Some of the recent messages about IDTI's management have been interesting. The contract with IBM as a second source for the C6+ has been questioned by several people. My response would be that without a second source as capable and reputable as IBM, no major first tier company would even CONSIDER using the C6+. Look at it this way: if a fire or explosion shuts down IDTI's Oregon fab, no other factory exists capable of second sourcing the part. With an IBM factory building the C6+ in a different state, an assured source exists that is very dependable. Note that Intel long ago told its customers it would be its own second sources. It does that by having large, physically separated production facilities that fabricate the same mask set using the same technology and process.
For a commodity item such as a DRAM or SRAM, many different companies build them to the same electrical specifications that are hammered out as industry standards by the JEDEC council. All kinds of second sources exist for DRAMs, and they are so plentiful and cheap to buy their profit margins are almost zero. That's why companies such as AMD and CYRIX have tried for years to get into supplying the Intel X86 microprocessor line. The copyright Intel has, and their technical prowess have made that extremely hard to do profitably. With a decent design, IBM as a second source, and an outlay of upfront money for the expensive and specialized equipment necessary to fabricate, assemble, and test the WinChip, IDTI has a good chance to move into the sub thousand dollar computer area, which Intel neglected in order to keep its profit margins high. The next six months to a year will tell just how well they've done. Give them credit for having a go at it, and I think IDTI has wisely diversified its product line from dependence on SRAMs into several areas, only one of which is the WinChip.
Beyond that, the talk of productivity can be deceiving. Long ago, I asked someone how it was defined. He didn't know, but it generally means the total dollar output of a factory(sales) divided by the number of people working there. The same applies for total sales of a company divided by total employees. Let me take a for instance in this strange semiconductor industry. The price of DRAMs has roughly halved in one year, that is, a memory circuit that cost $8.00 now costs $4.00. If a company makes the same number of DRAMs using the same number of people in those two years, its productivity fell by a factor of two. If the company, using the same labor force, actually doubled the number of units it made and sold, its productivity would remain the same. In the second case, the cost of making twice as many good die, and assembling and testing them would add greatly to the costs. Consequently, earnings would drop badly. The factory that second year would have doubled the number of units out the door, the employees would have been killing themselves to do it, their productivity would not have increased, and their earnings would have been awful. Well, DRAM companies such as Micron, Samsung, NEC, Hitachi, and others face that problem right now.
To be a bit more fair, I prefer to define productivity as the number of widgets a plant makes a year divided by the number of people working there. That number is independent of the unit sales price. It's not done that way, however, so always keep in mind fluctuating prices can greatly affect what we call productivity. For automobiles, wash machines, and refrigerators, whose prices don't fall too much, if at all, total dollars divided by total number of employees works fairly well. Don't use that definition this year on PCs, where the average sale price will fall drastically. To make things worse today, currency exchange rates also are varying, with a buck now buying 133 yen and an ungodly number of Indonesian rupiahs. The dollar is very strong, imports consequently are very cheap, American exports are under great pressure, and our stock market is nuts with Lucent now selling at a PE ratio of 234. You try to figure out how to guarantee a nice tidy, growing profit in this business! I can take the long view, and will wait for IDTI's earnings to increase and analysts to raise their opinion of it. As with some of the other patient investors, I expect the stock to hit around twenty bucks in twelve months, unless the entire house of cards tumbles down. MM's concept of using R&D investment money to compute a modified PE ratio makes sense when the research pays off with products that sell well. If the wrong products are designed, R&D expenses simply add to the loss.
If anyone has a specific definition of productivity other than mine, please share it with me. I can't help but feel that in a country where Alan Greenspan's, an unelected official, unintentional phrase at the wrong time can shake the DOW, logic and reason don't mean too much. |
