Thread - Bull/Bear arguments from SmartMoney....
August 26, 1998
THE BULLS have
rescued Intel
(INTC), driving it
up as high as $90
from a trough of
$65.65 earlier this
summer. But it
seems they can
carry it no farther.
On Wednesday,
the stock closed
near $83. And
even at that level,
Intel might be
overpriced. The
chipmaker is selling
at 21 times last
year's diluted earnings per share of $3.87. That's 30% higher
than the company's 11-year average P/E of 16, according to
SmartMoney's own calculations. And that's at a time when
profit margins are shrinking in the company's main business
of supplying semiconductors for desktop PCs.
Could Intel be set for another fall? A large faction on Wall
Street, including Piper Jaffray analyst Ashok Kumar, says
no. "Intel is becoming a one-stop shop for all the
[micro]processors that the industry needs," he explains.
"They are the engine of this industry. And we feel they
deserve a premium for that." Indeed. His price target of 105
would give the company a trailing P/E of 33 based on his
expectation that Intel earns $3.13 this year. That P/E is 1.65
times the company's expected earnings growth rate of 20%
over the next five years.
Of course, there are some stocks that command premiums
like that. The trailing P/E of networking leader Cisco
Systems (CSCO) is more than twice its 30% projected
earnings growth rate. But Intel's future doesn't look as
easygoing as that of Cisco. In order to combat the growing
market share of low-end PCs, the company must sell more
chips just to maintain its revenue. Consider that just a year
ago, the average selling price of Intel's chips was $250.
Today, it is between $208 and $212. Meanwhile, PC sales
growth has slowed. Last year, desktop PC sales increased
16% worldwide. In the first half of 1998, the figure was only
about 14%.
Merrill Lynch's Tom Kurlak has seen companies in this
position before. And the prognosis isn't good. "If you think
about Xerox (XRX) or Polaroid (PRD) or DEC [now part
of Compaq (CPQ)] or Data General (DGN), they've all
had their 30 years of life terminated by pricing pressure," he
observes. "[Intel's] facing that point in its life cycle. It's like a
midlife crisis, where they can't let price drive them into a
lesser role the way some of their predecessors have all gone.
They've got to get into new markets." With industry
dynamics like that, Kurlak doesn't believe Intel deserves any
premium. He says the stock is worth $65.
Things certainly look bad for Intel in 1998. The company is
projected to earn between $3 and $3.13 per share, down
about 20% from last year. How quickly things improve
depends on two things. First, the ability of Intel to increase
its market share in the sub-$1,000 PC market. And an
increase in sales of new, high-end products like the 450
megahertz Pentium II chip and the Xeon chip for the server
market.
On the bright side, PC sales growth is forecast to improve in
the second half of the year. Dataquest PC analyst Martin
Reynolds says growth could be more than 15% in the third
and fourth quarters of 1998, spurred by higher sales of
laptops and the new Pentium II computers. That should help
prevent Intel's gross margins from eroding further -- they're
already down to 52% from 60% or better in the past. But
most analysts agree that the burden is on new chips like the
Xeon to bring those margins back up to their historic levels.
The Xeon chip, which was just introduced this year, goes for
$1,100 a pop. That compares with a price of about $200
for a chip bound for a sub-$1,000 PC. But can Intel sell
enough of these chips to offset the profit decline in its PC
business? The market for chips for high-end servers is just a
fraction of that of desktop PCs. And it currently accounts for
just 3% of Intel's revenue.
No matter, says Scott Nirenberski of CS First Boston. First,
he explains, the market for high-end data servers is growing
at 30% per year and could potentially represent 10 million
units for microprocessors by the year 2000. At $1,000 per
processor, it would add up to a $10 billion market, of which
Intel could own up to 70%. That, he says, should add some
luster to Intel's price.
Says Nirenberski: "The premium for Intel's stock has been
inching up in the past few years, and I think it's both because
people understand this company is not going away and
because it's just become a greater component of the S&P
500." In other words, Intel's move into server products gives
it the kind of staying power one associates with blue chips
like General Electric (GE), currently trading at a P/E of
about 32, or double its secular growth rate of 14%.
All well and good, but despite the excitement over Xeon and
the anticipation of growth in PCs, neither development is
assured. And that's reason for caution. Greg Mischou of
Warburg Dillon Read writes in a recent report that Intel
should continue to trade in the $60 to $80 range for several
reasons. First, back-to-school sales of PCs and a buying
spurt in Europe are likely to be short-lived. Second, the
Xeon and Intel's other new products have yet to prove
themselves. Lastly, sales expectations for both the low- and
high-end products could yet prove unrealistic.
The folks who track the server-chip market are lukewarm
about Intel's prospects. Dataquest's Nathan Brookwood
expects the market for chips costing $1,000 or more to
shrink from the current level of $7 billion to about $6.5
billion by 2002 as Intel cuts prices. Intel will likely boost its
share from 30% to 45% of that market, he predicts, but its
dollar volume will remain steady. Can Intel win a larger
chunk? "It's doubtful," Brookwood says. "Sun's going to be
a competitor and those mainframes aren't going away."
Kurlak is also dubious. "Hopefully Intel can get to 8 to 20
million units [of server chips]," he says. "But I can't believe
they're ever going to sell 30 or 40 million units of any
microprocessor at $1,000. Without that market, their
revenue goes down." And don't count on new products like
DVD players and set-top boxes to make up the shortfall,
says Kurlak. "The PC is one of the few products that ever
got to a hundred million [units], other than the television," he
argues. "Even if Intel has 50% of some of these other
markets, you're talking about hundreds of millions [of dollars
in revenue]. They need billions."
Is Intel going to disappear? Not hardly. The company has
manufacturing prowess that continually enables it to
outproduce and underprice all of its competitors.
Nonetheless, the desktop PC market -- Intel's cash cow of
the past 15 years -- is running dry. Until it is clear that
management can replace those lost profits, the bulls will have
a hard time arguing this giant deserves a premium P/E.
-- By Tiernan Ray
John
|
