Hendrik Bessembinder of Arizona State University's W.P. Carey School of Business opens his landmark study, "Do Stocks Outperform Treasury Bills?"1, with a disarming admission: "The question posed in the title of this paper may seem nonsensical." Everyone knows the answer, after all. The equity premium is so well documented that its very size is called a puzzle. But Bessembinder's answer, drawn from every CRSP common stock from 1926 to 2016, is that most individual stocks do not outperform. In fact, most do not even survive the comparison.
The Median Stock Is a Disappointment
The headline findings are stark. Just 42.6% of common stocks generate lifetime buy-and-hold returns, dividends reinvested, that beat one-month Treasury bills over matched horizons. More than half deliver negative lifetime returns outright. And the single most frequent lifetime outcome? Bessembinder writes that the modal result "is a loss of 100%." Roughly 12% of all stocks suffered essentially complete losses.
The market as a whole prospers anyway because a small cohort does the heavy lifting. As Bessembinder puts it, "the positive performance of the overall market is attributable to large returns generated by relatively few stocks." The typical stock is not the story. The exceptional stock is.
Skewness, Not Contradiction
Is this at odds with risk-averse investors demanding a premium? Not necessarily. Asset pricing models speak to mean returns; this evidence speaks to medians. Bessembinder notes the results "are not necessarily at odds with the implications of standard asset pricing models." The reconciliation is positive skewness. A few enormous winners pull the mean far above the median.
Critically, that skewness is manufactured in part by time itself. Even if monthly returns were perfectly symmetric, "the compounding of random returns induces positive skewness in the multi-period return distribution," and more so for volatile stocks. His simulations show skewness of ten-year returns exploding as monthly volatility rises. With individual stock volatility running at 18.1% monthly, long-horizon lottery-ticket dynamics are not an anomaly. They are arithmetic.
The Wealth Creation Ledger
The concentration findings are the paper's most quoted, and deservedly so. Measured against a Treasury bill benchmark, the roughly 25,300 CRSP firms created $34.82 trillion in shareholder wealth through 2016. Five firms, Exxon Mobil, Apple, Microsoft, General Electric, and IBM, account for 10% of it. Ninety firms, about one-third of 1%, account for half. And 1,092 firms, just over 4%, account for all of it. The remaining 96%, Bessembinder writes, "collectively generate lifetime dollar gains that matched gains on one-month Treasury bills."
His bootstrap simulations drive the point home. A strategy of holding one random stock per month over the full 90 years underperformed the value-weighted market in 96% of trials and underperformed Treasury bills in 73%. Diversification repairs this quickly: with 100-stock portfolios, 93.1% of decade returns beat T-bills. Yet even diversified random portfolios beat the value-weighted market less than half the time, which, as he observes, helps "to explain why active portfolio strategies most often underperform benchmarks" even before fees.
Where the Damage Concentrates
Underperformance is not evenly distributed. It clusters in small caps, in exchange-delisted names (only 6.8% of which beat T-bills), and in stocks listed after the mid-1960s. The median lifetime return is negative for every listing cohort since 1977, consistent with Fama and French's evidence on riskier, less profitable new listings. Public equity, particularly at the small and recent end, increasingly resembles venture capital: many losers funding a handful of spectacular wins.
To be fair, Bessembinder does not condemn concentration outright. The results "highlight the potentially large gains from active stock selection" for anyone with genuine skill at identifying the home runs in advance. Whether such skill can be reliably identified, he concedes, "remains" the key question.
Five Key Takeaways for Advisors and Investors
1. Diversification is not just variance reduction; it is winner capture. Concentrated portfolios risk omitting the 4% of stocks that generate all net wealth creation.
2. Expect index-relative underperformance from concentrated active strategies, even absent fees or behavioral error. Skewness alone stacks the odds.
3. The median stock loses to cash. Client conversations about single-stock positions, including concentrated employer stock, should start here.
4. Time amplifies, not heals, single-stock risk. Compounding increases skewness with horizon, so "buy and hold forever" logic does not rescue undiversified bets.
5. Skewness preference is legitimate but must be priced. Investors seeking lottery-like upside should understand the most likely outcome is underperformance.
When the facts are this asymmetric, portfolio discipline is not optional. It is the whole game.
Footnote:
1 Bessembinder, Hendrik. "Do Stocks Outperform Treasury Bills?" Journal of Financial Economics, forthcoming, May 2018. SSRN, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2900447.
