In 2018, Hendrik Bessembinder published a study in the Journal of Financial Economics with a title that sounds almost naive: "Do Stocks Outperform Treasury Bills?" The answer, for most individual stocks, turned out to be no.
He took every US common stock in the CRSP database between July 1926 and December 2016, 25,967 of them, and measured the buy-and-hold return over each stock’s entire life as a listed company. Only 42.6% of them beat the return on one-month Treasury bills over the same period. Slightly more than four out of every seven did worse than cash.
The second finding is the one that reframes the first. The best-performing 4% of listed companies account for the entire net wealth the US stock market created above Treasury bills since 1926. The other 96%, taken together, matched T-bills. And 86 stocks alone produced about half of the total.
This is what a positively skewed distribution looks like in practice. The median stock is unremarkable or worse, the average is dragged upward by a handful of extreme winners, and the market index outperforms because it holds those winners by construction.
The practical consequence is uncomfortable for concentrated portfolios. If wealth creation lives in a small tail, a portfolio of ten names is a bet that you own part of that tail. Bessembinder makes the link explicit: the result helps explain why poorly diversified active strategies underperform market averages more often than not.
For a classroom, the study is useful precisely because it is counterintuitive. Students arrive believing that stocks beat cash, which is true of the index and false of the typical stock. Holding both facts at once is the beginning of thinking in distributions instead of averages.
Educational material. Not investment advice.