Time for another poll: Agree or Disagree with What's the Rush?
biz.yahoo.com
Thursday September 26, 3:35 pm ET
By Russ Mitchell
This article was originally published on SmartMoney Select on 9/19/02.
SEMICONDUCTOR STOCKS SUNK to new lows this week. They can't fall much further, some analysts believe. It looks like bargain time for bottom fishers.
But be warned: The semiconductor industry is undergoing fundamental change, and low prices won't matter much if chip stocks are stuck in the mud.
The biggest danger for chip-stock investors is repeating what's worked in the past: buy into semiconductors low in a cycle, watch the stocks rocket at first sign of a business pick-up and stay high for years. Historically, semi stocks rise quick and strong at the nearly imperceptible start of a fresh cycle, says Banc of America Securities chip analyst Mark Fitzgerald. The biggest gains are often made in hours or days. Be out of the market, or hesitate just a bit, and the sweetest opportunities vanish fast.
Investors "don't want to miss the three- to five-fold return," says Fitzgerald. "It's 20 years of learned behavior." That behavior won't work, not for the next several years at least. "Unfortunately," he says, "most investors haven't adjusted their thinking."
Stocks of many chip makers and semiconductor-equipment manufacturers are indeed at or near bottom he believes. But, he concludes, so what?
Look at demand: For 15 years or more, the personal computer was the chip industry's prime driver. The PC boom started in the mid-1980s and reached its peak a couple years ago; now it's flatlining, and a return to previous growth rates is unlikely, bordering on impossible. The next biggest market is communications; growth there might return, but with overcapacity so gross, it will be a long time coming. Add PCs and their peripherals to communications chips, Fitzgerald says, and you get 70% of semiconductor revenue. "I don't see demand rising for most of the sector," he says. He projects industry growth at 5% to 10%, several years out.
Drew Peck, a veteran chip analyst now with the Capital Ventures investment firm, is as pessimistic as Fitzgerald. Maybe more so. "Perhaps there are a few bargains here," he says. "But there's still plenty of room for future deterioration in fundamentals. The real danger is people stepping in with the assumption that things can't get worse."
The dim future for PC growth is bad news for Intel (NASDAQ:INTC - News); it's no surprise that Intel stock now trades at a five-year low. But anyone who thinks Intel is a bargain better think twice, Peck says. "The problem with Intel is that the company is in the midst of a dramatic change of fortune. The margins are going to look very different." The company draws heavy press for its move into analog and mixed-signal chips, the most promising market for chip makers. But competition will be fierce. "The Intel monopoly is over," according to Peck. "A lot of investors haven't picked up on that yet."
Analog chips and mixed-signal chips (analog-digital) are widely considered a semiconductor investor's best bet. It's taking far slower than originally advertised, but higher-bandwidth media delivered through computer networks is beginning to take off for consumers. Because images and sound are analog by nature, chips are needed to convert them to digital, and back to analog again. They're essential for everything from multimedia cell phones to DVDs to HDTV to TiVo (NASDAQ:TIVO - News).
Consumer devices don't seem so hot right now. Earlier in the week, Prudential Securities lowered earnings estimates for many chip stocks, partly based on poor projections for device sales this Christmas. Because of the bad economy, Peck says, the chip market is in the "happily rare" state where all chip stocks suffer at once. The analog niche will emerge as the best performer once recovery comes, he says. "Those stocks really may have hit bottom."
Market growth isn't the only attraction for analog stocks. They also boast high margins — when customers are buying. Unlike memory chips and microprocessors, which are stamped out by the millions, analog chips tend to be more customized, made in smaller lots. Customers pay more for a chip that's just right for their product. Manufacturing costs are lower, too. A wafer fabrication plant for analog chips might cost $100 million or so, compared with, say, Intel's $2.5 billion microprocessor factories.
Among the analog companies, Fitzgerald advises sticking with "quality names." Those include Linear Technology (NASDAQ:LLTC - News), Maxim Integrated Products (NASDAQ:MXIM - News) and Semtech (NASDAQ:SMTC - News). Peck says Texas Instruments (NYSE:TXN - News) is also worth considering. It's big and broad-based, but it's far less dependent on PCs than, say, Intel or Advanced Micro Devices (NYSE:AMD - News). Converting digital to analog and vice versa is a TI specialty. "Among the big companies it's very well positioned," says Peck. If I were going to buy a big name, Texas Instruments would be on top of the list — but that's not saying much." (Fitzgerald doesn't own shares of Linear, Maxim or Semtech; Banc of America doesn't perform investment-banking services for those companies. Peck doesn't own shares of Texas Instruments; Capital Ventures doesn't perform investment-banking services for the company.)
No, it's not saying much. Any semiconductor recommendation, it seems, comes off half-hearted these days. Even investors in analog companies should prepare for a long haul — maybe several years long. Ultimately, they've got a good chance of paying off. Ultimately.
There's another way to make money in chip stocks right now. Because so many chip-stock true believers are poised to leap from the sidelines, some piece of positive news will spark a major rally. "It'll go off like a hand grenade," he says. "Stocks will be bid up quickly, and they'll be very tradable at those levels." But unlike past years, the rally won't be able to sustain itself, Fitzgerald says.
The greater-fool theory is a risky way to invest. But so is investing in semiconductor stocks at all, even if you get them cheap.
Russ Mitchell is a veteran technology journalist based in San Francisco. |
