Read an Excerpt
CHAPTER 1
Past Performance of Fund Managers Is Not a Predictor of Future Performance
Even if you identify the managers who have good past performance, there's no guarantee that they'll have good future performance.
— George Sauter, Vanguard Group, Wall Street Journal June 17, 1997
Yesterday's masters of the universe are today's cosmic dust.
— Alan Abelson and Rhonda Brammer, Barren's, October 5, 1998
In investment performance, the past is not prologue.
— Charles Ellis, Winning the Loser's Game
The biggest investment mistake people make is focusing on last year's mutual fund performance and not on what really drives returns.
— Barbara Raasch, partner at Ernst and Young, Business Week, February 22, 1999
I do not believe that they (investment advisors) can identify, in advance, the top-performing managers — no one can! — and I'd avoid those who claim they can do so.
John Bogle, Common Sense on Mutual Funds
One of the most popularly held beliefs is that past performance is a predictor of future performance. For example, when a business seeks to hire a new CEO it will look at candidates who have been successful at similar positions. When a baseball team looks to add a cleanup hitter to its lineup, it will try to trade for, or sign as a free agent, someone who has previously performed well in that role. In both cases the rationale is that the person being sought has clearly demonstrated the skills necessary to do the job. While there is no certainty of success, the rationale is sound.
Using the same logic, investors choose active mutual fund managers based on past success. Financial trade publications know that in order to create a bestselling magazine all they have to do is place on the cover a lead like "The Ten Hot Funds to Include in Your Portfolio." Unfortunately for investors, the logic used to hire the next CEO or to find the next cleanup batter doesn't apply to identifying the active managers that will outperform the market in the future. For example, you would think that if anyone could beat the market, it would be the pension funds of the largest U.S. companies. They have access to the best and brightest portfolio managers, all of whom are clamoring to manage the billions of dollars in these plans. Presumably, these pension funds rely on the excellent track records of the "experts" they eventually choose to manage their portfolios.
Piscataqua Research, in a study covering the period 1987 through 1996, found that only 10 (7%) out of 145 major pension funds outperformed a portfolio consisting of a simple 60%/40% mix of the S&P 500 Index and the Lehman Bond Index, respectively. A 60% equity/40% fixed income allocation was used, since that is estimated to be the average allocation of all pension plans. The Piscataqua Research study provides evidence against not only the strategy of choosing managers based on past performance, but also the use of active managers in general.
In 1998, fewer than 20% of all equity funds outperformed the S&P 500 Index. That figure drops to 11% over the previous 10 years and to just 4% over the previous 15 years. And, as Fortune magazine put it: "Despite the solemn import that fund companies attribute to past performance, there is no evidence that the 4% who beat the index over the 15-year period owe their record to anything other than random statistical variation." (Given the number of players in the "game," the number that do succeed is far less than would be randomly expected.) "The whole industry is built up around a certain degree of black magic." Fortune concluded: "Despite volumes of research attesting to the meaninglessness of past returns, most investors (and personal finance magazines) seek tomorrow's winners among yesterday's. Forget it. ... The truth is, much as you wish you could know which funds will be hot, you can't — and neither can the legions of advisers and publications that claim they can." Despite being the very same magazine that glorifies the latest hot fund manager, they added: "We have learned that past investment records make lousy crystal balls."
As H.L. Mencken said: "The most costly of follies is to believe in the palpably not true."
Investors simply have no way of identifying ahead of time the few active managers that ultimately turn out to outperform the market over the long term. Jonathan Clements, columnist for the Wall Street Journal, stated: "I believe the search for top-performing stock funds is an intellectually discredited exercise that will come to be viewed as one of the great financial follies of the late 20th century." One example of this is a study that found that over two consecutive decades a remarkable 99% of top quartile funds moved closer to, or even below, the market mean from the first decade to the second. In fact, there was only one exception — the Fidelity Magellan fund. Its performance since then is evidence of the folly of believing that past performance of an actively managed fund is a useful predictor of future performance. "Make no mistake about it, the record is clear that top performing funds inevitably lose their edge."
Even Smith Barney, which heavily advertises its funds that have received five-star ratings on the basis of past performance, had this to say: "One of the most common investor mistakes is choosing an investment management firm or mutual fund based on recent top performance." This was the conclusion they drew from their own study covering 72 equity managers with at least 10-year track records. The managers studied covered the full spectrum of investment styles: large-cap to small-cap, value to growth, and domestic to international. In fact, the study found: 'The investment returns of top quintile managers tended to plunge precipitously while the returns of bottom quintile managers tended to rise dramatically."
Perhaps most revealing is what John Rekenthaler, research director of Morningstar, had to say when asked how to pick a winning equity mutual fund: "We should have more answers." He also added that there is "surprisingly little" that we can say for sure about how to find top-notch stock funds. More recently Rekenthaler stated that actively managed funds are beginning to show up on his cultural radar as a "marketing scam for suckers." This chapter documents the results of investment strategies that rely on past performance to choose actively managed funds. After reading this chapter I hope that you will never again want to read an article with a title like "The Ten Best Funds."
Investors Rely on False Premises When Choosing Mutual Funds
Periodically investors go through the ritual of evaluating their mutual fund holdings. The usual process results in selling the poor performers and purchasing shares of the latest hot funds. Let's examine the methodology and results of the various approaches based on this "buy winners strategy."
Perhaps the most popular approach is to rely on fund ratings by such services as Morningstar, which rates funds using a star system similar to the one used by film critics. Evidence of the popularity of this approach is that it is estimated that over 80%) of inflows into mutual funds go to four- and five-star rated funds. Unfortunately for investors, this approach has consistently produced below-benchmark returns for both bond and equity funds. For example, a Financial Research study covering the period January 1, 1995 through September 30, 1998, revealed that two- and three-star funds outperformed their four- and five-star counterparts for the entire period. The study's conclusion: "the linkage between past performance and future realizations is tenuous if not nonexistent." A similar study, by Christopher R. Blake, associate professor of finance at Fordham University's Graduate School of Business, and Matthew Morey, assistant professor of finance at Fordham, found that for the five-year period ending December 31, 1997, five-star funds underperformed the market by almost 4% per annum. The study also found that the differences between the performances of the three-, four-, and five-star funds are so small as to have very little statistical significance. Morningstar even stated that there is no connection between past and future performance and stars, historic star ratings, or any raw data and that the stars should not be used to predict short-term returns or to time fund purchases. Here is an interesting observation. Expenses play such an important role in determining returns that investors could have done a better job of predicting performance than did Morningstar's star system simply by ranking funds by their respective expense ratios. Morningstar itself conducted a survey of management fees and their effect on mutual fund returns. They found that mutual funds that charge the lowest fees also generate the highest returns over time. "The simple fact is costs take a bite from investor returns so the higher costs lower returns," said Russell Kinnel, head of equity research at Morningstar.
Investors should ask themselves the following question: If Morningstar, with all of its resources, can't identify with any degree of success the future winners, is there any rational reason to believe that I can?
Another popular approach is to rely on the recommendations of trade publications such as Forbes. Unfortunately, this approach produces the same poor results as does following the star system. For example, a 1991 Yanni Bilkey Investment Company study of the Forbes "Hall of Fame" list of mutual funds, which recommends the funds individuals should buy, found that for the seven five-year periods beginning in 1980, only once, and by the smallest of margins, did the group beat the S&P 500 Index. However, they never once beat the average equity fund. Another study on this famous list covered the period 1983 through 1990 and found that a portfolio of the Hall of Fame funds would have returned 10.46%, versus 16.43% for the S&P 500 Index. Following the recommendations of other popular financial magazines produced similar results. Here is another interesting point. If past performance had predictive value, then the "buy" lists of publications would contain the same names year after year. If the same funds don't repeat, how valuable is the list?
Another often-used approach is to buy the funds that have produced the best returns during the recent past. In his book A Random Walk down Wall Street, Burton G. Malkiel reported that he did extensive testing on whether an investor, by choosing the "hot" funds, could outperform the market. The results showed the ineffectiveness of a strategy that chose the top 10, 20, 30, or more funds on the basis of the performance of the previous 12 months and then one year later switched to the new top performers. Since 1980 this strategy produced results that were not only below the average mutual fund but also below that of the S&P 500 Index. Similar results were found when Malkiel tried ranking funds by their past 2-, 5-, and 10-year track records. Even more compelling evidence of the risk of buying yesterday's winners comes from Mark Carhart. Carhart, now cohead of quantitative research at Goldman Sachs Asset Management, studied the performance of mutual funds all the way back to 1962. He came to the amazing conclusion that the top 10% of performers in any one year are more likely to fall to the bottom 10% than repeat in the top 10%. In other words, investing in last year's top performers is a crap shoot.
Investors also fall prey to what is known as the gambler's fallacy, the idea that winners ride "hot streaks." Of course, there is no factual basis for that idea, either in gambling or investing. As further proof that selecting mutual funds that have previously beat the market is a loser's game, witness the effort of Mark Hulbert, publisher of the Hulbert Financial Digest, a newsletter that tracks the performance of investment newsletters. He constructed a portfolio of "market beaters" by choosing managers who had outperformed the market in the preceding year. That portfolio earned a 99% return over the next 15 years. Not a bad return, except for the fact that a portfolio of "market losers," those funds that lagged the market in the previous year, returned 350% over the same period. In contrast to these seemingly impressive returns, the stock market as a whole rose about 600% over the same period.
While the press ignores the poor and inconsistent performance of active managers, it has not gone unnoticed by large institutional investors. Philip Halpern, the chief investment officer of the Washington State Investment Board (a very large institutional investor), along with two of his coworkers, wrote an article on their investment experiences. They wrote the article because their experience with active management was less than satisfactory and they knew, through their attendance at professional associations, that many of their colleagues shared and verified their own experience. They quoted a Goldman Sachs publication: "Few managers consistently outperform the S&P 500. Thus, in the eyes of the plan sponsor, its plan is paying an excessive amount of the upside to the manager while still bearing substantial risk that its investments will achieve sub-par returns." The article concluded: "Slowly, over time, many large pension funds have shared our experience and have moved toward indexing more domestic equity assets."
Meir Statman advised that investors would do well to heed the advice of noted economist Paul Samuelson. While having kind words for a few active investment managers such as Warren Buffett and John Templeton, Samuelson drew the following conclusion. "Ten thousand money managers all look equally good or bad. Each expects to do 3% better than the mob. Each has put together a convincing story. After the fact, hardly 10 out of 10,000 perform in a way that convinces an experienced student of inductive evidence that a long-term edge over indexing is likely. ... It may be the better part of wisdom to forsake searching for needles that are so small in haystacks that are so very large."
If after all the evidence just presented, you still believe that you can select the future winning mutual funds based on past performance, I suggest that you ask yourself the following question. "What different process to select these future winners will I be using from that of others who have tried to do the same thing, relying on the same paradigm — of whom all failed?" If you can't identify any differentiation, then what logic is there in believing you will succeed where so many others have failed?
Top-Performing Actively Managed Funds and Encores
While it may be a good way to predict which batters are the most likely to hit .300 in a given year, unfortunately, past performance is a very poor indicator of the future when it comes to mutual funds and their managers. A study done by NationsBank, using Lipper Analytical Services data, provided further powerful evidence that investors relying on past performance as a predictor of future performance are playing a loser's game.
NationsBank examined the performance of mutual funds in the three asset class groups of growth, growth and income, and small company, for 10 one-year periods beginning in 1986. They calculated the percentage of funds that were ranked in the top 25% that repeated in the following year. Keep in mind that we would randomly expect 25% of the previous winners to repeat. If it were a winner's game, a majority would do so. For Lipper's growth category the highest percentage of repeat performers was just 12%, and the average number of repeaters was just 8.4%. For the growth and income category the figures were just 10.5% and 7.2%. For the small company category the figures were just 17%; and 8.4%. With just an average of 8% of all top performers repeating the following year, how is an investor to know which of the previous winners will repeat?
A study published in the April 2000 issue of the Journal of Finance shed further interesting light on the issue of encore performances. Have you ever seen an ad for a mutual fund that advertises its poor track record? Of course not. The ads are a form of selection bias — you only see ads for the winners, never the poor performers. Why do funds advertise their winning records? The presumption is that investors will extrapolate past success into the future — despite the warning/disclaimer that the SEC requires, that past performance is not an indicator of future performance.
(Continues…)
Excerpted from "What Wall Street Doesn't Want You To Know"
by .
Copyright © 2001 Larry H. Swedroe.
Excerpted by permission of St. Martin's Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
