The data for the historical analysis in this report is derived from the Annual Social and Economic Supplements (ASEC) of the Current Population Survey (CPS), which are conducted in March of every year. The specific files used in this report are from March 1971 to March 2023 and contain data on the annual income of households from 1970 to 2022. Conducted jointly by the U.S. Census Bureau and the Bureau of Labor Statistics, the CPS is a monthly survey of about 60,000 households and is the source of the nation’s official statistics on unemployment. The ASEC survey in March typically features a larger sample size (about 75,000 in recent years) and is the source of the annual income and poverty estimates reported by the Census Bureau.
The historical analysis for Asian Americans is limited to CPS data for 2010 and 2023. That is because Asian Americans alone were not identified in the CPS until 2003. Prior to that year, and starting only in 1988, the CPS reported data for Asians and Native Hawaiians and Pacific Islanders (NHPI) combined. But the sample size for the NHPI population is only about 300 households in 2010 and 2023. For that reason, we do not report historical trends for the NHPI population. We also do not report historical trends for American Indian and Alaska Native (AIAN) or multiracial populations. These two populations are not uniquely identified in the CPS.
Methodological and other revisions to the CPS may have an impact on estimated trends. For example, the 2015 ASEC introduced a redesigned set of income questions, and definitions of key socioeconomic categories, such as race and educational attainment, have changed over time.
The demographic analysis is principally based on the data from the 2022 American Community Survey (ACS). The public-use version of the ACS is a 1% sample of the U.S. population, or more than 3 million people. It is designed to collect the detailed information previously collected in the long form of the decennial census. But the ACS data are available only from 2005 onwards and are less suitable for long-run historical analyses.
The ACS is conducted in every month of the year, with data collected from about one-twelfth of the total sample in each month. The monthly responses are combined to form an annual portrait of the nation and of smaller geographic units. Because of its large sample size, the ACS is a better source than the CPS for analyses of subnational or subgroup demographic and income data. Nonetheless, we do not report the distribution of NHPI workers across income tiers by occupation and industry because of small sample sizes.
In some instances, the demographic analysis uses data from the 5-year ACS for the period 2018-2022. This file contains data from the five ACS surveys conducted over this period and represents a 5% sample of the U.S. population, or more than 15 million people. We exclude the data for 2020 because the COVID-19 pandemic had a significant impact on data collection in that year. The resulting file with four years of data – about 12 million people – was used to study smaller demographic groups, such as Korean Americans or Pakistani Americans. But we are still limited in the extent to which we can analyze the distribution of the AIAN and NHPI populations across income tiers in U.S. metropolitan areas.
The CPS and ACS microdata used are the Integrated Public Use Microdata Series (IPUMS) provided by the University of Minnesota. The IPUMS assigns uniform codes, to the extent possible, to data collected in the CPS and ACS over the years. More information about the IPUMS, including variable definition and sampling error, is available at IPUMS CPS and IPUMS USA.
Income
Household income is the sum of incomes earned by all members of the household ages 15 and older. Income is defined as money income received (exclusive of certain money receipts, such as capital gains) before payments for such things as personal income taxes, Social Security, union dues and Medicare deductions. Noncash transfers such as SNAP benefits (food stamps), health benefits, subsidized housing and energy assistance are not included. More detail on the measurement and collection of income in government data can be found in the documentation of the CPS or in the documentation of the ACS.
The CPS collects data on income received by the household in the calendar year preceding the date of the survey. The ACS is a rolling monthly survey, and the household income data refer to income received during the 12 months preceding the survey month. In other words, a household surveyed in January 2022 is expected to report income received from January 2021 to December 2021, a household surveyed in February 2022 is expected to report income received from February 2021 to January 2022, and so on. Households surveyed in December 2022 report income received from December 2021 to November 2022. Thus, in the 2022 ACS, the income data refer to the period from January 2021 to November 2022, a span of 23 months.
Because the income data collected in the ACS does not refer to a calendar year, the Census Bureau provides an adjustment factor that converts reported incomes to the levels they would have been had they been earned during a calendar year. Although this adjustment factor has its limits, we apply it to the income data in the 2022 ACS to convert reported incomes to their projected levels in the 2022 calendar year. In the 5-year ACS files, the reported incomes have already been standardized by the Census Bureau to dollars as valued in the final year of data included in the file.
The data on income are adjusted for inflation and reported in 2023 dollars in this report. We use the price index series published in the Census Bureau’s 2022 income report. From 2000 to 2022, this series is the same as the Chained Consumer Price Index for all Urban Consumers (C-CPI-U) published by the Bureau of Labor Statistics (BLS). From 1978 to 1999, this series is the BLS’s Consumer Price Index for all Urban Consumers Retroactive Series (R-CPI-U-RS). For years prior to 1978, the Census Bureau uses an experimental price index series from the BLS known as the CPI-U-X1. We apply the C-CPI-U from the BLS to extend the Census Bureau’s price index series from 2022 to 2023.
The choice of a price index does not affect the allocation of households into lower-, middle- or upper-income categories at a point in time. That is because the same price index applies to the incomes of all households and does not affect their income-based rank. However, the choice of a price index does affect measures of absolute progress over time. For example, from 2000 to 2023, the price level rose either 78% (R-CPI-U-RS) or 67% (C-CPI-U). This means that someone who earned $10,000 per year in 2000 would need to earn either $17,800 (using the R-CPI-U-RS) or $16,700 (using the C-CPI-U) in 2023 to be just as well off as in 2000.
Households in census data
The Census Bureau defines a household as the entire group of persons who live in a single dwelling unit. A household may consist of several persons living together or one person living alone. It includes the household head and all of their relatives living in the dwelling unit and also any lodgers, live-in housekeepers, nannies and other residents not related to the head of the household.
Adjusting income for household size
Household income data reported in this study are adjusted for the number of people in a household. That is done because a four-person household with an income of, say, $50,000 faces a tighter budget constraint than a two-person household with the same income. In addition to comparisons across households at a given point in time, this adjustment is useful for measuring changes in the income of households over time. That is because average household size in the U.S. decreased from 3.1 persons in 1970 to 2.5 persons in 2023, a drop of about 20%. Ignoring this demographic change would mean ignoring a commensurate loosening of the household budget constraint.
At its simplest, adjusting for household size could mean converting household income into per capita income. Thus, a two-person household with an income of $50,000 would have a per capita income of $25,000, double the per capita income of a four-person household with the same total income.
A more sophisticated framework for household size adjustment recognizes that there are economies of scale in consumer expenditures. For example, a two-bedroom apartment may not cost twice as much to rent as a one-bedroom apartment. Two household members could carpool to work for the same cost as a single household member, and so on. For that reason, many researchers make adjustments for household size using the method of “equivalence scales.”
A common equivalence-scale adjustment is defined as follows:
Adjusted household income = Household income / (Household size)N
By this method, household income is divided by household size exponentiated by “N,” where N is a number between 0 and 1.
Note that if N=0, the denominator equals 1. In that case, no adjustment is made for household size. If N=1, the denominator equals household size, and that is the same as converting household income into per capita income. The usual approach is to let N be some number between 0 and 1. Following other researchers, this study uses N=0.5. In practical terms, this means that household income is divided by the square root of household size – 1.41 for a two-person household, 1.73 for a three-person household, 2.00 for a four-person household and so on.
One issue with adjusting for household size is that while demographic data on household composition pertain to the survey date, income data typically pertain to the preceding calendar year or the preceding 12-month period. Because household composition can change over time, for example, through marriage, divorce or death, the household size that is measured at the survey date may not be the same as that at the time the income was earned and spent.
Once household incomes have been converted to a “uniform” household size, they can be scaled to reflect any household size. The income data reported in this study are computed for three-person households, the closest whole number to the average size of a U.S. household since 1970. That is done as follows:
Three-person household income = Adjusted household income * [(3)0.5]
Adjusting for household size does have an effect on trends in income since 1970. However, it is important to note that once the adjustment has been made, it is immaterial whether one scales incomes to one-, two-, three- or four-person households. Regardless of the choice of household size, the same results would emerge with respect to the trends in the well-being of lower-, middle- and upper-income groups.
Adjusting income for the cost of living in metropolitan and other areas
In our analysis of the ACS data, “middle-income” Americans live in households with incomes that are two-thirds to double the national median, after incomes have been adjusted for household size and the cost of living in their area. Their area of residence may be in a known metropolitan area, an unidentified metro area or outside of a metro area.
A metropolitan area consists of at least one urbanized area with a population of 50,000 or more people, plus neighboring areas that are socially and economically integrated with the core. Metropolitan areas may cross state boundaries, such as the Washington-Arlington-Alexandria, DC-VA-MD-WV area.
Our analysis of the state of the middle class for all Americans encompasses 222 of 381 metropolitan areas in the United States, as defined by the Office of Management and Budget in its 2013 classification of metro areas. The 222 metropolitan areas included are the maximum number of areas that could be identified in the ACS (IPUMS) data that also had a sample of at least 500 households. Together, these areas accounted for 72% of the nation’s population in 2022. Overall, the 2022 ACS identifies 248 metro areas.
The cost-of-living adjustment for all areas is based on price indexes published by the U.S. Bureau of Economic Analysis. These indexes, known as Regional Price Parities (RPP), compare the prices of goods and services in an area with the national average prices for the same goods and services. The latest available estimates are for 2022.
RPPs are available for three types of areas. The RPPs for metropolitan areas were used to adjust the incomes of households in metro areas identified in the ACS data. RPPs for the metropolitan portions of a state were used to adjust the incomes of households known to live in a metro area in that state but for whom the identity of the metro area was not known. Finally, RPPs for all nonmetro areas in a state were used to adjust the incomes of households located in nonmetro areas in that state.
Race, ethnicity, nativity, educational attainment and marital status
In our analyses, White, Black and Asian Americans, American Indians or Alaska Natives, and Native Hawaiians or Pacific Islanders are people who identified with a single major racial group and who are not Hispanic. Asian includes people who identified with more than one Asian race or origin, e.g., Japanese and Filipino. Native Hawaiians or Pacific Islanders include people identifying with more than one Micronesian race. Multiracial includes people who identified with more than one major racial group, e.g., White and Black, Black and Korean, or Chinese and Hawaiian, and are not Hispanic. Hispanics are of any race.
U.S. born refers to individuals who are U.S. citizens at birth, including people born in the 50 U.S. states, the District of Columbia, Puerto Rico or other U.S. territories, as well as those born elsewhere to at least one parent who is a U.S. citizen. The terms foreign born and immigrant are used interchangeably in this report. They refer to people who are not U.S. citizens at birth.
High school graduate refers to those who have a high school diploma or its equivalent, such as a General Education Development (GED) certificate, and those who had completed 12th grade, but their diploma status was unclear (those who had finished 12th grade but not received a diploma are excluded). Adults with some college include those with an associate degree and those who attended college but did not obtain a degree. In the estimates for 1971, adults with a bachelor’s degree or higher level of education are those who completed at least four years of college.
Unmarried includes never married, divorced, separated and widowed. Married includes those whose spouse may be absent.
Region of birth and ancestry
Region of birth is defined using traditional geographic boundaries as identified in the IPUMS USA variable BPL. North America and U.S. territories includes the U.S., American Samoa, Guam, Puerto Rico and the U.S. Virgin Islands, Canada, and Atlantic Islands. Central America and the Caribbean includes Mexico, the rest of Central America and Caribbean islands. South America is all countries in South America, and Europe is all countries in Europe, including Russia. Asia is geographic Asia, including the Middle East, and Africa refers to all countries on that continent. Oceania encompasses Australia, New Zealand and various Pacific Islands.
Ancestry refers to a person’s self-reported ancestry or ethnic origin and is based on the IPUMS USA variable ANCESTR1. Respondents may provide more than one ancestry. We use the first-reported ancestry in our analysis. North American refers to U.S. or Canadian origins and American Indian and Alaska Native tribes. Central and South American and other Hispanic ancestry refers not only to places of origin, such as Mexico, Puerto Rico and Peru, but also to ethnic origins, such as Mexican American, Spaniard and Hispanic. Caribbean includes the West Indies and non-Hispanic Central and South America.
South Asian includes origin groups from Afghanistan, India and neighboring countries, except Burma. Other Asian encompasses groups from Burma to the countries on the Pacific rim of Asia. Middle Eastern & North African origin groups are from countries ranging from Morocco to Iran. Sub-Saharan African groups are from Africa, excluding countries in North Africa. Pacific Islanders trace their origin to Australia, New Zealand and Pacific Islands, including Hawaii. Western European and Soviet or Eastern European ancestries are defined along familiar geographic classifications of these regions, except that the latter category also includes Europeans not elsewhere classified in the source data.
Statistical significance
Comparisons between most estimates are tested for statistical significance using the replicate weights in the source data. These tests for statistical significance are conducted using 95% confidence intervals. However, replicate weights are not available in the CPS for years prior to 2005. Thus, when comparisons are made between CPS estimates from 1971 and 2023, or in other instances where replicate weights are not available, the standard errors are determined using the “pweight” option in Stata, and tests of significance are conducted using 99% confidence intervals.
