Profunds index and sector trading

[frenchee]

OT: Advice to Bottom Fishers: Cut Bait

By ERIC J. SAVITZ

RACKING MY BRAIN FOR A GOOD STORY idea recently, I recalled a conversation I'd had with a portfolio manager about Credence Systems, a leading player in test equipment used by semiconductor makers. Credence looked insanely inexpensive: With the stock around $7, and estimates for the company's October 2005 fiscal year averaging north of a buck, the P/E stood in single digits. Cheap, right?

But on the very day I started my research on the company, it announced earnings for its fiscal third quarter ended July 31... and unveiled a disturbingly gloomy forecast for its fiscal fourth quarter, putting earnings at 5 to 10 cents a share, versus the Street consensus at about 25 cents. The crux of the matter was a sudden and sharp slowdown in orders from Taiwanese contract chip manufacturers. No matter, I thought, the stock will go down, and look even cheaper. Then I checked in with my portfolio manager friend. Turned out, he'd already dumped his shares. So I did more checking, and abandoned ship.

The clincher came from Jim Covello, Goldman Sachs' outspoken chip-equipment analyst. For starters, he blew away any fantasies I had about Credence's earning a dollar in fiscal '05: He now expects profits of just a nickel a share. More importantly, Covello makes a sobering case that the equipment sector, though already depressed, is headed for another down leg.


The Chips Are Down: The Nasdaq Composite dropped 0.9% for the week, to 1844.48. Intel slashed its third-quarter forecast, overshadowing better-than-expected jobs data. Chip maker Altera also cut its quarter sales outlook.


The semiconductor industry, he says, is expected to expand current production capacity 13% in 2005. But semiconductor unit sales, Goldman estimates, will rise only 7% in 2005. (And revenues will go up by less than that, given expected price declines.) The inevitable conclusion, Covello says, is that capacity-utilization rates in chip making are headed south. The decline is already under way, he adds, but will likely worsen in the next few quarters. Now running a little north of 90%, Covello figures capacity utilization will drop to the mid-to-high 80s. And when that kind of thing happens, orders for semiconductor manufacturing and test equipment inevitably go south.

Think he's wrong? Then re-read what Intel said at its mid-quarter earnings update last week. The big chip maker dropped its revenue and margin forecasts for the quarter, citing weak end demand and rising inventories. The news prompted a Friday selloff in chip stocks. And Intel's problems aren't all company-specific; Altera issued a mid-quarter warning Thursday afternoon.

The question is how long -- and how severe -- the downturn will be. At least for the equipment makers, Covello says, it "will either be more severe or last longer" that the Street generally expects. He notes that every chip-equipment downturn since 1982 -- there have been four -- has lasted two years.

So if you're tempted to bottom-fish, Covello has some advice: Resist. "We would avoid overweight positions in the group," he says diplomatically. So would he own any of the equipment stocks? "Not if I didn't have to," he replies.

Many analysts feel the same way about chip stocks, too. Merrill Lynch's Joe Osha observed in a Friday morning research note that, while Intel is suffering from overcapacity, it isn't likely to slow its construction of new cutting-edge chipmaking plants. Rather than mothball older but still viable factories, he figures, Intel will go after non-microprocessor markets. Concludes Osha: "The prospects for the graphics, chipset, and flash-memory companies hoping to compete with Intel look grim in 2005."