As all investors know there are lies, damned lies and past performance statistics. What the aca...
As all investors know there are lies, damned lies and past performance statistics.
What the academics and statisticians seem unable to agree on is whether past performance has any value. As first reported in last week's Investment Week, Autif has research which concludes that it can have value and previous work suggesting otherwise is flawed. Immediately after this Virgin issued a statement arguing the association's research was itself incorrect. Virgin, had previously commissioned a report on past performance, and this conclusively proved that past performance is not a guide for the future.
Despite the somewhat aggressive tone of Virgin's response, this was Autif's point. It had commissioned a report into the accuracy of past research into past performance ' a slight difference. In examining all the most recent data done on this particularly thorny subject, the study put together by Charles River Associates (CRA) says that most research has overlooked some key points. These being the fact that the research of the past has been done under certain assumptions, such as persistency in fund performance is the result of fund manager skill. The CRA study issued last week says this is not the point. The important factor is whether persistency, for whatever reason, exists or not and, apparently, it does.
For fund managers and intermediaries alike the study is a challenge to the FSA, which has consistently refuted the idea that past performance should have any relevance to fund selection in its comparative tables.
If fund selection for consumers is left to be judged just by cost then it will be the low-cost companies that would be the beneficiaries of it. Intermediaries have long known that many factors go into recommending a portfolio, much of which is the fund manager's skill. After all, who would hire a builder with a reputation for shoddy but cheap work?
This is also why many intermediaries have followed fund managers from one portfolio to another ' something that has been difficult to track in any research done into past performance. It seems a rather practical and logical thing to do ' take the information provided, such as the quants and costs, add in the qualitative such as whether the fund manager is good at what he or she does, based on a myriad of factors such as experience, reputation and ability, and apply this to the decision made. And yet the FSA has been content to leave consumers with just the costs, with its final say on past performance being its oft-quoted 'past performance is no guide to the future.'
Perhaps we could now add 'we think but we're really not sure' to the front of that disclaimer.
Slow progress in improving diversity
Share purchase deal with assets of £28m
Came into effect in January
Three examples of compensation rule issues
Buying in baskets