The Older Workers and Retirement Chartbook Chapter 2. Retirement

What challenges do workers face in preparing for retirement?

Half of American workers face a sharp drop in living standards at retirement based on a conservative measure that assumes retirees will draw down all their savings and use all their assets in retirement, including the equity in their homes (Munnell, Chen, and Siliciano 2021; Munnell et al. 2020). America’s looming retirement crisis is often attributed to longer life spans and workers’ failure to save and invest optimally, but workers in other countries face similar longevity and financial planning challenges yet achieve better retirement outcomes (Mitchell and Lusardi 2015; OECD 2021a, 2021b).

The Older Workers and Retirement Chartbook

A joint project of EPI and the Schwartz Center for Economic Policy Analysis

The U.S. retirement system is sometimes described as a three-legged stool made up of mandatory Social Security, voluntary employer pensions, and private savings. The three-legged stool simile is offered by way of explaining Social Security’s modest benefits, but the other legs are so wobbly that the value of Social Security benefits far exceeds that of other retirement savings and benefits for at least half of older Americans (Sabelhaus and Volz 2020, Figure 4). Because many workers are not covered by a workplace plan, they may continue working into old age to make ends meet, including some who are already receiving Social Security benefits. Among seniors age 65 and older, one-fourth of whom are still working, four in 10 rely on Social Security payments for at least half their income (Dushi and Trenkamp 2021).

Jump to the charts

The Organisation for Economic Co-operation and Development (OECD) estimates that U.S. workers with average earnings will replace 39.2% of their earnings with Social Security benefits if they work steadily from age 22 to 67 (67 is Social Security’s normal retirement age for workers born after 1959). By contrast, the average replacement rate from mandatory pensions is 51.8% in OECD countries, based on an average OECD normal retirement age of 66.1 years (OECD 2021a, Chapter 4). If U.S. workers also participate in a 401(k) plan with a 9% contribution rate from age 35 to 67, the OECD estimates that they will replace 65.7% of their earnings in retirement, but such steady participation is unrealistic for the many workers who lack access to a workplace plan (OECD 2021a, Chapter 4; OECD 2021c).

Relying on employers to offer retirement benefits has never served U.S. workers well, leaving roughly half of private-sector workers ages 25–64 without coverage; this was true even in the heyday of traditional pensions (Munnell and Quinby 2009). The U.S. retirement system went from bad to worse when private-sector employers began switching from traditional defined benefit pensions to 401(k)-style defined contribution plans in the 1980s and early 1990s (EBSA 2021), shifting most of the cost and all the risk of retirement benefits to workers. Despite making it easier for employers to offer a plan, the 401(k) revolution did not expand participation in retirement plans but simply replaced one type of plan with another.

Employers are not required to contribute anything to a 401(k) plan, though most do. In a typical plan, the employer contributes 3% of pay—the maximum matching contribution—if the worker contributes 6%. However, at least 23% of 401(k) participants are enrolled in plans where the maximum employer contribution is less than 3% of pay, including plans that offer workers no employer contribution at all (authors’ analysis of Brightscope and ICI 2022, Exhibits 1.2 and 1.8).

Higher-paid workers are more likely to have access to 401(k) plans, and those plans tend to be more generous. The median employer contribution for workers in the top fourth of the wage distribution who maximize their employer match is 4%, versus 3% for workers in the bottom three-fourths of the wage distribution (BLS 2020, Table 34). This nominal difference understates the actual difference in employer contributions going to highly paid and lower-paid workers because it does not account for differences in participation rates, which range from 22% of workers in the bottom fourth of the wage distribution to 59% of workers in the top fourth of the distribution, according to the same employer survey (which includes workers who do not have access to a plan).

Most of the difference in participation rates reflects a difference in access—that is, whether an employer offers a plan and whether a worker meets age and tenure criteria for eligibility, which tend to be stricter for low-wage workers (BLS 2020, Table 30). Low-wage workers are also more likely to have high-turnover and part-time jobs that make them ineligible for benefits, sometimes the result of deliberate strategies adopted by employers to minimize benefit costs for low-wage workers without running afoul of rules intended to prevent employer discrimination against these workers (Kristal, Cohen, and Navot 2018; Kolhatkar 2022). By any measure, most workers with access to a plan choose to participate, though it is difficult to reliably estimate take-up rates in voluntary defined contribution plans because access and participation rates vary across surveys (Radpour, Papadopoulos, and Ghilarducci 2021).

Low-wage workers not only receive less help from employers, they also receive less help from the government. The federal tax subsidy for retirement plans, ostensibly designed to encourage saving, is a function of the income tax that would otherwise be owed on investment earnings, and many low-wage workers owe little or no income tax (though they face higher payroll tax rates than high earners and, in most states, higher state and local tax rates) (Marr, Frentz, and Huang 2013; Wiehe et al. 2018). The Tax Policy Center estimates that only 3.8% of taxpayers in the bottom fifth of the income distribution receive a tax benefit from participating in a retirement plan, and the benefit they receive is only 0.3% of the total, most of which goes to taxpayers in the top income fifth (Tax Policy Center 2020).

Traditional defined benefit pensions are more efficient than 401(k)-style defined contribution plans because they can average costs across workers with different life spans and who retire during a variety of market conditions, including both bull and bear markets. Because pension funds engage in risk pooling and earn higher net investment returns, contributions to 401(k) plans need to be almost twice as large as contributions to defined benefit pensions to provide similar retirement security (Morrissey 2009; Morrissey 2019; Rhee and Fornia 2014). In practice, contributions to retirement plans have not increased to make up for the risks associated with 401(k)s, even as retirement wealth has become more skewed toward wealthy households (Munnell, Aubry, and Crawford 2015; authors’ analysis of Federal Reserve 2022). Thus the 401(k) revolution has not led to an increase in funds set aside for retirement but instead has made retirement more precarious and unequal.

Large disparities in retirement preparedness are the result of both random factors, such as stock market gyrations that affect 401(k) balances, and systemic failures, such as upside-down tax incentives that provide the most help to those who need it least. The lifelong hardships faced by low-wage workers—including many Black, Hispanic, disabled, LGBTQ, and women workers—continue into old age because of an employer-based retirement system that is stacked against low-wage workers. As a result, Black and Hispanic seniors ages 65 and older have more than double the poverty rates of white seniors in the same age group (Li and Dalaker 2021), Nearly half of seniors with disabilities live in or near poverty and are unable to afford basic necessities (United for ALICE 2022); older women, especially women of color, have higher poverty rates than men; and widowed, divorced, and never-married women are at even greater risk of old-age poverty than married women (Justice in Aging 2020; National Women’s Law Center and Justice in Aging 2021; Entmacher and Matsui 2013; Copeland 2022). Though fewer data are available on older LGBTQ adults, they also appear to have higher poverty rates than non-LGBTQ older adults, though the difference is not statistically significant owing to a small sample size (Badgett, Choi, and Wilson 2019, Table 6).

Black workers are more likely to work in industries and occupations that have low rates of access to retirement plans (Rhee 2021). However, an exception to the rule that Black workers and other disadvantaged groups tend to have jobs lacking retirement benefits is Black workers’ overrepresentation in public-sector jobs, which typically pay less but offer secure defined benefit pensions (Madowitz, Price, and Weller 2020; Thompson and Volz 2021). Among workers ages 45–65 in 2001–2019, 26% of Black workers participated in defined benefit pensions, compared with 24% of white workers and 13% of Hispanic workers. However, because older Black workers were less likely to participate in defined contribution plans than older white workers, their overall participation in retirement plans was somewhat below that of older white workers, 56% versus 59% (Sabelhaus and Thompson 2021, Table 10).

Among households headed by someone ages 40–59, defined benefit pensions accounted for 40% of Black households’ total wealth in 2019, when assessed using a measure of wealth that includes the estimated value of future pension and Social Security benefits. Including Social Security benefits (25%) and savings in defined contribution plans (12%), retirement wealth accounted for 76% of Black households’ total wealth. In contrast, retirement wealth accounted for 50% of white households’ total wealth, of which defined benefit pensions accounted for 20%, Social Security benefits for 16%, and defined contribution plan savings for 14% (authors’ analysis of Thompson and Volz 2021, Table 3). Because the wealth distribution is highly unequal, these breakdowns of Black and white wealth holdings do not reflect the wealth holding of a typical (median) Black or white household, for which the largest share of wealth is in the form of Social Security payments (Thompson and Volz 2021, Table 4). Though not included in measures of retirement wealth, Black retirees are also more likely to receive financial support from family or friends than white retirees (Copeland and Greenwald 2021).

The Black–white gap in retirement plan participation is exacerbated by Black workers having fewer years of employment, a problem linked to employment discrimination, health disparities, caregiving responsibilities, and incarceration. The unemployment rate of older Black workers ages 55–64 is typically 1.5 to 2 times that of older white workers (Wilson and Darity 2022). Black Americans are nearly five times as likely to be incarcerated as white Americans, with dire consequences for future employment and retirement outcomes (Schmitt and Kandra 2021; Chiteji 2020). In addition to higher rates of job loss and incarceration, Black workers are more likely than white workers to leave the workforce for health reasons or to care for family members, including grandchildren (Copeland and Greenwald 2021; Johnson 2020; Lahey 2018). Among those ages 45–65, 32% of Black and 23% of white adults were not working when surveyed in 2001–2019 (Sabelhaus and Thompson 2021, Table 10).

Hispanic workers are less likely to participate in retirement plans than either white or Black workers, a finding that holds across age and educational attainment groups. For example, Hispanic workers ages 45–65 were less likely to participate in retirement plans (35%) than their white (59%) or Black (56%) counterparts in 2001–2019. Unlike older Black workers, older Hispanic workers also had lower defined benefit pension coverage (13%) than older white workers (24%) (Sabelhaus and Thompson 2021, Table 10).

Despite low rates of retirement plan participation, retirement wealth accounted for 68% of Hispanics’ total wealth among households headed by someone age 40–59 in 2019, of which defined benefit pension benefits accounted for 30%, Social Security benefits for 28%, and defined contribution plan assets for 10%. This finding reflects low levels of nonretirement wealth rather than high levels of retirement wealth. As in the case of Black and white households, averages are shaped by the wealth holdings of better-off households. The typical (median) Hispanic household has neither a defined benefit pension nor a defined contribution plan and its most important asset is in the form of Social Security benefits (authors’ analysis of Thompson and Volz 2021, Tables 3 and 4).

Hispanic workers cluster in low-wage occupations that offer few benefits (Rhee 2021). Lower educational attainment, poor health, language barriers, and other known factors explain some but not all of their low pay and low retirement plan participation. Foreign-born Hispanics are generally lower-paid and less likely to participate in retirement plans than Hispanics born in the United States. Foreign-born Hispanic men have higher employment rates than white, Black, or U.S.-born Hispanic men, but foreign-born Hispanic women have lower employment rates than their white, Black, or U.S.-born Hispanic counterparts. Immigrants who came to the United States later in life may not qualify for Social Security or other retirement benefits, a problem also faced by those who lack legal authorization to work (Johnson 2016).

One explanation for Hispanics’ low levels of participation in retirement plans may be their greater reliance on family support (Richman et el. 2012). However, this may be as much a response to as a cause of limited access to retirement plans. Hispanic seniors age 65 and older are more likely to live in multigenerational households (44%) than are Black (33%) or white (16%) seniors (Johnson 2016). Multigenerational living arrangements help families economize on living expenses and allow grandparents to assist with child care and adult children to assist frail parents and grandparents. Hispanic retirees are also more likely to receive financial support from family or friends than white retirees (Copeland and Greenwald 2021).

Blacks and Hispanics also face a greater risk of developing disabling health conditions as they age (Goodman, Morris, and Boston 2019; Weaver 2020). Adults with disabilities, regardless of race and ethnicity, face significant challenges in the workforce. They are less than half as likely to be employed as nondisabled adults, even though most do not meet the overly stringent eligibility standards for receiving Social Security Disability Insurance (SSDI) benefits. By one estimate, workers with disabilities earn 74 cents for every dollar earned by their nondisabled peers (Vallas et al. 2022). Poor employment prospects and health-related expenses lead to high poverty rates and other measures of financial hardship, such as high rates of food and housing insecurity, difficulty paying medical bills, and forgoing needed medical care (Goodman, Morris, and Boston 2019; Weaver 2020; Vallas et al 2022; CBPP 2021).

For adults with disabilities, employment challenges are exacerbated by significant barriers to accessing benefits. This is the case even for disabled workers who should qualify based on stringent eligibility standards, a problem aggravated in recent years by inadequate and poorly allocated administrative funding (Weaver 2021). Half or fewer applicants for Social Security Disability Insurance (SSDI) benefits are ultimately accepted, and many are accepted only after a lengthy appeals process (CBPP 2021; Weaver 2020). Even a study cited by supporters of tightening eligibility requirements found that 72% of SSDI applicants who were denied benefits based on their supposed capacity for gainful employment were earning little or nothing two years later (Maestas, Mullen, and Strand 2013; Morrissey 2015).

Disabled adults’ low employment rate and low earnings affect their retirement savings and benefits, leading to financial hardship in old age. Employment problems are compounded by the slow and difficult process of obtaining disability benefits, as many workers who should be eligible for full SSDI benefits instead apply for reduced Social Security retirement benefits at age 62, before they reach full retirement age, often unaware that they may qualify for SSDI benefits even if they have already applied for retirement benefits. Accepting reduced Social Security retirement benefits rather than full disability benefits also means these workers will not qualify for Medicare until age 65, whereas SSDI recipients qualify for Medicare two years after receiving SSDI (Cloyd 2021).

All these problems are made worse by woefully inadequate Supplemental Security Income (SSI) benefits for low-income seniors and people with disabilities. A half century after SSI was signed into law, it does not provide enough income to keep beneficiaries out of poverty. Its income and asset limits are not indexed to inflation, let alone to a modern standard of living (Nuñez 2022). SSI’s meager benefits disproportionately hurt women, who make up half (50%) of disabled adults ages 18–64 and two-thirds (65%) of seniors age 65 and older receiving SSI benefits (SSA 2020).

Though working women have caught up with working men in terms of retirement plan participation, they are still at greater risk of financial hardship in old age because of lower pay, longer life spans, and more years spent out of the paid workforce or working part-time while caring for family members. Women are more likely than men to be single at older ages because they often outlive older husbands. Women are also more financially affected than men by late-in-life divorce. One reason for this discrepancy is that 401(k) savings are not well protected in divorce, unlike traditional defined benefit pensions. Traditional pensions are generally better for women because they provide a secure income, including spousal benefits, until beneficiaries die, and women tend to live longer and rely more heavily on spousal income and benefits than men (Matsui 2021; Stein 2021; Lin and Brown 2021; Entmacher and Matsui 2013).

Women shoulder more caregiving responsibilities than men, a factor in their lower lifetime earnings and greater retirement insecurity. Caring for young children and then for grandchildren or frail adult family members reduces the time women spend in the paid workforce. More than one-third (35%) of adults ages 50–64 provide unpaid personal care to adults who need such help. The majority (60%) of caregivers of adults are women (AARP and National Alliance for Caregiving 2020). Following women in their 50s as they aged over the 1992–2010 period, Fahle and McGarry (2018) found that caring for spouses, parents, and parents-in-law reduced employment by 8% and was associated with 1.3 fewer hours of paid work per week for women remaining in the workforce. Providing financial support to parents and caring for frail spouses are factors correlated with lower retirement wealth for older women, though it is difficult to differentiate between lower-wealth families relying more on unpaid care as opposed to caregiving having an impact on wealth (Bond, Saad-Lessler, and Weller 2020).

Caregiving is also a concern for older LGBTQ adults, who are less likely to have children than non-LGBTQ adults and who often find themselves caring for aging parents, spouses, and partners, but who have fewer younger family members they can rely on when they need help. Though friend networks play an important caregiving role in LGBTQ communities, LGBTQ seniors, especially men, are more likely to age alone, a problem exacerbated by the fact that many LGBTQ seniors do not feel welcome in many senior centers, assisted living and long-term care facilities, and health care settings (SAGE and National Resource Center on LGBT Aging n.d.; MAP and SAGE 2010; Schmidt 2022; Choi and Meyer 2016).

For LGBTQ workers as for other disadvantaged and discriminated-against groups, employment disparities carry over into retirement (Fredriksen-Goldsen et al. 2017). While data on pay and benefits are limited, the evidence suggests that LGBTQ men earn less than non-LGBTQ men, while LGBTQ women, like their non-LGBTQ counterparts, earn less than men (Badgett, Carpenter, and Sansone 2021). Though the Social Security Administration and some employers have taken steps to provide access to retirement benefits for same-sex partners, spouses, and survivors, including those who would have qualified for spousal benefits in the absence of prohibitions against same-sex marriage before the Supreme Court’s 2015 Obergefell v. Hodges decision, these changes came too late for many LGBTQ couples (SSA n.d.; Choi and Meyer 2016). Such factors, along with health disparities, contribute to higher indebtedness and lower retirement confidence for LGBTQ adults compared with non-LGBTQ adults (Badgett, Choi and Wilson 2019; Emlet 2016; Copeland and Greenwald 2022).

Point-in-time measures give an incomplete picture of retirement security, which is not just a function of the retirement savings and benefits an individual manages to accumulate while working and whether these funds will be sufficient to cover normal living expenses in retirement. Retirement security is also influenced by past disparities, such as inherited wealth and educational advantages or lack thereof, and future responsibilities and risks, such as the need to support family members or cover health-related expenses.

The U.S. tax system plays a significant role in augmenting disparities in retirement security. While lower-income workers are poorly served by a do-it-yourself retirement system, wealthy households benefit disproportionately from federal subsidies for retirement saving. This is not just because they are more likely to have access to employer-based retirement accounts and their account balances are larger but also because they can maximize the tax advantage by allocating riskier assets with higher expected rates of return to retirement accounts that allow investment income to grow tax-free (Mitchell 2022; Wamhoff 2021; Morrissey 2022).

Thus, the employer-based retirement system serves to magnify inequality, including racial and ethnic gaps in retirement wealth. Because Black and Hispanic households have historically been shut out of many housing markets and other asset-building opportunities (Rothstein 2017), they are disadvantaged by a system that rewards risk taking rather than incentivize new saving. Black and Hispanic households are often exposed to financial risk not by choice but by circumstances. They were disproportionately affected by plummeting housing prices during the Great Recession of 2008–2009, for example (Wolff 2018).

Financial burdens also weigh more heavily on Black and Hispanic families. For example, Blacks and Hispanics are more likely than whites to cite paying off debt or paying for a child’s education as barriers to saving. They are also more likely than whites to prioritize supporting family and friends over saving for retirement (Copeland and Greenwald 2021), a difference likely based on Black and Hispanic families’ greater immediate needs and reliance on reciprocity rather than on complacency about the future.

In contrast to employer-based plans, Social Security’s progressive benefit formula replaces a higher share of earnings for low-wage workers. Social Security benefits therefore serve to reduce wealth inequality, including the racial wealth gap. Researchers at the Center for Retirement Research, for example, estimated that among older households ages 51–56 between 1992 and 2016, the typical Black household had 46% of the retirement wealth of the typical white household, while the typical Hispanic household had 49%, if the future value of Social Security benefits was included in the wealth calculation. Without Social Security, however, these figures would fall to 14% and 20%, respectively (Wou and Sanzenbacher 2020). Wolff (2018) and Thompson and Volz (2021) found similarly wealth-equalizing effects of Social Security and defined benefit pensions for different age groups. However, Social Security’s progressivity has been somewhat eroded by growing disparities in life expectancy between low and high earners that affect the value of lifetime benefits (National Academies 2015; Rutledge 2018).

The shift from secure pensions to risky and inadequate 401(k) plans unfortunately coincided with cuts to Social Security enacted in 1983, when the system faced an imminent shortfall. These cuts took the form of a gradual increase in the normal retirement age from 65 to 67 and the taxation of some Social Security benefits, with the tax revenue reverting to the system. The increase in the normal retirement age is equivalent to a 13.3% across-the-board cut in benefits for workers born in 1960 or later, with smaller cuts for workers born between 1938 and 1959. Factoring in the taxation of benefits, GenXers will experience cuts averaging around 22% relative to what they would have received absent the 1983 reforms. Later generations will experience even deeper cuts as an increasing share of benefits is taxed (Morrissey 2019; Purcell 2015).

Meanwhile, longer life spans and slower economic growth have made funding retirement more expensive, whether that funding happens through intergenerational transfers, as with Social Security, or depends on investment returns, as with traditional pensions and retirement savings plans. Falling birth rates, increased life expectancy, and reduced immigration have resulted in an older U.S. population and older workforce. Adding to these challenges is the fact that most workers have seen slow or stagnant wage growth as their wages have fallen behind productivity growth (Goss 2022; American Academy of Actuaries 2020; EPI 2022). As corporate profits and high earners have captured a growing slice of the economic pie, this redistribution away from low- and middle-income families has hurt consumer demand and slowed growth while hindering many workers’ ability to save for retirement (Bivens and Banerjee 2022).

What do the charts tell us?

Only half of prime-age (25–54) workers participate in an employer-based retirement plan, a share that drops to one-third for workers ages 65 and older (Chart 2A). The main reason so few workers participate in an employer-based retirement plan is that employers are not required to offer one, and those that do can limit eligibility to workers meeting minimum age, tenure, and hours worked requirements. Most workers who do have access to a plan either are enrolled automatically, as in the case of traditional pensions, or choose to participate, as in voluntary 401(k) plans (Chart 2B).

Participation might be expected to increase with age as workers pay off student loans, settle into career jobs, and approach retirement. This does not happen, however. Retirement plan participation is little higher for workers ages 55–64 than for prime-age workers, and it is significantly lower for workers age 65 and older (Chart 2A). Participation is especially low for older workers without four-year college degrees, women age 65 and older, and older Hispanic workers (Charts 2C–2E).

It is normal for high earners to save more in retirement accounts than low earners when the goal is to replace a target share of earnings at retirement. But there is no good reason for retirement plan participation to depend on earnings. Among workers ages 55–64, however, only 1 in 5 in the bottom income quintile participates in a retirement plan, whereas four in five in the top income quintile participate. The pattern is similar—but participation rates are even lower—for workers age 65 and older (Chart 2F).

Participating in a retirement plan does not guarantee that workers will be able to maintain their standard of living or avoid hardship in retirement. However, workers with traditional defined benefit plans are generally better prepared for retirement than those with 401(k)-style defined contribution plans because in defined benefit plans, employers are responsible for funding a guaranteed benefit in the form of a monthly pension check. Older Black workers are more likely than older white or Hispanic workers to participate in these more secure plans because they are more likely to work in the public sector, where defined benefit pensions are more common. However, older Black workers are less likely to participate in defined contribution plans—or in any type of plan—than white workers. Hispanic workers are less likely than either Black or white workers to participate in either type of plan (Chart 2G).

Another gauge of retirement preparedness is whether a household has savings in a tax-favored retirement plan from a current job, a past job, or an individual retirement account (IRA). We would expect most households approaching retirement to have savings in one of these accounts, since participation in 401(k)-style defined contribution plans has exceeded participation in defined benefit pensions since 1992, and most IRA funds were rolled over from 401(k) plans (EBSA 2021; Copeland 2020). However, households ages 55–64 are no more likely to have retirement account savings than those ages 25–54. Among households with retirement accounts, older households do have more saved than their younger counterparts, but not much more than the equivalent of a year’s income (Chart 2H).

Among households age 55 and older, only college-educated, married, and non-Hispanic white households have substantial savings in retirement accounts (Charts 2I-2K), and they are a minority of older households.

What can we do to improve retirement security?

For some workers, delaying retirement can help close the retirement income gap by allowing more time to increasing savings and accrued benefits while shortening the amount of time spent in retirement itself, when assets are spent down. Deferring retirement can be especially beneficial for women whose careers have been delayed or interrupted by childrearing (Maestas 2018). However, expecting workers to work into old age is neither a feasible nor an equitable solution to the retirement crisis because the increase in life expectancy has been concentrated among higher earners with less physically demanding and onerous jobs. Americans already work more hours per year and more years than workers in most peer countries (OECD 2021d, 2022).

Though retirement challenges appear complex, the solution is simple, if politically challenging: Expand Social Security, SSI, and other social insurance programs. Social Security expansion should include both across-the-board and targeted benefit increases, including a caregiver credit that boosts benefits for workers whose caregiving responsibilities cause them to earn less than the median worker, and a higher special minimum benefit for workers who earn low wages over long careers. These are among the common elements of Social Security expansion plans proposed in recent years by Rep. John Larson (D-Conn.), Sen. Bernie Sanders (D-Vt.), Sen. Elizabeth Warren (D-Mass.), and others that have garnered strong support in the Democratic caucus. However, Social Security expansion has encountered resistance from Republican lawmakers despite support from Republican voters (Morrissey 2019; Data for Progress 2022; Altman 2022).

Older workers with caregiving responsibilities would have been helped by expanded funding for Home and Community Based Services in Medicaid, a major component of the Biden administration’s Build Back Better plan that, unfortunately, did not make it into the Inflation Reduction Act of 2022 (Justice in Aging 2020; Coalition on Human Needs 2022) The Supplemental Security Income (SSI) Restoration Act introduced by Sen. Sherrod Brown (D-Ohio) would also offer critical support to many vulnerable older Americans, including those with disabilities, by upgrading the financial eligibility rules and bringing them in line with current economic realities (Romig 2021).

Expanding Social Security would reduce the reliance on employers to voluntarily provide benefits, an aspect of the U.S. retirement system that has proven to be a failure. A complementary approach would be to require employers to offer pension benefits or contribute to a secure low-cost retirement plan, such as the Guaranteed Retirement Account plan (Ghilarducci 2008). The two approaches could be combined. For example, employers could be required to contribute to retirement savings accounts, which would enable workers to defer filing for Social Security benefits and so realize a larger monthly benefit when they did file (Koenig, Fichtner, and Gale 2018).

More incremental steps to expand access to retirement plans, lower costs, and promote fairness include automatic IRA plans set up by state and local governments and refundable tax credits for retirement saving (Pew 2016; Fisher and Ghilarducci 2017; Georgetown CRI 2020; Georgetown CRI 2021; Batchelder, Goldberg, and Orszag 2006). It is also important to defend traditional pensions, which work for those workers lucky enough to have them, including public-sector workers, whose employers are well suited to take on long-term liabilities (Morrissey 2019).


Retirement plan participation remains low despite 401(k) revolution: Share of workers who participate in a retirement plan at a current job, by age, 1992–2019

2A
Year Ages 25–54 Ages 55–64 Age 65+
1992 54.6% 61.4% 29.4%
1995 55.5% 51.6% 33.5%
1998 54.7% 57.6% 37.9%
2001 56.1% 55.7% 21.0%
2004 54.1% 63.8% 32.6%
2007 53.1% 60.2% 40.4%
2010 52.5% 57.8% 33.2%
2013 50.1% 57.2% 39.5%
2016 51.8% 60.5% 42.2%
2019 53.4% 57.2% 36.9%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. 

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of 1992 to 2019 Survey of Consumer Finances microdata (Federal Reserve 2022).

Roughly half of prime-age workers (ages 25–54) and older workers approaching retirement (ages 55–64) participate in a retirement plan, such as a traditional defined benefit pension plan or 401(k)-style defined contribution plan. This share has not changed noticeably in recent decades despite a change in the tax code in the late 1970s that set the stage for 401(k) plans. Under a defined contribution plan such as a 401(k), employee and/or employer contributions go into an investment account. These plans shift much of the cost and all of the risk of retirement onto workers (Morrissey 2019). Defined contribution plans overtook defined benefit pension plans in 1992, when the number of participants in defined contribution plans exceeded the number of participants in defined benefit pension plans (DOL 2021). Yet as this chart shows, making it easier and cheaper for employers to offer retirement plans boosted participation in these easier and cheaper defined contribution plans but did not expand overall participation in retirement plans.

The chart also shows that retirement plan participation is much lower for workers age 65 and older. Some of these workers describe themselves as retired from career jobs and have part-time or temporary jobs with fewer benefits. Participation in workplace retirement plans has increased somewhat among these workers in recent decades as more are retiring at older ages from jobs with benefits. One factor contributing to later retirement is the shift to 401(k) plans. Unlike traditional pension plans, 401(k) plans do not specify a “normal” retirement age and provide less secure retirement income (Munnell 2015).

Share this chart:

Copy the code below to embed this chart on your website.

Only about half of workers approaching retirement age participate in a retirement plan, largely owing to lack of access: Share of workers who have access to and participate in a retirement plan at a current job, by age, 2019

2B
Category Yes No
Access 63.0% 37.0%
Participation 53.4% 46.6%
Access 66.5% 33.5%
Participation 57.2% 42.8%
Access 46.3% 53.7%
Participation 36.9% 63.1%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. 

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Workers with access to a plan are workers who answered “yes” to questions about whether their employer offered any such plans and whether they were eligible to be included in any of those plans and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of 2019 Federal Reserve Board of Governors Survey of Consumer Finances microdata (Federal Reserve 2022).

Only 57.2% of older workers approaching retirement (ages 55–64) participate in a retirement plan. This is only slightly more than the 53.4% of prime-age workers (ages 25–54) who participate. These data call into question the common assumption that workers catch up on retirement preparation as they get older and settle into career jobs. 

This chart shows that lack of access is the biggest factor depressing worker participation in retirement plans. In most cases, workers who do not participate lack access to a plan, either because their employer does not sponsor one or because they do not meet eligibility requirements based on hours worked or tenure.

Retirement plan access and participation are much lower among workers age 65 and older. Some workers in this age group are semiretired and continue working for noneconomic reasons. But others need to work longer and save more to make up for insufficient retirement savings, often because of a lifetime of lower-income work. Lack of access to retirement plans at work makes saving for retirement more difficult. This is true even for workers who can stay employed and postpone their retirement (see Ghilarducci, Papadopoulos, and Webb 2022).

Share this chart:

Copy the code below to embed this chart on your website.

Older workers without a college degree are less likely to have access to and participate in a retirement plan: Share of older workers who have access to and participate in a retirement plan at a current job, by educational attainment and age, 2019

2C
Yes No
Access rate 61.6% 38.4%
Participation rate 48.6% 51.4%
Access rate 73.9% 26.1%
Participation rate 69.9% 30.1%
Access rate 42.1% 57.9%
Participation rate 27.4% 72.6%
Access rate 51.6% 48.4%
Participation rate 48.8% 51.2%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s.

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Workers with access to a plan are workers who answered “yes” to questions about whether their employer offered any such plans and whether they were eligible to be included in any of those plans and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of 2019 Survey of Consumer Finances microdata (Federal Reserve 2022).

Older workers without a bachelor’s degree or more education are much less likely than their college-educated peers to have access to and participate in a retirement plan. For example, seven in 10 (69.9%) of workers ages 55–64 with a bachelor’s degree participate in a plan, compared with less than half (48.6%) of their counterparts without a bachelor’s degree. Workers without a bachelor’s degree represent well over half of workers (59.7%) in this age group (not shown in the chart). (The distribution of workers by education level also comes from Survey of Consumer Finances microdata accessed from the Federal Reserve 2022).

Not having a bachelor’s degree is also associated with lower retirement access and participation rates among workers age 65 and older. Among workers in this age group, 48.8% of workers with a bachelor’s degree participate in a plan, nearly twice the participation rate of workers without a bachelor’s degree (27.4%). Though not shown in the chart, workers with a bachelor’s degree make up a larger share of workers age 65 and older (44.5%) than workers ages 55–64 (40.3%). The gap stems from the fact that college-educated workers are more likely to remain in the workforce at older ages, according to EPI and SCEPA analysis of SCF microdata (Federal Reserve 2022). However, many workers in the 65+ age group, including those with a bachelor’s degree, work part-time and/or supplement Social Security payments with income from lower-paid jobs taken as a transition to retirement. Often these “bridge” jobs do not provide retirement benefits (Cahill and Quinn 2020; Johnson and Kawachi 2007).

Share this chart:

Copy the code below to embed this chart on your website.

Working women age 65 and older are the least likely among older workers to participate in a retirement plan: Share of older workers who have access to and participate in a retirement plan at a current job, by gender and age, 2019

2D
Yes No
Access rate 65.6% 34.4%
Participation rate 54.7% 45.3%
Access rate 67.5% 32.5%
Participation rate 59.8% 40.2%
Access rate 50.6% 49.4%
Participation rate 41.4% 58.6%
Access rate 41.1% 58.9%
Participation rate 31.5% 68.5%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s.

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Workers with access to a plan are workers who answered “yes” to questions about whether their employer offered any such plans and whether they were eligible to be included in any of those plans and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Despite some progress, gender inequities in retirement persist. Working women approaching retirement (ages 55–64) are as likely to have access to a retirement plan as their male counterparts and are more likely to participate. However, working women age 65 and older are less likely to have access to a plan than working men in their age group or workers ages 55–64 (both men and women).

This limited access to retirement for working women age 65 and older is partly due to their work in part-time jobs, which are less likely to provide retirement benefits (BLS 2019). Because women make up nearly two-thirds of caregivers for adults (typically elder parents or spouses), their caregiving responsibilities often limit them to part-time jobs, which are far less likely to offer any retirement plan (AARP and National Alliance for Caregiving 2020). Even if women had equal access to retirement plans, they would be at greater risk of hardship in old age. That is because they earn less, live longer, and are more likely to be single or widowed than men, among other factors (Enda and Gale 2020; Entmacher, Waid, and Veghte 2016; Ghilarducci, Jaimes, and Webb 2018; Weller, Saad-Lessler, and Bond 2020).

Share this chart:

Copy the code below to embed this chart on your website.

Hispanic workers are the least likely among older workers to participate in a retirement plan: Share of workers age 55 and older who have access to and participate in a retirement plan at a current job, by race and ethnicity, 2019

2E
Yes No
Access rate 63.1% 36.9%
Participation rate 54.4% 45.6%
Access rate 63.8% 36.2%
Participation rate 50.3% 49.7%
Access rate 40.1% 59.9%
Participation rate 30.3% 69.7%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. Hispanic refers to Hispanic of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks.

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. Hispanic refers to Hispanics of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Non-Hispanic Asian Americans and Pacific Islanders and those not included in the other racial groups are not shown in the graph due to small sample sizes. Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Workers with access to a plan are workers who answered “yes” to questions about whether their employer offered any such plans and whether they were eligible to be included in any of those plans and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Older Black workers (age 55 and older) are slightly more likely to have access to a retirement plan than older white workers, though their participation rate is lower. Older Hispanic workers are much less likely than either Black or white workers to have access to or participate in retirement plans. Separate analysis of Survey of Consumer Finances and Current Population Survey data reveals some factors at play behind the differing access rates. Older Black workers are more likely to work full-time, more likely to work in the public sector, and less likely to be self-employed than older white workers, all factors associated with greater access to retirement benefits (Federal Reserve 2022; Flood et al. 2021). On the other hand, older Hispanic workers are much less likely than Black or white workers to have jobs in the public sector.

The lower participation rates for Black and Hispanic workers are due in part to industry characteristics. Black and Hispanic workers are much more likely to work in such sectors as the Accommodation and Food Services sector, where pay is low and benefits are meager (Rhee 2021). A 2022 survey ranked this sector last in the quality of 401(k) plans offered (Godbout 2022). Low-income workers, including many Black and Hispanic workers, also receive little or no tax benefit from participating in a plan because they owe payroll and other taxes but not income tax. Low pay, low employer matches, and low or no tax benefits discourage participation in defined contribution plans that require workers to contribute to the plan and assess a penalty if workers need to access funds before age 59½ (Morrissey 2019).

Share this chart:

Copy the code below to embed this chart on your website.

High-earning older workers are three times as likely as low-earning older workers to have access to a retirement plan: Share of older workers who have access to and participate in a retirement plan at a current job, by earnings group and age, 2019

2F
Access rate Participation rate
Top fifth 82.3% 79.5%
2nd from top 80.6% 73.4%
Middle fifth 75.0% 63.4%
2nd from bottom 56.7% 41.0%
Bottom fifth 27.6% 18.3%
Top fifth 68.1% 67.6%
2nd from top 62.2% 52.3%
Middle fifth 56.4% 47.8%
2nd from bottom 53.6% 41.7%
Bottom fifth 22.6% 11.0%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s.

Retirement plans include traditional defined benefit pension plans and retirement savings accounts such as 401(k)s. The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in a plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Workers with access to a plan are workers who answered “yes” to questions about whether their employer offered any such plans and whether they were eligible to be included in any of those plans and workers whose spouse or partner answered “yes” for them.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

High earners are much more likely than low earners to have access to and participate in retirement plans. Among workers approaching retirement age (ages 55–64), those in the top earnings fifth are three times as likely to have access and more than four times as likely to participate as those in the bottom earnings fifth. 

Workers ages 55 to 64 are more likely to have access and participate than workers age 65 and older. The gap is due in part to the fact that workers age 65 and older are more likely to be working part-time. Among workers age 65 and older, those in the top earnings fifth are three times as likely to have access and more than six times as likely to participate as those in the bottom fifth.

Share this chart:

Copy the code below to embed this chart on your website.

While traditional pension coverage of older Black workers outstrips that of other older workers, retirement coverage for Hispanics lags: Share of older workers who participate in a retirement plan at a current job, by race, ethnicity, and plan type, 2019

2G
Race/ethnicity Yes No
Any plan 54.4% 45.6%
Traditional pensions and other defined benefit  18.4% 81.6%
401(k)s and other defined contribution  46.2% 53.8%
Any plan 50.3% 49.7%
Traditional pensions and other defined benefit  24.8% 75.2%
401(k)s and other defined contribution  39.1% 60.9%
Any plan 30.3% 69.7%
Traditional pensions and other defined benefit  10.4% 89.6%
401(k)s and other defined contribution  27.2% 72.8%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Defined benefit pension plans include traditional pension plans as well as other retirement plans in which employers are responsible for funding promised benefits. Retirement savings accounts such as 401(k)s are referred to as defined contribution plans because employer contributions (if the employer contributes at all), rather than retirement benefits, are determined in advance. Hispanic refers to Hispanic of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks.

Defined benefit pension plans include traditional pension plans as well as other retirement plans in which employers are responsible for funding promised benefits. Retirement savings accounts such as 401(k)s are referred to as defined contribution plans because employer contributions (if the employer contributes at all), rather than retirement benefits, are determined in advance. Hispanic refers to Hispanic of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks.

The sample includes survey respondents and spouses or domestic partners who are employees or self-employed workers (which would include small business owners). Participants in any plan are workers who answered “yes” to a question about whether they were included in any “pension, retirement, or tax-deferred savings plans” connected with their job (not including IRA or Keogh plans) and workers whose spouse or partner answered “yes” for them. Participants who answered yes were then asked to identify which type or types of plans they were included in. Non-Hispanic Asian Americans/Pacific Islanders and those not included in the other groups are not shown in the graph due to small sample sizes. The share of individuals with any type of plan may be less than the sum of the share of individuals with defined benefit pension plans and the share of individuals with defined contribution plans because some individuals have both types of plan.

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Among workers age 55 and older, Hispanic workers are less likely to participate in retirement plans than white and Black workers. Black workers in the 55 and older age group are more likely than white workers to participate in traditional pension plans and other defined benefit plans and less likely to participate in 401(k)-style defined contribution plans. Although not shown in the chart, Black workers are more likely than white or Hispanic workers to be employed in the public sector, where traditional pensions are the norm (BLS 2019). Pensions are especially important to Black workers because discriminatory policies and practices have historically relegated these workers to jobs lacking secure benefits, have prevented Black families from accumulating home equity, and have generally hindered Black Americans from achieving financial security and transferring wealth to younger generations (Rothstein 2017).

Share this chart:

Copy the code below to embed this chart on your website.

Many older households have no savings in retirement accounts, and the typical household with savings has relatively little saved

2H
Share of households with retirement account savings, by age, 2019
Age Has retirement account savings No retirement savings
Ages 25–54 55.1% 44.9%
Ages 55–64 54.5% 45.5%
Age 65+ 43.7% 56.3%
ChartData
Median account balance of households with retirement savings, by age, 2019

Age Conditional median retirement account balance (2019$)
Ages 25–54 $ 44,500
Ages 55–64 $134,000
Age 65+ $125,000
ChartData

Notes: Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs but not in defined benefit pension plans. 

Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs but not in defined benefit pension plans. A household’s age is based on the age of the reference person, who, in households of more than one person, is defined by the Survey of Consumer Finances as the male in a mixed-sex couple or the older person in a same-sex couple. The median household retirement account savings includes reported savings in accounts held by both spouses or partners, in the case of married or partnered couples, and—less commonly—the retirement account savings of other people who are financially interdependent with the “economically dominant” individual or couple in the household (Bhutta et al. 2020).

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Just over half (54.5%) of households approaching retirement age (ages 55–64) have savings in a 401(k)-type account or an IRA. That is about the same share (55.1%) as prime-age households (ages 25–54). Households age 65 and older are less likely to have retirement account savings. But, though not shown in the chart, households age 65 and older are more likely than their counterparts approaching retirement age (ages 55–64) to have traditional pensions or other defined benefit plans from a current or past job. Half (49.7%) of household age 65 and older have defined benefit plans, versus 32.9% of households ages 55–64 (Federal Reserve 2022).

Among households with retirement account savings, account balances are higher for both groups of older households than for prime-age households, as would be expected. However, the median retirement account balance for households ages 55–64 is $134,000, only somewhat higher than their median income of $97,739. (Incomes, not shown in the chart, also come from the Survey of Consumer Finances data.) These figures suggest that even among the relatively privileged households with retirement account savings, account balances are only a fraction of what they would need to supplement Social Security income in retirement. Unless these households also have defined benefit pensions or other sources of income or wealth, many are likely to experience a sharp drop in their standard of living in retirement.

There is no consensus among experts about how much money people need to save in retirement accounts. Needed savings depends, among other things, on what other resources they have, such as Social Security income, home equity, or a traditional pension. However, few experts would suggest that a typical household will be able to maintain its standard of living in retirement with only one to two years’ worth of income saved in a retirement account, unless a household member has a traditional pension in addition to Social Security income.

Share this chart:

Copy the code below to embed this chart on your website.

Only college-educated older workers have significant retirement savings

2I
Share of older households with retirement account savings, by educational attainment and age, 2019

Educational attainment Has retirement account savings No retirement savings
No college degree 43.9% 56.1%
Bachelor’s degree or more 75.5% 24.5%
No college degree 30.6% 69.4%
Bachelor’s degree or more 66.0% 34.0%
ChartData
Median account balances of older households with retirement account savings, by educational attainment and age, 2019

Conditional median retirement account balance (2019$)
No college degree $74,000
Bachelor’s degree or more $295,000
No college degree $70,000
Bachelor’s degree or more $200,000

 

ChartData

Notes: Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans.

Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans. A household’s age and educational attainment are based on the age and education level of the reference person, who, in households of more than one person, is defined by the Survey of Consumer Finances as the male in a mixed-sex couple or the older person in a same-sex couple. The median household retirement account savings includes reported savings in accounts held by both spouses or partners, in the case of married or partnered couples, and—less commonly—the retirement account savings of other people who are financially interdependent with the “economically dominant” individual or couple in the household (Bhutta et al. 2020).

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Among older households, only households with a bachelor’s degree or more education are more likely than not to have retirement account savings. And only these households have a substantial dollar amount of savings in these accounts. These college-educated households are in the minority—constituting 33.4% of households in the 55–64 age group and 36.9% of households in the 65 and older group, according to the 2019 Survey of Consumer Finances microdata (Federal Reserve 2022) (not shown in the chart).

The low shares of non-college-educated households with retirement account savings and the low balances in those accounts suggest deep inequities in retirement preparedness. Roughly eight in 10 non-college-educated households approaching or entering retirement between ages 55 and 64 either have nothing (56.1%) or less than $74,000 (22.0%) saved in these accounts.

Share this chart:

Copy the code below to embed this chart on your website.

Older couples are more likely to have retirement account savings than older single people, especially older single women

2J
Share of older households with retirement account savings, by gender/marital or domestic partnership status, and age, 2019
Gender and marital status Has retirement account savings No retirement savings
Couples 64.4% 35.6%
Single men 36.1% 63.9%
Single women 44.7% 55.3%
Couples 55.8% 44.2%
Single men 27.5% 72.5%
Single women 33.9% 66.1%
ChartData
Median account balances of older households with retirement account savings by gender/marital or domestic partnership status, and age, 2019

Conditional median retirement account balance (2019$)
Couples $197,000
Single men $97,000
Single women $59,000
Couples $174,000
Single men $172,000
Single women $50,000

 

ChartData

Notes: Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans.

Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans. A household’s age is based on the age of the reference person, who, in households of more than one person, is defined by the Survey of Consumer Finances  as the male in a mixed-sex couple or the older person in a same-sex couple. The median household retirement account savings includes reported savings in accounts held by both spouses or partners, in the case of married or partnered couples, and—less commonly—the retirement account savings of other people who are financially interdependent with the “economically dominant” individual or couple in the household (Bhutta et al. 2020).

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

While older single women are more likely than older single men to have savings in a 401(k)-type account or an IRA, couples are more likely than either to have retirement account savings. Older couples and older single men with retirement accounts have more savings than older single women with retirement accounts.

Even among couples approaching retirement (ages 55–64), who are the most likely among all older age groups to have retirement accounts, the median account balance is only one and a half times their median income of $133,373. (Income data, not shown in the chart, also come from the Survey of Consumer Finances.) There is no consensus among experts about how much money people need to save in retirement accounts, which depends, among other things, on what other resources they have. However, few experts would suggest that a typical household will be able to maintain its standard of living in retirement with only one to two years’ worth of income saved in a retirement account unless a household member has a traditional pension in addition to Social Security income.

Share this chart:

Copy the code below to embed this chart on your website.

Older Black and Hispanic households have much lower retirement account savings

2K
Share of older households with retirement account savings by race and ethnicity, and age, 2019
Race/ethnicity Has retirement account savings No retirement savings
White   61.8% 38.2%
Black   35.5% 64.5%
Hispanic 29.5% 70.5%
White   48.2% 51.8%
Black   20.8% 79.2%
Hispanic 15.0% 85.0%
ChartData
Median account balances of older households with retirement account savings by race and ethnicity, and age, 2019

Conditional median retirement account balance (2019$)
White   $156,000
Black   $62,000
Hispanic $105,000
White   $150,000
Black   $38,600
Hispanic $41,000
ChartData

Notes: Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans. Hispanic refers to Hispanic of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks.

Retirement account savings include funds in 401(k)-style defined contribution plans and in IRAs, but not in defined benefit pension plans. Hispanic refers to Hispanic of any race, while white and Black refer to non-Hispanic whites and non-Hispanic Blacks. A household’s age, race, and ethnicity are based on the age, race, and ethnicity of the reference person, who, in households of more than one person, is defined by the Survey of Consumer Finances as the male in a mixed-sex couple or the older person in a same-sex couple. Non-Hispanic Asian Americans/Pacific Islander households and those not included in the other groups are not shown in the graph due to small sample sizes. Account balance calculations for Hispanic households age 55 and older and Black households age 65 and older should be interpreted with caution because of small sample sizes of such households with retirement account savings. The median household retirement account savings includes reported savings in accounts held by both spouses or partners, in the case of married or partnered couples, and—less commonly—the retirement account savings of other people who are financially interdependent with the “economically dominant” individual or couple in the household (Bhutta et al. 2020).

Source: Economic Policy Institute (EPI) and Schwartz Center for Economic Policy Analysis (SCEPA) analysis of Survey of Consumer Finances 2019 microdata (Federal Reserve 2022).

Older Black and Hispanic households are far less likely than older white households to have savings in a 401(k)-type account or an IRA. Only 15.0% of Hispanic households and 20.8% of Black households in the age 65 and older group have retirement account savings. Even among retirement savers in this age group, the typical (i.e., median or 50th percentile) account balance is much lower for Black households ($38,600) and Hispanic households ($41,000) than for white households ($150,000).

Among households in the 55–64 age group with retirement account savings, the typical account balance for Hispanic households ($105,000) is higher than the typical balance for Black households ($62,000). But a lower share of Hispanic households have retirement savings in the first place. As with the 65 and older age group, white households in the 55–64 age group are much more likely to have retirement account savings (61.8%) than either Black households (35.5%) or Hispanic households (29.5%). And white households with retirement account savings also have the highest median account balances ($156,000).

Share this chart:

Copy the code below to embed this chart on your website.


See more work by Monique Morrissey, Siavash Radpour, and Barbara Schuster