Analysts Divided on Intel's outlook
sjmercury.com
Posted at 3:33 p.m. PDT Sunday, May 10, 1998
Two heavy-hitter analysts divided on Intel's outlook
May 11, 1998
BY ADAM LASHINSKY
Mercury News Staff Writer
MERRILL Lynch & Co.'s Thomas P. Kurlak and Mark Edelstone of Morgan Stanley Dean Witter are having a $64 billion disagreement over Intel Corp. (Nasdaq, INTC).
One of these research analysts is going to be wrong, and anyone who cares about Intel, semiconductors, personal computers or tech stocks ought to be paying attention. What's more, their intellectual spat provides a rare window on how analysts affect stock prices through the sheer force of their recommendations.
Kurlak follows semiconductors for Merrill in New York and is perhaps the most prominent bear on Intel. Edelstone analyzes chips for Morgan in San Francisco and thinks people like Kurlak are crazy for abandoning Intel just because the semiconductor market is a little soft.
Does their little quibble matter? You bet.
Kurlak and Edelstone work for tremendously powerful brokerages with a vast clientele that carefully watches their stock pronouncements. It's also noteworthy because the two are so far apart -- their difference in estimated per-share earnings for 1999 is $1.35, or more than $2 billion. And they're staking their reputations on perhaps the most important tech-stock position of our day.
Kurlak thinks Intel's stock is heading south, from its current $84.06 to about $70, and possibly less. Edelstone sees a share of Intel trading for $110 within a year.
The difference in the valuations they're projecting, based on Intel's 1.6 billion shares outstanding, is roughly $64 billion. (Kurlak doesn't use a precise downside price target, but he suggested in a report that continued personal-computer slowdowns could drive down the stock to ''a new low . . . in the 60s.'')
Kurlak, 51, has worked at Merrill Lynch since 1978 and has a record of accurately calling near-term shifts in semiconductor stocks. He thinks Intel's in big trouble. He sees the PC industry having hit a saturation point where Intel can't keep up its game of charging top dollar for more powerful chips as prices plummet for less powerful ones.
Specifically, he worries that the advent of the sub-$1,000 PC will continue to erode Intel's profitability.
''We find it hard to believe that a market sector that hardly existed one year ago, and is now 27 percent of U.S. retail will not keep increasing,'' he told clients. ''Therefore, we expect Intel's average sales prices to remain under pressure.''
Kurlak thinks Intel, which earned $3.87 per share in 1997, will earn $2.95 this year and $2.90 in 1999. And he's not buying for a moment the company's guess that its business will turn later this year.
''Management's expectation of resumed growth in the second half seems to hang by a thin thread,'' he says.
Edelstone, 36, sees things completely differently. ''We're at a turning point for the company,'' he says, arguing that the Pentium II microprocessor will give the company new momentum later this year.
''Intel's going to grow a wall of glory right now,'' he says, adding that the company will ''distance itself from the competition'' and surprise Wall Street in the process, just as it did in 1996, after the last semiconductor-industry slowdown.
Edelstone is betting Intel will earn $3.25 per share this year and $4.25 next. For comparison, a consensus of 32 Wall Street analysts guesses Intel will earn $3.16 and $3.82 in 1998 and 1999 respectively, according to First Call Corp.
So if these two heavy-hitters are so far apart, how are investors supposed to judge? There's no easy answer, and even professional investors are split.
''If you look back at everything Kurlak has written about various stocks, he's had an annoying habit of being right,'' says Kevin Landis, a former semiconductor analyst himself who manages the Technology Value fund in San Jose.
Landis recalls buying shares of programmable logic device maker Xilinx Inc. (Nasdaq, XLNX) in 1994 as Kurlak's negative commentary drove down the stock from $60 to $30 (before a three-for-one stock split in 1995.
The stock went down in part because Kurlak made it go down. But Xilinx's decline also created a buying opportunity. Since the stock turned, it's grown more than fourfold in value.
That's how some would play Intel if its stock does head south.
''Kurlak is fundamentally right, but he underestimates the power of the Intel franchise,'' argues Howard M. Love Jr., who manages Love Capital in Burlingame. ''If the market were perfectly rational, it probably would be in the 60s.'' But that analysis fails to take into account Intel's strong following among retail and institutional investors and its monopoly power, says Love, who was an Intel buyer in the 70s. ''I was hoping it would get to 60 because I'd back up the truck.''
Investors far from Silicon Valley have a similar view.
''Even great companies like Intel have product cycles,'' argues Ben A. Hock Jr., research director for the John Hancock Funds in Boston, which holds Intel stock. ''Even Coca-Cola has product cycles. If you're trying to build a long-term position, you've got to own Intel.''
Meanwhile, both analysts are sticking to their guns.
Kurlak's most recent report to clients -- when Intel last traded for about $84 -- was titled, ''Expect Rally to Fail.''
For his part, Edelstone counsels keeping the faith.
''This is a classic example of people jumping off the train at the wrong time,'' he says, of those who'd abandon Intel now.
Check back in 12 months to see who's right.
--------------------------------------------------------------------------------
Contact Adam Lashinsky at the San Jose Mercury News, 750 Ridder Park Drive, San Jose, Calif. 95190 or siliconstreet@sjmercury.com or (408) 271-3782. |
