• SUPPORT CATO
• FOR STUDENTS
• FOR MEDIA
• FREE NEWSLETTERS

• HOME

• ABOUT

• BLOG

• BOOKSTORE

• EVENTS

• EXPERTS

• MULTIMEDIA

• PUBLICATIONS

Books

Reviews & Journals

Public Filings

Policy Studies

Opinion & Commentary

• RESEARCH AREAS

• SPEAKERS BUREAU

• SUPPORT CATO

• CATO UNBOUND

• DOWNSIZING
  GOVERNMENT.ORG

• ELCATO.ORG

Briefing Paper No. 38

July 20, 1998

Greater Financial Security
for Women
With Personal Retirement Accounts

by Darcy Ann Olsen

Darcy Ann Olsen is an entitlements policy analyst at the Cato Institute.

---

Executive Summary

As privatizing Social Security becomes more likely, people are raising questions about how women would fare under a new system. Would poor women be able to weather market downturns? Would they be capable investors? What about women's aversion to risk?

Although the current Social Security system does not differentiate between men and women, on average, women receive lower benefits than do men because women tend to have lower wages and fewer years in the workforce. Thus poverty rates are much higher among elderly women than elderly men. While Social Security alleviates some poverty, there is room for improvement. Virtually every woman would be better off financially under a system of fully private, personal retirement accounts with earnings sharing.

Since the success of privatization is sensitive to the contribution rate, the greater that rate, the greater the benefits. Thus, a fully private system would benefit women more than a partly private two-tiered system.

Inequity in the Present System

By law, the Social Security system treats all workers equally. Yet the system has a disparate impact on women because they typically earn less, work fewer years, and live longer than do men. In particular, Social Security punishes married women who work and favors married women who do not work. A married women who works her entire adult life may not receive any more benefits than a married woman who has never worked; a couple with two breadwinners may get fewer benefits than a couple with one breadwinner and identical lifetime earnings, and widows of two-earner couples may get less than widows of one-earner couples.

Those inequities result from the way benefits are calculated. A spouse can receive benefits in one of three ways. First, she can receive benefits based on her own work history. Second, she can receive benefits based on her husband's work history. By law, a woman is automatically entitled to benefits equal to 50 percent of her husband's benefits, whether or not she has ever worked or paid Social Security taxes. Third, she can receive benefits based on a combination of the two.

When a woman qualifies for benefits both as a worker in her own right and as a spouse (or surviving spouse) of a worker, she is subject to the "dual entitlement rule." That rule prevents her from collecting both her own retirement benefit and her spousal benefit. Instead, she receives only the larger of the two. And because the typical woman earns less and works fewer years than her husband, 50 percent of her husband's benefits is often larger than the benefits she would be entitled to receive in her own right. Consequently, she receives benefits based on only her husband's earnings--she receives no credit or benefits based on the payroll taxes she has paid. A woman who never worked at all receives exactly the same benefits.

The second inequity that results under Social Security's dual-entitlement rule is that a couple with two breadwinners may get fewer benefits than a couple with one breadwinner and identical lifetime earnings. Table 1 illustrates this point.

Each of the three couples has the same total earnings, yet couple A with one breadwinner receives higher benefits than do couples B and C with two breadwinners. During one year, couple A will receive $1,116 more than couple C. After 10 years, couple A will have received more than $11,000 more in retirement benefits than couple C. As men and women who reach age 65 are expected to live to age 80 or beyond, that inequity can have a significant impact on a couple's quality of life for many years.

While the dual-entitlement rule has a negative impact on many two-earner couples during their retirement years together, its most pernicious impact is often felt after a husband dies. Social Security's survivor benefit rules can leave widows with up to 50 percent less income than the couple was receiving when the husband was alive. That is one reason why the poverty rate among widows is 19.2 percent, two times greater than among widowers. And, in general, the more of the couple's earnings the widow earned, the smaller the share of the couple's retirement benefit she receives after her husband dies. Table 2 illustrates this point.

Table 1
Couples’ Benefits under Dual Entitlement

	Monthly earnings ($)	Monthly benefit ($)
Couple A
Husband	1,000	573
Wife (no income)	0	287
Total	1,000	860
Couple B
Husband	500	413
Wife	500	413
Total	1,000	826
Couple C
Husband	667	467
Wife	333	300
Total	1,000	767
Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 4.
Note: Monthly Earnings is the Average Indexed Monthly Earnings (AIME), which is found by adding the 35 years of a worker's highest indexed earnings and dividing by 420 (the number of months in 35 years). In this example, it is assumed that the workers' combined earnings equaled $1,000. The Monthly Benefit is the Primary Insurance Amount (PIA). The following progressive benefit formula is applied to the AIME to determine the PIA (1996): 90% of the first $437 of AIME, 32% of AIME from $437 to $2,635, and 15% of AIME over $2,635.

Each of the three couples has the same total earnings, yet the widow who never worked (A) receives higher benefits than the widows who worked (B and C). Widow A's benefits are approximately 16 percent higher than widow B's and 10 percent higher than widow C's. As Tables 1 and 2 indicate, the one-earner couple (couple A) receives the highest retirement benefits while the husband is alive, and the widow receives the highest survivor's benefit. In addition, the widow who made as much money as her husband receives less than the widow who earned only half as much as her husband.

Anna Rappaport of William M. Mercer found that the situation for low-income widows who worked is even worse. For example, the wife of a couple with $34,200 in annual pay gets $1,082 as a widow if she never worked, while the wife who brought home half that paycheck gets a widow's benefit of only $674. That is a difference of $408 per month.

The Social Security Administration reports that 24 percent of married and widowed women have their benefits slashed by the dual-entitlement rule. That number is projected to increase to 39 percent by 2040, as increasing numbers of women earn higher wages and work more hours. As Jonathan Barry Forman, former tax counsel to Sen. Daniel Patrick Moynihan (D-N.Y.), puts it, "In short, the Social Security system takes billions of payroll tax dollars from these working women and gives them no greater Social Security benefits in return."

Table 2
Widows' Benefits under Dual Entitlement

	Monthly Earnings ($)	Couple's Benefit ($)	Widow's Benefit
Couple A
Husband	1,000
Wife (no income)	0	860	573
Total	1,000
Couple B
Husband	500
Wife	500	826	413
Total	1,000
Couple C
Husband	667
Wife	333	767	467
Total	1,000
Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 5.
Note: Monthly Earnings is the Average Indexed Monthly Earnings (AIME), which is found by adding the 35 years of a worker's highest indexed earnings and dividing by 420 (the number of months in 35 years). In this example, it is assumed that the workers' combined earnings equaled $1,000. The Monthly Benefit is the Primary Insurance Amount (PIA). The following progressive benefit formula is applied to the AIME to determine the PIA (1996): 90% of the first $437 of AIME, 32% of AIME from $437 to $2,635, and 15% of AIME over $2,635.

The negative impacts of the dual-entitlement rule are exacerbated by a handful of other factors that make women disproportionately dependent on Social Security benefits. As a result of lower earnings and fewer years of work, women, on average, earn less Social Security benefits than do men. In 1995 male retirees received $810 in monthly benefits while women received only $621, on average. Lower earnings and part-time employment also make it more difficult for women to accumulate private savings for retirement. In addition, women are less likely than men to have employer-provided pension plans. Even if they do have pension plans, they generally save too little because of their moves in and out of the job market.

Those gender-specific issues aside, women, like men, face the larger problem of Social Security's looming debt and declining rate of return. Federal Reserve Board chairman Alan Greenspan estimates that Social Security's unfunded liability is roughly $9.5 trillion. If the government intends to make good on its promise to pay retiree benefits, it will have to raise taxes or cut benefits in order to meet that revenue shortfall. The Social Security board of trustees has estimated that it would take a tax hike of at least 6.3 percentage points to put the program in the black. A tax hike of that size would force workers to pay one-fifth of their wages to Social Security. Of course, cutting benefits is no solution either; benefit cuts would give workers an even worse deal than does the current system. Many of today's young workers can expect to get a negative rate of return from Social Security, according to the nonpartisan Tax Foundation. And, as the American Association of Retired Persons has pointed out, women would suffer most under reform proposals that reduce retiree benefits.

Benefits of Full Privatization

Women, like men, want to know what would be the likely results under a private system in which payroll tax payments were redirected into personal accounts similar to individual retirement accounts or 401(k) plans. Would private accounts increase the overall level of women's retirement benefits? Would private accounts address poverty among widows? Would private accounts end discrimination against working wives?

To answer those questions, Ekaterina Shirley and Peter Spiegler, graduates of Harvard University’s Kennedy School of Government, conducted two empirical analyses. The first is a retrospective analysis using actual earnings histories of 1,585 men and 1,992 women who retired in 1981. The researchers compared Social Security benefits with the benefits that hypothetically would have accrued under a private plan with a 7 percent contribution rate, a 6.2 percent rate of return, and 50-50 earnings sharing between spouses where applicable. Earnings sharing lets married couples split their contributions 50-50 before depositing them into each person's account.

Shirley and Spiegler found that all but .11 percent (approximately 3) of the women collecting benefits would have been better off under the private system. For those women, the difference between the plans was exactly zero--no woman was worse off under the private system. For 3.7 percent (approximately 110) of the women, the difference was less than $2,000. Even though the absolute dollar difference appears small, it is significant relative to total benefits from Social Security. Overall, the median value of the accrued difference between benefits from Social Security and benefits from a privatized plan was $30,000 for single women, $26,000 for wives, $21,000 for divorcees, $23,000 for surviving divorcees, and $20,000 for widows. As a percentage of Social Security benefits, that difference is substantial. The median values of that percentage are 58 percent for single women, 208 percent for wives, 67 percent for divorcees, 58 percent for surviving divorcees, and 97 percent for widows.

In the second analysis, Shirley and Spiegler conducted a prospective simulation since the cohort of women in the workforce today has significantly different labor and marital characteristics than the one that retired in 1981. As complete lifetime earnings histories for women who are currently in the workforce do not exist, the research team simulated the effects of various retirement plans. They compared Social Security; a fully private system; and a two-tiered, or partly private, system. Under the fully private system, the assumed contribution rate is 10 percent. The partly private approach would channel 5 percentage points of payroll taxes into a personal account, with the remaining 7.4 percentage points going to Social Security to finance a "flat benefit" and survivor's and disability insurance. The flat benefit is equal to 2/3 of the poverty rate. As they did in the retrospective analysis, the researchers assumed a 6.2 percent rate of return on invested contributions.

As Figure 1 shows, in every case the fully private system with a contribution rate of 10 percent would bring all women--whether collecting benefits based on their own earnings or as wives, divorcees, or widows--with full earnings histories more than twice the retirement benefits of Social Security.

Moreover, the greater the contribution rate, the greater a woman's financial security in retirement. Thus, the fully private system generates significantly higher retirement benefits than does the partly private, two-tiered system. The partly private system provides only slightly greater benefits than Social Security. The results are similar for women who have moved in and out of the job market, as Figure 2 shows.

In every case, the fully private system brings all women significantly greater benefits than does either Social Security or the partly private system. For example, the fully private plan gives married, divorced, and widowed women (with full or interrupted earnings histories) at least $200,000 more in retirement benefits than does Social Security or the partly private system. That's better than $10,000 per year.

To answer questions about the potential impact of privatization on women with low to moderate incomes, Shirley and Spiegler ran a simulation using half the national mean wage level of women. Figures 3 and 4 show that women with low to moderate incomes (with full or interrupted earnings histories) would do far better under a fully private system than under either Social Security or the partly private system.

Figure 1
Accrued Retirement Income of Mean-Income Women with Full Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 12.

Figure 2
Accrued Retirement Income of Mean-Income Women with Interrupted Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 12.

Figure 3
Accrued Retirement Income of Low-Income Women with Full Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 13.

 

Figure 4
Accrued Retirement Income of Low-Income Women with Interrupted Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 14.

For example, a married woman with a low income who has moved in and out of the workforce could expect to gain roughly $125,000 more in benefits under the private system than under Social Security. That's nearly $550 more per month than Social Security would provide. Even women in the worst-case scenario--low-income single women who do not benefit from the earnings sharing provision and who have moved in and out of the workforce--would receive greater benefits under full privatization than under Social Security, nearly $100 more per month.

One potential concern is that the positive benefits under privatization are simply the result of high contribution rates. To address that concern, Shirley and Spiegler calculated how much each program gives in return for each tax dollar invested. In other words, they wanted to find out whether women were getting their money's worth under each program. For example, Figure 5 shows that a widowed woman with a complete work record would get approximately

$1 in Social Security benefits for each $1 she contributed; under the fully private plan, she would get approximately $2.50 for each $1 contributed.

Figure 5
Money's Worth to Mean-Income Women with Full Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 12.

 

Figure 6
Money's Worth to Low-Income Women with Interrupted Earnings Histories

Source: Adapted from Ekaterina Shirley and Peter Spiegler, "The Benefits of Social Security Privatization for Women," Cato Institute Social Security Paper no. 12, July 20, 1998, p. 14.

 

In a final simulation, the researchers used half the national mean wage level of women in each category to examine whether low-income women were getting their money's worth from each program. As Figure 6 shows, full privatization would give low-income women with interrupted earnings histories much more value for the dollar than would either Social Security or the partly private plan.

The money's worth calculations demonstrate that, even taking into account Social Security's "progressive" benefit structure, all categories of women would still get more for their money under a fully private plan.

Shirley and Spiegler's retrospective, prospective, and value-for-the-dollar calculations show how a fully private system with a contribution rate of 10 percent would be able to bring all women--single, married, divorced, or widowed with low to moderate or average income--greater financial security than does Social Security. The implications are real and significant for women, yet many important questions still exist.

Concerns about Women and Private Accounts

People have raised more concerns about women and private accounts than can be addressed in this short paper. However, a brief discussion of a few of the most important concerns should alleviate much uneasiness about personal accounts.

Low-Income Women as Investors

"Because women are more likely to be living in low-income households, they generally have less access to good investment advice." It is unlikely that low-income women would not have adequate access to good investment advice. As the American Association of Retired Persons points out, "Lots of good information on saving and financial planning is free--from AARP, investment companies, the Internet, and your local library. Also, free or low-cost seminars specifically designed for women are offered in many communities." A market-based retirement system will undoubtedly increase investment companies' outreach efforts to women. What is perhaps even more important, however, is that full privatization does not require that every participant be an intelligent or experienced investor. The history of 401(k) plans has demonstrated that workers of all income groups can do well by entrusting their pension benefits to experienced investors who, for the most part, have fulfilled their responsibilities. Under a well-structured system with fully private accounts, low- and high-income workers could expect to receive guidance from fund managers in much the same way. As Melissa Hieger and William Shipman of State Street Global Advisors point out, "There is no need for a worker who chooses the market-based system to know how markets work as long as the pension system is properly structured and sensible guidelines are followed. In fact, most proposals for a privatized national retirement system have regulatory elements that restrict investment strategies that are either too risky or that would be insufficiently aggressive to provide needed retirement benefits."

Low-Income Women and Market Downturns

"Low-income women . . . would be less able than more affluent women to weather market downturns." One way to address that concern is to see how low-income women would have fared historically under a market-based system. That can be done by comparing Social Security benefits with simulated market benefits for low-income workers. For example, Hieger and Shipman compared an initial monthly Social Security benefit with an initial balanced-fund (60 percent equities, 40 percent bonds) benefit for low-income workers born in 1950 and 1970. The low-income worker born in 1950 could expect a monthly Social Security benefit of $668; the balanced-fund benefit would be $1,514. The low-income worker born in 1970 could expect a monthly Social Security benefit of $799; the balanced-fund benefit would be $1,431. In both cases, the market affords low-income workers higher retirement benefits than does Social Security. Those results are consistent with other studies that show higher retirement benefits from markets than from Social Security. If, however, the worst-case scenario arose leaving a worker with insufficient benefits upon retirement, the government could finance a safety net from general revenues. Moreover, if one believes the market-based system is inferior to Social Security, most privatization plans would allow workers the option of staying in the present system. The freedom to choose is particularly important to low-wage women who do not earn enough to save and invest on their own. That inability to invest is largely due to high payroll tax rates. Forcing women to stay in a system that takes 12.4 percent of their wages only to cheat them of a secure retirement is simply unjust. Low-income women should have the freedom to invest their earnings in a way that will increase their chances for a financially secure retirement.

Risk Aversion of Women

"With individual accounts, women may fare worse than men because they are more risk averse." There is some evidence that women tend to be more conservative investors than men; however, many studies that purport to show that did not control for education levels or investment-specific knowledge--factors that may account for some differences in investment behavior. According to the General Accounting Office, people who are given information about their investment choices and potential returns are more likely to invest more than those who do not receive such information. Investment companies compete vigorously to educate and attract female clients. For example, OppenheimerFunds has distributed a 160-page booklet called "A Woman's Guide to Investing." Merrill Lynch, PaineWebber Group Inc., and Smith Barney have similar marketing strategies. In addition, women who have been investing for a long time pursue investment strategies that are very close to those of men. Certainly, experience shows that it is unreasonable to suggest that women, simply because of their gender, are not capable of becoming perfectly competent investors. Finally, in a well-structured private system, women, as well as men, could expect to rely on investment managers to help with investment decisions.

Conclusion

Shirley and Spiegler's research demonstrates how full privatization with earnings sharing can address the shortcomings of the current Social Security system. First, both the retrospective and the prospective analysis show that fully private accounts would significantly improve the retirement incomes of women--single, married, divorced, or widowed with low-to-moderate, moderate, or average income--which would begin to address the problem of poverty that exists under Social Security. Second, fully private accounts would end the discrimination currently faced by women under Social Security by ensuring that every dollar earned by a woman had a strictly positive effect on her retirement income. Finally, changing Social Security into a fully funded system would help ensure that future generations grow up without being saddled by unnecessary debt and grow old with financial security.

Notes

I am grateful to Ekaterina Shirley and Peter Spiegler who did the research on which much of this analysis is based. Special thanks to Lea Abdnor for her constructive comments, to Carrie Lips for her expert research assistance, and to Michael Tanner for his support and direction. I take full responsibility for all errors.

© 1998 The Cato Institute
Please send comments to webmaster