The IMA's soon-to-be-published report into whether past performance provides an identifiable guide t...
The IMA's soon-to-be-published report into whether past performance provides an identifiable guide to future performance found what we have all known. Past performance can be a useful element in fund selection. It seems a point so obvious it is almost impossible to believe that there is still a debate on the subject.
It is not an infallible guide, of course, but anyone buying into funds on behalf of clients knows they would be a fool to ignore it. In fact, the report was a little disappointing.
What its authors, Charles River Associates, revealed was that most commonly, a top quartile fund selected on 12 months' past performance has, over the past 21 years, roughly a 33% chance of still being there 12 months later. Statistically that is well above random, but it is hardly satisfactory. So, what the IMA report shows is dog funds are just slightly more likely to remain dog funds than chance would have it. And top-performing funds continue to perform well against their peers, at least over the near future, in roughly the same proportion. It is hardly a day to wave flags for the active fund management industry.
The report was commissioned as a rebuttal to the FSA's stance that past performance is not useful to retail investors picking funds. That was important as it was seeking to put pressure on the FSA to include past performance data in its online comparative funds tables. The FSA's stance, based on its own reviews of past performance literature, was, the IMA argued, untenable. The IMA, of course did not go so far as to say that such a stance certainly saved the regulator a lot of time and money in collecting, verifying and publishing fund performance statistics and keeping its website up to date. To include past performance in its web-based product tables would have required additional resources that the FSA does not seem to have.
But does the IMA report actually put pressure on the FSA to change the tables? The answer is probably not. One point advisers should note about the report, however, is its finding that of all the periods of past performance you can look at simple three-year figures provide the worst guide to future performance. Buy a fund that was top quartile 12 months ago and you have a good chance of it outperforming in the coming 12 months. That chance falls to near random using three-year figures. An interesting finding, and one that should shake brokers not looking at past performance in a range of different ways, to rethink the weight they give to this traditionally used time frame.
Developed by industry-wide group
Joined in 2002
'Educate clients' children'
Raised £15m earlier this week
From 8pm Friday 19 October