US Median Household Income

Short-answer box

US median household income was $87,460 in 2025, according to the US Census Bureau. That figure is the midpoint of the household income distribution: half of households had more income and half had less. It is a pretax money income figure, adjusted for inflation, from the Census Bureau report Income in the United States: 2025. It rose 2.6 percent in real terms from 2024 and was the highest on record since the series began in 1967. The median is not an average salary, not income per person, and not take home pay. It describes a typical household, not any particular household, and on its own it says nothing about wealth, debt, hours worked, or how far that income goes in a particular place.

What the headline figure actually means

Three words do most of the work in this statistic: median, household, and income. If every household were lined up from lowest income to highest, the median would be the household in the middle. That differs from the mean, or average, which adds up all incomes and divides by the number of households. Income is not spread evenly, and a small number of very high incomes can pull a mean sharply upward while leaving the middle almost unchanged. The median resists that pull: a billionaire moving into a town changes the town mean a great deal but moves the median hardly at all, because it adds only one household above the middle. When the question is what a typical household experiences, the median is usually the more informative single number, even though it hides a great deal about the spread around it.

A short worked example shows the arithmetic. These numbers are illustrative only, invented to show the calculation, and they are not survey results. Imagine five households with annual incomes of $30,000, $50,000, $80,000, $90,000, and $500,000. The median is $80,000, the middle value in order. The mean is $150,000, because the five incomes add up to $750,000 and $750,000 divided by five is $150,000. Real data contain millions of households, so the gap is proportionally smaller than in this deliberately lopsided example, but the logic is the same.

In Census Bureau usage, a household consists of all the people who occupy a housing unit, regardless of relationship: one person living alone, a married couple, a family with children, several unrelated roommates, or more than one family living together. A family is narrower: two or more people, one of whom is the householder, related by birth, marriage, or adoption and living in the same housing unit. That breadth matters, because household income adds together the income of the people in the household. A household with two earners will usually report more than a single person household, even if each individual earner makes the same amount. A rise in median household income can therefore reflect more earners per household, a shift toward larger households, or higher pay, and the headline number alone cannot tell you which happened.

The third word is income, and the definition is more specific than everyday speech. The headline figure is money income: pretax cash income that does not count the value of in kind transfers, such as assistance received as goods rather than cash. It is not the same as post tax income, which the Census Bureau also publishes. It is not wealth, which is what a household owns minus what it owes. And it is not spending power after housing costs in a particular city, because the national figure is not adjusted for local prices. Finally, the figure is a survey estimate, not a count of every household. The 2025 report states that its comparative statements are statistically significant at the 90 percent confidence level unless otherwise noted, so small year to year movements may not be statistically meaningful.

The latest figure, with its year attached

The most recent headline figure at the time of writing is for calendar year 2025. US real median household income was $87,460 in 2025, up 2.6 percent from the previous year and the highest on record since 1967, when the Census Bureau began tracking this measure. The Census Bureau published that result in Income in the United States: 2025, based on data collected in the 2026 and earlier CPS ASEC. The margin of error reported alongside the 2025 median was plus or minus $1,040, so the estimate should be read as approximately $87,460 rather than as exact to the dollar. By simple arithmetic, $87,460 per year is about $7,288 per month before tax, a division of the annual median, not a separate survey result.

The comparison points must be stated in the same inflation adjusted terms. The Census Bureau reported that the median household income estimate of $85,210 for 2024 was not statistically different from the estimate in 2019, the last year before the COVID-19 pandemic and the record high to that point. The 2025 figure of $87,460 was higher than both 2019 and 2024. In other words, 2025 was the first year in which the inflation adjusted median clearly moved above its pre pandemic peak on this measure.

A reader who followed the previous release may notice a different 2024 number. In September 2025, the Census Bureau announced that real median household income was $83,730 in 2024, not statistically different from the 2023 estimate of $82,690. The 2025 report gives $85,210 for 2024 instead. The 2025 report itself explains the reason: estimates for 2024 in that report may not match those published the previous year because Vintage 2025 population controls were implemented. Survey estimates are revised when the population weights used to scale the sample up to the whole country are updated, and revised figures supersede earlier ones for the same year. This is normal survey practice, not a contradiction, but it is why a figure should always be cited with its income year, its report, and its survey.

The Census Bureau also publishes a post tax version of the same concept. Post tax median household income was $76,060 in 2025, up 3.1 percent from $73,760 in 2024, but it was not a record high: it did not surpass the levels of 2020 or 2021, when households received stimulus payments and other credits in response to the COVID-19 pandemic. A household comparing its own after tax income with the headline median is comparing two different definitions, and the post tax figure is the closer comparison for that purpose.

Real income, nominal income, and inflation adjustment

Nominal income, also called current dollar income, is income in the dollars of the year it was received, with no adjustment for price changes. Real income, also called constant dollar income, is income after adjustment for inflation, so amounts from different years are expressed in dollars of equal purchasing power. Suppose a household earned $50,000 in one year and $55,000 ten years later. In nominal terms, income rose by 10 percent. But if consumer prices rose by 20 percent over the same period, that household could buy less than before, so its real income fell. Any statement that incomes have risen or fallen over time, without saying whether prices have been taken into account, is incomplete.

The Census Bureau adjusts its historical CPS ASEC estimates using price series from the Bureau of Labor Statistics: the C-CPI-U from 2000 onward, the R-CPI-U-RS between 1978 and 1999, and the CPI-U-X1 series, an experimental series that preceded the R-CPI-U-RS, for 1967 to 1977. Those series are combined into one continuous series, and the 2025 report expresses its historical estimates in real, or 2025, dollars on that basis. The ACS follows a slightly different convention, using the R-CPI-U-RS for all years. The practical rule is simple: when comparing two years, make sure both figures are in the same dollar year, or that the change quoted is explicitly a real, inflation adjusted change.

How the Census Bureau measures it: CPS ASEC versus ACS

The United States does not have a single income survey. It has several, designed for different purposes, and they can give different answers to what looks like the same question. The two that matter most here are the Current Population Survey Annual Social and Economic Supplement, known as CPS ASEC, and the American Community Survey, known as ACS.

The CPS ASEC is the official source for annual estimates of income and poverty at the national level. The Current Population Survey is sponsored jointly by the Census Bureau and the Bureau of Labor Statistics, and it is the primary source of labor force statistics for the United States, including the official monthly unemployment data. The Annual Social and Economic Supplement is added every February, March, and April, collects data on work experience, income, household composition, health insurance coverage, and receipt of noncash benefits, and asks about the prior calendar year. That timing explains the lag that often confuses readers: a report released usually in September of one year describes income earned in the previous calendar year. The headline national median, including the $87,460 figure for 2025, comes from the CPS ASEC.

The American Community Survey is an ongoing survey conducted by the Census Bureau since 2005, collecting detailed social, economic, housing, and demographic information across the 50 states, the District of Columbia, and Puerto Rico, covering more than 40 topics including income. Because of its large sample size, the ACS is exceptionally useful for subnational analysis. It provides single year estimates for places, counties, and metropolitan areas with a population of at least 65,000, as well as for the nation and the states, and five year pooled estimates for smaller geographies.

The two surveys also differ in reference period. The CPS ASEC asks about income in the prior calendar year. The ACS asks about income in the past 12 months relative to the interview, and interviews are spread across the year, so an ACS single year estimate blends months from two calendar years. The surveys differ in sample size, in how they handle nonresponse, and in their inflation adjustment, as noted above. A concrete illustration helps: a summary of ACS data reported a US median household income of $81,600 in 2024, while the CPS ASEC release for 2024 reported $83,730 in its original vintage, a gap of about $2,100. That gap is the normal consequence of different surveys, reference periods, and methods. One further note on quality: the September 2025 press release reported a weighted response rate of 62.0 percent for the 2025 CPS ASEC, compared with 61.7 percent the previous year, with weights adjusted for nonresponse. Those adjustments reduce bias but do not eliminate it, which is why a change described as “not statistically different” should not be reported as a rise or a fall.

Trends over time: a long rise, an uneven distribution

From 1967 to 2025, pretax household income at the 10th percentile and at the 50th percentile, the median, each increased by about 56 percent in real terms. Over the same period, income at the 90th percentile rose by about 121 percent. The ratio of the 90th percentile to the 10th percentile widened from 9.23 in 1967 to 13.06 in 2025, so households at the 90th percentile had income about 13 times greater than households at the 10th percentile in 2025, compared with about 9 times greater in 1967. The gap between the top and the bottom grew over the long run, even as the middle rose in real terms. That is why the median alone is an incomplete summary: a rising median can coexist with a widening spread, with stagnation at the bottom, or with rapid gains at the top. A reader who wants to know whether growth was broadly shared should look at the 10th, 50th, and 90th percentiles together.

The shorter run story has its own shape. The inflation adjusted median reached its pre pandemic high in 2019, then fell or stood still through the pandemic years on the pretax measure, and only in 2025 moved clearly above both 2019 and 2024. The post tax story differs, because pandemic era payments and credits lifted post tax income in 2020 and 2021 above its 2025 level. At the top, the 2025 report notes a third consecutive year over year increase in both pretax and post tax income at the 90th percentile. At the bottom, by contrast, pretax and post tax income at the 10th percentile in 2025 did not differ significantly from 2024.

Variation across the country and across groups, in general terms

A national median is a summary of a very diverse country, and it should never be the end of the comparison. The Census Bureau publishes breakdowns by state, by household type, by the age, education, race, and Hispanic origin of the householder, and by other characteristics, and those breakdowns show wide variation. This article describes that variation in general terms: the detailed subgroup tables of the 2025 report were not reproduced in the Census Bureau summary pages consulted for this draft, so subgroup dollar figures are not quoted here from secondary summaries. The Sources section gives the authoritative tables, and a future update can add subgroup figures once they have been checked directly against them.

State medians differ by tens of thousands of dollars. As an ACS based illustration, a summary of 2024 ACS data placed state medians from $104,828 in Massachusetts to $59,127 in Mississippi, with the District of Columbia higher still at $109,700. Those state figures come from the ACS, not from the CPS ASEC national headline, because state comparisons belong to the survey designed for subnational analysis. The spread reflects differences in local job markets, industry mix, education levels, age structure, household size, and the cost of living. It does not, on its own, show that a household is better off in a high median state, because the same income buys less in an expensive metropolitan area than in a lower cost area, and the median is not adjusted for those price differences.

Household type matters at least as much as place. Family households, which contain people related by birth, marriage, or adoption, generally report higher medians than nonfamily households, which include people living alone and unrelated people sharing a home. That is largely arithmetic: a family household is more likely to contain two earners, and a one person household has only one potential earner by definition.

Age of the householder shapes the pattern in a predictable life cycle way. Households headed by people in their prime working years tend to report higher medians than households headed by very young adults, early in their careers, or by retired people, whose income more often comes from pensions and Social Security than from earnings.

Education is one of the strongest correlates of household income in the published breakdowns, because attainment is closely associated with occupation, with full time year round work, and with earnings over a working life. Households headed by someone with a bachelor’s degree or higher report substantially higher medians than households headed by someone without a high school diploma. That ordering is a correlation across groups, not a promise to any individual, and it does not separate the contributions of education, occupation, place, age, household size, and hours worked, nor the cost, debt, or time involved in obtaining a qualification.

Race and Hispanic origin breakdowns show persistent gaps between groups, and they require the most careful reading of any breakdown in this article. The gaps reflect a long history of differences in education, occupation, wealth, place of residence, household structure, and access to opportunity, and they cannot be reduced to any single cause. Two cautions are essential. First, a group median describes the middle household in that group, not every household in it. Second, a change in a group median from one year to the next can reflect changes in who is in the group, in household composition, or in survey methods, as well as changes in earnings, and subgroup samples are smaller, so their margins of error are larger. The practical lesson is to compare like with like: households of similar size and type, in a similar place, using the same survey, definition, and year.

What the median does not show

The median does not show wealth: a retired household with a modest annual income but a paid off home and substantial savings can be financially secure, while a high income household with large debts and no savings can be financially fragile. Income is a flow over one year; wealth is a stock accumulated over a lifetime. It does not show how many hours were worked, or by how many people: $87,460 earned by one person working full time year round is a different working life from the same total earned by two people each working part time. It does not show what the income can buy where the household lives, because the national median is not adjusted for regional price differences. It does not show income after tax, after housing costs, or after other unavoidable spending, except in the separate post tax series, so the headline pretax median will overstate take home pay for most households. It does not show inequality within the middle. It does not show individual prospects: a median describes a population in a past year, not a prediction for any person. This article is a statistics explainer. It offers no financial advice and makes no claim about what any individual reader will earn, should earn, or should do with their income. Finally, it does not show the future: the 2025 figure describes income earned in 2025, and the next release may revise the picture.

Common misreadings, and how to avoid them

“The average American earns $87,460.” No. The figure is a median for households, not a mean for people, and the mean household income is higher because very high incomes pull it up. If the question is what a typical worker earns, the relevant figure is median earnings for a person, by work experience, not median household income. “Half of Americans earn less than $87,460.” Not quite: half of households have household income below $87,460 and half have more, which is a statement about households, not about individual Americans. “Incomes rose 2.6 percent, so everyone is 2.6 percent better off.” No: a rise in the median says that the middle household did better in real terms, not that every household did better. In 2025, income at the 10th percentile did not differ significantly from 2024, even as the median rose. “The median is the same as middle class.” The Census Bureau does not have an official definition of the middle class. Commentators define it in different ways, often as a range around the median, but those are choices made by the commentator, not by the Census Bureau. “A higher median state is a better place to live.” A state median reflects its job market and its population, not its quality of life, and it is not adjusted for what income buys there. “Last year’s figure was wrong because this year’s report revised it.” Revisions from updated population controls, improved methods, or corrected weights are a normal part of survey statistics, and comparisons should use the revised series consistently rather than mixing vintages. “My household is below the median, so we are doing something wrong.” A median is a dividing line, not a standard. By construction, half of households fall below it, including many managing well on a single income, in a low cost area, or at a stage of life when earnings are naturally lower.

How to use the figure well

Used with its limits in mind, the median is a genuinely useful tool: a quick, comparable answer to what a typical household’s money income looked like in a given year, in a given place, on a stated definition. Name the survey, CPS ASEC for the national headline or ACS for state and local figures. Name the income year, not just the publication year. Name the definition, pretax money income or post tax income. Name the dollar year for historical comparisons, or quote an explicitly real change. State the unit, household, and do not convert it silently into a person, a family, or a worker. Note the margin of error when a small change is the story, and do not report a change described as not statistically different as though it were a real movement. Checking a figure against the table it came from is the single most reliable habit a reader of income statistics can develop.

Frequently asked questions

What is US median household income right now?

The latest headline figure at the time of writing is $87,460 for calendar year 2025, from Income in the United States: 2025. It is a pretax money income figure, adjusted for inflation, based on the CPS ASEC.

Is median household income the same as average income?

No. The median is the middle value in the income distribution, with half of households above it and half below it. The mean, or average, adds up all incomes and divides by the number of households, and it is pulled upward by very high incomes.

Is it the same as a salary?

No. A salary is what one person is paid for one job. Median household income adds together the money income of all the people in a household, which can contain more than one earner, or no earner at all.

What counts as a household?

A household consists of all the people who occupy a housing unit, regardless of relationship. It can be one person living alone, a family, or unrelated people sharing a home. A family is narrower: two or more people, one of whom is the householder, related by birth, marriage, or adoption and living in the same housing unit.

What counts as income in the headline figure?

The headline figure is money income, which is pretax cash income and does not count the value of in kind transfers. Post tax median household income was $76,060 in 2025, compared with $87,460 pretax. The two figures answer different questions, and they should not be mixed.

Why do different websites give different median figures?

Mostly because they quote different surveys, years, or definitions without saying so. The CPS ASEC figure, the ACS figure, a state figure, a post tax figure, and an earlier vintage of the same report can all differ by thousands of dollars while each being correct on its own terms.

How is the figure adjusted for inflation?

Historical CPS ASEC estimates are expressed in real dollars using Bureau of Labor Statistics price series: the C-CPI-U from 2000 onward, the R-CPI-U-RS between 1978 and 1999, and the CPI-U-X1 series for 1967 to 1977. ACS historical estimates use the R-CPI-U-RS for all years.

Does the median show whether people are better off?

Only in part. A rising real median shows that the middle household’s purchasing power rose. It does not show what happened at the bottom or the top, what happened to wealth or debt, how many hours were worked, or what income buys in a particular place.

Can I compare my household with the median?

Yes, as orientation, provided the comparison is like with like. Compare your pretax income with the pretax median, or your post tax income with the post tax median, and prefer households of similar size and type and the median for your state or area from the ACS. The median says nothing about your spending, debts, savings, health, or local costs.

When will the next figure be published?

The Census Bureau publishes its annual income report usually in September following the income year, based on CPS ASEC interviews conducted in February, March, and April about the prior calendar year. The next release may revise the 2025 figure if population controls or methods are updated.

Keep reading: What Is a Pension?

Sources & references

• US Census Bureau, "Median Household Income in 2025 Surpassed Previous Highs Before and After the COVID-19 Pandemic," Census Bureau story, 2026: https://www.census.gov/library/stories/2026/09/median-household-income.html . Used for the 2025 median of $87,460, the 2.6 percent real rise, the 2024 estimate of $85,210, the post tax median of $76,060 and its 2024 comparison, the 1967 to 2025 percentile growth and ratio, and the 90th and 10th percentile movements in 2025.
• US Census Bureau, "Income in the United States: 2025," Current Population Reports P60-289: https://www.census.gov/library/publications/2026/demo/p60-289.html . Used for the money income and post tax income definitions, the Vintage 2025 population controls revision note, the real 2025 dollar basis, the C-CPI-U and R-CPI-U-RS adjustment, and the 90 percent statistical testing standard.
• US Census Bureau, "Income, Poverty and Health Insurance Coverage in the United States: 2024," press release, September 9, 2025: https://www.census.gov/newsroom/press-releases/2025/income-poverty-health-insurance-coverage.html . Used for the original vintage 2024 median of $83,730, the 2023 estimate of $82,690, the CPS ASEC collection months and prior year reference, and the 2025 CPS ASEC response rate of 62.0 percent.
• US Census Bureau, "Frequently Asked Questions (FAQs) about Income": https://www.census.gov/topics/income-poverty/income/about/faqs.html . Used for the CPS ASEC as the official national source, the ACS as the source for subnational analysis, the household and family definitions, the median and mean definitions, and the statement that the Census Bureau has no official definition of the middle class.
• US Census Bureau, "Current versus Constant (or Real) Dollars": https://www.census.gov/topics/income-poverty/income/guidance/current-vs-constant-dollars.html . Used for the current and constant dollar definitions and the C-CPI-U, R-CPI-U-RS, and CPI-U-X1 series and their date ranges, including the ACS use of R-CPI-U-RS for all years.
• USAFacts, "What is the income of a US household?": https://usafacts.org/answers/what-is-the-income-of-a-us-household/country/united-states/ . Used only for the ACS based state illustration for 2024, Massachusetts $104,828 to Mississippi $59,127 and the District of Columbia $109,700, and for the ACS based national figure of $81,600 in 2024 used to illustrate CPS ASEC and ACS differences. USAFacts states that it sourced its data from the Census Bureau American Community Survey.