Siegel's Long-Run Evidence
Photo by Maxim Hopman on Unsplash
Jeremy Siegel's 1994 book Stocks for the Long Run assembled two centuries of US financial data to make one of the most important empirical claims in personal finance: over every 20-year holding period in US history since 1802, equities have outperformed bonds, cash, and gold in real (inflation-adjusted) terms. Not in most periods — in every period. The long-run real return on US equities has been approximately 6.5–7% per year, a figure that has remained remarkably stable across wildly different economic eras, including wars, depressions, technological revolutions, and monetary regime changes.
This stability is striking. It suggests that equity returns are not merely the product of a particular era's economic conditions but reflect something more fundamental — the long-run return to bearing the risk of productive enterprise.
Global Confirmation
Photo by Andrew Stutesman on Unsplash
Dimson, Marsh, and Staunton's 2002 book Triumph of the Optimists extended Siegel's US analysis to 16 countries over the full 20th century. The conclusion held globally: equities outperformed bonds in every developed country studied over the century. The global real equity return averaged around 5% per year; the real bond return averaged around 1%.
This confirmation matters because it rules out the possibility that the US result is an artefact of one particularly successful economy. Across countries with very different histories, equity ownership rewarded patient investors.
The Survivorship Bias Caveat
Photo by Andrew Ridley on Unsplash
The most important qualification to the long-run equity story is survivorship bias. The Dimson, Marsh, and Staunton dataset covers countries that survived the 20th century as functioning capital markets. But Russia's stock market was wiped out by revolution in 1917. China's market was closed after 1949. Germany and Japan both saw complete equity wipeouts after World War II. An investor in 1900 with no foreknowledge of which countries would succeed might have allocated across all major markets — and the returns from the portfolio as a whole would look considerably worse than the returns from the surviving winners.
The honest assessment: equities have the best long-run expected return of any major asset class, but this expectation is a probability distribution, not a guarantee. The distribution has a tail in which a country's equity market delivers permanent capital impairment.
Japan: A Warning
Photo by Kumiko SHIMIZU on Unsplash
Japan's Nikkei 225 index peaked at 38,957 in December 1989, the result of one of history's most dramatic asset bubbles. Thirty years later, the index remained more than 40% below that peak in nominal terms and further still in real terms. An investor who had retired in 1989 counting on Japanese equity returns never saw their portfolio recover. The Japan experience is not an anomaly to be explained away — it is a reminder that no single country's equity market is a guaranteed long-run compounder, and that global diversification matters.