Carl, two issues, the first, is the $75/share (now "extended to happen before the end of 2002). My reasoning is very simple, while the sales of semi chip has actually declined yoy (right now the forecast for this year is $135 Billions), the number of chips shipped is still increasing. For instance, in the DRAM business, the average bit count has increased 80% yoy and is expected to continue at this pace over the next three years. Granted, reduction of features allows for greater packing density of bits per unit of Si real estate, but this reduction does not match the increase in bit count. Thus, growth in Si real estate must be assumed to continue. (WFR actually reported a slight increase in shipment (if memory serves), but of course, sales declined due to decline in ASP). It used to be that the cost of Si was about 5% of the total selling price of chips, I wonder what it is now. I would guess that in the DRAM business due to the extremely rapid fall in ASP, wafer costs are actually greater then 5%. But let us assume it remains constant and this level. My basic assumption is that shipment of chips will resume its growth (I have seen some numbers citing 17% growth rate annually between 2000 and 2003), and the magic $200 billions annually will be breached before the end of 2002. That should provide for the soaking of excess capacity and thus better pricing environment. I think that WFR's break even point is now at about $200 MM quarterly (give or take about $5 MM), and thus they should be quite profitable at $300 MM quarterly sales (which will be 12% of the total market, not an impossibility for the second largest producer), a sales rate I expect to see exceeded toward the peak of the next cycle (profits probably in the range of $.8 to $1.25 per share quarterly). Since that will probably be the time at which the 300 mm program would be at the beginning of its "j" curve, I also think that a lot of euphoria will be around. Based on these assumptions, I do not think that $75 is out of the picture. Of course, they could screw up, Veba could decide to throw in the towel and Murphy might have their "better parts". But right now, with Veba raising their line of credit by another $100 MM (I think it stood at $931 MM of which $680 MM or so was used) survival through a full year of lousy sales is assured, and thus my optimistic forecast. Between here and there, I still can see at least a retest of the lows around 3 and possibly a breach of this low, so, I am still waiting.
As for FERO. The errors with the pullers' strategy were committed more than 10 years ago. They should have aggressively pursued the "consumable" part of the crystal growing business (crucibles and graphite and even possibly even the ethyl "granular polysilicon" business) to level off a little the major up and downs that are expected in the pullers business. Right now, I do not see any major growth in the business left and certainly not the huge up and downs of the past. I think the market is pricing them fairly and would not be a buyer.
Zeev |
