Seven facts about the economic security of older adults

Social Security is vital to the economic security of many older Americans. However, a Social Security solvency crisis is anticipated in 2033, jeopardizing future assistance and requiring reforms to the program. This analysis assesses the current state of older adults’ financial wellbeing and the role that Social Security reform might play in strengthening the economic security of older adults.

With data from the March 2023 Current Population Survey (CPS) Annual Social and Economic Supplement (ASEC), which represents calendar year 2022, we group older adults into five quintiles based on the ratio of their family income to the poverty threshold. In addition, we divide the first income quintile into two categories: the first represents individuals who are in poverty (50.9% of the quintile), while the second represents the remaining individuals in the first quintile with incomes above the poverty line. Additional information regarding our measure of income to poverty is available in the Methods section.

Using this measure of income as a ratio of the poverty threshold, this analysis examines the demographics and economic security of older adults across the income distribution. We describe the characteristics of low-income older adults, including their demographics and living arrangements (facts 1, 2, and 3). In addition, we evaluate the importance of social programs in maintaining the economic security of older adults (facts 4, 5, and 6) and the magnitude of older Americans’ medical expenses relative to their income (fact 7). Taken together, these facts aim to contextualize the current role of social programs in undergirding economic security for older adults and highlight potential avenues for reform.

This analysis is likely impacted by reporting error in the CPS, which underestimates participation and income from retirement accounts and government assistance programs. See the Limitations section for more information.

In Table 1, we categorize the living arrangements of older adults into 7 mutually exclusive and exhaustive categories. We report the share of people in each living arrangement, disaggregated by income quintile.

We find that low-income older adults are significantly more likely to live alone, which is associated with greater financial insecurity and increased health risks. Nearly half (49.7%) of individuals in the lowest income quintile live alone, compared to 14.2% of the highest income quintile. This inequality is starkly divided across genders: while the 14.2% of highest-income older adults living alone is roughly evenly divided into women (7.6%) and men (6.6%), women represent a significantly higher share of people living alone in the lowest income quintiles. For example, 16.7% of all individuals in poverty are men living alone, while 33.0% are women living alone.

Our measure of income depends on the poverty threshold, which varies based on the number of people in each family. Compared to single individuals, couples have a higher income threshold to be considered low-income. As such, single older adults are not automatically considered to be lower income than older adults living with other earners.

We find that the highest-income older adults tend to be younger: the top income quintile has the highest share of people ages 65-74. This finding is likely driven by the fact that we are measuring current income, not lifetime earnings or wealth. The disproportionate representation of older adults aged 65-74 in the higher income quintiles likely reflects the fact that many individuals in this age group continue to work (see Figure 3).

The share of individuals in the oldest age group (85 or older) is roughly two times higher in the bottom income quintile than in the top. The overrepresentation of the oldest age group at the bottom of the income distribution may indicate that individuals in this group need further financial support in addition to existing benefits. This need may be compounded by the fact that individuals age 85 or older have higher health costs on average.

As demonstrated by Figure 2, white older adults are disproportionately represented at the top of the income distribution. The share of white individuals grows steadily across the quintiles, increasing from 59.6% of individuals in poverty to 83.0% of the top income quintile. Conversely, Black and Hispanic individuals are concentrated in the lower income quintiles. The “Other” category includes all respondents reporting multiple races and/or a non-Hispanic race other than Black or white. This category is relatively evenly distributed across the income quintiles.

Within the lowest income quintile, individuals in poverty are less likely to be white (59.6% compared to 63.8%) and more likely to be Black (16.7% compared to 14.2%) relative to non-poor individuals in the first quintile.

These racial/ethnic inequities in income likely reflect the cumulative effects of systemic disadvantages in educational and labor markets. Throughout the life course, people of color often have less access to economic opportunities such as generational wealth, higher education, stable employment, and employer contributions to retirement benefits. As such, individuals of color, particularly those who are Black and/or Hispanic, are less likely to have income from investments, pensions, and retirement accounts as older adults. Furthermore, many older people of color are immigrants and may not have worked long enough in the U.S. to qualify for or receive significant Social Security benefits. As of 2020, immigrants who arrived at age 50 or later represented almost half of all individuals who will never receive Social Security benefits, and 60% of never-beneficiaries are non-white (compared to 24% of current and future beneficiaries).

Figure 3 reports average incomes for each income quintile, disaggregated by source. We find that average annual family income in the top quintile is over $215,000, compared to a mere $8,600 for families in poverty. This difference is primarily driven by earnings, business, or farm income, as well as pension, retirement, and capital income. These income sources represent a comparatively higher share of income in the top quintiles. For example, only 22% of poor families receive income from investments, pensions, or retirement accounts, while 92% of families in the top quintile receive such income. Together, these forms of income account for approximately $177,000 (88%) of the average difference in total income between the lowest- and highest-income quintiles.

As previously noted, reporting error may influence these results: researchers have documented that measures of income from retirement accounts are often lower in the CPS than in administrative data sources.

Figure 4 reports the share of average income represented by each source, demonstrating that low-income older adults are relatively more dependent on Social Security and Supplemental Security Income (SSI) benefits. In the bottom income quintile, the average Social Security income is $10,844, representing 73% of average total income ($14,794). Conversely, Social Security benefits represent only 14% of average income in the top quintile ($29,839 of $215,181). Table 2 reports additional information regarding Social Security and SSI recipiency by income quintile. Across all quintiles, 85% of older adults live in a family receiving Social Security benefits. The percentage of families receiving Social Security peaks among families in the second income quintile (91.5%). It declines to 85.4% in quintile 4 and drops significantly in quintile 5 to 79.4%. This decline likely represents the young, more affluent older adults waiting until age 70 to collect benefits. The average value of Social Security benefits is similar among the top three quintiles.

Note that the reported figures for average Social Security and SSI income include observations with a value of zero, representing those who do not receive any income of the given type. To calculate the average benefit received by those families receiving a given income type, one must divide the average benefit by the share of the population receiving benefits. For example, for individuals in poverty the average benefit among Social Security recipients is $9,087 ($6,061 divided by 66.7%).

Figure 5 shows the share of older adults in families receiving various safety net benefits, disaggregated by income quintile. Because we are focusing on low-income older adults, we only show the first and second income quintiles. We find that a relatively low share of low-income older adults, particularly those in poverty, receive benefits from safety net programs.

First, there is a striking difference in Social Security recipiency between poor and non-poor older adults. While 90% of non-poor individuals in the first income quintile live in a family receiving Social Security benefits, that figure falls to 67% for poor individuals. Over 30% of poor individuals whose families do not receive Social Security income are immigrants, compared to 19% of poor individuals receiving benefits. Of these immigrants in poverty who do not receive Social Security benefits, 25% immigrated in the past two decades (after 2003) and approximately 35% immigrated after age 50.

In addition, Social Security recipients in poverty receive less generous benefits than non-poor individuals in the first income quintile. For the latter group, the average family benefit among recipients is $17,520 ($15,803 divided by 90.2%; see Table 2). Conversely, the average poor recipient’s family Social Security benefit is only $9,087.

As such, differences in average Social Security income are driven by disparities in both the share of individuals receiving benefits and the average benefit among recipients. Overall, these differences in Social Security income are significant, accounting for 77% of the overall income disparity between poor and non-poor individuals in the first income quintile. Indeed, Social Security may be the driving force that pulls many of the individuals in the first income quintile out of poverty.

Beyond Social Security, 29% of individuals in poverty report that at least one member of their family is on Medicaid. A mere 11% of individuals in poverty report receiving SSI benefits, and only 28% report receiving Supplemental Nutrition Assistance Program (SNAP) benefits (formerly known as the “Food Stamps” program). Accessing these benefits could significantly improve the economic security of current non-recipients: among poor SNAP recipients, the average value of their family’s SNAP benefit is over $2,400. This represents over a quarter of the average family income for poor older adults.

The low rate of SNAP recipiency is a missed opportunity: many older adults with family income at or below the poverty threshold should be eligible for SNAP benefits. However, because of low SNAP participation rates among older adults, many individuals are not receiving the benefits to which they are entitled.

Finally, only 17% of older adults in poverty report living in public housing or receiving rent subsidies from the government (local, state, or federal). Particularly given the current trend of high and rising housing prices, housing support represents another area in which further assistance for poor older adults should be provided.

As previously noted, these estimates are impacted by reporting error: the CPS significantly under-estimates participation and income from government assistance programs such as SNAP and SSI.

Despite the existence of Medicaid and other means-tested programs that subsidize health care costs for low-income individuals, older adults at the bottom end of the income distribution report spending a higher share of their income on health insurance premiums. Figure 6 graphs the average premium payments as a share of income, disaggregated by income quintile. For older adults in poverty, premium payments represent approximately 7% of family income on average, compared to 2% in the top income quintile.

Non-premium health care expenses also pose a disproportionate burden on low-income older adults. Figure 7 reports the share of older adults in each quintile with non-premium out-of-pocket medical expenses representing over 5% of their income. These figures are further disaggregated to show the share of older adults for whom these expenses represent 5%-10% and over 10% of their income, respectively.

Nearly half (45%) of poor older adults report spending over 5% of their income on out-of-pocket medical expenses. This higher share reflects the lower income of these older adults, as average out-of-pocket costs in dollars in the lowest income quintile are less than a quarter of those in the highest income quintile.

Low-income older adults face much higher risk from health costs as well: over 35% of the lowest income older adults report making out-of-pocket medical payments that represent over a tenth of their income, compared to 2.4% of individuals in the highest income quintile.

The facts above illustrate that, in general, Social Security plays a vital role in maintaining the financial health of older adults – particularly in the lower income quintiles. However, Social Security recipiency is relatively low among individuals in poverty. These income differences have equity implications: older adults of color, particularly Black and Hispanic individuals, are more likely to be low-income than their white peers. This analysis highlights the important role that Social Security and SSI reform might play in strengthening the financial security of older adults and narrowing racial/ethnic disparities in economic outcomes. It also illustrates the importance of ensuring that all eligible individuals receive the means-tested benefits to which they are entitled.

Methods

Using data from the March 2023 CPS ASEC, which represents calendar year 2022, older adults are divided into five quintiles based on the ratio of their family income to the poverty threshold. Each quintile contains approximately 11.6 million individuals aged 65 or over. The first quintile represents the lowest-income older adults. In the figures below, this quintile is split into two categories: the first represents individuals who are in poverty (50.9% of the quintile), while the second represents the remaining individuals in the first quintile with incomes above the poverty line.

The ratio of income to poverty is determined at the family level, following the rules used for the calculation of official poverty statistics. Vital to these rules is the distinction between a household and a family. In the CPS, each respondent is asked to describe all the inhabitants of their home. The person in whose name the house is owned or rented is the “household reference person.” All residents in the home are grouped into one “household.”

To be grouped into a “family,” individuals must be related by birth, marriage, or adoption. For the purposes of calculating poverty status, any related individuals in the household are grouped into one family. If no members of a family are related to the household reference person, they are considered an “unrelated sub-family” and their poverty status is calculated separately from that of the household reference person and their family. Poverty status is also separately determined for any unrelated individuals living in the home (e.g., guests, boarders, staff). As such, there may be multiple families living in one household, and the average number of people in a household is larger than the average number of people in a family (see Table 1).

To calculate an older adult’s ratio of income to poverty, we take the sum of all cash income in their family and compare it to the appropriate poverty threshold, which is determined based on family size and age.

Because the CPS ASEC does not provide an account of individuals’ financial assets, this analysis does not study wealth disparities among older adults and instead focuses entirely on annual income.

Limitations

This analysis may be impacted by reporting error in the CPS. Income from retirement accounts and government assistance programs, including SNAP and SSI, are underreported in CPS data. As a result, estimates based on the CPS may overestimate the share of income accounted for by Social Security benefits. For example, based on 2015 CPS data alone, the Social Security Administration (SSA) estimated that 52.5% of older adults received over half of their income from Social Security. However, when these data were supplemented with administrative records from the IRS, that estimate was reduced by roughly a fourth: the updated estimate was 39.9%.

Though reporting error is a notable factor impacting the current analysis, the principal findings are robust to this magnitude of error. Even if the findings from this analysis overstate the reliance of older adults on Social Security by the magnitude estimated by the SSA, the conclusion that Social Security constitutes a sizeable share of income for many older adults, particularly those in the lower half of the income distribution, would remain significant.