What Is the Average Stock Market Return? 10% Since 1928, but Only 4 of 98 Years Came Close

Picture a hypothetical 30-year-old saver, call her Saver A, filling in a retirement calculator for the first time. The box says "expected annual return." She has heard one number her whole adult life, so she types 10 and watches a small monthly contribution turn into a seven-figure balance. It feels less like a forecast and more like a promise.

The number is not wrong. The average stock market return for the S&P 500, dividends included, was 10.02% a year compounded from 1928 through 2025. What the calculator leaves out is how strange that average is: in 98 calendar years, the index finished within two points of 10% exactly four times. The record below shows what the 10% figure really measures, how wide the range around it has been, and where the most recent decade sits in that range.

A 1927 stock certificate from Allentown, Pennsylvania, issued one year before the data series in this article begins (illustration only). Unknown author, Wikimedia Commons, public domain

Key takeaways

• From 1928 through 2025 the S&P 500 with dividends compounded at 10.02% a year; the simple average of the 98 annual returns was higher, at 11.86%.
• Only 4 of those 98 years (1968, 1993, 2004 and 2016) landed between 8% and 12%. The index fell in 26 years and gained more than 20% in 36.
• Across 89 rolling 10-year periods, annualized returns ranged from -1.67% to 20.11%. The latest, 2016-2025, returned 14.68%, higher than 64 of the other 88.
• After inflation of about 3.04% a year, the 98-year compound return works out to roughly 6.8%.

Figures checked: November 13, 2026.

Belief: "The stock market returns 10% a year"

The fact is that there are two averages, and they answer different questions. The data here come from the historical returns table that Aswath Damodaran of New York University's Stern School of Business maintains, last updated January 5, 2026. It lists the calendar-year total return of the S&P 500, price change plus dividends, in U.S. dollars, for every year from 1928 through 2025.

Add the 98 annual returns and divide by 98, and the result is 11.86%. That is the arithmetic average: the best single guess for what one randomly chosen year looked like. Chain the 98 returns together instead, as an investor who stayed put would have experienced them, and the growth rate is 10.02% a year. That is the geometric, or compound, average. In the same table, $100 placed in the index at the start of 1928 grows to $1,157,598.95 by the end of 2025, which is what 10.02% compounding for 98 years produces.

The gap between the two exists because losses and gains are not symmetric. A portfolio that rises 50% and then falls 50% has an arithmetic average return of zero, yet it ends 25% below where it started. The more returns swing, the further the compound rate falls below the simple average. So "10%" is defensible only as a compound figure. Quoting 11.86% as what an investor earned over time overstates the record by almost two points a year.

Belief: "So a normal year looks something like 10%"

The fact is that a 10% year is one of the rarest outcomes in the table. Sorting the 98 annual returns into bands gives the following picture.

Calendar-year total return, 1928-2025	Number of years	Share of 98 years
Loss (below 0%)	26	26.5%
0% to under 10%	15	15.3%
10% to 20%	21	21.4%
Above 20%	36	36.7%
Memo: between 8% and 12%	4	4.1%

The single most common result was a gain above 20%, which happened in 36 years. The second most common was a loss, in 26 years, and 12 of those losses were worse than 10%. The best year was 1954, up 52.56%. The worst was 1931, down 43.84%; 2008 lost 36.55%. The median year, with half of all years above it and half below, returned 15.21%, well above either average, because a handful of deep losses drag the averages down.

Aswath Damodaran, the NYU Stern professor who maintains this table, opens his 2023 data update with the same reminder: "It is the nature of stocks that you have good years and bad ones." The 10% average is what is left after those years are blended together. It does not describe any single one of them.

The years that actually resembled the famous number were 1968, 1993, 2004 and 2016. An investor who expects 10% and gets it should be mildly surprised. The long-run figure is built almost entirely from years that looked nothing like it.

Belief: "Ten years is long enough to earn the average"

The fact is that a decade narrows the range but does not close it. The 98 annual returns form 89 overlapping 10-year periods, from 1928-1937 through 2016-2025. Their annualized returns ran from -1.67% for 1929-1938 to 20.11% for 1949-1958, with a median of 10.96%. Five of the 89 decades lost money before inflation: the three that began in 1928, 1929 and 1930, and the two that began in 1999 and 2000. Thirteen decades returned less than 5% a year. Twenty-two returned more than 15%.

This is where the present comes in. The most recent complete decade, 2016-2025, compounded at 14.68% a year. That ranks 65th from the bottom among the 89 periods, higher than 64 of the other 88. The last three calendar years were stronger still: 26.06% in 2023, 24.88% in 2024 and 17.78% in 2025, or 22.85% annualized. An investor whose whole experience is the past decade has lived in the upper third of the historical distribution, not in its middle.

History offers 23 earlier decades that were at least this strong and have a completed decade after them. In the 10 years that followed, annualized returns ranged from -1.4% (after 1989-1998) to 19.0% (after 1979-1988). The median was 9.3%, and 13 of the 23 came in below 10%. That is not a prediction of a weak decade ahead. Three of the strongest follow-up decades came right after decades that were already excellent. What the record does say is that a strong decade has not, on average, been followed by another one as strong.

Stretch the window to 20 years and the picture calms down. None of the 79 rolling 20-year periods lost money in nominal terms. The weakest, 1929-1948, still compounded at 2.37%, the strongest, 1980-1999, at 17.70%, and the median was 10.91%. The latest, 2006-2025, came in at 10.90%, almost exactly the middle. The 10% figure starts to behave like an expectation only at that horizon, and even then the spread between the worst and best 20-year outcomes is more than 15 points a year.

Belief: "Ten percent is what ends up in the account"

The fact is that 10.02% is a nominal return on an index, before inflation, fund costs and taxes. The Bureau of Labor Statistics' Consumer Price Index for All Urban Consumers, as published on the St. Louis Fed's FRED database (not seasonally adjusted), stood at 17.3 in December 1927 and 324.054 in December 2025. Prices rose about 18.7-fold, or 3.04% a year. Dividing one growth rate by the other leaves a real compound return of roughly 6.8% a year.

A worked example shows how much the choice of number matters. Take a hypothetical one-time $10,000 investment left alone for 30 years, with no fees, taxes or further contributions:

• At 11.86%, the arithmetic average: about $288,566.
• At 10.02%, the compound average: about $175,448.
• At 6.78%, the compound average after inflation: about $71,500 in starting-year purchasing power.

The first two lines differ by $113,118, and the only thing that changed was which "average" went into the box. The third line is not a loss; it is the second line expressed in dollars that buy what they bought at the start. Fund expenses and taxes on dividends and gains, which vary by account type and are not modeled here, would lower all three.

The editorial judgment here is that the 10% figure deserves less trust as a planning input than as a description of the past. If the goal is a single number for a multi-decade plan, the after-inflation compound rate is the more honest one, because retirement spending happens in future prices. The case against that view is real, though. A lower assumption can scare a new saver into contributing nothing, or into holding too little in stocks for a 30-year horizon, and 98 years of U.S. data include a depression and a world war that may not repeat. The opposite objection is just as fair: the United States was one of the most successful stock markets of the last century, so its record may flatter what any one country delivers next. Neither objection changes the arithmetic above. They change how much weight to put on it.

For Saver A and the calculator, the practical reading is narrow. Typing 10 is a statement about nominal dollars over a very long stretch. Typing 6.8 is a statement about purchasing power. Neither says anything about next year, when a loss, on this record, has been about as likely as a coin landing heads twice in a row.

FAQ

What is the average stock market return?

For the S&P 500 with dividends, the compound annual return from 1928 through 2025 was 10.02%. The simple arithmetic average of the 98 annual returns was 11.86%. The compound figure is the one that describes what a buy-and-hold investor earned.

What is the average stock market return after inflation?

About 6.8% a year. Consumer prices measured by the CPI-U rose roughly 3.04% a year between December 1927 and December 2025, and dividing the 10.02% nominal compound return by that inflation rate leaves a real return of roughly 6.8%.

How often does the stock market lose money in a year?

The S&P 500 had a negative total return in 26 of the 98 calendar years from 1928 through 2025, or 26.5% of the time. Twelve of those years were losses of more than 10%.

Has the stock market ever lost money over 10 years?

Yes. Five of the 89 rolling 10-year periods since 1928 had a negative annualized return before inflation: 1928-1937, 1929-1938, 1930-1939, 1999-2008 and 2000-2009. None of the 79 rolling 20-year periods did.

What was the average stock market return over the last 10 years?

From 2016 through 2025 the S&P 500 with dividends compounded at 14.68% a year. That is higher than 64 of the other 88 rolling 10-year periods in the data and well above the 10.96% median decade.

Related reading

• Does a CAPE Above 40 Mean a Crash Is Coming? What Shiller's Own Data Says About the Next Ten Years
• The S&P 500's Dividend Yield Just Hit an All-Time Low of 1.05%. Here's Why It's Not Quite 2000 Again
• How Much Stock Loss Can You Deduct? The $3,000 Limit Hasn't Changed Since 1978

Sources

• Aswath Damodaran, NYU Stern School of Business, Historical Returns on Stocks, Bonds and Bills (annual S&P 500 total returns 1928-2025; updated January 5, 2026)
• Aswath Damodaran, "A Shock to the System or a Return to Normalcy: A Look at 2022," NYU Stern data update, January 2023
• Federal Reserve Bank of St. Louis (FRED), Consumer Price Index for All Urban Consumers: All Items, not seasonally adjusted (source: U.S. Bureau of Labor Statistics)

All return figures are calendar-year total returns for the S&P 500 index in U.S. dollars through December 31, 2025, and the averages, bands and rolling-period statistics were calculated from that table for this article. An index cannot be bought directly, and past returns do not guarantee future results. This article is general investment education, not personalized financial advice, and is not a recommendation to buy or sell any security. Consider consulting a licensed financial professional about individual decisions.