|
5.3.3
Morningstar, Forbes’
“Mutual Fund Honor Roll,” and mutual fund advertisements
are some of the most well-known sources of investment information used
by manager pickers. Examining these sources shows why superior performance
cannot be correlated to past superior performance; because the future
cannot be predicted.
Morningstar, a well-known independent investment research firm, is one
of the many sources of information used by manager pickers. Mutual funds
are currently assigned stars based on three, five and 10-year performances,
risks and fees. When available, the 10-year performance has a 50% weighting,
while the three year has a 30% and the two year has a 20% weight. While
the long-term overall star rating formula seems to give the most weight
to the 10-year period, the most recent three-year period actually has
the greatest impact because it is included in all three rating periods.
Morningstar holds that the star ratings are designed to be a starting
point for investors and that the ratings themselves are not predictive
of future performance. Its goal is to help the individual investor make
better decisions. Nonetheless, since they are commonly contained in mutual
fund advertisements, it is my belief that many investors look to the stars
as a guide.
In a July 15, 2003 report on investor behavior, Dalbar stated that, “Motivated
by fear and greed, investors pour money into equity funds on market upswings
and are quick to sell on downturns. Most investors are unable to profitably
time the market and are left with equity fund returns lower than inflation.”
The report goes on to state that the average holding period for equity
mutual funds was a little more than two years. The average manager picker
is not grasping the concept that past performance has nothing to do with
future performance.
An April 5, 2005 report by John Waggoner, of USATODAY.com, illustrates
the problem Dalbar has identified. Waggoner tracked the asset flows into
the then highly rated Fidelity Aggressive Growth Fund, which had $23 billion
in assets in March 2000. Investors poured 65% of those assets, $15.1 billion,
into the fund the 12 months before the S&P 500 peaked in March of
2000. So, a $10,000 investment back in March 2000 would be worth $2,697
in April 2005—that’s a 73% loss.
Other once hot funds that have subsequently performed badly include Janus
Worldwide, once a $13-billion fund that has fallen 45% over the past five
years; Nasdaq 100 Trust, once a $10- billion fund that has fallen 67%,
and Janus Global Technology, a $10-billion fund in March 2000 that has
since plummeted 73%.
In total, investors put $228 billion into the 50 best-selling stock funds
in the 12 months before the market peaked, yet only two of the top 50
funds have shown a gain over the past five years: American Funds New Perspectives,
up 2.3%, and Vanguard Capital Opportunity, up 1.5%. In total, the active
investors who plowed money into the 50 hottest-selling funds five years
ago are down an average 42% since March 2000, according to Lipper Analytical
Services, a well-known mutual fund data firm. (Further details of this
active investor madness is recounted in the book American Sucker, by David
Denby.)
Another example of the difficulty of picking a winning fund manager is
found in “Selling the Future: Concerns About the Misuse of Mutual
Fund Ratings,” a May 16, 1994 study conducted by Lipper Analytical
Services, a well-known mutual fund data firm. In the study, Lipper selected
highly rated mutual funds from Morningstar at the beginning of a year
and then measured their performances in the following 12 months against
mutual fund averages. This study was conducted in four subsequent one-year
periods: 1990, 1991, 1992, and 1993. The study found that the majority
of highly rated stock mutual funds underperformed mutual fund averages
in each of the four subsequent years. This means that investors who select
mutual funds from the list of highly rated funds can often end up in the
wrong mutual funds at the wrong time. This not only demonstrates the unreliability
of investing based on past performance over a period as short as one year,
it also shows how consistently unpredictable mutual funds can be in outperforming
the market. The results of the Lipper study are depicted in Table
5-16.
The study also found
that at the end of 1990, after a long period of superior performance by
foreign-oriented mutual funds, only 32% (25 out of 77) of the total number
of highly rated stock funds were listed in the “international”
and “global” fund categories. Predictably, many investors
jumped into these funds, believing that their past superior performance
would be repeated in the future. Not surprisingly, every one of these
25 international and global funds subsequently underperformed the average
stock fund in the following 12 months. At the end of 1992, after foreign-oriented
mutual funds performed poorly for a year, no international or global funds
appeared on a highly rated funds’ list. Few investors were attracted
to these international and global funds because they were at the bottom
of the pile. The result: investors missed the superior performance of
international and global mutual funds that began at the end of 1992.
Morningstar, for its part, released a new star-rating system in July of
2002. The old system compared the historic risk and returns from a mutual
fund with the risk and returns of a broad group of funds. The new star
system attempts to compare a fund with a much smaller group of funds of
a similar style. According to a March 2004 paper published in The Journal
of Financial Planning by William Reichenstein, Ph.D., Morningstar’s
decision to change rating systems reflects a decade of studies on the
importance of investment style in explaining stock returns. As discussed
several times in this book, a Fama and French study in 1992 concluded
that stock returns vary systematically across two dimensions: size and
value-growth. Actually, Fama and French came to this conclusion after
completing two different studies. First, the researchers focused on returns
from a period from 1963 to 1990. In a later study they conducted along
with their colleague Jim Davis, the researchers looked at returns from
a longer period—1929 to 1997. The researchers discovered that stock
returns can best be explained when stocks are separated into portfolios
based on size as measured by market capitalization and value-growth as
measured by book/market ratios. Many studies over the past decade have
confirmed and reinforced Fama and French’s conclusion that returns
vary systematically across size and value-growth dimensions. This type
of fund analysis was pioneered by Dimensional Fund Advisors (DFA), where
Fama is the director of research. Fama also has written extensively on
the random walk and efficient market theories and is one of the world’s
most cited economists. DFA uses a Fama/French designed factor regression
analysis to show that active managers’ returns are attributable
not to skill, but to exposure to these risk factors. Consistent with the
Fama and French research, DFA offers no actively managed funds, but has
a complete assortment of passively managed index funds.
The Forbes Mutual Fund Honor Roll is hailed by the media as a dependable
way to find superior performing mutual funds. Each year since 1973, the
highly respected Forbes magazine has singled out 15 to 30 stock mutual
funds and elevated them to Forbes Honor Roll status. These funds are selected
on the basis of their total returns over at least a 10-year period, the
stability of their investment management over at least seven years, and
their relative performance in both bull markets and bear markets over
several market cycles.
A comprehensive 1992 study by John Bogle titled “Selecting Equity
Mutual Funds” examined the record of the Forbes Honor Roll covering
the period of 1974 to 1990. The study sought to answer two questions:
(1) did Honor Roll mutual funds continue to beat comparable non-Honor
Roll funds in subsequent years during the 1974 to 1990 time period and
(2) did Honor Roll funds continue to beat the market in the ensuing years
during this time period?
To answer the first question, the study found that there was a virtual
tie in performance between the Honor Roll funds and the average stock
mutual fund in subsequent years during this time period. To answer the
second question, the study found that after commission loads were taken
into account, the Honor Roll funds subsequently underperformed the market
by a significant amount over the 1974 to 1990 period. The cumulative returns
of the Honor Roll funds was 439.7% and the cumulative return of the market
was 633.4%. That’s a difference of 193.7%.
The study found that mutual fund winners from the past significantly underperformed
the market in the future for several reasons. For one thing, the superior
performance generated by an active fund manager’s investment style
is dependent on the time period in which the market favors that style.
Since the stock market unpredictably favors different investment styles
for different time periods, a manager’s past superior performance
is closely tied to a past time period in which the market happened to
favor his or her kind of investment style.
A good example of the link between superior performance and a certain
time period is found in the 1983 Forbes Honor Roll. In that year it contained
a large number of small company stock mutual funds because small company
stocks had generally outperformed large company stocks over the previous
six or seven years. In the following years beginning in 1984, small company
stocks began a dismal run in performance relative to large company stocks.
As a result, small company stock funds began to drop out of the Honor
Roll after 1983. Although the small company stock funds in the 1983 Honor
Roll had outstanding performance histories, their returns in 1984 and
in the following years were inferior on average to the market and to the
average stock mutual fund.
The Forbes Honor Roll study reached two conclusions: (1) investors can’t
pick a future winning mutual fund based on its past performance and (2)
over the long run, even highly rated active funds underperform their respective
benchmark.
Mutual fund advertisements are another source investors turn to when manager
picking. Unfortunately, they convey this false message: “Since these
funds have done well in the past, they will do well in the future, so
buy them today.”
Mutual fund advertisements carry an SEC mandated disclaimer stating: “Past
performance is no guarantee of future results.” There is a reason
why the SEC requires this — it’s true! According to Dalbar
it appears that many investors act as if this disclaimer is not true at
all. They continue to buy and sell mutual funds based on short-term past
performance falling for the implied message of mutual fund advertising.
5.3.4
There is one other
point regarding the futility of attempting to identify skillful money
managers. An old investment proverb observes that “markets make
managers.” This means that if the market favors a money manager’s
particular investment style anyone can achieve outstanding performance.
Markets can make a money manager look good or bad — a factor that’s
independent of their “skillful” stock picking or market-timing
abilities. An active money manager that an investor selects will usually
turn out to be a winning or a losing manager because of the behavior of
the market itself, rather than the manager’s skill at picking stocks
or timing markets. Active money managers play a game that’s almost
entirely random in conferring long-term investment success among them.
5.3.5
There are at least
three other problems associated with manager picking. For one thing, investors
are seldom aware that active funds or separate portfolios that have good
performance histories are always riskier than the indexes they outperform.
According to Modern Portfolio Theory, any portfolio of investments
that hold fewer stocks than the index in which it is invested must be,
by definition, underdiversified relative to that index portfolio. It follows
then that any mutual fund or separate portfolio that has turned in a market-beating
performance achieved it by holding investments that somehow were different
in kind or amount from those of the relevant index. Any mutual fund or
separate portfolio that boasts a superior performance history must therefore
be riskier.
A mutual fund manager with recent performance success has bet money and
concentrated it in specific stocks or bonds. The bet may pay off, but
people are too blinded by the “brilliant investment insight”
to understand that the bet was too risky in the first place. Peter Lynch,
the legendary manager of Fidelity’s Magellan mutual fund, concentrated
about 25% of the fund’s holdings in foreign stocks in the 1980s.
These stocks turned out to be top performers, and Magellan widely outpaced
the S&P 500. The irony is that these stocks weren’t even represented
in the S&P 500.
Lynch’s performance was not measured against an appropriate benchmark
comprised of a proportionately weighted mix of U.S. and foreign stocks.
It was measured against the wrong benchmark, the S&P 500. Using an
appropriate benchmark would have reduced, perhaps even eliminated, his
successful performance during this period. Lynch’s bet was nevertheless
deemed a winner by popular acclaim, and he was widely hailed as the leading
investment guru of the decade.
Had Lynch’s bet turned out wrong and Magellan underperformed, Lynch
would have been widely criticized as a fool for making such a risky bet.
Right or wrong, it was still a risky bet because Magellan had a greater
amount of diversifiable risk than was represented in the benchmark by
which it was measured.
There are two lessons to be learned from this. First, any active investment
strategy is inherently risky, but is not considered risky in hindsight
if it turns out to be a winner. Second, a mutual fund’s outstanding
performance history is nothing more than the market’s reward for
exposure to excessive investment risk. Due to the unpredictable nature
of the market, the same excessive risk that produces outstanding performance
today can turn and produce miserable performance in the future. Once the
market begins to favor sectors other than those a manager is invested
in, his or her luck has run out.
Yet another problem with manager picking is that outstanding performance
histories can be surprisingly fragile. Few investors realize that the
most important factor separating a winning performance history from a
losing one is the choice of starting and ending dates. Fidelity’s
Magellan beat the S&P 500 for the decade ending in mid-1995. Lengthening
the ending date by one year to mid-1996 would have painted a very different
picture. Fidelity’s Magellan underperformed the S&P 500 for
that 11-year period.
Lastly, outstanding performance histories don’t always reflect taxes
or commission loads. Published mutual fund ratings are often pre-tax returns
that disguise their true after-tax performance in taxable accounts. Fidelity’s
Magellan generated an average annual pre-tax return of 18.3% over the
10-year period from mid-1985 to mid-1995. Once the taxes and commission
loads were factored in, the net return dropped to 12.7%. At first glance,
this fund appeared to widely outperform the market. A closer look reveals
that Fidelity’s Magellan came very close to underperforming it.
However, an investor may never know this because mutual fund advertisements
often feature only pre-tax and/or pre-commission load returns. Tax-adjusted
returns are now available from Morningstar on the Internet at www.morningstar.com.
Morningstar’s tax-adjusted returns only account for federal income
taxes, but not state income taxes. Investors should also consider that
state income taxes need to be deducted in order to see a complete picture
of how all taxes impact investment performance, especially relative to
a tax-efficient index fund.
The inception date for the Magellan fund was May 1, 1963. For a 47 year, 11 month comparison of the Fidelity Magellan Fund to the IFA Indexes and Index Portfolios, see here.
5.3.6
Indexes such as the S&P 500 or Wilshire 5000 are often used to evaluate
the performances of active money managers. Given the Fama and French findings,
the use of such benchmarks is often misleading. Because these indexes
are weighted heavily towards large company stocks and high priced stocks,
the performances of managers investing more heavily in small company stocks
or low priced stocks won’t be accurately measured by them. Instead,
customized benchmarks are needed to provide accurate measurements of the
contributions to performances made by active money managers.
The Fama and French Three Factor Model is a superior way to evaluate the performances
of active money managers. It shows whether a manager achieves returns
in excess of index returns. After all, an active manager shouldn't
be rewarded just for buying value stocks—that’s something
that can be done inexpensively with an indexing strategy.
The place where a portfolio is positioned or structured on the cross hair
map in Figure below determines the vast majority of its return. The cross hair
map doesn't plot the market risk factor since all stock portfolios
take similar market risk and are plotted relative to the stock market.
So, there’s no need for a separate axis; instead, the stock market
sits right at the cross hairs of the map. The cross hair map has two dimensions.
The size dimension is plotted along the vertical axis, and the value (BtM) dimension
is plotted along the horizontal axis. The axes represent exposure to these
two risk factors. Portfolios that take on a lot of size risk appear higher
along the size axis, and portfolios that take on a lot of value risk appear
further along to the right on the growth/value axis.
 Throughout the IFA Website, there are many assertions regarding the expected returns of different asset classes, and there are many warnings against attributing outperformance to active managers based on a few years of hot returns. One could easily make the mistake of thinking that IFA is merely expressing opinions without any mathematical support. In fact, however, nothing could be further from the truth. IFA has always relied on the scientific method of statistical analysis.
Perhaps the single most important application of statistics lies in the realm of hypothesis testing. Virtually all of the experimental and observational scientific studies (across all fields) take the approach of proposing a “null hypothesis” and then either rejecting or failing to reject the null hypothesis, based on the probability of making the recorded observations, assuming that the null hypothesis is actually true.
The specialty within economics known as “asset pricing theory” lends itself quite readily to hypothesis testing because for every asset (or asset class), there is usually a set of historical returns that can be tested. A common test would propose the null hypothesis that the expected return of the asset class is no different from the risk-free rate of return. Given a series of historical returns, we can calculate a parameter known as the “t-statistic” which will give us a quantitative indicator of the probability of observing these returns under the assumption of the null hypothesis. A higher value of the t-statistic indicates a lower probability of the occurrence of the observed returns, and vice versa. The t-statistic is commonly used to construct a 95% confidence interval around the observed mean, and if this confidence interval does not contain the value assumed in the null hypothesis, then we can reject the null hypothesis with a 95% level of confidence. Generally, a 95% confidence level is associated with a t-statistic of 2.
The value of the t-statistic depends on 3 separate parameters: The number of observations (N), the average of the observations, and the standard deviation of the observations. The t-statistic is directly proportional to both the square root of the number of observations and the average of the observations. It is inversely proportional to the standard deviation, so the more volatile the asset class, the less likely we will be able to draw firm conclusions.
To see exactly how the calculation of a t-statistic operates, we will use the returns for the last five calendar years of the top three mutual funds used in 401k plans (according to Brightscope.com )
The three funds to be evaluated are:
- AGTHX - American Funds Growth Fund of America (benchmarked against the S&P 500 Index)
- PTTAX - PIMCO Total Return Fund (benchmarked against Barclays Aggregate Bond Index)
- AEPGX - American Funds EuroPacific Growth Fund (benchmarked against the MSCI EAFE Index)
The results are summarized below: (1/1/2006 to 12/31/2010) |
| Fund |
Average
Excess Return |
Standard
Deviation |
T-Statistic |
# of Years Needed to get a T-stat of 2 |
| AGTHX |
0.73% |
5.60% |
0.29 |
235 |
| PTTAX |
1.82% |
3.38% |
1.20 |
14 |
| AEPGX |
2.54% |
4.96% |
1.15 |
15 |
For all three of these funds, we are unable to reject the null hypothesis that their expected returns are no higher than their benchmarks at a 95% confidence level. 401k Plan sponsors who have incorporated these funds based on short-term performance have no right to be surprised if the future returnscaptured by plan participants fall short of expectations.
The interactive chart below has the formula for a t-stat of 2, based on return and risk values built into the chart. As you roll your mouse to a coordinate of return and risk, the line that highlights represents the number of samples needed to obtain a t-stat of 2. A data box representing each point on the line provides the 3 values for that position.
FigureT-stat Chart
Here is the formula for the t-stat and a calculator that will allow you to do your own calculations. Don't believe the data without a t-stat of 2. If the calculator freezes up, refresh the page.
FigureT-stat Calculator
In the next example, we will use the t-stat calculation to evaluate an asset class that has seen an enormous increase in popularity in the last few years, commodities. Using 19 years of calendar year returns data for the Dow Jones UBS Commodity Index, we will test the hypothesis that the expected return of commodities is no different than the risk-free rate (One-Month T-Bills).
1/1/1992 to 12/31/2010 (19 years) |
| |
Average Excess Return |
Standard Deviation |
T-Statistic |
# of Years Needed to get a T-stat of 2 |
| DJ UBS Commodity Index |
5.20% |
18.94% |
1.20 |
53 |
Again, we are unable to reject the null hypothesis at a 95% confidence level, so those investors who have poured their hard-earned money into commodities (or commodity futures) might be setting themselves up for bitter disappointment.
Lastly, we will examine IFA’s statistical justification for tilting a portfolio towards the factors of small cap and value. Based on 83 years of data, we will test the hypothesis that the expected return of the IFA US Small Cap Value Index is no different than the IFA US Large Company Index.
1/1/1928 to 12/31/2010 (83 years) |
| |
Average Excess Return |
Standard Deviation |
T-Statistic |
| IFA US Small Cap Value Index |
5.29% |
16.64% |
2.90 |
Since the t-statistic is greater than two, we are able to reject the null hypothesis and conclude that small value has a higher expected return than the large blend segment of the market. In this case, however, we can offer an explanation that small value stocks are riskier and thus should carry a higher expected return. In general, no conclusion should ever be drawn from data alone, because as we all know, if the data is tortured for long enough, it will confess to anything. It is crucial to have a sound explanation for the observed data.
Although it is possible to find actively managed funds that have shown outperformance with a t-statistic greater than 2, IFA strongly cautions investors against throwing their money at these managers even when there appears to be a statistical justification for doing so. The reason, quite simply, is that since there are thousands of active managers, by chance alone, we expect to see some that have outperformed their benchmark after expenses. The problem is that the number that we actually do see is, in fact, no higher than what we would expect from chance alone (i.e., it is no higher than what we would observe if all active managers were monkeys throwing darts at the Wall Street Journal) . This means that when an active manager appears to exhibit outperformance, there is no reliable way to determine if it was due to luck (i.e., a false positive) or skill. Two papers that elegantly address this point are:
- “False Discoveries in Mutual Fund Performance: Measuring Luck in Estimating Alphas” by Barras, Scaillet, and Wermers which evaluated 2,076 fund managers over 32 years and found that 99.4% of active fund managers showed no genuine stock-picking ability.
- “Luck versus Skill in the Cross Section of Mutual Fund Alpha Estimates” by Fama and French which evaluated 819 actively managed funds over 22 years and found that 97% could not be expected to beat a risk-appropriate benchmark.
IFA has always encouraged investors to obtain as much education as possible so that they can make informed decisions. Most investors will find that having a good understanding of statistics is incredibly helpful. Whenever they come across an advertisement such as “Fund XYZ beat its 5-year Lipper average”, they would do well to ask, “What is the t-statistic behind that number?” Odds are, it will not be included in the advertisement, and investors should not waste their time or their money on such spurious claims.
5.4
The solution for manager
pickers is to stop being fooled by randomness, stop believing in Santa
Claus, and give up the hope that a fund manager can be selected in advance
to consistently beat a market in the future.
5.5
Statisticians have
stated their case saying they need at least 20 years worth of risk and
return data to establish skill in a manager. The real problem is choosing
those managers at the beginning of the period. Therefore, index funds
are a far better choice for investors because of their 80-year
track records.
5.6
1. Statisticians tell you that you need a minimum number of years
of performance data on mutual funds to draw conclusions about future risks
and returns. How many years are required?
a. 1 year
b. 5 years
c. 10 years
d. 20 years
2. The problem with picking a manager to beat the appropriate
index is that:
a. they can’t pick next year’s winning stocks
b. they can’t pick the best time to be in or out of the market
c. they can’t determine which style of investing is the best
d. there is no persistence in manager performance
e. all of the above
3. A Dalbar study found manager pickers changed their managers
every:
a. 7 months
b. 5.3 years
c. 2.6 years
d. 15.5 years
4. There are overlooked factors when investors review the past
performance of managers. They include:
a. improper benchmarks
b. after-tax returns in taxable accounts
c. exact same time periods
d. commission charged on the purchase of the fund
e. all of the above
5. According to the mutual fund tracking service, Lipper, the
top 50 hottest selling mutual funds in March 2000 were reviewed again
in March 2005. On average, the top 50 funds had a total change in value
of:
a. up 83%
b. down 42%
c. up 5%
d. down 10%
e. up 22%
|
