Investors who held U.S. equities over the past century earned roughly 10 percent a year on average, a figure that includes reinvested dividends. That long-run average, however, conceals drawdowns severe enough to cut portfolios in half and test the resolve of even disciplined savers. With stock price records stretching back to the 1870s and academic reconstructions reaching as far as 1802, the data trail is long enough to show that punishing declines are not exceptions to the pattern but part of it.
How the 10 percent average conceals severe drawdowns
The familiar 10 percent figure depends on a specific definition. As described in a prospectus filing hosted on SEC EDGAR, the S&P 500 Total Return Index reflects both price movements and the reinvestment of dividend income, reported under Bloomberg ticker SPTR. Strip out dividends and the long-run number drops sharply, because dividends historically supplied a large share of total gains. That distinction matters for anyone comparing brokerage statements to headline return figures.
The tension behind the headline is straightforward: earning 10 percent a year requires staying invested through stretches that feel nothing like 10 percent a year. Bear markets during the early 1930s, the mid-1970s, the early 2000s, and 2008 each erased years of accumulated gains in months. Investors who bought near peaks often faced a decade or more before inflation-adjusted balances recovered. The math of compounding means that a 50 percent loss demands a subsequent 100 percent gain just to break even, so the path of returns can matter as much as the average.
Periods of elevated retail participation, enabled by low-cost brokerage accounts and social media–driven narratives, have raised questions about whether crowd behavior amplifies both upside and downside. In theory, more frequent trading and higher leverage could deepen future maximum drawdowns, particularly when valuations start at historically rich levels. In practice, the historical record offers only suggestive evidence. Each major crash has arrived with its own cocktail of credit expansion, policy mistakes, and speculative fervor, making it difficult to isolate the impact of participation alone. What is clear from past cycles is that volatility tends to spike just when new investors have the least experience with severe losses, reinforcing the temptation to sell at the worst possible time.
Primary data sources behind long-run U.S. equity returns
Several independent datasets anchor the claim that U.S. stocks have compounded near 10 percent over very long periods. William Schwert’s reconstruction of early markets, released as an NBER working paper, assembled stock price indexes from 1802 to 1987 and documented the quirks of thinly traded 19th-century securities. Schwert highlighted survivorship bias, sector concentration, and limited reporting as key challenges, underscoring that the earliest numbers carry wider error bands than modern index data.
An earlier foundation for historical analysis came from Frederick Macaulay, whose 1938 volume on interest rates and asset prices included a detailed set of long-run charts for U.S. stocks and bonds. Macaulay’s painstaking work with pre-1900 quotations provided one of the first systematic attempts to measure equity performance over multiple generations. Although later researchers refined his series, his approach to documenting methodology and data quality still shapes how financial historians treat early price records.
For the period beginning in the late 19th century, the Cowles Commission and Standard & Poor’s created a more continuous index of U.S. common stocks. The Federal Reserve Bank of St. Louis distributes this history through FRED as an index of stock prices starting in the 1870s, giving researchers a monthly backbone for studying long-run returns. Because this series predates the modern S&P 500, it captures a broader evolution of the market from railroads and utilities to industrial and technology leaders.
Complementing these U.S.-focused efforts, the Global Investment Returns work by Dimson, Marsh, and Staunton at Cambridge Judge Business School has compiled equity, bond, and bill returns for many countries since 1900. Their cross-country comparisons show that while the United States sits near the top of the historical equity league table, other markets have delivered lower averages, longer droughts, or even permanent losses for buy-and-hold investors. This broader perspective reinforces that the U.S. experience, though well documented, is not guaranteed to repeat indefinitely.
Implications for today’s investors
For individuals planning over decades, the main lesson from this body of research is not that 10 percent is an ironclad forecast, but that realizing anything close to historical averages has required tolerance for deep interim losses. The same datasets that make long-run compounding look attractive also reveal stretches of 10 to 20 years when real returns were flat or negative. Portfolio choices, contribution rates, and withdrawal plans all need to account for that possibility.
Practically, that means distinguishing between the total return of a diversified index with dividends reinvested and the price-only performance of a handful of popular stocks. It also means treating historical averages as scenario inputs rather than promises. The record from Schwert, Macaulay, Cowles, and later researchers shows that severe drawdowns are embedded features of equity investing, not anomalies to be engineered away. Investors who understand that history may be better prepared to stay the course when the next downturn makes the long-run average feel remote.
David M. Keller is a finance writer based in Columbus, Ohio, covering personal finance and consumer-focused economic topics. He earned his degree in journalism from Ohio University and began his career reporting on local business and economic trends for a regional media outlet. Since then, he has contributed to a variety of online publications, focusing on clear, practical coverage of topics such as cost of living, debt, and everyday financial decision-making.
