Why do mutual funds underperform? Because most "skill" is actually luck
- Robin Powell
- 43 minutes ago
- 7 min read
Why do mutual funds underperform? Not because managers are incompetent. Most of what looks like skill is actually luck, and luck has a habit of running out.
Terry Smith was untouchable. For years, his Fundsmith Equity fund was the one British investors pointed to when they wanted proof that active management could work. Top Morningstar ratings. Billions pouring in. A performance record that made index fund advocates squirm.
Then came 2021. And 2022. And 2023. And 2024.
Four consecutive years of trailing the MSCI World Index. Last year was particularly brutal: Fundsmith returned around 8.9% while the benchmark delivered 20.8%. Morningstar downgraded the fund from Gold to Silver, then to Bronze. Investors who'd piled in during the good years headed for the exits.
So what happened? Did Smith forget how to pick stocks?
Here's the uncomfortable possibility: maybe there was less edge than anyone thought. Maybe the years of outperformance weren't proof of exceptional skill.
New research from the University of Sydney puts hard numbers on a question the fund industry would rather not answer. Those numbers suggest Smith's story isn't unusual. It's practically inevitable.
New research shows why mutual funds underperform
Jiali Gao and Juan Yao, finance researchers at the University of Sydney Business School, set out to answer a deceptively simple question: when a fund beats the market, how much is skill and how much is luck?
Their answer, drawn from 3,662 US mutual funds over more than three decades, is striking. At least 55% of the variation in fund performance is attributable to luck. Under some statistical models, the figure rises to 99%.
When you look at a fund's track record and see years of market-beating returns, more than half of what you're seeing is noise. The manager may be talented. They may be working incredibly hard. But the results you're using to judge them are dominated by factors outside their control.
Think of it like watching a roulette table. One player has won five spins in a row. Is she skilled? Or is she the statistical inevitability that emerges when enough people spin the wheel?
The maths says the latter. And the maths applies to fund managers too.
Why yesterday's winners become tomorrow's losers
Here's where the research takes an unexpected turn. Luck doesn't just fail to persist. It predicts future underperformance.
This sounds counterintuitive. Surely a lucky streak is random, with no bearing on what comes next? But in fund management, luck leaves footprints. And those footprints lead somewhere problematic.
When a fund gets lucky, two things happen. The portfolio's holdings appreciate, mechanically increasing the fund's size. And performance-chasing investors pile in, attracted by the strong numbers.
Lucky funds get bigger. Much bigger.
Bigger is a problem. Academic research has long documented "diseconomies of scale" in fund management — the more money a manager oversees, the harder it becomes to generate excess returns. Large funds move markets when they trade. They're forced into more liquid, more efficiently priced securities. Strategies that worked at £500 million become impossible at £5 billion.
Luck triggers a vicious cycle. Strong performance attracts capital. Capital increases fund size. Larger size erodes the capacity to outperform.
Luck triggers a vicious circle. Strong performance attracts capital. Capital increases fund size. Larger size erodes the capacity to outperform.
The pattern holds across nearly every fund size category. As Figure 1 shows, "good luck" funds consistently maintain higher assets under management than "bad luck" funds — year after year, decade after decade.
Gao and Yao's data confirms this. Funds that experienced lucky periods significantly underperformed over the following 12 to 24 months. Yesterday's winners didn't mean-revert. They fell below average.
Back to our roulette player. She's won five spins. The crowd has gathered. Feeling confident, she increases her bets. But the wheel doesn't care about her streak. The odds haven't changed. And now she's risking more on each spin.
Five-star funds significantly underperform one-star funds
If luck dominates performance, what does that mean for the ratings millions of investors rely on? Nothing good.
Morningstar's star ratings are the most widely used fund evaluation tool in the world. Advisers reference them. Platforms display them prominently. Investors treat five stars as a seal of approval.
But star ratings are backward-looking. They rank funds by past risk-adjusted returns. If past returns are mostly luck, star ratings are mostly measuring luck.
Gao and Yao calculated exactly how much. Morningstar's star ratings have 78.6% exposure to luck, versus only 48.4% exposure to genuine "funding status" — whether the fund is sized appropriately to generate alpha. The ratings capture nearly twice as much noise as signal.
The consequence: five-star funds significantly underperformed one-star funds in subsequent periods. Funds that looked best went on to disappoint. Funds that looked worst fared better.
This pattern was first documented by Blake and Morey in 2000. Morey called it the "kiss of death" in a 2003 follow-up. Gao and Yao explain why it happens. Star ratings identify funds that got lucky. Lucky funds attract inflows. Inflows increase size. Size erodes future performance.
The five-star badge isn't predicting quality. It's marking funds about to disappoint.
The next chart shows the trap in action. Investors continue pouring money into "overfunded" funds that happen to be lucky (orange dashed line), while pulling capital from "underfunded" funds that have been unlucky (blue solid line). The rational response would be the opposite: underfunded funds have room to generate alpha, while overfunded funds are structurally disadvantaged. But investors chase recent returns, not funding status.
The evidence is just as damning for UK funds
The Gao and Yao research focuses on US mutual funds. But UK investors shouldn't take comfort from that.
British academics have run similar analyses for years. The findings are, if anything, worse.
David Blake and colleagues at Bayes Business School examined 552 UK equity funds between 1998 and 2008. Using bootstrap methods to separate skill from luck, they found that 95% of fund managers failed to outperform the luck distribution. That was before fees. After fees, the failure rate hit 100%.
A 2008 study by Keith Cuthbertson, Dirk Nitzsche and Niall O'Sullivan analysed 935 UK equity funds over nearly three decades. While a tiny handful of top performers showed evidence of genuine skill, the vast majority did not. Their verdict: "If you choose your active funds by throwing darts at the Financial Times' mutual fund pages, then you are highly likely to choose a fund which has no skill."
Dart-throwing. That's what decades of UK fund selection has amounted to.
If you can't identify skill, stop trying
So where does this leave investors?
Stop playing the game.
The research doesn't prove skilled managers don't exist. Some almost certainly do. The problem is that skill is nearly impossible to identify in advance. The signal is too weak, the noise too strong. By the time a track record looks convincing, luck has already done its damage — attracting capital, bloating the fund, making future outperformance structurally unlikely.
This isn't despair. It's liberation.
If you can't reliably pick winners, you don't have to try. Low-cost index funds guarantee you'll match the market return, minus minimal fees. No guessing. No chasing. No hoping this manager, this time, has the magic touch.
The average actively managed UK equity fund charges around 0.75% to 1% annually. A comparable index fund might charge 0.1% or less. That gap compounds relentlessly. Over 30 years, it can consume a quarter of your portfolio's potential value.
What do you get for that extra cost? A worse than random chance of beating the index.
Terry Smith probably isn't a worse stock-picker than he was five years ago. He may well possess genuine skill. But the research tells us we can't know. The performance that made him famous was inseparable from luck, and the success that luck delivered made future success harder. The billions that flowed into Fundsmith weren't rewarding skill. They were chasing randomness.
Investors now leaving after four difficult years are caught in the same trap. They're reacting to recent performance, just as they did when they arrived. The direction has reversed, but the behaviour hasn't.
You don't have to keep playing. Own the market at the lowest possible cost, ignore the star ratings, and let compounding do its work without fees paid for illusory skill.
The fund industry would prefer you didn't know that most of what glitters is luck catching the light.
Now you do.
Resources
Gao, J. & Yao, J. (2025). Luck and skill in the world of diseconomies of scale. Working paper, University of Sydney Business School.
Blake, D., Caulfield, T., Ioannidis, C. & Tonks, I. (2017). New evidence on mutual fund performance: A comparison of alternative bootstrap methods. Journal of Financial and Quantitative Analysis, 52(3), 1279–1299.
Cuthbertson, K., Nitzsche, D. & O'Sullivan, N. (2008). UK mutual fund performance: Skill or luck? Journal of Empirical Finance, 15(4), 613–634.
Recently on TEBI
Are you a financial adviser?
If you're not providing regular educational content, your clients and prospects are getting it elsewhere (probably from someone selling expensive active funds). Let us create newsletters, articles, and resources that keep clients engaged with evidence-based thinking. Email Robin or message him on LinkedIn to discuss.