Dian's Column
Dian's Archive

Lavine/Liberman Archive




Lipper
Muriel Siebert & Co.


Beware of Past Performance Information

- Alan Lavine and Gail Liberman



Are you using Morningstar or other mutual fund data services to evaluate, pick or buy and sell mutual funds? If so, be aware that the information isn't the holy grail.

Recent financial research shows that investors and their financial advisers who depend too much on Morningstar and other data may be making a mistake. The reason: They may be relying too much on past performance data. The worst-performing manager in one period may rebound in the future, while the best-performing managers today could lag behind tomorrow.

Morningstar and other sources of risk and return statistics have flaws, says S. P. Kothari, finance professor at the Massachusetts Institute of Technology, coauthor of the study, "Evaluating Mutual Fund Performance."

Kothari's study, published in CPA Journal, found that standard, widely-used measures of mutual fund performance are inaccurate and unreliable, and can lead to faulty conclusions by investors. The problem: It is easy to detect abnormal performance and market timing ability when none exists.

There are other problems. The Capital Asset Pricing Model is the method used to produce many of the statistics needed to evaluate a mutual fund's risk and return. These are complicated mathematical calculations used by analysts. But John Graham, finance professor at Duke University, claims the Capital Asset Pricing Model has serious short-comings.

"It is helpful to use Morningstar's data," Graham said. "You can eliminate bad funds and you might reduce the odds that you will have a real loser. But if you buy or sell a fund, you still must keep a close eye on the investments."

Bill Goslee, investment strategist at Nationwide Financial, Columbus, Ohio, agrees that you can't rely on historical statistics. No one can predict the future. The best anyone can do, he stresses, is to make informed professional decisions about how a fund is currently managed.

"You have to go beyond the quantitative perspective," he said. "A fund can have a great track record based on statistics. We talk to the fund managers about their holdings. We look at the management. We look at manager changes and whether a manager is sticking to his investment style."

A study by two Harvard Business School finance professors, Randolph Cohen and Joshua Coval, say it's better to look at a fund's current holdings and timing of trades rather than past returns and rankings. The two draw the conclusion in their study, "Judging Fund Managers by the Company They Keep," a working paper published by the Social Science Research Network (http://:papers.ssrn.com). The professorsí performance evaluation approach judges a fund manager's skill based on whether his or her investment decisions resemble fund managers with outstanding performance records.

#

Alan Lavine and Gail Liberman are husband and wife columnist and authors of The Complete Idiot's Guide To Making Money With Mutual Funds, (Alpha Books).


To read more columns, please visit the column archive.




[ top ]