BLS Consumer Spending 2022 PDF

January 2023 Report 1101
Consumer Expenditures in 2021

ERRATA
Consumer Expenditures in 2021
The year 2020 was marked by the onset of the COVID-19 pandemic and an associated decline in consumer spending. As COVID-19 restrictions were repealed and the
economy began to open, one would expect 2021 to have been welcomed as the beginning of a strong recovery to pre-pandemic levels for expenditures on
entertainment, travel, food, apparel, gasoline, healthcare, and more. However, while certain constraints (e.g., government-mandated business closures) were loosened
from the early pandemic days, consumers faced a number of challenges from the pandemic that carried into 2021. Among those issues were rising prices, global supply
chain constraints and microchip shortages, and further sweeping COVID-19 waves.1
How did these new challenges affect the spending patterns of the typical household? This report analyzes integrated data from the Diary and Interview Survey
components of the Consumer Expenditure Surveys (CE). (For details about these components, including how their results are integrated in publication, see the technical
notes section.) The data in this report is drawn from a collection of CE tables. The incomes and expenditures shown in the tables throughout this report are expressed as
nominal values, representing spending in U.S. dollars as reported by surveyed consumers.2 Through the CE, the Bureau of Labor Statistics (BLS) collects information from
the reference person of consumer units (CUs).3
Given the unusual conditions facing consumers in 2020 and 2021, the next section provides context for the economic conditions that influenced expenditure trends
before and during 2021.
Effects of inflation and supply chain disruptions in 2021

Prices rose 4.7 percent from 2020 to 2021, as measured by the Consumer Price Index (CPI) for All Urban Consumers (CPI-U, U.S. city average, all items, base period
1982–84=100), compared with 1.2 percent from 2019 to 2020. In each successive month of 2021, inflation had been on a continual upward trend, beginning with a 12-
month percent change of 1.4 percent in January and reaching 7.0 percent in December. Key expenditures that have been especially affected by inflation in 2021 include
gasoline, used cars and trucks, and food at home and away from home. The average annual change from 2020 to 2021 in gasoline prices alone was 36.0 percent. Other
notable price increases include a 26.6-percent increase for used cars and trucks and a 3.9-percent increase for food. For a snapshot of how inflation could impact
measures of consumer spending in 2021, take a theoretical family that spent 5.0 percent more on goods and services relative to 2020.
The 5.0 percent represents their nominal spending growth. Nominal spending is the amount, measured in dollars, that the consumer provided in exchange for the good
or service of interest; all data in the CE are reported and published in nominal terms. As mentioned previously, average annual inflation rose 4.7 percent. Real spending
growth, defined as the rate of nominal spending growth minus the rate of inflation, was 0.3 percent (i.e., 5.0 minus 4.7 percent), meaning that much of the increase in
the family’s spending can be attributed to higher prices rather than to an increase in quantities purchased or living standards.
Two factors that simultaneously accompanied and contributed to inflation in 2021 were an ongoing supply chain bottleneck and the associated semiconductor chip
shortage. Microchips are integral components of a number of goods a typical CU purchases on an annual basis. Whether the good is a cell phone, laptop, or a new
vehicle parked in the driveway, all utilize microchips. When critical components to essential electronic devices are hard to find, this naturally reduces the capacity to
produce those goods, which negatively affects the overall supply and leads to higher prices. Compounding the pressure on the average CU of inflation and supply chain
constraints, further waves of COVID-19 emerged in 2021. In the latter half of 2021, the more transmissible Delta and Omicron strains caused many companies to extend
their work from home policies and caused consumer expenditures on travel to remain moderately unstable.4 The emergence of more transmissible strains could have put
downward pressure on spending demand for the most-at-risk cohorts, at least for expenditures such as food away from home and entertainment. The confluence of
these three factors was certainly felt by consumers as they plotted their spending course for 2021.
Expenditure trends: 2018 to 2021
Table A depicts annual expenditure values and their associated percent changes for both total expenditures and its major components (i.e., expenditure categories such
as food) between 2018 and 2021. Average annual expenditures rose by 9.1 percent to $66,928 in 2021, up almost $5,000 from 2020 and up $3,900 from 2019—a
noticeable reversal from the 2.7-percent decline from 2019 to 2020. Income continued to grow from 2020 levels but had not reached the level of income growth seen in
2019, when income rose 5.4 percent. Average annual income rose 3.7 percent from $84,352 to $87,432 at the CU level.
Table A. Average annual expenditures by major category of all consumer units and
percent changes, Consumer Expenditure Surveys, 2018–21
2018 2019 2020 2021 Percent change
Item
2018–19 2019–20 2020–21
Number of consumer units (in thousands)[1] 131,439 132,242 131,234 133,595 [2] [2] [2]
Consumer unit characteristics:
Income before taxes $78,635 $82,852 $84,352 $87,432 5.4 1.8 3.7
Age of reference person 51.1 51.6 52.2 51.8 [2] [2] [2]
Average number in consumer unit:
People 2.5 2.5 2.5 2.4 [2] [2] [2]
Children under 18 .6 .6 .6 .6 [2] [2] [2]
Adults 65 and older .4 .4 .4 .4 [2] [2] [2]
Earners 1.3 1.3 1.3 1.3 [2] [2] [2]
Vehicles 1.9 1.9 1.9 1.9 [2] [2] [2]
Percent homeowner 63 64 66 65 [2] [2] [2]
Average annual expenditures $61,224 $63,036 $61,332 $66,928 3.0 -2.7 9.1
Food 7,923 8,169 7,310 8,289 3.1 -10.4 13.4
Food at home 4,464 4,643 4,935 5,259 4.0 6.3 6.6
Cereals and bakery products 569 583 640 672 2.5 9.8 5.0
Meats, poultry, fish, and eggs 961 980 1,070 1,115 2.0 9.2 4.2
Dairy products 449 455 474 492 1.3 4.2 3.8
Fruits and vegetables 858 876 976 1,033 2.1 11.4 5.8
Other food at home 1,627 1,749 1,776 1,947 7.5 1.5 9.6
Food away from home 3,459 3,526 2,375 3,030 1.9 -32.6 27.6
Alcoholic beverages 583 579 478 554 -.7 -17.4 15.9
Housing 20,091 20,679 21,417 22,624 2.9 3.5 5.6
Shelter 11,747 12,190 12,604 13,258 3.8 3.4 5.2
Owned dwellings 6,678 6,797 7,473 7,591 1.8 9.9 1.6
Rented dwellings 4,249 4,432 4,408 4,684 4.3 -.5 6.3
Other lodging 821 961 722 983 17.1 -24.9 36.1
Utilities, fuels, and public services 4,049 4,055 4,166 4,223 .1 2.7 1.4
Household operations 1,522 1,570 1,465 1,638 3.2 -6.7 11.8
Housekeeping supplies 747 766 837 803 2.5 9.3 -4.1
Household furnishings and equipment 2,025 2,098 2,346 2,701 3.6 11.8 15.1
Apparel and services 1,866 1,883 1,434 1,754 .9 -23.8 22.3
Transportation 9,761 10,742 9,826 10,961 10.1 -8.5 11.6
Vehicle purchases (net outlay) 3,975 4,394 4,523 4,828 10.5 2.9 6.7
Gasoline, other fuels, and motor oil 2,109 2,094 1,568 2,148 -.7 -25.1 37.0
Other vehicle expenses 2,859 3,474 3,471 3,534 21.5 -.1 1.8
Public and other transportation 818 781 263 452 -4.5 -66.3 71.9
Healthcare 4,968 5,193 5,177 5,452 4.5 -.3 5.3
Entertainment 3,226 3,090 2,909 3,568 -4.2 -5.9 22.7
Personal care products and services 768 786 646 771 2.3 -17.8 19.3
Reading 108 92 114 114 -14.8 23.9 0
Footnotes
[1] Data are rounded to the nearest thousands.
[2] Data not applicable.
Source: U.S. Bureau of Labor Statistics.
2018 2019 2020 2021 Percent change
Item
2018–19 2019–20 2020–21
Education 1,407 1,443 1,271 1,226 2.6 -11.9 -3.5
Tobacco products and smoking supplies 347 320 315 341 -7.8 -1.6 8.3
Miscellaneous 993 899 907 986 -9.5 .9 8.7
Cash contributions 1,888 1,995 2,283 2,415 5.7 14.4 5.8
Personal insurance and pensions 7,296 7,165 7,246 7,873 -1.8 1.1 8.7
Life and other personal insurance 465 520 486 473 11.8 -6.5 -2.7
Pensions and Social Security 6,831 6,645 6,760 7,400 -2.7 1.7 9.5
Footnotes
[1] Data are rounded to the nearest thousands.
[2] Data not applicable.
In 2021, spending was higher than 2020 in 12 of the 14 expenditure subcategories. For example, after declining more than 10 percent in 2020, food spending shot up
13.4 percent in 2021, with noticeable surges in both food at home and food away from home spending. Food at home spending climbed 6.6 percent, reaching an annual
amount of roughly $600 above pre-pandemic levels. This was mainly driven by increased spending in the protein (meat, poultry, fish, and eggs) and produce sectors. As
of December 2021, both expenditures have seen annual average percent changes in inflation of 6.6 percent and 3.2 percent, respectively.5 Food away from home
spending rose 27.6 percent but was still nearly $500 below 2019 levels, likely driven by reluctance of at-risk demographics to eat out amid the multiple COVID-19 waves
that emerged in 2021. (See table A.)
Transportation spending surged 11.6 percent in 2021, climbing from lows in 2020 and exceeding 2019 levels. Rising 37.0 percent in 2021, gasoline, other fuel, and motor
oil spending was a key driver of the increase, along with vehicle purchases, which rose 6.7 percent. The loosening of travel restrictions, gradual return to the office, and
rising gasoline prices surely influenced the rise in expenditures of gasoline, other fuel, and motor oil. Supply chain disruptions for semiconductors and microchips, both
used to manufacture new vehicles, induced a negative supply shock (i.e., lessened production of goods or services than in “non-shock” times) in the new vehicle market.
Any negative supply shock leads to higher prices, which in turn can lead to higher spending. The lack of new vehicle inventory might have also pushed would-be
purchasers into the used car market, raising the cost of used cars, and driving up vehicle purchase expenditures. While public and other transportation spending did
climb 71.9 percent, its smaller proportion of the transportation category, relative to gasoline and vehicles, would lead to it having less upward bias in transportation
spending. Public and other transportation has also yet to return to 2019 spending levels. (See table A.)
Housing expenditures, which account for the largest share of total spending, rose 5.6 percent in 2021. Run-ups in prices for both owned homes and rental units (such as
apartments) undoubtedly contributed, as shelter accounts for just over half of total housing expenditures. While spending on owned dwellings rose 1.6 percent, rental
properties jumped at almost four times that rate, or 6.3 percent, in 2021. Low nominal interest rates and rising housing demand in early 2021 created conditions in which
housing prices took off. According to the St. Louis Federal Reserve, the median sale price of a home reached $423,600 in the fourth quarter of 2021, almost $65,000
above the 2020 fourth quarter median price of $358,700.6 Concurrently, the trend in the 30-year fixed rate mortgage began to inch upward, fluctuating over the year, but
ending up 0.44 percentage points higher at the end of 2021 relative to 2020. Together, increases in mortgage rates and housing prices, pushed up owned dwelling
(mortgage) expenditures.7 Higher housing prices could have pushed prospective home buyers into the rental market, driving up rental rates. Double digit run-ups in out-
of-town lodging of 89.9 percent, other lodging (which includes hotel/motel, B&B accommodations, etc.) of 36.1 percent, and household furnishings and equipment of
15.1 percent further contributed to the rise in housing spending.
Entertainment and apparel and services both rose by over 20 percent, 22.7 and 22.3 percent, respectively (chart 1). Double-digit spending growth was also observed in
personal care products and services, alcoholic beverages, food, and transportation, all following the model of looser COVID-19 restrictions, gradual returns to in-person
work, and higher demand for entertainment and travel. Six subcategories grew at rates below 10 percent. Healthcare spending, for example, rose 5.3 percent, potentially
because of elective surgeries resuming where consumers were able to book screenings. One expenditure category did not change, and one declined slightly. Spending on
reading material remained constant at $114 per year, while spending on education fell by 3.5 percent, potentially because of reduced annual amounts paid by CUs on
tuition and student loan payments, as shown in our detailed level tables, which can be found in the technical notes section (detailed tables, 1984–21).
Chart 1. Annual percent change for major expenditure groups, 2020–21
Entertainment
Apparel and services
Personal care products and services

Alcoholic beverages
Food
Transporta on
Personal insurance and pensions
Miscellaneous
Tobacco products and smoking supplies
Cash contribu ons
Housing
Healthcare
Reading
Educa on
-5 0 5 10 15 20 25
Percent
Hover over chart to view data.

View Chart Data
Expenditure shares: how do consumers divide up their spending?

The concept of expenditure shares is critical in analyzing short and long run spending dynamics. In the short run, expenditure shares illustrate the relative importance of
the various expenditure categories in everyday life, for example, the percent of total spending being directed to food relative to healthcare or housing. In the long run,
changes in shares reflect changes in living standards. For example, CUs who allocate lower shares to food and shelter can spend larger shares on entertainment, cash
contributions, personal care products and services, and other discretionary goods and services that are not as affordable to those on tight budgets.
Chart 2A. Percentage of expenditure shares by selected categories, 2018–21
2018 2019 2020 2021
Housing
Transporta on
Food
Healthcare
Entertainment
Cash contribu ons
All other expenditures
0 5 10 15 20 25 30 35 40
Percent
Click legend items to change data display. Hover over chart to view data.
View Chart Data
The largest share of consumer spending is generally allocated to housing (chart 2A). From 2018 to 2021, the share ranged from 32.8 percent in 2018 and 2019 to 34.9
percent in 2020. Following housing, transportation, food, personal insurance and pensions, and healthcare round out the top five expenditure shares (chart 2B). For the
purposes of this analysis, the six expenditure categories to which consumers allocate on average the smallest shares of their total expenditures are combined into the all
other expenditures category (chart 2A). Expenditures in this category include alcoholic beverages, personal care and products, reading, education, tobacco products and
smoking supplies, and miscellaneous expenditures.
Chart 2B. Percent distribution of expenditure shares, 2021
Housing
Transporta on
Food
Healthcare
All other expenditures
Entertainment
Cash contribu ons
View Chart Data
Recent changes in expenditure shares show a mixed picture. Of the four expenditure categories, five rose, five remained unchanged, and four declined from their 2020
levels. As displayed in table B and chart 2, CUs in 2021 allocated slightly larger shares of their total spending toward entertainment, transportation, food, and apparel and
services, all rising between three and six tenths of one percentage point. The fifth category that rose was in the all other expenditures category. Personal care products
and services increased by a tenth of 1 percentage point.
Expenditure shares marginally declined in housing (1.1 percentage points), but it remains by far the largest spending category, where 33.8 cents for every dollar of
spending outpaces the next closest category (transportation) by 17.4 cents per dollar (chart 2B). Similarly, the average consumer contributed fewer cents on the dollar to
personal insurance and pensions, cash contributions, and healthcare in 2021. Consumers spent half a percentage point less on all other expenditures in 2021 (6.4 to 5.9
percent), leaving the remaining 94.1 percent of spending, or 94.1 cents on the dollar, to the top eight expenditure categories (chart 2A).
Table B. Percent distribution of total annual expenditures

by major category for all consumer units, Consumer
Expenditure Surveys, 2018–21
Spending category 2018 2019 2020 2021
Average annual expenditures 100.0 100.0 100.0 100.0
Food 12.9 13.0 11.9 12.4
Food at home 7.3 7.4 8.1 7.9
Food away from home 5.6 5.6 3.9 4.5
Alcoholic beverages 1.0 .9 .8 .8
Housing 32.8 32.8 34.9 33.8
Shelter 19.2 19.3 20.5 19.8
Utilities, fuels, and public services 6.6 6.4 6.8 6.3
Household operations 2.5 2.5 2.4 2.4
Housekeeping supplies 1.2 1.2 1.4 1.2
Household furnishings and equipment 3.3 3.3 3.8 4.0
Apparel and services 3.0 3.0 2.3 2.6
Transportation 15.9 17.0 16.0 16.4
Vehicle purchases (net outlay) 6.5 7.0 7.4 7.2
Gasoline, other fuels, and motor oil 3.4 3.3 2.6 3.2
Other vehicle expenses 4.7 5.5 5.7 5.3
Public and other transportation 1.3 1.2 .4 .7
Healthcare 8.1 8.2 8.4 8.1
Entertainment 5.3 4.9 4.7 5.3
Personal care products and services 1.3 1.2 1.1 1.2
Reading .2 .1 .2 .2
Education 2.3 2.3 2.1 1.8
Tobacco products and smoking supplies .6 .5 .5 .5
Miscellaneous 1.6 1.4 1.5 1.5
Cash contributions 3.1 3.2 3.7 3.6
Personal insurance and pensions 11.9 11.4 11.8 11.8
Life and other personal insurance .8 .8 .8 .7
Pensions and Social Security 11.2 10.5 11.0 11.1
Income and expenditures by income quintile

Even in times of price stability, income is arguably the most important predictor of expenditures. Given COVID-19 related disruptions to the economy in 2020 (increased
unemployment, etc.), it is especially important to examine changes in income from 2020 to 2021 before considering changes in spending patterns at that time.
In 2021, gains in nominal income were observed across the board for CUs in all income quintiles except for the bottom quintile of earners, which fell by 0.4 percent. This
reduction in nominal income can be partly attributed to a decline in wages and salaries of 12.3 percent and in Social Security, private retirement, and government
retirement of 2.5 percent. All other cohorts saw gains between 0.6 and 3.8 percent, with the general trend being, the higher the income cohort, the greater the
percentage gain in before-tax income (table C).
Table C. Dollar change and percent change in average annual expenditures on major categories by income quintiles, 2020–21
Lowest quintile Second quintile Third quintile Fourth quintile Highest quintile
Item Over-the-year change Over-the-year change Over-the-year change Over-the-year change Over-the-year change
Dollar Percent Dollar Percent Dollar Percent Dollar Percent Dollar Percent
Income before taxes -$54 -.4 $217 .6 $1,792 3.0 $3,306 3.4 $8,195 3.8
Total 2,145 7.5 4,028 10.1 4,371 8.5 3,342 4.6 13,373 11.6
Food 776 18.9 409 7.6 1,067 16.9 875 10.3 1,728 14.1
At home 425 13.7 195 5.1 524 12.4 132 2.3 310 4.0
Away from home 351 35.1 213 13.5 543 26.2 744 26.6 1,419 32.0

Lowest quintile Second quintile Third quintile Fourth quintile Highest quintile
Item Over-the-year change Over-the-year change Over-the-year change Over-the-year change Over-the-year change
Dollar Percent Dollar Percent Dollar Percent Dollar Percent Dollar Percent
Alcoholic beverages 58 38.7 65 27.7 39 10.5 113 22.3 103 9.1
Housing 537 4.4 902 5.7 1,438 7.8 1,108 4.6 1,881 5.1
Apparel and services 181 23.8 343 39.8 277 21.9 311 21.3 485 17.2
Transportation -90 -2.1 1,770 28.5 1,096 11.9 374 3.0 2,408 14.3
Healthcare 303 10.9 106 2.6 -162 -3.3 272 4.5 807 10.2
Entertainment 201 16.9 169 8.9 511 25.1 410 11.7 1,946 32.8
Personal care products and services 81 27.1 106 23.3 53 9.2 168 22.8 208 17.8
Reading -9 -13.7 -8 -9.3 -22 -22.2 25 21.9 16 7.6
Education -127 -20.0 63 17.4 -148 -21.7 -90 -7.9 47 1.3
Tobacco products and smoking supplies 54 19.1 16 4.4 20 5.6 48 14.6 -4 -1.5
Miscellaneous 59 14.0 -4 -.7 1 .1 27 2.4 303 19.3
Cash contributions 195 26.0 42 3.1 -205 -11.4 -716 -25.0 1,303 28.0
Personal insurance and pensions -71 -11.7 48 2.3 0 8.8 416 4.6 2,144 10.7
Total expenditures rose across the board for all income groups, with increases ranging from 4.6 percent for the fourth income quintile to 11.6 percent for the highest
income quintile. The second and highest income quintiles' expenditure patterns outpaced change in expenditure by all consumers of 9.1 percent (table A), while the
lowest, third, and fourth lagged behind anywhere from 0.6 to 4.5 percentage points relative to the 9.1 percent for all CUs. (See chart 3.) The 11.6-percent increase
observed for the highest income group is attributed to a handful of main expenditure categories, including a 32-percent increase in food away from home, a 32.8-percent
increase in entertainment, a 14.3-percent increase in transportation, and a 10.7-percent increase in personal insurance and pensions. (See table C.) These four
categories alone account for 59.2 percent of the highest quintile’s total spending jump of over $13,000. The highest quintile follows the spending pattern seen in table A
in which food, transportation, and entertainment all rebounded noticeably in 2021 from their 2020 COVID-related declines. Unlike all other quintiles, the lowest saw a
reduction in transportation spending of 2.1 percent; the other quintiles saw jumps of 3.0 to 28.5 percent. Individuals in the lowest income quintile reduced spending on
new cars and trucks. This was partially offset by higher spending on used cars and trucks and greater use of public and other transportation.
Chart 3. Percent change in average annual expenditures by income quintile, 2020–21
Average annual expenditures Income before taxes

Percent
12.5
10
7.5
2.5
-2.5
All consumer units Lowest Second Third Fourth Highest
View Chart Data
Analysis of selected expenditures

The following section analyzes trends in individual expenditures in closer detail. The sections will cover ways in which the pandemic caused consumer spending patterns
to shift. For housing, and to a lesser extent food spending, the pandemic caused shifts in consumers’ preferences for two substitutes. For housing, the pandemic caused
consumers to reevaluate whether it was in their interest to rent or buy. In food, pandemic-related concerns about safety and government-mandated closures made many
consumers rethink their mix of food at home and food away from home. In categories in which consumers had a greater number of choices, the focus is on how the
pandemic affected consumers’ level of spending in that category.
Dining out or cooking at home: how are consumers spending food dollars?
The COVID-19 pandemic led to changes in consumer spending patterns for many goods and services, food included. There are two general types of food expenditures:
food at home and food away from home. Food at home includes food purchased from grocery stores or similar venues, for which preparation is needed. Food away from
home includes food from restaurants or similar venues, where the consumer pays for the service of food preparation. It may also include delivery to a consumer’s home
or table or the implicit costs of nonessentials—such as convenience, ambiance, and so forth. Therefore, during the pandemic period, when general health was threatened
by engagement in public activities and unemployment markedly increased, one would expect a decline in expenditures for food away from home and an increase in
expenditures for food at home.
To assess the magnitude of these changes, this section examines how the composition of spending on food changed in the COVID-19 era. In addition, this section
reveals other changes in food spending patterns, undoubtedly related to the pandemic, that are apparent in the CE data.
With the onset of the COVID-19 pandemic in 2020, total food expenditures fell, before recovering and surpassing 2019 levels in 2021. The components tell the story.
Note that expenditures for food at home climbed 6.5 percent on average each year from 2019 to 2021. Concurrently, food away from home plummeted 32.6 percent in
2020, before recovering modestly in 2021, up 27.6 percent (table A). Undoubtedly, the 2020 figures can be explained as reactions to the pandemic.
Between state and local government mandates forcing many restaurants to close for varying periods in 2020 and increased consumer caution in response to the
pandemic, consumers substituted food away from home with food at home throughout that year. Interestingly, as mandates began to recede, 2021 saw increases in both
food away from home and food at home. The recovery in food away from home in 2021 is not surprising. In order to meet the emerging market of consumers with
isolation fatigue, many restaurants—even high-end establishments—started offering delivery, curbside pickup, and other options that did not exist before the pandemic.
Yet despite the growing number of social-distanced food away from home options, food at home expenditures increased as well. The reason does not appear to be price
change. The CPI shows that prices for food at home increased 3.5 percent in both 2020 and 2021, compared with the 6.4-percent increase in expenditures in each year.
At the same time, the CPI for food away from home increased 4.5 percent in 2020 and 2021. These changes affected the allocation of the total food dollar, as shown in
chart 4.
Chart 4. Shares of total expenditures to food at home vs. food away from home, 2018–21
Food away from home Food at home
2021
2020
2019
0 2 4 6 8 10 12 14
Percent
View Chart Data
Note that food away from home expenditures accounted for more than 43.2 percent of food expenditures in 2019, plunging to 32.5 percent in 2020, and recovering to
just over 36.5 percent in 2021. (See chart 5.)
Chart 5. Allocation of food dollars spent, 2019–21
Food away from home Food at home

Percent
100
75
50
25
0
2019 2020 2021
View Chart Data
Once again, with restaurants initially closed in 2020, and consumers exercising prolonged caution, it is not surprising that the frequency of food away from home
purchases declined in this year. In 2021, many restaurants reopened, but the Omicron and Delta COVID-19 variants were likely to be responsible for the lack of full
recovery in food away from home expenditures in 2021.
Interestingly, food at home expenditures exhibited the same pattern as food away from home. The answer is likely stocking behavior. That is, consumers likely made
fewer trips to the grocery store, but purchased food in bulk on these trips. Consumers who typically shopped once a week might have switched to biweekly or even
monthly shopping trips.
Finally, these changes in food expenditure—at the total and component level—affected the allocation of total expenditures. In the pre-COVID-19 pandemic period (2019),
food away from home represented roughly 43 percent of total food spending, or 5.6 percent of total expenditures in both 2018 and 2019 (chart 4). As expected, with the
onset of the COVID-19 pandemic in 2020, consumers shifted toward purchasing food at home, most likely because of government mandated shutdowns of restaurants
and similar venues reduced seating capacity, and to comply with social distancing recommendations and pandemic-related health concerns.
As a result, in 2020, food at home accounted for 67.5 percent of total food spending. The next year, 2021, saw a gradual return to the 2019 allocations to food spending
at and away from home. Food at home declined to 7.9 percent and food away from home reached 4.5 percent of total expenditures. Shares of total expenditures for food
at home remain 0.5 percentage points higher than its 2019 level and food away from home remains 1.1 percentage points less than its 2019 level (chart 4). Multiple
variables can potentially account for this. As noted above, while food prices at home did not rise as much as food prices away from home, CPI data show that food at
home prices rose by 3.5 percent from 2020 to 2021. Furthermore, those most at risk for COVID-19 might still have been hesitant to dine outside the home and thus may
have remained on the sidelines from the restaurant scene. (See chart 4.)
Table D. Diary survey percent

reporting for food expenditures,
2019–21
Percent reporting 2019 2020 2021
Food, total 93.0 89.0 91.2
Food at home 85.3 81.1 82.7
Food away from home 74.4 60.2 64.8
According to results of the Diary Survey, percent reporting changed within these categories (table D). Note that 93 percent of consumer units reported a weekly food
expenditure of some kind in 2019. This fell to 89 percent in 2020 and rose to 91 percent in 2021. Again, the components are revealing, and follow the same pattern;
percent reporting fell in 2020 and recovered partially, in 2021. This pattern is most evident for food away from home; percent reporting fell from 74 to 60 percent, then
rose to 65 percent.8
Housing
Like food, housing is largely a dichotomous topic. People must decide whether they want to rent or own their dwelling. Their choice is, among many other factors,
influenced by prevailing interest rates for a mortgage and the overall level of housing prices. The onset of the pandemic had a strong impact on economic activity and
behavior, heavily influencing the housing market and, by extension, a CU’s decision to rent or buy.
Renting vs. homeownership

After rising for 2 of the last 3 years and recording a 2-percent rise at the beginning of the COVID-19 pandemic, homeownership rates slightly declined to 65 percent in
2021 (chart 6). This marks the first reduction in homeownership rates since 2015, when homeownership rates dropped to 62 percent, the lower bound of the 17-year
time series from 2004 to 2021. The corresponding rental rate rose to 35 percent in 2021, returning to its average for the time series. Much of this can be explained by
the movements of mortgage interest rates.
In the beginning of 2021, interest rates on 30-year mortgages remained high, yet, as the year progressed, interest rates began to fall. As mortgage rates decreased, the
gap between the number of potential homebuyers and the number of houses supplied grew. Due to this imbalance, there was a shortage of owned homes available for
first time buyers, crowding them out and pushing the youngest CUs into the rental market.
Average annual shelter prices rose 2.7 percent from 2020 to 2021, according to CPI data. As noted in the 2020 CE Annual Report, a change in data collection
methodology may have contributed to the rise in homeownership status in that year.9
Chart 6. Homeownership rate, Consumer Expenditure Survey (CE), 2004–21
Percent
70
68
66
64
62
60
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Hover over chart to view data.

View Chart Data
Homeownership and rent spending by region

Among the major spending categories, housing remains the single largest component. Housing is generally thought of in terms of geographic region. In the CE, the
United States has four regions: Northeast, South, Midwest, and West.10 Examining rent spending by region can elicit some interesting comparisons: for example, are
there more CUs renting in the Northeast than the West? Region itself is a broad metric and it might not pick up subtle differences that can be explained on a state or
even metropolitan statistical area (MSA) level.
In general, housing expenditures increased 5.6 percent from 2020 to 2021, following a 3.5-percent increase from 2019 to 2020 (table A). Even though housing
expenditures increased, the shares of housing expenditures decreased by 1.1 percent to 33.8 percent. Yet, despite this decrease, 2021 shares remain 1.0 percentage
point higher when compared with 2019 levels (table B). The overall homeownership rate decreased by 1.0 percentage point to 65.0 percent in 2021, following a 2.0-
percentage-point increase from 2019. (See chart 6.)
Chart 7A. Homeownership rate, Consumer Expenditure Surveys, 2019–21
Northeast Midwest South West

Percent
75
70
65
60
55
2019 2020 2021
View Chart Data
In each of the four regions of the United States, the homeownership rate increased from 2019 to 2020. However, each region differed slightly from 2020 to 2021. The
Northeast was the only region to show growth in homeownership rates at 1.0 percentage point. The Midwest had the highest homeownership rate, which remained
stable at 70.0 percent (chart 7A). The South and West experienced similar declines; as homeownership rates fell (and, therefore, rental rates rose) in 2021, the
homeownership rate in the South remained 2.0 percentage points above its 2019 level. In contrast, the West, which already had the lowest homeownership of the four
regions, was the only region in which 2021 homeownership rates fell below 2019 levels.
Homeownership rates in the West fell from 61.0 percent in 2020 to 58.0 percent in 2021. Trends of homeownership in the West are consistent with a past study of
homeownership rates, where it corroborated this deviation from trend in the West.11 The converse nature of homeownership and rental rates means that the West had
the highest proportion of renters to homeowners in 2021. It should be noted that this analysis focuses on the short term: 2019 to 2021.
Chart 7B. Homeownership rate of key California metropolitan statistical areas, 2016–21
Los Angeles San Francisco San Diego

Percent
62.5
60
57.5
55
52.5
50
47.5
2016-17 2018-19 2020-21
View Chart Data
To analyze the homeownership dynamics in the western United States, it is important to consider its major population centers. The most populous state in the West and
in the broader United States is California, with the three most populous cities in the state being Los Angeles, San Francisco, and San Diego. Data variations in the West
can, in theory, be explained by movements of these three cities. At the MSA level, the homeownership rates for San Diego and Los Angeles declined 6.0 and 2.0
percentage points in the 2018–19 collection period. Associated rebounds of 4.0, 7.0, and 8.0 percentage points for Los Angeles, San Francisco, and San Diego,
respectively, were observed in the following collection period of 2020–21. That runs counter to the directional trends in the broader West (chart 7B). Some further mining
found gold in Phoenix, AZ and Honolulu, HI. Over the same collection period, homeownership rates rose sharply in 2016–17 and 2018–19 before plunging 9.0 and 6.0
percentage points, respectively, mirroring the trend in chart 7A (chart 7C). MSA data are represented in multi-year tables with 2-year collection periods as opposed to
single-year collection periods in detailed level tables because of data limitations.
Chart 7C. Homeownership rates in additional western metropolitan statistical areas, 2016–21
Phoenix Honolulu
Percent
75
70
65
60
55
50
2016–17 2018–19 2020–21
View Chart Data
Apparel spending by size of consumer unit

Average annual expenditures in apparel and services increased by 22.3 percent in 2021 (to $1,754) compared with a 23.8-percent drop in 2020 (to $1,434) and yet
remained lower than in 2019 ($1,883).12 (See table A.) Within the apparel and services category, the biggest increase in a major component, 48.6 percent, occurred in
the other apparel products and services expenditure category. This increase is attributable to a 119.4-percent increase in the subcategory of jewelry expenditures, rising
from an increase in wedding bookings and other formal functions. Expenditures for general apparel items exhibited increases that were closer to the overall average
increase: 32.6 percent for boys ages 2 to 15, 29.2 percent for men ages 16 and over, and 20.4 percent for women ages 16 and over. Footwear expenditures increased by
7 percent in 2021, mainly because of a 125-percent jump in boys’ footwear. Interestingly, women’s footwear decreased by 11.6 percent during the same period. Children
under the age of 2 was the only age group to experience a decrease in apparel expenditures, at 2.9 percent. Despite this decrease in total apparel and service
expenditures, for the category of boys and girls from 2 years to 15 years of age, expenditures in 2021 were higher than in 2019. The gradual return to in-person learning
for children might explain this trend.
Although expenditures on apparel were still below pre-COVID spending, apparel spending increases in 2021 were consistent across CUs of all sizes. (See chart 8.)
Percentage increases in spending by size of the CU ranged from 8.8 percent for four-person CUs to 57.4 percent for three-person CUs. While small CUs (one or two
persons) had increases in expenditures at the smaller end of the range (about 11 percent each), for the largest CUs (five or more persons) expenditures rose 46.9
percent as lockdown and pandemic restrictions lifted. This contrasts to decreases for all CU sizes in 2020.
Chart 8. Spending on apparel and services by consumer units, 2019–21
Apparel and services 2019 Apparel and services 2020 Apparel and services 2021
Percent change 2019–20 Percent change 2020–21
Dollars Percent
4,000 70
60
50
3,000 40
30
20
2,000 10
-10
1,000 -20
-30
-40
0 -50
One person Two people Three people Four people Five or more people
View Chart Data
New and used vehicles by highest educational attainment of consumer unit

The purchase of a vehicle is one of the largest made by a CU. As stated earlier, according to the CPI, average prices for new and used cars and trucks were up 5.9 and
26.6 percent, respectively, from 2020 to 2021. (See the Effects of inflation and supply chain disruptions in 2021 section.) These inflated prices have only been
exacerbated the last 2 years by the ongoing semiconductor and microchip shortage.13 New cars have 150+ chips contained inside them, and with a limited supply of
microchips being imported fewer, new automobiles can hit the market.14 This reduced supply is likely one of the many reasons that new cars were even more expensive.
If CUs view new and used automobiles as substitutes, this reduction in supply for new cars raises demand for used cars. According to detailed level tables, new and used
vehicle spending rose 5.8 and 8.3 percent, respectively.
Despite the increase in overall spending on new and used cars, it is worth noting that the level of expenditure is not uniform across subgroups. For instance, vehicle
purchases vary by CU’s highest education attainment. Education subgroups have wide variation in income, occupation, and other demographics.
Chart 9A. New car and truck prices paid by highest education of consumer unit, 2019–21
2019 2020 2021

Dollars
50,000
40,000
30,000
20,000
10,000
0
Less than high High school High school Associate's degree Bachelor's degree Master's,
school graduate graduate graduate with some professional,
college doctoral degree
View Chart Data
With reduced supply in the new vehicle market, expectations would dictate higher prices and thus mean higher quarterly expenditures for vehicles among all education
groups. Chart 9A shows mean quarterly expenditures for CUs who purchased at least one new car or truck by highest education level within the respective consumer
unit.15 The trend of yearly expenditure increases holds true for all the subgroups except for the less than high school graduate group. Several factors are likely to cause
swings in mean quarterly expenditures on new cars for those in this subgroup.
The total number of CUs in the less than high school graduate subgroup is far below that of all other education subgroups. Only 7.9 of the total 133.6 million CUs occupy
it, roughly half as many as the next smallest share, associate's degree graduates. Moreover, percent reporting in 2021 is the lowest and most volatile of this 2019–21
period, so any swing in percent reporting will have a more profound effect on price paid. (Detailed level interview tables that include percent reporting are available upon
request; see the technical notes for more information.)
For those with at least a high school degree, the mean quarterly expenditure for a new car or truck rose gradually from 2019 to 2021. Prices paid in 2021 hovered
between $37,075 and $42,607, respectively, reflecting the negative supply shock. Mean quarterly expenditures jumped from 2020 to 2021. Bachelor’s degree holders
experienced the strongest increase and levels were elevated for master’s, professional, and doctoral degrees, too. Average annual incomes of over $100,000, which are
generally associated with higher levels of education obtainment, allow for the purchase of more luxury vehicles, pushing up the mean quarterly expenditure. (Income
levels can be found in the Calendar year Tables by Education.)
Table E. Interview survey percent reporting for new cars and trucks by education, 2019–21
Less than high school High school High school graduate with some Associate's Bachelor's Master's, professional, doctoral
Year
graduate graduate college degree degree degree
2019 .63 .96 1.49 1.95 1.60 2.08
2020 .26 1.05 1.16 1.62 1.88 1.82
2021 .42 .68 .93 1.70 1.47 2.23
Table F. Interview survey percent reporting for used cars and trucks by education, 2019–2021
Less than high school High school High school graduate with some Associate's Bachelor's Master's, professional, doctoral
Year
graduate graduate college degree degree degree
2019 3.60 4.40 4.37 5.94 4.39 4.12
2020 4.39 3.74 4.63 5.07 3.55 3.26
2021 4.60 4.56 4.22 4.89 3.59 3.18
With new cars and trucks in shorter supply, for those who had to or wanted to purchase a vehicle, this incentivized substitution to the used car market. Among the six
education subgroups, three saw upticks in percent reporting for used vehicles: high school, less than high school graduates, and bachelor’s degree holders (table F). This
uptick in reporting the purchase of an automobile could be tied to the fact that positions with an education obtainment of at most a bachelor’s degree were more likely to
be deemed “essential workers.”16 Alternatively, people with at most a bachelor’s degree could have put additional wear on their cars by working part-time as drivers in
the rapidly growing food delivery app industry, made popular by the pandemic. Master’s, professional, and doctoral degree holders had the highest mean quarterly
expenditure for used cars and trucks of $21,033, roughly $3,055 more than the next closest group of associate's holders at $17,978 (chart 9B). CUs in this subgroup
have higher income levels, allowing them to absorb higher prices more easily or more likely to purchase multiple cars for the purpose of leisure, thus driving up
expenditures.
Chart 9B. Used car and truck prices paid by highest education of consumer unit, 2019–21
2019 2020 2021

Dollars
25,000
20,000
15,000
10,000
5,000
0
Less than high High school High school Associate's degree Bachelor's degree Master's,
school graduate graduate graduate with some professional,
college doctoral degree
View Chart Data
Gasoline spending
As Americans adjusted to the ongoing pandemic, 2021 should have been the year in which consumers hit the open road en masse, filling up their cars, trucks, and SUVs
with great frequency. This could be to drive to the office (given the gradual return of in-person office work) or to take the family on that 1-year delayed road trip. What
most consumers likely did not anticipate was the surge in gasoline prices from the half-decade low of $2.24 per gallon seen in 2020 to a half-decade high of $3.13, as
the CPI data show. (See chart 10A.) With higher driving demand and higher gasoline prices, gasoline spending and prices rose 29.1 and 36.0 percent in 2021,
respectively, as measured by CE and the CPI. (See chart 10B.)
Chart 10A. Gasoline price and quantity purchase, 2008–21
Quan ty purchased per week Gasoline price per gallon

Gallons Dollars
20 4.0
3.5
15
3.0
10 2.5
2.0
1.5
0 1.0
2008 2010 2012 2014 2016 2018 2020
View Chart Data
Just like any other good or service, two factors determine total expenditures for gasoline: the price of the good and the quantity purchased. All else equal, any change in
gasoline prices will directly affect the resulting annual gasoline expenditure for a given year. To illustrate this idea, imagine a basic equation where total gasoline
expenditures of G is a function of the annual price paid by consumers of P, multiplied by the annual quantity of gallons purchased of Q. That is, G=P × Q. Strictly based
off chart 10B, it is not certain whether this rise in expenditures is a result of changes in price or quantity purchased.
Price elasticity of demand is a measure of how much consumers change the amount of a good they buy when the price of that good changes. Price elasticity of demand
can be influenced by factors such as the percentage of total expenditures for which a good (gasoline in this case) accounts, how many close substitutes exist for the
good, and whether the prices of the substitutes for the good of interest are higher or lower than the new price of the good of interest. For instance, if the price of
gasoline were to get too high, if people see that it is cheaper for them to take alternative modes of transportation (like public transportation or use ride-sharing apps),
they may substitute gasoline for one of those modes of transportation instead. Price elasticity of demand in this case is how much less gasoline people buy when it
becomes more expensive.
Regardless of reason, both prices and quantities can drive spending behavior. How much expenditures are affected by each depends on the magnitude of each respective
change. Price can rise, quantity consumed can fall, and expenditures can still rise if the percent price increase is larger than the percent quantity decrease. This also
works in the reverse fashion; prices can fall, and expenditures rise, when the percent increase in quantity purchased is larger than the percent decrease in prices.
Regardless, this was not the case in 2021, where it appears that prices increased, and quantities purchased declined slightly (indicating noticeable price inelasticity of
gasoline with respect to price).
Chart 10B. Percent change in gasoline spending and CPI-gasoline, 2008–21
Gasoline (CE) CPI-U Gasoline

Percent
40
20
-20
-40
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
View Chart Data
To analyze this concept graphically, chart 10A shows annual gasoline prices and weekly gallons of gasoline purchased. Weekly quantity of gallons of gasoline are
computed by first taking CE estimates for total gasoline expenditures, dividing that value by average price data for gasoline from the CPI, and then dividing by 52 to get
a weekly estimate. Presenting quantity in a weekly figure reflects the frequency of which consumers fill their tanks and how consumers typically frame gasoline spending:
on a weekly and not annual basis. In addition, note that BLS generates price indexes and average price estimates strictly for urban consumers. Therefore, charts 10A and
10B include expenditure data only from urban consumers. Furthermore, CPI average price estimates strictly capture gasoline and not motor oil or other fuel prices.
Estimates in published CE tables capture motor oil and other fuels. Therefore, CE gasoline data are extracted from the detailed level type of area tables (unpublished, but
available on request) described in the technical notes section of the report. (See section titled, “Detailed tables, 1984–21.”)
Analyzing charts 10A and 10B together provides the clearest picture for gasoline expenditure trends from 2008, the first year of the most recent recession before
COVID-19, to 2021. The components of chart 10A influence chart 10B. From 2020 to 2021, prices and expenditures rose at a close 36.0 and 29.1 percent, respectively.
(See chart 10B.) Weekly quantity purchased fell slightly by roughly nine tenths of a gallon per week (0.93 gallons). As stated, prices displayed are for urban areas only
and some of said urban areas experienced higher prices relative to others.
Thus, some consumers, potentially on the lower end of the income distribution, would not have been able to maintain quantities purchased once prices rose. That is,
such consumers might reduce miles driven each week, and therefore gasoline purchases, accordingly. Couple this with many companies keeping telework policies in place
through much of 2021, and the conditions were present for fewer weekly gallons purchased.
While expenditures on gasoline and the price of gasoline usually follow similar patterns, nearly identical concurrent movement in levels had not been seen since 2012 and
2009. In the 2008-–09 price change, when the United States was in a recessionary period (which the National Bureau of Economic Research (NBER) denotes as having
occurred from December of 2007 through June 2009), expenditures and prices fell 26.8 and 26.9 percent, respectively.17 In 2012, gasoline saw concurrent falls in prices
and expenditures of 3.3 and 3.9 percent, respectively.
Healthcare spending by age

Based on the human lifecycle, it would be reasonable to expect that healthcare expenditures are both related to, and increase with, a person’s age. That is, younger
people generally need less care, and may allocate most of their health care dollars to preventative care and occasional acute needs (such as the sudden onset of a
temporary illness). Older consumers are more likely to need more costly and extensive care, medical equipment, prescription drugs, and other healthcare goods and
services to maintain quality of life. Analyzing healthcare spending among the age groups can show whether this theoretical relationship exists in the data from 2020 to
2021.
All age groups spent more on healthcare in 2021 than in 2020. This change was driven by double digit increased spending in the medical services subcategory for all age
groups except two: those 45–54 years old and those 55–64 years old. CUs with reference persons ages 45–54 had the highest increase in expenditures in the drugs
subcategory whereas CUs with reference persons ages 25–34 had the greatest increase in medical supplies spending, but experienced declines in drug spending. Overall,
the largest changes in healthcare spending were in the medical services subcategory for those 35–44 years old (62.2 percent) and under 25 (58.3 percent). Increased
spending in medical services could be driven by the opening of elective surgeries that were in large part suspended in 2020. Elective and pandemic-related medical
services together could have worked in tandem to boost medical services spending.
Chart 11. Percentage change in average expenditure on healthcare and its components, 2020
to 2021, by age of reference person
All consumer units Under 25 years 25–34 years 35–44 years
45–54 years 55–64 years 65–74 years 75 years and older
Healthcare
Health
insurance
Medical
services
Drugs
Medical
supplies
-30 -20 -10 0 10 20 30 40 50 60 70
View Chart Data
Health insurance spending increased for each group 45 years and over but decreased for each age group under 45 years old. The largest decrease was 12.1 percent by
the under 25 age group. Spending on medical services increased for every age group in 2021, though the smallest increase was 3.2 percent by those 45–54 years old.
Spending on drugs fell in two groups: 25–34 years old (6.5 percent) and 65–74 years old (1.3 percent). The largest increase in spending on drugs was by those 45–54
years old, rising 15.3 percent. Spending on medical supplies declined for the under 25 age group and those 75 and older. The largest increase in medical supplies
spending in 2021 was by those 25–34 years old, who spent 39.4 percent more than the previous year (chart 11). The biggest component of medical supplies, eyeglasses,
increased by 17.6 percent. (See detailed level tables, available by request.)
Entertainment spending by age and income

In 2021, the rate of increase in entertainment expenditures for the average CU (22.7 percent) outpaced the rate of increase in total expenditures (9.1 percent). In
contrast, in 2020, the first year of the pandemic, entertainment expenditures (-5.9 percent) fell more steeply than total expenditures (-2.7 percent). (See table A.) The
2021 growth in entertainment expenditures was driven by increases in the subcategories of other supplies, equipment, and services; and fees and admissions (a category
that includes recreational vehicles and campers, motorized boats, and sports equipment). (See detailed level tables for a breakdown of each category.)
Chart 12A. Entertainment expenditures, average CU, U.S. dollars, 2013–21
Fees and admissions Pets Toys, hobbies, and playground equipment

Dollars Audio and visual equipment and services Other supplies, equipment, and services
1,250
1,000
750
500
250
0
2013 2014 2015 2016 2017 2018 2019 2020 2021
View Chart Data
In 2021, expenditures in the fees and admissions category as well as in other supplies, equipment, and services grew by 53.9 and 59.8 percentage points, respectively.
The entertainment categories of pets; audio and visual equipment and services; and toys, hobbies, and playground equipment increased steadily across the timeseries.
This increase persisted in the 2020 pandemic and 2021 post-pandemic recovery period (chart 12A).18
Entertainment expenditures differed greatly by age group in 2021. CUs with reference persons ages 45–54 had the highest levels of, and largest increase in, expenditures
in 2021. Levels of expenditures ranged from $1,700 for CUs with reference persons under 25 to $4,695 for CUs with reference persons ages 45-54. In contrast, levels of
expenditures in 2020 ranged from $1,266 for CUs with reference persons under age 25 to $3,628 for CUs with reference persons ages 35–44. However, CUs with
respondents ages 45-54 increased the most, by 47.8 percent. In 2020, this was also the age group whose expenditures fell the most (17.7 percent) (chart 12B).
Chart 12B. Entertainment expenditures by age group, U.S. dollars, 2013–21
All CUs Under 25 years 25–34 35–44 45–54 55–64

65–74 75 or over
Dollars
5,000
4,000
3,000
2,000
1,000
2019 2020 2021
View Chart Data
As noted above, other supplies, equipment, and services grew at a higher rate relative to all other entertainment components, at 59.8 percent annually from 2020 to
2021. This warranted further data exploration. It raised the question, could there be an underlying shift by income level in the other supplies, equipment, and services
subcategory?
Those CUs making $200,000 and above spent large sums of income on the purchase of motorized recreational vehicles, pushing up total spending in this subcategory
(table H). In 2020, respondents in this income group spent an average of $1,527 on motorized vehicles, more than all the other groups’ average expenditures on this
category combined. The income group with the next largest expenditure on motorized recreational vehicles, respondents earning $150,000 to $199,999, spent less than
a third in the same category, an average of $387. (See table G.)
Table G. Other supplies, equipment, and services category by income before taxes, 2020 [1]
All
Item consumer Less than $15,000 to $30,000 to $40,000 to $50,000 to $70,000 to $100,000 to $150,000 to $200,000
units $15,000 $29,999 $39,999 $49,999 $69,999 $99,999 $149,999 $199,999 and more
Entertainment $3,567.89 $1,432.49 $1,582.16 $1,863.13 $2,453.74 $2,456.39 $3,445.78 $4,454.15 $6,020.11 $10,811.51
All consumer Less than $15,000 to $30,000 to $40,000 to $50,000 to $70,000 to $100,000 to $150,000 to $200,000 and
Item
units $15,000 $29,999 $39,999 $49,999 $69,999 $99,999 $149,999 $199,999 more
Other entertainment
supplies, equipment, and 924.51 246.46 311.58 237.67 527.01 397.29 714.63 1,378.83 1,483.60 3,841.59
services
Un-motored recreational
167.32 102.40 10.21 16.78 64.85 93.56 119.78 452.49 422.77 281.58
vehicles
Motorized recreational
347.00 11.85 185.18 62.97 102.34 39.97 250.66 311.76 180.00 2,385.39
vehicles
Purchase of motorized
144.52 [2] 175.48 [2] 8.50 [2] 205.70 22.88 139.46 858.02
campers
Purchase of other vehicles 33.96 11.85 9.70 19.75 7.56 12.43 19.06 49.62 22.44 184.12
Purchase of boats with

168.52 [2] [2] 43.22 86.28 27.54 25.90 239.26 18.10 1,343.26
motor
Rental of recreational
31.92 6.88 7.04 5.47 30.70 30.40 15.00 42.75 99.20 92.50
vehicles
Docking and landing fees 14.32 .68 4.08 1.19 5.16 4.35 11.20 9.16 58.97 62.57
Footnotes
[1] This table does not contain a full list of the "other supplies, equipment, and services" subcategories. For a full list, see the detailed tables available at https://www.bls.gov/cex/tables/calendar-
year/mean/cu-all-detail-2021.pdf.
[2] Data not available.
In 2021, when asked how much they spent on motorized recreational vehicles in the last year, CU's making over $200,000 averaged $2,385 in expenditures, continuing
to outspend all other income groups.
Table H. Other supplies, equipment, and services category by income before taxes, 2021 [1]
All
Item consumer Less than $15,000 to $30,000 to $40,000 to $50,000 to $70,000 to $100,000 to $150,000 to $200,000
units $15,000 $29,999 $39,999 $49,999 $69,999 $99,999 $149,999 $199,999 and more
Entertainment $2,912.15 $1,074.97 $1,395.23 $2,100.65 $1,810.52 $2,004.36 $2,694.25 $4,331.07 $5,236.02 $7,653.62
Other entertainment
supplies, equipment, and 578.90 111.38 97.79 156.39 185.69 287.87 581.96 915.32 1,191.20 2,409.60
services
Un-motored recreational
94.27 15.96 .62 14.32 11.00 42.63 106.41 293.19 189.08 188.93
vehicles
Motorized recreational
212.23 8.88 6.62 23.68 15.45 76.44 173.45 193.59 387.18 1,526.90
vehicles
Purchase of motorized
18.76 [2] [2] [2] [2] 20.60 33.28 14.04 80.76 51.52
campers
Purchase of other vehicles 54.10 5.80 6.62 10.18 [2] 36.22 91.43 118.06 109.78 119.30
Purchase of boats with

139.37 3.08 [2] 13.50 15.45 19.61 48.74 61.49 196.64 1,356.08
motor
Rental of recreational
14.14 4.23 3.54 3.12 2.88 4.75 16.12 28.16 32.26 45.11
vehicles
Docking and landing fees 12.54 9.65 .10 17.05 6.60 3.01 1.85 18.30 17.31 62.92
Footnotes
The 2021 increase in expenditures for the highest income group on motorized recreational vehicles amounts to a 56.2 percent increase in spending from the previous
year. (See table I.) While this increase may not seem significant given the large percent changes seen in the other income groups’ expenditures in the motorized
recreational vehicles category, it is a rather large increase. The 56.2-percent increase amounted to an increase of $858.50, greater than the increases in this category for
all the other income groups combined. Thus, while the expenditures in this category of some of the other groups may have more than tripled, the other groups’
expenditures in this category were still far less.
Table I. Percent Change in other supplies, equipment, and services by income before taxes, 2020–21 [1]
Less than $15,000 to $30,000 to $40,000 to $50,000 to $70,000 to $100,000 to $150,000 to $200,000 and
Item
$15,000 $29,999 $39,999 $49,999 $69,999 $99,999 $149,999 $199,999 more
Footnotes
Less than $15,000 to $30,000 to $40,000 to $50,000 to $70,000 to $100,000 to $150,000 to $200,000 and
Item
$15,000 $29,999 $39,999 $49,999 $69,999 $99,999 $149,999 $199,999 more
Other entertainment supplies,

135.08 213.79 81.28 341.32 109.42 132.67 463.51 292.40 1,431.99
equipment, and services
Un-motored recreational vehicles 86.44 9.59 2.46 53.85 50.93 13.37 159.30 233.69 92.65
Motorized recreational vehicles 2.97 178.56 39.29 86.89 -36.47 77.21 118.17 -207.18 858.49
Purchase of motorized campers [2] [2] [2] [2] [2] 172.42 8.84 58.70 806.50
Purchase of other vehicles 6.05 3.08 9.57 [2] -23.79 -72.37 -68.44 -87.34 64.82
Purchase of boats with motor [2] [2] 29.72 70.83 7.93 -22.84 177.77 -178.54 -12.82
Rental of recreational vehicles 2.65 3.50 2.35 27.82 25.65 -1.12 14.59 66.94 47.39
Docking and landing fees -8.97 3.98 -15.86 -1.44 1.34 9.35 -9.14 41.66 -.35
Footnotes
Travel expenditures
With the onset of the COVID-19 pandemic, expenditures allocated to out-of-town trips declined by 56 percent in 2020, falling from $2,100 to $926 (table J). Despite a
rise of 95 percent in 2021 to $1,803, they were still 14 percent lower in 2021 than in 2019. At the height of the COVID-19 pandemic, in 2020, four of the five major
components of spending on out-of-town trips (food, alcohol, lodging, and entertainment) declined between 49 and 54 percent. However, the transportation component
was the hardest hit, falling 65 percent in 2020. Similarly, transportation on out-of-town trips recovered least in 2021, with expenditures ($597) 30 percent lower in that
year than in 2019. (See detailed level tables for spending on components of travel expenditures.) Expenditures for the remaining subcategories remained 2 to 7 percent
lower in 2021 than in 2019. The exception was entertainment on trips, for which expenditures were 2.4 percent higher in 2021 than in 2019.
Table J: Allocation of out-of-town trip

expenditures, 2019–21
Item 2019 2020 2021
Travel, total, component shares $2,100 $926 $1,803
Food on trips 20.1% 22.4% 21.8%
Alcohol on trips 3.3% 3.8% 3.7%
Lodging on trips 29.5% 34.4% 33.5%
Transportation on trips 40.4% 32.4% 33.1%
Entertainment on trips 6.6% 7.0% 7.9%
Regardless, the allocation of dollars spent within the travel category varied little, despite the pandemic in 2020 and 2021. The two exceptions were two subcategories:
lodging and transportation (table J). In 2019, lodging accounted for about 3 in 10 travel dollars spent, while transportation accounted for about 4 in 10 travel dollars
spent. In 2020, the share for lodging rose, while the share for transportation fell, with both categories accounting for about 1 in 3 travel dollars. In 2021, this allocation
was nearly unchanged.
Income quintile
Table K and charts 13A and 13B illustrate how expenditures for out-of-town trips relate to annual income before, during, and after the onset of the COVID-19 pandemic.
Regardless of period, one would expect expenditures for out-of-town trips to be strongly related to income, with the highest income groups spending the most.
Nevertheless, it is interesting to see how the pandemic affected these expenditures for each income quintile. For example, just because the highest income group had
the most ability to afford travel does not mean they chose to travel during a pandemic, when doing so required more thought than just considering the potential strain on
a budget.
Table K. Expenditures by
income quintile, 2019–21
Travel, total 2019 2020 2021
Lowest 20 percent $624 $268 $466
Second 20 percent 911 368 746
Third 20 percent 1,361 589 1,339
Fourth 20 percent 2,291 1,045 1,903
Fifth 20 percent 5,297 2,370 4,540

Chart 13A. Total travel expenditures by quintile, 2019–21
2019 2020 2021

Dollars
6,000
5,000
4,000
3,000
2,000
1,000
0
Lowest 20 percent Second 20 percent Third 20 percent Fourth 20 percent Fi h 20 percent
View Chart Data
Intriguingly, at least in percentage terms, spending declined at about the same rate in 2020 regardless of quintile; between 54.4 and 59.6 percent. Within this narrow
range, the highest two quintiles experienced the smallest declines: 54.4 percent by the fourth and 55.3 percent by the fifth quintile. Though not by much, the largest
decline of 59.6 percent was for the second quintile, with the first and third nearly tied at about 57 percent each (chart 13B).
Chart 13B. Total travel expenditures, percent change by quintile, 2019–21
2019-21 2019-20 2020-21

Percent
150
100
50
-50
-100
Lowest 20 percent Second 20 percent Third 20 percent Fourth 20 percent Fi h 20 percent
View Chart Data
The rebound pattern of 2021 was quite different, with the largest increase (128 percent) for the third quintile, and with the percentages tapering by quintile in the lower
direction but increasing in the higher direction. That is, the increase for the second quintile was 103 percent, while the first quintile exhibited an increase of 74 percent,
the smallest of any quintile. In contrast, the increase for the fourth quintile (82 percent) was smaller than both that for the third, and that for the fifth (92 percent).
Notes
1A May 2022 CPI/PPI Monthly Labor Review article affirms the complications from supply chain disruptions. See “PPI and CPI seasonal adjustment during the COVID-19
pandemic,” Monthly Labor Review, May 2022, https://www.bls.gov/opub/mlr/2022/article/ppi-and-cpi-seasonal-adjustment-during-the-covid-19-
pandemic.htm#:~:text=In%202020%2C%20many%20PPIs%20and,percent%20in%20April%202020%2C%20respectively.
2Unlike real dollar expenditures, nominal dollar expenditures are not adjusted for price change over time but reflect prices at the time of purchase. The terms “nominal”
and “real” are identical to the terms “current” and “constant,” which are also used to describe expenditures, incomes, or other items denominated in dollar terms.
3 A consumer unit consists of either: (1) all members of a particular household who are related by blood, marriage, adoption, or other legal arrangements; (2) a person
living alone or sharing a household with others or living as a roomer in a private home or lodging house or in permanent living quarters in a hotel or motel, but who is
financially independent; or (3) two or more persons living together who use their income to make joint expenditure decisions.
The reference person is the first member mentioned by the respondent when asked to "Start with the name of the person or one of the persons who owns or rents the
home." It is with respect to this person that the relationship of the other consumer unit members is determined.
4 The Center for Disease Control provides an aggregate and up to date timeline on the developments of COVID-19 relating to response, waves of infection, etc.
https://www.cdc.gov/museum/timeline/covid19.html.
5Annual percent changes in inflation from December 2020 to December 2021 are derived from the historic CPI report released on January 12, 2022, for the month of
December 2021. https://www.bls.gov/news.release/archives/cpi_01122022.htm
6 For data on the median sale price of a home in the United States by quarter, please see https://fred.stlouisfed.org/series/MEDDAYONMARUS#0.
7 For time series data on the 30-year fixed rate mortgage, please see https://fred.stlouisfed.org/series/MORTGAGE30US.
8 The percent reporting data are taken from the Diary Survey. While most items included in either food at home or food away from home are taken from the Diary
Survey, there is one item included in food at home that is derived from the Interview Survey: Food prepared by consumer units on out-of-town trips. However, pandemic
or not, these expenditures are both small in value (ranging from $50 to $69 annually in this period) and percent reporting (ranging from 8 to 12 percent quarterly in this
period). The percent reporting data shown above represent what happened “in one’s own neighborhood,” rather than out of town.
9 On March 19, 2020, CE in-person data collection ceased for both the Interview and Diary Surveys, and all in-person interviews were transitioned to telephone
interviews. While data could be collected in person starting in July of 2020, the initial contact attempt was to be over the phone. This restriction, as well as limited
availability of telephone numbers, may have contributed to measurement issues with renters in the CE.
10 For more information on the regions, please see https://www.bls.gov/cex/csxgeography.htm#regions.
11 See Geoffrey Paulin, "Housing and expenditures: before, during, and after the bubble," Beyond the Numbers, vol. 7, no. 10 (June 2018), www.bls.gov/opub
/btn/volume-7/housing-and-expenditures-before-during-and-after-the-bubble.htm.
12 Services in this context capture expenditures such as apparel alterations (sewing, quilting, etc.), laundry services, clothing rental, etc.
13For context on the prolonged semiconductor and chip shortage, please see Benjamin Preston, “Global chip shortage makes it tough to buy certain cars,” Consumer
Reports, May 6, 2021, https://www.consumerreports.org/buying-a-car/global-chip-shortage-makes-it-tough-to-buy-certain-cars-a8160576456/.
14 For context on the quantity of microchips in even the most basic of cars, please see Sean Tucker, “Customers paying full price, dealers making fortunes, investors
nervous anyway – chip shortage rocks car market,” Kelly Blue Book, April 26, 2021, https://www.kbb.com/car-news/customers-paying-full-price-dealers-making-fortunes-
investors-nervous-anyway-chip-shortage-rocks-car-market/#:~:text=The%20average%20new%20vehicle%20uses,electronics%2C%20but%20not%20new%20cars.
15 Charts 9A and 9B are constructed from data sourced from the Interview detailed level tables (available upon request). Data in the table is reported as average annual
expenditures, where the reported number is often far below what the typical CU would pay for a car. This stems from the fact that cars are infrequently purchased (i.e.,
low percent reporting) and those CUs who did not purchase a car have their value entered in as a zero, driving down the average. To account for this, average annual
expenditures are converted to average quarterly expenditures to reflect the quarterly waves of the interview survey. This number is then divided by percent reporting
over 100. This reflects the mean quarterly expenditure for those CUs which purchased at least one car or truck. Percent reporting is the percentage of CUs who reported
purchasing a product, in this case, a new or used car and truck. In addition, it is possible that a CU purchased more than one vehicle.
16 An article from the BLS Career Outlook program details “essential worker” positions and the minimum required education to hold these positions. See Elka Torpey,
"Essential work: Employment and outlook in occupations that protect and provide," Career Outlook, https://www.bls.gov/careeroutlook/2020/article/essential-work.htm.
17For information on how recessions are timed, i.e., from peak to through in the business cycle, please visit the National Bureau of Economic Research (NBER) data page
on U.S. Business Cycle Expansions and Contractions at https://www.nber.org/research/data/us-business-cycle-expansions-and-contractions.
18As published in CE, entertainment expenditures contain four main components, with the fourth—pets and toys, hobbies, and playground equipment—split into two
separate categories (pets and toys, etc.) to allow special attention for these subcomponents.
Statistical Tables
[+] Table 1. Income before taxes: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 2. Quintiles of income before taxes: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 3. Deciles of income before taxes: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 4. Region of residence: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 5. Population size of area of residence: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 6. Type of area: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 7. Composition of consumer unit: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 8. Highest education level of any member: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 9. Housing tenure: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 10. Number of earners in consumer unit: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 11. Size of consumer unit: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 12. Age of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 13. Generation of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 14. Selected age of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 15. Hispanic or Latino origin of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 16. Occupation of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
[+] Table 17. Race of reference person: Average annual expenditures and characteristics, Consumer Expenditure Surveys, 2021
Technical Notes
The primary goal of the Consumer Expenditure Surveys (CE) is to collect, process, and publish data on the purchasing habits of U.S. consumers. CE data are used by a
wide variety of stakeholders, such as government agencies, the private sector, and academia to generate cutting edge research. In addition, the data are required for the
regular revision of the Consumer Price Index (CPI) market basket. To craft a complete picture of how illuminating the CE is, it is important to consider its extensive
history; individual components; how seamlessly it interacts with other stakeholders within the Bureau of Labor Statistics (BLS) and outside of the agency; and practical,
published applications of the data.
CE History: 134 years of insight into the U.S. consumer’s spending habits
The CE has a rich and dynamic history spanning 134 years of survey collection and administration. The CE was first conducted in 1888. At the time, the survey was
administered on a roughly 10-year basis, leading to noticeable gaps in spending data. The infrequency of survey collection meant that CE missed the opportunity to
collect data on several time periods: the initial phase of the Great Depression from 1929–33, spending in WWII (for which an article was published in 2015) and the
immediate postwar period, and other key historic junctures of the first four-fifths of the twentieth century. This policy would hold constant through the 1972–73 survey.
The 1972–73 collection cycle marked the division of the then-singular survey into two, the Diary and Interview surveys, each with independent samples.
The Diary Survey is completed in two distinct 1-week time periods while data for the Interview Survey (with the removal of the initial “bounding” interview starting in
February 2015) is collected in four distinct waves over a 10-month period.1 The Interview Survey is spaced out in 3-month intervals to capture quarterly spending habits.
A given consumer unit (CU) under this scheme would be visited in January, April, July, and October. However, the 1970s were marked by high inflation, energy shortages,
and recessions, among other economic hardships for consumers. This led to the recognition that more frequently collected data allowed for better understanding of how
economic conditions—“boom” or “bust”—affected consumers. This demand for more frequent and timely spending data drove BLS toward annual collections for the CE.
As a result, data collection on a continuing basis began in late 1979, with the U.S. Census Bureau conducting the surveys for BLS.
As the dot-com boom of the 1990s and early 2000s set in, the CE continued to collect the Diary and Interview surveys purely on paper: without the use of digital aides.
The year 2003 marked the first shift in that direction with the adaptation of an electronic form of data collection. CE introduced a computer assisted personal interviewing
(CAPI) instrument for the two surveys. Among other short and long-term goals, it sought to minimize respondent burden and nonresponse bias and move the survey
administration process into the modern era. In 2021, spurred by the onset of the COVID-19 pandemic and to provide more timely data, mid-year CE Public Use Microdata
(PUMD) were released. Mid-year data are released in the second quarter of a given year and provide a partial snapshot into spending trends in anticipation of the annual
data release. The year 2022 saw another milestone regarding digital implementation with the full adoption of a revamped and modern online interface for the Diary
Survey.
The Diary and Interview surveys

As noted above, the CE is composed of two distinct surveys: the 2-week Diary Survey and the quarterly Interview Survey. The Diary Survey is designed to capture
expenditures on small, frequently purchased items that are normally more difficult for respondents to recall purchasing multiple months ago. It is considerably easier to
recall the purchase of fruit from a grocery store over a 2-week reference period than a 12-week reference period. Respondents keep detailed entries of consumption for
several key expenditure categories. A major category that the Diary Survey contains is that of food and beverage spending, both for food at home and food away from
home. The former comprises spending at grocery stores, public markets, and corner convenient stores. The latter focuses on food purchased at restaurants (full service,
fast food, etc.), or employer and school cafeterias. Other items collected in the Diary Survey include tobacco; housekeeping supplies; nonprescription drugs; daily
transportation needs, such as gasoline, taxi fares, and tolls; personal care products and services; and apparel. Although the Diary Survey was designed to collect
information on expenditures that are not recalled easily over an extended reference period, respondents are asked to report all expenses that the CU incurs during the
survey week, except those incurred while traveling on an out-of-town trip. For example, a New York-based CU would record spending on a taxi or limousine within
Manhattan but not for one from the airport to the hotel at the start of an out-of-town trip.
The Interview Survey is designed to capture data on expenditures centered around large-scale purchases and recurring payments. Unlike the Diary Survey, respondents
for the Interview Survey are asked to report purchases that can be recalled for at least 3 months, recorded by the interviewer. As stated, expenditures that consumers
can be expected to recall in a typical interview wave can be sorted into two main categories, the first being large-scale purchases. Examples of large-scale purchases that
a CU would report in the Interview Survey include the purchase of a new or used automobile, major appliances, and the maintenance of housing property. Recurring
payments reported in the Interview Survey include items such as rent, utility payments, and insurance premiums. In addition, unlike the Diary Survey, the Interview
Survey collects expenditures on out-of-town trips (excluding spending for business purposes). These data alone account for 60–70 percent of total expenditures for the
average CU. After adding in respondents’ global estimates for spending on food, alcohol, and tobacco products, estimates rise to about 95 percent of expenditures that
are covered in the Interview Survey.2 Nonprescription drugs, household supplies, and personal care items are excluded from Interview Survey collection.
Survey design, administration, and scope

Beyond the expenditure categories included in each survey, it is important to lay out the overall design, how each survey is administered, and what other implications
they have. For the Diary Survey, the U.S. Census Bureau draws a sample of 17,800 addresses per year for the CE, with approximately 15,000 of those addresses found to
be occupied housing units. The other 2,800 addresses are ineligible for the survey as generally, they are nonexistent, nonresidential, or vacant. Approximately 6,700 of
those occupied housing units respond to the survey, yielding a response rate of roughly 45 percent of eligible units. Each household is then asked to complete two
weekly diaries, recording all expenditures in that 2-week window. This yields a total of 13,400 weekly diaries per year. In a similar fashion, for the Interview Survey, the
U.S. Census Bureau visits 13,175 addresses on a quarterly basis, with 11,000 of those addresses having eligible, occupied housing units. Out of the 11,000 occupied
units visited, approximately 5,000 of those housing units respond to the survey for a corresponding quantity of interviews per quarter, yielding a response rate of 45
percent. Each CU is interviewed once per quarter, for four consecutive quarters. Given that the panel of CUs interviewed rotates each quarter, a consistent quantity of
CUs rotate in and out of the survey. Data from both the Diary and Interview Surveys are collected on an ongoing basis in 91 Primary Sampling Units (PSUs), which are
defined geographic areas of the United States. Integrated data from the Diary and Interview Surveys provide a more complete accounting of consumer expenditures and
income than either survey is designed to do on its own. One reason is that data on some expenditure items are collected in only one of the surveys. For example, the
Diary Survey does not collect information on reimbursements or expenses incurred by CU members while on an out-of-town trip. Examples of expenditures for which
reimbursements are included in the Interview Survey are medical care; automobile repairs; and construction, repairs, alterations, and maintenance of property. Similarly,
the Interview Survey collects only global, not detailed, expenditures on food, at or away from home.
Source selection and data changes

For expenditure items that are unique to one survey or the other, the choice of which survey to use as the source of data is obvious. However, there is considerable
overlap in coverage between the two surveys, and when dealing with integrated data, careful analysis and thought must be given when choosing the appropriate survey
from which to select specific expenditure items. When data are available from both surveys, BLS, in conjunction with members of the BLS CPI program, conduct
statistical analyses to decide the most reliable survey source. The team relies on three key statistical methods to decide which survey is more reliable including: counts
sufficiency; statistical significance through the transmission mechanism of weighted z-scores; and, starting in 2020, the introduction of edit rates to gauge data quality. In
this way, some items are selected from the Interview Survey, and others are selected from the Diary Survey. BLS reviews the survey sources every 2 years, with the
intent to move to an annual evaluation period to minimize lag time if there is a clear change in source selection in the year between evaluations. For source selection
details, see “CE Source Selection for Publication Tables” in the Consumer Expenditure Survey Anthology, 2011 (BLS Report, no. 1030).
For the year 2021, the CE and CPI team made five source selection changes. Four of the changes involved expenditure data at the Universal Classification Codes (UCCs)
level moving from the Diary to the Interview Survey including: videogames, hardware, and accessories; telephones and accessories; power tools; and athletic gear, game
tables, and exercise equipment. One UCC switched from Interview to Diary, that being Girls’ skirts. In addition to the five source selection changes mentioned above, the
team adjusted a substantial number of clothing UCCs. As of the second quarter of 2021, three global clothing UCCs were added to the Interview Survey STUB file.3 At
the same time, 20 clothing UCCs moved from the Interview to the Diary STUB file, to better reflect optimal survey placement. Additional details and specifics on the
movement of said UCCs is available upon request.
Additional key CE and CPI information

As noted above, the CPI market basket is revised in accordance with results from the CE, however, the population coverage of the CE differs from that of the CPI. For
example, the CE data cover the total population, including rural areas, whereas the CPI covers only the population in urban areas. With the latest CPI geographic
revisions, BLS is making efforts to align the urban areas sampled by the CPI and the CE.
Definitions of components also differ between the CE and CPI. For example, homeownership is treated differently in the two surveys. Actual expenditures of
homeownership (mortgage payments) are reported in the CE, whereas the CPI uses a rental equivalence approach that estimates the change in the cost of obtaining, in
the rental marketplace, services equivalent to those provided by owner-occupied homes. (For an overview of the CPI methodology, see the Handbook of Methods section
on the Consumer Price Index.)
Interpreting CE data
When interpreting CE data, the user should keep a few key points in mind. Expenditures are averages for CUs with specified characteristics, regardless of whether any
individual CU with those characteristics incurred an expense for a given item during the survey collection. Take new cars for example. The average price of a new car,
according to Kelly Blue Book, was $47,077 for 2021.4 Conversely, the average annual expenditure for new cars and trucks in the year 2021 was just under $2,200. Thus,
the average expenditure shown for an item may be considerably lower than the average expenditure by the CUs who purchased the item. The less frequently an item is
purchased, the greater the difference between the average for all CUs and the average for those purchasing the item. Similarly, an individual CU may spend more or less
than the average. Factors such as income, age and number of family members, and the geographic location of the CU influence its expenditures. For example, a CU in
the fifth income quintile will likely spend more on food on out-of-town trips than a CU in the first income quintile due to the fifth income CU’s higher disposable income.
In addition, even within groups with similar characteristics, the distribution of expenditures varies substantially, as consumer tastes and preference, prices, and other
factors not collected in the Consumer Expenditure Surveys (CE) (e.g., health status of CU members) also influence expenditure patterns.
Such points should be considered when comparing reported averages with the circumstances of any individual CU of interest. Users of these survey data should also
keep in mind that the data reflect conditions at the time they were collected, which may be different than current circumstances. For one thing, prices may have
changed. All prices, as measured by the CPI-U, increased 4.7 percent from 2020 to 2021 (annual average index). Undoubtedly, some prices rose more than the average
4.7 percent, some rose less, some stayed the same, and some may have even declined, either within this period or after 2021.
In addition, sample surveys are subject to two types of error: sampling and non-sampling. Sampling errors arise from the constraint of not being able to survey the entire
population. Whenever data are collected from a subset of a larger entity and are extrapolated to apply to all elements of that entity, there is bound to be error in the
data. Therefore, the mean of the sample may differ from the mean that would be obtained if data from the entire population were available. On the other hand, non-
sampling error results from data collection constraints and inconsistencies. Any of the following are sources of non-sampling error: the inability or unwillingness of
respondents to provide correct information regarding spending patterns, differences in interviewers’ abilities to harvest information from respondents, mistakes in
recording or coding, and other processing errors. For additional information on these types of errors, see the sampling and non-sampling errors question on the CE FAQs
page.
Data products
Standard tables, 1960–61, 1972–73, 1984–2021
Tables in this report include data integrated from the Diary Survey and Interview Survey components of the CE. Integrated data enables users to paint a more complete
picture of CU spending dynamics by pulling key components such as food and apparel from the Diary Survey and rent, utilities, and household appliances from the
Interview Survey. Integrated tables also provide expenditure breakdowns by the 17 demographic characteristics collected in the CE, allowing for analyses of
comprehensive spending trends by such characteristics as education level, occupation, and housing tenure. For more detail than is provided in this report, see the CE
tables archive. This archive provides tables from as far back as 1960–61.
In addition to annual tables, CE offers standard tables with a 2-year collection period. The 2-year tables are available in two broad categories: cross-tabulated tables and
geographic area tables. Cross-tabulated tables analyze the intersection of two demographic characteristics. Examples of both 2-year table categories are listed below.
Cross-tabulated tables
income before taxes, cross-tabulated by age, CU size, or region.
region of residence by income before taxes or housing tenure
single consumers by sex of reference person, cross-tabulated by either income (of the CU) or age (of reference person).
Geographic area tables
selected states by income before taxes including California, Florida, New York, New Jersey, and Texas.
selected metropolitan statistical areas (MSAs), such as New York City, Chicago, Washington D.C., and San Francisco.
Detailed tables, 1984–2021

To complement the standard tables, BLS also produces detailed level CE tables on request. Detailed level tables contain additional sub-categories of spending by
demographic characteristics, allowing users to not only view how much a CU spends on “public and other transportation,” but also spending levels on airline tickets,
intracity mass transit spending, and local transportation on out-of-town trips. However, while detailed level tables provide a more robust breakdown of spending data,
they have larger variances relative to other tables. For additional information on variability concerns, please see How does the variability of Consumer Expenditures affect
your analysis. A detailed table showing results at the all consumer units level, at even the most detailed breakdown of expenditures, is available at www.bls.gov/cex
/tables.htm#topline. Detailed level tables broken down by demographics are available by request by reaching out to the following email address: CEXInfo@bls.gov.
New additions to CE tables

For the 2019 release, CE adjusted how it presents urban and rural expenditures in tabular form. Urban and rural data had been available since 1984, but as a component
of the housing tenure table. For the 2019 release, urban and rural data was separated into a “new” table known as “area type.” The purpose of this most recent
subgroup is to offer the ability to conduct analyses of spending habits by whether a CU is geographically located in an urban or rural setting. For collection years 2019
and 2020, BLS used its own unique definition of what constitutes “urban” and “rural,” on the basis of metropolitan and micropolitan statistical areas (MSA’s and MiSA’s).
For collection year 2021, the switch was made to the most current definitions of what constitutes an urban and rural area (as designated by the Census Bureau), while
simultaneously adopting the Office of Management and Budget’s (OMB) definition of “principal city.” For the 2021 CE data release, CE produced the new type of area in
table 1721.
Public use microdata

The 2021 CE public use microdata (PUMD), including the Interview Survey files, Diary Survey files, and paradata (information about the data collection process), are
available at www.bls.gov/cex/pumd.htm. The Interview Survey files contain expenditure data in three distinct formats: MTBI, FMLI, and EXPN files. MBTI files present
monthly values in an item-coding framework based on the CPI-U pricing scheme. FMLI files present user-friendly summary expenditure variables. Finally, EXPN files
present detailed data files that organize expenditures by the section of the Interview questionnaire in which they are collected. Expenditure values on detailed data files
cover different time periods depending on the specific questions asked. In addition, these files also contain relevant non-expenditure information not found on the MTBI
files. The Diary Survey files contain expenditure data in two different formats: EXPD files that present weekly values in the same item-coding framework based on the
CPI-U pricing scheme and FMLD files that present user-friendly summary expenditure variables. Documentation of the CE PUMD, its conventions, files, sample code, and
methodology, can be found at www.bls.gov/cex/pumd-getting-started-guide.htm.
New estimates of local, state, and federal tax liabilities were included in the 2013 PUMD release and thereafter. The CE introduced these estimates to improve the quality
of surveyed tax liabilities, which suffered from low response rates. For more information on the aforementioned improvements and for recent applications of said tax
estimates, please see Improving Data Quality in the Consumer Expenditure Survey with TAXSIM and New Estimates of Personal Taxes in the Consumer Expenditure
Survey. Free public use microdata are available at the PUMD data files webpage for SAS, STATA, SPSS, and Excel for the years 1980 to 2021.
Recent publications highlighting CE data

To highlight the far-reaching scope of the data, economists in BLS frequently produce research for both agency and public audiences. BLS staff members actively
published articles in the following BLS publications: The Economics Daily, Spotlight on Statistics, Beyond the Numbers, and the Monthly Labor Review.
The Economics Daily provides readers with high-level, short highlights and headlines of current developments using BLS data. Recent articles published in The Economics
Daily from CE economists include “Consumer spending on public transportation rebounded in 2021 after a large decline in 2020,” “Single men outspent single women on
entertainment in 2019–20, but single women spent more on pets,” “Changes to consumer expenditures during the COVID-19 pandemic,” and “Consumer expenditures on
travel declined sharply from 2019–2020.”
Articles in Spotlight on Statistics present their stories via a collection of charts, focusing on the visual aspect of presenting CE data to the viewer. The most recent CE
publication is “Meal Appeal: Patterns of expenditures on food away from home.”
Beyond the Numbers articles take a more in-depth look at a topic of interest, focusing on more in-depth analysis and commentary, with a selection of related charts and
tables. As of December of 2022, the most recent Beyond the Numbers articles highlighting CE data include “How have healthcare expenditures changed? Evidence from
the Consumer Expenditure Surveys” and “Receipt and use of stimulus payments in the time of the COVID-19 pandemic.”
The most comprehensive analyses of CE data can be found in the Monthly Labor Review. Articles in this journal exhibit more technical concepts and use statistical
techniques not found in other publications. The most recent articles highlighting CE data in the Monthly Labor Review include "SNAP participation and food-at-home
expenditures through the Great Recession: United States and the New York Area" and “Assessing Consumer Expenditure Surveys data quality through the lens of data
use.”
In addition, the CE research library includes general articles and research papers using CE data, including instructional and how-to documents. The CE data quality and
comparisons profile provides users with a comprehensive analysis on how CE data compares with other seminal outlets that release expenditure data on a frequent basis.
Outlets that have CE comparison profiles include but are not limited to: the American Community Survey, Personal Consumption Expenditure, and Panel Survey of
Income Dynamics. For those interested in information on the methodology used to calculate and collect CE data, including the CE data quality profile, please reference
the CE Handbook of Methods.
Other survey information

Other survey information available on the website includes answers to frequently asked questions and a glossary of terms. Beginning with the CE 2000 data, estimates of
standard errors for integrated Diary and Interview Survey data are also available.
Upcoming events
BLS sponsors a microdata users’ workshop. The event is free, although registration is required. The microdata users’ workshop is intended to provide attendees with a
stable foundation of microdata knowledge that they can build on. The workshop starts with presentations for first time PUMD users, whether it be calculating basic
sample means or taking a dive into the MTBI or FMLI files. The workshop includes presentations and exercises that build upon the established baseline knowledge with
more complex assignments.
The workshop also features presentations from researchers not affiliated with BLS, who describe the nature of their projects, the specific files they use, the variables they
use, the problems they encountered, the solutions they used when working with the data, and any other relevant topics. Finally, the workshop features opportunities to
meet with an expert from the CE staff to discuss any aspect of a current or potential project, general or specific, about which the attendee has questions or concerns.
The next CE Microdata Users’ Workshop will be held in the summer of 2023. More information about these events is available on the CE website (https://www.bls.gov
/cex/ceworkshopthankyou.htm). Reports on these events (2009 through 2020) are also published in the Monthly Labor Review.
Contact information
For more detailed information on the availability of current and earlier data, contact the Division of Consumer Expenditure Surveys, Office of Prices and Living Conditions,
Bureau of Labor Statistics, 2 Massachusetts Avenue, NE, Washington, DC 20212-0001; call (202) 691-6900; email: CEXInfo@bls.gov; or visit the Consumer Expenditure
Surveys page. Information in this publication is in the public domain and, with appropriate credit, may be reproduced without permission. If you are deaf, hard of
hearing, or have a speech disability, please dial 7-1-1 to access telecommunications relay services.
Additional notes
1In 2015, the Interview Survey moved from a five to four panel wave. This was done to reduce non-response rates in subsequent waves, increase rates of reporting,
and shorten Interview length. The “bounding” interview referred to the first wave of the five-wave sequence.
2 A global question is one that collects broad, general information on the item of interest. For example, the Diary Survey collects detailed information on purchases of
food at home, such as rice or chuck steak. In contrast, the Interview Survey asks the global question, “what has been (your/your household’s) usual WEEKLY expense for
grocery shopping?”
3 STUB files show “hierarchical groupings,” which include a description of each UCC along with its hierarchical standing within each expenditure or income category for a
given year. Hierarchical groupings are not available for years prior to 1996. For years 1996 and forward, three file types are available:
“Integrated groupings” lists UCCs that the CE tables use, and identifies the survey source for the UCCs. These files use this naming convention: CE-HG-Integ-2017.
“Interview groupings” lists the UCCs from the Interview Survey. These files use this naming convention: CE-HG-Inter-2017. Not available for 1996.
“Diary groupings” list the UCCs from the Diary Survey. These files use this naming convention: CE-HG-Diary-2017. Not available for 1996.
4 As reported by Kelly Blue Book in January of 2022, the average price of a new car that rolled off the lot was $47,077, considerably more than the average annual
expenditure for a car at the all consumer units level. See Sean Tucker, “Average new car price tops $47,000,” Kelly Blue Book, January 14, 2022, https://www.kbb.com
/car-news/average-new-car-price-tops-47000/.
U.S. BUREAU OF LABOR STATISTICS Division of Information and Marketing Services PSB Suite 2850 2 Massachusetts Avenue NE Washington, DC 20212-0001
Telephone:1-202-691-5200 Telecommunications Relay Service:7-1-1 www.bls.gov/OPUB Contact Us

BLS Consumer Spending 2022 PDF

Uploaded by

Copyright:

Available Formats

BLS Consumer Spending 2022 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BLS Consumer Spending 2022 PDF

Uploaded by

Copyright:

Available Formats

January 2023 Report 1101

Consumer Expenditures in 2021

Consumer Expenditures in 2021

Effects of inflation and supply chain disruptions in 2021

Consumer unit characteristics:

Average number in consumer unit:

People 2.5 2.5 2.5 2.4 [2] [2] [2]

Children under 18 .6 .6 .6 .6 [2] [2] [2]

Adults 65 and older .4 .4 .4 .4 [2] [2] [2]

Earners 1.3 1.3 1.3 1.3 [2] [2] [2]

Vehicles 1.9 1.9 1.9 1.9 [2] [2] [2]

Percent homeowner 63 64 66 65 [2] [2] [2]

Food 7,923 8,169 7,310 8,289 3.1 -10.4 13.4

Food at home 4,464 4,643 4,935 5,259 4.0 6.3 6.6

Dairy products 449 455 474 492 1.3 4.2 3.8

Alcoholic beverages 583 579 478 554 -.7 -17.4 15.9

Housing 20,091 20,679 21,417 22,624 2.9 3.5 5.6

Shelter 11,747 12,190 12,604 13,258 3.8 3.4 5.2

Owned dwellings 6,678 6,797 7,473 7,591 1.8 9.9 1.6

Rented dwellings 4,249 4,432 4,408 4,684 4.3 -.5 6.3

Other lodging 821 961 722 983 17.1 -24.9 36.1

Household operations 1,522 1,570 1,465 1,638 3.2 -6.7 11.8

Housekeeping supplies 747 766 837 803 2.5 9.3 -4.1

Apparel and services 1,866 1,883 1,434 1,754 .9 -23.8 22.3

Transportation 9,761 10,742 9,826 10,961 10.1 -8.5 11.6

Healthcare 4,968 5,193 5,177 5,452 4.5 -.3 5.3

Entertainment 3,226 3,090 2,909 3,568 -4.2 -5.9 22.7

Reading 108 92 114 114 -14.8 23.9 0

Education 1,407 1,443 1,271 1,226 2.6 -11.9 -3.5

Miscellaneous 993 899 907 986 -9.5 .9 8.7

Cash contributions 1,888 1,995 2,283 2,415 5.7 14.4 5.8

Chart 1. Annual percent change for major expenditure groups, 2020–21

Personal care products and services

Hover over chart to view data.

Expenditure shares: how do consumers divide up their spending?

Chart 2A. Percentage of expenditure shares by selected categories, 2018–21

2018 2019 2020 2021

Personal insurance and pensions

Cash contribu ons

Apparel and services

All other expenditures

Chart 2B. Percent distribution of expenditure shares, 2021

Table B. Percent distribution of total annual expenditures

Average annual expenditures 100.0 100.0 100.0 100.0

Food 12.9 13.0 11.9 12.4

Food at home 7.3 7.4 8.1 7.9

Food away from home 5.6 5.6 3.9 4.5

Alcoholic beverages 1.0 .9 .8 .8

Housing 32.8 32.8 34.9 33.8

Shelter 19.2 19.3 20.5 19.8

Utilities, fuels, and public services 6.6 6.4 6.8 6.3

Household operations 2.5 2.5 2.4 2.4

Housekeeping supplies 1.2 1.2 1.4 1.2

Household furnishings and equipment 3.3 3.3 3.8 4.0

Apparel and services 3.0 3.0 2.3 2.6

Transportation 15.9 17.0 16.0 16.4

Vehicle purchases (net outlay) 6.5 7.0 7.4 7.2

Other vehicle expenses 4.7 5.5 5.7 5.3

Public and other transportation 1.3 1.2 .4 .7