Ref- <The Intel apologists (or bad behavior enablers) on this thread will tell us that it is not easy to predict revenues - because there are so many variables. Doesn't Cisco, Lucent, Msft, Yahoo, AOL, and others all have the same set of problems. Yet they always seem to meet earnings expectations.>
No! No! No! Mary, Cisco, Lucent, Msft etc do not have the same set of problems. The only uncertainty for these companies is predicting the unit demand for the next quarter. The ASP(Average Selling Price) and the Average Cost of Goods is very well defined. For chip companies, the Average Cost and ASP are in a continuos free fall. Unfortunately the exact amount of this decline is not exactly predictable. Lately the BTO( Build to Order Model) has aggravated the ability to forecast revenues and profits. For the last two years Dell and others place no long term orders(>30 days). They place the order and want the shipment tomorrow.[ Now this has caused a shortage since demand has exceeded forecasts].
Thus investing in a chip company requires nerves of steel. The market has always given chip companies lower earnings multiples because of the above uncertainty. The system companies have commanded higher multiples, because they have an easier time hitting their numbers. Over the years I remember companies like DEC, Data General, Burroughs,Polariod, Xerox,IBM,Kodak,Apple had a PE 2 to 4X that of chip companies. However an investment in Intel has been much more rewarding with a much lower multiple. The chip companies do much better long term, but are subject to significant short term volatility. Intel has the least volatility of any chip company.
Thus focus on the long term. If you are comfortable with the long term, the few percent error in short term earnings should be inconsequential. |
