The University Record, August 14, 1995
Are fund tables helpful?
The last issue of the Record (July 24) illustrates a principle neatly stated by Herbert Simon, a Nobel Prize winner:
"In a world where information is relatively scarce, and where problems for decision are few and simple, information is almost always a positive good. In a world where attention is a major scarce resource, information may be an expensive luxury, for it may turn our attention from what is important to what is unimportant."
American Economic Review, May 1978
I refer to your table giving the experience of a set of mutual funds. Would that information really help predict which funds will dowell in the future? Does the most recent three years even predict backward to the whole five- or 10-year period, where we get some correlation because we are relating part of an experience to the whole? Why are so few funds more than five years old?
What is important is something different. First, there is risk as well as yield, as the first-page report on the Common Fund (where someone was gambling with investors' money) illustrates. International funds can appear to have high yields, since high interest is paid when there is inflation, but the inevitable depreciation of the currencies wipes out most of them.
Second, there is the "aha" factor, or the selection bias. There are now more mutual funds than common stocks. A company can issue hundreds, then select the few that (happen to) do well and promote them. It reminds one of the race track tout who promotes a different horse to each section of the stands, then collects tips from the one where he guessed right, and avoids the others.
Third, there is the cost of having someone manage your portfolio. It is revealing that "index funds," which simply allocate funds across a representative list of stocks, persistently do better than the vast majority of funds, and it is a changing minority that may seem to be doing better. Index funds do no trading, hence have much smaller management fees. All funds persistently sell for less than the market value of the stock they own, when you buy and when you sell, reflecting the management costs that they incur.
Finally, there is the "new fund" phenomenon. A rapidly growing new fund can do well because it can adjust its portfolio by what it buys without having the double cost of selling some and buying others. Hence, while it is growing it can look very good. Again, does this help us predict the future? Perhaps, if you can predict when a fund will stop growing!
Remember that a major phenomenon in the world is "regression to the mean." A set of random number time series will exhibit negative autocorrelation, that is, those higher than average are more likely to fall and those lower are likely to rise. Indeed, this is exactly opposite of the naive assumption that good records will persist.
What most of us want is to reduce the risk, yet take advantage of the higher yields in common stocks. The obvious sensible answer is an index fund, or a very large stable fund like CREF, rather than trying to outguess the insiders and predict the future. Gambling may be fun, but not with one's retirement funds!
I suggest you take a full listing of mutual funds as of five years ago and see how well their previous five years' experience predicts their subsequent five years, including the ones which disappeared.
professor emeritus of economics, and research scientist emeritus, Institute for Social Research