Saturday, July 24, 2021

SEP Licensing Negotiation Groups -- Part I: analogy to patent pools entails false symmetry between facilitating and complicating automotive patent licenses

This is the first part of a trilogy on licensing negotiation groups: automotive industry cartels that would collectively negotiate standard-essential patent (SEP) licenses with major patent holders and pools.

Earlier this month, the European Commission levied a € 875 million fine on Volkswagen (that company has to pay more than half a billion euros alone) and BMW--Daimler got away unscathed only because it had blown the whistle--for restricting competition in emission cleaning for new diesel passenger cars. That decision stands in the tradition of regulatory findings of cartel law violations by car makers and their suppliers. Just a few recent examples from the EU that beg the question of whether such conduct is deeply ingrained in that industry's DNA:

One of my core principles in competition policy is that even the worst offender might have a valid concern or bring a meritorious complaint. There was a time when Microsoft was synonymous with antitrust violations in the technology industry (over issues that were not even 1% as serious as what Apple has been doing for more than a decade). Prior to Microsoft, IBM had that reputation. IBM was then behind efforts to instigate investigations against Microsoft, and Microsoft later did the same against Google. Par for the course.

But when a company that just got slapped with a half-billion-euro fine over a cartel keeps lobbying the European Commission, the U.S. Department of Justice, and possibly other regulators to obtain permission for forming another automotive cartel, its proposals and its arguments need to be closely scrutinized.

I've looked at this topic from multiple angles: whether the issues Volkswagen has identified are real; whether the proposed measures are necessary and proportionate; whether licensing negotiation groups would be more or less constructive than individual parties; whether SEP holders would still be reasonably capable of enforcing their rights; and whether the upside would outweigh the downside in competition policy terms.

Unfortunately, it turns out that this is not only a solution in search of a problem but--far worse--a kind of Pandora's box.

If it's as bad an idea as I say it is, how come it is being talked about at all? Well, that's a mystery to me, but I have a plausible explanation: policy makers' predilection for redressing the balance and for bringing about symmetry. In this case, however, it's a fallacy. A false symmetry.

Pooling patent portfolios that belong to different licensors is fundamentally different from "pooling" licensees. The asymmetry begins with respect to transparency:

  • A whole lot is publicly known about Avanci, a cellular SEP pool that contains patents from more than three dozen holders and provides a one-stop licensing option to car makers (which the U.S. DOJ approved last year).

  • The proposal of forming licensing negotiation groups ("LNGs") is largely being kept under wraps, the sole exception amounting to a couple of slides in a presentation delivered by Volkswagen's chief patent counsel Uwe Wiesner as part of a mid-April European Commission webinar.

Mr. Wiesner is undoubtedly the thought leader in the automotive sector when it comes to patent policy. But that doesn't mean his ideas are better than those of organizations who were dealing with cellular SEP licensing issues even before Mr. Wiesner joined the patent bar.

BMW, Volkswagen, and--unconfirmed rumor has it--Tesla have taken Avanci licenses at the car level. By now there's empirical evidence that Avanci's existence does not foreclose all sorts of bilateral license deals:

Those are just some publicly-known examples. There are other license deals that have been signed but not announced--at the car level as well as at the component level. If major smartphone makers held out as long as automotive companies (even Volkswagen's limited Avanci license falls far short of what it actually needs), there'd be dozens of large-scale disputes (like Nokia v. Oppo) pending. But Daimler's about-face (taking a car-level license after years of arguing that its suppliers should take the prerequisite licenses) marks a turning point, and it's too important for me to ignore. I stand by my advocacy of component-level license deals, and I celebrated the ones that were announced. That said, I recognize market realities. With the three German car makers (representing well over a dozen brands in the aggregate) and--if true--Tesla having taken car-level licenses, neither they nor their competitors can claim anymore that there's anything wrong with licensing cellular SEPs at the end-product level, just like the mobile handset industry has been doing it for ages.

In the aforementioned presentation, Volkswagen criticized the fact that Avanci's per-car license fee doesn't differentiate between, for example, a Tesla and a low-cost compact car. But the connectivity provided is essentially the same. Car makers are free to content themselves with 3G network access to save a few dollars in license fees (whether that makes sense is, of course, another question). Also, that argument is a slippery slope as automotive companies would rather pay the same per-unit royalties as the makers of the cheapest smartphones. Volkswagen's criticism of a consistent rate directly contradicts the arguments Apple always makes in its SEP licensing negotiations and in policy debates. The iPhone maker has put a lot more effort into the devaluation (just using the terminology of an Apple-internal document that surfaced in its Qualcomm litigation) of cellular SEPs than the entire automotive industry.

Volkswagen also raises the concern of double-dipping. Actually, if everyone consistently licensed SEPs at the car level, there wouldn't be any double-dipping issue--and cars could even contain more than one component implementing the covered standards and pay only a single license fee. The only risk remaining here would be that you might have a SEP holder in the supply chain, such as LG providing a network access device. True, the first authorized sale has an exhaustive effect, though that one may also be territorially limited anyway (patent exhaustion across borders is complicated, and German courts are particularly disinclined to recognize it). Also, the amount in question would not be huge even if a Huawei, Samsung or LG provided a component. And the problem could always be solved through good-faith negotiations.

It is hard to see what problem(s) licensing negotiation groups would be in a better position to solve than individual parties.

What is clear, though, is that a patent pool like Avanci making an optional one-stop licensing offer doesn't result in a concentration of market power or in coordinated misconduct. Deals of all sorts happen, and more often than not, an Avanci contributor is involved.

If Volkswagen's LNG proposal became mandatory in the sense that SEP holders facing a licensee group couldn't insist on a bilateral deal (and, if necessary, bring infringement actions to accelerate the process), it would be a recipe for disaster. It would invite group boycott. "LNG" would then mean "Licensing Never Group" until SEP holders do business on the implementers' preferred terms.

If SEP holders could bypass the LNG as they please, I guess they would and there'd be an additional layer of bureaucracy that would add no value to anyone (unless one considers it "valuable" to create an opportunity for parties to legally organize collective hold-out).

It's a misconception that the share of the patents declared essential to a standard and being contributed to a single pool has anything to do with market power. As an expert witness for then-Google-owned Motorola Mobility once famously wrote, "it only takes one bullet to kill." A single truly essential patent (meaning that it can't be worked around) confers gatekeeper-like market power on its owner. Any additional bullets would just hit a body that's already dead. In reality, the larger a pool is, the more likely it is to discipline other SEP holders by dissuading them from trying to charge more on a per-patent basis. The "market share" of the pool does, of course, have a bearing on transaction costs. But that's a potential efficiency gain that, in an idealized economic scenario, would be split fairly between licensors and licensees.

What's key is to have reasonably balanced SEP enforcement. Whether a pool holds one patent out of 10,000, or even all 10,000 that read on a standard, is a non-issue if SEP enforcement works. In the next part of this trilogy, I'll look at the implications of the LNG proposal for enforcement in more detail. The third and final part will then focus on antitrust and cartel considerations (which are so serious that even the much less radical "Proposal 75" of the European Commission's SEP Expert Group Report comes with a variety of caveats), and I'll summarize the key reasons for which I believe regulators and policy makers should flatly reject the LNG proposal and prioritize more workable and less problematic approaches.

Share with other professionals via LinkedIn: