Christophe Merkle and colleagues at the University of Mannheim asked a sample of educated and well-informed investors to look at funds' returns and say whether they thought they were due to luck or skill. They found that they tended to make two errors.
One is a form of selection effect. People fail to see that, in a large sample of funds, a few will outperform consistently, or by a lot, simply by chance even if there's no skill. For example, a fund with no skill has a 50:50 chance of beating the market in any 12-month period. If returns are serially independent then it has a 3.1 per cent chance (0.5 to the power five) of outperforming in each of five successive years. That might seem like a small chance. But as there are over 3,000 funds in Trustnet's database, we'd expect almost 100 to have such consistent outperformance - even if not one had any skill at all.
Secondly, they found investors didn't discount returns sufficiently to allow for their volatility. They are more likely to attribute skill to a fund with high but variable returns than to one with less spectacular but more stable returns - even if the latter really does have skill. This suggests that investors are prone to the intentionality bias - a tendency to attribute outcomes to peoples' intentions and abilities even when they are in fact due to luck. "Investors underestimate the probability that a track record was generated by pure chance, especially in large fund populations and when fund managers take excessive risks," says Professor Merkle.
These, though, are not the only mistakes investors make in picking funds. Earlier research at the University of Mannheim found that sophisticated investors are overconfident about their ability to spot good fund managers. And US research has found that investors are prone to an anchoring effect; they fail to appreciate how much funds' fees compound over time.
All this helps explain a long-standing paradox - that investors continue to hold actively managed funds even though most underperform their benchmarks over the long term; research at Vanguard Asset Management shows that, over the last 15 years 85 per cent of global equity funds have either underperformed, closed or merged. (It's important to take account of fund closures in assessing the overall performance of fund managers, because otherwise survivorship bias - the tendency for only good funds to survive - will overstate the number of funds that do well.)
The efficient market hypothesis - the theory that investors cannot systematically beat the market without taking on extra risk - has been widely attacked in recent years. Nevertheless, the evidence suggests that, when they are thinking of buying funds, investors should presume that the theory is true, unless they have strong evidence to the contrary.
