The US stock market has generated an average annual return of 10% over the past 100 years. That number is widely cited, deeply reassuring, and almost entirely misleading for anyone trying to pick individual stocks. Behind the aggregate sits a more unsettling truth: less than half of roughly 30,000 US stocks studied between 1926 and 2025 posted positive returns, and just 41.17% outperformed the US Treasury bill.
This is the central provocation of Morningstar Indexes strategist Dan Lefkovitz's article, "Why Most Stocks Aren't Worth Owning," which frames the research of Hendrik Bessembinder, professor at Arizona State University, as essential reading for anyone who touches a portfolio. The verdict, in Lefkovitz's own words, is blunt: "Most stocks stink."
The Wealth Concentration Problem
The headline figure from Bessembinder's updated research is staggering in its specificity. Just 46 stocks accounted for half the $91 trillion of wealth created across the entire US market from 1926 to 2025. That concentration of wealth creation had increased since his original study. In 2017, he found that 89 stocks accounted for half the wealth created.
The direction of travel matters as much as the number itself. Wealth creation is not merely concentrated; it is becoming more concentrated. The superstar effect is intensifying over time, not diffusing. The list of the biggest wealth creators spans across generations and sectors. It includes young companies like Tesla, Alphabet, and Nvidia, as well as stalwarts such as Coca-Cola, General Motors, and Johnson and Johnson. What they share is not sector affiliation or founding era. The question of shared attributes is precisely what Lefkovitz intends to press Bessembinder on at this year's Morningstar Investment Conference.
The CRSP Origin Story and Why It Matters Now
Lefkovitz grounds the research in its institutional history. CRSP was created at the University of Chicago in 1960 with an initial grant from Merrill Lynch. Professors Lawrence Fisher and James Lorie were tasked with developing a database of US stock returns. The brokerage house wanted to see the case made for equity investing. In 1960, investors were just a generation removed from the Great Depression.
The irony is complete. A database built to validate equity investing ultimately produced research that demolishes the case for undiversified equity selection. CRSP's database now goes back a century and, updated through the end of 2025, shows an average annual return over the past 100 years of 10% for a market-capitalization-weighted aggregate of US-listed stocks. That figure, born from Merrill Lynch's desire to make stocks look compelling, obscures the distribution underneath it.
What This Means for Stock Picking
Bessembinder's 2017 paper, "Do Stocks Outperform Treasury Bills?", was the original shock to the system. His 2026 update, "One Hundred Years in the U.S. Stock Markets," tightens the argument with another decade of data. The compounding implication is direct: identifying tomorrow's 46 superstar stocks from a universe of thousands is not a skill most investors possess, and the evidence suggests most professional selectors do not possess it either.
The article highlights that just 41.17% of US stocks outperformed Treasury bills, making broad diversification look less boring and more necessary. Lefkovitz's framing at the Morningstar Investment Conference is not accidental. He positions Bessembinder's findings as a case for indexing, not a counsel of despair. There are ways to improve the odds of picking a stock worth owning, with Morningstar equity analysts explaining why moats and dividends are a strong combination. The structural quality of a business, measured through competitive advantage, is one of the few identifiable attributes that correlates with durable outperformance.
The Morningstar-CRSP Dimension
Lefkovitz has a secondary interest beyond research synthesis. He will be joined at the conference by colleague Alex Poukchanski, who came over to Morningstar as part of its February 2026 acquisition of the Center for Research in Security Prices, a former affiliate of the University of Chicago. That acquisition brought the CRSP Market Indexes, benchmarks for over $3 trillion in US equities, into the Morningstar Indexes family. Bessembinder's research was built on CRSP data. Morningstar now owns CRSP. The institutional alignment between the research and the index products it supports is explicit.
The Argument for Broad Market Exposure
The article does not frame Bessembinder's findings as a reason to avoid equities. It frames them as a reason to hold all of them. When wealth creation is concentrated in 46 stocks drawn from a universe of 30,000, missing those 46 is catastrophic. Holding the market ensures you capture them. Active selection, without systematic edge, is probabilistically a losing exercise. The data across a century say so.
5 Key Takeaways for Advisors and Investors
- The 10% long-run average is a portfolio-level fact, not a stock-level expectation. Most individual stocks underperform cash over their lifetimes. Clients need to understand the difference between market returns and stock returns.
- Wealth creation is ruthlessly concentrated and becoming more so. In 1926 to 2025, 46 stocks generated half of all market wealth. Missing them through active selection has compounding consequences that are almost impossible to recover from.
- Broad diversification is the structural response to a positively skewed return distribution. The market's aggregate return is earned by capturing outliers. Active stock selection reduces the probability of capturing those outliers without systematic advantage.
- Economic moats and dividend stability are among the few identifiable attributes linking superstar stocks across history. Businesses with durable competitive advantages are less likely to cut dividends and more likely to compound wealth over the long term.
- The institutional history of CRSP matters for advisors. The very database commissioned to make the case for equities has produced the most powerful evidence against undiversified equity selection. The research and Morningstar's index infrastructure are now unified, with direct implications for how benchmarks and factor-based strategies are constructed.
Footnote:
1 Lefkovitz, Dan. “Why Most Stocks Aren’t Worth Owning.” Morningstar, Inc. N.p., n.d. Web. 24 June 2026.
