Tracking the COVID-19 Economy’s Effects on Food, Housing, and Employment Hardships
While employment is rising and strains on household budgets have eased in recent months, the employment rate remains below pre-pandemic levels, and millions still report that their households did not get enough to eat or are not caught up on rent payments. We are able to track the extent of the nation’s progress against hardship thanks to nearly real-time data from the Census Bureau’s Household Pulse Survey and other sources.
Key hardship indicators show strong improvement since December 2020, Census Bureau data show, aided by job growth and government benefits. Hardship rates fell especially fast after the enactment of the American Rescue Plan on March 11, 2021, which included $1,400 payments for most Americans as well as other assistance to struggling households. (See Figure 1.) Food hardship among adults with children also fell after the federal government began issuing monthly payments of the expanded Child Tax Credit on July 15, along with improvements in food assistance. Still, some 20 million adults live in households that did not get enough to eat, 10 million adult renters are behind on rent, and some of the progress from late March appears to have stalled.
The impacts of the pandemic and the economic fallout have been widespread, but remain particularly prevalent among Black adults, Latino adults, and other people of color. These disproportionate impacts reflect harsh, long-standing inequities — often stemming from structural racism — in education, employment, housing, and health care that the current crisis has exacerbated. Households with children also continue to face especially high hardship rates. Considerable evidence suggests that reducing childhood hardship and poverty would yield improvements in education and health, higher productivity and earnings, less incarceration, and other lasting benefits to children and society.
Census Bureau Data Show High Rates of Hardship
The Census Bureau’s Household Pulse Survey, launched in April 2020, has provided nearly real-time data on how the unprecedented health and economic crisis is affecting the nation. Data from this and other sources, such as unemployment data from Census’ Current Population Survey and the Department of Labor, show that millions of people are out of work and struggling to afford adequate food and pay the rent. The impacts on children are large (see Figures 2, 7, and 8).
For more on our methodology and data by state, see tables 1-4 at the end of this document.
Difficulty Getting Enough Food
Some 20 million adults — 9 percent of all adults in the country — reported that their household sometimes or often didn’t have enough to eat in the last seven days, according to Household Pulse Survey data collected September 15-27. When asked why, 79 percent said they “couldn’t afford to buy more food,” rather than (or in addition to) non-financial factors such as lack of transportation or safety concerns due to the pandemic.
Adults in households with children were likelier to report that the household didn’t get enough to eat: 11 percent, compared to 8 percent for households without children. (See Figure 2.) And 7 to 12 percent of adults with children reported that their children sometimes or often didn’t eat enough in the last seven days because they couldn’t afford it. Households typically first scale back on food for adults before cutting back on what children have to eat. (The 7-12 percent range reflects the different ways to measure food hardship in the Household Pulse Survey.)
Also, analysis of more detailed data from the Pulse Survey collected September 1-13 shows that between 5 and 8 million children live in a household where children didn’t eat enough because the household couldn’t afford it. These figures are approximations; the Pulse Survey was designed to provide data on adult well-being, not precise counts of children.
Black and Latino adults were more than twice as likely as white adults to report that their household did not get enough to eat: 18 percent for Black adults and 15 percent for Latino adults, compared to 7 percent of white adults. Adults who identify as American Indian, Alaska Native, Native Hawaiian, Pacific Islander, or as multiracial, taken together, were more likely than white adults to report that their household did not get enough to eat, at 12 percent. (See Figure 3.)
The share of adults reporting that their households did not get enough to eat rose in the fall of 2020, reaching a peak of nearly 30 percent in mid-December of 2020. It fell sharply in March 2021 after the enactment of the December relief package and the mid-March enactment of the American Rescue Plan. Food hardship among adults with children improved significantly following the issuance of the first monthly Child Tax Credit payment on July 15, as well as continuing economic growth and improvements in food assistance.
Inability to Pay Rent or Mortgage
The Household Pulse data also show that millions are not caught up on their rent or mortgage payments. Unfortunately, there are two concerns with the housing questions. First, the Census Bureau reworded the rent payment question starting with the late-August 2020 survey, making the results non-comparable to earlier weeks of the survey. Second, Census at the same time made the entire survey longer, which led more respondents to skip questions toward the end of the survey, including the housing questions. This “non-response” is higher among groups that are younger, have lower levels of education, and identify as Black or Latino — groups that are more likely to struggle to afford rent, due to long-standing inequities often stemming from structural racism in education, employment, and housing. For these reasons, the Pulse data likely understate the number of people struggling to pay rent.
Even with these issues, however, the data indicate that millions are having difficulty paying rent. An estimated 10.3 million adults living in rental housing — 14 percent of adult renters — were not caught up on rent, according to data collected September 15-27. Here, too, renters of color were more likely to report that their household was not caught up on rent: 23 percent of Black renters, 17 percent of Latino renters, and 21 percent of Asian renters said they were not caught up on rent, compared to 9 percent of white renters. The rate was 17 percent for American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and multiracial adults taken together. (See Figure 5.)
The number of adult renters reporting to the Census Bureau that their household was not caught up on rent has fallen from a peak of 15 million people — 1 in 5 adult renters — in January 2021 but has remained above 10 million people — about 1 in 7 adult renters — since the end of March. (See Figure 6.) These households, particularly those who have lost employment during the pandemic, may be accumulating debt from multiple months of back rent and late fees. Renters of color and families with children have consistently reported higher rates of rent hardship throughout 2020 and 2021.
The December relief package and the American Rescue Plan included over $46 billion in emergency rental assistance, designed to help people who are struggling to pay their rent and avoid eviction. However, this emergency aid is still making its way to people behind on rent. States and localities are working to get these funds to renters in need, but many communities did not have adequate systems in place to distribute emergency rental assistance funds quickly. As a result, some states and localities are now building the infrastructure for people to apply for and receive emergency assistance. The Supreme Court’s ruling ending the Centers for Disease Control and Prevention’s eviction moratorium places people behind on rent who have not received the much-needed aid at immediate risk of losing their homes.
In addition, 21 percent of renters who are parents or otherwise live with children reported that they were not caught up on rent, compared to 10 percent among adults not living with anyone under age 18. (See Figure 7.)
Children in renter households also face high rates of food hardship. One in 5 children living in rental housing live in a household that didn’t have enough to eat, according to data for September 1-13. And over 1 in 3 children living in rental housing live in a household that either isn’t getting enough to eat or is not caught up on rent. (See Figure 8.)
While households that don’t rent their homes but have mortgage payments typically have higher incomes than renters, they, too, can face difficulties, especially if they have lost their jobs or seen their incomes fall significantly. An estimated 7.6 million adults are in a household that is not caught up in its mortgage payment.
Difficulty Covering Household Expenses
Since late August 2020, the Household Pulse Survey has provided data on the overall number of adults struggling to cover usual household expenses such as food, rent or mortgage, car payments, medical expenses, or student loans. Some 63 million adults — 28 percent of all adults in the country — reported it was somewhat or very difficult for their household to cover usual expenses in the past seven days, according to data collected September 15-27.
The share of adults reporting difficulty covering usual expenses rose through the fall of 2020, reaching a peak of 38 percent in mid-December. (See Figure 9.) This likely reflected, in part, weaknesses of the relief packages enacted in the spring of 2020, including increased jobless benefits that expired over the summer, stimulus payments whose impact faded later in the year, and inadequate nutrition and housing assistance.
In early 2021, the share of adults with trouble covering expenses stabilized as aid from the end-of-year relief package — including renewed jobless benefits and another round of stimulus payments — reached households.
Following the enactment of the American Rescue Plan on March 11, and as the economy added jobs, the share of adults who had trouble covering usual expenses fell dramatically. The share ticked upward in May, likely due to the fading impact of the third round of stimulus payments, but it has remained statistically unchanged in recent months.
Adults in households with children were more likely to report difficulty paying for usual expenses than those without children: 35 percent, compared to 24 percent. Financial hardship can have serious effects on children’s long-term health and education, research shows.
Black and Latino adults reported difficulty covering expenses at higher rates: 44 percent and 39 percent respectively, compared to 23 percent for white adults and for Asian adults. (See Figure 10.) The rate was 38 percent for American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and multiracial adults taken together.
Adults with a disability were more than twice as likely to report difficulty paying for usual expenses than adults without a disability: 51 percent compared to 23 percent, according to detailed Pulse data collected September 1-13. In addition, LGBT adults were likelier than non-LGBT adults to live in households with difficulty covering expenses: 36 percent compared to 27 percent.
An estimated 36 percent of children live in households that have trouble covering usual expenses, according to our analysis of detailed data from the Pulse Survey collected September 1-13. They include 52 percent of children in Black households, 49 percent of children in Latino households, 28 percent of children in white households, and 22 percent of children in Asian households. (The Pulse Survey asks the race of the adult respondent, not the children.)
Many Workers Remained Sidelined, With Job Losses Concentrated in Low-Paid Industries
The unemployment rate jumped in April 2020 to a level not seen since the 1930s — and still stood at 4.8 percent in September 2021, compared with 3.5 percent in February 2020. The official unemployment rate, however, understates current job losses.
While the economy has added jobs in recent months, there were still 5.0 million fewer jobs in September 2021 than in February 2020. The majority of jobs lost in the crisis have been in industries that pay low average wages, with the lowest-paying industries accounting for 30 percent of all jobs but 56 percent of the jobs lost from February 2020 to September 2021, according to the latest month of Labor Department employment data. Jobs were down nearly twice as much in low-paying industries (5.3 percent) as in medium-wage industries (3.2 percent) and more than six times as much as in high-wage industries (0.8 percent) during this period. (See Figure 11.)
Black and Latino workers have experienced a far slower jobs recovery than white workers — reflecting historical patterns rooted in structural racism.  Some 7.9 percent of Black workers and 6.3 percent of Latino workers were unemployed in September, compared to 4.2 percent of white workers. Workers who were born outside the United States (this includes individuals who are now U.S. citizens) have experienced larger job losses than U.S.-born workers.
Data from the Census Bureau’s basic monthly Current Population Survey released September 8, 2021, provide more detail on unemployed workers and their family members. Some 22.3 million people either met the official definition of unemployed (meaning they actively looked for work in the last four weeks or were on temporary layoff) or lived with an unemployed family member in August. This figure includes 5.7 million children.
The official definition of unemployed leaves out many workers deprived of pay amid the pandemic, including some 1.5 million workers in August who reported they did not look for work because of the pandemic, according to the Labor Department. The official definition also omits over 200,000 workers who reported that they had a job but that they were absent from work without pay and lost pay in the last four weeks “because their employer closed or lost business due to the coronavirus pandemic,” we calculate.
When family members are considered, some 26.3 million people in August, including 6.8 million children, lived in a family where at least one adult did not have paid work in the last week because of unemployment or the pandemic, we estimate.
While policymakers expanded unemployment insurance eligibility and enhanced benefits during the COVID-19 emergency, these measures were temporary. Some 31 percent of all unemployment claims for the week ending September 18, 2021, comprising temporary, pandemic-related programs, regular state unemployment insurance, and other permanent programs were in the pandemic-related programs, which terminated on September 6. Permanent reforms are needed to fix an underlying system in which too many unemployed workers get inadequate benefits or no benefits at all.
State-by-State Food, Housing, and Employment Hardship Data
Data by state show that hardship is widespread. The following tables provide state-level data on:
- the share of adults reporting that their household didn’t have enough to eat (Table 1);
- the share of adults saying children in their household were not eating enough because they couldn’t afford enough (Table 1);
- the share of adults not caught up on rent (Table 2);
- the share of adults saying their household had difficulty paying for their usual expenses (Table 3); and
- the three-month moving average unemployment rate and recent jobless claim data (Table 4).
For data from the Pulse Survey we average data collected September 1-13 and September 15-27 to improve the accuracy of the state estimates.
Differences in Pulse hardship rates between states may reflect sampling error, so we suggest not drawing strong conclusions from modest differences between states. The data do show, however, that high levels of hardship are widespread across the country.
Difficulty Getting Enough Food
The Pulse Survey asks adult respondents if their household did not have enough to eat and if children in the household were not eating enough because the household couldn’t afford it.
|High Shares of Households Report Difficulty Getting Enough Food|
Among adults; data collected September 1-27.
How to read this table: In the United States, some 19 million adults reported that their household sometimes or often didn’t have enough to eat in the last seven days. This represents 9 percent of all adults in the country. Nearly 9 million adults living with children reported that “the children were not eating enough because we just couldn’t afford enough food.” This represents 12 percent of adults living with children.
|Adults Reporting That Household Didn’t Have Enough to Eat||Adults Reporting That Children in Household Weren’t Eating
Enough Because Household Couldn’t Afford Enough
|State||Number||As a Percent of Adults||Number||As a Percent of Adults Living with Children|
|District of Columbia||30,000||6%||15,000||10%|
Not Caught Up on Rent
Table 2 shows the estimated number of adults whose household was not caught up on rent by state. The Census Bureau reworded the Pulse Survey’s rent payment question starting with the late-August 2020 survey, so these rent hardship figures results are not comparable to data from earlier weeks of the Pulse Survey. In addition, Census at the same time made the entire survey longer, which led more respondents to skip questions toward the end of the survey, including the housing questions. Non-response is higher among groups that are younger, have lower levels of education, and identify as Black or Latino — groups that are more likely to struggle to afford rent, due to longstanding inequities that often stem from structural racism in education, employment, and housing. Therefore, the Pulse data likely understate the number of people struggling to pay rent.
|One in 7 Renters Nationwide Not Caught Up on Rent|
|Among adults in rental housing; data collected September 1-27|
|Not Caught Up On Rent|
|District of Columbia||34,000||12%|
Difficulty Covering Usual Household Expenses
The Pulse Survey asks adult respondents if their household had difficulty paying for usual expenses such as food, rent or mortgage, car payments, medical expenses, or student loans in the last seven days. Table 3 shows the estimated number and percent of adults reporting that it was somewhat or very difficult for their household to pay for their usual expenses in the last seven days.
|Over 1 in 4 Adults Nationwide Have Difficulty Covering Usual Household Expenses|
|Among adults; data collected September 1-27|
|Difficulty Covering Usual Household Expenses|
|District of Columbia||111,000||22%|
Table 4 provides state-by-state data on the unemployment rate over the June–August 2021 period and data on ongoing unemployment benefit claims.
|Unemployment, Jobless Claims Remain Elevated Across Most of the Country|
|States||Unemployment rate (June–August average)a||Current jobless benefits claims for week ending September 18b|
|District of Columbia||6.7||14,000|
 Federal surveys generally ask respondents whether they are “of Hispanic, Latino, or Spanish origin.” This report uses the term “Latino.”
 Claire Zippel and Arloc Sherman, “Bolstering Family Income Is Essential to Helping Children Emerge Successfully From the Current Crisis,” CBPP, updated February 25, 2021, https://www.cbpp.org/research/poverty-and-inequality/bolstering-family-income-is-essential-to-helping-children-emerge.
 The Pulse Survey does not provide data for these groups individually.
 The latest Pulse Survey estimates that 7.3 million adults live in households not caught up on rent. To adjust for non-response in the survey, we apply the share not caught up on rent (14.1 percent) to the total number of adult renters (73 million) in the March 2020 Current Population Survey to calculate an adjusted estimate.
 The latest Pulse Survey estimates that 6.7 million adults are in households not caught up on their mortgage. To adjust for non-response in the survey, we apply the share not caught up on their mortgage (7.6 percent) to the total number of adult homeowners (about 100 million) in the March 2020 Current Population Survey to calculate an adjusted estimate.
 Ajay Chaudry and Christopher Wimer, “Poverty is Not Just an Indicator: The Relationship Between Income, Poverty, and Child Well-Being,” Academic Pediatrics, Vol. 16, Issue 3, April 1, 2016, https://www.academicpedsjnl.net/article/S1876-2859(15)00383-6/fulltext.
 Starting in mid-April 2021, the Pulse survey asks respondents whether they have difficulty seeing, hearing, remembering or concentrating, or walking or climbing stairs. In this report, adults with a disability are those who reported “a lot of difficulty” with, or “could not do at all”, one or more of these four activities. This definition, like others, may not accurately reflect the identities and experiences of all disabled people.
 Starting in late July 2021, the Pulse survey asks respondents about their sex assigned at birth, gender identity, and sexual orientation. The Census Bureau categorizes Pulse respondents as LGBT if they identify as gay, bisexual, transgender, or as having a gender identity that doesn’t align with the sex they were assigned at birth. Respondents whose sex at birth aligns with their gender identity and who identify as straight are categorized as non-LGBT.
 Chad Stone, “Robust Unemployment Insurance, Other Relief Needed to Mitigate Racial and Ethnic Unemployment Disparities,” CBPP, August 5, 2020, https://www.cbpp.org/research/economy/robust-unemployment-insurance-other-relief-needed-to-mitigate-racial-and-ethnic.
 Many analysts reach a similar conclusion using a slightly different approach, noting that the official unemployment rate is too low because it omits workers who have exited the labor force in the last 12 months and are no longer looking for work, and because it ignores workers whom the Labor Department says are improperly classified as employed in its survey data but are in fact absent from work. When these two factors are corrected using an approach recommended by Federal Reserve Chair Jerome Powell, the unemployment rate for August 2021 could be as high as 7.8 percent. Jerome H. Powell, “Recent Economic Developments and the Challenges Ahead,” speech at the National Association for Business Economics Virtual Annual Meeting, October 6, 2020, https://www.federalreserve.gov/newsevents/speech/powell20201006a.htm.
 For programs that terminated September 6, unemployed workers could still submit claims after that date for weeks of unemployment that predated the programs’ expiration.
 Chad Stone, “Congress Should Heed President Biden’s Call for Fundamental UI Reform,” CBPP, May 5, 2021, https://www.cbpp.org/research/economy/congress-should-heed-president-bidens-call-for-fundamental-ui-reform.