maudlin letter
More Lies, Damned Lies and Statistics
Trust the System, Luke
The X Factor
The Nigerian Connection
Short Selling Rally
Retirement and Demographics
And Much More!
By John Mauldin
Today we look at why oil prices are likely heading down, how
brilliant economists come to the wrong conclusions and of course a
few comments on the recent market action. Let's jump right in.
More Lies, Damned Lies and Statistics
Since I criticized the trade policies of President Bush last week,
maybe I can work up enough nerve to critique a recent article by one
of the more well-known economists in the country, Arthur Laffer, of
Laffer Curve fame.
Last week, there was a Wall Street Journal op-ed piece by Laffer
which was so over the top in its market analysis that I found it
hard to believe the Journal would publish it. Except for the fact
that it was written by someone as famous (and brilliant) as Laffer,
I doubt it would have seen the light of day. I hesitate to critique
someone of Laffer's stature, but as I mused on it over the weekend,
I realized it was a perfect illustration of how statistics can be
very misleading. As investment analysis, this article was of dubious
value. As a teaching tool, it is most excellent.
To cut to the chase, Laffer did a set of calculations to show that
the market was undervalued by 48%. That was on Monday July 15, as
the Dow was at 8700 or thereabouts. Presuming he wrote the article
a week earlier, by his statistical analysis, the Dow should be at
13,000+. As the Dow dropped below 7600, I assume by the same
calculation this made the market undervalued by 75%! Abby Joseph
Cohen is downright bearish by comparison.
Analyzing how he came to this conclusion is instructive, and is
useful to illustrate how we should evaluate investment models. For
new readers, I spend a great deal of my professional time analyzing
investment managers and the systems they use. Naturally, I focus
more time on firms that have good track records than those who have
poor track records. The trick is to look past their track records
and to analyze what they did to attain their performance.
Trust the System, Luke
How many times have you invested money with a manager or fund, only
to have them go into the tank just after you invested? Studies show
that this is a very common experience.
That is why every investment offering you read has the words "Past
performance is not indicative of future results." It is more
important to know why a fund made money than how much it has made.
Most investment managers use some type of system: "When indicator A
is in the seventh house and Indicator B is aligned with Mars, we
short the market" or "We buy only value stocks defined by certain
set of criteria" or "Our proprietary trading model looks for short
term opportunities."
The manager/trader has found some connection between a certain set
of variables that he feels will produce positive and consistent
results. I can't tell you how many times I have seen graphs which
show near perfect past correlation between certain sets of variables
and market direction.
Condition A will produce Effect B. Trust the System, Luke. These
managers will tell you they are disciplined traders and promise to
always follow their system.
The X Factor
Then the system blows up, shortly after you give them money. What
happened? It could be several things. First, the variables could
simply be random coincidence. I read once of a study of hundreds of
pricing variables that showed the best predictor of the S&P 500 was
the price of butter in Bangladesh. If the study isn't real, it
should be. There are so many statistics used be economist to show
some connection between the markets and certain events. They have no
more real connection than the price of butter in Bangladesh or of
tea in China has with the US stock market. But there are any number
of variables which for a time could seem to track with the market.
Secondly (another possibility), there was a missing variable in the
connection between Condition A and Effect B. While A and B seem to
move together, there is a third variable, let's call it the X
factor, which is outside of the system and is the real connection
between A and B.
As an example, many advisors told us in January of 2001 that lower
interest rates would make the stock market rise. "Look," they
pointed out, "stock markets always rise when the Fed starts lowering
the interest rates. Now is a good time to buy." Condition A - the
Fed lowering rates - will lead us to Effect B, a rise in stock
prices. I specifically remember Jim Cramer making that argument, but
he had a lot of company.
Except that the X factor was the already over-valued stock market
and a coming recession. The X factor was a more important influence
on the direction of the market than Condition A.
As a third example, it could have been "being in the right place at
the right time." Just as every dog has its day, no market stays down
or goes up forever. The managers will tell you it was due to their
superior system, but they happened to be in a market that was going
up. Think Janus 20. Here was a group of people who said with a
straight face that the powerhouse results of their fund were due to
their superior skills at selecting tech stocks. It had a lot more to
do with their being invested in a bubble.
Finally, it could have been sheer luck. Some traders have simply
flipped heads 7 times in a row. They then put on a suit, call
themselves a hedge fund or a commodity pool and invite you to invest
in their coin flipping skills. They will tell you they have a
proprietary system. It is always a proprietary system. With few
exceptions, I simply walk away if a manager will not open his black
box and show me how his system works.
Now, let's get back to Mr. Laffer's treatise. Basically, he
constructs a new measure of market value he calls capitalized
economic value. He takes the profits from 5,000,000 US companies and
essentially divides them by the average ten year US treasury note
yield. Supposedly, this is what an investment banker would do to get
the price at which he would offer the shares of USA Inc. to the
market.
He shows a graph of this statistic compared with the S&P 500 and
tells us, "Just reviewing the plot of the two series illustrates the
remarkable closeness of fit." And indeed, with a few exceptions,
which he notes, the two lines do seem to closely follow each other,
up until recently, as the market has dropped. He makes the argument
that the market will therefore rise to once again coincide with his
capitalized economic profit (CEP for short).
As an investment advisor using this system, he would tell you to
jump into the market buying with both fists, then go to the golf
course and work on your game while the market rises. Don't worry
about day to day fluctuations. His system clearly indicates that the
market is going up, and rather dramatically.
The question is whether we should give some of our hard earned money
to a manager which uses this system? Is it really valid? Let's
analyze his assumptions.
Let's set aside for the moment that pessimists would tell you the
implication of the graph is that profits are going to come crashing
down to the current stock market price. That is too quick and easy.
First, let's look at the variables in his system. He starts with the
profits from 5,000,000 US companies, gathered by a reliable source
from government data. I will cede the accuracy of the data. But I am
not concerned with 5,000,000 companies. I am concerned with a few
thousand public companies.
As noted recently, private companies have little reason to pump up
their earnings they report to the US government. For a variety of
reasons I have noted in my last three columns (see the archives at
www.2000wave.com), the earnings of US public companies are suspect,
and are not likely to rise at anywhere close to the current
estimates of mainstream analysts.
But let's assume, for fun, that this number is representative of the
earnings of public companies. Then we look at the second variable,
the interest rate. The lower the interest rate, the higher the his
new CEP will be.
The theory is that if rates are low, I should be more willing to
invest in stock, and should be willing to pay a higher price for a
given stream of earnings from stocks.
And I ask, "Why?" Isn't this a version of the "invest in the stock
market because the Fed is cutting rates" argument? That system
didn't work too well last year.
Further, if this model is truly what an investment banker would use
to calculate the value of a stock, given the appalling track record
of investment bankers of late, why should I care what they tell me
the value of a stock is?
More to point, why should I not get a risk premium for the
uncertainty of a stock value against the certainty of a return on my
US bond? Maybe I am missing something, but Laffer seems to be
telling me I should be willing to value a stock at the same risk
level as I do a US bond. This is the same reasoning that led
Glassman and Hassett to write their book called DOW 36,000. They
also had an op-ed piece in the Wall Street Journal.
There are a lot of variables that come between the connection
between a CEP and the value of the markets. One that comes to mind
is the difference between risk and uncertainty. As investors, we
take risks. But we like to be able to quantify those risks. When
things become uncertain, we usually take less risk.
Investor sentiment or consumer sentiment is a measure of the belief
of the public in the level of uncertainty they feel about the
future. The more certain you are about your personal economic
future, the more willing you are to take some risk.
One would only be willing to value a stock like a bond in times of
great certainty. If you absolutely knew you could count on 15%
compound earnings growth for the next 20 years, you would be foolish
to invest in a bond which pays 5%. But the fact is, you can't be
certain, and the less certain you are, the more you value the
smaller but guaranteed income stream.
The direct one to one relationship between interest rates and market
values is tenuous. An indicator which uses interest rates as its
primary component of predicting future market values will be at best
very volatile and at worst could be disastrous, especially in
deflationary periods, which we may be heading into.
In essence, Laffer is telling the market to conform to his view of
what fair value should be; based upon a model he thinks is rational.
While perhaps not as futile as trying to stop the tide, getting the
market to conform to your view of rational is quite difficult.
As a market predicting system, I think I will pass on Laffer's CEP.
When someone approaches you with an investment fund, and shows you
pretty past performance graphs, always check the cause and effect
connection between the past performance and the statistics. If you
can't figure out the rationale, it might be because there isn't one.
(Full disclosure requires that I admit that on more than one public
occasion I have been fooled by statistics. I have invested and lost
money based upon past historical performance systems that seemed at
the time to be reasonable. Just because I am now a skeptical analyst
doesn't mean that I can't be burned. I must confess that I have also
used pretty graphs of past performance, and will do so again. It
comes with the territory.)
The Nigerian Connection
There is trouble brewing in the oil patch, and I think we could see
a drop in oil prices over the next few years.
The price of oil is delicately balanced between supply and demand.
World demand is roughly 70,000,000 barrels per day. If demand were
to drop by 1,000,000 barrels, or supply increase, then in a short
time there would be a glut of oil on the market, with nowhere to
store it. Prices would begin to drop.
Traders are fixated upon oil reserves in storage, and short term
market moves can be dramatic if reserves rise or fall more than
expected. With that background, let's look at developments in the
oil patch.
There is a clear dispute between the Oil Ministry and the Finance
Ministry in Venezuela. The Finance minister shows a budget which is
predicated upon increased oil production, well beyond what their
OPEC limit is. The Oil Minister says, "No way. We are sticking to
our quotas."
Given the upheaval in that country, and the clear need for cash,
care to make a bet who wins that debate?
Dennis Gartman tells us of reports that Nigerian officials are
making noise that they want to leave OPEC. This is possibly a ploy
to pressure OPEC to allow them to raise their production levels, but
it has caused some concern in OPEC circles.
I ran this by a high ranking government official involved in the oil
industry (the proverbial un-named source), and he informed me that
Nigeria is privately negotiating with the US to become a very direct
supplier of oil to the US, and is looking for US companies to be
involved in increasing their oil production.
This makes sense for Nigeria and for the US. It would also put
pressure on supplies, presumably because Nigerian wants to increase
its revenues.
Finally, demand for oil is weakening, as growth in the world is
slowing. Greg Weldon tells us that Japanese demand is down sharply,
for instance.
With Russia bringing on new oil reserves, the downward pressure on
oil prices is going to increase. This will cause some of the poorer
countries to cheat in order to keep much needed revenues stable.
Cheating on quotas is the one constant in the oil industry.
Barring the abrupt demise of oil production (think Iraq), I think
the clear momentum from recent events is that energy prices will be
going down over time.
Short Selling Rally
As predicted last week, we saw a nice rally yesterday, clearly
helped along by short selling. We will probably see more of these
quick updrafts, as for the first time in history, the public has
more open short interest than institutions and traders. These
rookies are not yet familiar with how fast traders can push a stock
with too much short interest. Shorting can be a good investment
technique, but it is also dangerous if you don't know what risks you
are taking.
Retirement and Demographics
I am looking for serious economic studies that analyze what the
effects of too many people trying to retire all at once can do to an
economy, and specifically what the equilibrium point is between too
many retirees and too few workers. I have been meditating upon the
coming retirement of the Baby Boomers, and it seems to me that it
will not be possible for all of us to comfortably retire over the
next 25 years without some degree of social discomfort. I want to
investigate this, but find very little in the way of academically or
statistically sound work. I can find lots of conjecture. If you know
of a book or paper, I would like to hear of it.
Next week, I hope to have a study finished that I am working on with
Steve Blumenthal of CGM on the true value of the NASDAQ after new
accounting changes are factored in, and where this index might be
going. It should make for interesting and controversial reading.
I am off to Orlando in a few minutes with my son. That is why you
are getting the e-letter a day early. After a day there in an air-
conditioned office reviewing statistics, my son and I will be out
and about. If someone would please turn on the air conditioning at
the theme parks, I would appreciate it.
Your getting ready to sweat off a few pounds analyst,
John Mauldin
John@2000wave.com
Copyright 2002 John Mauldin. All Rights Reserved |
