The Evidence-Based Investor

There will always be winners

Posted by Robin Powell on July 12, 2017

“Journalism attracts people with exceptional linguistic talent,” William Bernstein wrote in The Four Pillars of Investing, “but I have found that very few have the mathematical sophistication to appreciate the difference between skill and luck.”

I must say, the experience of this particular journalist — a history graduate, by the way, with a modicum of linguistic ability, who still can’t believe he scraped a B in O Level Maths — is very similar. On a daily basis I come across articles in the financial media that simply wouldn’t get written if the writer had a basic grasp of the laws of probability.

A typical example was a piece the other day from FE Trustnet, with the headline The UK funds that have consistently beat (sic) the market over the long run. The article featured six funds in the Investment Association’s UK All Companies sector which, sure enough, do appear to have outperformed the market after costs.

The impression given, of course, is that here are six funds which have skilful managers and which should continue to beat the market in the future. But let’s add some perspective, courtesy of Dr Tim Edwards, a far better mathematician than I, at S&P Dow Jones Indices in London.
 
“There are currently 253 funds in the “IA UK All Companies” sample, says Dr Edwards. “Let’s suppose (for purposes of discussion) they have a 50% chance of outperforming in any given 5 years.
 
“Then, over 15 years we would expect roughly 1/8th to outperform in all of the three 5-year sub-periods. (1/8 = ½ x ½ x ½)
 
“In other words, we expect around 32 funds to be in possession of a perfect record, just through chance alone.
 
“Of course, that ignores the fact that of those 253 funds, quite a few might have launched less than 15 years ago. But if there were roughly the same number of funds available 15 years ago, and the ones that outperformed survived, then we still should still be able to find those 32 winners.
 
“Now, the exact analysis was over 5-year rolling periods and asks for 75% or more outperformance, and they found 22 funds. But you see the point. There will always, ex-post, be plenty of funds with great track records.”

The SPIVA data collated by Dr Edwards’ colleagues shows time and again that, over the long term, no more funds outperform their benchmarks than you would expect from random chance (in actually, slightly fewer funds do). That data is wholly consistent with other major studies carried out by, for example, University of Chicago Booth School of Business and Cass Business Business School in London, and the proportion of winners is broadly similar around the world.

Remember, the confidence trick the fund industry plays on us depends on it constantly launching and promoting exciting new funds, and quietly merging dud funds with other funds or closing them altogether.

There will always be funds that have outperformed in the past, and those are precisely the funds that you read about on trustnet.com, in investment magazines and the weekend money pages. They are also the ones that get promoted at adviser conferences and you see advertised on taxi cabs and train station billboards.

But research has repeatedly shown that long-term outperformers are almost impossible to spot in advance, and the funds that outperform in the future are unlikely to be those that outperformed in the past.

Nobody to my knowledge — no adviser, no investment consultant, no fund analyst — has produced a system for identifying future winners in advance with any degree of consistency. When someone does, that really will be a story worth writing.

  • Share article:

Robin Powell

A journalist and broadcaster, Robin Powell is a campaigner for positive change in the investing industry. He runs a communications company for the evidence-based investing sector called Regis Media.

Disclaimer: All content is for informational purposes only. I make no representations as to the accuracy, completeness, suitability or validity of any information on this site and will not be liable for any errors or omissions or any damages arising from its display or use. Full disclaimer.