There is a question every serious investor must eventually confront: what actually determines long-run investment outcomes? The answer, when examined through the lens of a century of data, is both humbling and liberating. It is not stock selection. It is not macroeconomic forecasting. It is not the quality of one’s fund manager.
It is time.
The Mathematics of Duration
The U.S. equity market, as proxied by the S&P 500 and its predecessor indices, has compounded at approximately 10.5% per year in nominal terms since 1926. In real, inflation-adjusted terms, the figure is closer to 7.0% annually. These numbers are well-known. What is less intuitive — and therefore more instructive — is what those rates mean across different holding periods.
Consider $10,000 invested in 1994, thirty years ago. At 10.5% per year, that capital would have grown to roughly $200,000 today. Not through any great act of skill or foresight. Simply by owning a broad index of American businesses and leaving it alone.
The compounding curve is not linear. It accelerates. The investor who holds for 30 years does not simply get three times the benefit of the 10-year holder. They get far more. The final decade of a 30-year holding period contributes more absolute dollar gains than the first two decades combined, because the base on which returns compound is so much larger.
This is the engine of wealth creation that most market commentary ignores entirely, because it operates on a timescale that is commercially inconvenient for the financial media industry.
What the Data Says About Risk
A persistent misconception in popular financial thinking is that equities become safer over time in proportion to diversification across asset classes. In reality, the strongest risk reduction mechanism available to equity investors is diversification across time.
Examining all rolling one-year periods in the S&P 500 data from 1926 through 2024, the market has posted positive returns approximately 73% of the time. That means roughly one in four years produced a loss. Uncomfortable for most investors — and sufficiently uncomfortable that many abandon equities precisely when staying is most important.
Extend the window to rolling five-year periods, and the picture changes substantially: positive outcomes occur in approximately 88% of cases, with the worst five-year stretch producing an annualized loss of around 12.5%.
At ten years, the figure reaches 95%. At fifteen years, approximately 99%.
At twenty years — one of the most important numbers in long-term finance — the historical record shows zero instances of a negative nominal return for a U.S. equity investor who held a broad index from 1926 through 2024. Zero. The worst twenty-year stretch on record produced an annualized gain of approximately 3.1% per year. The best produced nearly 18%.
The Asymmetry of Patience
What these statistics reveal is a profound asymmetry. The investor with a one-year horizon faces genuine, substantial, and approximately 27% probability of a loss in any given year. The investor with a twenty-year horizon, on the historical evidence, has never lost money in nominal terms and has rarely seen real returns below the rate of high-quality bonds.
This asymmetry is not a coincidence. It reflects the underlying economics of what equities represent: fractional ownership of productive businesses. Over short horizons, equity prices are dominated by sentiment, liquidity flows, and changes in the discount rate applied to future earnings. Over long horizons, what matters overwhelmingly is the actual growth in corporate earnings and dividends — driven by productivity, innovation, and economic expansion. These forces are far more stable and predictable than short-term psychology.
Professor Jeremy Siegel’s research at Wharton, documented extensively in Stocks for the Long Run, demonstrates that over sufficiently long holding periods, stocks have beaten bonds, cash, gold, and virtually every other investable asset class — not occasionally, but systematically, in every major national market for which long-run data exists.
The True Cost of Market Timing
The counterargument to this view is that it ignores the possibility of permanent impairment — that one might invest at a peak and face years of losses before recovering. This concern is legitimate and worth taking seriously.
The historical data, however, suggests this concern is frequently overstated. Even an investor with catastrophically bad timing — who invested every dollar at the absolute peak of the market in 1929, just before the most severe crash in U.S. history — would have recovered their losses within approximately fifteen years. And from that point forward, the power of compounding would have carried them to substantial real wealth.
The far more common and damaging error is the opposite: selling after a large decline, missing the recovery, and permanently forfeiting the returns that patient capital would have earned. Studies of investor behavior consistently find that the average mutual fund investor earns substantially less than the funds they are invested in, due to the human tendency to buy after markets rise and sell after they fall.
An Institutional Perspective
Institutions that manage capital across genuinely long time horizons — sovereign wealth funds, university endowments, pension funds with 30-year liability windows — structure their investment programs accordingly. They accept short-term volatility as the price of participation in long-run equity returns. They resist the pressure to react to quarterly performance. They measure success in decades.
This posture, difficult for most individuals to maintain due to the psychological discomfort of unrealized losses, is the primary source of competitive advantage for long-horizon capital. It is, in a meaningful sense, the only freely available edge in public markets.
The good news is that it requires no special information, no analytical skill, and no access to restricted investment vehicles. It requires only patience — and the intellectual conviction that comes from understanding the historical record.
The century of data is unambiguous: time is not merely a component of successful long-term investing. It is the entire strategy.
All return figures cited refer to the S&P 500 total return index and its predecessor indices, sourced from the Robert Shiller data library at Yale University and Dimensional Fund Advisors. Past performance is not indicative of future results.