In the 100 years since the ratification of the 19th Amendment, women have made substantial and well-publicized gains—both in absolute terms and relative to men—in educational attainment, employment, and earnings. Their status in retirement, however, has received far less attention.
Variations across individuals or groups in economic status in retirement can be traced to differences in labor market earnings, retirement saving derived from those earnings, and other factors including marital status, life expectancy, risk aversion, and financial acumen.
These factors help explain women’s resources in retirement. For a variety of reasons, women earn less on average over the course of a lifetime than men do. Lower lifetime earnings make it harder for women to save for retirement. Exacerbating these differences, women are on average longer lived, more risk averse, less financially literate, and more likely to have greater caregiving responsibilities than men.
Public policies that aim to boost women’s status in retirement should focus on the ways women participate in the labor market and in wealth accumulation programs as well as on specific retirement programs and benefits.
For several interrelated reasons, average lifetime earnings are substantially lower for women than for men. Primarily, women provide a majority of unpaid family caregiving, which can lead women to interrupt their careers, seek part-time jobs in the market, or work in low-wage occupations to maintain flexibility. Caregiving provided during women’s 20s and 30s, when careers are formed and when age-earnings profiles are relatively steep, creates career-long earning losses. One study found that a woman with one child earns 28 percent less on average over her career than a woman without children, partially as a result of time out of the work force. (In sharp contrast, becoming a father typically does not reduce a man’s earnings.) Each additional child reduces average women’s earnings by another 3 percent. Women are also more likely than men to care for their aging parents—a responsibility that predominantly falls on women over the age of 50. People who leave the labor force early to care for a parent or other elderly relative lose an average of $142,000 in wages.
Even within occupations, however, women receive unequal pay for similar work. Among full-time, year-round workers in 2018, median women’s earnings equaled 81.1 percent of median men’s earnings. Controlling for age, education, job tenure, occupation, job title, location, and industry, the figure rises to 94.6 percent. Because this estimate controls for job title and occupation, it obscures important facets of gender inequality—the social constraints and biases that limit women’s opportunity for career advancement (i.e., promotions) and pressure women into lower paying jobs. The wage differential varies across occupations but is almost always to women’s detriment.
Progressive income taxation of family income provides significant disincentives for married women to work in the paid labor force. By taxing the first dollar of a second earner’s income at the same marginal tax rate as the last dollar of the primary earner’s income, the tax system discourages work among married women.
From earnings to retirement wealth
Lower lifetime earnings lead to lower retirement wealth. The most important link is through Social Security, which provides over half of family income to 52 percent of the elderly and at least 90 percent of income to 25 percent of the elderly. Women receive Social Security benefits that are, on average, 80 percent of those men receive. Benefits are based on a person’s 35 highest earning years. Women with long career interruptions risk not having 35 years with positive earnings, and the wage gaps noted above further reduce women’s benefits. The motherhood penalty applies here, too: having a first child reduces a woman’s Social Security benefits (through reduced earnings) by an average of 16 percent. Each additional child increases the gap by 2 percent. Women who leave work to care for an elderly family member not only lose wages, they also lose an average of $131,000 in lifetime Social Security benefits. Spouses (or ex-spouses, if the marriage lasted more than 10 years) can choose to receive benefits based on their own earnings history or to receive half of their spouse’s benefit. Given the trends in employment, earnings, and marriage, women are increasingly choosing to receive their own benefits.
Lower lifetime earnings also reduce the amount of wealth women can accumulate from employer-sponsored retirement plans, which come in two forms—defined benefit (DB) and defined contribution (DC) plans. Typically, in a DB plan, retirement benefits are paid as an annuity and depend on the worker’s job tenure and salary. In DC plans, an employee contributes funds to an individual account (with the employer potentially making a matching or other contribution). DC plan benefits come from the contributions and returns that accrue over time. Over the past several decades, DC plans (including 401(k)s) have increasingly substituted for DB plans, which are common now only for government and unionized workers.
The shift away from DB plans has helped women in a key sense: DB plans are designed for those with long, uninterrupted careers. Benefits generally do not vest immediately and are usually “back-loaded.” That is, they require several years of job tenure to receive any benefit, and long job tenure is rewarded with disproportionately larger benefits. While the gap between job tenure for men and women has fallen over time, men still have longer job tenure on average. DB plans were designed to benefit workers with career jobs (who, several decades ago, were almost exclusively men). As a result, retired women are two-thirds less likely to be receiving annuity and employer-based pension income, and among recipients, women’s benefits average two-thirds of the average for men.
DC plans tend to be less structured around job tenure and are portable. Workers are immediately vested in their own contributions and can take their accounts with them at job separation, which better fits women’s work patterns. As of 2012, men and women working full-time, year-round in the private sector had nearly equal access to DC plans (45 and 46 percent, respectively) and equal take-up rates (81 percent). Vanguard data show that in 2016 the average account balance for women was two-thirds the average for men, but almost all of this difference was the result of wage differences rather than saving behavior given earnings.
But managing DC accounts poses some new obstacles, too. First, women, on average, express more risk aversion and tend to hold overly conservative portfolios, which reduces the returns they earn on savings. In practice, the prevalence of target-date funds as default investments has helped women assume similar levels of risk (and returns) as men.
Second, DC plans often require participants to make active decisions about their contribution, investment, withdrawal, distribution, and rollovers. Automatic approaches to enrollment, escalation, investment allocation, etc., have helped people manage these accounts, but workers are still responsible for ensuring their own savings adequacy. As a group, however, women are less financially literate than men, though their financial literacy increases after their husbands’ deaths.
Third, DC plans rarely pay benefits in the form of a stable monthly income stream. This hurts women, who—as discussed below—have longer life expectancies than men. Under a DB plan, women and men with identical earnings histories receive equal monthly payments, even though women are expected to live longer.
From retirement wealth to retirement security
For a given amount of wealth, women face several obstacles relative to men in maintaining living standards in retirement.
Women tend to live longer than men and thus often have to draw down their retirement wealth over a longer period of time. In 2020, average life expectancy at age 65 is 21.1 years for women and 18.6 years for men. (The gap between men and women is fairly consistent across racial and ethnic groups.) As a result, for a given level of retirement wealth at age 65, women can afford to consume about 7 percent less per year than men. Women are more likely to run out of retirement savings, especially because older women are more likely to be the surviving partner, living on less Social Security income and with their partner’s medical bills. Perhaps in response, women over the age of 55 are increasingly remaining in the workforce. (Older men are, too, but the increase is smaller.) Working longer gives people more time to save for retirement and reduces time they need to depend exclusively on retirement savings.
Poverty rates for women rise with age—from 8.6 percent among women aged 65 to 69 to 13.5 percent among women aged 80 or older—and are closely tied to family circumstances. In 2017, among elderly women, the poverty rate was 4.3 percent for those who were married, 13.9 percent for widows, 15.8 percent for divorced women, and 21.5 percent for never married women. More and more women reaching retirement are divorced or unmarried. In every marital status group, women with children had higher poverty rates than women without children, a pattern that does not hold for men.
Although gender inequality is a pervasive problem with deep cultural roots, narrow economic policies can help reduce disparities between men and women. But fixing a retirement system that was not designed to accommodate women’s experiences will require significant changes, not just in retirement policy but in labor market practices and policies as well.
Making paid work more amenable to women is a key step. First, a robust federal paid family and medical leave program would let people save for retirement and earn Social Security credits while providing care to children and relatives. Under such a program, employers would be required to provide employees a certain amount of paid time off for caregiving, parental leave, and periods of illness. Eight states and the District of Columbia already have such programs. Second, under a Social Security caregiver credit, the government would assign a value to caregiving work that would be used as part of Social Security benefit calculations. The U.K., France, Germany, and Sweden provide caregiver credits for public pensions and both President Trump and former Vice President Biden have called for the U.S. to adopt a version of this policy. Third, subsidizing high-quality childcare would enable more mothers to remain in the workforce. Fourth, reforming the tax code either to provide a second-earner tax credit or to tax individuals rather than families would improve incentives for married women to work.
The retirement saving system presents many opportunities for reform. While women are increasingly working for employers that offer retirement plans, women still comprise a majority of the estimated 55 million U.S. workers who are not eligible for or covered by an employer-sponsored plan. Adoption of a nationwide automatic IRA program could make most of these workers (as well as other part-time and lower-wage workers) eligible to participate in a tax-preferred workplace retirement program. Already, an increasing number of states have launched their own auto-IRAs. Expanding the saver’s tax credit to make it refundable and to provide higher effective matching rates would raise the return to saving for women with low and moderate incomes.
Strengthening the social safety net is a third major component for reform. For example, boosting Supplemental Security Income benefits to close the gap between Social Security income and the poverty threshold could lift nearly 5 million elderly people, a majority of whom are women, out of poverty. Increased support from Medicare and Medicaid for end-of-life care could alleviate the stress on widows. Finally, a thorough re-examination and reform of divorce laws could help women be more adequately prepared to meet financial needs.
While differences in economic outcomes between men and women are sizable, they are falling over time. Policymakers can accelerate that change and help bring about gender equity.
This piece is part of 19A: The Brookings Gender Equality Series. Learn more about the series and read published work »