SOCIAL SECURITY ACT (1935)

338

Families in Society: The Journal of Contemporary Social Services | www.familiesinsociety.org | Copyright 2005 Alliance for Children and Families

The passage of the Social Security Act has been heralded as one of the United State’s greatest policy successes. However, the current focus is less on preserving Social

Security’s social welfare benefits for all workers and more on its long-term financing and maximizing individual ben- efits, which reflects a difference in the fundamental values underlying Social Security. This shift is most vividly illus- trated by President Bush’s decision to make the privatiza- tion of Social Security for younger workers a central focus of his second-term domestic agenda. Yet Social Security is indeed the safety net for many older adults, especially women and people of color. It is credited with removing more individuals from poverty than all other governmental programs combined. The more than threefold decline in

poverty among the older population over a 40-year period is due largely to Social Security. In 1959, the poverty rate among adults age 65 and older was 35.9% compared with 10.2% today (U.S. Census Bureau, 2001, 2004a ). Yet even with this policy success, approximately 3.6 million older Americans still fell below the official federal poverty line in 2003, and an additional 2.2 million older adults were classi- fied as near poor (income between the poverty level and 125% of this level; U.S. Census Bureau, 2004a). A dispro- portionate number of these older Americans living in poverty are women. Indeed, the stark reality is that almost 70% of poor older adults are women, in particular women of color and those older than 85 years and living alone (Fitzpatrick & Entmacher, 2000).

Reducing Poverty Among Older Women: Social Security Reform and Gender Equity Judith G. Gonyea & Nancy R. Hooyman

ABSTRACT

The authors document the higher poverty rate of older women, especially women of color, com-

pared with older men—a pattern created and maintained by the intersection of the structural

factors of age, race, and marital status. They then review how the U.S. Social Security program

generally benefits older women and reduces their late-life economic vulnerability. A persistent

gender inequity, however, is that women are more likely to disrupt their paid employment to

meet family care responsibilities, which may increase the number of zero-earnings years and

reduce the amount paid into Social Security. Current proposals to privatize the Social Security

system are critiqued in terms of their gender inequities. Three relatively revenue-neutral propos-

als that could increase Social Security’s protection against poverty and differentially affect low-

income women are briefly discussed.

DIVERSITY

This article is part of “The Future of Social Work With Older Adults,” a special issue of Families in Society with guest editor Carol Austin. www.familiesinsociety.org

Gonyea & Hooyman | Reducing Poverty Among Older Women: Social Security Reform and Gender Equity

339

As the nation’s older population has continued to grow, along with a dramatically increasing federal deficit, conflict- ing viewpoints have emerged regarding public or collective responsibility for income security in old age. Moreover, the level of this debate has intensified with the growing realiza- tion that members of the baby boomer generation will soon begin to enter the ranks of the 65-plus population. The U.S. Census Bureau projects that by 2030, the time at which the youngest members of the baby boomer generation will turn 65, the percentage of the older population will reach 20% (i.e., 1 of every 5 Americans; U.S. Census Bureau, 2001). Much of the current debate about privatizing Social Security has, therefore, been framed around the program’s financial solvency and returns on individual investment. Less atten- tion has been focused on the inadequacies of Social Security to protect low-income individuals from moving into poverty in old age. However, both issues—the long-term fis- cal balance and the antipoverty effectiveness of Social Security—are critical to promoting older women’s eco- nomic security.

We begin with an examination of the experience of poverty among older Americans. We not only highlight gender differences in the risk of late-life poverty but also examine how the poverty risk varies within the female pop- ulation based on structural characteristics such as age, socioeconomic class, race, and marital status. The ways in which women’s domestic and labor force roles contribute to their late-life economic vulnerability and the salience of Social Security to their lives are explored. Current propos- als to privatize Social Security are critiqued in terms of their gender inequities. In contrast, we argue for both the pro- tection of Social Security’s core principles and expansion of its antipoverty protection.

The Differential Risk of Poverty in Old Age

Lack of attention to the plight of older adults who are liv- ing in poverty may reflect a growing societal view that older adults are faring better financially relative to other age groups in the United States, particularly children. As noted, the poverty rate for persons 65 and older was 10.2% in 2003, which is lower than the 10.8% rate for working age adults and the 17.6% rate for children (U.S. Census Bureau, 2004a). It is increasingly recognized, however, that the annual cross-sectional poverty statistics produced by the Current Population Survey (CPS) do not present a com- plete picture of the economic status of older Americans (Wu, 2003). Using longitudinal data from the national Panel Study of Income Dynamics (PSID), for example, Wu (2003) tracked the poverty status of the same individuals from 1981 to 1992.

To explore the phenomenon of persistent poverty, Wu asked two important questions: (a) Do older persons face a greater risk than younger persons in falling into poverty for a long period, and (b) Do older adults compared with

young adults experience more difficulty escaping from poverty after they enter it? His findings reveal that the poverty experience does, in fact, vary by life stage. During a 5-year period (1988–1992), 24.3% of the 65-plus popula- tion experienced poverty for at least 1 year and 5.6% were poor for all 5 years; in contrast, 20.1% of the under-65 pop- ulation were poor for at least 1 year and only 3.6% lived in poverty throughout the 5 years. During the 12-year period (1981–1992), only 35.2% of the 65-plus population who spent 1 year in poverty escaped from economic hardship compared with 40.3% of the under-65 population. Wu concluded that “the majority of older adults who spent more than four consecutive years in poverty will stay in poverty for a long time, and some of them will remain poor until death” (2003).

The PSID longitudinal data reinforce a finding consis- tently documented in the CPS cross-sectional data: Women are at a much greater risk of falling into poverty in later life than men. From 1998 to 1992, 27.8% of older women expe- rienced at least 1 year of poverty compared with 17.6% of men (Wu, 2003). In fact, analysis of the CPS data on the percentage of older Americans living in poverty by age, gender, race, and Hispanic origin dramatically underscores the greater vulnerability of women of color.

As reflected in Table 1, approximately 25% of older African American or Hispanic women now live below the federal poverty level. Marriage often protects women against experiencing poverty in old age. Whereas less than 5% of older married women face poverty, as shown in Table 2, 17% of unmarried older women are poor (Federal Interagency Forum on Aging-Related Statistics, 2002; Older Women’s League, 2003).

Women’s Economic Vulnerability Across the Life Course

The difference in women’s greater economically vulnerabil- ity in old age compared with men’s is largely a consequence

TABLE 1. Percentage of Older U.S. Adults Living in Poverty, by Age, Gender, Race and Hispanic Origin in 2001

TOTAL WHITE BLACK HISPANIC ORIGIN a

Sex and Age Both Sexes

65 to 74 years 9.2 7.8 20.2 21.8

75 and older 11.2 10.2 24.2 22.0

Males

65 to 74 years 6.8 5.7 14.3 17.5

75 and older 7.3 6.4 18.1 19.4

Females

65 to 74 years 11.2 9.6 24.5 25.0

75 and older 13.6 12.5 28.3 23.7 aHispanic can be of any race. Source: U.S. Bureau of Labor Statistics and Bureau of the Census. Current Population Series, Annual Demographic Survey March 2001 Supplement. Table 1. http://ferret.bls.census.gov/macro/032002/pov/new01_001.htm

FAMILIES IN SOCIETY | Volume 86, No. 3

340

of the domestic division of labor and women’s position in the labor market. Exploring the linkages among the phases of the life course, rather than their distinctiveness, reveals how feminization of poverty occurs in old age. Although more women have entered the labor force in the last several decades, their wages, even for workers with the greatest employment effort, continue to lag behind those of men who are working full time and year-round. One important reason for this earnings gap is gender segregation in the labor market (i.e., the division into traditional women’s jobs and men’s jobs). In 2003, a gender-based comparison of fully engaged workers (e.g., continuous, full-time employ- ment) revealed that women earn, on average 75.5 cents for every dollar earned by men (U.S. Census Bureau, 2004b).

The measurement of an annual male–female wage gap (similar to the annual poverty statistic) does not, however, offer a complete picture of the gender-based wage differen- tial. Using the PSID to track the same men and women across a 15-year period, Rose and Hartman (2004) found that prime-age (26–59 years) employed women earned only 38% of that of prime-age men. The long-term effects of this earning differential are large and can be devastating:

Across the fifteen years of the study, the prime age

working woman earned only $273,592 while the aver-

age working man earned $722,693 (in 1999 dollars).

This gap of 62% is more than twice as large as the

23% gap commonly reported. (Rose & Hartman

2004, p. iii)

Moreover, the gender-based division of domestic responsibilities results in women, more often than men, reducing their time in the paid labor force in order to take on child and elder care and household management. Across the 15-year time span, Rose and Hartman found that slightly more than half (52%) of women had at least 1 calendar year without any earnings compared with just 16% of men. Similarly, women are more than twice as likely as men to work part time (i.e., fewer than 25 hr per week). In 2002, approximately 25% of employed women worked part-time compared with 11% of employed men (U.S. Bureau of Labor Statistics, 2003). The long-term earnings data underscore that women’s time spent performing fam- ily care often profoundly limits their economic resources in later life. In fact, motherhood has been identified as the sin- gle greatest risk factor for poverty in old age (Rappaport,

2004). Finally, women’s longer average life expectancy compared with men’s means that they may be required to stretch more limited financial resources over a greater number of years. This pattern is also true of married women who outlive their spouses. More than half (59%) of women enter their later years of life not married, even if they once were; as indicated in Table 2, these women face a fourfold greater chance of being poor (Administration on Aging, 2002).

The Importance of Social Security to Older Women’s Lives

Social Security is a near-universal old age social insurance program; 9 of every 10 older citizens are beneficiaries. In fact, it is, in most respects, a highly successful program (American Association of Retired Persons [AARP], 2005). Because of women’s longer life expectancies, they comprise 58% of all Social Security beneficiaries age 62 and older and approximately 71% of beneficiaries aged 85 and older. Without Social Security, it is estimated that more than 50% of our nation’s current population of older women would be poor (Moody, 2002;Older Women’s League, 2003; Weir, Willis, & Sevak, 2002).

From its beginnings, Social Security was never intended to be the only source of retirement income; rather, it was viewed as providing a foundation, along with incomes from pensions, savings, and investments, in the creation of an eco- nomically secure old age for U.S. citizenry. Yet for too many older Americans who lack private pensions or extensive

TABLE 2. Percentage of Older American Women Living in Poverty, by Marital Status in 2001

MARITAL STATUS POVERTY RATE (%)

Married 4.3

Widowed 15.9

Divorced 20.4

Never married 18.9

Source: U.S. Bureau of Census, Current Population Survey, March 2002.

17.0%

27.2%

0%

40.9%

25.2%

Less than 50% of Income

50% to 89% of Income

90% to 99% of Income

100% of Income

10%

20%

30%

40%

50%

60%

70%

80%

90%

100% Men Women

30.5%

16.5% 11.7%

31.2%

FIGURE 1. Gender comparison of Social Security as a percentage of income for older Americans, 2001.

U.S. Census Bureau, 2002, Current Population Survey, March Supplement as prepared by AARP Public Policy Institute, 2003.

Gonyea & Hooyman | Reducing Poverty Among Older Women: Social Security Reform and Gender Equity

341

assets, Social Security is their sole income source. As Figure 1 reveals, Social Security represents 90% to 100% of retirement income for almost 59% of women and 29% of men. Analysis by race further reveals that women of color currently rely even more heavily on Social Security for their income in old age than do White women. Social Security provides more than half the retirement income for over 80% of nonmarried older African American and Hispanic women compared with 73% of older White women. Similarly, for more than 50% of African American and Hispanic women, Social Security represents 90% or more of their retirement income compared with 40% of White women (National Women’s Law Center, 2003).

The heavy reliance on Social Security by America’s poor- est elders is again underscored, as shown in Table 3, through a comparison of the income sources for the lowest income quintile of the 65-plus population, which is disproportion- ately occupied by women, with that of the highest income quintile for this age group. Greater workforce attachment (e.g., full-time employment), job stability (e.g., longer job tenures), and higher incomes are associated with a greater accumulation of retirement resources. Women, as a result of their different employment histories, are thus much less likely than men to receive a private pension in old age. Yet even when an older woman does have pension income, it is typically much smaller than that of an older man’s. Using CPS data from 1999 to 2001, Lee and Shaw (2003) found that only about 30% of older women received pension income compared with almost 47% of men. Women’s median annual pension income was about half that of men: $5,600 versus $10,340 (in 2000 constant dollars).

For both genders, the most common reason for not par- ticipating in a pension plan is that the employer simply does not offer one. Almost equal percentages of men and women—39% and 35%, respectively—reported the lack of an employer-sponsored plan as the primary reason for nonparticipation. However, significant gender differences in pension participation exist among employees working for companies that offer pension plans: Female (24%) employees were almost twice as likely as male employees (13%) to report that their nonparticipation was due to not working a sufficient number of hours to qualify for enroll- ment (Shaw & Hill, 2001). This finding is of particular

concern given the dramatic expansion of part-time and temporary employment in the United States during the past few decades. Nearly 25% of the U.S. workforce—more than 30 million Americans—is now engaged in part-time employment with few employment-based benefits; women of all races and minority men disproportionately fill these positions (Hudson, 2000).

Three features of the Social Security program are partic- ularly salient to women’s economic status. First, the Social Security benefit formula is progressive. Benefits are deter- mined based on workers’ earnings; thus, workers with higher earnings pay more taxes and receive higher benefits than those with lower earnings. However, the progressive benefit formula means that Social Security replaces a greater proportion of lower earners’ past income than of higher earners’ past income (although higher income ben- eficiaries will receive a larger benefit in dollars because they have paid more into the system). As noted in our previous discussion, because women typically earn less than men, the progressive formula replaces a greater proportion of their lifetime earnings. The progressive benefit formula is of particular importance to women of color, who tend to be heavily concentrated in low-paying occupations. African American and Hispanic women who are full-time workers earn, on average, only 65% and 56%, respectively, of the earnings of White men and 74% of the earnings of White women(Older Women’s League, 2003). For workers who retired at age 65 in 2000, the replacement rate for what they had paid into Social Security was 53% of preretirement income for lower earners, 40% for average earners, 32% for higher earners, and 24% for those with the maximum tax- able earnings (Anzick & Weaver, 2001).

A second feature of the Social Security system that is salient for women is that workers’ dependents have access to benefits. Under Social Security law, a married woman or qualified divorced woman (after a marriage of at least 10 years) is entitled to the higher of two benefits: a benefit cal- culated based on her own employment history or a benefit that is 50% of her husband’s (or former husband’s) benefit. A widow or divorced widow is also entitled to the higher of her own worker benefit or her husband’s (or ex-husband’s) full benefit as long as she meets requirements related to length of marriage and if her divorced husband has lived long enough to collect benefits. The current reality is that women are more likely to receive Social Security benefits as a dependent—a spouse or widow—than men because their lower lifetime earnings mean that their benefits are typi- cally higher as a spouse or a widow versus as an employee. Although almost all (95%) men receive a benefit based fully on their own employment histories, only 37% of women garnered worker benefits in 1997 (Anzick & Weaver, 2001).

Third, Social Security is more than a worker retirement program; it offers both life insurance and disability insur- ance for workers and their families. Moreover, these aspects of the Social Security program are particularly crucial for

TABLE 3. Source of Income Among Persons Age 65 and Older in the Lowest and Highest Income Quintiles, 2001

INCOME SOURCE (%) LOWEST QUINTILE HIGHEST QUINTILE

Social Security 83.0 20.0

Asset Income 2.0 19.0

Pensions 4.0 21.0

Earnings 1.0 38.0

Public Assistance 9.0 0.0

Other 1.0 2.0

Total 100.0 100.0

Source: Social Security Administration Income of Aged Chartbook, 2002.

FAMILIES IN SOCIETY | Volume 86, No. 3

342

women. Despite increasing life expectancies, 1 of every 7 Americans (disproportionately men) will still die before reaching age 67. Many of these individuals, who are the family primary wage earners, lack life insurance policies. Through Social Security, family members are entitled to survivor’s benefits; currently, of the approximately 47 mil- lion Americans who are Social Security recipients, 7 million are the spouses and children of deceased workers. In fact, for the average wage earner with a family, the Social Security insurance benefit is estimated to be equivalent to a $322,000 life insurance policy (National Committee to Preserve Social Security and Medicare, 2004). Long-term disability may also jeopardize an individual’s ability to be employed. Although the vast majority of workers lack long-term dis- ability insurance, about 3 of every 10 young adult workers will become disabled before reaching age 67. Fortunately, Social Security offers protection to families and workers with major disabilities that prevent them from being able to work. For the average wage earner with a family, the Social Security insurance benefit is calculated to be equivalent to a $233,000 disability insurance policy (National Committee to Preserve Social Security and Medicare, 2004).

These nonretirement features of the program are criti- cally important to women across the life span because only slightly more than 33% of female recipients of Social Security receive benefits solely as retired workers compared with more than 80% of male recipients (National Women’s Law Center [NWLC], 2002). Analyses by race underscore that the Social Security’s disability and survivor benefits are critical particularly to the economic status of women of color and their families. Approximately 20% of African American and Hispanic beneficiaries are younger than 55 years compared with 10% of White beneficiaries. On the basis of calculations with Social Security Administration data, the NWLC (2003) found that African American women rely disproportionately on these nonretirement aspects of the Social Security program, given their higher rates of disability and their likelihood of surviving their husbands. The NWLC reports that

While African Americans make up 9% of all female

beneficiaries, African American women constitute

18% of female disabled worker beneficiaries.…

Whereas 7% of all Social Security beneficiaries are

children, 15% of African American beneficiaries are

children. In fact, African American children are almost

four times more likely to be lifted out of poverty by

Social Security than White children. (2003, p. 2)

Finally, the annual cost-of-living adjustment (COLA) fea- ture of Social Security helps all beneficiaries cope with rising costs such as utilities and prescription drugs. This third fea- ture is especially valuable for women given their longer life expectancies. Without this inflation protection feature, Social Security benefits would buy considerably less over

time. For example, with a 3% annual inflation rate and with- out the COLAs, it is estimated that benefits would buy 25% less after 10 years (National Women’s Law Center, 2002).

Gender Inequities Inherent in Proposals to Privatize Social Security

Social Security is a successful program precisely because it remains the most important source of retirement income for older Americans, especially low-income older women. Although concerns have been raised about Social Security’s longer term financial solvency program, efforts to reform it should not undermine the protections it currently offers our nation’s oldest citizens. Immediate fiscal reforms to address a crisis appear to be unnecessary because the Social Securities actuaries conservatively project that the trust fund balance will not be depleted until 2042. Yet, even after this date, Social Security would not be bankrupt; instead, annual collections from payroll taxes would be sufficient to meet more than 70% of benefits (Board of Trustees, Old Age and Survivors Insurance and Disability Insurance, 2004). The Congressional Budget Office’s model estimates the possible trust fund depletion date as 2052 and only a 1% payroll gap between income and benefits over the next 75 years, assuming no changes in the Social Security pro- gram (Congressional Budget Office, 2004).

Although prior presidents and members of Congress have discussed the private mechanisms of incentives to save, the likelihood of privatization has increased dramati- cally with the 2004 reelection of President Bush. Such a shift is congruent with the beginning privatization of Medicare through the 2003 Medicare Prescription Reform in this current era of market and private or individual responsibility (Binstock, 2002). Privatization would divert payroll taxes (or general revenue income tax credits) to new personal investment accounts among workers younger than 55 years. This model assumes a strong economy and stock market, discretionary resources to invest, and indi- vidual knowledge and skills to make informed investment decisions. The volatility of these assumptions, especially related to the stock market, will result in both increased individual risk and greater federal budget deficits. Regardless of any particular model, privatization reflects the following values: Free markets, not social insurance, are the most efficient and fair way to distribute resources; employment success is rewarded; and individual responsi- bility and freedom of choice (or risks) are paramount. These values contrast dramatically with the values of uni- versalism, mutual responsibility, cross-generational bene- fits, and earned right underlying the origin of Social Security (Smallhout, 2002).

Privatization of Social Security would negatively affect women, especially those of color, more than men (Older Women’s League, 2002). Although the specifics of President Bush’s current privatization model are not yet

Gonyea & Hooyman | Reducing Poverty Among Older Women: Social Security Reform and Gender Equity

343

fully known, the costs of moving to a privatization scheme have been acknowledged in prior analyses by the Social Security Administration and even by the Moynihan Commission’s Report on Social Security. Under privatization, the progressive benefit formula of Social Security, which replaces a higher percentage of earnings for lower income workers than higher income workers, would be lost. Moreover, as low-income and part-time employees, many women would have smaller private accounts to invest. With more limited financial resources, women typically avoid higher risk investments; therefore, it is anticipated that the yield of their accounts would generally be below average. In fact, women might find a larger share of their private accounts going to administrative costs. Many policy experts now estimate that the administrative costs—perhaps as high as $1 tril- lion—would be dramatically higher in individual account systems (AARP, 2005). For example, if transac- tion fees involve a flat per-account charge, administrative costs would consume a larger portion of the accounts of low- and moderate-income older adults (Diamond, 1998; Munnell, 1999). Ultimately, the burden for the manage- ment of the investment portfolio would fall squarely on the individual’s shoulders. In addition, older women who have historically received little training in financial man- agement may be at greater risk for faulty or poor invest- ment decisions.

Because of generally limited private investments, women are less likely than men to have sufficient income to last until death. Privatization means that there would no longer be a lifetime guarantee of a benefit; instead, when funds in the account are exhausted, the account ceases to exist. Given women’s longer life expectancy compared with men, coupled with their smaller accounts, women would face a greater prospect of outliving all of their savings and assets. Although women can purchase lifetime private annuities, such annuities, unlike Social Security, are monthly pay- ments based on gender-based life expectancies, resulting in women receiving a lower lifetime benefit even when their investments are equal to those of men.

Concerns are also raised that privatization would likely eliminate death and disability protection and the cost-of- living increase available through Social Security, all changes that would disadvantage women’s benefits. Women, com- pared with men, are much more likely to be responsible for children and themselves after a spouse’s disablement or death. How privatization would impact divorced women is unclear. Under Social Security, divorced spouses and divorced widows, after a marriage of 10 years, automatically receive the same benefits that married spouses and widows receive without any corresponding reduction in benefits to the worker or subsequent spouses. In a privatized system, however, the core benefit might be reduced, and division of the private account between husband and wife would fall under the jurisdiction of a divorce court.

The primary beneficiaries of privatization will be higher income unmarried workers who will not be born until 2025 and who will be largely Caucasian males. In the short term, women will bear the burden of transition and administrative costs, including the need to cut current Social Security bene- fits when funds are diverted into individual accounts and being taxed twice (e.g., paying for their own retirement through private accounts while continuing to pay for current beneficiaries; Cavanaugh, 2002; Favreault & Sammartino, 2002; National Committee to Preserve Social Security and Medicare, 2005; Williamson, 2002). Debates about privatiza- tion also need to consider other ways to prevent a shortfall in 2040, such as expanding the number of workers participat- ing in Social Security by requiring state and local govern- ment workers to participate, raising the cap on taxable income, reducing slightly future benefits or COLA increases, or allowing the government to invest the funds in equity markets. Several analysts, for example, have suggested that a 1.1% rise in the FICA tax would be sufficient to finance the Social Security system throughout the baby boomer genera- tion’s retirement years (Diamond & Orszag, 2003; National Committee to Preserve Social Security and Medicare, 2005; Quadagno, 1999). Although attention does need to be given to Social Security’s future, its solvency can be achieved through incremental changes and does not require privatiza- tion, which undermines its basic principles and reduces retirement income for women and persons of color.

Increasing Social Security’s Antipoverty Protection

The current debate on privatization has overshadowed dis- cussions of the plight of elders who continue to live in poverty. Despite the enormous success of Social Security in lifting generations of older Americans out of poverty, it remains a flawed antipoverty program (Callahan, 1999). Several proposals for programmatic reforms to Social Security have been advanced to reduce older women’s financial vulnerability, including raising the minimum Social Security benefit; increasing the survivors benefit for widows; and providing dependent care credits. A discussion of each of these options follows.

A Higher Minimum Social Security Benefit As we have shown, even a lifetime of employment does not guarantee a financially secure retirement, especially for the working poor. Raising the minimum Social Security bene- fit would be particularly valuable to women and persons of color, given their overrepresentation in the secondary labor market, a sector that is characterized by low-paying jobs with few benefits. Moreover, many women and persons of color are employed in physically demanding or taxing jobs (e.g., domestic, industrial, and farm labor) that lead to an earlier departure from the paid labor force. Lower income is also associated with a greater risk for earlier onset of a

FAMILIES IN SOCIETY | Volume 86, No. 3

344

number of chronic and disabling health conditions that might force earlier retirement decisions (Kijakazi, 2003). Using the data from the National Health and Retirement Study, Flippen and Tienda (2000) found that African Americans, Hispanics, and women experience more invol- untary job separation in the years immediately before retirement and that these periods of joblessness often result in permanent labor force withdrawal.

In fact, a special minimum Social Security benefit cur- rently exists for low-wage workers with a history of steady employment that provides these retirees with a higher monthly benefit than they would receive under the regular benefit formula. Few individuals, however, are currently eligible for the special minimum benefit because of its restrictive eligibility requirements. In 2000, only about 144,000 individuals, or 0.33% of Social Security beneficia- ries, received the special minimum benefit. Moreover, the maximum benefit amount remains at only 85% of the fed- eral poverty threshold for an adult aged 65 and older (Anazick & Weaver, 2001). A number of policy analysts, such as Wendall Primus of the Center on Budget and Policy Priorities, have offered proposals for a revised benefit for- mula and eligibility standards in order to both raise the minimum benefit amount and more effectively target these funds to the working poor (See Kijakazi, 2003, and Anazick & Weaver, 2001, for more detailed discussions of this pro- posed reform.) Finally, increasing the minimum Social Security benefit would particularly benefit poor employed women who either never married or were married fewer than 10 years and thus receive a benefit based solely on their own employment histories.

Increase the Survivor Benefit for Widows Women often experience a significant decline in their income with the death of their husband. Under the current Social Security system, a married couple is allowed to receive 100% of the higher earner’s income as well as a spousal benefit equal to 50% of the higher earner’s income (or her own earnings history if that would result in benefits higher than the spouse’s benefit). On her husband’s death, a woman receives 100% of her own benefit or 100% of the deceased spouse’s benefit. For most widows, the decline in Social Security income greatly exceeds the decline in their living expenses. The federal poverty threshold for a 1-per- son older household equals 79% of the federal poverty level for a 2-person older household. Thus, policy experts often suggest that the survivor’s benefit should be increased to 75% of the couple’s benefit; in other words, the surviving spouse’s benefit should not be reduced by more than 25% of the couple’s combined benefit (Anzick & Weaver, 2001; Burkhauser & Smeeding, 1994).

Raising the survivor benefit would provide gains for older widowed women, but the largest increases would be to wid- ows from families with higher lifetime earnings (Favreault, Sammartino, 2002). As Harrington Meyer (1996) notes,

middle- and upper-class White women are more like to receive noncontributory Social Security benefits. Because more women have entered the paid labor force and fewer women are married for the qualifying 10-year marriage, what was originally an important safety net for lower income retirees has greatest value for traditional families in higher income brackets (Harrington Meyer, 1996).

Offering Dependent Care Credits Care credits are often debated as a way to reward and rec- ognize women’s disproportionate responsibilities for rais- ing children. Rather than marital status as an eligibility criterion, women would receive benefits based on their contribution to the economy through both their labor force participation and their unpaid work of child care. Yet care credit reforms need to take account of race and socioe- conomic class differences within the female population. The most commonly debated type of care credit proposal is to remove zero-earnings years—when women have been out of the paid work force because of child care responsi- bilities—from women’s benefit calculation. This approach may further class and racial inequities, however, because upper income White married women, who can afford not to work for pay, are more likely to benefit than low-income married women of color who are employed out of eco- nomic necessity. Because most low-income women have to work, they are unlikely to have zero-earnings years in their benefit calculation. A second care credit model would drop additional low-earnings years (9 years) from the benefit calculations. Currently, workers can drop 5 low-earnings years between the ages of 22 and 62, which leaves them with 35 earnings years. Because the rewards for caregiving are directly tied to women’s earnings histories, women with high earnings would again fare better than women with low earnings. Placing a value on care is a third way to struc- ture care credits; such credits would be a set amount of earnings, which would substitute for a certain number of years of earnings that are below this level. To illustrate, if the care credit was $15,000 and a woman within her high- est 35 years of earnings had 2 years in which she earned only $8,000, she would be credited with an additional $7,000 for those years. This approach would benefit lower income women more than those with higher earnings (Herd, 2002). Care credits would be a more progressive way to distribute benefits than spousal benefits, because women would move onto the worker benefit and their lower incomes would be buffered by economic value being assigned to their unpaid care work. Generally, low-income women would be hurt most by a system that dropped more zero- or low-earnings years and would benefit most if half of their median wage were substituted into low-earnings years. Because care credits would eliminate spousal bene- fits, this approach would probably be revenue neutral for the government, unlike privatization proposals that carry heavy transition and administrative costs.

Gonyea & Hooyman | Reducing Poverty Among Older Women: Social Security Reform and Gender Equity

345

All three of the identified programmatic reforms would redistribute women’s benefits from earlier to later in life. However, it is evident that each of these programmatic changes would differentially impact various subpopula- tions of older women, given the complexity and diversity of their life experiences. In one of the most significant studies to date, Favreault and Sammartino (2002) use a dynamic microsimulation model on the 1990–1993 Survey of Income and Program Participation data to explore the impact of expanding the minimum benefit, increasing the survivor benefit, and offering dependent care credits. Their analysis underscores that programmatic reforms to improve Social Security’s adequacy and equity for current and future generations of women can be designed to be low-cost or revenue neutral. They conclude that

Policymakers should be careful not to rely on intu-

ition when designing reforms to shore up women’s

Social Security benefits, but rather to rely on rigorous

analyses … we have demonstrated that policymakers

can change the parameters in the existing system to

target the highest-risk low-income and older women.

Our analyses show how legislators can combine a

series of changes into packages that meet multiple

needs. (2002, p. ix)

Since its origin, the Social Security program has been amended a number of times to increase its antipoverty effectiveness. Amendments have included, for example, raises in benefit levels, the indexing of levels to inflation (COLAs), and shortening of the marriage duration require- ment from 20 to 10 years. Each of the identified reforms— raising the minimum benefit level, increasing the survivor benefit level, and offering dependent care credits—repre- sent critical programmatic reforms that would further strengthen Social Security’s antipoverty protection.

Unfortunately, the current focus on Social Security’s long-term solvency overlooks its centrality to older women’s relative economic security and protection from poverty, especially among low-income women of color. Even though a significant percentage of older women who receive Social Security benefits still remain poor or near poor, their economic status is likely to be at even greater risk under a system of privatization that assumes individu- als have adequate resources and investment capability. Given the relative invisibility of older women in our public policy-making process, their needs for protection from poverty under revenue-neutral proposals are unlikely to be heard compared with the financial gains from privatization for investment companies. Although the Older Women’s League has been a strong advocate for changes in Social Security to benefit older women, their voices are likely to be silenced by the powerful interest groups that characterize policymaking in this current era of free market and indi- vidual responsibility (Binstock, 2002).

The characteristics of future cohorts of women Social Security beneficiaries will differ markedly from current women beneficiaries. Changing marital, family, and labor force patterns suggest that a smaller proportion of women will be entitled to benefits solely as spouses or survivors and a growing proportion will receive worker-only benefits, dually entitled spouse benefits, and dually entitled survivor benefits. It is critical to recognize, however, that these trends will not eliminate concerns about the adequacy and equity of Social Security benefits. As the debate regarding how to “save” Social Security intensifies, progressives must effec- tively make the case that privatizing Social Security would mean less retirement income for the majority of Americans and would be particularly harmful to women. The Social Security program can be protected for future generations of retirees without introducing the risk of and high cost of individual private accounts. Further, progressives must advocate strongly that this current period of reform offers an opportunity not only to protect but also to raise the safety net of Social Security for older women who are at high risk for poverty.

The profession of social work has a long history of grass- roots advocacy and speaking out about the role of govern- ment in protecting our nation’s most vulnerable citizens. Social workers, individually and collectively through the National Association of Social Workers, can play a significant role in communicating concerns about the president’s pri- vatization plans, particularly for women. As a historically female profession, social work represents an important voice in working to preserve and strengthen Social Security not only for the current cohort of older women but also for our daughters, granddaughters, and great-granddaughters.

References Administration on Aging. (2002). A profile of older Americans: 2002.

Retrieved May 20, 2005, from http://www.aoa.gov/prof/Statistics/profile/2002profile.pdf

American Association of Retired Persons. (2005, February 12). AARP and Social Security: A background briefing. Retrieved May 20, 2005, www.aarp.org/socialsecurity

Anzick, M. A., & Weaver, D. A. (2001). Reducing poverty among elderly women. ORES Working Paper Series Number 87. Washington DC: Social Security Administration, Office of Research, Evaluation and Statistics.

Binstock, R. (2002). The politics of enacting reform. In S. H. Altman & D. I. Schatman (Eds.), Policies for an aging society (pp. 346–377). Baltimore, MD: John Hopkins University Press.

Board of Trustees, Old Age and Survivors Insurance and Disability Insurance. (2004). Annual report of the Board of Trustees of the federal Old Age and Survivors Insurance and Disability Insurance trust funds. Washington, DC: Author.

Burkhauser, R. V., & Smeeding, T. M. (1994). Social Security reform: A budget neutral approach to reducing older women’s disproportionate risk of poverty. Syracuse, NY: Maxwell School of Citizenship and Public Affairs/Center for Policy Research.

Callahan, D. (1999). Still with us: Elderly poverty in America. The American Prospect. 10(45), 74–77.

Cavanaugh, F. X. (2002). Feasibility of Social Security individual accounts. Washington, DC: Public Policy Institute, AARP.

Congressional Budget Office. (2004). The outlook for Social Security. Washington, DC: Author.

FAMILIES IN SOCIETY | Volume 86, No. 3

346

Diamond, P. A. (1998). The economics of Social Security reform. In R. D. Arnold, M. J. Graetz, & A. H. Munnell (Eds.), Framing the Social Security debate: Values, politics, and economics (pp. 38–64). Washington, DC: Brookings Institution Press.

Diamond, P. A., & Orszag, P. R. (2003). Understanding the legacy debt in Social Security. Washington, DC: The Brookings Institute.

Favreault, M. M., & Sammartino, F. J. (2002). Impact of Social Security reform on low-income and older women. Washington, DC: Public Policy Institute, AARP.

Federal Interagency Forum on Aging-Related Statistics. (2002). Older Americans 2000: Key indicators of well-being. Hyattsville, MD: Author.

FitzPatrick, C.S., & Entmacher, J. (2000). Increasing economic security for elderly women by improving Social Security survivor benefits. Retrieved October 1, 2005, http://www.nwlc.org/pdf/NASIwidows2.pdf

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

Harrington Meyer, M. (1996). Making claims as workers or wives: The distribution of Social Security benefits. American Sociological Review, 61, 449–465.

Herd, P. (2002). Care credits: Race, gender, class and Social Security reform. Public Policy and Aging Report, 12, 13–18.

Hudson, K. (2000). No shortage of “nonstandard” jobs. Washington, DC: Economic Policy Institute.

Kijakazi, K. (2003, June). Low-wage earners: Options for improving their retirement income. Retrieved October 1, 2005, http://www.centeronbudget.org/6-13-03socsec.pdf.

Lee, S., & Shaw, L. (2003). Gender and economic security in retirement. Washington, DC: Institute for Women’s Policy Research.

Moody, H. R. (2002). What is the future of Social Security? In H. R. Moody (Ed.), Aging: Concepts and controversies (4th ed., pp. 215–250). Thousand Oaks, CA: Sage.

Munnell, A. (1999). Reforming Social Security: The case against individual accounts. Boston: Center for Retirement Research at Boston College.

National Committee to Preserve Social Security and Medicare. (2004, October). Disability insurance & survivors’ benefits. Retrieved May 20, 2005, http://www.ncpssm.org/news/archive/vp_surviorsbene

National Committee to Preserve Social Security and Medicare. (2005, January). Social Security’s real ‘crisis’ is one of morals. Retrieved February 01, 2005, http://www.ncpssm.org/news/archive/nathansonoped.

National Women’s Law Center. (2002, May 10). Women and Social Security reform: What’s at stake? Washington, DC: Author.

National Women’s Law Center. (2003, January). Women of color and Social Security. Washington, DC: Author.

Older Women’s League. (2002). Social Security privatization: A false promise for women. Washington, DC: Author.

Older Women’s League. (2003). Women and retirement income. Washington, DC: Author.

Quadagno, J. (1999). Creating a capital investment welfare state: The new American exceptionalism. American Sociological Review, 64, 1–11.

Rappaport, F. S. (2004, May 21). Women and Social Security: Important issues for financial security of older Americans. Report presented to Mercer Human Resource Consulting, New York.

Rose, S. J., & Hartmann, H. I. (2004). Still a man’s labor market: The long- term earnings gap. Washington, DC: Institute for Women’s Policy Research.

Shaw, L., & Hill, C. (2001). The gender gap in pension coverage: What does the future hold? Retrieved October 1, 2005, http//www.iwpr.org/pdf/e507.pdf

Smallhout, J. H. (2002). Benefit design choices for personal Social Security accounts. Benefits Quarterly, 18, 44–64.

U.S. Bureau of Labor Statistics. (2003, October). Women at work: A visual essay. Monthly Labor Review, 46–50. Retrieved October 1, 2005, http://www.bls.gov/opub/mlr/2003/10/ressum3.pdf

U.S. Census Bureau. (2001, October). The 65 years and over population: 2000. Census 2000 brief. Washington, DC: Author.

U.S. Census Bureau. (2004a). Income, poverty, and health insurance coverage in the United States: 2003. Retrieved October 10, 2005, http://www.census.gov/prod/2004pubs/p60-226.pdf

U.S. Census Bureau. (2004b, August). Median earning of workers 15 years old and over by work experience and sex. Retrieved October 10, 2005, http://www.census.gov/prod/2004pubs/p60-203.pdf

Weir, D. R., Willis, R. J., & Sevak, P. A. (2002). The economic consequences of widowhood. Ann Arbor: University of Michigan, Michigan Retirement Research Center.

Williamson, J. B. (2002). What’s next for Social Security: Partial privatization? Generations, 26, 34–39.

Wu, K. B. (2003). Poverty experience of older persons: A poverty study from a long-term perspective. Washington, DC: AARP.

Judith G. Gonyea, PhD, is associate professor and chair, Boston University School of Social Research, Boston, MA. Nancy R. Hooyman, PhD, is endowed professor of gerontology and dean emeritus, University of Washington School of Social Work, Seattle, WA. Correspondence regarding this article may be sent to the first author at [email protected] or Boston University, School of Social Work, 264 Bay State Road, Boston, MA 02215.

Manuscript received: November 15, 2004 Revised: March 24, 2005 Accepted: March 25, 2005