Wages, income, and inflation
When did the middle actually get ahead?
PennyCalc · Analysis updated: · Source review: · Data through 2025
Averages say Americans earn more than they used to. The more interesting question is when the middle gained ground, and how much it shared in that progress. We put five annual measures on the same inflation basis to look beyond the endpoints.
The surprise is how uneven the progress was. From 1974 to 2000, median earnings for full-time, year-round workers rose just 2.0% after inflation, while the mean rose 20.4%. The middle gained more in later years. Yet by 2025, median worker earnings were only 3.2% above their pre-pandemic 2019 level.
That changes how we read an income headline. A higher average can coexist with modest gains in the middle, and a comparison starting near a recession can make recent progress look much larger. Neither tells us who captured the gains without a closer look at the distribution.
Follow the whole path, not just the final number
The default view gives every line the same starting value: 100 in 1974, after inflation. A value of 120 means 20% more buying power. Both panels use the same vertical scale. Switch to dollars to compare actual amounts, or start in 2019 to compare with the year before the pandemic.
Showing 1974 to 2025: growth from the starting year, after inflation. Both charts use the same vertical scale.
Over this period, median worker buying power changed +20.2%, median household buying power +41.6%, and the federal minimum benchmark -34.3%.
One worker: minimum, median, and mean earnings
Median and mean cover full-time, year-round workers. The federal minimum is an illustrative 40-hour week for 52 weeks at each year-end rate.
- Federal minimum, annualized
- Median worker earnings
- Mean worker earnings
Scroll the chart sideways to see every year. Exact amounts are in the annual data table.
One household: median and mean income
Household income combines money income across household members, including sources beyond earnings.
- Median household income
- Mean household income
Scroll the chart sideways to see every year. Exact amounts are in the annual data table.
Where the story changes
One percentage for the whole period misses the stretches that shaped it. These intervals separate the long early period, the weak years around the Great Recession, the later recovery, and the years since 2019. They are descriptive windows, not estimates of what any one policy caused.
| Period | Federal minimum, annualized | Median worker earnings | Mean worker earnings | Median household income | Mean household income |
|---|---|---|---|---|---|
| 1974-2000 | -18.2% | +2.0% | +20.4% | +19.1% | +38.6% |
| 2000-2012 | +8.8% | +2.5% | +4.3% | -6.1% | -3.6% |
| 2012-2019 | -8.3% | +11.5% | +13.0% | +23.5% | +26.2% |
| 2019-2025 | -19.5% | +3.2% | +3.7% | +2.5% | +4.0% |
1974-2000: the average improved much more than the middle
Median worker earnings ended this period only 2.0% above the starting point after inflation. Mean earnings rose 20.4%. So an account based only on the average would describe a much stronger improvement than the change at the middle of the distribution. The median moved along the way; it did not rise steadily for 26 years.
2000-2012: household income lost ground
Median household income fell 6.1% in real terms between these endpoints. Worker earnings held up better, but that does not mean job losses were harmless. The earnings series only counts people who worked full time and year round, so someone losing that status leaves this population.
2012-2019: a stronger stretch, with measurement breaks
Median worker earnings rose 11.5%, and median household income rose 23.5%. This is a stronger run in the published estimates, but part of the comparison spans changes to the survey questionnaire and processing. The Census footnotes matter here: we cannot treat the entire increase as an economic gain on an unchanged measurement system.
2019-2025: recent progress looks smaller than the long-run comparison
Starting just before the pandemic, median worker buying power grew 3.2% and median household buying power grew 2.5%. Those are positive changes in the estimates, but far smaller than the gains from 2010. Meanwhile, the unchanged federal minimum lost 19.5% of its buying power. These national measures also leave out the different pressures facing renters, parents paying for childcare, or households in expensive regions.
Why the 2020 worker spike needs special care
A rising median can reflect a change in who is counted. Census found that lower-paid workers disproportionately left full-time, year-round employment in 2020, pushing up the earnings median for the workers who remained in that group. Reading that spike as a broad pay rise would miss the people who lost work. This is one reason we use 2019 as the recent reference point.
The average is now much further above the middle
The mean was 13.1% above median worker earnings in 1974. By 2025, it was 38.4% above. The household gap widened from 16.9% to 44.9%. Because both numbers in each ratio use the same year's prices, this gap is not an inflation illusion.
- Solid green: workers
- Dashed blue: households
The middle can become better off while higher incomes pull the average further away. That is the important distinction here: “people earn more” and “the gains were broadly shared” are different claims. The first can be true without establishing the second.
To find out who received the gains, we need income shares, not just a mean and a median. Our income-distribution history follows the top 1% and the rest using CBO data, including the effect of taxes and benefits. It also examines explanations for shifts in the distribution. Its broader income definition should not be spliced into these Census lines.
What to do with this comparison
Use the worker median as a starting point for pay comparisons
If you are asking whether pay has improved for a typical full-time, year-round worker, the real worker median is the closer benchmark. Average household income answers a different question and can make progress look larger. For your own pay, compare the same job, hours, and location where possible; this national series cannot account for those differences.
Check the starting year before accepting a growth claim
A comparison from 2010 includes a long recovery. A comparison from 2019 asks what changed from just before the pandemic. Neither is automatically wrong, but they answer different questions. Use the chart's start-year control to see how much of a headline depends on the chosen window.
Separate an income gain from the room left in your budget
More real income is useful progress. It does not establish that a household has more left after tax, rent, childcare, or debt payments. The inflation calculator checks how a dollar amount changed with national prices; the take-home pay map compares modeled current taxes across states. Local housing costs and your own spending still need a separate check.
The next question raised by these charts is where the gains went. Start with the income-share evidence before concluding that the top 1% captured them. These earnings and income measures do not establish a transfer of wealth between classes.
What changed since 2010?
We use 2010 here because it was the first full calendar year with a $7.25 federal minimum. It was also just after a recession, so it is a starting point with consequences. Use the chart controls to see how the comparison changes over a longer period.
| Measure | 2010 | 2025 | Change |
|---|---|---|---|
| Federal minimum, annualized | $21,455 | $15,080 | -29.7% |
| Median worker earnings | $59,650 | $66,620 | +11.7% |
| Mean worker earnings | $78,410 | $92,220 | +17.6% |
| Median household income | $70,110 | $87,460 | +24.7% |
| Mean household income | $95,880 | $126,700 | +32.1% |
A much bigger paycheck can mean a modest gain in buying power
Before adjusting for prices, median worker earnings rose from $41,920 to $66,620, or 58.9%. After inflation, the increase was 11.7%. Both numbers describe the same change. The first is the growth in the paycheck; the second asks how much more it could buy using the Census price index.
The federal minimum tells a much narrower story
A worker earning the federal floor for 2,080 hours would receive $15,080 before taxes. That dollar amount has not changed since 2009. It buys less now, but it is not a measure of what all low-paid workers earn: many states and cities set higher minimums, and coverage and exceptions matter. Our federal minimum-wage history shows the changes behind this line.
On this inflation measure, matching the buying power of the 2010 federal minimum would take about $10.32 an hour in 2025. That is a price-adjusted comparison, not a proposed wage or an estimate of a local living wage.
For savings tied to inflation, our I Bond comparison shows another place the starting year matters: the year of purchase locks in a different lifetime fixed rate.
Sources and methodology
The annual series runs from 1974, the first year shared by the selected earnings table, through 2025, the latest completed income year in the Census release checked on October 13, 2026. We keep individual earnings and household income in separate charts because their populations and income definitions differ.
- Individual earnings: Census Table P-43, “Worked Full-Time, Year-Round,” both sexes. Earnings include wage and salary income and net self-employment earnings. This is not total personal income. The Census definition uses 35 or more hours per week during a majority of weeks worked, and at least 50 weeks in the year; it does not imply that everyone worked exactly 2,080 hours.
- Household income: Census Table H-5, all races. Money income is measured before personal income taxes and excludes capital gains and noncash benefits. The median divides the population in half; the mean is total income divided by the number of households or workers in the selected population.
- Federal minimum: the Department of Labor’s historical rate in effect at each year-end, multiplied by 40 hours and 52 weeks. This is an annualized benchmark, not an estimate of earnings actually received in years with a midyear increase. Earlier coverage expansions had different rates; this line follows the rate for workers originally covered by the 1938 Act.
- Inflation: the income lines use Census’s published 2025-dollar columns. We convert the minimum-wage benchmark using the same Census annual price series: annualized pay × 2025 index ÷ that year’s index. It combines C-CPI-U from 2000, R-CPI-U-RS for 1978-1999, and CPI-U-X1 for our earlier years. The 2025 index includes Census’s approximation for the missing October observation. Published rounding can produce small differences when reconstructing the income columns.
- Survey breaks: we select the redesigned-questionnaire estimate for 2013 and the updated-processing estimate for 2017. We do not splice the series to remove those breaks. Changes in age coverage, population controls, collection, and processing also affect comparisons. The Census historical footnotes document these changes; the JSON retains the source labels and row numbers.
The growth view sets each line to 100 in the selected starting year after inflation, using the same vertical scale for both panels. Dollar views use separate scales. The mean-premium chart divides each year's mean by its median, subtracts one, and multiplies by 100; it is a gap in levels, not an income share. These are repeated population estimates, not a panel tracking the same people. We have not calculated confidence intervals, and the charts cannot isolate the effect of a policy or explain what caused a change.
Our minimum-wage history page uses CPI-U and includes a labeled 2026 estimate. This comparison stops at 2025 and follows Census’s inflation method throughout, so its real-dollar figures will differ. Future updates will replace the source tables together and recheck revisions before changing the “last updated” date.
See the annual data, 1974-2025
All amounts below are annual 2025 dollars, regardless of the chart setting. Nominal amounts and source notes are in the downloads.
| Year | Federal minimum, annualized | Median worker earnings | Mean worker earnings | Median household income | Mean household income |
|---|---|---|---|---|---|
| 2025 | $15,080 | $66,620 | $92,220 | $87,460 | $126,700 |
| 2024 | $15,452 | $65,250 | $90,570 | $85,210 | $123,600 |
| 2023 | $15,852 | $64,580 | $87,560 | $84,730 | $120,400 |
| 2022 | $16,472 | $65,620 | $87,740 | $81,460 | $116,200 |
| 2021 | $17,741 | $66,440 | $90,060 | $83,270 | $120,400 |
| 2020 | $18,534 | $69,160 | $92,490 | $83,590 | $119,800 |
| 2019 | $18,727 | $64,580 | $88,960 | $85,320 | $121,800 |
| 2018 | $19,004 | $63,830 | $85,210 | $79,620 | $113,400 |
| 2017 | $19,387 | $63,970 | $84,060 | $78,600 | $112,700 |
| 2016 | $19,728 | $61,720 | $83,730 | $77,230 | $108,800 |
| 2015 | $19,902 | $61,190 | $81,650 | $74,590 | $104,600 |
| 2014 | $19,888 | $59,350 | $78,180 | $70,760 | $99,880 |
| 2013 | $20,171 | $58,840 | $78,730 | $71,670 | $100,600 |
| 2012 | $20,415 | $57,940 | $78,730 | $69,070 | $96,490 |
| 2011 | $20,809 | $58,080 | $78,860 | $69,070 | $96,150 |
| 2010 | $21,455 | $59,650 | $78,410 | $70,110 | $95,880 |
| 2009 | $21,750 | $59,830 | $79,000 | $71,790 | $98,040 |
| 2008 | $19,571 | $58,940 | $77,430 | $72,260 | $98,290 |
| 2007 | $18,120 | $60,040 | $76,780 | $74,810 | $100,700 |
| 2006 | $16,361 | $58,790 | $77,910 | $73,620 | $101,700 |
| 2005 | $16,836 | $58,010 | $77,120 | $72,810 | $99,560 |
| 2004 | $17,323 | $58,570 | $76,510 | $71,700 | $97,790 |
| 2003 | $17,757 | $59,340 | $76,950 | $71,810 | $97,920 |
| 2002 | $18,127 | $59,290 | $77,010 | $71,770 | $97,900 |
| 2001 | $18,353 | $57,630 | $76,810 | $72,350 | $99,730 |
| 2000 | $18,767 | $56,510 | $75,490 | $73,560 | $100,100 |
| 1999 | $19,414 | $57,070 | $73,750 | $73,760 | $99,200 |
| 1998 | $19,816 | $56,920 | $72,350 | $71,930 | $95,930 |
| 1997 | $20,086 | $56,450 | $70,890 | $69,390 | $93,180 |
| 1996 | $18,923 | $54,200 | $69,950 | $67,980 | $90,260 |
| 1995 | $17,379 | $53,370 | $68,400 | $66,990 | $88,340 |
| 1994 | $17,790 | $53,680 | $68,600 | $64,930 | $86,800 |
| 1993 | $18,178 | $53,720 | $67,670 | $64,240 | $85,190 |
| 1992 | $18,629 | $54,460 | $64,870 | $64,560 | $81,850 |
| 1991 | $19,102 | $54,390 | $64,280 | $65,100 | $81,940 |
| 1990 | $17,700 | $53,920 | $64,680 | $67,050 | $83,760 |
| 1989 | $16,362 | $54,790 | $66,640 | $67,880 | $85,760 |
| 1988 | $17,081 | $54,500 | $65,400 | $66,740 | $83,390 |
| 1987 | $17,712 | $54,710 | $65,110 | $66,250 | $82,390 |
| 1986 | $18,311 | $55,050 | $64,690 | $65,430 | $80,830 |
| 1985 | $18,613 | $53,970 | $62,720 | $63,090 | $77,640 |
| 1984 | $19,245 | $53,210 | $61,200 | $61,910 | $75,850 |
| 1983 | $20,051 | $51,970 | $60,520 | $60,100 | $73,090 |
| 1982 | $20,892 | $51,760 | $60,500 | $60,480 | $72,890 |
| 1981 | $22,156 | $52,260 | $60,230 | $60,650 | $72,460 |
| 1980 | $22,461 | $53,390 | $61,050 | $61,690 | $73,370 |
| 1979 | $23,332 | $54,040 | $62,920 | $63,670 | $75,630 |
| 1978 | $23,341 | $54,730 | $63,350 | $63,790 | $75,080 |
| 1977 | $21,643 | $54,870 | $62,610 | $61,400 | $72,840 |
| 1976 | $23,043 | $54,670 | $62,360 | $61,100 | $71,870 |
| 1975 | $22,238 | $54,730 | $61,520 | $60,080 | $70,150 |
| 1974 | $22,944 | $55,420 | $62,680 | $61,760 | $72,220 |