Suppose that you had invested your wealth in a broadly diversified set of stocks, starting in January 1871, with the dividends being rolled back into your portfolio, and with your portfolio being rebalanced every January to maintain diversification. If you had also paid no taxes and incurred no fees, you would have had 65,004 times your initial investment, as of this past January. By contrast, if you had performed the same experiment with long-term U.S. Treasury bonds, you would have only 41 times your initial wealth. That is the difference between an average annual inflation-adjusted return of 7.3% for stocks and 2.5% for bonds—4.8 points per year, or what Rajnish Mehra and Edward C. Prescott called the “equity premium puzzle.”
Of course, neither of these strategies is possible in real life, since one also must pay commissions and deal with the price pressures of rebalancing one’s portfolio by selling winners and buying losers. Taxes, too, would take a big cut. They would be levied on your interest income from bonds, on your realized capital gains from stocks, and – back when stock buybacks were highly disfavored—on the dividends that you received. Together, these costs would reduce your real return by perhaps one-third, leaving an equity premium for stocks of around 3.2 percentage points per year. That means a stock-market investor could make twice as much as the bond-market investor in 22 years, on average.
Before Edgar L. Smith published Common Stocks as Long-Term Investments in the 1920s, this basic fact about stock and bond investors was not widely known. Most people considered stocks highly “speculative,” because they focused more on the returns from individual stocks and on the high likelihood that any given corporation would fail to maintain its position in the marketplace over time. Betting on individual stocks was best left to gamblers, insiders with special information, or those who truly believed they had special insights into the business cycle. But most retail stock-pickers suffer from the Dunning-Kruger effect (thinking that you are smarter than you really are), which is why their losses have long powered the gains of successful professional equity traders.
This perspective is not incorrect. But while picking individual stocks may be a fool’s game, assembling a large, properly diversified portfolio of stocks is something else entirely. By spreading the risk across companies, one can essentially eliminate it most of the time. Moreover, long-term diversified stock-market investments tend to have other risk-reducing advantages that are missed by investors focused on short-term stock performance of individual companies.
The workings of the semi-regular business cycle mean that low cashflows from a company this year will probably be offset by higher cash flows three, five, or 10 years from now. Equally, changes in valuation ratios—the multiple of average expected future earnings and dividends that the market is willing to pay—will also probably be reversed in the future. Hence, it follows that a truly long-term diversified investor should ignore market fluctuations and transitory earnings blips, and simply place his trust in businesses’ long-term profitability.
The success of the diversified, long-term approach to stocks also leads us to ask whether bonds are as safe as they are assumed to be. After all, bonds are extraordinarily vulnerable to inflation, and if there are ever conditions that do substantial permanent damage to business profitability, they will probably derange government finances even more.
Another notable feature of asset markets is that this stock-bond gap has been persistent across the generations. A stock investor ending their 40-year career in 1910 would have accumulated three times more wealth (excluding taxes and transaction costs) than a bond investor, and so, too, would a stock investor ending their 40-year career in 1950, and again in 1990. Past performance is never any guarantee of future results, but it remains the case that in a typical year, business earnings are at least 4% of stock-market equity value, whereas bond investors are lucky if returns are two points above the inflation rate.
If stocks are such a good deal over the long run, why are U.S. stock-market investors not richer? One answer is that some of them are: just look at Warren Buffett’s career. But more importantly, it takes time for the law of averages to work itself out—for “average” to become truly “typical.” If you experience an episode where you lose your entire stake, you will not have time to get it back. Mathematically, your strategy may have had a high expected return. But if you are holding and rebalancing your portfolio from January to January, there could come a year when things go spectacularly wrong. That happened in 1931 and 2009, and it may be only a matter of time before it happens again.
J. Bradford DeLong, Professor of Economics at the University of California, Berkeley, is a research associate at the National Bureau of Economic Research and the author of Slouching Towards Utopia: An Economic History of the Twentieth Century (Basic Books, 2022). He was Deputy Assistant U.S. Treasury Secretary during the Clinton Administration, where he was heavily involved in budget and trade negotiations. His role in designing the bailout of Mexico during the 1994 peso crisis placed him at the forefront of Latin America’s transformation into a region of open economies, and cemented his stature as a leading voice in economic-policy debates.