Loading...
Loading...
The S&P 500 has averaged roughly 10% a year, yet rarely delivers 10% in any single year. See what a century of data really shows.
The S&P 500 is an index that tracks 500 of the largest publicly traded companies in the United States. It is weighted by company size, so larger companies have more influence on its movement. Because it covers a large share of the US stock market's total value, it is the most common benchmark for how "the market" is doing.
You can't invest in the index directly, but you can buy index funds and ETFs that aim to match it. The numbers in this lesson are the index's total returns, meaning price changes plus reinvested dividends, so they show what a fund tracking the index would have earned before its own costs. As we covered in our lesson on dividends and total return, leaving dividends out would understate the results.
The figure you'll hear most often is that the S&P 500 has returned about 10% a year over the long run. The exact number depends on the start date, the end date, and whether it's an average or a compound rate. Here is the same history described three ways, using calendar-year total returns from 1926 through 2025.
| Measure | Approximate Value | What It Tells You |
|---|---|---|
| Simple average of yearly returns | 12.4% | The plain average of 100 separate years; overstates what an investor actually earned |
| CAGR (compound annual growth rate) | 10.5% | The steady yearly rate that would produce the actual growth |
| Real (inflation-adjusted) return | Roughly 6.5% to 7% | Growth in purchasing power after inflation, based on published estimates since 1957 |
The gap between 12.4% and 10.5% is the same effect we covered in our lesson on CAGR: big losses hurt more than equal-sized gains help, so the simple average runs higher than the compound rate. And the gap between 10.5% and about 6.5% to 7% is inflation, explained in our lesson on nominal vs real returns. When someone says "stocks return 10%," ask which of these they mean.
A quick note on the data: the index began in its current 500-company form in 1957. Earlier years use its predecessor index, which tracked fewer companies. The figures here are nominal, before taxes and fees.
Say Sara invested $10,000 in an S&P 500 index fund at the start of 2006 and left it alone through the end of 2025, with dividends reinvested. Her balance would have grown to about $80,600, a CAGR of roughly 11.0%. That's a strong result, and it includes the 37% loss in 2008 and the 18% loss in 2022.
These are illustrations of past data, not predictions or guarantees, and they ignore taxes, fees, and inflation. You can test your own timeline with our CAGR Calculator and Compound Interest Calculator.
An average of 10% a year does not mean 10% every year. In fact, a calendar year with a return close to 10% is fairly ordinary rather than typical. Here is how the 100 calendar years from 1926 to 2025 were distributed.
| Yearly Total Return | Number of Years (out of 100) |
|---|---|
| Worse than −20% | 6 |
| −20% to −10% | 6 |
| −10% to 0% | 14 |
| 0% to 10% | 14 |
| 10% to 20% | 22 |
| 20% to 30% | 18 |
| 30% or better | 20 |
In this record, the index finished higher in 74 of 100 years and lower in 26. The best year was 1933 at about +54%, and the worst was 1931 at about −43%. Positive years are common, but a loss in roughly one year out of four is part of the deal.
| Year | S&P 500 Total Return |
|---|---|
| 2008 | −37.0% |
| 2009 | +26.5% |
| 2020 | +18.4% |
| 2022 | −18.1% |
| 2023 | +26.3% |
| 2024 | +25.0% |
| 2025 | +17.9% |
Notice how far these are from a steady 10%. A −37% year was followed by +26.5%, and an −18% year by two years above 25%. Anyone who tried to guess which years would be good or bad would have had to be right on both sides of some very large moves.
The single most useful lesson from the data is that the longer you stay invested, the narrower the range of outcomes becomes. Here are the results of every rolling period in the 1926–2025 record, measured as annualized total return.
| Holding Period | Periods Measured | Periods With a Loss | Worst Result (annualized) | Best Result (annualized) |
|---|---|---|---|---|
| 1 year | 100 | 26 | −43.3% (1931) | +54.0% (1933) |
| 5 years | 96 | 12 | −12.5% a year (1928–1932) | +28.6% a year (1995–1999) |
| 10 years | 91 | 4 | −1.4% a year (1999–2008) | +20.1% a year (1949–1958) |
| 20 years | 81 | 0 | +3.1% a year (1929–1948) | +17.9% a year (1980–1999) |
Over one year, results ranged from a 43% loss to a 54% gain. Over 20 years, every period in this record ended with a positive return, and the range narrowed to roughly 3% to 18% a year. That's the practical case for a long time horizon: it doesn't remove risk, but it shrinks the spread of results. Past results are not a guarantee, and all figures are nominal, so a 3% annualized result over 20 years could still lag inflation.
| Period | Annualized Total Return | $10,000 Becomes |
|---|---|---|
| 2000–2009 (sometimes called the "lost decade") | About −1.0% | $9,088 |
| 2016–2025 | About +14.8% | $39,833 |
| 1926–2025, the full record | About +10.5% | — |
An investor who started in 2000 lived through two major downturns and earned close to nothing for ten years, even with dividends reinvested. An investor who started in 2016 earned far above the long-run average. Neither decade looks like the 10% average, which is why it's risky to assume a recent decade's result will repeat, in either direction.
| Limitation | Why It Matters |
|---|---|
| Past results are not a forecast | The future can be better or worse than any historical period |
| Returns are nominal | Inflation reduces what those returns can buy |
| The index has no costs | Real funds charge fees and investors pay taxes, as shown in our lesson on fees and taxes |
| It is one country's market | The US has had an unusually strong history, and other markets have had different results |
| It assumes you stayed invested | Selling during a downturn and buying back later can produce very different results |
Many planners choose an assumption below the historical average, both to leave room for a weaker future and to reflect fees, taxes, and inflation. For example, a long-term plan might use a nominal return in the 6% to 8% range rather than the full 10%, or express the assumption as a real return of around 3% to 5%. There's no single right number, and the choice should reflect your own goals and comfort with risk.
A practical way to use the data is to run several scenarios: a weak one, a middle one, and a strong one. If your goal only works at the strong assumption, it may deserve a second look. You can try different rates in our 401(k) Projector and see how inflation changes the picture with our Inflation Calculator.
1. Assuming the market returns 10% every year. The 10% is a long-run figure. Only a small share of individual years land near it, and about one in four calendar years has been negative.
2. Confusing the average return with the compound return. The simple average of yearly returns was about 12.4% in this record, but the compound rate was about 10.5%. Plan with the compound rate.
3. Ignoring inflation, fees, and taxes. The 10% headline is nominal and before costs. Your actual real, after-fee, after-tax result will be lower.
4. Extrapolating a recent hot streak. A decade like 2016–2025 at about 14.8% a year is well above the long-run average, and a decade like 2000–2009 was slightly negative. Neither is a reliable guide to the next ten years.
5. Judging the market over too short a window. One-year results ranged from about −43% to +54%. Twenty-year results were far more tightly clustered, so measure over horizons that match your goals.
Key Takeaway: The S&P 500 has compounded at roughly 10% a year over the long run, or about 6.5% to 7% after inflation, but the path has been bumpy, with a loss in about one year out of four, and the record is a guide to the range of outcomes rather than a promise of future returns.
Over the long run, it has been about 10% a year in nominal terms with dividends reinvested, and roughly 6.5% to 7% after inflation. The exact figure depends on the period measured and on whether an average or a compound rate is used.
Usually because one figure is adjusted for inflation and the other isn't. The roughly 10% figure is nominal, while a figure around 6.5% to 7% is the inflation-adjusted, or real, return.
Yes. The figures in this lesson are total returns, which combine price changes with reinvested dividends. The headline index level you see in the news usually shows price return only, so it looks lower.
In the 100 calendar years from 1926 to 2025, it finished lower in 26 of them, or roughly one year in four. Losses have varied from small dips to declines of more than 40%.
In this record, 4 of the 91 rolling 10-year periods had a negative annualized return, including 1999–2008. None of the 81 rolling 20-year periods did, though the worst still earned only about 3% a year. History doesn't guarantee future results.
Not the index itself, but you can buy index funds or ETFs that track it through most brokerage accounts and retirement plans. Compare their expense ratios, since fees reduce the return you receive.
Disclaimer: This article is for general educational purposes only and does not constitute personalized financial, investment, tax, or legal advice. Figures, rates, and rules mentioned may change over time — verify current details with an official source or a qualified professional before making financial decisions.