Arik Hesseldahl

Recent Posts by Arik Hesseldahl

The Problem With Those Rumors of an AMD Buyout

It all seems so simple. At chipmaker Advanced Micro Devices, a sudden and unexpected sweeping away of management–starting with CEO Dirk Meyer, followed within weeks by COO Robert Rivet and Marty Seyer, senior VP for corporate strategy–has left the company looking disorganized and vulnerable, the thinking goes.

And while a search for Meyer’s replacement is underway, I’m told it could easily extend into the summer.

It didn’t take long for rumors about “takeover chatter” concerning AMD to emerge, and briefly yesterday, Dell was mentioned as a possible buyer. AMD shares traded up 4 percent for part of the day but closed down 3 cents during the regular session. Dell more or less shot down the rumor. During its earnings conference call, CEO Michael Dell, answering a question on acquisitions, said, “…we’re looking for relatively smaller sized ingredient acquisitions where we can leverage them with our substantial customer access and distribution.” With AMD currently trading at a valuation north of $6 billion with about $2.2 billion in long-term debt, it’s not the kind of target that would qualify as “smaller sized.”

There will always be rumors of this sort about the perennial number two in the PC microprocessor business. Those who trade on them don’t get something fundamental about AMD: That it would be a complicated company to buy and to own.

Any deal to acquire AMD will necessarily include a third party: Intel. For decades Intel and AMD have operated under a series of patent cross-license agreements that give AMD access to the crown jewels of Intel’s intellectual property, including the x86 instruction set. These patents are on the technology that make a PC a PC, and they are fundamental to the success, or failure, of both companies.

When AMD first sought to spin off its manufacturing operations into the company that became GlobalFoundries, Intel asserted that AMD couldn’t assign access to these patents to a third party without its say-so. This dispute ultimately got the two companies talking and resulted in what I like to call the Treaty of Maui, the settlement of a sweeping antitrust dispute in 2009, a story I reported at the time for BusinessWeek.

There are, however, some limits governing Intel’s conduct in this scenario. When it settled an antitrust case against it last year, Intel agreed to hold off on suing any company that buys one of its competitors for a year, in order to hold “good faith negotiations” over the terms of that patent cross-license agreement. What this all means is that any company that first concludes a deal to buy AMD will then have to pivot and face the possibility of lengthy negotiations with Intel that could, if not successful, end in a costly and distracting patent lawsuit.

Intel may turn out to be willing to play ball, and cut a reasonable deal with any new owner, but the fact remains that every so often the cross-license arrangement has to be renewed. And that’s not to say a determined buyer couldn’t ultimately cut through all this and get a deal done. Dell has $15 billion in cash and could conceivably get a deal done, and being an AMD customer could arguably benefit from owning AMD over the long term, but it has signaled that it’s not interested, and probably never was in the first place.

There are other considerations: AMD is 20 percent owned by the Mubadala Development Company, the investment arm of the Arab Emirate of Abu Dhabi, which changes the potential deal dynamic a bit. Then there’s the big question concerning the wisdom of competing with Intel. As AMD’s prior CEOs will tell you, simply grappling with Intel in the marketplace is a dangerous, thankless job.

But the complication of the Intel cross-license agreement alone should be enough to give any company mulling an AMD buyout serious pause. At the same time it should serve as food for thought for anyone wanting to trade on the latest AMD buyout rumor. This surely is not the last.

Latest Video

View all videos »

Search »

Another gadget you don’t really need. Will not work once you get it home. New model out in 4 weeks. Battery life is too short to be of any use.

— From the fact sheet for a fake product entitled Useless Plasticbox 1.2 (an actual empty plastic box) placed in L.A.-area Best Buy stores by an artist called Plastic Jesus