Excellent post, Wes:
You said, "if chip orders were to just remain flat (no growth, no decline), then chipmakers in theory would not have to increase capacity, in which case, equipment orders go to ZERO."
Capacity orders would go to near zero (unless there is a rash of fires in chip plants), but technology orders would continue. And technology orders are a non-trivial fraction of the total. And we are at the initial stages of several critical technology shifts, which give a significant advantage to those who adopt them. That is, in 1998 the entire semi industry could put their 300mm plans on hold, but today I don't think they can. The transition is too far along now. It's one of those threshhold events, and the reaction has achieved "activation energy." Once several companies have installed the equipment, and are producing on 300 mm pilot plants, they have committed to ramping up to volume production. And, as production costs per chip fall, their competitors have to buy the 300mm equipment also, or get out of that chip market. So, IMO, technology purchases will continue, even if (as seems very likely), chip demand is less than expected. Of course, companies who are using junk bonds rather than cash to build their fabs are not going to be able to make any purchases.
The next critical questions are: will the economy have a hard landing, and how closely is chip demand tied to the overall economic cycle?
My opinion is that chip demand is now much larger than just PC demand, and is now in so many different products, that cycles in chip demand will fairly closely follow overall economic demand. Cars, for instance, are a classic "cyclic" industry. And the fraction of a car's cost, represented by chips, has been steadily increasing, and, IMO, will continue to increase. Same thing for most toys: almost everything I'm buying this Chrismas for my kids (5, 7, and 9 years old) has chips in it, or needs chips to use it.
So, if we get a recession in 2001, then I agree, we'll see AMAT at or below 30 (BTW, 27-30 seems to be most posters target price, given this scenario). In 1998, we bottomed at a P/S of 2.2; a fall to that valuation in 2001 would mean a further 50% haircut (more like beheading), from 40 to about 20. IMO, even if everything possible goes wrong, we may not get that low. The 1998 low valuation was a lot higher than the 1996 low valuation, and a 2001 valuation low will (again, IMO) most likely be around a P/S of 3 rather than 2. Margins are higher, the company is better known, the company is in more semi-equip sectors, and has increased market share. These things deserve a higher valuation. Unless anyone thinks I'm starting to sound like a PermaBull, notice I am justifying a P/S of 3-4, not a P/S of 13-14.
So, the only thing I think that might possibly get AMAT to 20, would be a recession in 2001. You're right, if we get a recession, my recent purchase at 40 may be underwater till 2002. I can handle that, although I'd rather not. Could you tell me why you think we'll get a recession? I was very encouraged by Greenspan's speech today. He essentially took back his "irrational exuberance" speech. And he signalled a change in bias, soon. |
