| ||||||||||||||
|
KEY POINTS:
 1850 M Street NW Suite 300 Washington, DC 20036 202-223-8196 | www.actuary.org Craig Hanna, Director of Public Policy © 2017 American Academy of Actuaries. All rights reserved. |
||||||||||||||
| Members of the Social Security Committee include: Timothy Leier, MAAA, FSA, EA—chairperson; Mark Shemtob, MAAA, ASA, EA—vice chairperson; Robert Alps, MAAA, ASA; Eric Atwater, MAAA, FSA, FCA, EA; Douglas Eckley, MAAA; Ronald Gebhardtsbauer, MAAA, EA, FSA; Jay Jaffe, MAAA, FSA; Leslie Lohmann, MAAA, FSA, FCIA, EA, ECA; Timothy Marnell, MAAA, ASA, EA; John Nylander, MAAA, FSA; Brendan O’Farrell, MAAA, EA, FSPA, FCA; Jeffery M. Rykhus, MAAA, FSA; Joan Weiss, MAAA, FSA, EA; Paul Withington, MAAA, EA, FSA; Ali Zaker-Shahrak, MAAA, FSA. |
Women and Social Security
Social Security provides beneï¬ts on a gender-neutral basis. Beneï¬ts are based on an individual’s earnings record, employment history, and family composition. However, gender-related differences in the American work culture mean that, in reality, Social Security provides different levels of retirement security for women and men. When the current beneï¬t structure was set up, the traditional roles of men in the family as primary wage earners and women as primary child care providers were predominant. The system provides higher beneï¬ts relative to contributions to married couples with a primary wage earner compared to unmarried individuals and couples where both spouses work and earn about the same total amount. As women have increasingly assumed roles as single heads of families or as co-equal wage earners in their families, situations frequently arise where Social Security provides lower beneï¬ts for the same contributions, or requires signiï¬cantly higher contributions with little, if any, increase in beneï¬ts compared to the “traditional” family.
Other gender-related factors that can cause differences in the benefits women and men receive from Social Security include:
Employment history and earnings. Women tend to have more frequent breaks in employment due to child-bearing, child care, or caring for elderly parents or relatives. When they are working, women on average earn less than their male counterparts. Thus, women generally receive smaller monthly Social Security benefits based on their own earnings histories, as compared to men.
Life expectancy and assets. Women, on average, live longer than men and are more likely to outlive accumulated personal assets and may therefore be more dependent on Social Security.
Life expectancy and lifetime Social Security benefits. Because of their longer lifetimes, women, on average, receive a greater amount in total lifetime Social Security benefits than men with the same work histories.
Marital status. Women, on average, are more likely than men to become widowed before or during retirement as a consequence of longer life expectancy, and the fact that women are often younger than their spouses. In addition, older women who lose a spouse are less likely than men to remarry.1 These factors mean women are likely to spend longer periods during retirement unmarried, and these periods are likely to fall at more advanced ages. Social Security survivor benefits are designed to provide retirement security to dependent children, widows and widowers, but provide no protection for the never-married or for divorcees if the marriage lasted less than 10 years.
Social Security recognizes differences in financial circumstances among the workers and dependents it covers. As such, the current Social Security program provides certain safety net (or social adequacy) features to assist the less well-off in having enough income for retirement security. Although these features are not targeted at women, women often benefit from them more than men because of the factors outlined above. Some Social Security reform options that have been or are being contemplated would change the social adequacy components of the program, with potentially disproportionate effects on women relative to men.
This issue brief discusses the relative differences in the factors affecting men and women, such as those noted above, as well as how these differences contribute to disparate benefits under Social Security. This includes the impact on an average woman of (a) the social adequacy provisions of the current Social Security program and (b) various proposals to reform Social Security.2
Background
In general, an individual’s Social Security benefit is based on the average indexed earnings of a worker’s 35 highest-paid years. Social Security’s rules are gender-neutral, so that a woman who retires with the same average lifetime earnings as a similarly situated man will receive the same monthly benefit. However, on average, some of the program’s features affect women differently because the average woman’s work history is not the same as that of the average man.
Women are more likely than men to be out of the work force, or to have breaks in employment. Even with the narrowing gender gap in the rates of labor force participation, women often leave temporarily or permanently for pregnancy, child care, and other family care responsibilities. As a result, women tend to have shorter work histories and thus smaller benefits than men.
In 2014, the participation rates in the labor force were as follows:3
| Percentage of U.S. Men and Women in the Workforce | ||
| Ages | Women | Men |
| 25 – 54 | 74% | 88% |
| 55 – 64 | 57% | 70% |
In 2013, about 65 percent of women with children under age six were in the labor force.4
Women on average earn less than men. Based on their own earnings records, women on average receive lower Social Security benefits than men with the same number of years of covered earnings due to differences in earnings between men and women. This particularly affects women who live alone or are the primary wage earners in their families. Married women, even in dual-earner families, may receive higher benefits based on their husbands’ work histories than on their own.
- From 1972 to 2014, the median annual women’s earnings for full-time year-round workers rose from 58 percent to 79 percent of the average for similarly situated men.5
- Despite gains in education, professional and managerial jobs, and business ownership, women have not achieved wage parity with men. In 2012, the median covered wage reported to the Social Security Administration for all workers was $31,205 for men and $21,914 for women, giving a ratio of women’s earnings as a percentage of men’s of 70 percent.6
Women live longer on average than men and will, therefore, need more assets in retirement. Because women generally have less income from other sources, and non-Social Security assets (lower on average than for men) must be spread out over a longer expected lifetime, Social Security benefits are a more significant component of a women’s retirement security.
- The average life expectancy at age 65 is 18.1 years for males and 20.6 years for females.7
- In addition, most married women have older spouses. As a result, current elderly women are much more likely to become widows than women a generation younger and not have a spouse able to assist them financially in their most elderly years.
- Since women live longer and are less likely to have a spouse to care for them, they are much more likely to need paid assistance with the tasks needed for daily living from family in-home care agencies and long-term care facilities. These services often require significant assets. About 23 percent of women age 62 and older (but only about 18 percent of similarly aged men) depend on Social Security for 90 percent or more of their family income.8
| Percentage of Income From Social Security | ||
| Males age 62 and older | Females age 62 and older | |
| SS<50% of Income | 60.4% | 53.4% |
| SS is 50% to 90% of Income | 21.8% | 23.4% |
| SS>90% of Income | 17.8% | 23.2% |
Of the women over age 62 collecting benefits as dependents, about 23 percent receive these benefits based on spousal benefits or earnings only. This has declined from 57 percent in 1960.9
Women are more likely than men to be single, widowed, or divorced in retirement. Widow(er)s can continue to be eligible for survivor benefits, but these benefits may be reduced or curtailed upon remarriage before age 60. Furthermore, those who have been divorced before 10 years of marriage may not be entitled to survivor benefits at the time of the Social Security beneficiary’s death.
- Women are more likely than men to be unmarried at older ages.
| Percent Married by Age Group10 | ||
| Age Group | Women | Men |
| 65 – 74 | 47% | 71% |
| 75 – 84 | 29% | 67% |
| 85 and older | 12% | 49% |
- In 2014, the largest group of women at ages 75 to 84 were widowed (47 percent), compared with married (34 percent), divorced or separated (14 percent), and never married (4 percent).
- The poverty rates for never-married women over age 65 are some of the highest for any subgroup in the country. For example, only 4.7 percent of married couples over age 65 have incomes below the federal poverty line; but for a never-married woman over age 65 that figure is 17.6 percent.11
This combination of factors means that the average woman has a higher risk of having insufficient income or depleting her savings in her retirement years.
Social Adequacy Features of Current System
The current Social Security program contains safety net, or social adequacy, features that help address retirement security issues for women.
Progressive Benefit Formula: A worker’s basic benefit, called the primary insurance amount (PIA), is determined using a progressive benefit formula that applies a declining scale of percentages as the average indexed monthly earnings (AIME) increases. Under the current PIA formula, the percentages (90, 32, and 15) remain the same from year to year but the amounts of AIME where they apply (bend points), increase each year based on the national average wage index.12
The PIA formula was designed so that lowerpaid workers receive a greater benefit as a percentage of wages than higher-paid workers. The PIA formula also provides proportionately greater benefits to a l l workers (not just lowerpaid workers) who have been absent from the labor force for significant periods than to those who have worked in covered employment throughout their adult lives. This is due to the fact that anyone with less than 35 years of covered employment will have years of zero earnings included in the calculation of their AIME and will have a lower AIME than someone at the same earnings level but who worked all years. Their PIA won’t be more but the PIA as a percentage of AIME will be higher. These features help women, who are usually lower-paid and likely to be absent from the labor force longer than men.
Benefits for Spouses: Under Social Security, a worker’s spouse is eligible to receive 50 percent of the worker’s benefit while the worker is alive, and 100 percent after the worker’s death (provided the spouse is not entitled to a higher benefit based on his or her own earnings history or on another spouse’s earnings history). Benefits are also payable to divorced spouses to whom the worker was married at least 10 years, without any reduction in benefits to the worker or to other family members. If a worker dies before becoming eligible for retirement, but was fully insured at death, Social Security pays benefits to the worker’s spouse and dependent children. As a result of spousal and survivor benefits, women with low earnings or little work history can still receive Social Security benefits based on a spouse’s earnings history.
Greater Expected Lifetime Benefits for Women Due to Their Longer Lifetimes: Social Security is a traditional retirement benefit where the retirement income is payable for life. This is different from an account-type system that accumulates a fund for an individual and then amounts are paid out at the discretion of each individual. These amounts may not last the full lifetime for people who live a long time or whose fund becomes depleted for other reasons. Since women’s lifetimes are on average longer than men’s lifetimes, women will on average receive more lifetime benefits from Social Security than similarly situated men. This feature of Social Security is an advantage for women.
Differences in Benefits and Taxes Within the Current System
Due to some of the social adequacy features of the current system, benefits are not the same proportion of Federal Insurance Contributions Act (FICA) taxes paid for all participants within the system. Specifically, one-earner couples (who were predominant when Social Security was created) receive benefits as a greater percentage of taxes paid than twoearner couples and single persons. The example in the chart that follows shows that there can be significant differences in retirement benefits and surviving spouse’s benefits for families with the same earnings histories and payroll tax contributions, but with different splits of earnings between the spouses. One-earner households receive a higher benefit from Social Security than two-earner couples with the same total household earnings. Moreover, the survivor of the two-earner couple receives a much smaller survivor benefit than the survivor of a one-earner couple with the same total earnings.
In the case of the two-earner couple where the primary earner has the same income as the oneearner couple, the secondary earner’s benefit is only marginally higher than if she or he did not work at all. Compared to the one-earner household, the survivor of the two-earner couple in this case receives no additional benefits, even though he or she may have worked for many years and contributed payroll taxes to the program. Stated differently, the secondary earner’s income is effectively taxed at a higher rate than the primary earner’s income, which has an adverse effect on women who are the secondary earner.
Current-Law Example
| Impact on Social Security Benefits of Different Family Circumstances | |||
|
One-earner couple (Example 1) |
Two-earner couple with equal earnings (Example 2) |
Two-earner couple with unequal earnings (Example 3) |
|
| Spouse A earns | $50,000 | $25,000 | $50,000 |
| Spouse B earns | $0 | $25,000 |
$25,000
|
|
Annual Social
Security tax (6.2%)
|
$3,100/year | $3,100/year | $4,650/year |
| Total monthly benefit at retirement* |
$1,770 spouse A
+ $885 spouse B
($2,655 total)
|
$1,120 spouse A + $1,120 spouse B ($2,240 total) |
$1,770 spouse A + $1,120 spouse B ($2,890 total) |
| Total monthly benefit to survivor | $1,770 | $1,120 | $1,770 |
*Estimates assume both are age 66 and retired in 2014
As more women continue to enter the work force and earn higher wages, more will have retirement benefits greater than 50 percent of their spouses’ benefits, so the spousal benefit will be provided to a declining proportion of women. Thus the disparity illustrated by Examples 1 and 2 will occur less frequently. However, as long as women continue to earn less on average than men, their retirement benefits will be less on average than their spouses’ benefits, so the survivor benefit (100 percent of the higher wage earner’s benefit) will continue to apply for a high percentage of women. The disparity in taxes paid illustrated by Examples 1 and 3 will continue to occur.
Proposals for Changes to Social Security
There have been many proposals in recent years to change Social Security, some of which would have particular impact on women. The intent of this section is to provide an analysis of some of the options and their impact. The discussion of the options has been separated into 1) options that address specific challenges faced by women, followed by 2) options that address Social Security’s funding challenges, which also impact women because of gender-related factors.
Options That Address Challenges Faced by Women
Modify the Computation Period for Benefits: Social Security currently uses a 35-year averaging period in the formula for calculating a benefit. If a worker is out of the workforce to care for family members for a period of years, her (or his) Social Security benefit could be adversely affected by those years with zero earnings. To address this issue, one reform option is to impute, or credit, income for a certain number of years for the stay-at-home care of a child. Another option is to reduce the 35-year period for every year of stayat-home care. This latter approach is already in use, to a limited extent, for disabled individuals where the 35-year average is reduced to 30 years under certain circumstances.
Earnings Sharing: Under an earnings sharing proposal, the spousal benefit would be eliminated. Instead, a couple’s earnings would be added together and split evenly each year. This would help dual-earner couples in which both spouses have substantial (but unequal) employment histories as well as divorced women—especially those who were married for less than 10 years. However, married households in which only one spouse has a substantial employment history, and which currently receive spousal benefits for little or no additional tax contributions, would be adversely affected. This proposal would also reduce the benefits of workers who shared their earnings, and then divorced after retirement.
Provide a Minimum Benefit: A proposal from the National Commission on Fiscal Responsibility and Reform would protect low-income workers, disproportionately women, against poverty by guaranteeing that an individual who worked at least 30 years would retire with a minimum income of 125 percent of the poverty line in 2017 and wage-indexed thereafter.13 A smaller minimum benefit would also be provided for those with between 10 and 30 years of service. This proposal would shift the balance toward social adequacy and away from individual equity (equivalent returns on contributions for all contributors).
Change Spousal Benefits: A proposal from the 1994-1996 Advisory Council on Social Security would reduce the 50-percent spousal benefit to 33 percent and would increase survivor benefits for two-earner couples to 75 percent of the total benefit paid to them when both were alive (or 100 percent of either worker’s benefit if greater).14 As the example comparing this proposal to current law illustrates, this proposal shifts benefits from retired one-earner couples to survivors of two-earner couples, predominantly women.
Advisory Council Proposal Spousal Benefit: 33 percent of insured worker benefit. The current law spousal benefit is 50 percent of the insured worker benefit.
Advisory Council Proposal Widow(er) Benefit: 75 percent of the total benefit paid when both spouses were alive (or 100 percent of either worker’s benefit if greater). The current-law widow(er) benefit is 100 percent of the insured worker benefit.
Current Law Spousal Benefit: 50 percent of insured worker benefit.
Current Law Widow(er) Benefit: 100 percent of insured worker benefit.
This proposal would also make it more likely that working spouses will be entitled to retirement benefits solely on their own work records, rather than as spousal benefits. However, it does reduce the benefits to low-earning and nonworking women to 33 percent of the benefit of the working spouse. This could be particularly problematic for divorced women, who have the highest poverty rate of the elderly.
| Advisory Council Proposal | |||
| One-earner couple |
Two-earner couple with equal earnings
|
Two-earner couple with unequal earnings | |
| Spouse A earns | $50,000 | $25,000 |
$50,000
|
| Spouse B earns | $0 | $25,000 |
$25,000
|
|
Total monthly benefit at
retirement*
|
$1,770 spouse A
+$590 spouse B
(Total $2,360)
|
$1,120 spouse A
+$1,120 spouse B
(Total $2,240)
|
$1,770 spouse A
+$1,120 spouse B
(Total $2,890)
|
|
Total monthly benefit
to survivor
|
$1,770 | $1,680 | $2,168 |
Earnings Sharing: Under an earnings sharing proposal, the spousal benefit would be eliminated. Instead, a couple’s earnings would be added together and split evenly each year. This would help dual-earner couples in which both spouses have substantial (but unequal) employment histories as well as divorced women—especially those who were married for less than 10 years. However, married households in which only one spouse has a substantial employment history, and which currently receive spousal benefits for little or no additional tax contributions, would be adversely affected. This proposal would also reduce the benefits of workers who shared their earnings, and then divorced after retirement.
Provide Additional Benefit for the “Oldest Elderly”: As retirees age, many liquidate their other resources and become relatively more dependent on Social Security for their support. The National Commission on Fiscal Responsibility and Reform report proposed a 5 percent increase in the monthly benefit to recipients after 20 years from initial eligibility. The bump up would be phased in over five years, 1 percent per year. This would have a disproportionate effect on women who make up approximately 60 percent of the population over age 80.
Options that Address Social Security’s Funding Challenges
Modify the Computation Period for Benefits: Most proposals in this category would phase in a change to lengthen the 35-year averaging period to help improve Social Security’s financial condition and to encourage people to work for more years. For example, current proposals would increase the averaging period from 35 to 38 or 40 years. In general, this option could affect women disproportionately, if additional years add lower-paid or zero-earnings years into the calculation of the average earnings upon which the benefit is based, unless they were allowed drop-out years for care giving outside the paid workforce. The effect on benefits of an increase to 40 years (phased in from 2017 to 2025) would depend on the worker’s earnings history but would reduce scheduled benefits, on average, by about 3 percent.15
Reduce Benefits Across the Board: One proposal would reduce all Social Security benefits by 3 percent.16 This reform option is consistent with individual equity but can be seen as hurting social adequacy (the benefit adequacy and social insurance aspects of the program). Such a change would have a greater impact on women, especially very elderly women without spouses, because of their longer life expectancy and greater reliance on Social Security relative to other types of retirement income.
Change the Benefit Formula: There are several approaches to changing the benefit formula. One would reduce the formula percentages (90 percent, 32 percent, and 15 percent for the three tiers of earnings) each year by a constant factor—for example, 1 percentage point per year. Such a change would maintain the formula’s progressivity but would gradually reduce benefit adequacy, particularly for low-wage earners, and would have a greater impact on women. Reducing only the percentages applicable to the higher-wage levels (i.e., the 32 percent and the 15 percent levels, but not the 90 percent level) would make the formula even more favorable to low-income workers relative to medium and high-income workers while maintaining program adequacy for very-low-wage earners because the 90 percent level would not change.
A similar proposal involves indexing the earnings tiers in the Social Security benefit formula to the Consumer Price Index (CPI), rather than the wage index. This has the effect of gradually reducing the (inflation adjusted) PIA below current levels since the wage index is generally higher than the CPI. Initially, this proposal would affect the highest earners the most, but it eventually could lead to a large shift in the balance between individual equity and social adequacy.
Increase the Full Retirement Age: The full (unreduced) retirement age is the earliest age at which full benefits are payable and currently ranges from ages 66 to 67, according to the worker’s year of birth. Benefits before the full retirement age may begin as early as age 62, but such benefits are permanently reduced to make lifetime benefits approximately actuarially equivalent, regardless of retirement age. Increasing the full retirement age as life expectancy increases is a means for solving the system’s financial problems by reducing benefits relative to current law and encouraging healthy workers with jobs to continue to work. Adopting this reform impacts benefit adequacy for those who cannot find or continue in employment in older ages. The increase in the full retirement age might be accompanied by increases in the initial eligibility age (62) and by changes in the disability benefit to protect those in jobs that don’t lend themselves to later retirement. These changes can be significant for women who rely heavily on Social Security for income in old age because increasing the full retirement age is functionally equivalent to reducing benefits.
Lower COLAs: Social Security benefits increase each year by a cost-of-living adjustment (COLA) equal to the annual change in the CPI. However, the cost of living of the elderly does not change at the same rate as the CPI because the elderly not only consume a different mix of goods and services than the population as a whole, but they may be less willing or able to substitute as product prices change.
A lower annual COLA is cumulative over time and the reduction in benefits will eventually disproportionately affect the oldest elderly, a group that can be projected to consist of mostly women and have a higher rate of poverty.
Increase the Payroll Tax: The FICA tax rate is 12.4 percent, split equally between employers and employees or the full amount paid by those who are self-employed. Increases to the tax rate place a disproportionate impact on lower-paid workers but would provide increased retirement benefits relative to current law. Women on average fall more into this category.
The impact of an increase in the payroll tax can be mitigated if it is phased in over a long period and if incomes for women are rising. For example, increasing the payroll tax rate by 0.1 percentage points per year for a sufficiently long period would improve Social Security finances while the after-tax income of workers could continue to increase. This scenario would require real wage increases over the same period. However, the rising costs of other social insurance programs that benefit women, particularly Medicare and Medicaid, may also require tax increases or benefit reductions or a combination of both.
Increase the Limit on Taxable Earnings: In 2015 about 83 percent of earnings in covered employment were below the 2015 limit on taxable earnings of $118,500.17 This limit also applies to earnings taken into account in the benefit formula. One proposal is to raise the earnings limit gradually so that Social Security taxes about 90 percent of all earnings in covered employment as was the case both in the beginning years of Social Security and after the last major reform of Social Security in the early 1980s.18 Such a change would require raising the limit by about 110 percent in addition to annual adjustments based on increases in average wages. Women on average would be impacted less than men by an increase in the limit on taxable earnings.
Conclusion
The current Social Security law is genderneutral, and contains spousal and subsidized benefit provisions that mitigate, but do not eliminate, factors that produce lower benefits for women: lower earnings, shorter work histories, longer life spans, greater dependency on spouses, divorce, prior death of spouses, etc. It is also important to note that the present system provides a lower level of benefits relative to Social Security taxes paid for two-earner families where the second earner has significant income. Policymakers who are evaluating various options to reform the Social Security system should not only address its financial problems, but also consider that Social Security remains an extremely important source of retirement income for women. This reliance on Social Security suggests that proposals to directly or indirectly address the unique issues of women and dual-earner families should be studied carefully and modeled to show the impact on families/beneficiaries in a variety of situations before being adopted.
| 1. | Four-in-Ten Couples are Saying “I Do,” Again; “Chapter 2: The Demographics of Remarriage”; Pew Research Center; accessed on May 10, 2017. |
| 2. | See Social Security Reform Options; American Academy of Actuaries; March 2014 for a comprehensive summary of program features and reform proposals. |
| 3. | “U.S. Bureau of Labor Statistics, 2014 Current Population Survey, Average Annual Data, Table 3”; U.S. Bureau of Labor Statistics; accessed on May 10, 2017. |
| 4. | BLS Reports – Women in the labor force: a databook; U. S. Bureau of Labor Statistics; December 2015. |
| 5. | “Facts & Figures: Women and Pay Inequality”; Moyers & Company; accessed on May 10, 2017. |
| 6. | Annual Statistical Supplement to the Social Security Bulletin; Social Security Administration; April 2015. Note that this 70 percent ratio includes the effects of both differing rates of pay and differing percentages of full-time employment. |
| 7. | “2015 OASDI Trustees Report, Table V.A3.—Period Life Expectancy”; Social Security Administration; accessed on May 10, 2017. |
| 8. | Income of the Population 55 and older, 2012 Survey, Table 8.B1; Social Security Administration; April 2014. |
| 9. | Fast Facts & Figures About Social Security, 2015; Social Security Administration; September 2015. |
| 10. | “2014 American Community Survey, Sex by Marital Status by Age”; U.S. Census Bureau; accessed on May 10, 2017. |
| 11. | Fast Facts & Figures About Social Security, 2015; op. cit. |
| 12. | For more on how the PIA and AIME are calculated as well as the current bend point levels, refer to the Primary Insurance Amount section of the Social Security Administration website. |
| 13. | The Moment of Truth: Report of the National Commission on Fiscal Responsibility and Reform; National Commission on Fiscal Responsibility and Reform; December 2010. |
| 14. | “1994-96 Advisory Council Report: Findings, Recommendations and Statements“; Social Security Administration; accessed on May 10, 2017. |
| 15. | Letter to the Honorable Jason Chaffetz, “Office of the Chief Actuary, November 9, 2011”; Social Security Administration; accessed on May 10, 2017. |
| 16. | See Proposals B7.1 and B7.2 in “Provisions Affecting Level of Monthly Benefits”; Social Security Administration; accessed on May 10, 2017. |
| 17. | Fast Facts & Figures About Social Security, 2016; Social Security Administration; August 2016. |
| 18. | “Provisions Affecting Payroll Taxes”; Social Security Administration; accessed on May 10, 2017. See E3.1. |
Share