Dear Editor:

Mr. Williams’ article makes the case that the recipe for investment success is an index fund. His argument is strengthened if he can show that skilled managers don’t exist, or that if they exist, you can’t find them.

In my first letter, I pointed out that the coin-flipping analogy Mr. Williams offered in support of his argument assumes that skilled managers do not exist because he uses a coin that has an equal chance of turning up heads or tails.

Mr. Williams explains that he purposely constructed a world in which skill does not exist (he says, “precisely because skill is not a factor in coin-tossing”) in an attempt to prove his point that it is difficult to distinguish luck from skill when comparing past track records. Within a world where skill does not exist, he succeeds in proving that luck can look a lot like skill.  But if the purpose of this analogy was to support the argument for index funds, this point seems tangential.

His analogy ultimately provides much less support for index funds than it first appears because it leaves unanswered the questions of whether skilled managers exist and whether they can be found by examining track records.

In a world where there is no skill, I think we both agree that track records cannot help you find something that does not exist. But, when the analogy is changed to allow for skill it becomes clear that track records can increase the odds of finding skill.

Mr. Williams then questions whether examining track records to find skilled managers is simple and would work in reality. He references a study in which only 4 of the top 20 equity mutual funds from 1983-1993 “maintained their stellar record in the next decade,” and “16 failed to even match the market return” as evidence that it can’t be done.

I will concede that it is not so simple that it can be reduced to a single metric that needs to be assessed just once each decade.  Frankly, I don’t know anyone who thinks that choosing the top 20 performing mutual funds over the last 10 years, and holding onto them, no matter what, for the next decade, is a good idea. For one thing, fund managers change a lot (the average tenure is less than five years). If you chose a fund based “solely” on its past 10-year track record, why would you stay in the fund after the manager who built that track record leaves? It doesn’t make sense because at that point the track record of the fund tells you nothing about the skill of a completely different manager.

Studies like the one cited by Mr. Williams that are based on an analysis of mutual fund track records share a common problem. Mutual fund track records often include returns attributable to previous managers. If you are trying to determine whether a manager is skilled it makes no sense to credit him for returns generated by a previous manager who is no longer there. Perhaps the strongest conclusion we can draw from such studies is that a mutual fund’s past performance is not a reliable indicator of the future performance of the current fund manager especially if the manager has changed. Actually, I don’t need a study to convince me of that.

The evidence Mr. Williams provided in favor of index funds looks weaker upon examination than it first appeared. You cannot prove that skilled managers don’t exist with an analogy that assumes they don’t.  And, you cannot prove that you can’t find skilled managers with studies of the type referred to by Mr. Williams. If we want to find skilled managers, we need to examine manager track records not mutual fund track records. The question of how to do this in practice is beyond the scope of this reply.  But, if there is interest, I’d be happy to contribute my thoughts to this debate.

Ken Kam
President
Marketocracy Capital Management, LLC

Display article as PDF for printing.

Would you like to send this article to a friend?