Tuesday, March 1, 2022

Fifth Circuit says automotive suppliers can't complain if patent holders license only car makers: Continental lacks standing against Avanci pool, Nokia, others

Even Apple's and Tesla's amicus briefs didn't help: Continental, which is generally known for making tires but also in the telematics control unit (TCU) business, just suffered a defeat in an appeals court that has intercontinental implications for the car industry's debate over the appropriate licensing level(s) of standard-essential patents (SEPs). On Monday, the United States Court of Appeals for the Fifth Circuit handed down its Continental v. Avanci et al. opinion following an early October hearing. As opposed to "Conti" getting the dismissal of its case against the Avanci patent pool and some of its key members (most famously Nokia) reversed, or at least a chance to amend its complaint once again, the automotive supplier is actually in a worse position now than before the appeal. It's technically even one step further away from winning (which it was never really close to), and policy makers and courts around the globe may adopt the Fifth Circuit's reasoning that SEP holders may choose to license only end-product makers, at least as long as they don't cause any direct injury through patent enforcement against suppliers they deny a license.

Short of someone actually trying to prevent Conti from making TCUs, there's no antitrust standing, as the United States District Court for the Northern District of Texas already held in 2020--but the Fifth Circuit went beyond and found that Conti even lacks basic Article III standing because it is at best an incidental beneficiary of SEP holders' FRAND commitments and, therefore, in no position to enforce such rights:

"No evidence suggests that Patent-Holder Defendants and SSOs intended to require redundant licensing of third parties up the chain, which is unnecessary to effectuate the purpose of the FRAND commitments and reduce patent hold-up. [...] Continental does not appear to be an intended beneficiary contractually entitled to a license on FRAND terms. And as an incidental beneficiary, it would have no right to enforce the FRAND contracts between the Patent-Holder Defendants and the SSOs."

Pure pragmatism: if it works, it works. If it ain't broke, don't fix it. It's also an effects-based approach. What is the FRAND pledge about? Enabling the making and distribution of standard-conformant products. Does it matter what happens upstream? Not as long as what comes out at the bottom is a licensed standard-conformant product (and the upstream is left alone).

Yesterday's appellate decision is a ringing endorsement of the "access to all" position taken by companies like Qualcomm, Ericsson, and Nokia (all of which are among Avanci's 49 licensors): major SEP holders who decline to license chipmakers and other suppliers argue that if the end-product maker has a license, the supply chain is all set and should shut up. The product can be made, license fees can be paid, customers get what they want.

Like the Ninth Circuit in FTC v. Qualcomm, the Fifth Circuit has effectively determined that SEP holders are free to choose the licensing model that best suits them, and others can't impose their preferences on them. The cases and reasons are different, but the shared philosophical foundation is that patents are property and others can't just force patent owners to do business on any particular terms only because they are the implementers' preferred terms.

The Avanci pool, which Conti chose to attack, simply can't offer component-level licenses unless its licensors authorize it to. It doesn't preclude them from entering into bilateral licenses at any level of the supply chain, and such deals have indeed happened. In the end, all that Conti can accuse Avanci of is that it has (after years of hard work) established a successful licensing model, which obviously does nothing to discredit end-product-level licensing, but Conti can't fault Avanci for not granting a license the actual patent holders have never authorized it to grant. What does Conti want Avanci to do? Should Avanci not offer a pool license that some of Conti's own customers have meanwile embraced, only because Conti has a different vision?

MPEG LA, a pool administrator that is actually different from Marconi (the company behind Avanci) in various ways, supported Avanci with an amicus brief because it was rightly concerned about the damage that a decision in Conti's favor could have done to the patent pool business at large. MPEG LA must be relieved. High fives in the patent pool industry for he Fifth Circuit opinion.

At a time when Avanci is working on a 5G SEP pool, it clearly has momentum on multiple fronts:

By contrast, Conti's legal department has wasted many millions on a U.S. litigation campaign that it mismanaged from the start. The basis on which the Fifth Circuit decided potentially spells doom for a later-filed Continental v. Nokia case in Delaware. Not only has Clueless Conti been throwing good money after bad for years, trying to salvage a case that was fundamentally flawed from the start (I already raised serious doubts in 2019), but its Fifth Circuit appeal has even been counterproductive: if Conti had simply accepted the district court ruling, that would not only have been cheaper but also a less devastating outcome than yesterday's Fifth Circuit decision. Normally, if a party loses in district court, files an appeal, and the appeal fails to defibrillate the case, it's merely been an unproductive exercise. Here, it's even proved counterproductive because

  • the legal basis of the decision has upped the ante with a view to Conti's own Nokia case and any case Conti or other suppliers might have planned to bring (it's like someone just put an extra moat around a castle when Conti couldn't even overcome the castle walls before there was the moat), and

  • one of the most influential and reputable U.S. appeals courts has adopted the purely pragmatic reasoning of the "access to all" camp, which makes this not only a legal but also a political defeat for Conti and, by extension, its amici curiae, above all Apple and Tesla.

In 2019, Conti was even seeking an antisuit injunction in this U.S. case that would have barred Nokia from enforcing German SEP injunctions against Daimler. The Munich I Regional Court entered an anti-antisuit injunction to thwart that attempt to torpedo Nokia v. Daimler, which the Munich Higher Regional Court affirmed--and by now anti-antisuit injunctions are routinely granted in Munich and slowly but surely expanding to other jurisdictions. So if there's one positive thing to say about Conti's various failures in patent litigation and patent policy, it's that some of its actions have led to decisions of transcendental importance, just that not Conti but its targets, above all Nokia, have benefited--and will continue to benefit--on the bottom line.

What can Conti do now? It has the right to seek a rehearing en banc (full-court review). But none of the three judges on the panel thought Conti had a case. One of them, Judge Ho, would have resolved the case at the level of Sherman Act claims instead of standing, but the outcome would have been identical and that's why he didn't even bother to write a dissent or concurrence (it's just a footnote, literally). Can Conti get the Supreme Court interested? If Apple and Tesla aren't tired of losing by supporting Conti, they might try, but they'd have to convince the Supreme Court that a question of Article III standing in a narrow set of cases in which there has been zero cognizable injury (just conspiracy theories and, at best, premonition) is worthy of the top U.S. court's attention. Frankly, it would be one of the worst and weakest stories ever for a cert petition.

After that bird's-eye view, let me show you the ruling and discuss the legal logic in more detail (below the document):

22-02-28 Continental v. Ava... by Florian Mueller

I have to take a step back because with all that was going on, I didn't even find the time to write about the October 7, 2021 appellate hearing in New Orleans (which by the way got postponed because of a hurricane's landfall).

In 2019, the case started in the Northern District of California, but Judge Koh (now sitting on the Ninth Circuit) exercised her discretion to grant a motion to transfer the matter to the Northern District of Texas. In September 2020, a Rule 12 motion by defendants succeeded, i.e., the pleading deficiencies were considered so fundamental that the case could be resolved even far ahead of a summary judgment.

The case also involves claims under state law, but the federal court wasn't going to deal with them if the related federal antitrust claims failed. Now, for there to be an actional--meritorious or not, but at least actionable--Sherman Act case, there are three initial requirements:

  1. Article III standing (without it, you won't get your day in court)

  2. Antitrust standing (a more specific requirement on top of Article III standing)

  3. a Sherman Act Section 1 and/or Section 2 claim

It's an AND combination: if any one of the three is missing, it's game over. And the earliest game-over scenario is the first one, but Judge Barbara Lynn (the Chief Judge of the Northern District of Texas, the largest city of which is Dallas) wasn't going to be all that strict. So she said that Conti had just about managed to plead an injury in the form of being denied what Conti would consider a FRAND license to certain SEPs, and considered this sufficient--though she didn't seem overwhelmed either--to concede that Conti had Article III standing. We now know that the Fifth Circuit disagrees and would have cut the whole story short at that point. But Judge Lynn then went on to address antitrust standing, which Conti lacked in her opinion because the alleged injury (denial of a license) is no antitrust injury, and Conti wasn't the correct plaintiff either (as it's not a car maker). In order to get this part reversed, Conti would have had to persuade the appeals court that it had suffered an antitrust injury and was the correct plaintiff--questions the Fifth Circuit's relatively short (just 14 pages) opinion didn't even (have to) reach.

But even if the Fifth Circuit had agreed with Judge Lynn on basic Article III standing and had reversed her (for which I saw no indication at the hearing) on antitrust-specific standing, Conti would still have had to overcome the dismissal of its Sherman Act claims. Judge Lynn held that there was no Section 1 claim because the Avanci agreement does not preclude licensors from entering into bilateral license agreements with suppliers, and without such restriction, there is no unreasonable restraint of trade. The Section 2 claims were thrown out because even if Conti was right (which the judge seemed to doubt anyway) that the actual patent holders had made fraudulent FRAND declarations at the time of participating in standardization, the competitive process itself wouldn't have been harmed--and the absence of such harm was also fatal for Conti's claims of a conspiracy to create a monopoly. (By the way, the DOJ's Business Review Letter relating to Avanci's future 5G program was clearly a blow to any conspiracy theory.)

It would have been dispositive to just hold that Conti lacked antitrust standing. Judge Lynn went beyond the call of duty by reaching the Sherman Act claims nonetheless, which she may have done not only to make her decision appeal-proof but also because she thought those claims were just blatantly meritless.

Conti probably knew that it might not be able to solve all those problems in one fell swoop. Its Plan B was to be granted leave to amend its complaint again, which is a possibility when deficiencies don't appear incurable and would have been a way for the Fifth Circuit to duck certain questions if it had wanted to. After the hearing, I was pretty certain that Conti was going to lose, but there was some residual doubt that maybe they'd be granted another bite at the apple if they were lucky, especially since Judge Stewart (the most senior judge and only Democrat on the panel) appeared cautiously receptive to the idea, unlike the two Trump appointees (Judges Ho and Engelhardt).

It then took the Fifth Circuit almost five months to converge on a decision. Judge Stewart wrote the opinion, which is fully supported by Judge Engelhardt and which Judge Ho can live with, though he'd have been the only one of the three to even reach any antitrust-specific questions (with Conti ultimately losing anyway).

The technical effect here is that the district court's judgment is vacated as the appeals court overruled the lower court on the question of Article III standing, and remanded to Dallas with the direction to dismiss the case for lack of Article III standing. It's not a "right for the wrong reasons" holding if one thinks it through, as the Fifth Circuit didn't conclude that the reasons for which Judge Lynn dismissed the case were wrong per se: it's that she reached the right result but unnecessarily reached questions she didn't have to. It's more like a "could have been even more right (and at an even earlier stage of analysis)" type of affirmance of the outcome, though it does say "the district court erred" because it even reached questions it never had to reach. It's like Conti's case shouldn't survive for a second according to the Fifth Circuit's reasoning, but the district court gave Conti a second and a half, which is too long for that kind of claim.

Avanci and its co-defendants would have been just fine with affirmance of Judge Lynn's decision, but the Fifth Circuit said the proper first step of the analysis was to determine whether it had jurisdiction in the first place. Conti's Article III theories of injury came down to the potential pass-through of supra-FRAND royalties. Take Daimler, for instance: that company took a car-level license from Nokia, and then one from Avanci. Conti couldn't--and presumably still can't--point to a single case of actually having had to indemnify a company like Daimler after it took that kind of license. All they said is that it might happen, but conjectural injury doesn't count here.

The other point--and the one that also seemed more important to the district judge-- was the refusal of exhaustive component-level licenses requested by Conti. The legal question here is whether patents holder who declined to extend such a license denied Conti property to which it was entitled. Being denied property one is entitled to would be a cognizable injury in fact, but that's not the case here in the Fifth Circuit's opinion:

Unlike Microsoft in its dispute with Motorola (over WiFi and video codec SEPs) and Broadcom in its Qualcomm case, Conti is not another one of those "parties that must adopt a standard in order to conduct their business." Microsoft was a member of the standard-setting organizations in question; Broadcom was a direct competitor of Qualcomm. And "crucially, it does not need SEP licenses from Defendants-Appellees to operate; Avanci and Patent-Holder Defendants license the OEMs that incorporate Continental’s products."

What follows then is the passage I quoted further above: FRAND works even if Conti doesn't get the license it demands. That reasoning stands on an additional pillar: even if Conti was an intended third-party beneficiary,

"the contracts have not been breached because the SEP holders have fulfilled their obligations to the SSOs with respect to Continental. The supplier acknowledges that Avanci and Patent-Holder Defendants are 'actively licensing the SEPs to the OEMs[,]' which means that they are making SEP licenses available to Continental on FRAND terms. As it does not need to personally own SEP licenses to operate its business, it has not been denied property to which it was entitled. And absent a 'denial of property to which a plaintiff is entitled,' Continental did not suffer an injury in fact."

What sets Conti's claims apart from cases in which parties were denied something they were entitled to and, therefore, had Article III standing, Conti can't demonstrate injury because even without the license it seeks, Conti "receive[s] the benefit of a right conferred by contract even if the contractual right [continues to be] 'denied' directly." As Conti's customers get licensed, there is no problem to be solved here, the Fifth Circuit effectively says.

The Fifth Circuit doesn't take a definitive position on whether OEMs might have standing if they disagreed with Avanci's license terms, or maybe even standardization bodies. But Conti is neither a car maker nor a standard-setting organization.

The appellate decision also involves some technicalities, such as why certain submissions made by Conti would not be considered. For example, a complaint cannot be cured by an opposition to a motion to dismiss. Another example is that Conti itself claimed at some point that its complaint didn't have to be amended. Conti has no one to blame but itself. Both its strategy and its execution were inadequate.

With Ericsson being an Avanci licensor (and a longstanding proponent of end-product-level licensing) and Apple having unsuccessfully supported Conti, and given that Apple and Ericsson disagree on the royalty base (end product vs. smallest salable patent-practicing unit) for cellular SEPs implemented by smartphones and other products, this case here is a proxy war between Apple and Ericsson on issues that are adjacent to, and slightly overlap with, the ones in dispute between the two, and Ericsson has prevailed. Apple is not a big fan of the Fifth Circuit and would rather give the Federal Circuit appellate jurisdiction over the U.S. part of its FRAND dispute with Ericsson. Whether that will happen depends on Judge Gilstrap, who can easily just sever Apple's patent-specific declaratory judgment claims from the FRAND issues.

Share with other professionals via LinkedIn: