Here is another view on inventory (thanks to les horowitz)
Message 20618669
Intel's recent mid-quarter update on Sept. 2 was interesting from the perspective of inventory levels. Intel has an inventory problem (although management is trying to take our eyes off that), and I think the situation can get worse. The Sept. 3 column examined Intel's the mid-quarter update in detail. Meanwhile, the inventory build is something that most don't want to acknowledge, and which can have a significant ripple effect:
For Intel it's bad, because to drive revenue growth it will have to forgo margins -- unless demand picks up substantially. Intel may need to move out parts and become less dogmatic about what they're "worth." This could prop up sales at the expense of margins, or we could see write-downs at Intel.
For competitor AMD, my sense is the inventory issue isn't as bad. AMD has built some inventories but not to the extent Intel has.
For OEMs, it's fairly positive because they'll keep part of the price reduction and pass on the rest to the customer. When suppliers publicly admit they have inventory it's always a bonanza for those who buy and use chips. Pricing becomes negotiable versus the previous year or so that parts have been on "allocation." Compounding the risk to the sector is buyers now have an incentive to wait, since prices seem likely to fall even further.
For consumers it's good, because OEMs will slash pricing, and in theory the average Joe could see lower PC prices.
For IT distributors (such as Ingram Micro, Tech Data, etc.) the impact is tougher to determine and possibly too new to rate. It's a difficult area to get insight via broadline distribution, because it's not a main item, although chips of some type are in virtually every product sold. There are a lot of cross currents in PC-land: a fall off in demand could well be offset by a price war among Hewlett-Packard, IBM, Dell, Toshiba, Sony, and others that could keep unit sales moving. Pricing and margins then become issues, but with the various crosscurrents, the impact on distributors is too tough to call right now.
I think inventory will continue to be a problem for Intel. The company's production efficiencies are causing output to grow faster than demand. Unless demand growth picks up significantly and stays up, Intel could continue to be trapped. AMD will be squeezed, but not as badly.
Inventory is the elephant in the room nobody wants to see.
For those who buy and use chips in their products, suppliers publicly admitting they have inventory is always a bonanza. Pricing becomes negotiable versus the last year or so that parts have been on "allocation." Compounding the risk to the sector is buyers now have a direct incentive to wait, since prices seem likely to come down further.
In terms of absolute levels, I believe Intel's $3.2 billion is higher than after the last peak. Clearly there are excess PCs and components in the channel. Cisco admitted extra inventory as far back as May, so networking seems to also have excess. Rumors abound of excess inventory in many of the cell phone channels -- although end demand has been a lot stronger for handsets than it has for PCs, so I'm inclined to think the inventory problem wouldn't be as great as it is with CPUs.
So far, the CNBC hype cycle has succeeded in pushing the clearly false claim that excess inventory is "needed" to support growth. If this was true then why has the last decade been spent buying IT gear, which was supposed to eliminate excess inventory in the supply chain? What happened to Just in Time (JIT) inventory management? The CNBC argument is so totally wrong that people think they must be missing something. If they believe the "spin" they are missing something.
The next "spin" will be that sales are up, which supposedly excuses margin declines. This assumes suppliers choose to move inventory at a discount rather than write it off completely and hope to use it as "free money" in future quarters. (I've spoken with one sales manager at a publicly held firm tell me this was the firm's explicit strategy in 2001 when the business last turned down.)
If Intel had set the bar lower, it would have to admit there was no way in hell it could get rid of the inventory it has. By setting it in the $9 billion range, it can continue claiming it has a plan to start working off inventory. (Although I think they'd need to have an $11 billion to $12 billion quarter in order to completely get through the excess inventory without major production cuts.) To some degree, Intel is trapped: set the bar lower and the question becomes "How in the world will you get rid of that inventory without substantial price cuts?" Set it high, and the judgment day is somewhat put off.
Which means that the numbers Intel's management has been touting don't add up at all. |
