RESEARCH & ANALYSIS
Economic Well-Being of
U.S. Households in 2021
May 2022
BOARD OF GOVERNORS OF THE
FEDERAL RESERVE SYSTEM
The Federal Reserve System is the central
bank of the United States. It performs five key
functions to promote the effective operation
of the U.S. economy and, more generally, the
public interest.
The Federal Reserve
■ conducts the nation’s monetary policy to promote maximum employment
and stable prices in the U.S. economy;
■ promotes the stability of the financial system and seeks to minimize
and contain systemic risks through active monitoring and engagement in
the U.S. and abroad;
■ promotes the safety and soundness of individual financial institutions
and monitors their impact on the financial system as a whole;
■ fosters payment and settlement system safety and efficiency through
services to the banking industry and U.S. government that facilitate
U.S.-dollar transactions and payments; and
■ promotes consumer protection and community development through
consumer-focused supervision and examination, research and analysis of
emerging consumer issues and trends, community economic development
activities, and administration of consumer laws and regulations.
To learn more about us, visit www.federalreserve.gov/aboutthefed.htm.
RESEARCH & ANALYSIS
Economic Well-Being of
U.S. Households in 2021
May 2022
BOARD OF GOVERNORS OF THE
FEDERAL RESERVE SYSTEM
Contents
Executive Summary
......................................................................................................... 1
Overall Financial Well-Being ................................................................................................ 2
Income ............................................................................................................................. 2
Employment ...................................................................................................................... 3
Dealing with Unexpected Expenses ..................................................................................... 3
Banking and Credit ............................................................................................................ 3
Housing ............................................................................................................................ 4
Education .......................................................................................................................... 4
Student Loans ................................................................................................................... 5
Retirement and Investments ............................................................................................... 5
Overall Financial Well-Being
.......................................................................................... 7
Current Financial Situation ................................................................................................. 7
Changes in Financial Situation over Time ........................................................................... 12
Local and National Economic Conditions ............................................................................ 14
Overall Life Satisfaction .................................................................................................... 15
Income
............................................................................................................................. 17
Level and Source ............................................................................................................. 17
Child Tax Credit ................................................................................................................ 19
Income Variability ............................................................................................................. 20
Employment
.................................................................................................................... 23
Reasons for Not Working .................................................................................................. 23
Working from Home .......................................................................................................... 25
Job Changes .................................................................................................................... 28
COVID-19 Precautions at Work .......................................................................................... 29
Part-Time Jobs, Temporary Jobs, and Irregular Schedules .................................................... 31
The Gig Economy ............................................................................................................. 31
Dealing with Unexpected Expenses
........................................................................... 35
Small, Unexpected Expenses ............................................................................................ 35
Health-Care Expenses ...................................................................................................... 38
Hardships from Natural Disasters ..................................................................................... 39
Banking and Credit
........................................................................................................ 43
Unbanked and Underbanked ............................................................................................. 43
Overdraft ......................................................................................................................... 46
Credit Outcomes and Perceptions ..................................................................................... 47
Credit Cards .................................................................................................................... 48
Buy Now, Pay Later ........................................................................................................... 49
iii
Housing
............................................................................................................................ 53
Living Arrangements ......................................................................................................... 53
Moving ............................................................................................................................ 54
Homeownership and Mortgages ........................................................................................ 55
Renting and Evictions ....................................................................................................... 56
Education
........................................................................................................................ 61
Modes of Learning in Primary and Secondary School .......................................................... 61
Perceptions of Children’s Performance in School ................................................................ 64
Modes of Learning in Higher Education .............................................................................. 65
Overall Value of Higher Education ...................................................................................... 66
Look Back on Education Decisions .................................................................................... 68
Student Loans
................................................................................................................ 71
Incidence and Types of Education Debt .............................................................................. 71
Student Loan Payment Status ........................................................................................... 73
Relation to Financial Well-Being ......................................................................................... 75
Relation to Self-Assessed Value of Higher Education .......................................................... 75
Retirement and Investments
....................................................................................... 77
Current Retirees .............................................................................................................. 77
Retirement Savings among Non-Retirees ........................................................................... 79
Financial Literacy and Experience with Investments ............................................................ 82
Description of the Survey
............................................................................................. 85
Survey Participation ......................................................................................................... 85
Targeted Outreach and Incentives ...................................................................................... 85
Survey Questionnaire ....................................................................................................... 86
Survey Mode ................................................................................................................... 86
Sampling and Weighting ................................................................................................... 87
Item Non-response and Imputation .................................................................................... 88
Acknowledgments
......................................................................................................... 91
Corrections
..................................................................................................................... 93
iv Economic Well-Being of U.S. Households in 2021
Executive Summary
This report describes the responses to the 2021 Survey of Household Economics and Decision-
making (SHED). The Federal Reserve Board has fielded this survey each fall since 2013 to under-
stand the wide range of financial challenges and opportunities facing families in the United
States.
1
The findings in this report primarily reflect financial circumstances in late October and
early November of 2021, before the increase in coronavirus (COVID-19) cases from the Omicron
variant.
Despite persistent concerns that people expressed about the national economy, the survey high-
lights the positive effects of the recovery on the individual financial circumstances of U.S. fami-
lies. In 2021, perceptions about the national economy declined slightly. Yet self-reported financial
well-being increased to the highest rate since the survey began in 2013. The share of prime-age
adults not working because they could not find work had returned to pre-pandemic levels. More
adults were able to pay all their monthly bills in full than in either 2019 or 2020. Additionally, the
share of adults who would cover a $400 emergency expense completely using cash or its equiva-
lent increased, reaching a new high since the survey began in 2013.
Parents with children at home, who had been disproportionally affected by the pandemic in 2020,
exhibited notable improvements in their financial well-being in 2021. After declining in 2020, par-
ents’ assessments of their financial circumstances rebounded in 2021. This improvement is con-
sistent both with reduced childcare burdens as schools returned to in-person classes, as well as
additional financial resources provided to parents such as the enhanced child tax credit (CTC).
Most parents also said that their child was doing better academically, socially, and emotionally in
2021 than they were a year earlier.
The report also highlights several new topics added to the survey in 2021, such as disruptions
from natural disasters, rental debt, and employer vaccine mandates. These new questions provide
additional context on the experiences of U.S. adults in handling unexpected expenses, paying for
housing, and navigating ongoing changes in the labor market.
To better understand consumer experiences with emerging products, cryptocurrencies and “Buy
Now, Pay Later” (BNPL) products were included on the survey for the first time. While most adults
did not use cryptocurrencies in the prior year, cryptocurrency use as an investment was far more
1
The latest survey interviewed over 11,000 individuals in October and November 2021. The anonymized data, as well as
appendixes containing the complete SHED questionnaire and responses to all questions in the order asked, are also
available at https://www.federalreserve.gov/consumerscommunities/shed.htm.
1
common than use for transactions or purchases. However, while transactional use of cryptocurren-
cies was low, those using cryptocurrencies for purchases rather than as investments frequently
lacked traditional bank and credit card accounts.
The report also provides insights into long-standing issues related to individuals’ personal finan-
cial circumstances, including returns to education, housing situations, and retirement savings. In
many cases, the report finds that disparities by education, race and ethnicity, and income per-
sisted in 2021.
Key findings from the survey include the following:
Overall Financial Well-Being
In the fourth quarter of 2021, the share of adults who were doing at least okay financially increased
relative to 2020. With these improvements, overall financial well-being reached its highest level since
the survey began in 2013.
•
Seventy-eight percent of adults were either doing okay or living comfortably financially, the
highest share with this level of financial well-being since the survey began in 2013.
•
Parents experienced particularly large gains in financial well-being over the prior year. In 2021,
three-fourths of parents said they were doing at least okay financially, up 8 percentage points
from 2020.
•
Forty-eight percent of adults rated their local economy as “good” or “excellent” in 2021. This
share was up from 43 percent in 2020 but well below the 63 percent of adults who rated their
local economy as “good” or “excellent” in 2019, before the pandemic.
Income
The majority of parents received additional income in 2021 through the monthly CTC. Most higher-
income parents primarily saved this money, while most lower-income parents primarily spent it on
housing, items for their children, or food.
•
Three in 10 CTC recipients with income less than $50,000 used the largest portion of their
credit on housing expenses, just over 2 in 10 spent the largest portion on their child, and
15 percent spent the largest portion on food.
•
Fifteen percent of adults with income less than $50,000 struggled to pay their bills because
of varying monthly income. This challenge was even more acute among people who were par-
ents in this income range, of whom 27 percent struggled to pay their bills because of income
variability.
2 Economic Well-Being of U.S. Households in 2021
Employment
Many people switched jobs in 2021, and those who did generally said that their new job was better
than their old one. Most employees also said that their employer was taking about the right amount
of COVID-19 precautions, although some people not working indicated that concerns about the virus
contributed to the choice not to work.
•
Fifteen percent of workers said they were in a different job than 12 months earlier. Just over 6
in 10 people who changed jobs said their new job was better overall, compared with 1 in 10
who said that it was worse.
•
Seventy-seven percent of employees said their employers were taking the right amount of pre-
cautions against COVID-19. Those who did not were almost evenly split between thinking their
employers were taking too many and too few precautions.
•
Seven percent of all prime-age adults said that they were not working and that concerns about
getting COVID-19 contributed at least in part to their decision not to work.
•
Among those working from home, the share of employees who would look for another job if their
employer required they work in person was similar to the share who would look after a
pay freeze.
Dealing with Unexpected Expenses
The overall share of adults who would cover a small emergency expense using cash or its equivalent
increased to the highest level since 2013, when the survey began. Financial preparedness is an
important buffer for those who encounter unexpected events, such as medical expenses or disrup-
tions from natural disasters.
•
Sixty-eight percent of adults said they would cover a $400 emergency expense exclusively using
cash or its equivalent, up from 50 percent who would pay this way when the survey began in
2013.
•
Twenty percent of adults had major, unexpected medical expenses in the prior 12 months, with
the median amount between $1,000 and $1,999.
•
Sixteen percent of adults experienced a financial disruption or hardship from a natural disaster
or severe weather event in the prior year.
Banking and Credit
Most adults had a bank account and were able to obtain credit from mainstream sources in 2021,
but notable gaps in access to basic financial services still exist among Black and Hispanic adults
and those with low income.
Executive Summary 3
•
Six percent of adults did not have a bank account. Black (13 percent) and Hispanic (11 percent)
adults were more likely not to have a bank account than adults overall.
•
Eleven percent of adults with a bank account paid an overdraft fee in the previous 12 months,
with higher shares of low-income adults having overdrafted over this period.
•
Three percent of adults used cryptocurrency for purchases or money transfers. Among these
transactional users of cryptocurrencies, 13 percent did not have a bank account.
Housing
Low mortgage rates resulted in a continuation of the wave of refinancing in 2021, although high-
income borrowers were primarily the beneficiaries of this opportunity to reduce monthly housing
costs. The share of renters who had been behind on their rent in the prior 12 months was higher
than before the pandemic, and many still owed back rent at the time of the survey.
•
Nearly one-fourth of all homeowners with a mortgage refinanced their mortgage in 2021. This
includes nearly 3 in 10 mortgage holders with an income of at least $100,000, but a lower
16 percent of those with income under $50,000.
•
Seventeen percent of renters were behind on their rent at some point in 2021, including 8 per-
cent who were behind at the time of the survey in late 2021. Among those still behind in late
2021, the total outstanding back rent was between $9.3 billion and $10.9 billion.
Education
At the time of the survey, most parents of primary or secondary school students reported that their
children were attending classes completely in person. Most parents also said that their child was
doing better academically compared with a year earlier. In contrast to the experience of K–12 stu-
dents, online education remained prevalent at higher education institutions in the fall of 2021.
•
Ninety-three percent of parents with a child in public or private school said their youngest child
who was enrolled in K–12 education was attending classes completely in person, compared
with 27 percent attending completely in person in 2020.
•
Fifty-six percent of parents with a child in public or private school said that their child’s aca-
demic performance improved in 2021, compared with 7 percent who said it declined.
•
Seventy-six percent of higher education students in 2021 said they prefer online or
hybrid education, given the situation with the pandemic.
4 Economic Well-Being of U.S. Households in 2021
Student Loans
The share of student loan borrowers who were behind on their payments in the fall of 2021 declined
relative to before the pandemic. These borrowers also saw increases in their financial well-being
compared with prior years.
•
Twelve percent of borrowers were behind on their payments in 2021, a significant decline from
the 17 percent who were behind in the fall of 2019.
•
Seventy-three percent of those who went to college and have student loans for their own educa-
tion were doing at least okay financially in 2021, up from 65 percent before the pandemic.
Retirement and Investments
Among non-retirees, a higher share reported they felt like their retirement savings were on track than
in either 2020 or 2019. However, a sizeable share of recent retirees said COVID-related factors
affected the timing of their retirement decision.
•
Forty percent of non-retirees thought their retirement saving was on track, up from 36 percent in
2020 and 37 percent in 2019.
•
Twenty-five percent of adults who retired in the prior 12 months, and 15 percent of those who
retired one to two years ago, said factors related to COVID-19 contributed to when they retired.
Executive Summary 5
Overall Financial Well-Being
The share of adults doing at least okay financially rose to the highest level since the survey began
in 2013.
2
Although financial challenges and risks to the recovery remain, this generally positive
assessment of financial well-being was consistent with improved economic conditions and addi-
tional COVID-19 relief measures in 2021.
The increase in financial well-being occurred broadly across the population and was especially
large among parents. Even so, existing gaps by education and by race and ethnicity persisted.
Current Financial Situation
At the end of 2021, 78 percent of adults were
doing at least okay financially, meaning they
reported either “doing okay” financially
(39 percent) or “living comfortably” (39 per-
cent). The rest reported either “just getting
by” (16 percent) or “finding it difficult to get
by” (6 percent). The 78 percent of adults
doing at least okay financially in 2021 was up
3 percentage points from 2020 and was well
above the 62 percent doing at least this well
in 2013 (figure 1).
As further evidence of greater financial well-
being in 2021, the share of adults who said
they were living comfortably rose by 4 per-
centage points. This increase in financial well-
being aligns with improved economic condi-
tions and the additional COVID-19 relief
measures enacted in 2021.
3
2
The survey was fielded in October and November 2021 and results reflect financial situations at that time. References
to “in 2021” refer to the 12-month period before the survey rather than the precise calendar year.
3
The American Rescue Plan Act of 2021 became law in March 2021 and provided additional relief to most households to
address the continued impact of the COVID-19 pandemic; see https://www.congress.gov/bill/117th-congress/house-
bill/1319 and https://www.whitehouse.gov/wp-content/uploads/2021/03/American-Rescue-Plan-Fact-Sheet.pdf.
Figure 1. At least doing okay financially (by
year)
78
7575
75
74
70
69
65
62
202120202019201820172016201520142013
Percent
Note: Among all adults.
7
Adults with at least a bachelor’s degree continued to be much more likely to be doing at least okay
financially (91 percent) than those with less than a high school degree (49 percent). The 42 per-
centage point gap in well-being was little changed from the 44 percentage point gap in 2020
(figure 2). Moreover, looking over the past five years shows a steady and sizeable increase in
financial well-being among those with at least a bachelor’s degree (an increase of 9 percentage
points in the share doing at least okay from 2016 to 2021), while adults with less than a high
school degree have not experienced lasting gains in financial well-being.
Parents were one group that experienced particularly large gains in financial well-being over the
prior year. In 2021, three-fourths of parents said they were doing at least okay financially, up 8 per-
centage points from 2020 (figure 3).
Low-income parents saw even more substantial increases in their financial well-being in 2021.
Among parents with income under $25,000, the share doing at least okay financially rose by
13 percentage points, from 40 percent in 2020 to 53 percent in 2021. The share of parents with
income between $25,000 and $49,999 who were doing at least okay financially increased by
7 percentage points, while those with higher income exhibited more modest improvements.
A potential explanation for the large rise in financial well-being among parents is the expansion of
the CTC. The American Rescue Plan temporarily increased the CTC from $2,000 per child to
$3,000 per child ($3,600 for a child under age 6), increased eligibility among low-income families,
Figure 2. At least doing okay financially (by year and education)
Percent
Bachelor’s degree
or more
Some college/technical
or associate degree
High school
degree or GED
Less than a high
school degree
202120202019201820172016201520142013
89
88
87
85
82
80
7777
62
55
47
42
48
46
56
49
54
45
60
64
64
69 69
66
67
62
66
68
69
72
73
72
91
49
70
74
Note: Among all adults. Results for 2017 to 2019 differ slightly from previous reports because of adjustments in educa-
tion coding for consistency.
8 Economic Well-Being of U.S. Households in 2021
and paid the credit monthly (the “Income” section of this report discusses how parents used this
credit).
4
Many families also saw a return to in-person schooling in the fall of 2021, which may
have eased childcare responsibilities and allowed some parents to return to work or work
more hours.
The increase in financial well-being among parents in 2021 contrasts with the decline they experi-
enced from 2019 to 2020 (figure 3).
5
Parents were hit especially hard by the pandemic in 2020,
having experienced higher rates of job loss and having faced disruptions to childcare and in-person
K–12 schooling that affected their availability to work. (See the report Economic Well-Being of U.S
Households in 2020 for additional information).
6
4
More details on the enhanced CTC are available from U.S. Department of the Treasury, “Child Tax Credit,” https://
home.treasury.gov/policy-issues/coronavirus/assistance-for-american-families-and-workers/child-tax-credit.
5
Results in earlier years were updated for consistency with 2021 methods for classifying education status so may differ
slightly from earlier reports.
6
See Economic Well-Being of U.S. Households in 2020, https://www.federalreserve.gov/publications/files/2020-report-
economic-well-being-us-households-202105.pdf.
Figure 3. At least doing okay financially (by year and parental status)
Percent
Parents (living with own children under age 18)
All other adults
2021202020192018201720162015
79
78
77
76
75
71
70
65
68
71
72
71
67
75
Note: Among all adults.
Overall Financial Well-Being 9
Differences in financial well-being across racial and ethnic groups persisted in 2021. Eighty-
eight percent of Asian adults were doing at least okay financially, followed by 81 percent of White
adults, 71 percent of Hispanic adults, and 68 percent of Black adults (figure 4).
7
All racial and ethnic groups measured in the survey saw an increase in financial well-being over the
prior year, with Hispanic adults seeing a particularly sharp increase. In 2021, the 71 percent of
Hispanic adults who said they were doing at least okay was up 7 percentage points from 2020.
The increase in well-being for Hispanic adults was largely concentrated among parents, similar to
the pattern for adults overall. The share of Hispanic parents doing at least okay increased 14 per-
centage points (to 70 percent) in 2021. However, Hispanic adults not living with their own children
under age 18 saw a relatively slight increase (figure 5).
7
The reported categorizations reflect the largest statistical groupings but are neither exhaustive nor the only distinctions
important to understand. Sample sizes for other racial and ethnic groups and subpopulations are not large enough to
produce reliable estimates. Asian adults were separately identified for the first time in the survey in the Economic Well-
Being of U.S. Households in 2020, and in 2021 the Federal Reserve Board identified Asian adults in earlier years of the
survey. However, results for Asian adults are sometimes excluded when the sample size is insufficient to provide a reli-
able estimate.
Figure 4. At least doing okay financially (by year and race/ethnicity)
Percent
84
80
79
85
78
81
77
78
72
79
71
75
68
79
65
75
56
53
61
65
66
66
65
55
55
63
64
64
67
66
64
64
88
81
71
68
Asian
Hispanic
Black
White
202120202019201820172016201520142013
Note: Among all adults.
10 Economic Well-Being of U.S. Households in 2021
Other dimensions across which financial well-being differed include income, geography, LGBTQ+
status, and disability status (table 1). Fifty-five percent of adults with family income less than
$25,000 were doing at least okay financially, compared with 96 percent of adults with family
income greater than $100,000. People living in low- or moderate-income communities also had
lower levels of financial well-being than those living in middle- or upper-income communities.
8
Addi-
tionally, those living in metro areas were faring better than those in non-metro communities.
9
Other surveys have shown that adults identifying as LGBTQ+ were more likely to face economic
insecurity, suggesting LGBTQ+ status may be associated with financial well-being.
10
Consistent
with this evidence, the 2021 SHED found that 67 percent of adults identifying as LGBTQ+ were
doing at least okay financially, compared with 78 percent of the overall population.
11
Moreover, an
even lower 62 percent of adults who were transgender or nonbinary, or who reported their sexual
8
Neighborhood income is defined using the Community Reinvestment Act definition. Under this definition, low- and
moderate-income refers to communities that have a median family income of less than 50 percent of the area median
income. For details on the definition, see https://www.federalreserve.gov/consumerscommunities/cra_resources.htm.
9
Non-metro areas are defined throughout this report as being outside of a Metropolitan Statistical Area (MSA), and metro
areas are those inside of an MSA, as defined by the Office of Management and Budget. This definition differs from the
Census Bureau’s definition of urbanized areas. For details, see U.S. Census Bureau, “2010 Urban Area FAQs,” https://
www.census.gov/programs-surveys/geography/about/faq/2010-urban-area-faq.html.
10
For example, see U.S. Census Bureau, “Household Pulse Survey Shows LGBT Adults More Likely to Report Living in
Households with Food and Economic Insecurity than Non-LGBT Respondents,” https://www.census.gov/library/stories/
2021/08/lgbt-community-harder-hit-by-economic-impact-of-pandemic.html.
11
Survey respondents could report their sexual orientation and gender identity on a demographic profile survey previously
conducted by the survey vendor. Respondents are classified as LGBTQ+ based on responses to these questions.
Figure 5. At least doing okay financially (by year, race/ethnicity, and parental status)
2021202020192018
Asian Hispanic BlackWhite
2021202020192018
Percent Percent
84
84
82
66
67
68
67
69
68
79
80
82
88
70
71
82
83
87
79
63
62
53
66
62
56
76 76
73
89
61
70
79
Parents All other adults
Note: Among all adults. Parents are adults living with their own children under age 18.
Overall Financial Well-Being 11
orientation as something other than straight,
gay, lesbian, or bisexual, were doing at least
okay financially.
12
Finally, 60 percent of adults with a disability
were doing at least okay financially, markedly
lower than the overall population.
13
Prior to
2021, the SHED did not include disability
status, so we cannot observe how financial
well-being has evolved for adults with a dis-
ability through the pandemic. However, as dis-
cussed in the “Employment” section of this
report, other surveys find evidence of an
increase in employment among adults with a
disability in recent years.
Changes in Financial Situation
over Time
The survey also tracks overall financial well-
being by asking respondents whether they are
better or worse off financially than they were
12 months earlier. Measuring well-being in
this way helps track changes in perceived well-
being over time, as some individuals may feel
worse off financially than they were a year ear-
lier, for instance, even if they feel they are still
doing okay overall (or that their financial well-
being is improving even if they are still strug-
gling overall).
The share of adults who said they were worse off financially than a year earlier fell from 24 per-
cent in 2020 to 20 percent in 2021, yet remained much higher than the 14 percent seen in 2019,
12
Differences in financial well-being between adults identifying as LGBTQ+ and other adults were present across age
groups. For example, only 58 percent of LGBTQ+ adults ages 45 to 54 were doing at least okay, compared with 75 per-
cent among all adults in that age group.
13
Disability status is defined based on a five-question functional limitation sequence that asks about hearing, vision,
ambulatory, self-care, and independent living difficulties. This approach for determining disability status is similar to the
six-question sequence used for the American Community Survey (see U.S. Census Bureau, “How Disability Data Are Col-
lected from the American Community Survey,” https://www.census.gov/topics/health/disability/guidance/data-
collection-acs.html).
Table 1. At least doing okay financially (by
demographic characteristics)
Percent
Characteristic 2021
1-year
change
5-year
change
Family income
Less than $25,000 55 3 8
$25,000–$49,999 67 2 5
$50,000–$99,999 85 1 5
$100,000 or more 96 1 4
Disability status
Disability 60 n/a n/a
No disability 81 n/a n/a
LGBTQ+ status
Identifies as LGBTQ+ 67 -1 n/a
Does not identify as LGBTQ+ 79 2 n/a
Marital status
Married 86 4 9
Not married 67 0 6
Place of residence
Metro area 79 3 9
Non-metro area 72 3 4
Neighborhood income
Low or moderate income 66 4 6
Middle or upper income 82 2 9
Overall 78 3 8
Note: Among all adults. Low- or moderate-income neighbor-
hoods are defined here using the definition from the Community
Reinvestment Act. LGBTQ+ status was first identifiable in the
2019 survey and disability status was first identifiable in the
2021 survey. Here and in subsequent tables and figures, per-
centages may not sum to 100 because of rounding.
n/a Not applicable.
12 Economic Well-Being of U.S. Households in 2021
before the pandemic (figure 6). The share doing about the same as a year earlier increased 3 per-
centage points to 54 percent, while the share who said they were better off was unchanged at
25 percent.
When asked to compare their financial situation to two years ago, before the pandemic, nearly
one-fourth (24 percent) said they were worse off. Forty percent said they were doing about the
same, and 36 percent said they were better off than two years ago.
14
Those who were doing
worse off than before the pandemic were disproportionately adults with lower family income and
less education.
To get a longer-term perspective, individuals were also asked to compare their current financial cir-
cumstances to how they perceived their parents’ financial situation at the same age. Looking
across a generation shows evidence of economic progress over time, despite financial setbacks
during the pandemic. A majority of adults (57 percent) thought they were better off financially than
their parents were, up from 54 percent in 2020 and back to the pre-pandemic level from 2019.
Twenty-one percent thought they were worse off than their parents were at the same age.
14
A subset of respondents completed both the 2020 and 2021 surveys. Combining the one-year change in well-being
results in the 2020 and 2021 surveys for these repeat respondents leads to similar results. Twenty-four percent
reported that their financial well-being declined in one year and did not improve in the other, while 34 percent indicated
that their well-being improved in one year and did not decline in the other. The remaining 43 percent either said their
well-being was about the same in each year (35 percent) or had an improvement in one year and a decline in the other
(7 percent).
Figure 6. Financial situation compared with 12 months prior (by year)
Percent
Better off
Worse off
20212020201920182017201620152014
25
32
31
33
2727
29
21
19
17
15
13
14
25
20
24
Note: Among all adults.
Overall Financial Well-Being 13
People holding at least a bachelor’s degree were more likely to experience upward economic
mobility, relative to those with less education. This is particularly true among first-generation col-
lege graduates, among whom 70 percent thought they were better off financially than their par-
ents were.
15
Local and National Economic Conditions
Along with questions about their own financial circumstances, people were asked to rate their
local economy and the national economy as “excellent,” “good,” “only fair,” or “poor.” The share of
adults rating their local economy favorably increased from 2020 to 2021. Forty-eight percent of
adults rated their local economy as “good” or “excellent” in 2021, with the rest rating conditions
as “only fair” or “poor.” This share was up from 43 percent in 2020, but well below the 63 percent
of adults who rated their local economy as “good” or “excellent” in 2019, before the pandemic.
This pattern was generally similar across
racial and ethnic groups, with higher shares
rating their local economy favorably relative to
2020 but still below the share from 2019
(table 2). One exception was Black adults: the
share of Black adults rating their local
economy favorably increased 10 percentage
points from 2020 to 2021, and was much
closer to the pre-pandemic level than for other
groups. However, Black and Hispanic adults
remained the least likely to report that their
local economy was faring well.
People’s perceptions about their local
economy diverged in 2021 for metro and non-
metro areas. While perceptions of the local economy improved for residents of metro areas, per-
ceptions ticked down 1 percentage point for those in non-metro areas. Additionally, the 34 percent
of non-metro residents who rated their local economy as “good” or “excellent” remained far below
the 53 percent that did so in 2019.
15
First-generation college graduates are those who have at least a bachelor’s degree and who report that neither of their
parents completed at least a bachelor’s degree.
Table 2. Self-assessment of the local economy
as good or excellent (by race/ethnicity and
place of residence)
Percent
Characteristic 2019 2020 2021
Race/ethnicity
White 67 46 50
Black 46 32 42
Hispanic 57 39 44
Asian 72 44 61
Place of residence
Metro area 65 44 50
Non-metro area 53 35 34
Note: Among all adults.
14 Economic Well-Being of U.S. Households in 2021
Similar to people’s perceptions of their local
economy, the share rating the national
economy favorably fell precipitously from 2019
to 2020, after the onset of the pandemic
(figure 7). However, people’s perceptions of
the national economy continued to decline in
2021. Only 24 percent of adults rated the
national economy as “good” or “excellent” in
2021, down 2 percentage points from 2020
and about half the rate seen in 2019. This
trend contrasts starkly with people’s increas-
ingly favorable assessment of their own finan-
cial well-being.
Overall Life Satisfaction
In addition to questions on financial well-
being, the 2021 survey included a question on
overall life satisfaction to provide a broader
look at how people were faring. Respondents rated how satisfied they were with life as a whole on
a scale from 0 to 10. Fifty-eight percent of adults reported “high” life satisfaction (rating 7 to 10),
30 percent reported “medium” life satisfaction (rating 4, 5, or 6), and 11 percent reported “low”
life satisfaction (rating 0 to 3).
Life satisfaction was strongly associated with income. Nearly three-fourths (73 percent) of adults
with family income of $100,000 or more reported high life satisfaction, compared with 41 percent
among those with family income less than $25,000. Differences by education were also large, as
were those by disability status and LGBTQ+ status (table 3).
16
Differences in overall life satisfaction by race/ethnicity, on the other hand, were small. The shares
of White, Black, and Hispanic adults reporting high life satisfaction were all within 2 percentage
points of the share doing so for the overall population. Asian adults exhibited the largest differ-
ence from the overall population, with 63 percent reporting high life satisfaction.
16
The scales used to measure life satisfaction and financial well-being are not directly comparable.
Figure 7. Assessment of own financial
well-being, local economy, and national
economy (by year)
78
757575
74
41
51
50
26
24
57
64
63
43
48
Own financial
well-being
(doing at least okay)
Local
economy
(good or excellent)
National
economy
(good or excellent)
20212020201920182017
Percent
Note: Among all adults.
Overall Financial Well-Being 15
Table 3. Share of adults with high life satisfaction (by demographic characteristics)
Characteristic Percent
Family income
Less than $25,000 41
$25,000–$49,999 51
$50,000–$99,999 61
$100,000 or more 73
Education
Less than a high school degree 40
High school degree or GED 53
Some college/technical or associate degree 55
Bachelor’s degree or more 67
Race/ethnicity
White 59
Black 56
Hispanic 57
Asian 63
Disability status
Disability 41
No disability 62
LGBTQ+
Identifies as LGBTQ+ 46
Does not identify as LGBTQ+ 60
Parental status
Not living with own children under age 18 57
Parent (living with own children under age 18) 62
Place of residence
Metro area 59
Non-metro area 53
Overall 58
Note: Among all adults.
16 Economic Well-Being of U.S. Households in 2021
Income
Income is central to most people’s financial well-being. Recognizing this, the survey included a
series of questions on income level and sources, as well as monthly income volatility.
Most parents with a child under age 18 received additional income in 2021 from the CTC. Parents
used these monthly payments in a variety of ways, including saving them, spending them on food,
and spending them on rent, mortgage, and utilities.
Most adults had income that was roughly the same each month. For adults with varying monthly
income, 3 in 10 reported that the volatility caused financial challenges. Income variability was
more likely to result in financial challenges among those with lower income.
Level and Source
Family income in this survey is the cash
income from all sources, before taxes and
deductions, that the respondents and their
spouse or partner received during the pre-
vious year. Income is reported in dollar ranges
and not exact amounts. Just over one-fourth
of adults had a family income below $25,000
in 2021, and nearly one-third had $100,000
or more (table 4).
Family income varied dramatically by race and ethnicity in 2021. Forty-three percent of Black
adults and 40 percent of Hispanic adults had a family income below $25,000. This is at least
twice the rate among White and Asian adults. Conversely, White and Asian adults were dispropor-
tionately likely to have family income above $100,000.
Labor earnings were the most common source of income, but many people had other sources of
income as well. Sixty-seven percent of adults and their spouse or partner received wages,
Table 4. Family income (by race/ethnicity)
Percent
Race/
ethnicity
Less than
$25,000
$25,000–
$49,999
$50,000–
$99,999
$100,000
or more
White 20 16 27 37
Black 43 18 22 17
Hispanic 40 18 22 20
Asian 17 10 23 51
Overall 26 16 25 32
Note: Among all adults.
17
salaries, or self-employment income (collectively referred to here as labor income) (figure 8),
matching the share from 2020. Yet, 59 percent of adults and their spouse or partner received non-
labor income in 2021.
17
The share of adults receiving unemployment income in 2021 (9 percent) remained higher than
before the pandemic but was lower than in 2020, when the job losses caused by the COVID-19
pandemic peaked. Individuals who received income from unemployment insurance in 2021
reported that they were most likely to learn about their eligibility from their employer (45 percent),
followed by their own internet research (32 percent).
Assistance from nonprofits and private sources—including financial support from a friend or family
member living outside of their home—can also supplement family income. Fifteen percent of
adults ages 21 and older received at least one type of assistance from private or nonprofit
17
Non-labor income is defined as income from interest, dividends, or rental income; social security (including old age and
Disability Insurance (DI)); Supplemental Security Income (SSI), Temporary Assistance for Needy Families (TANF), or cash
assistance from a welfare program; unemployment income; or income from a pension. Non-labor income does not
include Economic Impact Payments, tax credits such as the Earned Income Tax Credit, or in-kind benefits.
Figure 8. Family income sources
2021
2020
Any non-labor income
Unemployment income
SSI, TANF, or cash assistance
from a welfare program
Social security
(including old age and DI)
Interest, dividends,
or rental income
Non-labor income
Wages, salaries, or
self-employment income
Percent
67
67
32
33
27
27
5
5
14
9
61
59
Note: Among all adults. Respondents could select multiple answers. DI is Disability Insurance; SSI is Supplemental
Security Income; and TANF is Temporary Assistance for Needy Families. Key identifies bars in order from top to bottom.
18 Economic Well-Being of U.S. Households in 2021
sources in 2021 (table 5). Nearly 1 in 10 adults received groceries or meals from a food pantry,
religious organization, or community organization, down 2 percentage points from 2020.
Seven percent of adults ages 21 and older received financial assistance from a friend or family
member living outside of their home, essentially unchanged from 2020. On the other hand, 15 per-
cent of adults provided support to others.
Adults with less education were more likely to receive at least one type of assistance from private
or nonprofit sources. More than 3 in 10 (34 percent) adults with less than a high school degree
received this type of assistance, compared with less than 1 in 10 adults with at least a bach-
elor’s degree.
Child Tax Credit
Starting in July 2021, most parents of children under age 18 saw their income supplemented by
the enhanced CTC. Eighteen percent of all adults, and 70 percent of adults living with their chil-
dren under age 18, reported receiving monthly CTC payments in 2021.
18
An additional 5 percent
of adults living with children under age 18 did not know if they received monthly CTC payments.
Parents who received monthly CTC payments most frequently saved the payments, spent them on
their child, or used them for necessities. Saving was the most common use of the monthly CTC
payments, with 43 percent of recipients saying they saved at least a portion of them.
18
This estimate corresponds to 27 million payments to 52 million children. Administrative data from the Treasury Depart-
ment find that 36 million payments were made in December 2021 for 61 million qualifying children (U.S. Department of
the Treasury, By State: Advance Child Tax Credit Payments Distributed in December 2021 (Washington, DC: Department of
the Treasury, December 2021), https://home.treasury.gov/system/files/131/Advance-CTC-Payments-Disbursed-
December-2021-by-State-12152021.pdf). The lower estimate of parents reporting the credit in the SHED suggests that
some parents either did not know their family received the payment or did not know that it was the CTC.
Table 5. Financial assistance received (by educational attainment)
Percent
Characteristic
Free groceries
or meals
Financial support
from religious
or community
organization
Financial support
from friends or
family not
in household
Received at least
one type of private
or nonprofit
support
Less than a high school degree 27 5 15 34
High school degree or GED 13 2 7 17
Some college/technical or associate degree 10 2 9 17
Bachelor’s degree or more 3 1 5 8
Overall 9 2 7 15
Note: Among adults age 21 and older. Respondents could select multiple answers.
Income 19
Other common uses were spending on their
child (40 percent); spending on food (31 per-
cent); and spending on rent, mortgage, or utili-
ties (29 percent) (table 6).
Respondents were also asked how they used
the largest portion of the monthly CTC pay-
ments. Saving the monthly payment was again
the most common response (36 percent), with
many others saying that they spent the largest
portion on their child or on rent, mortgage, or
utilities.
The ways people used the CTC payments varied by income. Higher-income adults were most likely
to save the largest portion of their credit, whereas lower-income adults were most likely to spend it
on housing. For instance, 54 percent of recipients with income of at least $100,000 saved the
largest portion of their credit, whereas only 18 percent of recipients with income less than
$25,000 did so (table 7).
Income Variability
Since many bills must be paid monthly, variations in monthly income can lead to financial chal-
lenges. Most adults had income that was roughly the same each month, but about 3 in 10 had
income that varied from month to month. This share was essentially unchanged from 2020.
Since income variability can result from either dips or spikes in monthly income, the survey asked
those who reported varying monthly income whether they struggled to pay bills as a result.
Thirty percent of those who experienced varying monthly income, representing slightly less than
Table 6. Uses for Child Tax Credit (CTC)
Percent
Purpose
Used any
for purpose
Used largest
portion
for purpose
Saved it 43 36
Paid off debt 21 10
Spent on child 40 20
Spent on rent, mortgage, or utilities 29 17
Spent on food 31 12
Spent on other things 12 5
Note: Among parents with a child under age 18 who reported
receiving CTC payments. Respondents could select multiple
answers.
Table 7. Use for largest portion of Child Tax Credit (by family income)
Percent
Purpose
Less than
$25,000
$25,000–
$49,999
$50,000–
$99,999
$100,000
or more
Saved it 18 15 33 54
Paid off debt 14 9 9 10
Spent on child 22 22 22 16
Spent on rent, mortgage, or utilities 29 33 18 6
Spent on food 13 19 15 7
Spent on other things 5 2 5 7
Note: Among parents with a child under age 18 who reported receiving CTC payments.
20 Economic Well-Being of U.S. Households in 2021
1 in 10 adults overall, said they struggled
to pay their bills at least once in the
past 12 months because of varying
monthly income.
Lower-income adults were more likely to have
varying monthly income and to report that they
struggled to pay their bills at least once in the
past 12 months as a result (table 8). Fif-
teen percent of adults with income less than
$50,000 struggled to pay their bills because
of varying monthly income. Among lower-
income parents, an even greater 27 percent
struggled to pay their bills because of income
variability.
Adults with and without a disability were simi-
larly likely to experience income variability.
However, when those with a disability experi-
enced such variability, it was more likely to
lead to a hardship.
Income variability also continued to differ greatly by industry in 2021. Workers in the leisure and
hospitality industry were the most likely to have varying monthly income (figure 9).
Table 8. Income volatility and related hardship
(by family income, race/ethnicity, and disability
status)
Percent
Characteristic
Varying
income,
causes
hardship
Varying
income,
no hardship
Stable
income
Family income
Less than $25,000 16 23 61
$25,000–$49,999 14 17 69
$50,000–$99,999 6 18 75
$100,000 or more 1 19 79
Race/ethnicity
White 7 19 75
Black 12 19 69
Hispanic 14 23 63
Asian 4 20 76
Disability status
No disability 7 20 72
Disability 13 15 72
Overall 9 20 72
Note: Among all adults.
Figure 9. Income varied at least occasionally from month to month (by industry)
Public administration
Armed forces
Professional and business services
Financial activities
Manufacturing
Information
Education and health services
Transportation and utilities
Other services
Construction
Wholesale and retail trade
Natural resources and mining
Leisure and hospitality
Percent
47
42
39
37
35
34
27
27
27
24
22
22
21
Note: Among adults who reported industry of employment.
Income 21
These workers also reported the highest rates of hardship because of their varying income. How-
ever, the prevalence of income variability within the leisure and hospitality industry was similarly
high both before and after the pandemic.
22 Economic Well-Being of U.S. Households in 2021
Employment
The share of adults who were working in late 2021 remained below the pre-pandemic level. Health
limitations, concerns about COVID-19, and family responsibilities were common reasons for not
working. Many also switched jobs in 2021, and those who did generally said their new job was
better than their old one.
Reasons for Not Working
Twenty-three percent of prime-age adults (ages 25 to 54) were not working in October 2021, down
from 26 percent in 2020, but up from 21 percent in 2019, before the pandemic.
19
Health limitations and concerns about getting
COVID-19 were commonly cited reasons for not
working. Twelve percent of all prime-age adults
were not working, at least in part, for one of
these reasons. Health limitations or disability
were cited by 9 percent and specific concerns
about COVID-19 were cited by 7 percent
(figure 10). Family responsibilities were also
commonly cited as reasons for not working.
The proportion of prime-age adults who said
that they were not working because they could
not find work fell from 9 percent in 2020 back
to 5 percent in 2021, the same as before the
pandemic.
20
Two percent of all prime-age adults said that
they were not working, at least in part,
because they didn’t want to lose access to unemployment insurance or other government ben-
19
This pattern is consistent with that observed by the Bureau of Labor Statistics, who reported 22 percent not working in
October 2021, down from 24 percent not working at the time of the survey in 2020, but up from 20 percent in October
2019. See U.S. Bureau of Labor Statistics, “(Seas) Employment-Population Ratio—25–54 yrs.,” https://data.bls.gov/
timeseries/LNS12300060.
20
Some of the decrease could have been due to a change in the questionnaire from 2020 to 2021 to include an addi-
tional reason that respondents could give for why they were not working, although respondents still could give mul-
tiple answers.
Figure 10. Reasons for not working among
prime-age adults
1
1
2
4
5
7
9
7
Percent
Retired
School or training
Access to government
benefits
Childcare
Could not find work
Family obligations
besides childcare
Concerned about
COVID-19
Health limitations or
disability
Note: Among adults ages 25 to 54. Respondents
could select multiple answers.
23
efits.
21
Among those whose family received government benefits in the prior year, 6 percent indi-
cated that benefit eligibility contributed to them not working.
22
However, even among benefit recipi-
ents, other factors including concerns about COVID-19 exposure (15 percent) and an inability to
find work (11 percent) were more likely to be cited as reasons for not working.
Prime-age women were particularly likely to say that they were not working because of childcare
and other family responsibilities. Six percent of prime-age women cited childcare as a reason for
not working, and 10 percent cited other family responsibilities, far exceeding that for men
(figure 11). However, the share of women not working because of childcare responsibilities did not
increase relative to that seen before the pandemic.
Another difference between prime-age men and women is that women were more likely to say they
were not working, at least in part, because of concerns about getting COVID-19. Eight percent of
21
The survey was conducted in the fourth quarter of 2021, after the expiration of expanded unemployment insurance pro-
grams in September.
22
Government benefit recipients include prime-age adults whose family received unemployment insurance, Social Security,
Supplemental Security Income, TANF, other cash welfare assistance, SNAP benefits, Medicare, or Medicaid.
Figure 11. Reasons for not working among prime-age adults (by gender)
Women
Men
Retired
School or training
Access to government
benefits
Childcare
Could not find work
Family obligations
besides childcare
Concerned about
COVID-19
Health limitations
or disability
Percent
8
8
10
5
5
5
5
1
1
1
6
2
2
2
2
9
Note: Among adults ages 25 to 54. Respondents could select multiple answers. Key identifies bars in order from top
to bottom.
24 Economic Well-Being of U.S. Households in 2021
prime-age women cited concern about getting the virus as a reason for not working, compared with
5 percent of prime-age men. Previous studies have found that occupations with more women
working in them before the pandemic had higher rates of COVID-19 exposure.
23
Working from Home
A major change in many people’s work lives since 2019 was the increased prevalence of working
from home, also known as remote work. During the week of the survey in late 2021, 22 percent of
adults who worked for someone else (“employees”) worked entirely from home. This share was
down from 29 percent in 2020, but well above the 7 percent who worked entirely from home in
2019, before the pandemic.
24
In addition, 17 percent of employees said they worked from home
some of the time during the week of the survey in 2021. For some workers, such as those with
disabilities, the increased prevalence of remote work in recent years may have facilitated greater
participation in the labor market (see box 1).
Employees with more education were much
more likely to work from home than those with
less education. Thirty-three percent of
employees with at least a bachelor’s degree
worked entirely from home, whereas 14 per-
cent of employees with some college, and
9 percent with a high school degree or less,
did so (figure 12).
25
One reason for the differences by education is
that employees with more education were
more likely to have a job where they could
work from home. Nearly three-fourths of
employees with a bachelor’s degree or more
either worked from home or said that they could if their employer would let them, compared with
29 percent of employees with a high school degree or less. Overall, 53 percent of employees
either worked from home or said that they could if their employer would let them.
23
Stefania Albanesi and Jiyeon Kim, “Effects of the COVID-19 Recession on the U.S. Labor Market: Occupation, Family,
and Gender,” Journal of Economic Perspectives 35, no. 3 (Summer 2021): 3–24, https://www.aeaweb.org/
articles?id=10.1257/jep.35.3.3.
24
The question asked in 2019 was different from 2020 and 2021. The 2019 survey asked where people worked in their
main jobs most of the time.
25
There is variation across industry in the likelihood of working from home, although even within an industry those with a
bachelor’s degree are generally more likely to work from home than are those with less education.
Figure 12. Amount of work done from home
(by education)
AllSomeNone
Bachelor’s degree
or more
Some college/
technical or
associate degree
High school
degree or less
81
9 9
14
33
11
24
75
43
Percent
Note: Among adults who worked for someone else.
Key identifies bars in order from left to right.
Employment 25
Box 1. Pandemic Employment Experiences of Adults with a
Disability
Individuals with a disability have long faced significant barriers in the labor market. Before the pan-
demic, in February 2020, the employment rate among individuals with a disability was only 19 percent,
compared with 67 percent among those without a disability. Most of this gap was due to individuals
with a disability being out of the labor force, meaning they were not looking for work.
1
The widespread
disruptions to the labor market during the pandemic created unique challenges for adults with disabili-
ties, but the restructuring of how work is conducted also created an opportunity for some to enter the
labor force and contribute in ways that were not previously possible.
Despite the additional pandemic-related hurdles, employers’ increased reliance on working from home
and remote work had the potential to expand employment opportunities for a wide array of workers with
a disability. Thirty-three percent of workers with a disability who had less than a bachelor’s degree
worked from home some of the time in the week before the survey, compared with 25 percent of their
peers without a disability (figure A). Among workers with at least a bachelor’s degree, a higher share
worked from home, and the shares were similar among workers with and without a disability. The
increase in remote work opportunities during the pandemic may have contributed to the more rapid
recovery in employment rates since the start of the pandemic among workers with disabilities observed
in other data.
2
Bachelor’s degree
or more
No disabilityDisability
Figure A. Worked from home at least some of the time (by education and disability status)
Note: Among adults who worked in the month prior to the survey. Key identifies bars in order from top to bottom.
Less than a
bachelor’s degree
Percent
33
25
60
58
(continued)
1
U.S. Bureau of Labor Statistics “(Unadj) Employment-Population Ratio—With a Disability, 16 Years and Over” https://data.bls.gov/
timeseries/LNU02374597; and U.S. Bureau of Labor Statistics, “(Unadj) Employment-Population Ratio—With No Disability, 16 Years
and Over,” https://data.bls.gov/timeseries/LNU02374593.
2
According to the U.S. Bureau of Labor Statistics, the employment to population ratio of workers with a disability in October 2021 was
1 percentage point above pre-pandemic levels from February 2020 while remaining 2 percentage points below pre-pandemic levels
among those without a disability.
26 Economic Well-Being of U.S. Households in 2021
Most employees who worked from home, or who said they could if their employer would let them,
would prefer to work from home. Eighty-nine percent said they would like to work from home at
least some of the time. Forty-one percent said
they would prefer to do so all of the time.
Employees who preferred to work from home
at least some of the time most commonly
cited less time commuting and work-life bal-
ance as reasons (figure 13). Both were cited
by 89 percent of employees who preferred to
work from home.
Another common reason employees said they
preferred to work from home was increased
productivity. Seventy percent of employees
who preferred to work from home said that
one reason for their preference was that they
were more productive at home. While lower,
Figure 13. Reasons employees prefer to work
from home
Percent
Able to live in a
different area
Work–life balance
Concerns about
COVID-19
More productive
working at home
Less time commuting
89
70
56
89
47
Note: Among employees who worked from home or
could work from home if their employer allowed, who
also said they preferred working from home at least
some of the time. Respondents could select multiple
answers.
Box 1—continued
Nevertheless, with the unprecedented health challenges that the pandemic presented to the general
population, adults with a disability faced significant difficulties that may be exacerbated by their dis-
ability. For some types of disabilities, contracting COVID-19 may present unique difficulties in going
about one’s day or receiving care. Further compounding matters, comorbidities are more prevalent
among adults with a disability, meaning a COVID-19 infection may be more likely to result in serious ill-
ness or death.
3
Consequently, workers with a disability may have different preferences on workplace COVID-19 precau-
tions. This, in turn, could affect their employment decisions. After controlling for their level of educa-
tion, employees with a disability were more likely to favor a vaccine mandate relative to those without a
disability.
Additionally, among those who were not working, people with a disability were more likely to say that
concerns about contracting COVID-19 contributed to their not working. Among adults with at least a
bachelor’s degree, 19 percent of those with a disability reported that COVID-19 concerns were a con-
tributing factor. This is nearly twice the frequency of these concerns among similarly educated people
without a disability. Among those with less than a bachelor’s degree, just over one-fifth of nonworkers
with a disability, and just under one-fifth of nonworkers without a disability, said that COVID-19 concerns
were a factor in their employment decision. Consequently, while employment rates among workers with
a disability have improved recently, health and safety concerns appear to be hampering their employ-
ment growth more than among other adults.
3
Sally-Ann Cooper, Gary McLean, Bruce Guthrie, Alex McConnachie, Stewart Mercer, Frank Sullivan, and Jull Morrison, “Multiple
Physical and Mental Health Comorbidity in Adults with Intellectual Disabilities: Population-Based Cross-Sectional Analysis,” BMC
Family Practice, 16, no. 1 (2015): 1–11.
Employment 27
56 percent of employees who preferred to work from home said that concerns about COVID-19
contributed to their preference.
Many of those who worked from home at least some of the time also said that they would actively
look for another job if their employer required them to work in person each workday. Forty-five per-
cent said they were at least somewhat likely to look for another job or leave their job if their
employer required them to work in person each workday.
26
Twenty-two percent were very likely.
Among those working from home full time, an even larger 55 percent said that they would be
somewhat or very likely to look for another job if required to report in person each workday.
For context on the importance of the ability to work from home in people’s job decisions, the
survey also asked respondents about their likelihood of looking for another job if their employer
froze their pay or cut their pay by various amounts.
27
The share of employees who were at least
somewhat likely to look for another job if their employer required they work in person was similar
to the share who would look after a pay freeze (figure 14).
Job Changes
In the fourth quarter of 2021, 15 percent of workers said they were in a different job than they
were 12 months earlier. Most people who changed jobs said that their new job was better than
26
However, 16 percent of those working from home who said that they would prefer to work in person also said that they
would actively look for another job in this situation. This suggests that at least some of these individuals either were
actively looking irrespective of the work location, or they value being given the choice of where to work even if they
chose not to work from home.
27
The order of the pay cut and telework questions was randomized in the survey, as was the amount of the pay cut that
respondents were asked about.
Figure 14. Likelihood of looking for a new job (by required in-person work and pay cuts)
Percent
Not at all likelyNot that likelySomewhat likelyVery likely
Pay cut of 10 percent
Pay cut of 5 percent
Pay freeze
Report in person
22 23
23
18
32
13
29
32
32
40
39
20
8
45
16
7
Note: Among adults working for someone else who worked from home at least some of the time. Key identifies bars in
order from left to right.
28 Economic Well-Being of U.S. Households in 2021
their old one. Over 6 in 10 people who
changed jobs said their new job was better
overall, compared with 1 in 10 who said that it
was worse.
Pay, opportunities for advancement, and
interest in the work were frequently seen as
better in the new job. Just over half of people
who changed jobs said that their pay and ben-
efits improved, compared with 20 percent who
said their pay was worse. Similarly, far more
people said that their work-life balance, oppor-
tunities for advancement, and interest in the
work improved than said these measures
declined (figure 15).
Of the characteristics considered, COVID-19
policies and exposure were the most likely to
be the same in the old and new jobs. Nearly
two-thirds (64 percent) of job changers said that COVID-19 policies and exposure were about the
same, while one-fourth said that they were better, and 11 percent said that they were worse.
28
Workers who experienced a layoff and changed jobs were less positive about their new positions
than other job changers.
29
Those who were laid off were substantially less likely than those not
laid off to say their new job was better overall.
30
They also were less likely to say that the pay and
benefits, opportunities for advancement, and interest in the job improved at their new position
(figure 16).
COVID-19 Precautions at Work
One factor in people’s decisions on whether to work and where to work is their perceptions of
workplace safety, including COVID-19 precautions. Employees mostly thought that their employers
were taking the right amount of precautions to prevent the spread of COVID-19. Seventy-seven per-
cent of employees said their employers were taking the right amount of precautions. Those who
28
Better COVID-19 policies could mean stricter or more lenient policies, depending on the preference of the respondent.
The survey did not define what better or worse policies meant.
29
Seven percent of adults said that they were laid off in the 12 months prior to the survey.
30
People’s perceptions of their old jobs likely also reflect the circumstances of their leaving, including negative feelings
from getting laid off. Moreover, while it is likely the laid-off worker is comparing their new job to the job they were laid off
from, it is also possible they had another job in-between.
Figure 15. Change in quality of job characteris-
tics after job change
BetterAbout the sameWorse
Work–life
balance
COVID-19 policies
and exposure
Physical
demands
Interest in
the work
Opportunities
for advancement
Pay or
benefits
Overall
10 28
62
51
44
29
41
20
15
48
3913
31
5713
256411
4043
16
Percent
Note: Among working adults whose main job was not
the same as it was a year ago. Key identifies bars in
order from left to right.
Employment 29
did not were almost evenly split between thinking their employers were taking too many (11 per-
cent) and too few precautions (12 percent).
31
In terms of specific precautions, just under one-fourth of employees said that their employer had a
policy requiring vaccination. An additional 19 percent of employees said that employees could
either be vaccinated or be tested regularly. Forty-nine percent said their employer had no specific
vaccine or testing requirement.
32
Employees were almost evenly split on whether they wanted vaccine requirements in their work-
places. Forty-nine percent of workers said that they wanted their employer to require vaccinations
of all employees, whereas 51 percent said that they did not. The share who wanted a vaccine
requirement was higher (59 percent) among employees who had a COVID-19 vaccine themselves.
Just 4 percent of workers who were not vaccinated wanted their employer to require vaccines.
31
One important aspect, however, is that people who had left an employer over their level of precautions would not have
been asked this question about the job that they left. In these cases, an individual will be asked about the precautions
at their new job or, if not working, would not be asked the question at all. Nevertheless, most who changed jobs said
that their new job’s COVID-19 policies were no better or worse than at their previous job.
32
The remaining 8 percent did not know if there were any requirements.
Figure 16. Share reporting that job characteristic is better after job change (by layoff status)
Not laid off
Laid off
Work–life balance
COVID-19 policies
and exposure
Physical demands
Interest in the work
Opportunities for
advancement
Pay or benefits
Overall
Percent
43
67
37
32
35
55
47
52
29
25
24
36
42
31
Note: Among working adults whose main job was not the same as it was a year ago. Key identifies bars in order from
top to bottom.
30 Economic Well-Being of U.S. Households in 2021
Part-Time Jobs, Temporary Jobs, and Irregular Schedules
Thirteen percent of adults worked part time at their main job, and 5 percent said that their main
job was a temporary position. The share of adults who were working part time declined 1 per-
centage point from 2020 to 2021. Part-time work was more common among women than among
men. Sixteen percent of women worked part time, while 10 percent of men did.
One indication of a stronger job market is that the share of part-time workers who said they
wanted to work more hours declined to 41 percent in 2021, from 51 percent in 2020. Differences
remained across the population, however. For example, a higher 58 percent of Hispanic part-time
workers said that they wanted more work.
People who had a part-time or temporary job reported more financial strain than people who
worked full time. Twenty-nine percent of part-time workers and 31 percent of temporary workers
said that they were either just getting by or finding it difficult to get by. A smaller 16 percent of per-
manent, full-time workers reported the same levels of financial strain.
Some employees also had irregular schedules, including 16 percent who had a work schedule that
varied based on their employer’s needs. These workers with irregular schedules that they did not
control tended to be under more financial strain. Twenty-seven percent of workers with a schedule
that varied based on their employer’s needs said that they were either just getting by or finding it
difficult to get by. This compares with 16 percent of workers with a fixed schedule or with a
schedule that they control.
The Gig Economy
Individuals who perform gig work or other gig activities may be contributing to the economy in ways
not observed through traditional employment measures. To understand this aspect of the
economy, including the effects of the gig economy on household finances, the survey includes a
series of questions about gig activities. Gig activities in this report include selling items at places
such as flea markets and garage sales or through online marketplaces, short-term rentals of
items or property, and freelance gig work such as ridesharing or other roles where people are paid
for specific tasks and generally have flexibility about when and how to work.
Overall, 16 percent of adults had performed gig activities over the prior month.
33
This includes
11 percent who sold things, 1 percent who offered short-term rentals, and 6 percent doing other
freelance or gig work (with some people performing more than one type of gig activity) (figure 17).
33
It is not possible to compare how frequently people did gig activities in 2021 with prior years because the gig economy
questions were revised substantially in 2021 to refine the definition of gig activities and to reduce respondent burden.
Employment 31
Most commonly, people sold things that they
owned for personal use, like clothing.
Eight percent of all adults sold a personal
item in the prior month. A smaller 3 percent
sold something that they purchased to resell,
and 2 percent sold something that they made.
Gig activities were typically not full-time jobs.
Sixty-four percent of those who performed gig
activities (10 percent of all adults) said they
spent less than 20 hours doing so over the
prior month. That said, those doing freelance
gig work were more likely than those selling or
renting items to spend at least 20 hours on it
over the prior month.
People performing gig activities often had another job. Over half of those performing gig activities
(54 percent) also had a job working for someone else. Even among those who spent at least
20 hours on gigs, 46 percent reported that they had a job working for someone else.
34
As a result, gig activities were rarely people’s main source of income. Only 2 percent of all adults
said they earned more than half of their income from gigs over the prior month. An even lower
1 percent of all adults said that they earned at least 90 percent of their income from gig activities.
People with lower financial well-being were
more likely to perform gig activities than those
who were faring better financially. One-fourth
of people who found it difficult to get by finan-
cially did gig activities, compared with 13 per-
cent of those who were living comfortably
(figure 18). At the same time, however, people
who performed gig activities were more likely
to say that they did it by choice (71 percent)
than out of necessity (29 percent).
People who did freelance gig work also varied
in how they felt their pay compared with what
34
Gig questions were asked separately from the standard employment questions. One percent of all adults said that they
were both not employed and spending at least 20 hours on gig activities in the prior month.
Figure 17. Gig activities performed
16
Percent
Any gig activity
Freelance or
gig work
Offering
short-term rentals
Selling items
11
1
6
Note: Among all adults. Respondents could select mul-
tiple answers.
Figure 18. Share performing gig activities (by
financial well-being)
Percent
Living comfortably
Doing okay
Just getting by
Hard to get by
25
17
16
13
Note: Among all adults.
32 Economic Well-Being of U.S. Households in 2021
they could earn from a more traditional job. Thirty-three percent of people who did freelance gig
work said that they earned more doing gig work than they could in a traditional job, while 39 per-
cent said they earned less. The remaining 28 percent said they earned about the same amount as
they could in a traditional job.
Employment 33
Dealing with Unexpected Expenses
The overall share of adults who would cover a small emergency expense using cash or its equiva-
lent increased to the highest level since 2013, when the survey began. Despite this positive trend,
many still faced difficulty paying monthly bills. Black and Hispanic adults, as well as adults with
lower income, disproportionately faced such challenges.
Small, Unexpected Expenses
Relatively small, unexpected expenses, such as a car repair or a modest medical bill, can be a
hardship for many families. When faced with a hypothetical expense of $400, 68 percent of all
adults in 2021 said they would have covered it exclusively using cash, savings, or a credit card
paid off at the next statement (referred to, altogether, as “cash or its equivalent”).
35
The
remainder said they would have paid by borrowing or selling something, or said they would not
have been able to cover the expense.
The share who would pay using cash or its
equivalent was up 4 percentage points from
2020 and was at the highest level since the
survey began in 2013 (figure 19). This
increase is consistent with the results on
overall financial well-being and may reflect
improving economic conditions and the addi-
tional COVID-19 relief measures enacted
in 2021.
Like the results for overall financial well-being,
parents saw a sharp increase in the share
who would cover a $400 expense with cash or
its equivalent—up from 56 percent in 2020 to
64 percent in 2021. Those not living with their
own children under age 18 saw a smaller
increase of 3 percentage points. One reason
that parents experienced this sharp increase
35
However, some who would not have paid with cash or its equivalent likely still had access to $400 in cash. Instead of
using that cash to pay for the expense, they may have chosen to preserve their cash as a buffer for other expenses (See
box 3 from the Report on the Economic Well-Being of U.S. Households in 2019, Featuring Supplemental Data from
April 2020 at https://www.federalreserve.gov/publications/files/2019-report-economic-well-being-us-households-
202005.pdf.)
Figure 19. Would cover a $400 emergency
expense completely using cash or its
equivalent (by year)
202120202019201820172016201520142013
50
53
54
56
59
61
63
64
68
Percent
Note: Among all adults.
35
may be the expansion of the CTC. The most
common way parents used their CTC pay-
ments was saving them, potentially improving
their ability to handle unexpected expenses.
36
Those who would not have covered a $400
expense completely with cash or its equivalent
(32 percent of adults) may have found it more
difficult to handle small, unexpected
expenses. For these adults, the most common
approach was to use a credit card and then
carry a balance, although many indicated they
would use multiple approaches (figure 20).
Eleven percent of all adults said they would be
unable to pay the expense by any means,
similar to the 12 percent seen in 2020.
To understand more about covering household expenses, the survey asked about adults’ ability to
pay their monthly bills. As of October and November 2021, 24 percent of adults indicated that
they had, or were close to having, difficulty paying bills for that month: 14 percent of adults had
one or more bills that they were unable to pay in full, and an additional 10 percent said they would
have been unable to pay their bills if faced with a $400 expense. The 24 percent having difficulty
(or close to having difficulty) paying bills was down 3 percentage points from 2020 and down
4 percentage points from 2019. These declines are consistent with improvements seen in overall
financial well-being.
37
Lower-income adults were especially likely to face difficulty paying bills. Half of adults with a family
income less than $25,000 had one or more bills that they were unable to pay in full that month or
were one $400 financial setback away from being unable to pay them, compared with 5 percent
for adults with a family income of $100,000 or more.
Black and Hispanic adults were much more likely than White or Asian adults to face difficulty
paying bills, and these differences were present at all income levels (figure 21). Forty percent of
Black adults and 35 percent of Hispanic adults had, or were close to having, difficulty paying bills,
compared with 19 percent of White adults and 11 percent of Asian adults.
36
More detail on the CTC is available in the “Income” section of this report.
37
Hispanic adults and parents were two groups that experienced particularly large declines in the share facing difficulty
paying bills, similar to results for financial well-being. (See the “Overall Financial Well-Being” section for results on finan-
cial well-being among Hispanics and parents.)
Figure 20. Other ways individuals would cover
a $400 emergency expense
Percent
Would not be able to pay
for the expense right now
Use a payday loan,
deposit advance,
or overdraft
Use money from a bank
loan or line of credit
Sell something
Borrow from a friend
or family member
Put it on a credit card
and pay it off over time
8
6
2
1
11
14
Note: Among all adults. Respondents could select mul-
tiple answers.
36 Economic Well-Being of U.S. Households in 2021
Looking across income levels shows that, even among adults with a family income of $100,000 or
more, Black and Hispanic adults were more than twice as likely as White or Asian adults to face
these challenges. Several interrelated factors, including discrimination or differences in credit
access, could have contributed to the differences by race and ethnicity. (See box 2 on “Racial and
Ethnic Discrimination” in the report Economic Well-Being of U.S Households in 2020 for a discus-
sion of discrimination, and the “Banking and Credit” section for differences in credit access.)
38
Some financial challenges, such as a job loss, require more financial resiliency than would an
unexpected $400 expense. One common measure of financial resiliency is whether people have
savings sufficient to cover three months of expenses if they lost their primary source of income. In
2021, nearly 60 percent of people said they had set aside money specifically as emergency sav-
ings or “rainy day” funds, the highest share since the survey began in 2013.
For those who did not set aside money for this purpose, some would have dealt with a loss of their
main source of income by borrowing, selling assets, or drawing on other savings. Fourteen percent
38
See Economic Well-Being of U.S Households in 2020, May 2021, https://www.federalreserve.gov/publications/files/
2020-report-economic-well-being-us-households-202105.pdf.
Figure 21. Not able to fully pay current month’s bills (by income and race/ethnicity)
After a $400 emergency expense
Currently
Asian
Hispanic
Black
White
$100,000 or more
Asian
Hispanic
Black
White
$50,000–$99,999
Asian
Hispanic
Black
White
Less than $50,000
Percent
21 17
33
20
21
9
7
11
9
3
2
3
2
1
27
23
7
16
12
4
3
8
7
1
Note: Among all adults. Key identifies bars in order from left to right.
Dealing with Unexpected Expenses 37
of adults said that they could have covered three months of expenses in this way. Twenty-
seven percent of adults indicated they could not cover three months of expenses by any means.
Health-Care Expenses
Out-of-pocket spending for health care is a common unexpected expense that can be a substantial
hardship for those without a financial cushion. As with the financial setbacks discussed earlier,
many adults were not financially prepared for health-related costs at the time of the survey.
Twenty percent of adults had major, unexpected medical expenses in the prior 12 months, with the
median amount between $1,000 and $1,999. Fifteen percent of adults had debt from their own
medical care or that of a family member (not necessarily from the past year).
Many went without medical care because they
could not afford it. Twenty-four percent of
adults went without some form of medical
care because they could not pay, ticking up
from 23 percent in 2020 but well below the
32 percent reported in 2013. Dental care was
the most frequently skipped, followed by vis-
iting a doctor (figure 22). Some people also
reported skipping prescription medicine,
follow-up care, or mental health visits.
The likelihood of skipping medical care
because of cost was strongly related to family
income. Among those with family income less
than $25,000, 38 percent went without some
medical care because they couldn’t afford it, compared with 9 percent of adults making
$100,000 or more.
Ability to afford health care may contribute to the finding that, as family income rises, the likeli-
hood a person reported being in good health increases substantially. Among those in families with
income less than $25,000, 75 percent reported being in good health, compared with 92 percent
for those in families with income of $100,000 or more.
Health insurance is one way that people can pay for routine medical expenses and protect against
the financial burden of large, unexpected expenses. In 2021, 91 percent of adults had health
insurance, a slight uptick from 2020. Those without health insurance were nearly twice as likely to
Figure 22. Forms of medical treatment skipped
because of cost during 2021
13
17
8
8
8
24
Percent
Any treatment
Mental health care
or counseling
Follow-up care
Prescripton medicine
Seeing a doctor or
specialist
Dental care
Note: Among all adults. Respondents could select mul-
tiple answers.
38 Economic Well-Being of U.S. Households in 2021
forgo medical treatment because they couldn’t afford it. Among the uninsured, 40 percent went
without medical treatment because they couldn’t afford it, versus 22 percent among the insured.
Hardships from Natural Disasters
Those without a financial cushion may face unique and particularly severe challenges in the event
of natural disasters. Natural disasters may cause disruptions to people’s ability to work, damage
to or displacement from their home, or higher bills for heating or cooling. These consequences all
require some financial resources to manage, such as a rainy-day fund; access to credit; or support
from family, friends, or the local community.
Almost one in six adults (16 percent) were
directly affected by a natural disaster during
the prior 12 months, meaning that they experi-
enced one or more of the following five events
as the result of a natural disaster or severe
weather event: (1) an income loss or work dis-
ruption, (2) property damage, (3) a temporary
evacuation, (4) longer-term displacement from
home, or (5) the injury or death of a family
member or close friend. The two most
common ways that people were affected by
natural disasters were property damage,
which affected 8 percent of adults, and an
income loss or work disruption, which affected
6 percent. Four percent of people had a close
friend or relative who was injured or killed by a
natural disaster (figure 23).
The effects from natural disasters are not uniform across segments of society. Adults with lower
income or less education were more likely to be affected by natural disasters than those with
higher income (table 9). Nearly 2 in 10 of those with income below $50,000 reported any
disaster-related hardship.
Looking at specific hardships, people with an income of less than $50,000 were more than twice
as likely to experience an income loss or work disruption, longer-term displacement from home, or
the injury or death of a friend or relative than someone with an income of $100,000 or more.
Similar differences were observed across education levels. Black or Hispanic adults were also
more likely to be affected by natural disasters than White or Asian adults, both overall and within
specific income or education categories.
Figure 23. Disruptions from natural disasters in
the prior 12 months
Percent
Any disruption
Injury or death of
family or friend
Long-term displacement
Temporary evacuation
Property damage
Income disruption
8
2
1
4
16
6
Note: Among all adults. Respondents could select mul-
tiple answers.
Dealing with Unexpected Expenses 39
Overall, most people did not expect their risk
of experiencing a natural disaster to change in
the near future. Fifty-eight percent of adults
assessed their likelihood of experiencing a
natural disaster in five years as “about the
same” as it is currently. However, 37 percent
of adults expected their risk of experiencing a
natural disaster to be higher in five years,
including 11 percent of adults who assessed
their natural disaster risk to be “much higher.”
Less than 5 percent of adults expected their
risk of experiencing a natural disaster to
decrease.
In contrast to the overall population who
expected their risk from natural disasters
would be about the same in five years, most
people affected by a natural disaster in the
past year expected their risk of a natural
disaster to be higher in five years. This
includes 20 percent who expected their risk to
be “much higher” (figure 24).
Some people undertook mitigation activities
to reduce their risks from natural disasters.
These mitigation activities differed substan-
tially by homeownership status. Nineteen per-
cent of renters investigated other places to
live, which was about twice the rate of home-
owners. Conversely, 18 percent of home-
owners improved their property to mitigate the
risks of disasters (figure 25).
While the risks from natural disasters appear
higher for people with less education, these
individuals were less likely to engage in
natural-disaster mitigation activities
(table 10). Twenty-two percent of renters with
a bachelor’s degree or more investigated other
Table 9. Disruptions from natural disasters (by
demographic characteristics)
Characteristic Percent
Family income
Less than $25,000 20
$25,000–$49,999 18
$50,000–$99,999 14
$100,000 or more 13
Education
Less than a high school degree 23
High school degree or GED 15
Some college/technical or associate degree 17
Bachelor’s degree or more 14
Race/ethnicity
White 14
Black 19
Hispanic 21
Asian 15
Age
18–29 17
30–44 17
45–59 17
60+ 14
Homeownership status
Own 15
Rent 18
Other 14
Overall 16
Note: Among all adults.
Figure 24. Expect risk of being affected by
natural disaster to increase in the next 5 years
(by whether affected in the prior 12 months)
Expect risk to be
somewhat higher
Percent
Expect risk to
be much higher
Overall
Affected by disaster
in past year
Not affected by
a disaster
9
26 10
19
46
31
27
20
11
Note: Among all adults. Key identifies bars in order
from left to right.
40 Economic Well-Being of U.S. Households in 2021
places to live in the prior 12 months because
of natural disaster risks, double the 11 per-
cent of renters with less than a high school
degree who did so. Homeowners with college
degrees were also more likely to improve their
properties to reduce natural disaster risks
than were those with less education.
Mitigation activities also differed significantly
by age. Younger adults—regardless of whether
they were renters or homeowners—were more
likely to investigate other places to live
because of natural disaster risk and more
likely to invest in property improvements than
older adults.
Mitigation activities were also much more
common among people who expected their
risk of experiencing a natural disaster to be
higher in five years. Among adults who
expected their natural disaster risk to be
higher in five years, 28 percent of renters
investigated other places to live, and 29 per-
cent of homeowners improved their
properties—each about 10 percentage points
higher than the comparable rates for all
renters and homeowners.
Fewer adults purchased additional insurance
to mitigate their natural disaster risk com-
pared with the other disaster mitigation activi-
ties asked about in the survey. Overall, 6 per-
cent of renters and 5 percent of homeowners
purchased additional insurance in the prior
12 months. While the rates were higher
among people who expected their risk of expe-
riencing a natural disaster to increase, they
were less than 10 percent for that
population.
Figure 25. Natural disaster mitigation
activities (by homeownership status)
Renter
Homeowner
Purchased
additional
insurance
Improved
property
Investigated other
places to live
Percent
10
19
18
7
5
6
Note: Among all adults. Key identifies bars in order
from left to right.
Table 10. Natural disaster mitigation activities
(by homeownership status, education, and age)
Percent
Characteristic
Investigated other
places to live
Improved property
to reduce risk
Home-
owner
Renter
Home-
owner
Renter
Education
Less than a high
school degree 7 11 12 9
High school degree or GED 6 15 13 5
Some college/technical or
associate degree 10 21 19 9
Bachelor’s degree or more 11 22 20 6
Age
18–29 15 21 20 8
30–44 12 21 18 6
45–59 10 16 17 5
60+ 6 10 18 6
Note: Among all adults.
Dealing with Unexpected Expenses 41
Banking and Credit
Access to financial services from banks and credit unions can be important for people’s financial
well-being. Most adults had a bank account and were able to obtain credit from mainstream
sources in 2021, but notable gaps in access to basic financial services still exist among Black
and Hispanic adults and those with low income.
Fewer applicants were denied credit in 2021, and the share of adults who were “very confident”
that they would be approved for credit if they applied increased to the highest levels observed
since the survey first asked this question in 2015. The use of alternative financial services such
as money orders, check cashing, and payday loans was essentially unchanged after a years-long
decline.
Unbanked and Underbanked
Most adults in the United States (81 percent) were “fully banked,” meaning that they had a bank
account and, in the past 12 months, did not use any of the alternative financial services asked
about in the survey. Such services include money orders, check cashing services, payday loans or
payday advances, pawn shop loans, auto title loans, and tax refund advances.
An additional 13 percent had a bank account but also made use of alternative financial services.
These adults are considered “underbanked” because the banking services they accessed appear
to have been insufficient to meet their financial service needs.
The rest of the adult population (6 percent) did not have a bank account. Less than half of these
“unbanked” adults used alternative financial services.
Unbanked and underbanked rates were higher among adults with lower income, adults with less
education, and Black and Hispanic adults. The largest differences were by education and income
level. Twenty-four percent of adults with less than a high school degree, and 17 percent of adults
with income below $25,000, were unbanked (table 11). The share of people with income under
$25,000 without a bank account far exceeded that of the two highest income levels. As a result,
79 percent of all unbanked adults had income below $25,000, and 91 percent had income below
$50,000.
Adults with less education and adults with lower income were also more likely to be underbanked.
Nearly one-fourth of those with less than a high school degree and 20 percent of those with
43
income less than $25,000 were underbanked. In addition to being more likely to use alternative
financial services, lower-income adults were also more likely to use cryptocurrencies for transac-
tion purposes, as discussed in box 2.
Between 2020 and 2021, the shares of unbanked, underbanked, and fully banked adults were
essentially unchanged. Looking at the longer-term trend, however, shows a 7 percentage point
decline in the underbanked rate since 2015. During this same period, the share of adults with a
bank account increased by 2 percentage points. As a result, much of the decline in the under-
banked rate likely reflects the decline in the share using the alternative financial services asked
about in the survey.
This decline does not necessarily reflect improvements in financial inclusion, however. The market
for financial products and services has been evolving during this time, particularly in the digital
space. As a result, while use of these alternative financial services have declined, people may
have substituted away from the products and services asked about in the survey to other nonbank
offerings that are harder to measure.
Table 11. Banking status (by family income, education, and race/ethnicity)
Percent
Characteristic Unbanked Underbanked
Fully
banked
Family income
Less than $25,000 17 20 62
$25,000–$49,999 4 20 76
$50,000–$99,999 1 12 87
$100,000 or more 1 5 94
Education
Less than a high school degree 24 23 53
High school degree or GED 10 15 75
Some college/technical or associate degree 5 17 79
Bachelor’s degree or more 1 7 92
Race/ethnicity
White 3 10 87
Black 13 27 59
Hispanic 11 18 71
Asian 2 6 92
Overall 6 13 81
Note: Among all adults.
44 Economic Well-Being of U.S. Households in 2021
Box 2. Conducting Financial Transactions Using
Cryptocurrencies
Cryptocurrencies are relatively new digital assets that may be held as an investment or used for con-
ducting financial transactions.
1
In 2021, most people using cryptocurrencies did so for investment pur-
poses. In 2021, 12 percent of adults held or used cryptocurrencies in the prior year. Eleven percent of
adults had held cryptocurrency as an investment, while a far smaller 2 percent of adults said that they
used cryptocurrency to buy something or make a payment in the prior 12 months, and 1 percent used it
to send money to friends or family.
2
Those who held cryptocurrency purely for investment purposes were disproportionately high-income,
almost always had a traditional banking relationship, and typically had other retirement savings.
Forty-six percent of those using cryptocurrencies only for investment had an income of $100,000 or
more, while 29 percent had an income under $50,000. Additionally, 99 percent of those investing in
cryptocurrency, but not using it for transactions, had a bank account, and 89 percent of nonretired crypto-
currency investors had at least some retirement savings (figure A).
Figure A. Share without a bank account, credit card, or retirement savings (by cryptocurrency use)
Note: Among all adults. Key identifies bars in order from left to right.
No cryptocurrency use
Cryptocurrency for investment only
Cryptocurrency for transactions
No bank account No credit card No retirement savings
Percent
7
17
27
13
6
1
27
11
29
(continued)
1
Cryptocurrencies are decentralized digital assets that have a distributed ledger and can be used for peer-to-peer payments. For
additional information on cryptocurrencies, see Board of Governors of the Federal Reserve System, Money and Payments: The U.S.
Dollar in the Age of Digital Transformation (Washington: Board of Governors, January 2022), https://www.federalreserve.gov/
publications/files/money-and-payments-20220120.pdf.
2
Because the survey is conducted online, the sample population may be more technologically connected than the overall population,
which could increase the share of adults reporting use of emerging technologies such as cryptocurrencies.
Banking and Credit 45
Moreover, the longer-term decline in use of alternative financial services has been similar among
both the unbanked and banked, providing more evidence that a wider availability of banking ser-
vices may not explain the decline in the underbanked rate.
Overdraft
Overall, 11 percent of adults with a bank
account paid an overdraft fee in the previous
12 months (table 12). Adults with income less
than $50,000 were three times as likely to
have paid an overdraft fee as people with an
income of $100,000 or more. Similarly, the
share of adults paying an overdraft fee tended
to be higher among people with less educa-
tion or of younger ages. Across races or eth-
nicities, a larger share of Black or Hispanic
adults paid an overdraft fee in the past
12 months than the population as a whole. In
contrast, 3 percent of Asian adults paid an
overdraft fee, a rate about one-third of the
population as a whole.
Adults with bank accounts who used credit
alternative financial services and those with
lower self-reported credit ratings were particu-
larly likely to have paid an overdraft fee in the
prior year.
39
Fifty-three percent of banked
adults who used credit alternative financial
39
Credit alternative financial services include payday loans or payday advances, pawn shop loans, auto title loans, and tax
refund advances.
Box 2—continued
The financial profiles of those who used cryptocurrency for transactions, however, were quite different.
Nearly 6 in 10 adults who used cryptocurrencies for transactions had an income of less than $50,000.
A far lower 24 percent of transactional users had an income of more than $100,000.
Transactional cryptocurrency users also were less likely to have a bank account. Thirteen percent of
those who used cryptocurrency for transactions lack a bank account, compared with 6 percent of
adults who did not use cryptocurrency. Similarly, 27 percent of transactional cryptocurrency users did
not have a credit card, exceeding the 17 percent of non-users without a credit card.
Table 12. Paid an overdraft fee on a bank
account in the prior year (by demographic
characteristics)
Characteristic Percent
Family income
Less than $25,000 16
$25,000–$49,999 16
$50,000–$99,999 10
$100,000 or more 5
Education
Less than a high school degree 16
High school degree or GED 11
Some college/technical or associate degree 13
Bachelor’s degree or more 7
Race/ethnicity
White 9
Black 20
Hispanic 14
Asian 3
Age
18–29 15
30–44 14
45–59 10
60+ 5
Overall 11
Note: Among adults with a bank account.
46 Economic Well-Being of U.S. Households in 2021
services also paid an overdraft fee in the prior
12 months, about five times the rate of the
banked population overall (figure 26).
40
Addi-
tionally, 35 percent of those rating their credit
as “poor” and 28 percent rating their credit as
“very poor” paid an overdraft fee, compared
with 3 percent of those rating their credit as
“excellent.”
One explanation for these patterns is that
those using credit alternative financial ser-
vices and those rating their credit as “poor” or
“very poor” were in a precarious financial posi-
tion, making them more likely to unintention-
ally overdraft their account. Another possibility
is that people without access to cheaper
forms of credit were intentionally using over-
draft as a form of short-term, albeit high-cost,
credit. Finally, use of online credit alternative
financial services themselves may directly
trigger an overdraft when the lender attempts to collect payment.
41
Credit Outcomes and Perceptions
Thirty-eight percent of adults applied for credit in 2021, a slight increase over the share who
applied in 2020. But among those who applied, the share who were either denied credit, or
approved for less credit than they requested, fell about 3 percentage points to 28 percent. Consis-
tent with the lower denial rates, consumer confidence about credit card applications improved.
Sixty-five percent of adults were “very confident” that their application would be approved, 4 per-
centage points higher than in 2020 and the highest share since this question was first asked in
2015. Similarly, only 12 percent of adults were “not confident” that their application would be
approved.
The share of adults who were denied credit, or approved for less than requested, differed by
income level and by race and ethnicity (figure 27). Almost half of credit applicants with income
below $50,000 experienced such actions, compared with 11 percent of those with income above
$100,000.
40
Overall, 4 percent of adults had a bank account and used credit alternative financial services.
41
See Consumer Financial Protection Bureau, Online Payday Loan Payments (Washington: CFPB, April 2016), https://
files.consumerfinance.gov/f/201604_cfpb_online-payday-loan-payments.pdf.
Figure 26. Paid an overdraft fee on a bank
account in the prior year (by use of credit
alternative financial services (AFS) and
self-reported credit rating)
Percent
Very
poor
PoorFairGoodExcellentUsed
credit
AFS
All
banked
adults
11
53
3
10
25
35
28
Note: Among adults with a bank account.
Banking and Credit 47
Denial rates also differed by race and ethnicity, with Black and Hispanic applicants being particu-
larly likely to report a denial or an approval for less credit than requested. For Black applicants,
this was also true within income levels, although for Hispanic applicants with income below
$50,000, the denial rates were comparable to others with similar incomes.
Credit Cards
People use credit cards in different ways. Some use credit cards as a convenient, if not necessary,
way to pay expenses, paying off their balances in full each month and avoiding any interest costs.
Others carry a balance and thus use credit cards as a source of credit to defer paying expenses.
Eighty-four percent of adults had a credit card in 2021. They were nearly evenly split between the
people who paid off their balances in each of the previous 12 months and people who carried bal-
ances from month to month at least once in the prior year. Among those who carried a balance at
least once, 73 percent were carrying a balance at the time of the survey.
Almost all people with income of at least $100,000 had a credit card. At lower income levels,
having a credit card was somewhat less common, though adults at these income levels who did
have credit cards were more likely to use them to carry balances from month to month. Conse-
quently, middle-income adults were the most likely to have a credit card that they used to finance
purchases by carrying balances from one month to the next. Almost half of people with income
between $25,000 and $99,999 carried a balance on a credit card at least once in the past
Figure 27. Denied credit or approved for less than was requested (by family income and race/ethnicity)
Hispanic
Black
White
All income levels
Hispanic
Black
White
$100,000 or more
Hispanic
Black
White
$50,000–$99,999
Hispanic
Black
White
Less than $50,000
43
60
47
20
39
33
8
22
18
22
46
37
Percent
Note: Among adults who applied for some form of credit in the past 12 months.
48 Economic Well-Being of U.S. Households in 2021
12 months, exceeding the shares of adults
with either lower or higher income levels who
did so (table 13).
Similar patterns were observed across educa-
tion levels, with more-educated adults being
both more likely to have a credit card and less
likely to carry a balance from one month to
the next. Credit card usage also differed by
race and ethnicity. Over 90 percent of Asian
adults had a credit card but just under one in
four of those with a credit card carried a bal-
ance at least once in the prior 12 months.
Black and Hispanic adults were much more
likely to carry balances on their credit cards
than other racial or ethnic groups.
The share of adults with outstanding debt who
were carrying less debt than 12 months ago
was comparable with, though slightly higher
than, the share carrying more debt. This pat-
tern is similar to that in recent years, with the
exception of 2020, when the share who had
reduced their credit card debt exceeded the
share who had increased it by 8 per-
centage points.
Buy Now, Pay Later
The 2021 survey introduced a series of ques-
tions about the use of BNPL services. BNPL
allows people to finance a purchase by
making a small number of equal payments, often without being charged interest. For example,
someone purchasing a $100 item, instead of paying the entire amount upfront, may instead be
able to make four monthly payments of $25, with the first payment due at the time of purchase.
Overall, 10 percent of people used a BNPL service in the previous 12 months. Seven percent were
making payments under a BNPL plan at the time of the survey, with about half paying on just one
purchase.
Table 13. Credit card access and usage (by
demographic characteristics)
Percent
Characteristic
Has a
credit
card
Carried a
balance
(among
credit
card
holders)
Carried a
balance
(among
adults)
Family income
Less than $25,000 57 57 33
$25,000–$49,999 84 59 49
$50,000–$99,999 94 50 47
$100,000 or more 98 36 36
Education
Less than a high school degree 52 57 30
High school degree or GED 76 57 43
Some college/technical or
associate degree 83 56 46
Bachelor’s degree or more 96 35 34
Race/ethnicity
White 88 42 37
Black 72 72 52
Hispanic 77 63 48
Asian 93 24 22
Self-reported credit rating
Very poor 33 66 22
Poor 45 77 35
Fair 73 81 59
Good 90 65 58
Excellent 97 31 31
Don’t know 40 39 16
Overall 84 48 40
Note: Among all adults. Carried a balance in the prior 12 months
includes adults who carried an unpaid balance from one month
to the next at least once in the 12-month period.
Banking and Credit 49
The most cited reasons for using BNPL services were convenience (78 percent) and not wanting to
use a credit card (53 percent) (figure 28). Just over half of people who used BNPL also indicated
that it was the only way they could afford their purchase (51 percent).
The use of BNPL was more common among people with lower income and less education
(table 14). Around 13 percent of those with income below $50,000 used BNPL in the prior year,
compared with 7 percent of those with an income of $100,000 or more. Similarly, 14 percent of
people with less than a high school degree used BNPL, compared with 8 percent of those with at
least a bachelor’s degree.
The reasons for using BNPL also differed with income and education levels. While convenience
was universally the most cited reason for using BNPL, around 60 percent of people with income
under $50,000 or with no more than a high school degree cited an inability to pay for the product
otherwise, compared with about one-fourth of people whose income was at least $100,000. More-
over, around 25 percent of people in these lower income and education groups reported using
BNPL because they lacked another accepted payment option, more than twice the rate for people
in the highest income and education categories.
People also differed in their use of BNPL according to their self-reported credit rating. Those with
lower credit ratings were more likely to use BNPL than were people who rated their credit as
“excellent.” Among those who used BNPL, adults with lower self-reported credit ratings were also
more likely to cite an inability to afford the purchase otherwise or a lack of other payment options
as reasons for using BNPL than adults who rated their credit higher.
Figure 28. Reasons for using Buy Now, Pay Later (BNPL) services
Other
Did not want to use
a credit card
Only accepted payment
method I had
Only way I could
afford it
More convenient
Cheaper
34
78
51
19
53
11
Percent
Note: Among adults who have used a BNPL service in the past year.
50 Economic Well-Being of U.S. Households in 2021
Most people who use BNPL make their pay-
ments on time. Overall, 15 percent of people
who used BNPL in the prior 12 months were
late making a payment. Late payments were
somewhat more common among people with
income less than $50,000. Late payments
were also more common among people with
lower self-reported credit ratings. Adults who
rated their credit as “poor” were over five
times as likely to have been late making a
BNPL payment as someone who rated their
credit as “excellent.” Among people who
would not have been able to afford their pur-
chase without BNPL, 23 percent paid late,
compared with 7 percent of people who did
not give that reason for using BNPL.
Table 14. BNPL service use (by demographic
characteristics)
Percent
Characteristic
Used a
BNPL service
Paid BNPL late
(among users)
Family income
Less than $25,000 12 23
$25,000–$49,999 14 18
$50,000–$99,999 11 11
$100,000 or more 7 6
Education
Less than a high school
degree 14 n/a
High school degree or GED 10 15
Some college/technical or
associate degree 12 17
Bachelor’s degree or more 8 10
Race/ethnicity
White 7 10
Black 20 21
Hispanic 15 19
Asian 7 n/a
Age
18–29 13 18
30–44 13 18
45–59 12 13
60+ 6 9
Credit card ownership
Has a credit card 10 14
No credit card 10 23
Self-reported credit score
Very poor 13 n/a
Poor 21 29
Fair 22 13
Good 12 13
Excellent 5 5
Don’t know 6 n/a
Overall 10 15
Note: Among all adults. Some results are not applicable
because of small sample size.
n/a Not applicable.
Banking and Credit 51
Housing
Housing—especially the cost of housing and housing tenure type—affects people’s economic well-
being. The majority of adults owned their homes. Adults who rented their homes were dispropor-
tionately lower-income, Black, or Hispanic. The share of renters who had been behind on their rent
in the prior 12 months was higher than the level before the pandemic. Among homeowners, the
refinancing wave continued, although high-income borrowers were primarily the beneficiaries of this
opportunity.
Living Arrangements
Living arrangements can affect family finances
and well-being. Eighty-six percent of adults
lived with other people, usually a spouse or a
partner and frequently their children under age
18 (table 15). More than half (52 percent) of
all adults lived in a household with a spouse
or partner or with a child under age 18 and
with no one else. Other types of living arrange-
ments were less common. Still, more than
one-fourth of adults (28 percent) lived in a
household that contains multiple generations
of adults, meaning that the adult respondents
either lived with their parents or adult children.
Older adults, and older women in particular, were the most likely to live alone. Twenty-one percent
of adults age 65 or older lived alone, and 27 percent of women age 65 or older lived alone. In con-
trast, young adults were very likely to live with their parents. But this rate drops significantly for
adults in their mid- and late 20s: 47 percent of 22- to 24-year-olds lived with their parents com-
pared with 27 percent of 25- to 29-year-olds. Fewer adults in older age cohorts live with their par-
ents: 13 percent of 30- to 44-year-olds live with their parents and 7 percent of 45- to 59-year-olds
live with their parents. Conversely, the share living with a spouse or partner increased with age,
from 26 percent of 22- to 24-year-olds to 52 percent of 25- to 29-year-olds.
42
42
Among adults ages 22 to 24, the majority of those living with a spouse or partner are not married, and a smaller share
are married. For the older age groups, the majority of adults living with a spouse or partner are married.
Table 15. Other people living in household
Category Percent
Live alone 14
Spouse or partner 66
Children under age 18 25
Adult children age 18 or older 16
Parents 13
Brothers or sisters 6
Other relatives 4
Other non-relatives 5
Note: Among all adults. Respondents (other than those who live
alone) can select multiple answers.
53
A substantial majority of young adults living
with their parents said that saving money was
a reason for their living arrangement. Overall,
three-fourths of people who lived with their
parents said this was to save money. However,
the direction of financial support flips for
many older adults who live with their parents.
Fifty-seven percent of adults ages 30 to 44
who lived with their parents said that providing
financial help was a reason, as did 61 percent
of those ages 45 to 59 (table 16).
Another common reason for living with parents
was to provide care, especially as people get
older. Twenty-one percent of people who lived with their parents gave this reason. Adults in their
30s, 40s, and 50s who lived with their parents were more likely to say that they lived with others
for caregiving reasons. Forty-five percent of 45- to 59-year-olds who lived with their parents said
they lived with others to provide care.
Moving
The share of people who reported moving in 2021 was unchanged from 2020. Nine percent of
adults said they moved to their home in 2021. The majority of individuals who moved remained in
the same state. Just over one in four adults who moved—2 percent of all adults—crossed state
lines in 2021.
Moving was generally associated with an increasing distance from family, friends, and other
informal supports. Thirty-two percent of people who moved in 2021 said they moved farther away
from family, while 24 percent said they moved closer (figure 29). Forty percent of adults who
Table 16. Reasons for living with parents
(by age)
Reason 22–24 25–29 30–44 45–59
To save money 92 88 63 48
To help them financially 31 44 57 61
To provide help with
childcare or
medical care 7 15 27 45
To receive help with
childcare or
medical care 13 16 20 15
Prefer living with others 46 40 36 23
Note: Among people living with parents. Respondents could
select multiple answers.
Figure 29. Distance to friends, families, and workplaces after moves
Farther awayCloser
Usual workplace
Friends
Extended family 24
18
27
32
40
34
Percent
Note: Among people who moved in 2021. Distance relative to one’s usual workplace is among employed adults. Key
identifies bars in order from left to right.
54 Economic Well-Being of U.S. Households in 2021
moved said they moved farther from friends.
Another moving pattern that continued in
2021 was employed adults moving farther
away from their usual workplace. Thirty-four
percent of movers who were employed moved
farther away from their usual workplace com-
pared with 27 percent who moved closer.
Homeownership and Mortgages
Nearly two-thirds of adults owned their homes,
though young adults, as well as Black and His-
panic adults, were less likely to own. Twenty-
nine percent of 18- to 29-year-olds owned
their homes, compared with 84 percent of
people age 60 and older. Within each age
group, there is substantial variation in the
homeownership rate by race and ethnicity. For
example, 4 in 10 Black adults and nearly 5 in
10 Hispanic adults ages 30 to 44 were home-
owners. Among White adults in this age range,
nearly 7 in 10 owned their home (table 17).
Many homeowners took advantage of the con-
tinued low interest rates in 2021 to refinance
their mortgages. Nearly one-fourth of all
homeowners with a mortgage refinanced their
mortgage within the prior year. Higher-income
homeowners were the predominant group who
opted to refinance (figure 30).
43
Nearly 3 in
10 mortgage holders with income of at least
$100,000 per year refinanced within the prior
12 months, compared with 23 percent of
those with income between $50,000 and
$99,999 and 16 percent of those with income
under $50,000.
43
This fact holds when considering individuals who refinanced in 2020 as well. Among those who have refinanced in the
past two years, high-income (those with an income of $100,000 or more) homeowners with a mortgage were about
twice as likely to have refinanced as those with an income of less than $50,000.
Table 17. Homeownership rate (by age and
race/ethnicity)
Characteristic Percent
18–29
White 35
Black 18
Hispanic 20
Asian 34
Overall 29
30–44
White 69
Black 40
Hispanic 49
Asian 71
Overall 61
45–59
White 83
Black 57
Hispanic 63
Asian 90
Overall 76
60+
White 88
Black 68
Hispanic 78
Asian 78
Overall 84
Note: Among all adults.
Figure 30. Share of homeowners with a
mortgage who refinanced in the prior year (by
family income)
Percent
$100,000 or more
$50,000–$99,999
$25,000–$49,999
Less than $25,000
16
15
23
29
Note: Among homeowners with a mortgage.
Housing 55
Mortgage holders with a higher mortgage pay-
ment were also more likely to have refinanced
during the prior year. More than one-third
(35 percent) of homeowners with a monthly
mortgage payment of $2,000 or more refi-
nanced, compared with 15 percent of home-
owners with a monthly payment from $500 to
$749 (figure 31).
44
Renting and Evictions
More than one in four adults rent their home
(27 percent). Benefits of renting include the
flexibility to move more easily as well as the
convenience of not having to manage repairs.
But renting can also lead to less-stable living
arrangements and less control over living
spaces and repairs. Many renters do not own
their home because of financial circumstances. (See box 3 for a discussion of renters who fell
behind on rent during the pandemic.)
Adults with lower income, and those who are Black and Hispanic, are more likely to rent their
homes. Forty-five percent of adults with a family income of less than $25,000 rent, compared with
10 percent of adults with family income of $100,000 or more (table 18). Forty-four percent of
Black adults and 37 percent of Hispanic adults rent, compared with 21 percent of White adults
and 23 percent of Asian adults.
Housing tenure varies by other demographic characteristics, including disability status and neigh-
borhood income. Adults with a disability had a greater likelihood of being renters than adults with
no disability. Forty-four percent of adults who live in low- and moderate-income neighborhoods rent.
This is over twice the rate among adults who live in middle- and upper-income neighborhoods.
44
While families with more income have higher mortgage payments on average, the increased prevalence of refinancing
among those with higher mortgage payments holds even when controlling for income. In part, this may reflect that the
potential savings from refinancing are greater for those with larger loans.
Figure 31. Share of homeowners with a
mortgage who refinanced in the prior year (by
current month mortgage payment)
Percent
7
15
19
19
26
30
35
24
Overall
$2,000 or more
$1,500–$1,999
$1,250–$1,499
$1,000–$1,249
$750–$999
$500–$749
Less than $500
Note: Among homeowners with a mortgage.
56 Economic Well-Being of U.S. Households in 2021
Renters with lower family income were fre-
quently cost burdened, meaning that they
spent more than 30 percent of their income
on rent payments.
45
About half of renters with
income between $25,000 and $49,999 had
rent payments that exceeded 30 percent of
their income.
Reflecting the flexibility that comes with
renting, most people who moved were renters.
Almost three-fourths of people who moved in
the prior year did not own their home before
the move. In general, these moves were to
another rental, rather than a home
purchase—just 26 percent of those who did
not own their previous house and moved in
the past year did so for a home that they pur-
chased.
Some of these moves, however, resulted from
an eviction. Slightly fewer than 1 percent of
adults, which is about 1.8 million people, said
they moved in the prior year because of an
eviction or the threat of an eviction.
46
This
represents approximately 8 percent of all
people who moved during this period.
45
Cost burdened is defined using the midpoint of both the monthly rent payment range and the family income range and
comparing the ratio to the 30 percent threshold. The Department of Housing and Urban Development has established
this “cost burdened” threshold of 30 percent. For details, see U.S. Department of Housing and Urban Development,
“Rental Burdens: Rethinking Affordability Measures,” https://www.huduser.gov/portal/pdredge/
pdr_edge_featd_article_092214.html.
46
In this report, people who experienced an eviction or the threat of eviction include those who reported they were evicted
or received an eviction notice; had a landlord tell them or a person they were staying with to leave; missed a rent pay-
ment and thought they would be evicted; or were living in a property that was condemned by the city, forcing them
to leave.
Table 18. Share who rent (by demographic
characteristics)
Characteristic Percent
Family income
Less than $25,000 45
$25,000–$49,999 36
$50,000–$99,999 24
$100,000 or more 10
Race/ethnicity
White 21
Black 44
Hispanic 37
Asian 23
Disability status
Disability 36
No disability 24
Metro status
Metro 28
Non-metro 21
Neighborhood income
Low or moderate income 44
Middle or upper income 21
Overall 27
Note: Among all adults.
Housing 57
Box 3. Pandemic’s Effect on Rent Payment
Many renters faced challenges paying their rent before the pandemic, but in the fall of 2021, after
nearly two years of economic disruptions from the pandemic, a higher share of renters reported they
had been behind on their rent in the prior 12 months. Moreover, many still owed back rent despite a
variety of government supports including unemployment benefits, stimulus checks, and rental assis-
tance.
More renters were behind on rent in 2021 than before the pandemic. When asked about their rent pay-
ments before the pandemic, 10 percent of renters reported they had missed a payment at some point
in 2019.
1
In the fall of 2021, a higher 17 percent of renters reported they had been behind on their
rent in the prior 12 months.
The pandemic increased the share of renters who were behind on rent for most racial and ethnic
groups. When compared with pre-pandemic levels, White, Black, and Hispanic renters all saw increases
in the share behind on rent sometime in the prior 12 months (figure A).
Figure A. Share of renters behind on rent during the year (by year and race/ethnicity)
Note: Among renters. Key identifies bars in order from left to right.
AsianHispanicBlackWhite
20212019
Percent
7
13
21
14
12
22
9
8
Black renters and Hispanic renters were more likely to be behind on rent payments, compared with
other renters. This disparity was present in 2019 and persisted through the pandemic. In 2021, over
one in five Black renters and Hispanic renters said they had been behind on rent in the prior year.
(continued)
1
This question was asked of renters in 2021 about their rent payments in 2019. It was not asked on the 2019 survey.
58 Economic Well-Being of U.S. Households in 2021
Box 3—continued
Layoffs during the pandemic likely contributed to difficulty paying rent. Among renters who were laid off
in the prior 12 months, 38 percent were behind on rent, compared with 15 percent of renters who were
not laid off. Differences in layoffs by race and ethnicity may contribute to differences in being behind on
rent since layoffs were more common for Black and Hispanic renters than other renters in 2020
and 2021.
Low-income renters were also hit particularly hard by the pandemic recession. The share of adult
renters with income below $50,000 reporting being behind on rent increased from 12 percent in 2019
to 23 percent in 2021. For renters with income of $50,000 or more, a smaller 6 percent reported in
the fall of 2021 that they had been behind on rent in the prior 12 months—similar to the 5 percent
who said they were behind in 2019.
Many renters who fell behind during the year still carried rental debt as of late 2021. Forty-five percent
of renters who were behind on their rent at some point in 2021 said that they still owed money for back
rent or fees at the time of the survey. This represents 8 percent of renters (2 percent of adults) who
still owed back rent or fees in late 2021. For this group of renters, the mean amount still owed was
$2,064 and the median was $1,200. This mean amount of back rent suggests total estimated back
rent for all renters as of late 2021 was between $9.3 and $10.9 billion.
2
These estimates are lower
than October 2021 estimates from the Federal Reserve Bank of Philadelphia ($16.8 billion) and
Moody’s Analytics and the Urban Institute ($16.7 billion).
3
2
These estimates are for adults who are renters. Because respondents who were married could have given the amount of back rent
for their household, or just their individual share, the lower estimate divides the amount of back rent in half for married couples to
account for married respondents who may have answered for their household rather than for just themselves.
3
The Federal Reserve Bank of Philadelphia’s estimate comes from a model simulating job loss and calculates total back rent for
renters who experienced a job loss during the pandemic (Federal Reserve Bank of Philadelphia, Household Rental Debt during
COVID-19: Update for August 2021 (Philadelphia: FRB Philadelphia, July 2021), https://www.philadelphiafed.org/-/media/frbp/
assets/community-development/briefs/updatedhouseholdrentdebt-final.pdf ). Moody’s Analytics and Urban Institute’s estimate
comes from Moody’s baseline economic forecast and calculates back rent for all renters, including those without a job loss during
the pandemic (Jim Parrott and Mark Zandi, “The Race to Save Millions from Eviction,” Urban Institute, September 2021, https://
www.urban.org/sites/default/files/publication/104762/the-race-to-save-millions-from-eviction.pdf). In contrast, the SHED estimates
rely on survey questions that ask directly about the amount of back rent owed.
Housing 59
Education
Education is widely recognized as a path to higher income and greater financial well-being. The
pandemic brought widespread education disruptions, including school closures for students of all
ages in 2020. In 2021, K–12 schools largely returned to in-person education. At the time of the
survey, most parents of primary or secondary school students reported that their youngest child
was attending classes completely in person.
This shift to in-person learning likely reduced childcare responsibilities, and most parents said
they preferred in-person classes over online or hybrid options. However, potentially reflecting
ongoing concerns about COVID-19 transmission, some parents whose children were attending
school in person in the fall of 2021 would have preferred online or hybrid classes for their child.
In contrast to the experience of K–12 students, online education remained prevalent at higher edu-
cation institutions in the fall of 2021. Most higher-education students preferred at least some
online classes.
Modes of Learning in Primary and Secondary School
In the fall of 2021, most parents of primary and secondary school students said their children had
returned to completely in-person education after the widespread reliance on online learning in
2020.
47
At the time of the survey, 93 percent of parents with children in school said their
youngest child enrolled in K–12 education was attending classes completely in person, compared
to 27 percent with completely in-person classes in 2020.
Even among children attending school in person, however, disruptions occurred because of the
pandemic. Over one-fourth (27 percent) of parents whose youngest child’s classes were com-
pletely in person said that at least once, since the start of the school year, their child was unable
to attend in person because of a pandemic-related disruption. For 7 percent of parents whose
child’s classes were completely in person, a disruption to in-person schooling led them to work
fewer hours or take unpaid leave from work.
Though nearly all parents of K–12 students said that their child’s classes were in person, lower-
income parents were less likely to report in-person K–12 education than higher-income parents.
Ninety percent of parents making less than $25,000 per year said that their child’s classes were
47
References to a child’s education in this section refer to the individual’s youngest school-age child. Parents of
school-age children are respondents who lived with their own children under age 18 who were enrolled in a public or pri-
vate K–12 school. Except where specified, parents who only home-school their children are excluded.
61
in person, compared with 97 percent of parents making $100,000 or more per year. Additionally,
Black and Asian parents were less likely to say their child was attending school in person than
White and Hispanic parents.
However, preferred modes of education also varied by income and race. Nearly 9 in 10 parents of
school-age children with an annual income of $100,000 or more said they prefer completely
in-person education, compared with fewer than 7 in 10 parents with an annual income under
$25,000. Eighty-seven percent of White parents with school-age children said they prefer com-
pletely in-person education, higher than that seen among Black, Hispanic, or Asian parents
(figure 32).
Additionally, parents’ preferred mode of learning differed by the type and level of the school. Par-
ents of children in public school were less likely (80 percent) than parents of children in private
school (89 percent) to prefer completely in-person education, and parents whose youngest child is
in middle or high school were less likely (76 percent) to prefer in-person class than parents with
children in elementary school (84 percent).
Most parents (72 percent) felt that their child’s school was taking the right level of COVID-19 pre-
cautions. Of those who did not, slightly more felt that the school was taking too few precautions
(17 percent) than too many (12 percent). This is consistent with the observation that a larger
share of students had in-person education than parents preferred.
Figure 32. Actual and preferred mode of learning for K–12 (by income and race/ethnicity)
Overall
Asian
Hispanic
Black
White
Race/ethnicity
$100,000 or more
$50,000–$99,999
$25,000–$49,999
Less than $25,000
Family income
Percent
90
68
9176
92
80
97
89
95
87
89
64
94
76
89
80
9380
Child’s school is completely in person!Prefers completely in-person education!
Note: Among parents with a child enrolled in public or private school. Based on the youngest child enrolled in public or
private school who lives with their parent.
62 Economic Well-Being of U.S. Households in 2021
Parents’ views on the precautions taken by their children’s schools varied along similar lines to
their preferences for in-person education. Low-income parents were more likely than high-income
parents to say their child’s school was not taking enough precautions in light of the pandemic, and
Black parents were over twice as likely as White parents to say this (figure 33).
Parents of children in public school were also more likely (18 percent) than those with children in
private school (9 percent) to say their child’s school was not taking enough precautions. However,
parents’ opinions on precautions at school did not vary by the age of their youngest child.
Some parents opted to home school their children in 2021. Just under 1 in 10 parents of
school-age children said that one of their children was home schooled and not enrolled in public or
private school. For most home-schooling parents, this was not because of COVID-19. Fifty-
three percent said that COVID-19 concerns and school safety policies did not contribute to the
decision. This is consistent with observations from other data that about half of the current rate of
home schooling predated the pandemic.
48
Of those who home schooled their children for COVID-19-related reasons, most did so because
of concerns about exposure at school. Thirty-seven percent of parents who home school at least
one child said they do so in part because of concern about COVID-19 exposure at school.
48
The Census Household Pulse Survey indicates that, by the fall of 2020, the share of households with school-age chil-
dren who home school their children was around 11 percent—about double the share who home schooled at the start
of the pandemic. See https://www.census.gov/library/stories/2021/03/homeschooling-on-the-rise-during-covid-19-
pandemic.html. By December 2021, the share of households with school-age children who reported home schooling in
the Census Household Pulse Survey was still around 11 percent, similar to the share who reported home schooling in
this year’s SHED.
Figure 33. Parents’ views on precautions taken by child’s school (by income and race/ethnicity)
Overall
Asian
Hispanic
Black
White
Race/ethnicity
$100,000 or more
$50,000–$99,999
$25,000–$49,999
Less than $25,000
Family income
25
18
16
12
13
28
19
15
17
10
7
12
14
16
5
8
5
12
Too many precautions!
Percent!
Not enough precautions
Note: Among parents with a child enrolled in public or private school. Based on the youngest child enrolled in public or
private school who lives with their parent. Key identifies bars in order from left to right.
Education 63
Fourteen percent said they decided to home school their child in part because the local school’s
safety measures were too strict.
Perceptions of Children’s Performance in School
An important factor for parents as they are making work, spending, and housing decisions is the
effects that these decisions have on their children. Parents may make changes in these areas that
have implications for their family finances if they feel that their child is falling behind. Conse-
quently, the survey asked parents how they felt that their child was faring as the educational envi-
ronment shifted through the pandemic.
Many parents in the 2020 survey felt that the quality of their children’s education had declined
amid the pandemic. In 2021, parents generally said their youngest child in K–12 school was doing
well, both academically and emotionally. Eighty-five percent of parents with a child in public or pri-
vate school said that their child was doing well academically, 84 percent said that they were doing
well socially and emotionally, and 84 percent said their child liked school. A slightly smaller
majority (76 percent) said that their child was prepared for this school year.
A majority of parents with a child in public or private school also said that their child was doing
better academically than in 2020. Fifty-six percent said that their child’s academic performance
improved, compared with 7 percent who said it declined. Similarly, a majority of parents (59 per-
cent) said their child was doing better socially and emotionally compared with a year earlier, while
many fewer parents (8 percent) said their child’s social and emotional performance was worse
than in 2020.
However, parents of children taking classes partially or completely online were less likely to say
their child has improved socially and emotionally. Forty-three percent of parents whose youngest
child was attending classes at least partially online said that their child was doing better socially
and emotionally than in 2020, compared with 60 percent of parents whose youngest child was
attending classes completely in person. In addition, parents of children in online or hybrid educa-
tion were less likely to say that their youngest child likes school or that their child was doing well
academically (figure 34). The share of parents who said their child had improved academically did
not differ significantly between those with completely in-person classes and those with online or
hybrid education.
While most parents said their youngest school-age child was doing well in school in the fall of
2021, parents’ assessments of their child’s educational performance varied by race and ethnicity.
Seventy-six percent of Black parents believed their youngest child was doing well academically,
lower than the share seen among the other racial and ethnic groups. Similarly, Black parents and
Hispanic parents were least likely to say that their youngest child was prepared for the school year.
64 Economic Well-Being of U.S. Households in 2021
However, the share of parents who said their
child likes school did not differ significantly by
race and ethnicity (table 19).
Parents’ assessments of their children’s per-
formance in school also increased with
income. Seventy-seven percent of parents with
under $25,000 in income said their child is
doing well academically, compared with
89 percent of parents with $100,000 or more
in income. Seventy-four percent of parents
earning less than $25,000 agreed that their
child was doing well socially and emotionally,
whereas 88 percent of parents with income of
$100,000 or more agreed with this statement.
Modes of Learning in Higher Education
While primary and secondary schools largely returned to in-person classes in 2021, online
learning remained prevalent at higher education institutions. More than three-fourths of students
enrolled in higher education said their classes were partly or completely online. Although many
postsecondary students in the 2020 survey expressed concern about the quality of online
classes, 76 percent of college students in 2021 said they prefer online or hybrid education.
Figure 34. Parents’ assessment of child’s performance in school (by mode of education)
Online or hybrid
In person
They were prepared
academically to start
the school year
They like school
They are doing well
socially and emotionally
They are doing
well academically
Percent
86
74
84
75
85
68
76
76
Note: Among parents with a child enrolled in public or private school. Based on the youngest child enrolled in public or
private school who lives with their parent. Key identifies bars in order from top to bottom.
Table 19. Parent’s assessment of child’s
performance in school (by race/ethnicity)
Percent
Assessment White Black Hispanic Asian
They are doing well
academically 87 76 83 82
They are doing well socially
and emotionally 86 76 83 81
They like school 84 81 84 86
They were prepared
academically to start the
school year 79 70 74 79
Note: Among parents with a child enrolled in public or private
school. Based on the youngest child enrolled in public or private
school who lives with their parent.
Education 65
Bachelor’s degree students were the least
likely to prefer completely online education,
given the situation with the pandemic. Just
fewer than 3 in 10 students enrolled in bach-
elor’s degree programs said they prefer com-
pletely online classes (table 20). In contrast,
47 percent of students in either technical and
associate degree programs or graduate and
professional degree programs prefer online-
only education. This may reflect that technical
and associate degree students, as well as graduate degree students, were more likely to be older
adults who may have other responsibilities. Sixty-four percent of technical and associate degree
students and 78 percent of graduate degree students were over age 24. Among bachelor’s degree
students, a far lower 36 percent were over age 24.
College students generally expressed satisfaction with the amount of pandemic-related precau-
tions taken by their school. Eight in ten postsecondary students said they thought their school was
taking about the right amount of precautions, while just more than 1 in 10 said their school was
not taking enough precautions.
Overall Value of Higher Education
At the time of the survey, 70 percent of adults had ever enrolled in an educational degree program
beyond high school, and 36 percent had received a bachelor’s degree. Self-reported financial well-
being rose strongly with education, although the effects differed across demographic groups
(see the “Overall Financial Well-Being” section of this report for details on financial well-being by
education).
Consistent with the higher rates of financial well-being among those who have more education,
more than half of adults who went to college said that the lifetime financial benefits of their higher
education exceeded the financial costs. Meanwhile, one in five said that the costs are higher. The
rest saw the benefits as about the same as the costs. These self-assessments of the value of
education have changed little in recent years.
The self-assessed value of higher education, while generally positive, depends on several aspects
of a person’s educational experience. Most importantly, those who completed their program and
received a degree were more likely to see net benefits than noncompleters. For example, among
those who went to college but did not complete at least an associate degree, 31 percent said the
benefits of their education exceeded the cost. This fraction jumped to 46 percent of those with an
associate degree and 67 percent of those with at least a bachelor’s degree.
Table 20. Prefer online only education (by type
of degree program)
Characteristic Percent
GED, technical, or associate degree 47
Bachelor’s degree 29
Graduate or professional degree 47
Overall 40
Note: Among students currently enrolled in higher education.
66 Economic Well-Being of U.S. Households in 2021
The self-assessed value of higher education also differed by age. Among those with at least an
associate degree, older adults were more likely than younger adults to see the benefits of their
education as greater than the costs (figure 35).
49
One explanation for this result could be that
older respondents have had a longer time to experience the benefit of their education than
younger respondents. This variation in views on the net benefit of college may also be driven by
the rising cost of higher education—people who attended college more recently likely faced a
higher cost than those who attended college further in the past.
50
Additionally, the gap in valuations of higher education across age groups was wider among those
with higher degree levels. Among those with an associate degree, fewer than 4 in 10 adults under
age 30 said the benefits of their education exceeded the costs, compared with nearly 6 in 10
adults age 60 and over. Among those with a bachelor’s degree or more, this gap was wider—
56 percent of adults under age 30 thought the benefits of their education exceeded the cost, com-
pared with 82 percent of adults age 60 and over.
One potential explanation is that younger adults are more likely to have taken out debt for their
education and to be paying down these loans. Consequently, the costs of education may be more
salient for them than for older adults (see the “Student Loans” section of this report for a discus-
sion of educational debt and the self-assessment of the value of higher education).
49
If adults currently enrolled in higher education levels are excluded, the share of adults who say the benefits outweigh the
cost increases with age at every education level.
50
From 1995 to 2015, net tuition, fees, room, and board rose 54 percent at public four-year institutions and 29 percent at
private, nonprofit, four-year institutions. See College Board, Trends in College Pricing 2014, https://research.collegeboard.org/
pdf/trends-college-pricing-2014-full-report.pdf. In the current school year, net tuition, room, board, and fees at public and
private nonprofit institutions are about the same as they were in the 2014–15 school year (see https://
research.collegeboard.org/pdf/trends-college-pricing-student-aid-2021.pdf).
Figure 35. Benefits of education exceed costs (by education and age)
Bachelor’s degree
or more
Associate degree
Percent
38
40
43
59
56
59
70
82
60+45–59
30–44
18–29
Note: Among adults who attended college. Key identifies bars in order from top to bottom.
Education 67
Another contributor to differences in how people viewed their education was the type of institution
attended.
51
Consistent with previous years of the survey, 69 percent of those with bachelor’s
degrees from public institutions and 63 percent with bachelor’s degrees from private not-for-profit
institutions saw their educational benefits as greater than their costs. However, 43 percent of
those with bachelor’s degrees from for-profit institutions felt their education was worth the cost.
Look Back on Education Decisions
Another way to assess the value of education is to consider what people would have done differ-
ently if given the chance. Most people value the education they have, but with the benefit of hind-
sight and life experience, it is also common to think that different educational decisions could
have been better.
Completing more education was the most common change that those with less education would
have made if they were able to make a change. Sixty-seven percent of those without a college
degree and 61 percent of those with an associate degree said they would like to have completed
more education (figure 36). For those with a bachelor’s degree or more, choosing a different field
of study (37 percent) was the most common change they would make to their education. Few
people of any education level said they would have completed less education if they could make
their decisions again.
51
Individuals do not self-report the type of institution in the survey. Instead, the institution type is assigned by matching
the name and location of the college reported by the individual with data from the Center on Postsecondary Research at
the Indiana University School of Education (https://cpr.indiana.edu/). For individuals who completed an associate or
bachelor’s degree, institution type is based on the school from which they received the degree. For other individuals, it is
based on the last school attended.
Figure 36. Changes would now make to earlier education decisions (by education)
Bachelor’s degree or more
Associate degree
Some college or technical degree
Attended a
different school
Chosen a different
field of study
Not attended college
or less education
Completed more
education
Percent
67
61
33
14
37
37
31
43
8
6
22
24
Note: Among adults who attended college and are not currently enrolled. Respondents could select multiple answers.
Key identifies bars in order from top to bottom.
68 Economic Well-Being of U.S. Households in 2021
Additionally, reassessments of educational decisions varied by the type of institution attended.
Slightly over half of those who attended a for-profit institution said they would have attended a dif-
ferent school, compared with 31 percent of those attending a private not-for-profit institution and
23 percent attending a public institution (figure 37). This difference remains even after accounting
for the selectiveness of the institution, level of education completed, the parents’ level of educa-
tion, and demographic characteristics of the student.
52
The changes adults who completed at least some college said they would now make to their edu-
cational decisions were also related to the field of study they pursued. In particular, the share who
said they would study a different topic in hindsight varied by fields of study. Forty-eight percent of
those who studied the humanities and arts, and 46 percent of those who studied social and
behavioral sciences, said they would choose a different field. In comparison, a lower share
(24 percent) of those who studied engineering said they would have chosen a different field
(figure 38).
52
Selective institutions, as defined by the Carnegie Classification, are those whose first-year students’ test scores are in
the middle two-fifths of baccalaureate institutions; more selective institutions are in the top one-fifth of baccalaureate
institutions. See also “The Carnegie Classification of Institutions of Higher Education,” web page, http://
carnegieclassifications.iu.edu/. The remainder are referred to here as “less selective” institutions.
Figure 37. Changes would now make to earlier education decisions (by institution type)
Private for-profit
Private not-for-profit
Public
Attended a
different school
Chosen a different
field of study
Not attended college
or less education
Completed more
education
Percent
49
37
58
9
23
38
36
40
5
14
31
52
Note: Among adults who attended college and are not currently enrolled. Respondents could select multiple answers.
Key identifies bars in order from top to bottom.
Education 69
Figure 38. Would now choose a different field of study (by field of study)
Percent
38
Overall
24
Engineering
32
Computer/information sciences
33
Health
33
Physical sciences/math
36
Law
36
Life sciences
37
Business/management
40
Education
42
Vocational/technical training
44
Undeclared/other
46
Social/behavioral sciences
48
Humanities/arts
Note: Among adults who completed at least some college who are not currently enrolled.
70 Economic Well-Being of U.S. Households in 2021
Student Loans
Education debt is prevalent among people who went to college, and especially among younger
adults. In 2021, many student loan borrowers continued to receive delays in payment due dates
for student loan bills because of ongoing pandemic relief measures. Consequently, the share of
borrowers from a range of backgrounds who were behind on their payments in the fall of 2021
declined relative to before the pandemic. Additionally, borrowers who had outstanding student loan
debt at the time of the survey reported higher levels of financial well-being compared with
prior years.
Incidence and Types of Education Debt
Thirty percent of all adults—representing over 4 in 10 people who went to college—said they
incurred at least some debt for their education. This includes 20 percent of college attendees who
still owed money and 22 percent who borrowed but fully repaid their education debts. Adults under
age 30 who attended college were more likely to have taken out loans than older adults, consis-
tent with the upward trend in educational borrowing over the past several decades (figure 39).
53
53
Student loan borrowing has declined since its peak in 2010–11 but remains substantially above the levels from the mid-
1990s. (Jennifer Ma and Matea Pender, Trends in College Pricing and Student Aid 2021 (New York: The College Board,
2021), https://research.collegeboard.org/pdf/trends-college-pricing-student-aid-2021.pdf).
Figure 39. Acquired debt for own education, including repaid debt (by age and highest degree
completed)
Some college
or technical
degree
Associate
degree
Bachelor’s
degree
Graduate
degree
18–29
30–44
45–59 60+
Percent
37
42
27
13
39
46
38
20
58
60
47
32
58
65
59
39
Note: Among adults who attended college. Key identifies bars in order from top to bottom.
71
The incidence of education debt varied by the type of institution attended. Among those who
attended public institutions, 40 percent either previously held debt or currently had debt at the
time of the survey, compared with 57 percent of those who attended private not-for-profit and
59 percent who attended private for-profit institutions. Among younger cohorts of students, those
who attended private for-profit institutions were also more likely to have taken out student loans
than those who attended either private not-for-profit or public institutions.
Not all education debt is in the form of stu-
dent loans. Ninety-six percent of those with
outstanding debt from their own education
had student loans, but many borrowers had
other forms of education debt as well
(table 21). This includes 19 percent who bor-
rowed with credit cards, 4 percent with a
home equity loan, and 11 percent with some
other form. Collectively, 24 percent of bor-
rowers had one or more forms of education
debt besides student loans for their own
education.
Most student loan borrowers owe less than $25,000 on their loans. The median amount of educa-
tion debt in 2021 among those with any outstanding debt for their own education was between
$20,000 and $24,999. One-quarter of student loan borrowers had less than $10,000 in out-
standing student debt (figure 40). Student debt balances vary across different demographic
groups. Borrowers with an income of less than $50,000 a year were more likely to carry lower bal-
ances of student loan debt.
Table 21. Type of education debt
Percent
Debt type
Own
education
Child’s or
grandchild’s
education
Student loan 96 88
Credit card 19 12
Home equity loan 4 9
Other loan 11 11
Note: Among adults with at least some debt outstanding for
their own education or a child’s or grandchild’s education. Some
people had more than one type of debt.
Figure 40. Distribution of adults with outstanding debt for their own education
$100,000
or above
$75,000–
$99,999
$50,000–
$74,999
$40,000–
$49,999
$30,000–
$39,999
$20,000–
$29,999
$10,000–
$19,999
Less than
$10,000
26
19
15
7
6
10
5
10
Percent
Note: Among adults who borrowed for their own education.
72 Economic Well-Being of U.S. Households in 2021
Some people also took out education debt to assist family members with their education through
either a co-signed loan with the student or a loan taken out independently. Although this is less
common than borrowing for one’s own education, 4 percent of adults owed money for a spouse’s
or partner’s education, and 4 percent had debt that paid for a child’s or grandchild’s education.
Like debt outstanding for the borrower’s education, debt for a child’s or grandchild’s education can
be in forms other than a student loan.
Student Loan Payment Status
The Coronavirus Aid, Relief, and Economic Security Act (CARES Act) and subsequent executive
orders in response to COVID-19 provided federal student loan payment relief throughout 2021, dra-
matically reducing the share of borrowers who were behind on their payments.
54
Among adults
with outstanding debt from their own education, 12 percent were behind on their payments in
2021, a significant decline from the 17 percent who were behind in the fall of 2019, before the
pandemic.
55
Consistent with previous years, borrowers with less education were more likely to be behind on
their payments. Twenty-three percent of borrowers with loans outstanding who completed less than
an associate degree reported being behind.
56
Among borrowers with an associate degree, 18 per-
cent were behind. The delinquency rate was even lower among borrowers with a bachelor’s degree
(6 percent) or graduate degree (5 percent).
Borrowers who said neither of their parents had completed a bachelor’s degree were more likely to
be behind on their payments than those with a parent who had completed a bachelor’s degree. In
2021, borrowers who did not have a parent with a bachelor’s degree were almost twice as likely to
be behind on their payments as those with a parent who completed a bachelor’s degree
(table 22). However, the difference in repayment status among these groups has narrowed since
the fall of 2019.
54
Beginning on March 27, 2020, the CARES Act granted relief to student loan borrowers by temporarily pausing
payments—including principal and interest—on federally held student loans. This payment pause for federal student
loan borrowers has been extended multiple times by executive orders during the COVID-19 pandemic through all of
2021 and into 2022. (See U.S. Department of Education at https://studentaid.gov/announcements-events/
coronavirus).
55
Borrowers could be behind on payments for student loans or other types of debt for their own education. Although the
federal student loan pause has been in effect since March 2020, findings from the 2020 survey did not show substan-
tial improvement in student loan repayment status among borrowers. This could be due to the uncertainty regarding the
policy and interpretation of the survey questions. Data from the Federal Reserve Bank of New York show a decline in
student loan delinquency in 2020 and 2021 (Federal Reserve Bank of New York, Quarterly Report on Household Debt
and Credit (New York: FRB New York, November 2021), https://www.newyorkfed.org/medialibrary/interactives/
householdcredit/data/pdf/HHDC_2021Q3.pdf).
56
Currently enrolled students are frequently not required to make payments so are less likely to fall behind. Among those
with less than an associate degree who are not currently enrolled, a larger 28 percent of borrowers are behind.
Student Loans 73
Difficulties with repayment also vary by race
and ethnicity. While Black and Hispanic bor-
rowers were still disproportionately likely to be
behind on their debt and were less likely to
have completely paid off their student loan
debts, these borrowers saw improvements in
their repayment status. In 2021, 17 percent
of Black borrowers and 18 percent of Hispanic
borrowers reported being behind on their stu-
dent loan debt, compared with 29 and 24 per-
cent in 2019, respectively.
While the percentage of borrowers behind on
payments declined over the prior two years,
disparities in payment status persist based on
the type of institution attended. Twenty-
three percent of borrowers who attended for-
profit institutions were behind on student loan
payments, versus 11 percent who attended
public institutions and 7 percent who attended
private not-for-profit institutions.
Greater difficulties with loan repayment among attendees of for-profit institutions may partly reflect
the lower returns on degrees from these institutions.
57
Indeed, when accounting for race and eth-
nicity, parents’ education, level of institution (two year or four year), and institution selectivity, the
relationship between for-profit institution attendance and being behind on student loan payments
persists. This suggests that the high payment difficulty rates for attendees of for-profit institutions
reflect characteristics of the schools and is not simply due to the characteristics of their students.
Although it is common to focus only on those with outstanding debt, many people who borrowed
for their education had repaid their loans completely. Excluding these people who have paid off
their debt could overstate difficulties with repayment. Indeed, the share of adults who were behind
on their payments is much lower when accounting for all who ever borrowed, including those who
had completely repaid that debt.
Among those who ever incurred debt for their education, 6 percent were behind on their payments
at the time of the 2021 survey, 42 percent had outstanding debt and were current on their pay-
57
See David J. Deming, Claudia Goldin, and Lawrence F. Katz, “The For-Profit Postsecondary School Sector: Nimble Critters
or Agile Predators?” Journal of Economic Perspectives 26, no. 1 (Winter 2012): 139–64, https://www.aeaweb.org/
articles?id=10.1257/jep.26.1.139, for a discussion of the rates of return by education sector.
Table 22. Behind on student loan payments for
own education (by parents’ education,
race/ethnicity, and institution type)
Percent
Characteristic 2019 2021 Change
Parents’ education
Parent has completed a
bachelor’s degree 9 8 −2
Neither parent has completed a
bachelor’s degree 22 15 −7
Race/ethnicity
White 11 9 −2
Black 29 17 −12
Hispanic 24 18 −6
Institution type
Public 15 11 −4
Private not-for-profit 11 7 −4
Private for-profit 27 23 −4
Note: Among adults with outstanding debt for their own educa-
tion. Change reported may not match difference between the
columns because of rounding.
74 Economic Well-Being of U.S. Households in 2021
ments, and 52 percent had completely paid off their loans. Nevertheless, the demographic and
educational characteristics of those who fall behind on payments remain similar when also incor-
porating those who have paid off their loans.
Relation to Financial Well-Being
Consistent with the student loan payment relief and improvements in payment statuses, self-
reported financial well-being among adults with outstanding debt has increased over the pan-
demic. Among all adults who went to college and had outstanding student loan debt, 73 percent
were doing at least okay financially in 2021. This is up from 65 percent who were doing at least
okay financially in 2019.
The improvement in financial well-being among
student loan borrowers occurred among bor-
rowers of all education levels. The 65 percent
of borrowers with an associate degree who
reported doing at least okay financially in
2021 was up 9 percentage points from the
56 percent who were doing at least okay in
2019 (table 23). Among borrowers with some
college education but no associate degree,
the improvement was 7 percentage points.
Among those with at least a bachelor’s
degree, the improvement in financial well-
being was 6 percentage points.
In contrast, adults who attended college and
either did not borrow or had already repaid
their student loan debts did not exhibit similar
improvements in financial well-being. For
those with an associate degree who never borrowed, 83 percent were doing okay financially in
2021, as were 76 percent of those who borrowed and paid off their debt. Each of these were
similar to or below the shares doing at least okay in 2019, standing in contrast to the improve-
ments seen among those with outstanding loans. This suggests that the changes in student loan
policies likely contributed to the increase in self-reported well-being among borrowers.
Relation to Self-Assessed Value of Higher Education
The self-assessed value of higher education was lower among those who had outstanding debt.
Among borrowers with outstanding debt, 40 percent said the benefits of their education exceeded
Table 23. At least doing okay financially
(by education and debt status)
Percent
Characteristic 2019 2021 Change
Some college, no associate degree
Never had debt 77 76 −1
Previously had debt, now repaid 71 74 3
Currently has debt 51 58 7
Associate degree
Never had debt 85 83 −2
Previously had debt, now repaid 79 76 −3
Currently has debt 56 65 9
Bachelor’s degree or more
Never had debt 92 94 2
Previously had debt, now repaid 92 94 2
Currently has debt 75 81 6
Note: Among all adults who attended college.
Student Loans 75
the cost. This was below the 63 percent of borrowers who completely paid off their debt and
51 percent of those who went to college but never had debt.
58
Student loan borrowers with out-
standing debt also were twice as likely as those who repaid their debt to say that the costs of their
education outweigh the benefits.
These gaps in perceptions of one’s higher education were particularly notable among those who
completed a degree. Approximately 3 in 10 adults who attended college but did not complete an
associate or bachelor’s degree said that the benefits of their education exceeded the costs,
regardless of their student loan status. However, substantial gaps in perceptions of higher educa-
tion emerged for those who completed a degree. Just over 3 in 10 associate degree recipients
with outstanding debt said that the benefits exceeded the costs, compared with half of those
without outstanding debt. Among bachelor’s degree recipients, the gap in perceptions between
those with and without outstanding student loan debt is even greater (figure 41). This gap indi-
cates the extent to which perceptions of higher education are linked to whether individuals had to
borrow for their education, and whether the returns on their education were sufficient for them to
repay their student loans.
58
Differences in the level of education within these debt status groups also contribute to the self-assessment of costs
and benefits. Those with a bachelor’s degree or higher make up 71 percent of those who attended college and previ-
ously had debt, compared with 42 percent of those who attended and never had debt.
Figure 41. Self-assessed value of higher education (by education and debt status)
Percent
Costs are greater
Costs and benefits are about the same
Benefits are greater
Currently has debt
Previously had debt,
now repaid
Never had debt
Bachelor’s degree
or more
Currently has debt
Previously had debt,
now repaid
Never had debt
Associate degree
Currently has debt
Previously had debt,
now repaid
Never had debt
Some college, no
associate degree
32
46
22
28
37 35
31
31 38
49
39 11
49
31
19
31
30
39
72
18
10
74
14 12
46
19
35
Note: Among adults who attended college. Key identifies bars in order from left to right.
76 Economic Well-Being of U.S. Households in 2021
Retirement and Investments
In 2021, retirees’ descriptions of their reasons for retirement and their income sources were con-
sistent with recent years. As was the case in 2020, a sizeable share of recent retirees said
COVID-related factors affected the timing of their retirement decision. Among non-retirees, a higher
share reported they felt like their retirement savings were on track, and a smaller share borrowed
against or cashed out retirement savings, compared with 2020. Yet, differences by age and race
or ethnicity in retirement preparedness among non-retirees remained similar to earlier years.
Current Retirees
Retirees represent a sizeable portion of the adult population. Twenty-seven percent of adults in
2021 considered themselves to be retired, even though some were still working in some
capacity.
59
Fourteen percent of retirees had done some work for pay or profit in the prior month.
Consequently, 4 percent of all adults considered themselves retired and were still working.
Retirees with more education were slightly more likely to work in retirement.
In deciding when to retire, most retirees indi-
cated that their preferences played a role, but
life events contributed to the timing of retire-
ment for a substantial share (figure 42). Forty-
nine percent of retirees said a desire to do
other things or to spend time with family was
important for their decision to retire, and
45 percent said they retired because they
reached a normal retirement age.
Nonetheless, 29 percent said that a health
problem was a factor in their decision to
retire, and 15 percent said they retired to care
for family members. One in 10 said they were
forced to retire or that work was not available.
Collectively, health problems, caring for family, and lack of work contributed to the timing of retire-
ment for 45 percent of retirees.
59
In this report, descriptions of current retirees include everyone who reported being retired, including those who also
reported that they are working.
Figure 42. Reasons for the timing of retirement
49
45
14
29
Forced to retire
or lack of work
Care for
family members
Health problem
Life events
Didn’t like the work
Reached normal
retirement age
Wanted to do
other things
Preferences
Percent
15
10
Note: Among retirees. Respondents could select mul-
tiple answers.
77
A sizeable share of recent retirees indicated that COVID-19 was a factor in their retirement deci-
sion. Twenty-five percent of adults who retired in the prior 12 months, and 15 percent of those
who retired one to two years ago, said factors related to COVID-19 contributed to when they
retired. Compared with other retirees, recent retirees whose retirement decision was related to
COVID-19 were more likely to say they retired because they were forced to do so or work was not
available. While the pandemic may be contributing to retirement decisions for some recent
retirees, the share of adults who consider themselves to be retired has remained relatively consis-
tent during the pandemic.
60
Social Security remained the most common
source of retirement income, but 79 percent
of retirees had one or more sources of private
income. This included 57 percent of retirees
with income from a pension; 43 percent with
interest, dividends, or rental income; and
32 percent with labor income (table 24).
61
Seventy-eight percent of retirees received
income from Social Security in the prior 12
months, including 92 percent of retirees age
65 or older.
While retirees as a group report a generally high
level of financial well-being and life satisfaction,
those who were not married and those with a
disability reported lower levels for these subjective measures (table 25).
62
In 2021, 81 percent of
all retirees said they were doing at least okay financially, and 60 percent reported high levels of
life satisfaction. On average, retirees who were not married were not doing as well, with just
68 percent saying that they were doing at least okay financially and 49 percent reporting high
levels of life satisfaction. Retirees with a disability, regardless of their marital status, were less
likely to report they were doing at least okay financially or that they had high levels of life satisfaction.
60
Other recent data have shown an increase in retirements during the pandemic. In part, the difference in findings is
because retirees in the SHED include some who are retired while also working in some capacity, as well as some who
are retired but provide other reasons—such as health limitations—as the reason for not working. An alternative defini-
tion of retirement focuses only on older adults who are not working and who say the reason they are not working is
because they are retired. By this measure in the SHED, 51 percent of adults age 55 or older were retired in 2021—a
share that edged up over the course of the pandemic from 48 percent in 2019 and 49 percent in 2020—consistent
with results from the Current Population Survey using a similar definition (Richard Fry, “Amid the Pandemic, a Rising
Share of Older U.S. Adults Are Now Retired,” web page, Pew Research Center 2021, https://www.pewresearch.org/fact-
tank/2021/11/04/amid-the-pandemic-a-rising-share-of-older-u-s-adults-are-now-retired/).
61
The type of pension was not specified, so pension income may include income from defined benefit plans, which pay a
fixed monthly amount, and defined contribution plans, such as 401(k) and 403(b) plans.
62
About one-third of retirees were not married, and about one-fourth of retirees had a disability.
Table 24. Sources of income in the prior
12 months among retirees (by age)
Percent
Source
Retirees
age 65
and older
All retirees
Social Security 92 78
Pension 66 57
Interest, dividends, or rental income 49 43
Wages, salaries, or self-employment 25 32
Cash transfers other than
Social Security 7 11
Note: Among retirees. Respondents could select multiple
answers. Sources of income include the income of a spouse
or partner.
78 Economic Well-Being of U.S. Households in 2021
Retirement Savings among
Non-Retirees
Although three-fourths of non-retired adults
had at least some retirement savings, about
one-fourth did not have any (figure 43). This
share has remained nearly unchanged since
2019. Among those with retirement savings,
these savings were most frequently in defined
contribution plans, such as a 401(k) or
403(b), with 55 percent of non-retired adults
having money in such a plan. These accounts
were more than twice as common as tradi-
tional defined benefit pension plans. Fifty-
two percent of non-retirees had retirement
savings outside of formal retirement accounts,
up from 48 percent of non-retirees who reported
having such accounts in 2020.
While most non-retired adults had some type
of retirement savings, only 40 percent of non-
retirees thought their retirement saving was
on track. Still, the share of non-retirees who
thought their retirement saving was on track
increased in 2021, from 36 percent who
thought their saving was on track in 2020 and
37 percent who thought their retirement sav-
ings were on track in 2019. Because retire-
ment saving strategies differ by circumstances
and age, survey respondents assessed
whether or not they felt that they were on
track, but they defined that for themselves.
Retirement savings and perceived preparedness differed across demographic groups. Younger
adults were both less likely to have retirement savings and to view their savings as on track than
older adults. Compared with all non-retirees, Black and Hispanic non-retirees were less likely to
have retirement savings and to view their retirement savings as on track, while White and Asian
non-retirees were more likely to have such savings and say they were on track (table 26).
Table 25. Financial well-being and life
satisfaction among retirees (by marital status
and disability status)
Percent
Characteristic
At least
doing okay
financially
High life
satisfaction
Married
No disability 92 70
Disability 77 53
Overall 88 65
Not married
No disability 75 54
Disability 58 40
Overall 68 49
Note: Among retirees.
Figure 43. Forms of retirement savings among
non-retirees
55
52
14
36
None
Business or real estate
Other retirement
savings
Defined benefit pension
IRA
Savings not in
retirement accounts
Defined contribution
pension
Percent
22
13
10
25
Note: Among non-retirees. Respondents could select
multiple answers.
Retirement and Investments 79
The lower rates of savings among Black and
Hispanic non-retirees partly reflect the fact
that Black and Hispanic adults were, on
average, younger than the non-retired popula-
tion overall. Even within age cohorts, however,
significant differences remained in retirement
savings by race and ethnicity, consistent with
patterns seen in previous years.
Non-retirees with a disability were also less
likely to have retirement savings and to view
their savings as on track. Among non-retirees
with a disability, only 49 percent had retire-
ment savings and 17 percent viewed their sav-
ings as on track. Adults with a disability have
a lower rate of employment compared with
adults without a disability. (See box 1 for more
on the employment experiences of adults with
a disability during the pandemic.) In addition,
adults with a disability who receive means-
tested benefits may face asset limits that
would deter holding any savings they may have
accrued.
63
Occasionally, retirement savings can also act
as a source of emergency funds for non-retirees who face economic hardships. Overall, 8 percent
of non-retired adults tapped their retirement savings—a slight decrease from 2020. Yet, 14 per-
cent of non-retired adults who had experienced a layoff in 2021 borrowed or cashed out funds
from their retirement savings. Even so, some non-retirees who may have a need for a reserve fund
to weather a hardship may not have retirement savings or may have already tapped such savings.
Forty-three percent of non-retirees who experienced a layoff in the prior 12 months did not have
self-directed retirement savings at the time of the survey, compared with 26 percent of non-
retirees who did not experience a layoff.
Non-retirees with smaller account balances were more likely to have borrowed from, or cashed out,
funds from their retirement accounts in the prior 12 months (figure 44). Twelve percent of those
63
SSI and Social Security Disability Insurance (SSDI) are federal programs to support adults with a disability who meet
medical and other requirements. SSI recipients must have limited income and resources, but SSDI recipients do not
have to meet income and resource limits to qualify for benefits. See Social Security Administration, Red Book: A Sum-
mary Guide to Employment Supports for Persons with Disabilities Under the Social Security Disability Insurance (SSDI) and
Supplemental Security Income (SSI) Programs, January 2020, https://www.ssa.gov/redbook/.
Table 26. Retirement saving and self-assessed
preparedness (by age, race/ethnicity, and
disability status)
Percent
Characteristic
Any
retirement
savings
Retirement
savings on
track
Age
18–29 62 30
30–44 75 39
45–59 84 45
60+ 87 52
Race/ethnicity
White 81 46
Black 64 26
Hispanic 61 25
Asian 85 52
Disability status
No disability 79 43
Disability 49 17
Overall 75 40
Note: Among non-retirees.
80 Economic Well-Being of U.S. Households in 2021
with account balances under $50,000 bor-
rowed from, or cashed out, these accounts,
compared with 7 percent of those with
account values of $50,000 or more. While
tapping retirement funds could result in
smaller account balances, or the exhaustion
of such reserves altogether, adults with lower
income (and likely with lower account bal-
ances) were more likely to experience shocks
that could prompt them to tap retirement
reserves early.
64
Self-directed retirement accounts frequently
have complex rules on withdrawals and rely on
individuals to have the skills and knowledge
required to manage their own investments. Non-retirees with self-directed retirement savings
varied in their comfort with making investment decisions for their accounts. Nearly 6 in 10 non-
retirees with self-directed retirement savings expressed low levels of comfort in making investment
decisions with their accounts.
Among those non-retirees with self-directed
savings, a higher share of men were comfort-
able managing their retirement investments
compared to women (figure 45). Sixty-four per-
cent of men with a bachelor’s degree were
mostly or very comfortable making investment
decisions, compared to 33 percent of women
with this level of education who were mostly or
very comfortable. In fact, the 33 percent of
women with a bachelor’s degree who were
comfortable investing was similar to the
37 percent of men with a high school degree
or less who expressed the same level of
comfort.
64
For more on early withdrawals and the relationship with economic shocks and income, see Robert Argento, Victoria L.
Bryant, and John Sabelhaus, “Early Withdrawals from Retirement Accounts during the Great Recession,” Contemporary
Economic Policy 33, no. 1 (March 2013), https://www.researchgate.net/publication/
254969212_Early_Withdrawals_from_Retirement_Accounts_during_the_Great_Recession.
Figure 44. Borrowed from or cashed out
retirement savings accounts in the prior 12
months (by amount of self-directed retirement
savings)
Percent
$50,000+
Less than $50,000
None
6
12
7
Note: Among non-retirees.
Figure 45. Mostly or very comfortable
investing self-directed retirement savings (by
gender and education)
Percent
Bachelor’s degree
or more
Some college/
technical or
associate degree
High school
degree or less
Women
Men
37
45
64
29
26
33
Note: Among non-retirees with self-directed retirement
savings. Key identifies bars in order from top
to bottom.
Retirement and Investments 81
Financial Literacy and
Experience with Investments
To get some sense of individuals’ financial
knowledge, respondents were asked three
questions—on interest, inflation, and risk
diversification, respectively—that are com-
monly used as measures of financial literacy
(figure 46).
65
Higher shares of adults provided correct
answers to questions about interest and infla-
tion than to the question on risk diversifica-
tion. The average number of correct answers
was 1.8 out of 3, and 34 percent of adults got
all three correct.
Self-assessed comfort in managing invest-
ments was correlated with these measures of
financial literacy. Among those with self-
directed retirement accounts, on average,
those who expressed comfort with managing
their investments answered a larger share of
questions correctly (78 percent) than those
who expressed little or no comfort (59 per-
cent) (table 27). Notably, the share of incor-
rect answers did not vary much with invest-
ment comfort. Instead, the number of “don’t
know” responses fell as investment comfort
rose. Overall, however, non-retirees with such
accounts still answered more financial literacy
questions correctly, on average, than either
non-retirees who did not have such accounts
or people who were already retired.
65
These questions were developed by Annamaria Lusardi and Olivia Mitchell (see “Financial Literacy around the World: An
Overview,” Journal of Pension Economics and Finance 10, no. 4 (2011): 497–508) and have been widely used to study
financial literacy. In the 2021 SHED, half of the respondents received the questions and answer choices developed by
Lusardi and Mitchell, and the results reported here reflect their responses. The other half of the respondents received
the same questions without the “don’t know” answer option. Full question wording is available in appendix A and results
from the group who received the alternative formulation are included in appendix B of the appendixes to this report.
Figure 46. Responses to financial literacy
questions
Incorrect
Percent
Don’t knowCorrect
Diversification
Inflation
Interest
69 19
12
64
26
10
43
53
4
Note: Among the one-half of respondents who were
asked the questions including “Don't know” as an
answer choice. Key identifies bars in order from left
to right.
Table 27. Financial literacy (by retirement
savings and comfort investing)
Percent
Presence of retirement savings
and level of investing comfort
Correct Incorrect
Don’t
know/
Refused
Has self-directed
retirement savings 67 7 26
Mostly or very comfortable
investing 78 7 15
Not or slightly comfortable
investing 59 8 33
No self-directed
retirement savings 32 11 56
Retired 61 8 31
Overall 59 8 33
Note: Among the one-half of respondents who were asked the
questions including “Don’t know” as an answer choice.
82 Economic Well-Being of U.S. Households in 2021
Gender differences in financial literacy mirrored differences in being comfortable with the invest-
ment decisions. Women, on average, answered a lower share of financial literacy questions cor-
rectly (52 percent) than men (66 percent). Women were also more likely to select “don’t know” or
to skip questions (39 percent) than men (26 percent). As a result, women, on average, had lower
levels of financial literacy by this measure. Some evidence suggests that one driver of this gender
difference may relate to different levels of experience with financial decisions.
66
Financial knowledge appears to be correlated with experience with other investments as well. On
average, people who own individual stock answered 77 percent of the financial literacy questions
correctly. Those who have self-directed retirement savings, but no individual stocks, answered
59 percent correctly, and those with no self-directed retirement savings or stock holdings
answered 32 percent correctly.
The high financial literacy scores among those who own individual stocks is particularly notable
since one of the questions asks if owning individual stocks is riskier than owning a stock mutual
fund. Sixty-three percent of stock owners correctly answered that it is riskier to own an individual
stock, whereas just 33 percent of people who do not own individual stocks answered this question
correctly. Consequently, many people who own individual stocks appear to be aware of this addi-
tional risk and either view the benefits as outweighing the risk or see the individual stocks as part
of their broader investment portfolio.
Financial w ell-being is higher among people with
higher rates of financial knowledge. Among
those who answ ered all three of the financial lit-
eracy questions cor rectly, 89 percent w ere doing
at least oka y financially , whereas a low er 64 per-
cent of those who did not answ er any of the
questions cor rectly w ere doing at least oka y
financially ( figure 47
). This positive relationship
between financial well-being and financial lit-
eracy remains even when looking at people
with the same level of education, although the
magnitude of differences shrinks. Neverthe-
less, at least a portion of the increase in well-
being with additional financial knowledge is
likely attributable to other factors rather than
to differences in financial knowledge alone.
66
Some of the gender gap in financial literacy may relate to specialization in financial tasks within a household, with
women being less likely to handle the finances. Joanne Hsu finds that women’s financial literacy increases after the
death of a spouse (see “Aging and Strategic Learning: The Impact of Spousal Incentives on Financial Literacy,” Journal of
Human Resources 51(4) (Fall 2016): 1036–67).
Figure 47. At least doing okay financially (by
number of financial literacy questions
answered correctly)
Percent
Zero
One
Two
Three
89
79
69
64
Note: Among the one-half of respondents who were
asked the questions including “Don’t know” as
an answer.
Retirement and Investments 83
Description of the Survey
The Survey of Household Economics and Decisionmaking was fielded from October 29 through
November 22, 2021. This was the ninth year of the survey, conducted annually in the fourth
quarter of each year since 2013.
67
Staff of the Federal Reserve Board wrote the survey questions
in consultation with other Federal Reserve System staff, outside academics, and professional
survey experts.
Ipsos, a private consumer research firm, administered the survey using its KnowledgePanel, a
nationally representative probability-based online panel. Since 2009, Ipsos has selected respon-
dents for KnowledgePanel based on address-based sampling (ABS). SHED respondents were then
selected from this panel.
Survey Participation
Participation in the 2021 SHED depended on several separate decisions made by respondents.
First, they agreed to participate in Ipsos’ KnowledgePanel. According to Ipsos, 10.1 percent of
individuals contacted to join KnowledgePanel agreed to join (study-specific recruitment rate). Next,
they completed an initial demographic profile survey. Among those who agreed to join the panel,
61.3 percent completed the initial profile survey and became a panel member (study-specific pro-
file rate). Finally, selected panel members agreed to complete the 2021 SHED.
Of the 18,322 panel members contacted to take the 2021 SHED, 11,965 participated and com-
pleted the survey, yielding a final-stage completion rate of 65.3 percent.
68
Taking all the stages of
recruitment together, the cumulative response rate was 4.0 percent. After removing a small
number of respondents because of high refusal rates or completing the survey too quickly, the
final sample used in the report included 11,874 respondents.
69
Targeted Outreach and Incentives
To increase survey participation and completion among hard-to-reach demographic groups, Board
staff and Ipsos used a targeted communication plan with monetary incentives. The target
groups—young adults ages 18 to 29; adults with less than a high school degree; adults with
67
Data and reports of survey findings from all past years are available at https://www.federalreserve.gov/
consumerscommunities/shed.htm.
68
Three hundred ninety-five respondents were not included in the analysis because they started, but did not complete, the
survey (known as break-offs). The study break-off rate for the SHED was 3.2 percent.
69
Of the 11,965 respondents who completed the survey, 91 were excluded from the analysis in this report because of
either leaving responses to a large number of questions missing, completing the survey too quickly, or both.
85
household income under $50,000 who are under age 60; and those who are a race or ethnicity
other than White, non-Hispanic—received additional email reminders and text messages during
the field period, as well as additional monetary incentives.
All survey respondents not in a target group received a $5 incentive payment after survey comple-
tion. Respondents in the target groups received a $15 incentive. These targeted individuals also
received additional follow-up emails during the field period to encourage completion. Additionally,
the incentives offered to some targeted individuals increased to $25 during the field period to
increase the incentive for completion.
70
Survey Questionnaire
The 2021 survey took respondents 21.6 minutes (median time) to complete.
A priority in designing the survey questions was to understand how individuals and families—
particularly those with low- to moderate-income—fared financially in 2021. The questions were
intended to complement and augment the base of knowledge from other data sources, including
the Board’s Survey of Consumer Finances. In addition, some questions from other surveys were
included to allow direct comparisons across datasets.
71
The full survey questionnaire can be
found in appendix A of the appendixes to this report.
Survey Mode
While the sample was drawn using probability-based sampling methods, the SHED was adminis-
tered to respondents entirely online. Online interviews are less costly than telephone or in-person
interviews and can be an effective way to interview a representative population.
72
Ipsos’ online
panel offers some additional benefits. Their panel allows the same respondents to be
re-interviewed in subsequent surveys with relative ease, as they can be easily contacted for sev-
eral years.
70
All participants received a pre-notification email before the survey launch. They also received four email reminders
during the three-week field period in addition to the initial survey invitation. Targeted respondents received two additional
email reminders over this period. Three days before closing the survey, the email reminder to targeted adults increased
the incentive for completing the survey from $15 to $25. Of the 5,733 respondents in a targeted group, 251 received
the higher $25 incentive payment and the rest received the $15 incentive payment.
71
For a comparison of results to select overlapping questions from the SHED and Census Bureau surveys, see Jeff Larri-
more, Maximilian Schmeiser, and Sebastian Devlin-Foltz, “Should You Trust Things You Hear Online? Comparing SHED
and Census Bureau Survey Results,” Finance and Economics Discussion Series Notes (Washington: Board of Governors
of the Federal Reserve System, October 15, 2015), https://doi.org/10.17016/2380-7172.1619.
72
David S. Yeager et al., “Comparing the Accuracy of RDD Telephone Surveys and Internet Surveys Conducted with Prob-
ability and Non-Probability Samples,” Public Opinion Quarterly 75, no. 4 (2011): 709–47.
86 Economic Well-Being of U.S. Households in 2021
Furthermore, internet panel surveys have numerous existing data points on respondents from pre-
viously administered surveys, including detailed demographic and economic information. This
allows for the inclusion of additional information on respondents without increasing respondent
burden.
73
The respondent burdens are further reduced by automatically skipping irrelevant ques-
tions based on responses to previous answers.
The “digital divide” and other differences in internet usage could bias participation in online sur-
veys, so recruited panel members who did not have a computer or internet access were provided
with a laptop and access to the internet to complete the surveys. Even so, individuals who com-
plete an online survey may have greater comfort or familiarity with the internet and technology
than the overall adult population, which has the potential to introduce bias in the characteristics of
who responds.
Sampling and Weighting
The SHED sample was designed to be representative of adults age 18 and older living in the
United States.
The Ipsos methodology for selecting a general population sample from KnowledgePanel ensured
that the resulting sample behaved as an equal probability of selection method (EPSEM) sample.
This methodology started by weighting the entire KnowledgePanel to the benchmarks in the latest
March supplement of the Current Population Survey along several geo-demographic dimensions.
This way, the weighted distribution of the KnowledgePanel matched that of U.S. adults. The geo-
demographic dimensions used for weighting the entire KnowledgePanel included gender, age, race,
ethnicity, education, census region, household income, homeownership status, and metropolitan
area status.
Using the above weights as the measure of size (MOS) for each panel member, in the next step a
probability proportional to size (PPS) procedure was used to select study specific samples. This
methodology was designed to produce a sample with weights close to one, thereby reducing the
reliance on post-stratification weights for obtaining a representative sample.
After the survey collection was complete, statisticians at Ipsos adjusted weights in a post-
stratification process that corrected for any survey non-response as well as any non-coverage or
under- and oversampling in the study design. The following variables were used for the adjustment
of weights for this study: age, gender, race, ethnicity, census region, residence in a metropolitan
area, education, and household income. Demographic and geographic distributions for the
73
This approach also may allow for the retroactive linking of information learned about respondents from other data, as
was done in 2021 to determine Asian respondents in earlier years of the survey.
Description of the Survey 87
noninstitutionalized, civilian population age 18 and older from the March Current Population
Survey were the benchmarks in this adjustment. Household income benchmarks were obtained
from the 2021 March Current Population Survey (CPS).
One feature of the SHED is that a subset of respondents also participated in prior waves of the
survey. In 2021, about one-third of respondents had participated in the fall 2020 survey. Prior
year case identifiers for these repeat respondents are available in the publicly available dataset,
along with weights for this subset of respondents. These weights use a similar procedure as
described above to ensure estimates based on the repeated sample are representative of the
U.S. population.
Although weights allow the sample population to match the U.S. population (excluding those in the
military or in institutions, such as prisons or nursing homes) based on observable characteristics,
similar to all survey methods, it remains possible that non-coverage, non-response, or occasional
disparities among recruited panel members result in differences between the sample population
and the U.S. population. For example, address-based sampling likely misses homeless popula-
tions, and non-English speakers may not participate in surveys conducted in English.
74
Despite an effort to select the sample such that the unweighted distribution of the sample more
closely mirrored that of the U.S. adult population, the results indicate that weights remain neces-
sary to accurately reflect the composition of the U.S. population. Consequently, all results pre-
sented in this report use the post-stratification weights produced by Ipsos for use with the survey.
Item Non-response and Imputation
Item non-response in the 2021 SHED was handled by imputation. Typically, less than 1 percent of
observations were missing for each question.
75
As a result, population estimates were not sensi-
tive to the imputation procedure and a simple regression approach was used.
76
74
For example, while the survey was weighted to match the race and ethnicity of the entire U.S. adult population, there is
evidence that the Hispanic population in the survey were somewhat more likely to speak English at home than the
overall Hispanic population in the United States. In the 2021 SHED, the 60 percent of Hispanic adults who speak
Spanish at home is below estimates from the 2019 American Community Survey. See table B16006 at https://
data.census.gov. See the Report on the Economic Well-Being of U.S. Households in 2017 for a comparison of results to
select questions administered in Spanish and English at https://www.federalreserve.gov/publications/2018-economic-
well-being-of-us-households-in-2017-preface.htm.
75
Because item non-response is very low in the SHED, 2021 estimates are comparable with prior years where item non-
response was handled differently.
76
A logit regression was used for binary variables, a multinomial logit for categorical variables, an ordinal logit for ordered
values, and a linear regression for continuous values. Typical predictors included income, education, race and ethnicity,
age, gender, and metropolitan status, but varied depending on how well they predicted the variable of interest and item
non-response. Additional predictors were included as appropriate.
88 Economic Well-Being of U.S. Households in 2021
The imputation procedure was carried out as follows:
1. Impute questions, like income and education, to be used in the imputation models throughout.
2. Continue at the beginning of the survey and impute missing values sequentially, question by
question.
In some cases, the imputation for one question affected later questions by switching an observa-
tion from out-of-universe to in-universe or vice versa. These cases were handled by imputing the
missing “downstream” question response or recoding it to missing, where appropriate.
Each variable in the publicly available SHED dataset has a corresponding imputation flag,
‘var’_iflag, which is set to 1 if the observation was imputed and 0 otherwise.
77
For example, the
first question of the survey about whether the respondent lived with their spouse or partner, L0_a,
has a corresponding imputation flag of L0_a_iflag. This question had 31 missing values that were
imputed, accounting for 0.3 percent of all observations.
77
The survey data can be downloaded at https://www.federalreserve.gov/consumerscommunities/shed_data.htm.
Description of the Survey 89
Acknowledgments
This survey and report were prepared by the Consumer and Community Research Section of the
Federal Reserve Board’s Division of Consumer and Community Affairs (DCCA).
DCCA directs consumer- and community-related functions performed by the Board, including con-
ducting research on financial services policies and practices and their implications for consumer
financial stability, community development, and neighborhood stabilization.
DCCA staff members Alicia Lloro, Ellen Merry, Kenneth Brevoort, Kayla Jones, Jeff Larrimore, Jacob
Lockwood, Anna Tranfaglia, Erin Troland, Douglas Webber, and Mike Zabek prepared this report.
Federal Reserve staff members Laura Benedict, Andrea Brachtesende, David Buchholz, Ellen Levy,
Madelyn Marchessault, Kirk Schwarzbach, and Pamela Wilson provided valuable feedback on the
report. The authors would like to thank Eric Belsky, Anna Alvarez Boyd, Eileen Divringi, Dan Gorin,
Simona Hannon, Geng Li, Tenisha Noel, Kamila Sommer, Sarah Stein, and Ann Thompson for their
contributions to new survey questions. The authors would also like to thank Bob Torongo,
Frances Barlas, Poom Nukulkij, and Alyssa Marciniak for their assistance fielding the survey.
If you have questions about the survey or this report, please email [email protected]v.
91
Corrections
The Federal Reserve revised this report on August 22, 2022, to reflect corrected data
described below.
On page 38, in the Health-Care Expenses section, figure 22, “Forms of medical treatment skipped
because of cost during 2021,” data were corrected for the entry “Dental care” from 13 percent to
17 percent and for the entry “Seeing a doctor or specialist” from 17 percent to 13 percent.
The Federal Reserve revised this report on February 2, 2023, to reflect the corrections
described below.
On page 12, in the Overall Financial Well-Being section, in table 1, “At least doing okay financially
(by demographic characteristics),” “LGBTQ+ status was first identifiable in the 2020 survey” was
corrected to “LGBTQ+ status was first identifiable in the 2019 survey.”
On page 54, in the Housing section, the title of figure 29 was corrected from “Distance to friends,
families, and workplaces in 2020” to “Distance to friends, families, and workplaces after moves”
and the bar lengths were adjusted to match the figure values.
On page 63, in the Education section, the bar lengths of figure 33, “Parent’s views on precautions
taken by child’s school (by income and race/ethnicity),” were adjusted to match the figure values.
On page 71, in the Education section, the bar lengths of figure 39, “Acquired debt for own educa-
tion, including repaid debt (by age and highest degree completed),” were adjusted to match the
figure values.
On page 76, in the Education section, text in the footnote of figure 41, “Self-assessed value of
higher education (by education and debt status),” was corrected from “Among adults ages who
borrowed for their own education” to “Among adults who attended college.”
On page 83, in the Retirement and Investments section, in figure 47, “At least doing okay finan-
cially (by number of financial literacy questions answered correctly),” the text “Key identifies bars
in order from left to right” was removed from the footnote.
93
Find other Federal Reserve Board publications (www.federalreserve.gov/publications.htm) or order
those offered in print (www.federalreserve.gov/files/orderform.pdf) on our website. Also visit the site for more
information about the Board and to learn how to stay connected with us on social media.
www.federalreserve.gov
0522
