100 years, 29,000 stocks, 46 winners: the case for indexing just got stronger
- Robin Powell
- 3 hours ago
- 8 min read
A landmark study covering a full century of US stock market data shows that wealth creation is concentrated in fewer companies than ever. Just 46 firms out of nearly 30,000 drove half of all shareholder gains — down from 89 in the original study. For anyone still tempted to pick stocks, the case for indexing has never been clearer.
Everyone knows the phrase 'looking for a needle in a haystack'. For stock pickers, it's always been a fitting description. The US stock market has listed tens of thousands of companies over the past century, and only a tiny fraction have driven the returns that matter.
But what if the haystack kept growing — and the number of needles kept shrinking?
That's the picture emerging from a new working paper by Hendrik Bessembinder of Arizona State University. Bessembinder is best known for his 2018 study, published in the Journal of Financial Economics, which showed that 4% of US stocks accounted for all the net wealth created by the stock market since 1926. The other 96% collectively matched the return of Treasury bills. It was one of the most cited findings in modern investment research, and a powerful early case for indexing.
He's now updated his analysis to cover a full 100 years, from January 1926 to December 2025, encompassing 29,754 common stocks. The findings, published as an initial draft on 18 March 2026, haven't been peer-reviewed yet — but the methodology is identical to the 2018 study, and the results are more arresting.
In the original, 89 firms accounted for half of $43 trillion in net shareholder wealth creation. In the update, that number has dropped to 46 — even as total wealth creation more than doubled to $91 trillion. The haystack got bigger. The needles got fewer.
Most stocks lose money — and it's getting worse
'The middle stock in the US market's 100-year history lost money.'
The typical stock is a disappointment. Across all 29,754 stocks in Bessembinder's dataset, the mean buy-and-hold return is a staggering 30,000%-plus. The median is -6.9%.
Read that again. The middle stock in the US market's 100-year history lost money.
Why such a gap? Returns aren't evenly spread. A small number of spectacular winners haul the average far above what most stocks deliver — a stock can rise by thousands of percent, but it can only fall by 100%. Statisticians call this positive skewness. For investors, it means the headline numbers are wildly misleading.
Only 48% of stocks generated a positive buy-and-hold return over their lifetimes. Fewer than 42% beat Treasury bills over the months they were listed. And 28% outperformed the broader market. Pick a stock at random from the past century and the odds were roughly three in four that it trailed the average.
Those are the century-long figures. The decade-by-decade picture is grimmer.
As the data in Table 3 shows, individual stock performance has deteriorated sharply since the mid-1980s. Across the first six decades (1926–1985), the median decade-horizon return averaged 63.6%, and roughly 61% of stocks beat T-bills. Across the four most recent decades (1986–2025), the median collapsed to 5.8%, and only 48% cleared the T-bill hurdle.
This happened while the overall market performed strongly. The value-weighted portfolio did very well from 1986 to 2025. Individual stocks, on the whole, did not.
What changed? Partly composition. As Fama and French (2004) documented, the 1970s and 1980s brought a wave of younger and smaller companies to US exchanges. Many were speculative and short-lived, dragging median outcomes down even as the biggest winners pulled further ahead.
The market has been brilliant. Most of the stocks in it haven't.
The winners' circle is shrinking fast
The concentration of stock market wealth in a few firms isn't new. What's new is how rapidly it's intensified — and how dramatically that strengthens the case for indexing.
Bessembinder measures this through shareholder wealth creation (SWC) — how much better off a company's shareholders ended up compared to what they'd have earned in Treasury bills, measured in dollars and accounting for dividends, buybacks and share issuances.
In his 2018 study, covering data through 2016, 89 firms accounted for half of $42.61 trillion in net SWC.
The updated numbers are something else. Total SWC has more than doubled to $90.96 trillion, but the number of firms driving half of it has nearly halved. 46 companies now account for 50% of a century's worth of net wealth creation. Apple and Nvidia alone account for 10%. Eight firms account for a quarter.
'46 companies now account for 50% of a century's worth of net wealth creation. Apple and Nvidia alone account for 10%.'
And 1,082 firms — 3.72% of every company that ever listed — generated all of the net gains. The other 96%? Collectively, they matched T-bill returns. A century of listing, and nothing to show for it beyond what you'd have got from the safest investment in the world.
The acceleration since 2017 has been stark. The top 30 wealth creators over the past nine years generated 61% of post-2016 SWC. Before 2017, the top 30 generated 31%. Nvidia alone created $4.51 trillion in shareholder wealth after 2016 — roughly a tenth of all post-2016 gains from a single chipmaker.
The haystack contains nearly 30,000 stocks. The needles? 46. That's 0.16%.
Why stock pickers can't beat these odds
'Talent isn't the issue. The odds are.'
When so few firms drive so much of the market's gains, missing even one can be devastating. And the fewer stocks you hold, the more likely you are to miss one.
If 3.72% of firms generated all net wealth creation, a stock picker building a portfolio of 30 or 50 holdings is fishing in waters where 96% of the catch delivered nothing better than T-bill returns. You don't need to pick good companies. You need to pick the right ones — the tiny handful that will produce outsized returns. And you need to hold enough of them.
Bessembinder drew this conclusion himself in 2018, noting that the results 'help to explain why active strategies, which tend to be poorly diversified, most often underperform'.
Terry Smith's Fundsmith Equity fund is a good example of the structural trap. Smith is one of the UK's most respected managers, with a disciplined focus on high-quality businesses. Yet he declined to invest in Nvidia — the single largest wealth creator since 2017. That's not a knock on Smith's ability. It's the kind of miss that even a seasoned professional with a clear philosophy will make when one stock can drive a tenth of all recent gains.
The majority of actively managed funds underperform their benchmarks over ten years or more, according to the SPIVA scorecards. Not because managers lack talent. Because the distribution of returns is so lopsided that concentrated portfolios are more likely to miss the winners than find them.
Talent isn't the issue. The odds are.
The quiet power of compounding — and patience
'Two extra percentage points a year, compounded over a century, created an eightfold gap.'
The biggest winners in stock market history didn't get there by being spectacular. They got there by being steady.
Bessembinder's data on the top 30 stocks by cumulative return is counterintuitive. The median annualised return among them was 13%. Not 30%. Not 50%. Thirteen. Good returns, comfortably above the long-run average, but nothing that would make headlines in any given year.
What made them extraordinary was duration. These stocks were listed for an average of 93.9 years. Compounding at a moderately high rate for that long produces results that look fictional. Altria Group (formerly Philip Morris) returned 16.53% annualised across the full century, turning $1 into $4.42 million.
Small annual differences, given enough time, become vast. Altria's annualised return was only 1.18 times Vulcan Materials' 14.35%. But Altria's cumulative outcome was 8.81 times as large. Two extra percentage points a year, compounded over a century, created an eightfold gap.
There's a striking corollary for anyone chasing hot stocks: not one of the 30 highest-annualised-return stocks (among those with 20+ years of data) also appears on the list of the 30 highest cumulative returns. The stocks that burned brightest didn't sustain it. The ones that built real wealth kept going, year after year, at rates that looked unremarkable at any given moment.
Time in the market beats timing. It always has.
Buy the whole haystack: the case for indexing in practice
Vanguard founder John Bogle had a characteristically blunt answer to the stock-picking problem: don't bother looking for the needle in the haystack — buy the whole haystack.
A century of data shows why. An index fund owns everything — Apple, Nvidia, and whatever unknown company will drive the next decade's returns. It also owns the 59% of stocks that destroy wealth. But the winners more than compensate. Over 100 years, the net result is $91 trillion in shareholder wealth creation.
William Sharpe formalised the logic decades ago in The Arithmetic of Active Management (Sharpe, 1991). Before costs, the average actively managed pound must match the market — because active investors collectively are the market. After costs, they must underperform. That's not a theory. It's arithmetic.
A single global index tracker does the job. No stock selection. No hit-rate anxiety. No agonising over whether you've missed the next big winner.
You don't need to find the needles. You own the haystack.
A century of data, one clear lesson
100 years ago, the haystack was smaller and the needles easier to spot — 89 firms drove half the gains across the first nine decades. Today, the needles have shrunk to 46. The direction of travel is clear.
Whether it continues is an open question. Bessembinder himself asks whether artificial intelligence will accelerate 'winner take all' dynamics or whether broad adoption of new technology might level the playing field. Nobody knows.
But the case for indexing doesn't depend on the answer. The more concentrated wealth creation becomes, the riskier it is to bet you can identify the handful of firms that will matter — and the more sensible it is to own them all.
You don't need to predict which 46 companies will drive the next century. You need to make sure you haven't left any of them out.
Resources
Bessembinder, H. (2026). One Hundred Years in the U.S. Stock Markets. Working paper, Arizona State University.
Bessembinder, H. (2018). Do Stocks Outperform Treasury Bills? Journal of Financial Economics, 129(3), 440–457.
Fama, E. F., & French, K. R. (2004). New Lists: Fundamentals and Survival Rates. Journal of Financial Economics, 73(2), 229–269.
Sharpe, W. F. (1991). The Arithmetic of Active Management. Financial Analysts Journal, 47(1), 7–9.
Finding the right adviser
Owning the whole haystack is the easy part. The harder questions (how much to invest, which tax wrappers to use, how to draw an income in retirement) are where good financial advice earns its keep. If you're looking for an adviser who understands the evidence and won't try to sell you on stock picking, our free Find an Adviser questionnaire can match you with one. It takes two minutes.
For advisers
At TEBI, our mission is to change the way the world invests and help investors achieve better outcomes. We do that through high-quality video and written content. If you'd like to commission your own content to add value for your current clients, attract new ones, or both, email Robin Powell or connect with him on LinkedIn.