Sunday, January 6, 2019

Code names referenced in FTC v. Qualcomm antitrust trial: sounds like a spy novel (at times)

After my bird's-eye view of the first FTC v. Qualcomm antitrust trial day ("In its courtroom chess match with Qualcomm, the FTC gains initial control over the center of the board"), there's more information to share. And there's still more to come after this post.

Toward the end of the day, Judge Lucy H. Koh requested a glossary from the parties given all the acronyms that came up, such as NDRC (China's National Development and Reform Commission, one of three government agencies sharing the responsibility for antitrust enforcement until the consolidation announced last year) and DG COMP (the European Commission's Directorate-General for Competition). The first trial day featured not only acronyms but also three code names: Eureka, Project Berlin, and Project Phoenix--code names of the kind one would see in a spy novel or movie.

Eureka = Qualcomm (in Apple docs)

According to what was said in court (without contradiction, at least so far), "Eureka" was a code name for Qualcomm that appeared in various Apple-internal documents.

As a former student of Ancient Greek, I primarily associate it with the exclamanation attributed to Archimedes. It's the first person singular of the present perfect of the Greek verb for "to find" (which, by the way, is also the etymological root of the word "heuristic").

But why Apple chose that particular code name for Qualcomm is unknown at this point. As Wikipedia indicates, "Eureka" has a wide array of meanings and uses. With Apple and Qualcomm both being headquartered in California, the northernmost major city of the state or the fact that "Eureka!" is the official state motto may very well have driven this choice. At least those two possibilities appear far more likely than any connection with the namesake cat in The Wizard of Oz or a burger chain...

Project Berlin and Project Phoenix: potential breakup of Qualcomm (separating chips from patents)

My first Qualcomm-focused blog post ever, back in 2012, was about a restructuring by which Qualcomm sought to avoid granting implicit patent licenses to open-source projects (with potentially far-reaching ramifications due to patent exhaustion).

What became known on Friday is that Qualcomm was actually analyzing, at two points in its recent corporate history, the question of whether shareholder value could be maximized through a voluntary breakup. If the objective is not jst a spinoff (where one entity would remain a major shareholder of the other) but a complete separation, the same shareholders end up owning shares in two companies instead of one. In Qualcomm's case, shareholders would then have held not only QCOM (Qualcomm's current stock ticker symbol) shares, but in two separate companies (a product and services business, and a patent monetization business). Once the second Qualcomm entity would have gone public, shareholders could then have been free to decide how to weight those two entities, relative to each other, in their portfolios.

According to what was said on the first trial day, the assumption was that the patent licnsing business represented roughly two thirds of Qualcomm's total enterprise value. This validates a BlackBerry lawyer's characterization of Qualcomm as "basically a patent house" in a meeting I had with them in Brussels in 2006.

It happens all the time that large publicly-traded companies consider breakups and spinoffs. In my observation, breakups are considered primarily when stakeholders are worried about negative synergies or cultural incompatibilities outweighing any positive synergies. Spinoffs are typically driven by "stock marketing" considerations. For an example, about two decades ago some major carriers such as Deutsche Telekom and Telefónica had their Internet service provider businesses (T-Online and Terra Networks) listed separately because such types of companies were trading at far higher multiples than carriers--and later reintegrated them.

Breakups can also be ordered by competition authorities. Microsoft once faced that threat. In Qualcomm's case, a proper breakup would remedy the "no license-no chips" problem, but the FTC has been very conservative and is, therefore, focusing on injunctive relief, rather than restructuring, to address the issue. By contrast, there are competition enforcers and policy-makers in Europe who are much less conservative and think aloud about potential breakups of certain U.S. tech giants...

I know analysts who actually would like Qualcomm to rethink its business model, now that it has come under so much pressure, and who would view a voluntary breakup favorably. Under Qualcomm's current business model, the fact that the same entity owns and controls the chipset and the patent licensing business is, however, absolutely key to leveraging the "no license-no chips" policy. The existence of that policy is undisputed; what Qualcomm's counsel indicated in his opening statement is that they're going to argue they never actually did cut off supplies. Some of the Friday testimony explained how Qualcomm can and according to those witnesses sometimes does threaten with a disruption of chipset supplies, forcing companies like Huawei and Lenovo to agree to license terms they'd otherwise consider excessive and reject (because they'd likely get a far better deal if Qualcomm had to bring infringement litigation and if a court of law determined a FRAND royalty rate).

"No license-no chips" wouldn't work after a breakup because any single shareholder who would hold shares in the chipset but not in the patent licensing business, or who would own a relatively larger share in the chipset than in the patent licensing business, could sue the chip company's management for breach of their fiduciary duty by harming their own company's business only to give a separate entity more leverage in licensing negotiations.

In what Qualcomm internally referred to as "Project Berlin" and "Project Phoenix," the upside and downside of a breakup were evaluated, and in both cases the ultimate conclusion (in one case validated by the Boston Consulting Group) was to refrain from such a structural change.

What's not clear (at least not yet) is what really motivated such analytical efforts in the first place. It may have appeared potentially attractive to separate the lower-margin chipset business from the patent licensing business, which would have had an extremely high gross margin (in fact, a threshold margin of 100%). But I suspect that Qualcomm was also thinking of the specter called patent exhaustion: while Qualcomm would always argue that the patents are already held by a legal entity that is separate from the chipset business, both are part of the same corporate group. Ultimately they may not be able to have their cake and eat it in terms of leveraging synergies through the "no license-no chips" policy while stressing the existence of separate legal entities in the patent exhaustion context. (With the Supreme Court's Lexmark ruling, end-runs around patent exhaustion are even harder.)

Some of the Friday evidence and testimony was about the conclusion reached by Qualcomm's boar of directors, executives, and consultants (BCG) that the patent licensing business would strike better deals as long as the same company also controls the chipset business.

Qualcomm's witnesses try to attribute those synergies just to the fact that if you have an "ongoing relationship" with a business partner and the relationship is about building products, there's a better climate for negotiating patent licenses. But it was crystal clear on Friday that the FTC has plenty of evidence it's about hard leverage ("no license-no chips") more than anything else. Of course, the analysis won't stop there: Qualcomm argues that extracting supra-FRAND royalties, should it have happened, wouldn't be an antitrust violation anyway, and that what happened "in the marketplace" doesn't suggest there was anticompetive harm.

Why Qualcomm chose those particular project names wasn't discussed. In the case of "Phoenix," it's hard to see. Qualcomm never needed to rise from the ashes. "Berlin" is almost certainly a reference to the city's temporary East/West division.

Share with other professionals via LinkedIn: