Qualcomm buying Intel makes no sense, unless…

It was widely reported late last week that Qualcomm has approached Intel about a friendly takeover bid. It looks like the WSJ got the story first, with Reuters hot on its heels, both inevitably relying on anonymous sources. They stressed things are at an early stage so this feels like a classic trial balloon story, leaked by at least one of the interested parties, to gauge market sentiment on the matter.

Despite both being US chip giants, Qualcomm and Intel are very different companies and the more you look at them, the less of a good match they seem. Qualcomm specialises in designing low-power chips and modems, which it then pays foundries to manufacture for use in mobile devices. Intel manufactures its own high-power chips that are mainly used in PCs and servers. Qualcomm has long been the number one player in its sector while historically dominant Intel has been struggling for years.

Mega-mergers are very high risk and, even if eventually successful, take years to fully play out. Qualcomm doesn’t need Intel’s manufacturing capability and is even starting to take market share from it in the PC and server markets. There are bits of Intel that would be complementary for Qualcomm, such as its PC and server relationships and channels, but none of those would justify the enormous cost and disruption a deal like this would cause. And that’s assuming it was approved.

As Richard Windsor from Radio Free Mobile asserts: “I see no reason to dilute shareholders to purchase revenues that could very well be coming one’s way anyway… Even if Intel recovers and becomes the leading semiconductor manufacturing company again, Qualcomm will still be able to access its fabs through Intel Foundry.” In other words, the business case for this acquisition is weak.

So the only way this makes sense is if the US government is playing a major part in the deal. The US has long made it clear that it considers the chip sector to be of paramount geopolitical strategic importance. To that end it has imposed a number of arbitrary restrictions on Chinese access to the latest semiconductor technology and is pulling out all the stops to make itself less reliant on regions bordering China for its chip supply.

Additionally, the US state has been chucking money at Intel in the hope it can pull its socks up and do a better job of competing with TSMC and Samsung on chip manufacture. But Intel recently announced the pause of a few manufacturing projects as part of a broad cost-cutting drive, so the US government must be panicking about the fortunes of this key front in its chip war with China.

While it seems likely that the two companies have had at least background support from the US state in this enterprise, it’s not obvious what substantial assistance it could offer to overcome the many arguments against such a deal. A massive loan is one possibility, but a politically awkward one at a time when public funds are being frittered away on subsidising proxy wars.

And then there’s the regulatory side. Even if the US government were somehow (and, presumably, illegally) able to guarantee the approval of US authorities for the deal, both are global companies. It has already expended a lot of political capital in its chip war and the EU at least is likely to bristle at the competition implications of such a move.

All things considered, the probability of this move taking place seems low, which is probably why both companies’ share prices only moved by a few percent on the news. Qualcomm has little reason to buy Intel and even the mighty US state probably doesn’t have the power to sweeten the deal sufficiently to overcome that. That doesn’t mean it won’t try, though, so desperate is the US to dominate the chip world.