|
 |
| Price |
7.810 |
|
| Change |
-0.220 |
| Research
Wizard
Add
to MSN Stock List
Message
Board
|
 |
| Price |
47.000 |
|
| Change |
-0.310 |
| Research
Wizard
Add
to MSN Stock List
Message
Board
Related Resources
Build
a screen of real estate stocks with our Stock
Screener
What
stocks are hot or not?
What
are the top-performing
industries?
The Speculator
Recent articles:
• Buy companies
buying their own stock, 4/18/2002
• A good
news/bad news book for optimists, 4/11/2002
• 10 rules to
squash the market competition, 4/4/2002
More...
|
|
 |
| sponsored
by: | |
The Speculator
Don't settle for 1,500,000%
profit
U.S. stocks returned that mammoth number in the past 100
years, but our studies show you can do even better with proper
stalking and patience. Also: why 164% of your portfolio should be in
stocks.
By Victor
Niederhoffer and Laurel Kenner
One dollar invested in U.S. stocks at the end of
1899 came to $15,000 at the end of 2001 with dividends plowed back
in. This 1,500,000%-a-century return was by no means atypical:
England’s was similar, while Sweden’s and Austria’s were even
greater.
These
amazing results are among the investment themes reported and
documented in full in the masterpiece we reviewed in our April 11
column, "The Triumph of the Optimists: 101 Years of Global
Investment Returns", by Elroy Dimson, Paul Marsh and Mike Staunton.
(See "A
good news/bad news book for optimists.")
Ideas found in
books of this greatness require savoring, elaboration and
experimentation, and we’ve been doing this with the 1,500,000%
theme.
We wondered whether you need to be in the market all
the time, or whether a little rhythmic variation, so usefully
employed in many other areas, might improve returns. Drawing
inspiration from the classical composers and the Botswana bush, we
invented two variations that may increase the effectiveness of a
long-term investment strategy. We also invited a good colleague,
trader Alix Martin, to join in the improvisation, and we report his
variation below.
In the classical era, variations on
well-known themes were a vehicle for improvisation by such
composer/performers as Mozart and Beethoven. Variations require
finesse in tone, dynamic and tempo from the performer and challenge
the listener to follow the tune in various manifestations of rhythm
and meter. The themes are generally simple and regular (i.e., four
bars, eight bars, 16 bars), but with a twist, to give the listener
the pleasure of recognition.
Far from the classical world, in
the African savannah, as Laurel recently had the pleasure of
observing, elephants spend 22 hours of the day chomping their daily
quota of 330 pounds of grasses, acacia bark and thorn bushes, only
5% of which actually provides nourishment. Cheetahs have developed
an extreme variation of mealtime strategy; they may stalk an impala
for hours before going for the kill in a 70 mph, 300-yard sprint.
They then can wait five days for the next opportunity. The art of
stalking and waiting seems particularly appropriate for investors
seeking improved returns.
Variation
No. 1: A stumble, a skip and a jump
If Cole Porter had put
the “Triumph” theme to the music of “Night and Day,” it might have
gone like this:
Like the steady support of a lifelong friend
Like the
flow of Victoria Falls
Like the steady state of electric
flow,
Like the steady rise of the Dow
Like the ching, ching,
ching of the cash flow
As the ships come into dock
So a
voice within me keeps repeating,
stocks, stocks,
stocks. Some readers might object that, far from being
steady, the stock market has ups and downs. Yet when you look at the
graph of what a buck invested at the start of 1899 would have come
to by 2001, the declines look like little blips -- even the move
down from 1929 to 1933. The crashes of 1987 and 2000-2001 are barely
visible.
Even
so, we found that the 1,500,000% return may be magnificently
improved by buying in the second year after any down year. From 1899
through 2001, negative returns occurred in 26 years. The return two
years later averaged 16%, with 23 years up and just two years down.
The standard deviation was a modest 18%.
The correlation
between returns in one year and the return two years later turns out
to be negative 0.25. That's very significant for 100 years of data,
with a probability of less than 1 in 100 of occurring by chance.
Stalking a better return
| Results of buying stocks in 2nd year
after a fall |
| Year |
Total return (%) |
Two years later |
Total return |
| 1903 |
-15% |
1905 |
22% |
| 1907 |
-29% |
1909 |
20% |
| 1910 |
-9% |
1912 |
7% |
| 1913 |
-8% |
1915 |
39% |
| 1914 |
-6% |
1916 |
6% |
| 1917 |
-19% |
1919 |
21% |
| 1920 |
-18% |
1922 |
31% |
| 1929 |
-15% |
1931 |
-44% |
| 1930 |
-28% |
1932 |
-10% |
| 1931 |
-44% |
1933 |
58% |
| 1932 |
-10% |
1934 |
4% |
| 1937 |
-35% |
1939 |
3% |
| 1940 |
-7% |
1942 |
16% |
| 1941 |
-10% |
1943 |
28% |
| 1946 |
-6% |
1948 |
2% |
| 1957 |
-10% |
1959 |
13% |
| 1962 |
-10% |
1964 |
16% |
| 1966 |
-9% |
1968 |
14% |
| 1969 |
-11% |
1971 |
18% |
| 1973 |
-19% |
1975 |
38% |
| 1974 |
-28% |
1976 |
27% |
| 1977 |
-3% |
1979 |
26% |
| 1981 |
-4% |
1983 |
23% |
| 1990 |
-6% |
1992 |
9% |
| 1994 |
0% |
1996 |
21% |
| 2000 |
-11% |
|
|
| 2001 |
-11% |
|
|
|
|
|
|
|
|
# of observations |
25 |
|
|
avg. return |
16% |
|
|
median return |
18% |
|
|
standard deviation |
19% |
|
|
% up |
92% | |
Guess
what? Two years ago, the stock market’s return was a negative 11%.
That’s very bullish for 2002. Our best guess, based on the
correlations, is up 12% this year. We're throwing caution to the
winds and putting our money in stocks.
As Porter might have
put it:
Up and down, that’s how to trade.
You buy after a down
year, sell after a gain.
Just make sure you skip a
year
Whether up or down when trading, dear
We’ve tested this
every way.
Variation No. 2:
Lucky 5
Seasonality is always a suspect in stock returns.
The authors of "Triumph" explore whether there are good months to
buy and sell. They review whether January is a good month to buy and
conclude that it might show a high return for U.S. small caps. They
believe the market may tend to rise in the summer -- the opposite of
the conventional wisdom on the “summer slump.” Such issues, the
authors suggest, will “continue to intrigue.”
They certainly
intrigued us, enough so to inspire a study of whether certain years
show high returns. We explored whether years ending in a particular
number, such as 5, might be good times to buy, and whether years
ending in other numbers, such as 0 or 7, are not so favorable.
As the table below shows, the average return for years
ending in 5 is 32%. The average return for years ending in 0 or 7 is
close to 1%.
| Average returns for years ending in 0
through 8 |
| Year Ending |
% Return |
| 0 |
1 |
| 1 |
6 |
| 2 |
10 |
| 3 |
10 |
| 4 |
12 |
| 5 |
32 |
| 6 |
11 |
| 7 |
0 |
| 8 |
25 | |
To
test whether these rather intriguing results were non-random, we put
all 100 returns into an urn. Then we chose 10 groups of 10 returns.
We repeated the process 1,000 times.
Next, we computed the
difference between the largest and the lowest returns. This turned
out to be above 30% just 2% of the time.
Thus, we conclude
that years ending in 5 are great times to buy, and years ending in 0
or 7 are bad times to buy.
Studies of the best years to buy
bonds, bills and the Consumer Price Index show completely random
results.
Variation No. 3:
Leverage
Alix Martin, a Paris-based student of statistics
and consultant to many European telecom companies, offered the
following variation on the 1,500,000% percent theme:
Wall
Street pundits typically recommend an asset allocation split between
stocks, bonds and cash. But over the last century, an investor who
would have had 70% in stocks, 20% in bonds and the remaining 10% in
T bills would have turned $1 into “only” $5,070, one-third of what
one would have made by being fully invested in stocks.
Is it
possible to improve on the “100% in stocks” strategy? It may not
have been in 1899, but it is today. One can buy stocks on margin, or
buy futures, to increase exposure to stock returns beyond 100%. If
one had had a 200% exposure to stocks during the last century,
instead of turning $1 into $15,000, he would have made
$1,723,781.
Is too much of a good thing possible? Yes: With a
300% exposure, our fictional investor would have lost everything in
1931. Even a winning blackjack player can go bust if his bet is too
large relative to his bankroll. The same phenomenon is at work here:
gambler's ruin.
However, an investor or a fund trying to
implement this strategy would face additional costs: he would have
to pay margin rates to be allowed to hold more stocks than he has
the cash for, and would need to buy or sell stocks at the end of
each period to maintain his exposure at the desired level.
To make the study more realistic, Alix subtracted T-bill
returns from stock returns for the part of the exposure that goes
beyond 100%, as an approximation of margin rates. Even with this
more restrictive hypothesis, additional exposure improves the fully
invested strategy. A 150% exposure would have more than tripled the
gains over the century, turning $1 into $47,228.
With
hindsight, the best asset allocation would have been 164% stocks.
This turns a dollar into $50,950, nudging up the compounded return
from 10% to 11.3%.
But that would have been a wild ride: from
1928 to 1932, one would have lost more than 90%. And anyone daring
enough to muster more than 225% exposure would have gone bust.
Nobody said this wasn't risky.
Final
note
We mentioned in our April 18 column that only 13
S&P 500 companies had announced stock buybacks, and said we
would buy all 13 -- but we listed only 11 names. An amazing number
of readers caught the disparity, and we thank them for their
attention. The two missing companies were
Xerox (XRX,
news,
msgs)
and Viacom (VIA.B,
news,
msgs).
They should be added to the list we planned to buy and are now long.
We omitted them because they did not report complete information on
the number of outstanding shares and the number of shares to be
bought back.
Readers also were eager to point out that we
didn’t take account of such things as whether the companies actually
did buy shares back or whether the buyback was a mere replacement
for options, and queried us about how we took account of value. Our
answer is that with a 30 percentage-point divergence in favor of the
buyback group, a real outperformance which has continued for those
announced after March 30, 2001, and a normal deviation corresponding
to the outperformance of 6, we did not feel it appropriate to
bifurcate further.
Our April 18 column cited a seminal 1995
Journal of Financial Economics study by David Ikenberry based on
U.S. stock buybacks in the 1980s, but failed to mention Ikenberry’s
two co-authors. They are Theo Vermaelen, professor of finance at
INSEAD, a business school in France, and Josef Lakonishok, a finance
professor at the University of Illinois. The professors have
continued to explore the subject of buybacks, and Vermaelen kindly
brought us up to date on a subsequent study on Canadian markets
published in the Journal of Finance in October 2000:
"One
of the interesting things about Canada is that companies have to
report each month how many shares they have actually repurchased.
What we find is that (1) quite a few companies announce buybacks but
don't do them; (2) companies that did not repurchase stock in year
after the repurchase experience positive abnormal returns during
that year, but not in the two following years; (3) companies that do
repurchase stock don't experience abnormal returns in the year
following the repurchase, but do experience positive abnormal
returns in the next two years.
"The whole story is consistent
with the argument that companies use share repurchase authorizations
to take advantage of an undervalued stock price. If the market
becomes smarter immediately, they don't do the buybacks. If the
market does not correct the undervaluation, they go ahead and
complete the repurchase.
"The point is that we now have
three independent findings supporting the buyback strategy: (1) the
1995 paper in the Journal of Financial Economics; (2) the 2000 paper
in the Journal of Finance; (3) the results of the KBC equity buyback
fund([a fund that Vermaelen manages for a Belgian bank, KBC)] If you
add to this the results of David Fried's newsletter, you must
conclude that there is indeed a way to beat the market: exploit
LEGAL insider trading by managers who repurchase stock when it is
cheap."
Of course, as with all such fixed systems, the
Speculators would caution that these ideas are good only until the
regime shifts. However, we note that Professor Vermaelen has
spearheaded a fund that actually holds such shares, and its
real-time record has been quite good. He has kindly agreed to field
readers’ queries about buybacks directly through us so that, as has
happened in the past, we do not succumb to substituting our own
amateurish ideas for those of the expert in fields other than squash
and music.
Vermaelen’s 50-stock portfolio consists of
companies that cite undervaluation as the reason for buying back
their shares; he reports a return of 40% since the fund’s inception
in July 1998. Those who wish additional information may contact him
by e-mail.
We
still have available the worksheet of yearly returns from 1900 to
the present for stocks, bonds and bills kindly provided by the
Dimson Group. Kindly contact us at gbuch@bloomberg.net with your
suggestions, encomia and comments.
At the time of
publication, Victor Niederhoffer and Laurel Kenner owned or
controlled shares in the following equities mentioned in this
column: Xerox and Viacom.
|
