Poverty Among the Population Aged 65 and
April 14, 2021
Older
Zhe Li
Among the aged population (persons aged 65 and older) in the United States, the poverty rate
Analyst in Social Policy
(the percentage of individuals living in poverty, or economic hardship characterized by low

income) has declined by over two-thirds in the past five decades. In 2019, approximately 8.9% of
Joseph Dalaker
aged individuals had income below the poverty thresholds (dollar amounts used to determine
Analyst in Social Policy
poverty status). However, the number of aged poor has increased since the mid-1970s as the total

number of the aged population has grown. In 2019, 4.9 million people aged 65 and older lived in
poverty.

The poverty rate for the aged population historically was higher than the rates for younger groups, but the aged have
experienced lower poverty rates than children under age 18 since 1974 and lower rates than adults aged 18-64 since the early
1990s. In 2019, the 8.9% poverty rate among individuals aged 65 and older was lower than the 9.4% poverty rate among
adults aged 18-64 and the 14.4% poverty rate among children under 18 years old.
Although the poverty rate has generally declined for the aged population in most demographic groups, certain subgroups of
the aged population still had disproportionately higher poverty rates as of 2019. For example:
 People aged 80 and older have a higher poverty rate than other aged people. Approximately 11.1% of
people aged 80 and older lived in poverty, compared with poverty rates of 9.2% among individuals aged
75-79, 7.4% among those aged 70-74, and 8.4% among those aged 65-69. Women aged 80 and older had
the highest poverty rate among older women and men in all age groups at 13.6% for women aged 80 and
older.
 Individuals aged 65 and older who were not married at the time of the survey generally had a higher
poverty rate than those who were married and living together with spouses. Among women aged 65 and
older, about 14.4% of widows, 15.8% of divorced women, and 16.9% of never-married women had total
incomes below the official poverty threshold compared with 4.7% of married women. Among individuals
aged 65 and older, poverty rates were also high among never-married men at 18.6%.
 Poverty rates vary by race and Hispanic origin. Hispanic origin is distinct from race, and people may
identify with one or more races. In 2019, the poverty rate was 18.0% among those identifying as Black or
African American compared with 17.1% for those identifying as Hispanics, 9.3% for the Asian population,
and 6.8% for the non-Hispanic White population.
The official poverty measure used in the United States is defined using cash income only, before taxes, and is computed
based on food consumption in 1955 and food costs in 1961, indexed to inflation. That definition prevents the official measure
from gauging the effects of noncash benefits, taxes, or tax credits on the low-income population, and it does not consider how
certain other costs, such as housing or medical expenses, might affect them as well. After decades of research, the
Supplemental Poverty Measure (SPM) was developed to address some of the official poverty measure’s limitations. The
SPM poverty rate for the aged population is higher than the official poverty rate (12.8% compared with 8.9% in 2019). This
higher poverty rate results largely from higher medical out-of-pocket costs among the aged.
Social Security and Supplemental Security Income (SSI) are the main federally funded programs that provide cash benefits to
the aged poor. They accounted for almost 90% of total money income received by the aged population whose incomes were
below the poverty thresholds in 2019. The federal government also provides certain noncash benefits to help the aged poor,
such as housing subsidies and the Supplemental Nutrition Assistance Program (SNAP). In 2019, the SPM poverty rate among
individuals aged 65 and older would increase by more than 32 percentage points if Social Security benefits were excluded
from their income resources, holding other economic behaviors constant. Among the other resources, eliminating SSI,
housing subsidies, or SNAP from income would each increase the SPM poverty rate by about one percentage point.
Congressional Research Service

link to page 25 Poverty Among the Population Aged 65 and Older

Contacts
Author Information ....................................................................................................... 21

Congressional Research Service

Poverty Among the Population Aged 65 and Older

Introduction
The aged population (persons age 65 and older) has been and continues to be of interest to
Congress for various federal policies. Several government programs have contributed to increased
incomes among the aged, including Old Age, Survivor and Disability Insurance (OASDI,
commonly known as Social Security) and Supplemental Security Income (SSI).1 However,
certain groups of the aged population—such as widows, divorced women, and never-married men
and women—are stil vulnerable to poverty. In light of those facts, Congress may be interested in
the incidence of poverty among the aged and the effect of existing programs that reduce poverty.2
To address that interest, this report presents time trends and the most recently available poverty
rates (percentages of individuals that are in poverty, or economic hardship characterized by low
income) among the aged population in the United States both as a whole and by demographic
group.3 This report also discusses how federal programs that may provide assistance to the aged
poor affect estimates of poverty among the aged.
Most of this report wil examine poverty among the aged population using the official measure of
poverty, because it has been in use for over 50 years and provides a consistent time series for
examining demographic trends. Analysis of aged poverty in this report is for the time period from
1966 to 2019, before the onset of COVID-19. The poverty rate among the aged has declined by
over two-thirds over the past five decades from 28.5% in 1966 to 8.9% in 2019.4 In 2019, 4.9
mil ion aged individuals had incomes below the poverty thresholds (dollar amounts used to
determine poverty status).5
The official measure has some limitations, among them a limited ability to gauge the impact of
federal programs on the wel -being of the poor. Researchers inside and outside the government
developed a newer measure, the Supplemental Poverty Measure (SPM), to address the

1 See CRS Report R42035, Social Security Primer; and CRS In Focus IF10482, Supplemental Security Income (SSI).
2 Congress has seen proposals to improve income among vulnerable groups of the older population, such as changing
the factors used in the Social Security benefit formula, revising the minimum benefit for long-term low-wage earners,
increasing Social Security benefits for certain survivors, increasing the Social Security cost -of-living adjustments, and
modifying the SSI program to increase payment amounts by changing income exclusions and resource thresholds.
3 For details on the definition and measurement of poverty, see CRS Report R44780, An Introduction to Poverty
Measurem ent.
4 In this report, numbers and percentages of those in poverty are based on the U.S. Census Bureau’s estimates.
National-level data in this report are obtained from the Current Population Survey Annual Social and Economic
Supplement (CPS ASEC) conducted by t he U.S. Census Bureau. Age is measured at the time of the interview
(February, March, or April), but annual income is based on the previous year. While the official poverty measure is
often regarded as a statistical yardstick rather than a complete descrip tion of what people and families need to live, it
offers a measure of economic hardship faced by the low-income population by comparing family income against a
dollar amount called a poverty threshold, a level below which a family is considered to be poor. Research indicates
that, due to COVID-19, the nonresponse to the 2020 CPS ASEC increased substantially compared to surveys in prior
years. T his nonresponse likely biased income estimates up and poverty statistics down. For more information, see
Jonathan Rothbaum and Adam Bee, Coronavirus Infects Surveys, Too: Nonresponse Bias During the Pandem ic in the
CPS ASEC, U.S. Census Bureau, September 15, 2020, https://www.census.gov/content/dam/Census/library/working-
papers/2020/demo/sehsd-wp2020-10.pdf.
5 T he poverty thresholds vary by family size and composition. For example, in 2019, a single individual under age 65
was considered to be living in poverty if that person’s income was less than $13,300; for an individual 65 and over,
$12,261; for a person 65 or older living with one child under 18, $17,555; for an adult (of any age) and two children,
$20,598; and for two adults and two children, $25,926. In all, 48 thresholds are used. T hese thresholds do not vary
geographically, but they are updated annually for inflation using the Consumer Price Index. For further explanation, see
CRS Report R44780, An Introduction to Poverty Measurem ent.
Congressional Research Service
1

Poverty Among the Population Aged 65 and Older

weaknesses of the official measure. Statistics on the aged population using the SPM are presented
at the end of this report in order to better il ustrate the impact of federal programs on the aged
poor. This report compares official poverty data for the aged population with data from the SPM
and provides statistics measuring the impact of federal cash benefits (mainly Social Security and
SSI), taxes, and in-kind benefits (such as housing, energy, and food assistance) on aged poverty.
While it received scrutiny from the academic and federal research community, the SPM is
considered a research measure—it addresses many, but not al , of the limitations of the official
measure, and its methodology is refined periodical y to make use of newer data sources and
improved measurement techniques.6
How the Official Poverty Measure Is Computed
Poverty status is determined by comparing a measure of a family’s resources against a measure of
its needs.7 Families whose resources are less than a dollar amount representing an austere level of
“needs” are considered to be in poverty.8
Defining resources and needs is not straightforward. The official poverty measure is based on 48
dollar amounts cal ed poverty thresholds that vary by family size and composition but not by
geographic area.9 These official thresholds were developed in the 1960s using data available at
the time and are updated annual y for inflation. Rather than by determining amounts for al major
goods and services a family could possibly need, the thresholds were based on the average
amount of a family’s budget that was spent on food. (As of 1955, this was one-third of a family’s
budget.) Food costs were represented in the thresholds by a U.S. Department of Agriculture food
budget designed for families in economic stress. The entire family budget was the cost of this
plan multiplied by three, with some adjustments for couples and individuals. As such, the
thresholds reflect a level of deprivation based on a restrictive food budget and their associated
costs but are not based on a full enumeration of al items families and individuals need. Family
resources are measured in dollars and are based on cash income before taxes. That is, income
from al sources—such as Social Security, earnings, pensions, and other cash benefits—are
included in their pre-tax amounts but not the value of noncash benefits such as housing subsidies
or the Supplemental Nutrition Assistance Program (SNAP). Al poverty data presented in this

6 A more thorough description of the SPM methodology may be found in CRS Report R45031, The Supplemental
Poverty Measure: Its Core Concepts, Developm ent, and Use.
7 T he CPS ASEC is conducted among the civilian, noninstitutional population of the United States. It does not include
residents of prisons or nursing homes or military personnel living on base. According to the National Center for Health
Statistics, an estimated 1.4 million persons resided in nursing homes in 2016. Of this number, almost 1.2 million (83%)
were aged 65 or older. See Centers for Disease Control and Prevention, National Center for Health Statistics, Vital and
Health Statistics, February 2019, T able 9, https://www.cdc.gov/nchs/data/series/sr_03/sr0343-508.pdf. For a thorough
discussion of poverty measurement, see CRS Report R44780, An Introduction to Poverty Measurem ent.
8 All members of a family have the same poverty status. For persons not living with family members, poverty status is
determined using their own needs and resources as if they were a “one-person family.”
9 T he measure described and used in this report is a statistical measure of poverty —the official poverty thresholds
published by the Census Bureau—and is different from another set of dollar figures called poverty guidelines published
by the U.S. Department of Health and Human Services. T he poverty guidelines are a simplification of the poverty
thresholds, are used for administrative (not statistical) purposes, and are sometimes referred to (somewhat
ambiguously) as the federal poverty level. Unlike the official poverty thresholds used by the Census Bureau, the
poverty guidelines include separate dollar amounts for Alaska and Hawaii. For details see CRS Report R44780, An
Introduction to Poverty Measurem ent.
Congressional Research Service
2

link to page 21 Poverty Among the Population Aged 65 and Older

report are estimates based on a survey, and like al survey estimates, they are subject to sampling
and nonsampling error.10
The poverty research community has discussed the official poverty measure’s limitations for
decades. Its use of pretax income general y does not al ow for estimating tax credits’ effects on
the low-income population. It also does not consider in-kind (noncash) benefits (e.g., housing
subsidies) as income and, as a result, cannot (on its own) il ustrate such benefits’ effects on the
poor population. Although the measure of need represented by the thresholds is updated every
year for overal inflation, it may not accurately reflect the current costs of needs. This is because
prices for goods and services related to basic needs may not rise at the same rate as prices for the
other items measured in the Consumer Price Index—the tool used to adjust the official poverty
thresholds for inflation. Since the official measure’s initial development, new data sources have
offered more detail on the goods and services families consume, but developing an approach that
defines basic needs and determines available resources for families to spend on those needs has
taken decades of research and discussion. The SPM resulted from that research and is described
briefly in the section “The Supplemental Poverty Measure.”11
Poverty Status of the Aged
Notwithstanding the official poverty measure’s limitations, for more than 50 years, it has
provided a consistent measure of poverty in the United States, with few methodological changes
over that time, and it is based on empirical measures of need (food budgets and food
consumption, albeit in 1961 and 1955, respectively).12 For these reasons, trends for the aged
population based on the official measure are discussed throughout the rest of the report.
The proportion of the aged population who lived in poverty has declined significantly in the past
50 years. In 1966, 28.5% of individuals aged 65 and older had family incomes below the poverty

10 Estimates computed using different survey samples will likely differ from one another and from the “true”
population value, even when the samples are drawn from the same population. T he margin of error is a measure of an
estimate’s variability due to sampling. T he larger the margin of error is in relation to the size of the estimate, the less
reliable is the estimate. T he Census Bureau’s estimates from the Current Population Survey (CPS), like estimates from
other surveys, likely contain nonsampling error—error due to causes other than the fact that a sample was used in place
of the entire population (for instance, respondents misremembering or misreporting income amounts, respondents
failing to answer the questionnaire, or errors during t he processing of the data file). For example, some researchers
have expressed concerns that the CPS ASEC records may underreport pension income to a certain extent, particularly
income from periodic (irregular) distributions from defined contribution plans and individual retirement accounts. See
Adam Bee and Joshua Mitchell, Do Older Am ericans Have More Incom e Than We Think? U.S. Census Bureau, July
2017, https://www.census.gov/content/dam/Census/library/working-papers/2017/demo/SEHSD-WP2017-39.pdf; and
Anqi Chen, Alicia H. Munnell, and Geoffrey T . Sanzenbacher, How Much Incom e Do Retirees Actually Have?
Evaluating the Evidence from Five National Datasets, Center for Retirement Research, November 2018,
http://crr.bc.edu/working-papers/how-much-income-do-retirees-actually-have-evaluating-the-evidence-from-five-
national-datasets/.
11 A more comprehensive discussion of how the SPM differs from the official poverty measure is available in CRS
Report R45031, The Supplem ental Poverty Measure: Its Core Concepts, Developm ent, and Use .
12 T he use of 1961 food budgets as the basis of the official poverty measure was documented in Betty Peterkin, The
Measure of Poverty: Food Plans for Poverty Measurem ent, T echnical Paper XII, Department of Health, Education, and
Welfare, 1976, p. 1, https://www.census.gov/content/census/en/library/publications/1976/demo/measure-of-
poverty.html. Documentation of Mollie Orshansky’s use of the 1955 Household Food Consumption Survey is available
in Gordon Fisher, The Developm ent of the Orshansky Poverty Thresholds and Their Subsequent History as the Official
U.S. Poverty Measure, U.S. Census Bureau, September 1997, https://www.census.gov/library/working-papers/1997/
demo/fisher-02.html.
Congressional Research Service
3

link to page 8

Poverty Among the Population Aged 65 and Older

thresholds.13 By 2019, the poverty rate among the aged population had dropped to 8.9% (see
Figure 1). One study suggests that increased Social Security retirement benefits explained most
of the decline in poverty among the aged that occurred during 1967-2000.14 However, whereas
the proportion of persons aged 65 and older who live in poverty has fal en over the past five
decades, the number of aged poor has increased since the mid-1970s as the total number of the
aged population has grown. In 1974, 3.1 mil ion people aged 65 and older lived in poverty,
whereas in 2019, 4.9 mil ion people aged 65 and older lived in poverty. This is partly because
many persons in the large cohort of baby boomers (people born between 1946 and 1964) have
moved into the 65+ age category and people are living longer than before, on average.15
Figure 1. Number of Individuals Aged 65 and Older Below Poverty and Poverty Rate,
1966-2019
Poverty rates in percentages, number of individuals in thousands. Shaded bars indicate recessions.

Source: U.S. Census Bureau, Current Population Survey, 1967-2020 Annual Social and Economic Supplements,
“Historical Poverty Table 3,” https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-
poverty-people.html. Recession dates obtained from National Bureau of Economic Research (NBER), “US
Business Cycle Expansions and Contractions,” http://www.nber.org/cycles/cyclesmain.html.
Notes: The Census Bureau implemented a change to the Current Population Survey Annual Social and
Economic Supplements (CPS ASEC) in 2014 to improve income questions on pension withdrawals and asset
income and to improve the accuracy of health insurance coverage estimates. This change was partial y
implemented with the 2013 data and ful y implemented for the 2014 data. Data for 2013 in the above figure
reflect the values from the CPS ASEC sample that received the redesigned income questions. Due to COVID-19,

13 A family is defined in the CPS as a group of two people or more related by birt h, marriage, or adoption and residing
together. Fam ily incom e in the CPS is the sum of income for all family members. T he person’s income is counted as
family income if the person is living alone.
14 Gary V. Engelhardt and Jonathan Gruber, Social Security and the Evolution of Elderly Poverty, National Bureau of
Economic Research (NBER), Working Paper no. 10466, May 2004, https://www.nber.org/papers/w10466. T he study
showed that the aged poverty rate declined rapidly as the Social Security program expenditures per capita grew quickly
in the 1960s and 1970s and then declined more slowly as program growth slowed in the 1980s and 1990s.
15 T he average remaining life expectancy at age 65 of Americans born in 1975 was 13.7 years for men and 18.0 years
for women. It has been estimated that those born in 20 19 will live for an average of 18.1 years after age 65 for men and
20.6 years for women. See Social Security Board of T rustees, The 2020 Annual Report of the Board of Trustees of the
Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds, 2020, T able V.A4,
https://www.ssa.gov/OACT /T R/2020/index.html.
Congressional Research Service
4

link to page 9

Poverty Among the Population Aged 65 and Older

the nonresponse to the 2020 CPS ASEC (regarding 2019 poverty status) increased substantial y compared to
surveys in prior years. This nonresponse likely biased income estimates up and poverty statistics down.
The poverty rate for individuals aged 65 and older historical y was higher than the rates for adults
aged 18 to 64 and children under the age of 18, but today it is the lowest among those three age
groups.16 (See Figure 2.) For example:
 In 1966, the poverty rate among persons aged 65 and older was 28.5%, compared
with 10.5% among adults aged 18-64 and 17.6% among children under the age of
18.
 In 1974, the aged poverty rate fel below the rate among children under the age of
18, and by the early 1990s, the aged poverty rate had fal en below the rate among
adults aged 18-64. The aged poverty rate has remained lower than the poverty
rate for adults aged 18-64 since that time.
 The poverty rate among individuals aged 65 and older was 8.9% in 2019, which
was lower than the 9.4% poverty rate among adults aged 18-64 and the 14.4%
poverty rate among children under 18 years old.17
Figure 2. Poverty Rates, by Age Group: 1966-2019
Poverty rates in percentages. Shaded bars indicate recessions.

Source: CRS using data from U.S. Census Bureau, Current Population Survey, 1967-2020 Annual Social and
Economic Supplements, “Historical Poverty Table 3,” http://www2.census.gov/programs-surveys/cps/tables/time-

16 See CRS Report R46000, Poverty in the United States in 2018: In Brief.
17 T he 2014 redesign of the CPS ASEC improved the income questions on pension withdrawals and asset inc ome and
reduced nonresponses. However, some studies of the redesign find modest improvements in median incomes and no
evidence of any change in poverty of the aged population. See Jessica L. Semega and Edward Welniak Jr., The Effects
of Changes to the Current Population Survey Annual Social and Economic Supplement on Estim ates of Incom e , U.S.
Census Bureau, January 2015, https://www.census.gov/content/dam/Census/library/working-papers/2015/DEMO/
ASSA-Income-CPSASEC-Red.pdf; and Joshua Mitchell and T rudi Renwick, A Com parison of Official Poverty
Estim ates in the Redesigned Current Population Survey Annual Social and Economic Supplement, U.S. Census Bureau,
January 4, 2015, https://www.census.gov/content/dam/Census/library/working-papers/2014/demo/SEHSD-WP2014-
35.pdf.
Congressional Research Service
5

link to page 11 Poverty Among the Population Aged 65 and Older

series/historical-poverty-people/hstpov3.xls, downloaded November 2020. Recession dates obtained from NBER,
“US Business Cycle Expansions and Contractions,” http://www.nber.org/cycles/cyclesmain.html.
Notes: The Census Bureau implemented a change to the CPS ASEC in 2014 to improve income questions on
pension withdrawals and asset income and to improve the accuracy of health insurance coverage estimates. This
change was partial y implemented with the 2013 data and ful y implemented for the 2014 data. Data for 2013 in
the above figure reflect the values from the CPS ASEC sample that received the redesigned income questions.
Due to COVID-19, the nonresponse to the 2020 CPS ASEC (regarding 2019 poverty status) increased
substantial y compared to surveys in prior years. This nonresponse likely biased income estimates up and
poverty statistics down.
Poverty Among the Aged by Demographic
Characteristics
Poverty status among the aged population general y varies across different demographic groups.
This section describes the aged population’s poverty status for selected demographic
characteristics based on age groups, sex, marital status, and race and Hispanic origin.
Age
The poverty rate for al age groups among the older population general y declined in the past four
decades. People aged 80 and older, however, have a higher poverty rate than aged individuals
under the age of 80. Figure 3 displays the percentage of the aged population who were in poverty
by age groups from 1975 to 2019. In 1975, the poverty rate among individuals who were in the
oldest age group (80 and older) was 21.5%, compared with 16.4% among those aged 75-79,
14.4% among those aged 70-74, and 12.5% among those aged 65-69. Poverty rates declined over
the past 40 years, and in 2019, approximately 11.1% of people aged 80 and older lived in poverty
(a 10 percentage point reduction from 1975), but the share was stil higher than the 9.2% poverty
rate among individuals aged 75-79, 7.4% among those aged 70-74, and 8.4% among those aged
65-69.18 Individuals aged 80 and older might be more vulnerable to income risks because they are
more likely to have lower or no earnings (as they phase out of the labor force),19 exhaust existing

18 Some legislative proposals have been introduced to increase income for those very old people (e.g., those aged 80
and older). For example, the Social Security Enhancement and Protection Act of 2019 (H.R. 5392, 116th Congress)
included a provision to provide additional benefits to certain very old or long-term Social Security beneficiaries.
19 In 2019, about 7% of individuals aged 80 and older had some earnings, com pared with 28% of individuals aged 65-
79. In the same year, the median earnings among those with any earnings was $24,440 for individuals aged 80 and
older, compared with $35,000 for those aged 65 -79. Data are based on CRS analysis of the 2020 CPS ASEC.
Congressional Research Service
6

link to page 12

Poverty Among the Population Aged 65 and Older

retirement resources,20 have reduced purchasing power in certain defined benefit pensions,21 and
incur higher medical expenses.22
Figure 3. Poverty Status of Individuals Aged 65 and Older, by Age Group, 1975-2019
Shaded bars indicate recessions.

Source: CRS analysis of data from the CPS ASEC. Recession dates obtained from NBER, “US Business Cycle
Expansions and Contractions,” http://www.nber.org/cycles/cyclesmain.html.
Notes: The Census Bureau implemented a change to the CPS ASEC in 2014 to improve income questions on
pension withdrawals and asset income and to improve the accuracy of health insurance coverage estimates. This
change was partial y implemented with the 2013 data and ful y implemented for the 2014 data. Data for 2013 in
the above figure reflect the values from the CPS ASEC sample that received the redesigned income questions.
Due to COVID-19, the nonresponse to the 2020 CPS ASEC (regarding 2019 poverty status) increased
substantial y compared to surveys in prior years. This nonresponse likely biased income estimates up and
poverty statistics down.
Women aged 80 and older had the highest poverty rate among older women and men in al age
groups (see Figure 4). In 2019, the poverty rate of women aged 80 and older declined to 13.6%,
compared with 7.6% among men in the same age group and 9.3% among women aged 65-69.

20 In 2019, about 35% of households headed by those aged 80 and older had some retirement assets, compared with
49% of households headed by those aged 65-79. T he median retirement asset level among households with any assets
was $62,000 for households headed by those aged 80 and older, compared with $146,000 for households headed by
those aged 65-79. Data are based on CRS analysis of the 2019 Survey of Consumer Finances, which includes the
following tax-advantaged accounts in retirement assets: defined contribution plans and Individual Retirement
Account s, Profit Sharing Plan, Supplementary Retirement Annuity, Cash Balance Plan, Portable Cash Option Plan, and
etc. For more information, see Federal Reserve, “ Codebook for 2019 Survey of Consumer Finances,”
https://www.federalreserve.gov/econres/files/codebk2019.txt ; and Federal Reserve, “ Macro-Variable Definitions,”
https://www.federalreserve.gov/econres/files/bulletin.macro.txt .
21 A cost-of-living adjustment to mitigate the effects of inflation is provided in Social Security but is not provided in
many defined benefit pensions. In defined benefit pensions where the benefits are not adjusted regularly by the cost of
living, the purchasing power of periodic benefit payments will be eroded over time.
22 T he average amount of personal health care spending was $32,903 for individuals aged 85 and older in 2014,
compared with $16,977 for those aged 65-84. See Department of Health and Human Services, Centers for Medicare
and Medicaid Services, NHE Fact Sheet, 2019, T able 7, https://www.cms.gov/Research-Statistics-Data-and-Systems/
Statistics-T rends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet .
Congressional Research Service
7

link to page 13 link to page 13 link to page 13

Poverty Among the Population Aged 65 and Older

Figure 4. Poverty Status of Individuals Aged 65 and Older in 2019, by
Age Groups and Sex

Source: CRS analysis of data from the 2020 CPS ASEC.
Note: Due to COVID-19, the nonresponse to the 2020 CPS ASEC (regarding 2019 poverty status) increased
substantial y compared to surveys in prior years. This nonresponse likely biased income estimates up and
poverty statistics down.
Poverty status among individuals aged 80 and older varies depending on whether the person is
living with other family members or nonrelatives.23 Poverty rates for those living with other
people in 2019 were about half the rates for those living alone. In 2019, the poverty rate for men
aged 80 and older was 6.4% if they lived with others and 11.2% if they lived alone (see Figure
5). In the same year, the poverty rate for women aged 80 and older was about 8.6% if they lived
with others and 18.9% if they lived alone.

23 Among the aged population, individuals younger than age 80 who lived with family members were more likely to be
married and living with spouses than those aged 80 and older. In 2019, about 81.8% of individuals aged 65 -79 who
lived with other family members or nonrelatives were married and living together with spouses, compare d with 69.5%
of individuals aged 80 and older. Because of the correlation between living arrangement and marital st atus for the
population aged 65-79, this report discusses poverty for this age group based on marital status only. For more
information, see the section “ Marital Status.”
Congressional Research Service
8

link to page 14

Poverty Among the Population Aged 65 and Older

Figure 5. Poverty Rates of Individuals Aged 80 and Older in 2019, by Living Status

Source: CRS analysis of data from the 2020 CPS ASEC.
Note: Living with others include those living with family members or nonrelatives. Due to COVID-19, the
nonresponse to the 2020 CPS ASEC (regarding 2019 poverty status) increased substantial y compared to surveys
in prior years. This nonresponse likely biased income estimates up and poverty statistics down.
Marital Status
Individuals aged 65 and older who were married and living together at the time of the survey
general y had a lower poverty rate than those who were not married, including those widowed,
divorced, never married, separated, or married but living apart (see Figure 6).24 In 1975, about
53.0% of individuals aged 65 and older were married and living together, and this percentage was
slightly higher at 56.8% in 2019. Approximately 8.2% of married persons aged 65 and older and
living together had family incomes below the federal poverty threshold in 1975, and this rate
declined to 4.5% in 2019. During the same period, the poverty rate among aged nonmarried
persons decreased from 23.4% to 15.0%.

24 In the 2017 CPS ASEC, the Census Bureau added separate opposite-sex and same-sex categories for the spouse and
unmarried partner categories. Married-couple families now include same-sex married couples. For details, see Census
Bureau, “Same-Sex Couples Working Papers,” https://www.census.gov/topics/families/same-sex-couples/library/
working-papers.2019.html. For an overview of the effects of the proposed changes on poverty statistics, see Ashley
Edwards, The Presence and Im pact of Sam e-Sex Married Couples on Poverty Rates in the Current Population Survey,
Census Bureau, April 27, 2017, https://www.census.gov/content/dam/Census/library/working-papers/2017/demo/
SEHSD-WP2017-25.pdf.
Congressional Research Service
9

link to page 15

Poverty Among the Population Aged 65 and Older

Figure 6. Poverty Status of Individuals Aged 65 and Older, by Marital Status,
1975-2019
Shaded bars indicate recessions.

Source: CRS analysis of data from the 1976-2020 CPS ASEC. Recession dates obtained from NBER, “US
Business Cycle Expansions and Contractions,” http://www.nber.org/cycles/cyclesmain.html.
Notes: Married individuals are those married and living together. Nonmarried individuals include those
widowed, divorced, never married, separated, and married living apart. The Census Bureau implemented a
change to the CPS ASEC in 2014 to improve income questions on pension withdrawals and asset income and to
improve the accuracy of health insurance coverage estimates. This change was partial y implemented with the
2013 data and ful y implemented for the 2014 data. Data for 2013 in the above figure reflect the values from the
CPS ASEC sample that received the redesigned income questions. Due to COVID-19, the nonresponse to the
2020 CPS ASEC increased substantial y compared to surveys in prior years. This nonresponse likely biased 2019
income estimates up and poverty statistics down.
Figure 7 shows the poverty rate in 2019 by sex and marital status at the survey time. Married
couples generally have significantly lower poverty rates than nonmarried individuals, and
widowed and divorced women aged 65 or older are more likely to be in poverty than their male
counterparts. Among women aged 65 and older, about 4.7% of married women (and living
together with the spouse) had total incomes below the official poverty threshold in 2019,
compared with 14.4% of widows, 15.8% of divorced women, and 16.9% of never-married
women.25 In contrast with the widowed and divorced men in this age group, who are less likely to
be poor than widowed and divorced women, poverty rates are also high among never-married
men at a rate of 18.6% in 2019.
Older widowed and divorced women have a higher observed poverty rate than their male
counterparts for several reasons.26 First, women on average live longer than men,27 and thus more
women are likely to be widowed than are men.28 As a consequence, women may spend more time

25 Recent Congresses have seen legislative proposals to increase Social Security benefits for certain widow(er)s,
disabled widow(er)s, and surviving divorced spouses, such as H.R. 2654, H.R. 4121, H.R. 1540, and S. 345 in the 116th
Congress. Note the definitions of widowed and divorced in this report are based on marital status, not Social Security
benefit types.
26 For more information, see CRS Report R41479, Social Security: Revisiting Benefits for Spouses and Survivors; and
CRS Report R46182, Social Security and Vulnerable Groups—Policy Options to Aid Widows.
27 Social Security Board of T rustees, 2020 Annual Report, T able V.A4.
28 Madonna Harrington Meyer, Douglas A. Wolf, and Christine L. Himes, Linking Benefits to Marital Status: Race and
Congressional Research Service
10

Poverty Among the Population Aged 65 and Older

in retirement and are more vulnerable to inflation and the risk of outliving other assets. The
widow may also confront depleted assets from the deceased spouse’s medical or long-term care
expenses. Second, women enter retirement with fewer retirement resources than men. Women are
more likely to have taken employment breaks to care for children or parents and thus had a lower
labor force participation rate than men.29 Breaks in employment result in fewer years of
contributions to Social Security and employer-sponsored pension plans and thus lower retirement
benefits.30 Moreover, the median earnings of women who are full-time wage and salary workers
are 82% of their male counterparts.31 Because Social Security and private pension benefits are
linked to earnings, this “earnings gap” can lead to lower benefit amounts for women than for
men. Lastly, a widow is also at risk of a substantial income reduction after the spouse’s death
compared with the couple’s total income prior to the spouse’s death. The widow’s Social Security
benefit may be 33%-50% lower than the combined couple’s Social Security benefit. The deceased
spouse’s pension from work may be lost or cut in half.
Figure 7. Poverty Status of Individuals Aged 65 and Older in 2019, by Marital Status
and Sex

Source: CRS analysis of data from the 2020 CPS ASEC.
Notes: Married persons are married and living with their spouses. Nonmarried persons may be divorced,
widowed, never married, separated, or married but living apart from their spouses. Due to COVID-19, the
nonresponse to the 2020 CPS ASEC increased substantial y compared to surveys in prior years. This
nonresponse likely biased 2019 income estimates up and poverty statistics down.

Dim inishing Access to Social Security Spouse and Widow Benefits in the U.S., Center for Retirement Research, March
2004. See also Christopher R. T amborini, Howard Iams, and Kevin Whitman, “ Marital History, Race and Social
Security: Spouse and Widow Benefit Eligibility in the United States,” Research on Aging, vol. 31, no. 5 (2009), pp.
577-605.
29 Bureau of Labor Statistics, Women in the Labor Force: A Databook, December 2019, T ables 1, 5 and 7,
https://www.bls.gov/opub/report s/womens-databook/2019/home.htm.
30 In December 2020, the monthly average Social Security benefits for retired workers were $1,714 for men and $1,378
for women. See Social Security Administration, the Office of Chief Actuary, “Number of Social Security Recipients at
the End of Dec 2020,” https://www.ssa.gov/cgi-bin/currentpay.cgi. T he Employee Benefit Research Institute found that
the median balance in individual retirement accounts was $38,842 for men and $28,616 for women in 2016 . See Craig
Copeland, Individual Retirem ent Account Balances, Contributions, Withdrawals, and Asset Allocation Longitudinal
Results 2010–2016: The EBRI IRA Database, Employee Benefit Research Institute, October 2018.
31 Bureau of Labor Statistics, Women in the Labor Force, T able 16.
Congressional Research Service
11

link to page 16 link to page 17

Poverty Among the Population Aged 65 and Older

In 2019, roughly 6% of individuals aged 65 and older lived in families with children under 18
years old. Poverty rates among aged men and women varied by the presence of children in the
family (see Figure 8), although not always in the same direction. Among married men and
women, a relatively higher share of those with children lived in poverty (8.7% for men and 10.1%
for women) than those without any children (4.1% for men and 4.4% for women). In contrast,
among widowed, divorced and never-married individuals, those living with children had
relatively lower poverty rates than those not living with children.
Figure 8. Poverty Status of Individuals Aged 65 and Older in 2019, by Marital Status,
Sex, and the Presence of Children

Source: CRS analysis of data from the 2020 CPS ASEC.
Notes: Married persons are married and living with their spouses. Children are those under age 18. Due to
COVID-19, the nonresponse to the 2020 CPS ASEC increased substantial y compared to surveys in prior years.
This nonresponse likely biased income estimates up and poverty statistics down.
Race and Hispanic Origin32
Poverty rates vary by race and Hispanic origin, as shown in Figure 9. In surveys, Hispanic origin
is asked separately from race. Accordingly, persons identifying as Hispanic may be of any race.
The poverty rate for the aged population has decreased among persons identifying as Black or
African American alone, non-Hispanic White alone, and Hispanic from 1975 to 2019. Among
aged African Americans, the poverty rate decreased from 36.3% in 1975 to 18.0% in 2019;
among the aged non-Hispanic White population, from 13.0% to 6.8%; and among the aged
Hispanic population, from 32.6% to 17.1%. During the period for which data are available (since

32 Since 2002, federal surveys have asked respondents to identify with one or more races. Previously they could choose
only one. T he groups in this section represent those who identified with one race alone. Another approach is to include
those who selected each race group either alone or in combination with one or more other races. T hose data are also
available in Jessica Semega et al., Incom e and Poverty in the United States: 2019 , Census Bureau, September 15, 2020,
https://www.census.gov/library/publications/2020/demo/p60-270.html, in Appendix B and accompanying historical
data tables.
Congressional Research Service
12

link to page 17

Poverty Among the Population Aged 65 and Older

1987), the poverty rate for the aged Asian population ranged between 7.4% and 16.7% with no
consistent directional trend.33
Figure 9. Poverty Status of Individuals Aged 65 and Older, by Race and
Hispanic Origin, 1975-2019
Shaded bars indicate recessions.

Source: U.S. Census Bureau, 1976-2020 CPS ASEC, Historical Poverty Tables: People and Families-1959 to 2019,
https://www.census.gov/data/tables/time-series/demo/income-poverty/historica l-poverty-people.html. Recession
dates obtained from NBER, “US Business Cycle Expansions and Contractions,” http://www.nber.org/cycles/
cyclesmain.html.
Notes: People of Hispanic origin may be of any race. Racial groups shown above include people of Hispanic
origin unless otherwise indicated (White alone, not Hispanic). Additional y, beginning in 2002, respondents may
identify with one or more racial groups. Previously they could choose only one. Except for “Hispanic,” the
remaining groups shown include those who identified with one race only. Data for Native Hawai ans and Other
Pacific Islanders, American Indians and Alaska Natives, and the population of two or more races are not shown,
though individuals who identify with one these racial groups (or two or more racial groups) may be included in
the share of individuals who identify as Hispanic. From 1987 to 2001, however, Pacific Islanders were included in
the Asian category. The Census Bureau implemented a change to the CPS ASEC in 2014 to improve income
questions on pension withdrawals and asset income and to improve the accuracy of health insurance coverage
estimates. This change was partial y implemented with the 2013 data and ful y implemented for the 2014 data.
Data for 2013 in the above figure reflect the values from the CPS ASEC sample that received the redesigned
income questions. Due to COVID-19, the nonresponse to the 2020 CPS ASEC increased substantial y compared
to surveys in prior years. This nonresponse likely biased 2019 income estimates up and poverty statistics down.

33 Margins of error tend to be larger for smaller demographic groups, and the Asian population is smaller than the other
demographic groups shown. T he larger margins of error help to explain why the line for the Asian population is less
smooth than for the other demographic groups: T he estimates are affected by sampling variability to a greater degree
than for the other groups. Estimates for the American Indian and Alaska Native population, the Native Hawaiian and
Other Pacific Islander population, and the population identifying with two or more races are not sho wn because the
CPS ASEC sample size is not large enough to produce reliable estimates for these groups. Estimates for poverty rates
by race and Hispanic origin are also available based on the American Community Survey (ACS), which is conducted
by the Census Bureau for a much larger sample size. (About 3.5 million addresses are sampled every year.) In 2019, the
proportion of individuals who lived in poverty was about 7.2% for the aged non -Hispanic White population, 17.6% for
the aged Black or African American alone population, 12.6% for the aged Asian alone population, 17.7% for the aged
population identifying as Hispanic or Latino, 16.4% for the aged American Indian and Alaska Native population,
11.9% for the aged Native Hawaiian and Other Pacific Islander p opulation, and 12.5% for the aged population
identifying with two or more races. ACS estimates are not available for the period shown in Figure 9 because it is a
newer survey. For more information about the ACS, see Census Bureau, “American Community Survey,”
https://www.census.gov/programs-surveys/acs.
Congressional Research Service
13

link to page 18 link to page 19

Poverty Among the Population Aged 65 and Older

As shown in Figure 10, among the racial and Hispanic origin groups, in 2019, the poverty rate
was lowest among the aged non-Hispanic White population (5.4% for men and 7.9% for women)
and highest among the aged Black population (14.9% for men and 20.2% for women).
Figure 10. Poverty Status of Individuals Aged 65 and Older in 2019, by Race, Hispanic
Origin, and Sex

Source: CRS analysis of data from the 2020 CPS ASEC.
Notes: People of Hispanic origin may be of any race. Additional y, respondents may identify with one or more
racial groups. Except for “Hispanic,” the remaining groups shown include those who identified with one race
only. Data for Native Hawai ans and Other Pacific Islanders, American Indians and Alaska Natives, and the
population of two or more races are not shown, though individuals who identify with one of these racial groups
(or two or more racial groups) may be included in the share of individuals who identify as Hispanic. Due to
COVID-19, the nonresponse to the 2020 CPS ASEC increased substantial y compared to surveys in prior years.
This nonresponse likely biased 2019 income estimates up and poverty statistics down.
Federal Programs for the Aged Poor
Social Security and Supplemental Security Income (SSI) are the two main federal programs that
provide cash benefits to the aged poor. Social Security is a federal social insurance program that
provides benefits to insured workers and their eligible family members, provided the workers
worked in jobs covered by Social Security for a sufficient number of years and meet certain other
criteria.34 SSI is a federal assistance program that provides monthly cash benefits to aged, blind,
and disabled individuals who have limited income and assets.35 In 2019, Social Security
accounted for 76.9% of total money income among aged individuals whose family incomes were
below 100% of the poverty threshold and 80.8% among those with family incomes below 125%
of the poverty threshold (see Table 1).36 In the same year, SSI and other cash public assistance
accounted for 9.8% of the total money income for aged individuals whose family incomes were

34 See CRS Report R42035, Social Security Primer.
35 See CRS In Focus IF10482, Supplemental Security Income (SSI).
36 Some studies find that the CPS ASEC underreported income from pensions (including retirement saving accounts)
and assets. T hus the share of total money income from Social Security might be overestimated. See Bee and Mitchell,
Do Older Am ericans Have More Incom e Than We Think?; and Chen, Munnell, and Sanzenbacher, How Much Incom e
Do Retirees Actually Have?
Congressional Research Service
14

link to page 19 link to page 19 link to page 20 Poverty Among the Population Aged 65 and Older

below 100% of the poverty threshold37 and 6.3% for those with family incomes below 125% of
the poverty threshold.38
Table 1. Share of Total Money Income from Specified Sources for Poor Individuals
Aged 65 and Older, 2019
Below 100% of the Poverty
Below 125% of the Poverty

Thresholda
Thresholdb
Total Percentage
100.0%
100.0%
Percentage of Income from—

Earnings
4.3%
5.4%
Social Security
76.9%
80.8%
Pensions
4.0%
2.9%
Asset Income
2.3%
2.1%
SSI and Other Public Assistance
9.8%
6.3%
Other Income
2.6%
2.5%
Number of Observationsc
2,404
3,575
Population (thousands)
4,858
7,285
Source: CRS analysis of data from the 2020 CPS ASEC.
Notes: Al values displayed in the table have been adjusted for population weights provided by the CPS. The
CPS ASEC was designed to col ect information on money income received on a regular basis. Money income is
measured on a pre-tax basis and does not include noncash benefits and transfers. Earnings are the sum of income
from wages and salaries and net income from self-employment. Social Security includes retired-worker benefits,
dependents’ or survivor benefits, and disability benefits. Pensions include Railroad Retirement income,
government employee pensions, and private pensions or annuities. Lump-sum or irregular distributions from
retirement saving accounts (employer-sponsored defined contribution plans and individual retirement accounts)
may not be properly captured in the survey. Asset income includes income from interest, dividends, rent,
royalties, and estates and trusts. Other public assistance includes cash public assistance payments to people with
low income, such as Temporary Assistance for Needy Families, general assistance, and emergency assistance.
Other income includes worker’s compensation, unemployment insurance, alimony, child support, and financial
assistance from friends and relatives. Due to COVID-19, the nonresponse to the 2020 CPS ASEC increased
substantial y compared to surveys in prior years. This nonresponse likely biased income estimates up and
poverty statistics down.
a. The official poverty thresholds in 2019 were $12,261 for single individuals aged 65 and older and $15,453
for couples aged 65 and older with no children. In larger families, thresholds are not distinguished by the
householder’s age: Persons 65 and older in larger families use the same thresholds as young families. For al
48 poverty thresholds see Census Bureau, “Poverty Thresholds,” https://www.census.gov/data/tables/time-
series/demo/income-poverty/historica l-poverty-thresholds.html.
b. The dol ar amounts representing 125% of the poverty threshold are the poverty thresholds multiplied by
1.25. For example, in 2019, for a single person aged 65 or older not living in a family, $15,327 was 125% of
the poverty level. More people fal below 125% of poverty than below 100% of poverty because the income
cutoff is higher.

37 Based on the reported information in the 2020 CPS ASEC, SSI accounted for 96% of total cash public assistance
received by individuals aged 65 and older. Other public assistance includes cash public assistance payments to people
with low income, such as T emporary Assistance for Needy Families, general assistance, and emergency assistance.
38 Some research has shown that survey respondents at the bottom of the income distribution frequently confuse Social
Security with SSI such that SSI plays a larger role among the low-income aged population than the survey suggests.
See Bee and Mitchell, Do Older Am ericans Have More Incom e Than We Think?
Congressional Research Service
15

Poverty Among the Population Aged 65 and Older

c. The number of observations is the number of individuals in the survey who are age 65 or older and below
the poverty cutoff shown.
Social Security is not designed solely for the poor, but benefits are weighted to replace a greater
share of career-average earnings for low-paid workers than for high-paid workers.
Social Security benefits alone, however, would not be sufficient to eliminate poverty for a large
number of older Americans. The poverty rate among Social Security beneficiaries aged 65 and
older was 6.8% in 2019.39 Although the Social Security program contains a special minimum
benefit provision that increases benefits to workers who have many years of low earnings and
meet certain other criteria, this provision has virtual y no effect on the benefits paid to today’s
new retirees.40 According to the Census Bureau’s analysis, in 2019, 35.8% of the aged population
would live in poverty without Social Security benefits, holding other resources and expenses
constant.41
SSI is intended to provide a minimum level of income to adults who have difficulty meeting their
basic living expenses due to age or disability and who have little or no Social Security or other
income.42 Some studies show that the SSI program does not provide effective income protection
for the oldest Americans. For example, the maximum SSI benefit in 2019 was 75% of the poverty
threshold for a single person aged 65 and older and 90% of the poverty threshold for a married
couple aged 65 and older.43 Thus, aged SSI recipients may stil be impoverished. Furthermore, the
maximum SSI benefit is more generous for married couples, who are less likely to need
assistance than are older single individuals.

39 CRS analysis of data from the 2020 CPS ASEC.
40 T he special minimum benefit (also known as the special minimum primary insurance amount) is linked to prices,
whereas the regular Social Security benefit is linked to wages. Because wages generally grow faster than prices, t he
special minimum benefit affects fewer beneficiaries every year. T he Social Security Administration estimates that the
provision will have no effect on workers who attain age 62 in 2019 or later. See CRS Report R43615, Social Security:
Minim um Benefits.
41 See Census Bureau, Impact on Poverty of Alternative Resource Measure by Age: 1981 -2019, September 2020,
https://www.census.gov/data/tables/2020/demo/income-poverty/p60-270.html. Social Security benefits are based on a
worker’s career-average earnings in Social Security–covered employment (i.e., jobs in which the worker’s earnings
were subject to Social Security payroll taxes). If the Social Security program had been abolished, people would not
receive Social Security income during their retirement years, but they would have also incurred fewer payroll taxes
during their working years—which in turn implies that they would have had additional money for other purposes such
as consumption, retirement savings, and investment. T he values presented above do not reflect the effect of reducing
payroll taxes or other behavioral changes.
42 See CRS In Focus IF10482, Supplemental Security Income (SSI). In 2019, the SSI program paid out nearly $55.4
billion in federal benefits, of which about 20% (or $11.3 billion) was paid to recipients aged 65 or older. See Social
Security Administration, “ SSI Monthly Statistics, 2019,” T able 6, https://www.ssa.gov/policy/docs/statcomps/
ssi_monthly/2019/table06.html.
43 In 2019, the monthly maximum federal SSI payment amount was $771 for an eligible individual (or $9,252 for the
year), which was about 75% of the poverty threshold for an individual aged 65 and older ($12,261). In the same year,
the monthly maximum federal SSI payment amount was $1,157 for an eligible couple (or $13,884 for the year), which
was about 90% of the poverty threshold for an aged couple with no children ($15,453). For more information, see
Center on Budget and Policy Priorities, 2014, “Introduction to the Supp lemental Security Income (SSI) Program,”
https://www.cbpp.org/research/introduction-to-the-supplemental-security-income-ssi-program; and Pamela Herd et al.,
“A T argeted Minimum Benefit Plan: A New Proposal to Reduce Poverty Among Older Social Security Recipients,”
Russell Sage Foundation Journal of the Social Sciences, vol. 4, no. 1 (February 2018), pp. 74-90,
https://www.urban.org/research/publication/targeted-minimum-benefit -plan-new-proposal-reduce-poverty-among-
older-social-security-recipients.
Congressional Research Service
16

Poverty Among the Population Aged 65 and Older

The federal government also provides certain noncash benefits to help the aged poor, such as
housing subsidies and SNAP benefits.44 Congress funds housing subsidy programs, such as public
housing and government subsidies to renters, to help poor and vulnerable populations meet their
housing needs.45 SNAP is designed primarily to increase the food purchasing power of eligible
low-income households to help them buy nutritional y adequate low-cost foods.46
Additional y, the aged population may also receive health care support from federal health
insurance programs, such as Medicare and Medicaid. Medicare is a federal insurance program
that provides coverage of health care services for qualified beneficiaries, including virtual y al of
the population aged 65 and older,47 while Medicaid is a joint federal-state program that finances
the delivery of primary and acute medical services, as wel as long-term services and supports, to
a diverse low-income population, including children, pregnant women, adults, individuals with
disabilities, and people aged 65 and older.48
Individuals aged 65 and older may also receive a smal portion of income from some other federal
programs, including refundable tax credits, school meals, Temporary Assistance for Needy
Families, the Low Income Home Energy Assistance Program, unemployment insurance, workers’
compensation, and the Special Supplemental Nutrition Program for Women, Infants and Children
(WIC).49 The official poverty measure is of limited value for analyzing various federal programs’
effects on poverty status among the aged population, but the SPM, discussed in the following
section, addresses some of those impacts.50
The Supplemental Poverty Measure (SPM)
The official poverty measure was developed in the 1960s and was established by the Bureau of
the Budget (later the Office of Management and Budget) for measuring the official poverty rate in
the United States.51 Under the official poverty measure, an individual is counted as poor if his or

44 Housing subsidies and SNAP benefits are provided to poor individuals of all ages, although the share of the po or
population that receives these benefits is usually smaller among individuals aged 65 and older relative to those under
age 65. See, for example, April Yanyuan Wu, Why Do So Few Elderly Use Food Stam ps? , University of Chicago,
Harris School of Public Policy Studies, October 2009, http://citeseerx.ist.psu.edu/viewdoc/download?doi=
10.1.1.167.6663&rep=rep1&type=pdf.
45 See CRS Report RL34591, Overview of Federal Housing Assistance Programs and Policy.
46 See CRS Report R42505, Supplemental Nutrition Assistance Program (SNAP): A Primer on Eligibility and Benefits.
47 See CRS Report R40425, Medicare Primer.
48 See CRS In Focus IF10322, Medicaid Primer.
49 For additional information on each program, see CRS Report R43783, School Meals Programs and Other USDA
Child Nutrition Program s: A Prim er; CRS Report R43634, Tem porary Assistance for Needy Fam ilies (TANF):
Eligibility and Benefit Am ounts in State TANF Cash Assistance Program s; CRS Report RL31865, LIHEAP: Program
and Funding; CRS Report R41777, Antipoverty Effects of Unem ploym ent Insurance; CRS Video WVB00014,
Introduction to Workers' Com pensation; and CRS Report R44115, A Prim er on WIC: The Special Supplem ental
Nutrition Program for Wom en, Infants, and Children .
50 T he SPM considers medical out-of-pocket expenses but does not address the full impact of Medicare or Medicaid on
people’s well-being or economic behavior. For a more thorough discussion of the components included in the SPM and
the rationale for SPM methodology, see CRS Report R45031, The Supplem ental Poverty Measure: Its Core Concepts,
Developm ent, and Use.
51 Census Bureau, “ Office of Management and Budget (OMB) in Statistical Policy Directive 14 (May 1978) ,”
https://www.census.gov/topics/income-poverty/poverty/about/history-of-the-poverty-measure/omb-stat-policy-14.html.
T his directive replaced Circular A-46, issued by the Bureau of the Budget in 1969. For a history of the official poverty
measure, see Census Bureau, “ T he History of the Official Poverty Measure,” https://www.census.gov/topics/income-
poverty/poverty/about/history-of-the-poverty-measure.html.
Congressional Research Service
17

Poverty Among the Population Aged 65 and Older

her family’s pretax money income fal s below the poverty threshold. One of the main criticisms
of the official poverty measure is that pretax money income excludes the value of government
noncash benefits (such as health insurance, SNAP, or housing assistance) provided either
privately or publicly.52 It also does not consider taxes paid to federal, state, or local governments
or tax benefits (such as the Earned Income Tax Credit) that families might receive.53
The Census Bureau’s SPM was designed to address the official poverty measure’s limitations and
has been published since 2011.54 The SPM poverty thresholds measure a standard of living based
on expenditures for food, clothing, shelter, and utilities and “a little more” for other expenses.55
Its thresholds—dollar amounts related to the level of need for a family—vary by whether the
family rents, owns a home with a mortgage, or owns a home without a mortgage (the latter of
which is more common among the aged population than it is among younger populations), as wel
as by metropolitan area and state (for nonmetropolitan areas). It computes the amount of
resources available after taxes, includes the values of noncash benefits, and subtracts some
expenses (such as work-related expenses and medical out-of-pocket expenses, the latter of which
tend to be higher among the aged than among younger populations).
In 2019, the most recent data available, the SPM poverty rate for persons aged 65 and older was
12.8%,56 compared with 8.9% using the official poverty measure.57 This higher poverty rate

52 Another criticism of the poverty measure’s use of money income is that income amounts are under reported in
household surveys. One proposed solution, not discussed in detail in this report, is to use a consumption -based poverty
measure. Some scholars argue that quantifying the value of goods and services people consume provides a clearer
picture into their economic well-being and, consequently, their povert y status, because people may have access to
credit, savings, or other resources besides income. However, as with income, survey data on consumption also come
with technical challenges, and other scholars have questioned consumption -based poverty measurement by comparing
the method’s theoretical strengths and weaknesses. For a discussion of one approach to consumption-based poverty
measurement, see Bruce D. Meyer and James X. Sullivan, Annual Report on U.S. Consum ption Poverty: 2017 ,
American Enterprise Inst itute, October 31, 2018, https://www.aei.org/wp-content/uploads/2018/11/2017-Consumption-
Poverty-Report-Meyer-Sullivan-final.pdf. For a comparison of the strengths and weaknesses of income-based and
consumption-based poverty measures, using examples of families in various economic situations, see David S.
Johnson, “Measuring Consumption and Consumption Poverty: Possibilities and Issues,” November 18, 2004,
http://www.welfareacademy.org/pubs/poverty/Johnson.pdf. For an analysis that presents serious criticisms of
consumption-based poverty measures, see National Academies of Sciences, Engineering, and Medicine, A Roadm ap to
Reducing Child Poverty (Washington, DC: National Academies Press, 2019), Appendix D, https://doi.org/10.17226/
25246.
53 Constance F. Citro and Robert T . Michael, eds., Measuring Poverty: A New Approach (Washington, DC: National
Academies Press, 1995); Rebecca Blank, “ Presidential Address: How to Improve Poverty Measurement in the United
States,” Journal of Policy Analysis and Management, vol. 27, no. 2 (Spring 2008), pp. 233-254; and T rudi Renwick
and Liana Fox, The Supplem ental Poverty Measure: 2015 , Census Bureau, September 2016, pp. 60-258.
54 T he SPM has been published as a research measure, per the guidance of an interagency technical working group
organized by the Office of Management and Budget. As the working group observed, “ T he SPM would not replace the
official poverty measure. T he Working Group has designed it as an experimental measure that defines thresholds and
resources in a manner different from the official poverty measure. T he SPM should be considered a work in progress,
with the expectation that there will be improvements to it over time. ” See “ Observations from the Interagency
T echnical Working Group on Developing a Supplemental Poverty Measure,” March 2010, published on the Census
Bureau’s website at https://www.census.gov/content/dam/Census/topics/income/supplemental-poverty-measure/spm-
twgobservations.pdf.
55 See CRS Report R45031, The Supplemental Poverty Measure: Its Core Concepts, Development, and Use.
56 Liana Fox, The Supplemental Poverty Measure: 2019, Census Bureau, September 2020, https://www.census.gov/
library/publications/2020/demo/p60-272.html.
57 T he SPM poverty rate for persons aged 18-64 was 11.2% in 2019, compared with an official poverty rate of 9.4% in
the same year. Liana Fox, The Supplem ental Poverty Measure: 2019 .
Congressional Research Service
18

link to page 24 Poverty Among the Population Aged 65 and Older

results largely from higher medical out-of-pocket costs among the aged, in spite of lower housing
expenses among the aged, who are more likely to have paid off their mortgages.
Income Sources’ Impact on Poverty of the Aged per the SPM
The data presented in Figure 11 il ustrate how changing the definition of the SPM to exclude a
particular resource or expenditure can affect the SPM poverty rate among the aged population.
The data do not consider the behavioral effects that may occur if the resource or cost were to be
eliminated in reality.
Social Security has the greatest effect, by far, on the poverty status of the aged population.
Removing Social Security as a resource while holding the other resources and expenditures
constant would increase the SPM aged poverty rate by about 32.1 percentage points. Among the
other resources, housing subsidies, SSI, and SNAP had the next largest impacts on the SPM
poverty rate but were a full order of magnitude smal er (around a single percentage point instead
of tens of percentage points). The remaining resources affected the SPM poverty rate by much
less than one percentage point. Three of the resources shown are related to child rearing (child
support, school lunch, and WIC), and tax credits are often targeted to families with children.
Households headed by people aged 65 and older are less likely than nonelderly households to
have children present in the family.58
Among the expenses considered in the SPM but not considered in the official measure, medical
out-of-pocket costs had the largest effect: Deducting those costs from family income raised the
SPM poverty rate by 4.0 percentage points. Given that the aged population tends to have greater
medical need and higher out-of-pocket health care costs than do younger populations,59 it is
perhaps not surprising that medical costs had a larger effect than the other costs shown in the
figure. The remaining costs were largely related to work, and, congruent with the aged
population’s lower likelihood to be working compared with younger populations,60 these costs
affected the aged population’s SPM poverty rate by less than one percentage point.

58 In 2019, the average number of children per family was 0.14 children among all households headed by people aged
65 and older and 0.33 children among poor households headed by those aged 65 and older. T he corresponding averages
for all households and poor households headed by those aged 25 -34 were 1.31 and 2.13 children, respectively. T he
corresponding averages for all households and poor households headed by those aged 35 -44 were 1.75 and 2.32
children, respectively. See Census Bureau, “ Mean Number of Related Children per Family, by Family Structure, Age
of Householder and Poverty Status: 2019,” https://www.census.gov/data/tables/time-series/demo/income-poverty/cps-
pov/pov-33.2019.html.
59 Greater health care utilization among the population aged 65 and older, com pared with younger populations, is
documented among several metrics in National Center for Health Statistics, Health, United States, 2019,
https://www.cdc.gov/nchs/data/hus/hus19-508.pdf. Also, see more information at U.S. Centers for Medicare and
Medicaid Services, “NHE Fact Sheet,” 2020, https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-
T rends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.
60 In 2019, nonworkers outnumbered workers among people aged 65 and older by a factor of more than three (41.9
million compared with 12.7 million). Among people aged 18-64, the ratio was reversed (154.6 million workers
compared with 42.9 million nonworkers). See Census Bureau, “ Work Experience During Year by Age, Sex, Household
Relationship and Poverty Status for People 16 Years Old and Over: 2019,” https://www.census.gov/data/tables/time-
series/demo/income-poverty/cps-pov/pov-22.html.
Congressional Research Service
19

Poverty Among the Population Aged 65 and Older

Figure 11. Effects of Resources and Costs on the SPM Poverty Rate for the
Population Aged 65 and Older: 2019
Poverty rates (in percentages) that would result if the SPM were computed without including the element
labeled at left. Each element’s numerical contribution to the SPM poverty rate is shown without
considering any behavioral changes that may result from its removal as an actual cost or resource.

Source: CRS, using data from Liana Fox, The Supplemental Poverty Measure: 2019, Census Bureau, September
2020, Table 6, “Effect of Individual Elements on SPM Rates: 2018 and 2019,” https://www.census.gov/library/
publications/2020/demo/p60-272.html.
Notes: The SPM poverty rate for the aged population (65 and older) was 12.76% in 2019.
SPM = Supplemental Poverty Measure.
SSI = Supplemental Security Income.
SNAP = Supplemental Nutrition Assistance Program.
LIHEAP = Low-Income Home Energy Assistance Program.
TANF = Temporary Assistance for Needy Families.
FICA = Federal Insurance Contributions Act payrol tax.
Additional Considerations
Poverty Not Measured for Certain Populations
Approximately 1.2 mil ion persons in nursing homes are aged 65 or older.61 Poverty status is not
measured for the institutionalized population, which includes persons in nursing homes or prisons
or military personnel living on base. This exclusion is not trivial considering that the population

61 According to the National Center for Health Statistics, an estimated 1.4 million persons resided in nursing homes in
2016. Of this number, almost 1.2 million (83%) were aged 65 or older. See National Center for Health Statistics, Vital
and Health Statistics, February 2019, T able 9, https://www.cdc.gov/nchs/data/series/sr_03/sr03_43-508.pdf. For a
thorough discussion of poverty measurement, see CRS Report R44780, An Introduction to Poverty Measurem ent.
Congressional Research Service
20

Poverty Among the Population Aged 65 and Older

in nursing homes is about one-fourth as large as the 4.9 mil ion persons aged 65 or older who
were in poverty in 2019.62
Health Status Not Directly Included in Poverty Measures
Poverty is used as a measure of wel -being, but it measures only economic wel -being and does
not directly include a person’s health status. Health status may influence the amount and types of
income a person receives (by affecting, for example, ability to work or receive disability benefits)
and is thus considered indirectly.63 However, the noneconomic aspect of wel -being that comes
from good health is not considered in the poverty measures discussed in this report. Furthermore,
in the SPM, medical out-of-pocket expenses are considered, but the overal value of health
insurance programs to the individual, which may wel exceed out-of-pocket costs for medical care
or insurance premiums, is not. Considering that Medicaid is an important vehicle for long-term
care, the benefits Medicaid provides to the aged population could be characterized as fulfil ing
needs that are not solely medical in nature but have economic value as wel .

Author Information

Zhe Li
Joseph Dalaker
Analyst in Social Policy
Analyst in Social Policy

Disclaimer
This document was prepared by the Congressional Research Service (CRS). CRS serves as nonpartisan
shared staff to congressional committees and Members of Congress. It operates solely at the behest of and
under the direction of Congress. Information in a CRS Report should n ot be relied upon for purposes other
than public understanding of information that has been provided by CRS to Members of Congress in
connection with CRS’s institutional role. CRS Reports, as a work of the United States Government, are not
subject to copyright protection in the United States. Any CRS Report may be reproduced and distributed in
its entirety without permission from CRS. However, as a CRS Report may include copyrighted images or
material from a third party, you may need to obtain the permission of the copyright holder if you wish to
copy or otherwise use copyrighted material.

62 According to a 2010 analysis, at least four-fifths of the aged population lack sufficient personal resources to live in a
nursing home for more than three years, and almost two -thirds cannot afford one year. Gretchen Engquist, Cyndy
Johnson, and William Courtland Johnson, System s of Care: Environm ental Scan of Medicaid -Funded Long-Term
Supports and Services, Center for Health Care Strategies, May 2010, cited in Loraine A. West et al., 65+ in the United
States: 2010, Census Bureau, June 2014, p. 51. For further information on how long-term care is financed, see CRS In
Focus IF10343, Who Pays for Long-Term Services and Supports?
63 According to a Kaiser Family Foundat ion analysis of the aged population, those reporting “fair” or “poor” health
have a poverty rate more than twice as high as the aged reporting “excellent” or “very good” health under both the
official poverty measure and the SPM. Juliette Cubanski et al., How Many Seniors Live in Poverty?, Kaiser Family
Foundation, November 19, 2018, https://www.kff.org/report -section/how-many-seniors-live-in-poverty-issue-brief/.
Congressional Research Service
R45791 · VERSION 5 · UPDATED
21