Compound Interest.

Historical Stock Market Returns Calculator

The S&P 500 returned 10.02% a year from 1928 to 2025 with dividends reinvested — but only 6.78% a year after inflation. Pick any date range below to see both numbers for the period you care about.

Historical S&P 500 Return Calculator

Pick any date range from 1928 to 2025 to see what the S&P 500 actually returned — the headline nominal number and the real number after inflation. Dividends are reinvested.

$

A lump sum invested at the start of 1928.

Nominal Return

10.02%

per year, 98 years

Real Return (After Inflation)

6.78%

inflation averaged 3.04%

Worst Year in Range

-43.8%

in 1931 · 26 down years of 98

$10,000 Grew To

$115,701,364

nominal, end of 2025

Worth in 1928 Dollars

$6,176,825

actual purchasing power

Lost to Inflation

$109,524,539

95% of the nominal balance

Growth of $10,000, 19282025

19282025
Real value (1928 dollars)Eroded by inflationLogarithmic scale — equal heights mean equal percentage gains.

From 1928 through 2025 the S&P 500 returned 10.02% a year with dividends reinvested, but inflation ran 3.04% — so the return that actually bought you anything was 6.78% a year. The ride included 26 down years, the worst being 1931 at -43.8% and the best 1954 at +52.6%.

Average S&P 500 Return by Decade

The long-run average hides how little any individual decade resembles it — not one of the decades below came within a full percentage point of the 10.02% nominal average. Each figure is annualized (compound), with dividends reinvested.

DecadeNominalInflationReal
1930s-0.92%-2.04%1.14%
1940s8.50%5.36%2.98%
1950s19.46%2.22%16.86%
1960s7.74%2.52%5.09%
1970s5.92%7.36%-1.35%
1980s17.34%5.10%11.65%
1990s18.05%2.93%14.69%
2000s-0.95%2.52%-3.39%
2010s13.44%1.75%11.48%
2020s (20202025)14.92%3.94%10.56%

The 2020s row covers 20202025 and is still in progress.

Two rows deserve attention, because they show that a nominal return can mislead in both directions:

  • The 1930s lost 0.92% a year nominally and still gained 1.14% a year in real terms. Prices fell 2.04% a year during the Depression, so each surviving dollar bought more. The decade looks like a catastrophe on a price chart and like a modest gain in a grocery store.
  • The 1970s gained 5.92% a year nominally and still lost 1.35% a year in real terms. Stagflation ran inflation at 7.36%. An investor who watched their statement rise every year was quietly getting poorer.

The 2000s were the only decade to lose money on both measures — -0.95% nominal and -3.39% real, bracketed by the dot-com bust and the financial crisis. If you want to see what that does to a balance, set the calculator above to the Lost decade preset.

Best vs. Worst 10-Year Periods

There have been 89 overlapping 10-year windows since 1928. Here are the extremes, annualized:

PeriodYearsNominalReal
Best nominal1949195820.11%17.95%
Worst nominal19291938-1.67%0.32%
Best real1949195820.11%17.95%
Worst real19992008-1.36%-3.78%

Notice that the worst nominal decade and the worst real decade are not the same period. 19291938 spans the Great Depression and lost -1.67% a year on paper, yet deflation left it slightly positive in purchasing power at 0.32%. The genuinely worst decade to have lived through was 19992008, which compounded at -3.78% a year in real terms — a decade that turned $10,000 into about $6,800 of purchasing power.

How often does a decade disappoint? 5 of the 89 windows finished negative in nominal terms, and 11 — roughly one in 8— finished negative after inflation. Ten years is a real horizon for real people, and it does not guarantee a gain.

Stretch the window to 30 years and the picture changes. None of the 69 overlapping 30-year windows since 1928 produced a negative real return. The weakest was 19651994 at 4.29% a year real; the strongest, 19321961, managed 10.10%. Even the worst 30-year stretch in this dataset roughly 3.5x'd an investor's purchasing power. Time, not timing, is what closes the gap.

How to Use Historical Returns in Compound Interest Planning

A projection is only as good as the rate you feed it. Four rules turn the history above into an assumption you can defend.

  • Never mix nominal returns with today's dollars. This is the single most common planning error. If you project 10% growth, you must also inflate your future costs — a $60,000 lifestyle will not cost $60,000 in thirty years. Either grow at 10% and inflate expenses, or grow at the 7% real rate and leave expenses in today's dollars. Both are correct; the hybrid is not.
  • Use the geometric average, not the arithmetic one. The compound rate that actually links a starting balance to an ending balance is 10.02%. Averaging the yearly percentages gives a higher number that no investor ever earned, because losses hurt more than equal-sized gains help.
  • Check that the return includes dividends. The 10.02% above is a total return with dividends reinvested. An index price chart excludes them and runs well below it.
  • Stress-test against the bad decades, not the average. Before you commit to a plan, rerun it at the 4.29% real rate of 19651994. If the plan only works at 7%, it is not a plan — it is a bet on an average showing up on schedule.

In practice: take the real return from the calculator above for whatever period you consider representative, drop it into the investment growth calculator as your annual return, and keep your contributions and goal in today's dollars. If you would rather work in nominal terms, use the inflation-adjusted returns calculator to translate the ending balance back into purchasing power.

Two things matter more than the rate you pick. The first is when you start: the gap between investing at 25 and at 35 dwarfs a percentage point of assumed return, as the 25 vs. 35 comparison shows. The second is what you compare against: the 6.8% real return above is the reward for accepting 26 down years out of 98, and cash gives up most of it for the privilege of never falling — see HYSA vs. CD vs. index fund.

Frequently Asked Questions

What is the average annual return of the S&P 500?

From 1928 through 2025, the S&P 500 returned an annualized 10.02% per year with dividends reinvested. That is the geometric (compound) average — the rate that actually turns a starting balance into an ending balance. The simple arithmetic average of the yearly returns is higher, but no investor ever earns it, because a 50% loss followed by a 50% gain leaves you down 25%, not flat.

What is the S&P 500's average return after inflation?

About 6.78% per year over 1928–2025. Inflation averaged 3.04% annually across the same period, which is what turns the headline 10.0% nominal return into a real return near 7%. Real return is what your money can actually buy, and it is the number to use for any long-term goal denominated in today's dollars.

What return should I assume for retirement planning?

Most planners use 10% nominal or roughly 7% real. The critical rule is not to mix them: if you project 10% growth, you must also inflate your future expenses, because a $60,000 lifestyle will not cost $60,000 in 30 years. If you project 7% growth instead, you can leave expenses in today's dollars. Both approaches give the same answer; using 10% growth against un-inflated expenses badly overstates what you will have.

What was the worst 10-year period for the S&P 500?

In nominal terms, 1929–1938 (the Great Depression) was worst at -1.67% per year. But the worst decade for purchasing power was 1999–2008, at -3.78% per year after inflation. The Depression window was rescued in real terms by deflation — prices fell, so each surviving dollar bought more. Adjusting for inflation reorders the history.

Has the S&P 500 ever lost money over 10 years?

Yes. Of the 89 overlapping 10-year windows since 1928, 5 ended with a negative nominal return and 11 — about 12% of them — ended with a negative real return. Ten years is not long enough to guarantee a gain. Over 30-year windows, no period in this dataset has ever produced a negative real return; the weakest was 1965–1994 at 4.29% per year.

Does the 10% average return include dividends?

Yes. The 10.02% figure is a total return: price appreciation plus dividends reinvested. Price-only returns run roughly 1.5 to 2 percentage points lower, because dividends have supplied a large share of the market's long-run gain. If you compare a calculator's assumed return against an index chart that only tracks price, you will understate your result.

Why isn't the real return just 10% minus 3% inflation?

Subtraction is an approximation that drifts as rates rise. The exact relationship is the Fisher equation: real = (1 + nominal) ÷ (1 + inflation) − 1. With a 10.02% nominal return and 3.04% inflation, that gives 6.78%, not the 6.98% you get by subtracting. The gap is small at low inflation and grows meaningfully at 1970s-style rates.

Can I expect 10% a year going forward?

Treat it as a long-run central tendency, not a forecast. Individual decades have ranged from -0.95% a year in the 2000s to 19.46% a year at the best, and 26 of the last 98 calendar years were down. Starting valuations matter, and no decade has repeated the average. Plan with the historical real return, save more than the model demands, and treat anything above that as a margin of safety.

Data sources. S&P 500 annual total returns (with dividends reinvested) from Aswath Damodaran, NYU Stern. Inflation is the December-to-December change in CPI-U (FRED: CPIAUCNS). Figures cover 19282025 and are annualized (geometric) unless stated otherwise. Past returns do not predict future returns.