Why Women Retire Poorer
American women reach 65 with smaller Social Security checks and smaller retirement accounts, then live longer on them. The gap traces to lower lifetime earnings and years out of paid work. It does not trace to timid investing.
By Teona Abuladze · 7 min read
In December 2024, women collecting a Social Security retired-worker benefit received an average of $1,780 a month. Men received $2,181. The agency publishes both figures in the same table of its 2025 Annual Statistical Supplement, one line above the other.
That $401 is the cleanest number in this whole subject, because Social Security is the one retirement system almost everybody is in. It is where the arithmetic of a working life shows up with no cushioning.
Of the 30.3 million women aged 65 and older receiving benefits that month, 60.8 percent were entitled solely on their own earnings record, and only 15.2 percent were paid purely as somebody’s spouse or widow. Among women 62 and older, the equivalent share has fallen from 57 percent in 1960 to about 15 percent now. Nearly everything about how American women reach retirement has changed, and the gap is still there.
The machine counts thirty-five years, and it counts zeros
Social Security does not average what you earned. It averages 35 years, whether or not you worked 35 years.
The formula indexes your lifetime earnings for wage growth, takes the highest 35 years, sums them, and divides by 420 months. If you have 30 years of earnings, the remaining five slots are filled with zeros and divided in anyway. That average then runs through a progressive formula that in 2026 pays 90 percent of the first $1,286, 32 percent of the amount up to $7,749, and 15 percent above that.
Here is what the zeros cost. Take a woman whose indexed earnings run $60,000 a year. With 35 years on the record, her monthly benefit at full retirement age works out to $2,345.80 under the 2026 formula. With 30 years and five zeros, it is $2,117.00. Five years out of paid work costs her $228.80 every month for the rest of her life, roughly $2,750 a year before cost-of-living adjustments.
Five years at home cost about $229 a month, forever. The formula does not ask why the years were empty.
How many zeros do women actually carry? An SSA research note published in November 2000 measured it. Among people born between 1926 and 1945, women averaged 12.5 zero years inside their highest 35; men averaged 3.7. Among baby boomers born 1946 to 1960, women’s figure fell to 3.7 and men’s to 2.8. That collapse is the most encouraging trend here, and it is also why the benefit gap is a lagging indicator: the women collecting checks today built their records in the 1960s and 1970s.
The ten-year rule, and the two-year rule nobody mentions
A divorced woman can claim on her ex-husband’s record. The Social Security Handbook sets the conditions out in section 311.
She must have been married to him for at least ten years before the divorce became final. She must be 62 or older, and currently unmarried. Her own retirement benefit must be less than half his primary insurance amount, since what she receives is a top-up to her own rather than a separate payment. And if he has not yet claimed, she must have been divorced from him for at least two continuous years.
One thing about this rule is routinely misunderstood: claiming takes nothing from him. SSA’s operating manual, section RS 00615.682, says a divorced spouse “is never reduced for the maximum,” and that benefits for his current wife and children are calculated without counting the divorced beneficiary at all. He is not notified and cannot object.
The ten-year line is brutal in its precision. A marriage that ends at nine years and eight months produces nothing. Divorce lawyers know this. Many women approaching a decade of marriage do not.
Survivor rules run on a parallel track. A surviving ex-spouse also needs ten years of marriage, must be 60 or older, and must not have remarried before 60. Remarriage at 60 or later is disregarded entirely, which is useful to know and almost never said out loud. A surviving spouse can receive up to 100 percent of the deceased worker’s benefit at full retirement age.
That last rule contains the widow’s trap. When a husband dies, the household keeps the larger of the two checks and loses the smaller one outright. Rent does not fall by the same proportion. In SSA’s tabulation of income data for people 65 and older in 2014, poverty ran 16.3 percent for widowed women against 9.2 percent for widowed men, and 18.4 percent for divorced women against 12.8 percent for divorced men. Married women and married men both sat near 5 percent.
Under the Supplemental Poverty Measure, averaged over 2020 to 2022, 12.4 percent of women 65 and older were poor against 10.2 percent of men, and the rate for everyone 80 and older rose to 14.3 percent. Median income in 2022 was $37,430 for older men and $24,630 for older women.
The caregiving number that keeps getting repeated
If you have read anything about women and retirement, you have seen the figure $324,044, presented as what caregiving costs a woman over her lifetime in lost wages, Social Security and pension.
It comes from a single 2011 report by the MetLife Mature Market Institute, built on 1,112 respondents aged 50 and older in the 2008 Health and Retirement Study: $142,693 in lost wages, $131,351 in lost Social Security, about $50,000 in pension. The authors were candid that the pension piece was an assumption rather than a measurement, and that the rest borrowed behavioral assumptions from other published work about reduced hours and early exit from the labor force.
MetLife shut the Mature Market Institute down in 2013. The number has not been updated in fifteen years, it was never peer reviewed, and it is now quoted in 2026 dollars as though somebody had just measured it. Treat it as a rough order of magnitude from one insurer’s think tank, and stop citing it as a fact.
The peer-reviewed work is less dramatic and more useful. Courtney Van Houtven, Norma Coe and Meghan Skira, using Health and Retirement Study panel data from 1992 to 2008, found that women caring for a parent took a wage penalty of about 3 percent, roughly 37 cents an hour. Women who kept caregiving across four or more survey waves took a 14 percent hit, about $3,000 a year. Male caregivers showed no wage penalty at all.
Their finding on Social Security should change the conversation: the effect on future benefits was minimal, because only 2.4 percent of caregivers would have qualified for a higher benefit had they not been caring for someone. Note the scope, though. That study covered people aged 50 to 64 caring for parents. It says nothing about the earlier, longer gaps taken to raise children, which is where most of the zeros in a woman’s 35 years come from.
Women are not the problem investors
The cliché says women are too cautious with money and that is why they end up with less. The data does not support it.
Vanguard’s How America Saves 2026, covering the 2025 plan year, reports deferral rates by sex and income band. Women contribute a higher percentage of pay than men in every single band: 6.2 against 6.0 percent below $15,000, 6.9 against 6.7 at $50,000 to $74,999, 9.4 against 8.8 at $100,000 to $149,999. Men still finish with about 30 percent more, an average balance of $194,597 against $146,476. Among participants earning $30,000 to $149,999, though, women’s balances came within 10 percent of men’s. The gap is a salary gap wearing a savings costume.
On trading, the evidence runs the other way from the stereotype. Brad Barber and Terrance Odean’s 2001 paper in the Quarterly Journal of Economics, using records from more than 35,000 households at a discount brokerage between 1991 and 1997, found men traded 45 percent more than women, and that trading cut men’s net returns by 2.65 percentage points a year against 1.72 for women. Single men traded 67 percent more than single women and trailed them by 1.44 points a year. Fidelity’s October 2021 analysis of 5.2 million self-directed retail accounts over the decade to December 2020 found women’s annualized returns ran 40 basis points ahead of men’s.
Women save a larger share of a smaller paycheck and trade less. Both are correct behavior. The shortfall is upstream of the brokerage account.
Upstream looks like this. In 2023, women working full time had median weekly earnings of $1,005 against $1,202 for men, 84 percent. Among workers aged 25 to 34 the ratio was 89 percent. By 55 to 64 it had fallen to 77 percent, and that is the band that sets the top of a 35-year earnings record.
What is getting better, and what a fix would cost
The gap is narrowing on several measures, and it is dishonest to pretend otherwise. Howard Iams, in the Social Security Bulletin in 2016, projected that 62 percent of wives born 1937 to 1945 would claim an initial benefit based only on their own earnings, rising to 82 percent for those born 1956 to 1965.
Caregiver credits are the fix most often proposed, and they come in two designs. One lets a parent of a child under six drop up to five extra years from the calculation, shrinking the divisor from 35 to 30. The other credits those years with earnings equal to half the national average wage. The Center for Retirement Research priced them in August 2018 at roughly 0.05 percent and 0.22 percent of taxable payroll over 75 years, and proposed paying for them by cutting the top benefit factor for high earners from 15 percent to 13 or 8 percent. Somebody has to say that out loud: this is a transfer from higher-earning retirees.
Targeting is a live objection too. Earlier research found dropout years disproportionately helped higher-status women who had chosen to work fewer years. SSA’s own projection of the credit design, run in February 2024, is more encouraging: by 2050 it would raise benefits for 48 percent of people in the lowest income quintile, and for 31 percent of women against 21 percent of men. Then there is the plumbing. American birth records are not linked to parents’ Social Security files, and that connection would have to be built first.
Facts worth carrying, from IRS Notice IR-2025-111 of 13 November 2025: for 2026 the 401(k) elective deferral limit is $24,500, the catch-up for workers 50 and over is $8,000, and for ages 60 through 63 it is $11,250. The IRA limit is $7,500, with a $1,100 catch-up.
The longevity figure sits underneath all of it and moves slowly. In 2024, life expectancy at 65 was 20.8 years for women and 18.4 for men, according to the National Center for Health Statistics, which published the estimate in January 2026. Women reach the line with a smaller pot and 2.4 more years to spend it.
Teona Abuladze is the founder and editor of Shansi Magazine.
Sources: Social Security Administration, “Annual Statistical Supplement, 2025 — Highlights and Trends” — https://www.ssa.gov/policy/docs/statcomps/supplement/2025/highlights.html; Social Security Administration, “Fast Facts & Figures About Social Security, 2025” — https://www.ssa.gov/policy/docs/chartbooks/fast_facts/2025/fast_facts25.html; Social Security Administration, “Benefit Formula Bend Points” and “Your Retirement Benefit: How It Is Figured,” 2026 — https://www.ssa.gov/oact/cola/piaformula.html and https://www.ssa.gov/oact/cola/Benefits.html; Social Security Handbook §311, “Divorced Spouse’s Benefits” — https://www.ssa.gov/OP_Home/handbook/handbook.03/handbook-0311.html; Social Security Administration, POMS RS 00615.682, “Divorced Spouse Benefits and the Family Maximum” — https://secure.ssa.gov/poms.nsf/lnx/0300615682; Social Security Administration, POMS RS 00207.003, “Effect of Remarriage on Widow(er)’s Benefits” — https://secure.ssa.gov/poms.nsf/lnx/0300207003; Social Security Administration, “Who can get Survivor benefits” and “What you could get from Survivor benefits,” 2026 — https://www.ssa.gov/survivor/eligibility and https://www.ssa.gov/survivor/amount; Social Security Administration, “Distribution of Zero-Earning Years by Gender, Birth Cohort, and Level of Lifetime Earnings,” Research Note No. 2000-02, November 2000 — https://www.ssa.gov/policy/docs/rsnotes/rsn2000-02.html; Social Security Administration, “Population Profile: Marital Status & Poverty” (2014 data) — https://www.ssa.gov/policy/docs/population-profiles/marital-status-poverty.html; Social Security Administration, “Projected effects of a proposal to credit earnings to caregivers’ records,” run February 2024 — https://www.ssa.gov/policy/docs/projections/policy-options/credit-for-caregivers.html; Howard M. Iams, “Married Women’s Projected Retirement Benefits: An Update,” Social Security Bulletin 76(2), 2016 — https://www.ssa.gov/policy/docs/ssb/v76n2/v76n2p17.html; KFF, “How Many Older Adults Live in Poverty?” (Supplemental Poverty Measure, 2020–2022 average) — https://www.kff.org/medicare/how-many-older-adults-live-in-poverty/; Administration for Community Living, “2023 Profile of Older Americans” (2022 income and poverty data) — https://acl.gov/sites/default/files/Profile%20of%20OA/ACL_ProfileOlderAmericans2023_508.pdf; National Center for Health Statistics, “Mortality in the United States, 2024,” NCHS Data Brief No. 548, January 2026 — https://www.cdc.gov/nchs/products/databriefs/db548.htm; U.S. Bureau of Labor Statistics, “Highlights of Women’s Earnings in 2023” — https://www.bls.gov/opub/reports/womens-earnings/2023/; MetLife Mature Market Institute, National Alliance for Caregiving and New York Medical College, “The MetLife Study of Caregiving Costs to Working Caregivers,” June 2011 — https://www.caregiving.org/wp-content/uploads/2011/06/mmi-caregiving-costs-working-caregivers.pdf; Courtney Harold Van Houtven, Norma B. Coe and Meghan M. Skira, “Effect of Informal Care on Work, Wages, and Wealth,” Center for Retirement Research at Boston College Working Paper 2010-23, and “The effect of informal care on work and wages,” Journal of Health Economics 32(1), 2013 — https://crr.bc.edu/wp-content/uploads/2010/12/wp_2010-23-508.pdf; Vanguard, “How America Saves 2026” (2025 plan-year data) — https://workplace.vanguard.com/insights-and-research/report/how-america-saves-2026.html; Brad M. Barber and Terrance Odean, “Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment,” Quarterly Journal of Economics 116(1), 2001 — https://faculty.haas.berkeley.edu/odean/papers/gender/boyswillbeboys.pdf; Fidelity Investments, “2021 Women and Investing Study,” press release 8 October 2021 — https://www.businesswire.com/news/home/20211008005269/en/; Center for Retirement Research at Boston College, “Modernizing Social Security: Caregiver Credits,” Issue Brief 18-15, August 2018 — https://crr.bc.edu/wp-content/uploads/2018/08/IB_18-15.pdf; Internal Revenue Service, IR-2025-111, “401(k) limit increases to $24,500 for 2026, IRA limit increases to $7,500,” 13 November 2025 — https://www.irs.gov/newsroom/401k-limit-increases-to-24500-for-2026-ira-limit-increases-to-7500
