The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Note: This report is available in its entirety in the Portable Document Format (PDF).
Foreword
Despite women's increased labor force attachment, rising earnings, and expanded pension coverage, Social Security is the mainstay of retirement income for older women, 90 percent of whom receive benefits from the program. Widowed, divorced, and never-married women are especially dependent on Social Security, which accounts for at least half of the income of nearly three-fourths of nonmarried women aged 65 and older. It is the only source of income for about
one-fourth of them.
Though the program is gender neutral, many of Social Security's provisions are particularly beneficial to women. Women are, for example, disproportionately represented among low-wage workers. Social Security's benefit calculation formula replaces a higher proportion of the earnings of low-wage workers than of higher-wage workers. Women's longer life expectancy is not taken into consideration when benefits are calculated, so men and women with identical work histories and earnings receive identical benefits. In addition, women's lower life-time earnings and discontinuous work histories mean that their spouse and/or survivor benefits are often higher than their own retired worker benefits. They receive the larger of the two benefits. A divorced women who had been married for at least 10 years also qualifies for benefits based on the earnings record of her former husband.
Over the years, the Social Security Act has been amended numerous times in ways that further enhance the financial well-being of women, in particular. The duration of marriage requirement, for example, has been shortened from 20 to 10 years. A divorced spouse who has reached retirement age can begin drawing Social Security benefits on the earnings record of a former spouse, even if the former spouse has not retired. Women who remarry after the age of 60 now have the right to collect Social Security based on their current or former husband's record, whichever produces higher benefits. Yet, despite all the ways in which women benefit from Social Security, millions of older women have incomes near or below the poverty level.
In the research reported on in this study, Melissa Favreault and Frank Sammartino of the Urban Institute examine the impact on low-income and older women of a number of additional reforms to Social Security, several of which, in one version or another, have been proposed in recent years. Increasing survivor benefits, for example, has long had its proponents, under the assumption that doing this could reduce substantially the number of women who fall into poverty upon widowhood. Using DYNASIM3, a dynamic microsimulation model, Favreault and Sammartino compare distributions of Social Security benefits under current law and under eight reform options that would increase benefits and under three options that would "pay" for some benefit improvements by combining benefit increases with benefit cuts. The analyses reveal widely varying and sometimes unanticipated impacts of the various reforms and should serve as a warning that, as the investigators observe, "using intuition alone to guide reform efforts can be dangerous."
Sara E. Rix, Ph.D.
Senior Policy Advisor
AARP Public Policy Institute
Executive Summary
Introduction
Social Security is vitally important to American women, keeping millions out of poverty in old age each year. Yet many analysts have raised concerns that it does not do enough for some needy older women, for example, low-income unmarried women. Others express concern that the program provides smaller benefits to some women who have contributed to the system than to other women who have never worked (for example, married women who do not work outside the home and who have high-earning husbands can receive more generous benefits than some low-earning workers). They argue that the system should be reformed to increase women's economic security in retirement, while treating women with different family arrangements more equitably. However, the Social Security system's enormous long-term fiscal deficit complicates the prospects for reforming the program to better meet the needs of contemporary women.
Purpose
This report first describes how women's Social Security benefits are likely to change over the next four decades. It then considers how proposed reforms to the Social Security system might affect American women's economic well-being. Proposals advanced by current movements to reform women's Social Security benefits fall into two major groups: The first includes structural reforms that would fundamentally change the system, for example, by introducing individual accounts (either added on to the current system or carved out of it). The second group deals with programmatic reforms that would alter parameters of the existing system without changing its basic features. This report concentrates on the latter type. We examine 11 incremental proposals that would modify the relationship between benefits and family status or age. These reforms include eight proposals that would increase benefits and three packages that would combine benefit increases for some with reductions for others, as follows:
Proposals That Increase Benefits
- Change the current level of survivor benefits by:
1. increasing the survivor benefit to 75 percent of a couple's benefit;
2. increasing the survivor benefit to 67 percent of a couple's benefit.
- Expand minimum benefits by:
3. providing a benefit equal to 60 percent of the wage-indexed poverty threshold for workers with at least 20 years of qualified earnings, and increasing the benefit by 2 percentage points for each additional year of qualified earnings to reach a maximum of 100 percent of the wage-indexed poverty threshold for workers with 40 or more years of earnings.
- Expand eligibility and benefit levels for divorced spouses and survivors by:
4. raising the divorced spouse benefit from 50 to 100 percent of the worker's Primary Insurance Amount (PIA);
5. raising the divorced spouse benefit from 50 to 75 percent of the worker's PIA;
6. reducing the required marriage duration for divorced spouse and survivor benefits to 5 years;
7. reducing the required marriage duration for divorced spouse and survivor benefits to 7 years.
- Recognize childcare by implementing credits, specifically by:
8. crediting parents with half of the average wage for up to five total years in which they have a child under age six in their care.
Packages That Combine Benefit Increases with Cuts
- Change the current mix of spousal and survivor benefits by:
1. combining an increase in the survivor benefit to 75 percent of a couple's benefit with a reduction in the spouse benefit to 33 percent of a worker's PIA and reductions of 5 percent each in the upper two bend percentages; and
2. combining an increase in the survivor benefit to 67 percent of a couple's benefit with a reduction in the spouse benefit to 33 percent of a worker's PIA and reductions of 1 percent each in the upper two bend percentages.
- Change the indexation of benefits in retirement, namely by:
3. cutting initial benefits by 12 percent but wage indexing subsequent benefits.
Methodology
We use a dynamic microsimulation model, DYNASIM3, to compare distributions of Social Security benefits under current law and under the alternative reforms. The model ages a representative population of more than 100,000 persons from the 1990-1993 Survey of Income and Program Participation (SIPP) in yearly increments, replicating family and individual economic and demographic processes such as birth, death, marriage, divorce, work, and earnings. Through this aging process, the model produces a file of life histories that analysts can use to compute Social Security benefits under current law and under various alternatives. We focus on results for women who are ages 62 and older in 2040, and pay special attention to differences in benefits, total income, and poverty status by age, lifetime earnings, and marital status.
Principal Findings
We find that the composition of women's current-law Social Security benefits will change markedly in coming decades. Fewer women will be entitled to benefits solely as spouses or survivors, and more will receive worker-only benefits, dually entitled spouse benefits, and especially dually entitled survivor benefits. This trend should reduce, though not eliminate, some concerns about the equity of Social Security benefits.
The distributional implications of the benefit increase proposals that we examine differ in significant ways. Increases in survivor benefits effectively target older and widowed women, but grant the largest increases to women in the highest lifetime family earnings quintiles. (Caps on the survivor benefits could address this concern.) Increased eligibility and benefit levels for divorced spouse benefits target more of their gains to women in the bottom family earnings quintiles, but, of course, do not reach those women who never marry, and only reach a small fraction of married or widowed women. Higher minimum benefits have a more global reach, not excluding any potential recipients on the basis of marital status, and target those with the lowest lifetime earnings quite effectively. Childcare credits have fairly modest effects but are likewise well targeted toward women at the bottom of the lifetime earnings distribution.
The three packages that combine benefit increases and cuts likewise have different effects. Although all three successfully redistribute women's benefits from earlier to later in life, the wage indexing proposal does so more progressively than the spouse/survivor options. The simulations of packages do suggest that reforms to improve adequacy and equity of the Social Security system for women could be designed to be low-cost or revenue-neutral, an important consideration given that the current system is underfunded.
Conclusions
Models can reveal important distributional differences in proposals aimed at improving the Social Security system's treatment of women. Policymakers should be careful not to rely on intuition when designing reforms to shore up women's Social Security benefits, but rather to rely
on rigorous analyses. Using DYNASIM3, we have demonstrated that policymakers can change parameters in the existing system to target the highest-risk low-income and older women. Our analyses also show how legislators can combine a series of changes into packages that meet multiple needs.
Note: This report is available in its entirety in the Portable Document Format (PDF).
PDF
Printer-friendly summary
