By Sam Mamudi
A DOW JONES COLUMN
It's perhaps the most quoted financial-services mantra: Past
performance is no indication of future results.
Unfortunately, human nature usually leads investors to think the
exact opposite.
Most investors look at a fund manager's returns before deciding
to invest. The question is: Just how much can we judge a fund by
its track record?
Several studies published last year suggest that picking a fund
based on its past is far from a sure thing. Among other findings,
the studies show that fund ratings don't help identify future
performance, that it's difficult to distinguish a manager's skill
from luck, and that for all but the very top funds, future
performance is unlikely to meet expectations.
In other words, it's a messy and largely unpredictable business.
But that doesn't stop investors from choosing their funds based
purely on track record.
"When you look at the data, there's a pretty slight and often
infrequent relationship between past and future performance, and
yet people still put an awful lot of stock in a fund's
performance," said John Rekenthaler, vice president of research at
Morningstar Inc. "In their hearts, most people believe that it
works."
Top Dogs Fall Down
Advisor Perspectives, an e-newsletter for financial advisers, in
December published a study which suggests that Morningstar Inc.'s
star ratings, which are based on past returns, don't provide much
predictive value. It also found that many five-star-rated funds
were likely to underperform their peers.
Robert Huebscher, chief executive of Advisor Perspectives,
randomly selected a fund with a particular rating, and then looked
at how that fund performed against randomly selected funds with
lower-star ratings from the third quarter of 2006 through the third
quarter of 2009. He found many cases where the lower-rated funds
were more likely to outperform those with higher ratings.
Yet, despite those findings, "the public pour money into funds
that get higher ratings," Huebscher said.
Part of the problem is how fund companies sell themselves, he
said. "The most concrete thing a fund can sell is its historical
performance, so they say, 'We did well in the past and you should
think that we'll do well in the future.' "
Matthew Morey, professor of finance at Pace University's Lubin
School of Business, conducted several studies on the predictive
value of Morningstar ratings. He took issue with some of
Huebscher's methodology, particularly the way it lumps all mutual
funds into five broad categories--U.S. stock, international stock,
taxable bond, balanced and municipal bond--and compares funds
within these broad categories. That doesn't account for the variety
of approaches that funds take, he said.
A better method, Morey said, would be to look at funds' relative
performance within each of Morningstar's 48 categories. And, he
said, Huebscher's study acknowledges that it doesn't include funds
that were liquidated or merged away during the period--a fact that
likely boosts the performance of the worst funds, since it's rare
for a four- or five-star fund to close.
Morningstar's Rekenthaler questioned the time frame the study
used. But while he disagreed with some of the details, he said he
didn't have an issue with the notion that it's hard to use past
performance to predict future results.
Luck Versus Skill
Morey said there's a case to be made that top managers can
repeat their performance.
"I do think good managers, who've been at a fund a long time,
produced consistently good performance and charge relatively low
fees are a good way to go," he said.
Morey added that about 10 years ago he was a firm believer in
index funds, but has recently come around to the idea that there
are some good managers out there.
A Morgan Stanley Smith Barney's Consulting Group study seems to
support that view. The study found that the very best managers--
the top 10%--can repeat their performance; at least, they did in
three-year periods from 1994 to 2007 covered in the study. But the
study also found that the worst 20% of performers are also likely
to outperform in the future.
The study's author, Frank Nickel, said this is because the
bottom funds benefit from changing market cycles--their holdings
were out of favor but then get hot and thus outperform. Nickel said
that rather than picking a fund based on manager ability--which his
study suggests does exist--he'd rather pick an unloved area of the
market and ride its moves to the top.
"I like buying things that are out of favor because they're
likely to be cheap," Nickel said in an interview. In other words he
thinks a surer way to pick funds is by market cycles rather than
manager, even a good manager.
Nickel, as well as Pace University's Morey, said that even funds
with good managers may not be wise investments in the future.
That's because the popularity of successful funds means over long
periods of time they often grow so big that their managers can't
repeat their performance: The large number of assets in the fund
makes it too hard to put all that money to effective use.
But while Nickel was willing to ascribe some outperformance to
manager skill, some researchers dispute even that claim.
A study published late last year by Eugene Fama, professor of
finance at the University of Chicago Booth School of Business, and
Kenneth French, professor of finance at Dartmouth College Tuck
School of Business, ran 10,000 simulations of what investors could
expect from actively managed funds. They found that outside the top
3% of funds, active management lags results that would be delivered
due simply to chance. In other words, aside from the top 3%,
there's no way of knowing whether performance is based on luck or
skill.
Fama is one of the pioneers of the efficient-market hypothesis--
the idea that securities are priced correctly because they
incorporate all the known information relating to a company--and so
it fits his views to suggest that very few people can consistently
beat the financial markets. But that doesn't lessen the relevance
of his research. Nickel said that, while he hasn't seen Fama and
French's study, it's possible that the consistency his study found
among the top 10% of funds were delivered by the top 3% of managers
who could be said to have skill.
Less Efficient Markets
Morey said markets have become less efficient in the past 10 to
15 years, thus creating an opportunity for good managers. He
pointed to increasingly unclear company accounts, the fact that
derivatives positions aren't fully reported and the recent failures
of the bond-ratings agencies as examples of how hard it's become
for investors to get full information about companies.
"We're more in the dark now than we were 20 years ago," he said.
In such a world, a good manager can make a difference, he
added.
Morey named T. Rowe Price Group Inc. (TROW) as an example of a
firm with good managers. It's not just about performance, but the
fact that T. Rowe seems to encourage and develop its managers, who
typically stay for many years.
"There's definitely a 'T. Rowe culture,' " said Rekenthaler,
echoing Morey's sentiments.
Rekenthaler said he'd also include American Funds, a unit of
Capital Group Cos., and Vanguard Group as fund companies that seem
to have the right culture--not just in manager development, but by
charging relatively low fees and having fairly conservative fund
strategies, as well as not launching funds in popular sectors just
to grow assets.
"I've moved in my thinking from being enamored with a fund to
liking a fund company," he said. "You need to surround yourself
with the best-quality of people."
So will investors look more at a fund firm's culture rather than
a top performing manager? Fund flows suggest it's unlikely.
"It's what we call the hot-hand fallacy," said Hersh Shefrin,
professor of finance at Santa Clara University's Leavey School of
Business, and a pioneer in the field of behavioral finance.
Think about watching a basketball game. If a player hits several
baskets in a row, we think he's hot, that he's more likely to make
his next shot. But statistics show that a player who's made his
recent buckets isn't more likely to make his next one.
"Our intuition is misguided," said Shefrin. "If we don't know
the odds and try to estimate them we tend to overweight recent
events."
Instead of taking a chance on an unpredictable fund manager,
Shefrin suggested investors review their entire approach.
"There are two needs we're talking about here, financial and
psychological," he said. "For your financial needs the best
long-term approach is buy-and-hold with a well-diversified set of
securities with [passive] bond and stock holdings that reflect your
risk profile.
"For your psychological needs--and we're all thrill-seekers--
just recognize it for what it is and protect yourself," Shefrin
said. "Treat your [hot stock] choices the way you would if you went
to Las Vegas--you don't go expecting to make money."
(Sam Mamudi writes for MarketWatch. He can be reached at
415-439-6400 or by email at: AskNewswires@dowjones.com)