Level One: Saved by Intel

Analysts say the purchase of Level One by Intel came right when the company was facing a slowdown in business.
Author:
Publish date:

SAN FRANCISCO -- When Intel (INTC) - Get Report announced last Thursday it would spend $2.2 billion to buy network chip maker Level One Communications (LEVL) - Get Report, Adams Harkness & Hill analyst Tim Kellis didn't take the news sitting down.

"I fell out of my chair," Kellis says.

Kellis downgraded Level One earlier that day from strong buy to accumulate, driving the stock down 16% to 27 1/8. Bailing out of the stock seemed an unfortunate move only a few hours later, when Intel said after the market close that it would pay the equivalent of $49 for each Level One share.

Rather than writing it off simply to rotten timing, Kellis defends himself by taking a humble tack. Since he doesn't work at a bulge bracket firm, Kellis says, he doesn't have the clout to cause the stock to drop as much as it did. Rather, it dropped because people looked into his suggestion that Level One's business was slowing and found it to be true.

Even the weight of Intel won't change that. The stock rebounded the next day based solely on Intel's announced purchase price, not on Level One's financials, he says. "Most people who look at what is going on agree with me on the downgrade," he says.

Kellis isn't the only one with concerns about Level One. Charlie Glavin of

Credit Suisse First Boston

downgraded Level One Friday and Drew Peck of

Cowen & Co.

is also bearish on the stock. If they're right, Intel might find market dominance of the networking chip market elusive. (Neither First Boston nor Cowen have an underwriting relationship with Level One or Intel.)

Based on its past history alone, Level One looks like a great buy for Intel. "Level One in its history has never grown less than 40% a year," says

NationsBanc Montgomery

analyst Clark Westmont, one of the biggest bulls on the company. "Last year they grew sales 68%." (Montgomery is an underwriter of Level One.)

But Level One won't likely see that kind of growth in 1999, says Glavin. That's because the company has lost ground in the market of telecommunications transmission chips which are geared for wide area networks, a market that accounts for 28% of its business. And it's facing brand new competition from

Lucent

(LU)

and

Broadcom

(BRCM)

in the market for local area networking chips, which makes up about 72% of its business.

As for growth expectations this year, roughly 40% of the new revenue generated is to come from two recent acquisitions -- that of

Jato Technologies

and

Acclaim Communications

, both of which make fast ethernet products. There's no guarantee that the integration of either company will go smoothly, Glavin says, noting that the acquisition of Acclaim is already behind schedule.

Meanwhile, Level One faces competition in the networking market from

National Semiconductor

(NSM)

, Broadcom, Lucent and ethernet pioneer

SEEQ

(SEEQ)

.

With its vast resources, Intel can shoulder a temporary slowdown in networking. "Level One is a pimple as far as Intel is concerned," he says. Long term, if Intel can connect all the pieces it has bought --

Shiva

, a developer of remote access technology,

Dayna Communications

, a maker of small business networks and

Case Technology

, which makes network packaging -- it will be a formidable player in the market.

And that is something Intel needs to do, Peck says, because it's becoming ever more clear that the microprocessor market isn't going to yield the fabulous riches it once did. The days of 30% revenue growth for Intel are over. Expectations this year are for growth of 15%.

"In the long run, communications components are an attractive market," Peck says. "It may not be as profitable a business as microprocessors once were, but it will be more profitable than microprocessors will be. If ever there was a time to start planning for the future, this is it. The microprocessor market ain't what it used to be."

As originally published, this story contained an error. Please see

Corrections and Clarifications.