Ex-Intel Board Members Make An Ill-conceived Case For Spinning Off Foundry

Comment A gaggle of ex-Intel board directors have called on the chipmaker to spin off its floundering foundry business while glossing over the fact that a company bleeding billions each quarter is unlikely to survive on its own.

Instead, the former directors, David Yoffie, Reed Hundt, Charlene Barshefsky, and James Plummer, argued in a letter published in Fortune on Tuesday that breaking Intel in two is the only way to save America's most important manufacturer.

During its disastrous Q2 earnings call, CEO Pat Gelsinger announced Intel reported $1.6 billion in net losses, planned to lay off 15 percent of its global workforce (about 15,000 people), and was cutting capital expenditures. Since then, the chipmaker has also announced it will spin off its foundry division as an independent subsidiary.

However, the former board members aren't convinced this will be enough. "Intel's management and board must take responsibility and move decisively to stem the bleeding. The actions they've taken to date will not suffice."

"Intel offered foundry services to the industry for two decades, never building a successful business," they added.

Intel offered foundry services to the industry for two decades, never building a successful business

If you're scratching your head wondering when exactly this happened, we're right there with you. Intel has offered custom foundry services to the likes of Altera before. But, from what we understand, that program only lasted five years between 2013 and 2018. You'd think a bunch of ex-board members would have a better memory of Intel's history, but perhaps they know something we don't.

The group argues that there are numerous advantages to spinning off Foundry, including alleviating concerns by competitors like Broadcom, Nvidia, and Apple, which they argue are desperate for an alternative to TSMC but are hesitant to do business with a potential competitor — something we've talked about before.

They also point to AMD's spin-off of its foundry unit in 2009, which became GlobalFoundries, but we're not quite sure this is the glowing example they think it is. GlobalFoundries hasn't produced leading edge silicon since abandoning its 7nm process node and lacks the scale to compete with TSMC or Samsung in any meaningful way. AMD, for its part, has all but forgotten GlobalFoundries, having moved nearly all of its production to TSMC. Spin off Intel Foundry before it's self-sufficient and there's a good chance Intel's Products division will be stuck on TSMC forever.

What the letter doesn't do a great job of addressing is how exactly an independent Intel Foundry is supposed to avoid going bankrupt. The ex-board directors argue that Intel's design business remains profitable so it should establish a long-term supply contract as part of a spinoff. "The new Intel design company would need to partially underwrite the foundry's losses and guarantee sales for several years," they wrote.

The problem is Intel is already doing this and it still can't turn a profit. Intel Products reported $2.9 billion in operating income in Q2, while Foundry posted a $2.8 billion operating loss. Between that, its other divisions and expenses, and the chipmaker reported $1.6 billion in losses last quarter.

As we've discussed before, part of this comes down to the fact Intel doesn't actually produce much of its portfolio in house anymore. It's pretty much just old Core-series processors, many of which have a critical design defect, and its datacenter Xeons, which aren't exactly driving sales right now either. DCAI reported $276 million in operating income on about $3 billion in revenues in Q2. Not exactly the hero customer, an independent foundry needs to keep the lights on.

Pretty much everything else Intel makes has been outsourced to TSMC. Intel does plan to pull many of these products back into its own fabs starting next year, but a spinoff of its Foundry unit prior to this could easily delay those plans and any hope of ever turning a profit in the process.

Intel also can't go back to writing Foundry a blank check for manufacturing services, as the ex-board members seem to suggest. Doing so could be seen as preferential treatment, which we all know regulators just love. And even if a spin off does help Intel win new contracts for American made chips, it will take several quarters of heavy bleeding to fulfill.

• It's about time Intel, AMD dropped x86 games and turned to the real threat
• Spectre flaws continue to haunt Intel and AMD as researchers find fresh attack method
• AMD pumps Epyc core count to 192, clocks up to 5 GHz with Turin debut
• Arrow Lake splashdown: Intel pins hopes on replacement for Raptors

The former board members do acknowledge some of this, saying that "Intel's design company alone cannot support an independent foundry. Yet, Intel's manufacturing operation is the only hope for maintaining the most advanced nodes on US soil."

The argument here seems to be: don't worry about Foundry's financials, it's too important to America's national security for Uncle Sam to let it fail.

"Today, Intel threatens to become this administration's Solyndra (the solar company, which went bankrupt after getting more than $500m in government funding). This would be disastrous, both for the government and Intel," the ex-directors argued. "The government should be very clear on what it is willing to finance, including Intel's corporate structure."

Perhaps it's a good thing these guys don't sit on Intel's board anymore, it seems they're far more concerned about protecting their investments from Gelsinger's ambitions than saving "America's most important manufacturer." ®