The S&P 500 (SNPINDEX:^GSPC) has a well-earned reputation as a long-term compounder. In the 69 calendar years from 1957 through 2025, it gained around 10.6% annually with dividends reinvested, according to yearly return data gathered by NYU Stern professor Aswath Damodaran.
With a long-run number like that, it’d be easy to think a typical year looks close to 10%. But that year almost never happens.
Missed AI’s “Act 1”? Act 2 Could Be 14x Bigger. Most investors think they missed the AI boat because they didn’t buy Nvidia in 2005. But according to our analysts, we’re only at the end of “Act 1″—the R&D phase. “Act 2” is the global rollout. Continue »
From the same data, I counted how many years ended with a total return between 8% and 12%. Just four did: 1968, 1993, 2004, and 2016. Meanwhile, 15 years lost money, and 25 finished up over 20%.
Image source: Getty Images.
A lopsided spread
Sort all 69 years by their returns, and more of them sit above the average than below it. Thirty-nine years finished over 12%. Twenty-six ended under 8%, including the 15 losing years. That leaves just the four in the middle.
The extremes are wide, too. The best year in the period was 1958, when the index returned around 44% with dividends. The worst was 2008, when it fell about 37% in the financial crisis.
Notably, the index climbed over 20% in five straight years from 1995 through 1999, then dropped in each of the next three.
Even widening the range to 5% to 15% only catches 14 of the 69 years, or around one in five.
The recent record looks the same. After 2016’s return of around 11.8%, the index climbed about 18% or more in seven of the next nine years. It fell in the other two, 2018 and 2022.
Losing years do the damage
Why doesn’t the average year look like the typical year? The median year (the one right in the middle when all 69 are sorted) returned around 15% — far above the 10.6% long-run number.
But losing years can be big ones, and a big loss takes an even bigger gain to recover. The 37% fall in 2008, for example, needed a gain of almost 58% just to break even. Losses like that pull the long-run compound return down to around 10.6%.
In other words, 10% has arguably never been a forecast for any one year. It’s what lots of good years and a few bad ones have added up to.
Long stretches pull returns toward 10%
Stretch the time frame out, though, and the average starts to show up. I measured the yearly return, with dividends reinvested, for every rolling five-year, 10-year, 20-year, and 30-year period in Damodaran’s data.
