience, despite the fact that the latter are often less prepared to engage in such work at the outset (Calasanti, 2006; Calasanti & King, 2007).

Russell (2007) finds that careworking husbands’ primary struggle is with the fact that their work goes largely unseen after lifetimes of recognition for work in public realms. Still, work that men are not expected to perform can draw kudos that caregiving wives seldom receive (Calasanti, 2006; Rose & Bruce, 1995). In fact, social recognition of their caregiving as well as a wife’s appreciation com- prise important rewards of caregiving for some hus- bands (Davidson, Arber, & Ginn, 2000; Ribeiro & Paul, 2008). By contrast, women have done care work throughout their marriages, and its lack of visibility to others does not surprise them. Instead, research suggests that caregiving wives are more likely to feel that their spouses make more unrea- sonable requests and that caregiving has increased the stress in their relationship (e.g., Lyons et al., 2009). Davidson and colleagues found that hus- bands who received care were indeed demanding and tried to retain control and their roles as house- hold heads, perhaps because their physical prob- lems meant weakness and thus threatened their authority as men. In addition, care-receiving hus- bands had not been responsible for domestic labor in their marriages, and they knew little of what the work entailed. By contrast, care-receiving wives understood the work involved in giving care, as they had long provided it. As a result, they were more grateful for their husband’s care work, and their husbands described them as “good patients.”

Hayes and colleagues’ (2009) study of care for spouses with Alzheimer’s disease and related dis- orders provides additional insights concerning gender relations and spousal care. Focusing on issues of intimacy within these couples, they found that the physical aspects of care work fatigued women and that this lowered their interest in sex- ual activity. But other issues arose for them as well. Having to be the one to initiate sexual intimacy was problematic as this was typically their hus- bands’ purview. In addition, “their husband’s inability to convey feelings of love, warmth, com- passion, and understanding as a precursor” to inti- macy troubled caregiving wives (Hayes et al., 2009, p. 54). It signaled a lack of emotional reci- procity in the relationship that was crucial for

them. Caregiving husbands were less bothered by their wives’ lack of emotional support and did not experience this as rejection.

The act of providing physical care had another dimension as well. Physical dependence or being “baby-like” made women feel that they were taking on mothering roles, which conflicted with those of lovers and wives. By contrast, such dependence did not stop caregiving husbands from seeing their wives as spouses; as a result, they could still initiate sexual intimacy. The authors’ explanation for these differ- ences point clearly to gender relations: “[W]ives . . . often had less status and power relative to their hus- bands throughout their marriage, and this changed little after wives’ diagnosis. . . . . In contrast, women caring for a cognitively impaired spouse were uncomfortable with continuing sexual relations when husbands’ status changed from competent male provider to child-like and dependent. The identity of husbands receiving care was tarnished more by the illness” (Hayes et al., 2009, p. 55).

These ways of living up to ideals of behavior, rooted in gender relations, also appear in research on 22 spouses caring for husbands and wives who are cognitively impaired. Although interviews did not focus on intimacy, discussions of difficulties with toileting behaviors emerged from women and men (for a longer discussion, see Calasanti, 2006). All but one of the care-receiving wives were incontinent or had difficulty with toileting, while only a few care-receiving husbands had such dif- ficulties. Yet, only caregiving wives mentioned these as reasons to institutionalize spouses. Men approached them as problems to be solved, time consuming and sometimes distasteful but not insurmountable. For women, poor toileting behav- iors had more relational import, ushering in the ends of marital relationships and inspiring grief. One woman spoke of her husband’s increased con- fusion concerning toileting through tears; she noted that while she had no difficulty giving pedi- cures, “cleaning up after the bowel movements” was the most emotionally difficult aspect of care work. “You wouldn’t think you would be doing that to a person that you are supposed to be shar- ing your life. . . . ” For her, this meant that “I am not a wife, I am someone who takes care of him . . . it’s like taking care of a little child, but worse” (Calasanti, 2006 , p. 286). Based on traditional gender relations in a marriage, incontinence repre- sented a much larger drop in status for men than for women. A husband’s loss of power and control makes all the more apparent the change in the

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist728

marriage. By contrast, women’s poor toileting behaviors did not signal the same kind of shift; wives are not assumed to be powerful, their places in marital relations defined by their greater physical prowess or control. At the same time, Hayes and colleagues (2009) find that incontinence made wives less sexually desirable to their caregiving husbands.

Their different response to incontinence does not mean that care work did not induce any stress in husbands. Indeed, husbands struggled with ten- sion and sought to relieve it by blocking emotions, focusing on tasks, and keeping busy in order to avoid feelings not aligned with masculine aggres- sion. Although expressions of anger are acceptable for men in most contexts, other unpleasant emo- tions, especially those that might indicate a level of stress that demonstrates an inability to handle problems or a lack of competence, are not. But such means of emotion management may not be effective, and the skills that are a part of their gen- der repertoire—task-oriented approaches to care work and suppression of bad feelings—do not pre- pare them to deal with stress in ways that satisfy them. As Jerry said,

Most men keep things to themselves. I am not going to tell everybody I know that I am stressing or I feel depressed or I feel I need somebody to hug . . . . I don’t know whether it’s a man thing or it’s my thing or whether it’s my thing being a man [but] I am not going to share them with people. I keep it to myself. And I think men do that. They keep a lot of things to themselves that will kill you.

When negative emotions surface, husbands may not know what to do with them; most of the men (but none of the women) in this study reported using prescription drugs and alcohol. As David said, “Fortunately Dr. Merlot came in at night and got me through” (Calasanti & King, 2007, p. 525).

This discussion of gender and spousal care is based on research generally limited to White, middle-class, and heterosexual couples. Research on domestic labor suggests labor less divided by gen- der among African American couples (Coltrane, 2000); as a result, some of the gender differences noted above might be altered. Furthermore, care- giving research takes for granted the ways that het- erosexuality shapes gender repertoires by assuming a marital relationship with different roles for hus- bands and wives. However, same-sex couples can- not fall back on notions of “husbands” or “wives” within their intimate relationships and must nego- tiate who will perform what domestic labor with

less recourse to naturalized ideals of sex-specific behavior. Thus, research finds that gay and lesbian couples tend to be relatively egalitarian (Heaphy, 2007). Gay men, for instance, may both develop the workplace skills and identities noted above, while they also learn aspects of domestic labor that husbands are relatively likely to forego because they cannot assign responsibility to wives. How such gender repertoires influence care work remains unknown but is an important area for future research.

Health and Bodies in Old Age

There is a longer history of research on gender inequities in health outcomes, as many gerontolo- gists have been concerned to understand, for instance, why women suffer higher rates of chronic diseases and men face more life-threatening condi- tions at younger ages. Scholars have observed rela- tionships between gender and health behaviors, including lifestyles and help seeking, as well as the meaning and experience of different conditions. However, as with some of the caregiver studies, researchers have sometimes only controlled for gender, noted differences between men and women, or emphasized early socialization; and interven- tions have tended to follow suit, devising strate- gies, or programs based on lists of sex-correlated traits. But increasing health knowledge by educa- tion does not necessarily change gender-based health behaviors (e.g., Solimeo, 2008a). To be effective, research and interventions need to under- stand gender relations as daily interactions shaped by the immediate forces of institutions. In order to keep from reinforcing gender inequalities, such research and practice should also be attuned to intersections between gender, age, and other inequalities. To illustrate both these points, I briefly discuss masculinity and health.

Recent scholarship has focused on those health behaviors that are to men’s detriment and has found that men engage in more risky behaviors are less likely to seek help for physical ailments and, indeed, may resist such help (Courtenay, 2000; Kaye, Crittenden, & Charland, 2008; Mansfield, Syzdek, Green, & Addis, 2008; O’Brien, Hunt, & Hart, 2005). The same is true of psychological services, the need for which is suggested by men’s higher rates of suicide and substance abuse. Indeed, older men are four times more likely to be treated for alcohol abuse than are women (Mansfield et al., 2008). Men are also less likely than women

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 729

to be concerned about healthful behaviors. For instance, focus group interviews of people aged 55 years and older found that although both men and women thought a healthy diet important for aging well, women took more responsibility for it. Women also assumed responsibility for getting and providing health care information (Wu, Goins, Laditka, Ignatenko, & Goedreis, 2009).

Many researchers have attributed such differ- ences to abstract ideals of masculinity, such as the importance of control, learned through socializa- tion (e.g., Mansfield et al., 2008). One such argu- ment, for instance, maintains that help seeking “may threaten men’s sense of control over their well-being . . . . [and indicate] that they are incapa- ble . . . ” (Kaye et al., 2008, p. 11). Reports from men also demonstrated beliefs in individual control, as when one notes that, “if you wake up in the morning you’ve got a choice of doing two things, you’re either happy or you’re sad. You’ve got that choice.” Or in relation to healthy eating, another argues that “It’s your personal choice if you’re willing to sacrifice taste for health” (Wu et al., 2009, p. S76).

Although such studies take important steps toward understanding gender and health, a gender relations approach goes further, as Thompson (2008) argues in his discussion of masculinity and health in later life. Noting that some researchers have begun to move beyond the emphasis on gen- der socialization, on “what men internalized,” he advocates greater focus on “how and why men ‘do gender.’ Through their decisions and everyday behavior men routinely affirm or challenge exist- ing power relations” (Thompson, 2008, p. 7). This shift toward a more fully theorized account of gender relations discounts neither socialization nor identity but views them instead as outcomes of routine struggles to maintain membership in a high-status group (adult men) and usually to achieve dominant positions relative to other men and to all women in gendered and age-graded institutions. In adopting this approach, I draw upon current scholarship on the use of bodies as naturalized markers of group status to see how these gender and age relations can shape health behaviors and experiences of health conditions.

Laz (2003, p. 507) notes that bodies serve as “highly personal and individualized resources that people make use of in accomplishing their own age and age in general.” This matters to people and to the larger maintenance of inequality because bod- ies serve as markers of age as well as gender, cue

people to categorize each other in routine interac- tions, and can thus serves as bases of inclusion/ exclusion to high-status social networks (Laz, 2003; West & Fenstermaker, 2005). Furthermore, because most people maintain views of bodies as governed mostly by natural forces, such categori- zations tend to occur without notice and to be very difficult to challenge. Given that ageism serves to exclude old people from full participation in Western societies and bodies serve as naturalized markers of age, people tend to want to present their bodies in age-appropriate ways that will not mark them as having “given into” old age.

At the same time, other systems of inequality intersect with age relations such that the categori- zation of bodies and the assessment of changes in them vary by gender, race, ethnicity, class, and sexual orientation. Upper body strength, density and location of body hair, and styles of clothing choices affect categorization and thus group mem- bership; those group memberships in turn affect interpretations of such changes as puberty or gray- ing of hair and wrinkling of skin, and such activi- ties as athletic feats, medical care, sexual display, and relaxation (Calasanti & Slevin, 2001; Laz, 2003). For instance, while similarities may accrue, definitions of debilitating illnesses vary with the age and gender of the sufferer and our collective imputations of what the condition implies. Thus, gender and age influence how we treat our bodies and how we respond to them; they shape health behaviors as well as how people experience differ- ent health conditions.

This framework allows us to see more signifi- cance in daily behaviors, both mundane and formal, as we think about socialization and the sex differ- ences in health behaviors that research has mea- sured. It leads us to focus more on context and how, for example, behaviors reinforce or challenge group memberships and power relations that people nego- tiate every day. Men and women can behave in very similar ways in relation to their health—ways that might seem contradictory to masculine or feminine ideals as defined in the abstract, in terms of stable sets of sex-specific traits. But those behaviors depend upon local organizational context for their sense of natural rightness. For example, and despite the liter- ature reviewed, men do assert themselves in seeking health care, under certain conditions (O’Brien et al., 2005). But this does not mean that gender does not matter; instead, it focuses our attention on how men may behave in ways that align them with high- status groups.

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist730

Rather than assume that orientations toward masculinity constrain men from seeking medical help per se, O'Brien and colleagues (2005) use data from focus groups of diverse Scottish men (by age, occupation, socioeconomic status, and current health) to produce findings that demonstrate both the importance of gender and age relations and context. They report that men express a reluctance to get medical help but are more willing if it is tied to an expression of dominance within some institu- tion: when concerned with sexual performance and when working such widely celebrated male- dominated jobs as firefighting. In the latter instance, their occupational status involved “having a fit, masculine body which enabled them to perform effectively at work . . . . [S]eeking help at the first sign of symptoms and asking for preventative health checks were key to ensuring that their ability to work effectively was not jeopardized . . . . Thus, unlike for other men, help seeking was a way of preserving, not threatening, masculinity for this group, a means of safeguarding their place in an archetypically masculine occupation” (p. 514). For a second group of men, those who had ME (myalgic encephalomyelitis or chronic fatigue syn- drome), receipt of a diagnosis helped them to explain why they were not fulfilling their bread- winner roles and thus “presented the prospect of restoring a masculine identity that was undermined by the nature of the symptoms of their illness” (p. 515).

Focus on gender relations can also help us to understand the ways in which groups experience similar health conditions such as debilitating illnesses in different ways. Solimeo (2008b) found that, although men and women have similar prev- alence rates of Parkinson’s disease (PD) and often present similar symptoms, men and women focused on different aspects of PD experiences that bore on their competence at tasks assigned by a gender division of labor. Women focused on aspects of PD that made it hard for them to per- form domestic labor tasks, such as meal prepara- tion or coordinating social events, tasks which helped women to maintain their positions in social relationships of most immediate value to them. Men with PD found other dimensions of the dis- ease more problematic. Although both men and women commented on their changing appearance, men focused on how others perceived them. A third of the men mentioned “loss of strength and ambition as their primary symptoms; no women did. Men described their unhappiness with what

they perceived to be emasculated and unreliable bodies, particularly with poor posture and loss of strength . . . that led them to avoid public spaces” (Solimeo, 2008b, p. S45). Men saw their masculin- ity as partly related to their ability to drive, lift heavy things, and do outside household activities, and they performed these more public displays of strength as long as they could.

Men and women also discussed similar symp- toms in different ways. Both men and women were talked about tremors, but women pointed to the difficulties they experienced in activities that involved others and to relational work; men focused on how tremors influenced their appear- ance and sense of competence. Solimeo (2008b) concludes that men and women will be most attuned to those impacts of symptoms that inter- fere with activities upon which they base gendered identities and status.

These studies also make clear that gender inter- acts with age relations in shaping health behaviors and experiences of health conditions. O'Brien and colleagues (2005) found that older men were more open to seeking medical care than were younger men, weighing masculine dictates to wait and tol- erate pain against age-based concerns that they could sustain greater physical damage if they do. As they conclude, “men do not uniformly avoid care, but they utilize services via a ‘hierarchy of threats to masculinity’” (O'Brien et al., 2005, p. 514) in which age relations play a role. Late life enactments of gender influence health (Solimeo, 2008a) differently than they do in younger years and often to keep bodies from appearing to others as old. This helps explain Gibb’s (2008) findings that middle-aged men feel threatened by arthritis because it limits employment and by the seeking of treatment for it because they regard it as an old (i.e., retired) person’s disease. By contrast, those men approaching retirement (aged 61 years and older) were more open to self-management pro- grams because they saw it as a way to keep work- ing and keep fit. In this sense, they worked to maintain membership in high-status groups not merely to live as socialized long before.

These studies direct us to understand health behaviors in terms of the categorization of people, often by reference to their bodies, to maintain gen- dered status and inequality. Groups do this in ways that are sensitive to contexts, within the larger sys- tem of intersecting inequalities. A woman may not object to using a walker for greater stability or to prevent falling, but she may if it interferes with her

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 731

ability to cook or do housework, as it poses a threat to her status within the family home. By contrast, men may object to being seen with walk- ers in public, in the groups in which their primary dominance rests, but may not mind using them at home. Similarly, increasing men’s knowledge con- cerning osteoporosis may not influence their pre- ventative behaviors (Solimeo, 2008a, p. 74) if such actions are seen to be treating their bodies in ways that would make them seem old and “more like women” (given that osteoporosis is seen to be pri- marily an older women’s disease).

Understanding these links between group status and health behaviors can result in more effective strategies, such as Kaye and colleagues’ (2008, p. 12) recommendation that health outreach should appeal to men by providing them “oppor- tunities to problem-solve and gather information about a topic as well as serve as an ‘expert’ to other men,” as opposed to more socially oriented oppor- tunities provided to women that allow them to talk about feelings. Similarly, Mansfield and col- leagues (2008, p. 17) promote a model for public health practitioners that encourages men to seek help by using norms of “toughness” and “empha- sizing that men should not hesitate to ‘step up to the plate’ and address their problems.” Such strat- egies are no doubt more immediately effective than those that do not take gender into account, but they may also serve to reinforce gender inequities. Appeals to men’s breadwinner role or higher sta- tus within families, for example, can encourage the continued sorting of people, in terms of natural- ized markers of gender, into groups of different status. A focus on inequality clarifies some of the ethical issues raised by recommendations for inter- ventions. Immediate effectiveness in altering health behaviors in order to preserve the health of a group is a laudable goal that weighs against the ameliora- tion of inequities that affect whole populations and give rise to many of the negative outcomes that our scholarship was intended to address.

Discussion

A gender relations approach begins with the observation that societies divide tasks, authority, and status on the basis of gender. In the process of performing gendered labor, people make differ- ences seem like products of nature (“women are just more nurturing”); they become a part of “the way we do should do things,” and what it means to be a man or a woman. This naturalized inequality

justifies divisions of resources that polarize in later life.

Gerontologists who adopt this framework con- sider the gender division of labor in paid and unpaid realms as they research why old men and women have different financial positions, experi- ences of care work, and health behaviors and outcomes. That women maintain primary respon- sibility for unpaid domestic labor, even as they engage in remunerative work, matters for how men and women experience later life. Because gender is an on-going accomplishment, we must explore the contexts in which people negotiate these divisions. As the section on spousal care demonstrated, the fact that men have typically not been responsible for care work does not mean that they cannot or will not do this labor; in fact, they often engage in what is considered gender-atypical behaviors. An understanding of gender relations helps us to see why it is that, despite oft-stated assumptions that spousal care work will be diffi- cult for husbands because it is “feminized activ- ity,” caregiving men still see themselves as “real men” (Ribeiro et al., 2007). They bring their gen- der repertoires to bear on this activity, and they perform it in ways that make sense to them as men (Calasanti & King, 2007; Ribeiro et al., 2007).

Gerontologists may design studies with a gen- der relations framework in many ways. At issue is neither the data collected nor mode of analysis but the theory that drives both questions asked and interpretations made. Controlling for gender in quantitative analysis, merely to reveal statistically significant variation, fails to explore how gender matters. Researchers may contribute to the more complex understanding by use of organizational records, self-reports, direct observations, or any other data as they track a wide range of variables including divisions of labor and income, attempts to maintain visages of youth or good health, objections to failures to live up to ideals of strati- fied group memberships, for instance. Whether collected en masse and analyzed in quantitative form, as by McCall (2005) and Herd (2005, 2006), or in small groups in qualitative fashion (e.g., Calasanti, 2006; Solimeo, 2008b), analysis of any such data, informed by theories of the ways in which people manage behavior with ideals of stratified group membership in mind, allow us to see how inequalities shape the problems of old age. The point of taking gender seriously is not to con- strain methods of social science or to prescribe variables to be collected in all studies but rather to

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist732

direct theoretical attention to propositions about the role of naturalized ideals of gender in the sort- ing of people into stratified groups.

To situate research and practice within a theo- retical framework of context-specific categoriza- tion of bodies by gender, age, and other dimensions of inequality affects our efforts to improve elders’ quality of life. If scholars impute an observed gen- der difference to prior socialization—say, women earn less because they conform to learned ideals of deference and acceptance of inequity—then they may recommend policies designed to retrain women. However, to understand gender relations as contemporaneous, context-specific negotiation of group membership suggests that we instead focus efforts on organizational accountability to standards of rational decision making and fairness, which have been shown to be effective in many cir- cumstances. For instance, in the workplace, this translates into analyzing jobs and making relevant job criteria clear, establishing mechanisms by which job candidates can make their interests and qualifications known and insuring accountability to higher ups. These strategies diminish the impacts of gender and racial stereotyping on employment (Bielby, 2000). Promoting behavioral change with- out understanding how gender is embedded in such organizational contexts as family and work will be ineffective—not just because people are not com- fortable with the changes but because social insti- tutions are based on and reinforce such inequalities, often through organizational routines and such formal mechanisms as Social Security benefit for- mulae. As a result, we must consider research and interventions at the organizational level as well as at that of prior socialization.

The roots of economic insecurity in later life are multiple and are tied to divisions of paid and unpaid labor, among other resources. As Herd (2009) argues, addressing some of these requires changes to Social Security that would ameliorate some of the gender and racial and ethnic biases in the program. Understanding the power relations that underlie these inequities also points to the misguided nature of targeting individuals to pro- vide financial incentives for staying in the labor force longer. By focusing on labor supply rather than demand, such efforts can fail to help women and racial and ethnic minority group workers (Flippen & Tienda, 2000). Such interventions and lines of research ignore the role of naturalized inequalities and assume a “neutral” economy in which individual choice is the key.

Similarly, whether or not men better handle caregiving stress is a question to be explored with measures sensitive to interactional mechanisms by which people respond to categorical identities and associate tasks, authority, and esteem with them. By those means, we can take into account men’s greater likelihood of leaving stress unexpressed (Courtenay, 2000). The findings may not translate across genders, especially if they do not attend to the pervasiveness of inequality across many con- texts, including family homes. Wives may have dif- ficulty ignoring husbands’ negative emotions, and they tend to resist behaviors that further reduce dependent husbands’ status. Husbands will not find useful support groups that expect them to describe in detail how they feel about their care work but will instead seek information that will help them complete tasks with little fuss. By the same token, providing men such information will not change the gender relations that result in the invisibility and stresses that they experience or that underlie women’s feeling that their care is inadequate.

In a market-based society such as ours, one straightforward method of increasing the value ascribed to any form of labor is to pay for it; thus, we should explore proposals and programs that pay relatives to provide care. To date, such pro- grams, nationally (such as the Money Follows the Person Demonstration programs implemented in 2007) and internationally, are often couched in terms of savings or for the consumers, that is, pay- ing relatives for care allows greater choice on the part of elderly care receivers. However, interna- tionally, such payments are sometimes imple- mented in recognition of the impact that informal care has on women’s reduced labor market participation and poverty (Glendinning, 2009). Among “cash-for-care” models, the Australian Carer Allowance is promising, as it pays caregivers in recognition of their care work and thus provides a universal, nontaxable benefit that is neither means-tested nor dependent upon their paid work status. Initial research indicates that caregivers under such programs feel that payment raises their status. However, all the programs—including that of Australia—tend to reinforce gender inequality by offer remuneration at lower rates than would be commanded if bought in the market place (Glendinning, 2009; Grootegoed, Knijn, & Da Roit, 2010).

The framework I have outlined challenges us to rethink naturalized gender differences such that,

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

Vol. 50, No. 6, 2010 733

for example, we would not take for granted that men are unconcerned with taking care of their bodies. If we shift our notions of health behaviors and bodily concerns away from an implicit com- parison with women, we see that many men do take care of their bodies—witness the many who spend hours each week building their muscles. And in later life, this truism is contradicted by the huge profits made by the pharmaceutical industry in supplying drugs for erectile dysfunction. This would not mean that gender does not matter, or denying research that shows that women are more likely to engage in caring for health, but would challenge us to view gender as a daily accomplish- ment and explore the ways that contexts shape gender behaviors.

Finally, gerontologists need to consider both the ways that gender intersects with other inequalities and its dynamic nature. The research discussed here relates to gender relations among the current cohort of old women and men; the extent to which this may hold for future experiences of aging is unknown, although it is unlikely that large changes will occur any time soon. Although recent evidence suggests that married men are engaging in more household labor, women still perform significantly more. Furthermore, free time appears to be an emerging area of gender inequality in terms of divi- sion of labor (Sayer, 2005). How these shifts in the division of labor will play out, and for whom, are empirical questions—ones that have implications not only for effectiveness but also for equality.

Once we consider the power relations that underlie gender inequalities, we also take responsi- bility for either challenging or upholding them. We cannot ignore our roles in reinforcing or changing present inequities when we focus on effective research or practices; it is possible for us to develop research agendas or derive strategies that achieve our short-term goals while simultaneously but- tressing power differences.

Acknowledgments

I am indebted to Neal King for his extensive comments on earlier drafts and to Thomas Ratliff for his research help.

References Administration on Aging. (2010). A profile of older Americans: 2009.

Retrieved March, 2010, from http://www.aoa.gov/AoAroot/Aging _Statistics/Profile/2009/

Arber, S., & Ginn, J. (1995). Gender differences in informal caring. Health and Social Care in the Community, 3, 19–31.

Baker, K. L., & Robertson, N. (2008). Coping with caring for someone with dementia: Reviewing the literature about men. Aging & Mental Health, 12, 413–422.

Becker, P. E., & Moen, P. (1999). Dual-earner couples’ work-family strat- egies. Journal of Marriage and Family, 61, 995–1007.

Bielby, W. T. (2000). Minimizing workplace gender and racial bias. Con- temporary Sociology, 29, 120–129.

Bureau of Labor Statistics. (2002). Highlights of women’s earnings in 2001. Retrieved September 7, 2010, from http://www.bls.gov/cps /cpswom2001.pdf

Bureau of Labor Statistics. (2010a). Highlights of women’s earnings in 2009. Retrieved September 7, 2010, from http://www.bls.gov/cps /cpswom2009.pdf

Bureau of Labor Statistics. (2010b). Economic news release. Table 2. Retrieved May 2, 2010, from http://www.bls.gov/news.release/wkyeng.t02.htm

Calasanti, T., & King, N. (2007). Taking ‘women’s work’ ‘like a man’: Husbands’ experiences of care work. The Gerontologist, 47, 516–527.

Calasanti, T. (2006). Gender and old age: Lessons from spousal caregivers. In T. Calasanti, & K. M. Slevin (Eds.), Age matters: Re-aligning femi- nist thinking (pp. 269–294). New York: Routledge Press.

Calasanti, T., & Slevin, K. F. (2001). Gender, social inequalities, and aging. Walnut Creek, CA: Alta Mira Press.

Carr, D. (2010). Golden years? Poverty among older Americans. Con- texts, 9, 62–63.

Cha, Y. (2010). Reinforcing separate spheres: The effect of spousal over- work on men’s and women’s employment in dual-earner households. American Sociological Review, 75, 303–329.

Coltrane, S. (2000). Research on household labor: Modeling and measur- ing the social embeddedness of routine family work. Journal of Mar- riage and Family, 62, 1208–1233.

Connell, R. W. (1995). Masculinities. Berkeley: University of California Press.

Courtenay, W. H. (2000). Constructions of masculinity and their influence on men’s well-being: A theory of gender and health. Social Science and Medicine, 50, 1385–1402.

Davidson, K., Arber, S., & Ginn, J. (2000). Gendered meanings of care work within late life marital relationships. Canadian Journal of Aging, 19, 536–553.

Flippen, C., & Tienda, M. (2000). Pathways to retirement: Patterns of labor force participation and labor market exit among the pre-retirement population by race, Hispanic origin, and sex. Journal of Gerontology: Social Sciences, 55(B), S14–S27.

DeNavas-Wait, C., Proctor, B. D., & Smith, J. C. (2009). Income, poverty, and health insurance coverage in the United States, 2008. U.S. Census Bureau, Current Population Reports, P60–236. Washington, DC: U.S. Government Printing Office.

Gibbs, L. (2008). Men and chronic arthritis: Does age make men more likely to use self-management services? Generations, 32, 78–81.

Glendinning, C. (2009). Cash for care: Implications for carers. Health and Ageing, 21, 3–6.

Glenn, E. N. (1992). From servitude to service work: Historical conti- nuities in the racial division of paid reproductive labor. Signs, 18, 1–43.

Glenn, E. N. (1999). The social construction and institutionalization of gender and race: An integrative framework. In M. M. Ferree, J. Lorber, & B. B. Hess (Eds.), Revisioning gender (pp. 3–43). Thousand Oaks, CA: Sage.

Grootegoed, E., Knijn, T., & Da Roit, B. (2010). Relatives as paid care- givers: How family carers experience payments for care. Ageing & Society, 30, 467–489.

Harrington Meyer, M., & Herd, P. (2007). Market friendly or family friendly? The state and gender inequality in old age. New York: Russell Sage Foundation.

Harrington Meyer, M., Wolf, D. A., & Himes, C. L. (2006). Declining eligibility for Social Security spouse and widow benefits in the United States? Research on Aging, 28, 240–260.

Hayes, J., Boylstein, C., & Zimmerman, M. K. (2009). Living and loving with dementia: Negotiating spousal and caregiver identity through narrative. Journal of Aging Studies, 23, 48–59.

Heaphy, B. (2007). Sexualities, gender and ageing: Resources and social change. Current Sociology, 55, 193–210.

Herd, P. (2005). Ensuring a minimum: Social Security reform and women. The Gerontologist, 45, 12–25.

Herd, P. (2006). Crediting care or marriage? Reforming Social Security family benefits. Journals of Gerontology: Social Sciences, 61B, S24– S34.

Herd, P. (2009). The problem of poverty among older people in America: Options for reform. Benefits: The Journal of Poverty & Social Justice, 17, 125–135.

by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from

The Gerontologist734

Hurd, L. (1999). ‘We’re not old!’: Older women’s negotiation of aging and oldness. Journal of Aging Studies, 13, 419–439.

Johnson, R. W. (2008). Women’s retirement security: The role of private wealth. Presented at the annual meetings of the Gerontological Society of America, Washington, DC, November.

Jones, J. (1985). Labor of love, labor of sorrow: Black women, work, and the family from slavery to the present. New York: Basic Books.

Kaye, L. W., Crittenden, J. A., & Charland, J. (2008). Invisible older men: What we know about older men’s use of healthcare and social services. Generations, 32, 9–14.

Laz, C. (2003). Age embodied. Journal of Aging Studies, 17, 503–519. Leonhardt, D. (2006, December 24). Gender pay gap, once narrowing, is

stuck in place. New York Times. Retrieved, from http://www.nytimes. com/2006/12/24/business/24gap.html?_r=1

Light, J. S. (1999). When computers were women. Technology and Cul- ture, 40, 455–483.

Lowenstein, R. (2005). A questions of numbers. Retrieved September 7, 2010, from http://www.nytimes.com/2005/01/16/magazine/16SOCIAL .html

Lyons, K. S., Stewart, B. J., Archbold, P. G., & Carter, J. H. (2009). Opti- mism, pessimism, mutuality, and gender: Predicting 10-year role strain in Parkinson’s disease spouses. The Gerontologist, 49, 378–387.

Mansfield, A. K., Syzdek, M. R., Green, J. D., & Addis, M. E. (2008). Understanding and increasing help-seeking in older men. Generations, 32, 15–20.

McCall, L. (2005). The complexity of intersectionality. Signs, 30, 1771– 1800.

Milkman, R. (1982). Redefining ‘women’s work’: The sexual division of labor in the auto industry during World War II. Feminist Studies, 8, 336–372.

O'Brien, R., Hunt, K., & Hart, G. (2005). “It’s caveman stuff, but that is to a certain extent how guys still operate”: Men’s accounts of mascu- linity and help seeking. Social Science and Medicine, 61, 503–516.

Purcell, P. (2009). Pension sponsorship and participation: Summary of recent trends. Congressional Research Service. Retrieved April 26, 2010, from http://assets.opencrs.com/rpts/RL30122_20090911.pdf

Ribeiro, O., & Paul, C. (2008). Old male carers and the positive aspects of care. Ageing & Society, 28, 165–183.

Ribeiro, O., Paul, C., & Nogueira, C. (2007). Real men, real husbands: Caregiving and masculinities in later life. Journal of Aging Studies, 21, 302–313.

Rose, H., & Bruce, E. (1995). Mutual care but differential esteem: Caring between older couples. In S. Arber & J. Ginn (Eds.), Connecting gen- der & ageing (pp. 114–128). Buckingham: Open University Press.

Roth, L. M. (2003). Selling women short: A research note on gender dif- ferences in compensation on Wall Street. Social Forces, 82, 783–802.

Russell, R. (2007). The work of elderly men caregivers: From public careers to an unseen world. Men and Masculinities, 9, 298–331.

Sanchez, L., & Thomson, E. (1997). Becoming mothers and fathers: Par- enthood, gender, and the division of labor. Gender and Society, 11, 747–772.

Sayer, L. (2005). Gender, time and inequality: Trends in women’s and men’s paid work, unpaid work and free time. Social Forces, 84, 285– 304.

Shaw, L., & Hill, C. (2002). Gender gap in pension coverage: What does the future hold? Institute for Women’s Policy Research. Washington, DC. Retrieved April 25, 2010, from http://www.iwpr.org/pdf/d447.pdf

Shirey, L., & Summer, L. (2000). Caregiving: Helping the elderly with activity limitations. Challenges for the 21st century: Chronic and dis- abling conditions, No. 7. Washington, DC: National Academy on an Aging Society.

Social Security Administration. (2009). Fast facts & figures about social secu- rity, 2009. Retrieved January 7, 2010, from http://www.socialsecurity .gov/policy

Social Security Administration. (2010a). Annual statistical supplement to the social security bulletin, 2009. Retrieved July 21, 2010, from http:// www.socialsecurity.gov/policy/docs/statcomps/supplement/2009 /supplement09.pdf

Social Security Administration. (2010b). Income of the aged chartbook: 2008. Retrieved April 19, 2010, from http://www.socialsecurity.gov /policy

Solimeo, S. (2008a). Osteoporosis in older men: Feelings of masculinity and a “women’s disease.”. Generations, 32, 73–77.

Solimeo, S. (2008b). Sex and gender in older adults’ experience of Parkinson’s disease. Journal of Gerontology: Social Sciences, 63B, S42–S48.

Thompson, E. H., Jr. (2000). Gendered caregiving of husbands and sons. In E. W. Markson, & L. A. Hollis-Sawyer (Eds.), Intersections of aging: Readings in social gerontology (pp. 333–344). Los Angeles, CA: Roxbury Publishing Co.

Thompson, E. H., Jr. (2008). Gender matters: Aging men’s health. Gen- erations, 32, 5–8.

West, C., & Fenstermaker, S. (1995). Doing difference. Gender & Society, 9, 8–37.

Williams, C. L. (1995). Still a man’s world. Berkeley: University of California Press.

Wu, B., Goins, R. T., Laditka, J. N., Ignatenko, V., & Goedreis, E. (2009). Gender differences in views about cognitive health and healthy lifestyle behaviors among rural older adults. The Gerontologist, 49, S72–S78. by guest on January 22, 2015 http://gerontologist.oxfordjournals.org/

D ow

nloaded from