|
PureBytes Links
Trading Reference Links
|
http://dai.multexinvestor.com/article.asp?docid=9667
The Earnings Contrarian
Out of the casino 2002-07-12
How to profit from widespread misuse of earnings information
by Marc H. Gerstein, Director of Investment Research
Visions and revisions
Before diving into the topic at hand--contrarian approaches to
earnings information--I have a confession to make: I'm no seer. I
can't tell you what any company will earn in the next quarter or the
next year. And I'm not alone. If analysts could really foretell
earnings, estimate revisions would be the exception, not the rule.
(Note that whatever I say regarding analysts' ability to forecast
applies equally to the corporate executives who guide them.) But as
of this writing, our database has current quarter earnings estimates
for 3,853 companies--and fully 84 percent of them have been tweaked
at least once over the past three months. So, in fact, changes to
earnings estimates are the rule, not the exception.
What analysis can tell us
Despite my limitations as a fortune teller, I can tell you
confidently that next month will be August. That's not a matter of
knowing the future in advance. It's an assumption based on what I
know about the present and past, including the fact that our culture
agreed a long time ago that we'd mark the passage of time according
to specific conventions and that at the end of the period we call
July, a new period we agreed would be called August will commence.
The facts that underlie my assumption are exceptionally stable;
there's no advocacy whatsoever for changing the way we mark time.
That's why I can say, without looking like a Cable TV psychic, that
next month will be August.
We do something similar when we predict corporate earnings--although
the assumptions we use are far less stable. We recognize that we
can't know ahead of time what will happen. But we can and do analyze
the past and present in such a way as to enable us to make reasonable
assumptions about what will happen down the road. I may look at a
housing-related company and study the way ebbs and flows of consumer
spending, savings rates, mortgage rates, population etc. have
impacted its sales in the past. I could then say, given what we
expect from the indicators going forward, I predict the company will
generate revenues equal to such-and-such. It's not necessary that the
historic relationships be stable. I may note that over time, the
company is generating more revenue from a particular set of economic
conditions than it did when comparable conditions prevailed at
earlier points in time. That might lead me to forecast the company
will do better in this business cycle than in earlier ones.
Naturally, I can make expense-related assumptions by engaging in a
similar process. How much of a workforce has the company historically
needed to support a particular level of sales activity (and are these
relationship improving or deteriorating)? What is the inflation
picture? How much raw material is needed, and what cost issues are
observable?
I may have a good handle on the facts upon which these assumptions
rest, so I may be confident in my forecast. But as good as I might
feel in such a situation, I know for sure that the underlying facts
are not nearly as stable as those which underlie my assumption that
next month will be August. How volatile are my facts? That depends.
For some companies in some industries, the volatility is modest. For
others, it's huge. (If you want to see how huge, send a resume to a
brokerage firm and tell them you want to be an airline analyst. Even
the living legend Warren Buffett once attributed a poor airline
investment to "a case of sloppy analysis.")
Part of the factual instability issue rests on the business cycle.
Remember my housing stock example. I mentioned a variety of economic
indicators I'd need to utilize. In fact, these are the end results of
a different sort of forecasting process engaged in by another
profession, economics. Economists will tell you it's a lot easier to
forecast when things are following a regular trend. But when the
trend is abnormal, as it is during downturns, it's very hard for them
to get a confident handle on the facts that underlie their
assumptions.
What you're seeing
Many in the market today are, or at least were, spoiled. They got
into it somewhere during the course of what was a nearly 20-year bull
market. We had some trend breaks along the way. But on the whole, a
certain set of trends was in place far more often than it was out of
place. So many got the idea that forecasting was easy. Analysts were
expected to predict earnings to the penny; if not, something was
badly wrong. Companies were expected to give perfectly accurate
guidance; if they didn't something was badly wrong.
Guess what. That was a fairy tale; a nice fairy tale, I'll admit, and
a pretty long one, but a fairy tale nonetheless. Now, we're all
getting hit with a harsh dose of reality. Forecasting earnings is
never a simple process, not for the analysts, nor for the executives
who guide them. Bull markets are great for making life look easier
than it really is. We had one for a long time. Now, we don't.
The result is that many are getting increasingly angry, distressed,
and downright frantic about all these negative surprises and
revisions. And many are reacting the only way they know how: by
selling stocks. The fact that some corporate insiders allegedly tried
to prolong the fairy tale with improper accounting adds to the
negative sentiment.
It doesn't have to be this way
Here is the most important thing you will ever learn about stock
analysis. You do not, I repeat, do not have to sell when earnings
disappoint, nor do you have to buy when earnings come in above
expectations.
Company valuation is a complex and fascinating topic and I touched on
it in a series of articles last winter. (Visit our Education Home
Page to see links to these articles.) Earnings are, of course,
extremely important, but in all cases, we're talking about a stream
of earnings that persists over a prolonged period. And given that
we're talking about the great unknown, we recognized that imprecision
is inevitable (and discussed how models requiring precision fell into
disuse). Moreover, we're never talking about a single quarter. And to
the extent we label an earnings trend good, bad, or neutral, you can
bet we're not doing so based on whether the numbers match
analyst/company predictions.
Business cycles can be unpleasant. But like death and taxes, business
cycles are a regular part of life. We can prolong the good and
minimize the bad, but the world still has not found a way to abolish
the cycle. This has a vital implication for investors. Companies,
even the best of companies, will perform badly form time to time.
The cycle isn't the only obstacle. Sometimes, a company will perform
badly, even though conditions are good, because it makes a mistake.
Yes, companies make business errors from time to time, even the best
of the best.
If you seek companies that never experience a bad spells and never
give guidance that turns out to be faulty, my advice is to stick CDs,
because you are going to wind up flitting about from one stock to the
next never finding what you want and continually selling low when
reality rears its head. And there's just so much of this you can do
before you find your portfolio getting dangerously close to zero.
Before you buy a stock, it's important that your level of conviction
about the company extend way beyond whether or not you think it
can "hit its numbers" over the next quarter or two.
The misguided notion that companies ought to always get it right--or
be punished when they don't--stems from an earnings momentum style
that became increasingly prominent after the 1960s. The idea was to
buy shares of companies that are doing well and avoid those that
aren't. The system, although theoretically ridiculous--it made no
allowance for valuation--actually worked very well for a long time.
Shares of companies with good earnings trends performed a bit better
than the market averages, and shares of companies with weak earnings
momentum underperformed the market. Investors who were good at
figuring out which companies and/or industries were likely to
experience especially good or bad earnings momentum tended to thrive.
The system cracked in the late 1990s, and has since crumbled under
the weight of its own success. Too many investors did the same things
with the same stocks at the same time in response to the same
earnings news. Now, companies whose earnings are weak don't see their
shares "underperform." Instead, these shares are trounced, stomped,
and pilloried--to the point where valuations are so far out of
proportion to underlying fundamental merit that it would be laughable
if the pain of the style's ill-advised adherents weren't so real, and
wide-spread.
That alone would be bad enough. But in fact, today's reality is much
worse. At first, the momentum game was based on the most recently
reported set of numbers. Then, as Wall Street research departments
grew in size and competence, the gem turned its focus away from the
recent numbers toward expectations for the next set of results. And
then, it really careened over the edge by comparing earnings, less to
the overall trend, and more to estimates. Companies that do well see
their shares get hammered if they do less well than analysts expect.
Companies with rotten numbers get rewarded if their numbers are less
dreadful than analysts feared. If this strikes you as more of a
casino (where you gain or lose chips depending on whether earnings
predictions are accurate or inaccurate) than an investment market,
I'd say you are correct.
You can succeed
The above may sound depressing. But if you think about it, you can't
buy low and sell high unless somebody out there is selling low. (And
you can't sell high unless you can find someone else willing to buy
high.)
We're now in our quarterly earnings season, the time when companies
release earnings for the latest quarter. This is a time when
investors who play the casino game overreact to the things they hear.
(For the record, the numbers to be reported will, for the most part,
cover the quarter that ended 6/30/02. But most of the action in the
stocks will be based on guidance relating to the upcoming 9/30/.02
period.) Considering where we are in the business/market cycle, a lot
of these overreactions are likely to be on the down side. That means
you're going to get a lot of chances to buy low. And you'll often
find yourself forced to decide whether or not you should sell low. My
suggestion: do the former, avoid the latter.
This does not mean you should naively buy or hold every stock that
gets hammered due to bad earnings news (or guidance). What it means
is that you should step back, take a deep breath, and do a calm,
objective analysis. If things feel too hectic on reporting day, then
wait a bit. Stock investing is far more contemplative than Hollywood
would have you believe. Pardon the pun, but you'd be amazed at how
often you can be burned by things that are hot off the press. Cool
down, then do your homework.
By all means, pay attention to the earnings information. But do not
worry about whether it met or exceeded expectations. Compare the
earnings to the company's overall fundamental picture, with an eye
toward determining whether the problem is transitory or potentially
permanent. If the latter looks like the case, then sell or avoid the
shares. But if the situation is just temporary, wait it out. I've
seen all the nonsense about how it takes a 100 percent gain to offset
a 50 percent loss. Anybody who has any reasonable level of experience
in investing knows full well that these supposedly rare 100 percent-
plus recovery gains happen quite often. All you have to do is peruse
price charts to see it.
If you study analysts research reports, look for whether the theme
is "if" or "when." The dangerous situation is where the analyst has
serious reservations about the "if" question. But if the analyst is
bothered by issues of "when," that is an encouraging signal.
This is not a new or innovative way to use earnings information. It's
the way we're supposed to use it; as an updated report card and a
prompt to focus our attention on key bigger-picture questions. The
other, more common, way to use earnings data (as the functional
equivalent of a roulette ball falling into the slot that tells you
whether you won or lost your expectations wager) is the aberration.
Successful investors recognize the difference, and tend to stick to
the basics.
----------------------------------------------------------------------
----------
Marc H. Gerstein joined Market Guide in 1999 as Director of
Investment Research.
He started working as an equity analyst in 1980 and covered stocks
spanning a wide variety of groups including Household Products,
Retail, Restaurant, Hotel/Gaming, Media, Natural Resources,
Homebuilding, Conglomerates, and Transportation. He also managed a
high-yield bond mutual fund and conducted seminars teaching investors
how to select stocks using screening software.
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Save on REALTOR Fees
http://us.click.yahoo.com/Xw80LD/h1ZEAA/Ey.GAA/zMEolB/TM
---------------------------------------------------------------------~->
To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|