To: Jim Oravetz who wrote (643 ) 10/16/2002 11:45:59 AM From: Jim Oravetz Read Replies (1) | Respond to of 912 With Deteriorating Prospects,ARM's Shares Remain Pricey By DAVID REILLY Staff Reporter of THE WALL STREET JOURNAL When a stock blows up, there's a temptation for investors to give it a look simply because it seems so cheap. ARM Holdings PLC, a designer of chips for mobile phones and other gadgets, highlighted the danger in this approach Tuesday when it reported third-quarter results two weeks after shocking the market with a big revenue and profit warning. In the wake of the warning, there were a raft of downgrades by analysts, but there also was talk that the market had overreacted by driving the company's share price from about 125 pence ($1.95 or €1.98) to about 45 pence. The rationale was that while ARM's prospects had deteriorated, things weren't nearly as bad as all that. It turns out they are that bad, and ARM's shares could get cheaper still. In speaking about the results, ARM executives said they had grown slightly more pessimistic between the time of the warning and this week. That prompted their decision to lay off 10% of the company's work force by year's end, a move that would have seemed unthinkable just a few months ago. ARM also said its outlook for a flat fourth quarter could extend into early 2003. "Industry conditions have not improved at all; if anything industry conditions have deteriorated," Chief Executive Warren East said during a conference call. ARM's shares took another hit on this news, falling 13% at one point Tuesday, although they bounced back later in the day to close down 3.8% at 45 pence in London Stock Exchange trading. Even at this level, ARM's shares are trading at about 20 times 2003 earnings estimates, which have come down significantly since the start of the month. Merrill Lynch is even more pessimistic and Tuesday cut its 2003 earnings per share expectation to 1.79 pence. That results in a price/earnings ratio of about 25 times. Taking either of these multiples, ARM is still richly valued considering its growth argument has lost credence for now. License sales are likely to remain anemic well into 2003 and royalty revenue shows no signs of picking up anytime soon. ARM primarily makes money in two ways: It licenses chip designs to semiconductor and mobile-phone makers, for example, and then collects royalties every time a chip is sold using that design. During the conference call, Mr. East and Chairman Robin Saxby said that while semiconductor companies are still working on new designs and products, companies are taking longer to make spending decisions or postponing them. If revenue continues falling next year, it is hard to argue the company should trade at such a high premium to the general market. It also doesn't deserve to continue trading at a premium to some of the larger software companies. In the past, bulls argued ARM deserved a 25% premium because of its lock on the chip-design market. "The argument had been that it's relatively immune to the cycle, a safe haven with quarterly revenue progression," said Steve Woolf, a technology analyst at Commerzbank Securities who has been downbeat about ARM's prospects for some time. "Unfortunately, that's not the case. So why should you attach a premium at this point?" That leads to another argument: whether ARM should be valued as a software company, since its business model is akin to one, or as a semiconductor play, since it is in the chip space. The reasons for a software-style valuation were that ARM's royalty revenue would provide growth even if licensing slowed, it isn't as capital intensive as an actual semiconductor maker and it isn't tied to the semiconductor cycle. The last point is one ARM is especially sensitive about. While acknowledging that the downturn in the semiconductor industry has finally caught up with it, Mr. Saxby also said, "I don't believe the ARM business model necessarily relates closely to the business cycle." But ARM's recent downturn shows it is subject to the semiconductor cycle, even if only at a very late stage, said Mr. Woolf. In a normal cycle, the company may not be affected because semiconductor companies cut into their research and development spending only as a last resort. When things go really bad, as they have now, they cut into R&D spending, and that is when ARM becomes cyclical. If that is the case, ARM perhaps deserves a multiple that falls between software valuations of about 20 times next year's earnings and the average of 13 times seen by European semiconductor companies. Even skewing toward a software-style valuation, the shares have room to fall further.