So far, this week has not been kind to AMD. Between the less-than-phenomenal Phenom launch and the sale of 8% of the company to an Arab investment firm, the stock price dropped almost 15% between Monday and Wednesday. The stock closed Wednesday at a four-year low.
While AMD management may be comfortable springing financing surprises out of their hats, the stock market isn’t.
At such times, wishful thinking starts turning into news stories. Recently, there’s been two of those.
The first was this article suggesting that Mr. Ruiz might be on his way out.
The problem with the story is that it’s not based on anything beyond, “People are grumbling.” More importantly, nothing more solid about the matter has emerged after the story came out.
The sad fact is that there’s no terribly rational reason to get rid of Mr. Ruiz, simply because it’s not going to help. Whatever bad decisions AMD has made are spilled milk, and reversing them would be a cure worse than the disease.
The ATI purchase is spilled milk. AMD paid top dollar on a company on the verge of going into a slump, but selling ATI now would probably be getting bottom dollar for a company on the verge of recovering.
AMD has two fabs. If you think they can’t afford three, you’re probably right, but it’s not like AMD has been spending its fab money in New York State. The stock market people would like AMD to stop spending money on fabbing, but that’s like telling someone that the only cure for their illness is to stop breathing.
The fab money this year has mostly been spent getting a trailing-edge fab up to full speed. What else were they supposed to do, make 90nm processors forever? In any event, the money on Fab 36 is spent. Next, they were supposed to renovate Fab30. It appears they’ve stopped doing that at the moment, but if they don’t renovate that, what are they supposed to do with an antiquated fab?
Some people think you can outsource fabbing like a call center. If AMD did that, that would be the greatest disaster of all. It makes no financial sense; outsourcing to a foundry makes sense only when your volumes aren’t enough to fill a fab with work. That’s not the case with AMD, outsourcing would cost them more, not less. It would be suicidal from a technological standpoint; if AMD and IBM can’t keep up with Intel, what makes you think a TSMC or Chartered will immediately do much better.
AMD’s problem isn’t that they’ve spent too much on capital expenditures and R&D; it’s that they’ve spent too little (or at least spent what they had ineffectively). If AMD could punch out 3.5-4GHz CPUs with their eyes close, and ATI chips were 20% faster, nobody would be talking about billions in losses or paying too much for ATI. They’d be wondering if Intel was going to make it.
Unfortunately, that’s not quite the case, and replacing the man on top isn’t going to change that.
Does that mean Mr. Ruiz is safe in his job? Not at all.
In sports, when a team does poorly, often the person who gets fired is the coach/manager, even when the players are really at fault. He is made the scapegoat for the team’s failure.
If Mr. Ruiz goes soon (and if it does, this will be dressed up as a “retirement” as of the date his contract expires in late April), he’ll be thrown to the wolves not because his successor will or even can do better, but because the wolves want vengeance on their scapegoat.
I don’t think this is a done deal yet. I think more K10 delays and a worse-than-expected fourth quarter would do it, but that’s probably not going to be the case.
The irony is that if Mr. Ruiz goes, he’ll go for the wrong reasons. As we’ve said in the past, AMD shouldn’t be faulted for believing they had to get big or wither away, that assessment was and still is a good one. AMD should be faulted for trying to do so without raising serious money while times were good and they had momentum, then using it to keep up the momentum and be able to withstand any short-term Intel price games.
The failure to do these things is why Mr. Ruiz should go, not because he built a second fab.
Nonetheless, right reason or wrong, his departure isn’t going to solve anything.
I will say this, though, if AMD has already decided Hector’s fate, stay or go, they ought to publicly announce that soon, certainly by the end of the year. Leaving him hanging might make some people feel good, but further FUD is the last thing Green needs right now.
Tomorrow, we’ll talk about the second instance of wishful thinking, the thought that Apple might take a stake in AMD.