Munich appeals court rejects German approach to standard-essential patents, tends to pave way for next ECJ ruling on FRAND

European Commission 1 – Retired Judge Meier-Beck 0.

While ip fray does not believe that HMD (a company often, and not incredibly, accused of hold-out) deserved the EC’s support, the problem is that too few SEP cases actually reach the German appeals courts. Patent hold-up forces even the most willing implementer one could imagine into a license agreement on supra-FRAND terms. Both hold-up and hold-out are a reality. A well-balanced legal framework would strike a reasonable balance. Germany, however, has a history of the pendulum swinging from one extreme to another:

• In the Orange-Book-Standard era, every implementer lost, and the reasons given by the courts were sometimes utterly unreasonable.
• Then came the first few years after Huawei v. ZTE where SEP holders found it incredibly difficult to obtain injunctive relief in Germany.
• After a while, German patent judges in different places decided to gut Huawei v. ZTE, and felt vindicated by Sisvel v. Haier. Virtually every implementer’s FRAND defense have been rejected over the course of several years.

The Munich appeals court would like to strike a balance. Presiding Judge Meinhardt and Judge-rapporteur Baumann (who looked rather young for an appeals court judge) may have surprised the parties and the European Commission by provioding their preliminary opinion, in preparation of the hearing, only sometime yesterday afternoon or evening. And some of what they said during the first part of the hearing looked like they were looking for a simple, not really well-thought-out exit. But it became clear that they actually are committed to the quest for a workable solution. The third judge, Judge Sander, did not say anything. She has the rank of a Local Court Judge, which is formally below the regional court from which this case originated.

With the greatest respect, this is not a “seasoned” appellate panel. Judge Meinhardt had virtually no patent expertise when he took over in early 2022, and the two side judges are very young. But they credibly came across as a panel that wants to address a problem that they have identified and that objectively exists. It’s just that unless they distance themselves from the fix they proposed today, they would simply trade in one set of issues for another:

Deposit or injunction?

The appeals court’s idea at the outset of the hearing was that an implementer who meets some minimum requirements ahead of trial (though HMD would now be allowed to do so after the appellate hearing as this is new law) just has to make a deposit to the amount of the royalty demanded by the SEP holder, and on that basis the court would proceed to a full-blown FRAND determination with expert witnesses and what have you.

The court’s idea is to enable licensing negotiations free from the threat of an injunction.

HMD’s lead counsel Lars Baum from Hoyng Rokh Monegier raised the legitimate concern that this might encourage SEP holders to ask for the moon, betting on an implementer’s inability to make an outsized deposit. Presiding Judge Meinhardt said that this was not the idea. In the event of a blatantly absurd royalty demand, they would dispose of the case on the basis of bad faith. HMD still expressed the concern that if someone faced multiple assertions (such as parallel litigation by multiple members of a patent pool) and every SEP holder asked for a supra-FRAND royalty, that could add up to a prohibitive aggregate amount.

VoiceAge EVS’s lead counsel Peter-Michael Weisse of the Wildanger firm was even less enthusiastic about this proposal than HMD. He warned against the risk of bad-faith implementers simply declining to take a license even on perfectly FRAND terms if they knew that they could always avert an injunction by merely making a deposit at trial time and delaying resolution of the case for a long time.

Mr. Weisse and, from the same firm, Jasper Meyer zu Riemsloh understandably advocated the interests of SEP holders in getting leverage over whatever the courts would consider to be unwilling licensees. They denied that the current system leads to license agreements on supra-FRAND terms as the result of hold-up (a problem that the appeals court correctly identified) and they downplayed other issues that exist. But it was indeed very interesting to hear from Mr. Weisse that total litigation expenses for the dispute between VoiceAge EVS and HMD were estimated to amount to 12 million euros, which he described as a substantially greater amount than the licensing costs that HMD would have incurred if they had accepted an offer from VoiceAge EVS three years ago.

Judge Meinhardt repeatedly brushed aside Mr. Weisse’s concern by saying that if the courts found an implementer to be a willing licensee under the current standard, they would also have to proceed to a FRAND determination. What Judge Meinhardt hopes is that the parties would typically prefer to reach an agreement at the negotiating table. He was disarmingly honest when he asked this rhetorical question:

Do you believe we’re eager to make a FRAND determination?

And he answered it in the negative, by himself.

Two fundamental issues with the appeals court’s idea that SEP holders would face

In a way, Judge Meinhardt sounded a bit like Mr Justice Birss in Unwired Planet v. Huawei with his famous “whatever terms are in fact FRAND” statement. In particular, Judge Meinhardt not only wants an implementer to commit to pay the royalty demanded by the SEP holder if the court ultimately validates it as FRAND, but also to accept the entire set of terms and conditions offered by the SEP holder.

Still, there is a fundamental difference between the hypothetical Munich approach and the way it works in the UK:

In the UK, an implementer seeking to avoid the enforcement of an injunction must commit to take a license on court-determined terms, so the question is not whether the terms offered by the SEP holder are frand, but what terms are in fact FRAND. The contemplated Munich approach, however, would result in a binary outcome:

• If the SEP holder prevails at the end of the FRAND proceedings, the implementer must take the license.
• In the other event, however, the SEP holder is denied an injunction and might have to file a new case, based on a revised offer, to reach the same point again. In the meantime, the patents-in-suit might even expire.

The other reason for which the appeals court’s ideas don’t sufficiently protect the interests of SEP holders is the fundamental difference between actual license fee payments (that physically arrive on a bank account and are processed by the accountants as revenues) and mere collateral:

• Publicly traded companies need revenues they can report. Deposits or bonds are ineligible for that.
• The same applies to companies that have institutional investors (be they venture capital funds or litigation funders) that can’t report to their own investors anything other than actual results.
• And companies that depend on licensing income in order to fund the next round of innovations need to physically receive money, too.

For those reasons as well as others, ip fray does not believe that the Munich Higher Regional Court is on the right track. But what was said today is not necessarily the last word.

How sequential is sequential?

Just like the appellate panel had limited exposure to SEP cases (Judge-rapporteur Baumann acknowledged that so far they had basically just seen emergency motions to stay enforcement), the European Commission’s in-house and outside counsel is also somewhat new to the topic.

Professor Moritz Lorenz of Arneth Sibecke Dabelstein has not been seen in German sEP litigation so far and had to consult with two lawyers from the European Commission’s Legal Service, who were seated right behind him, after virtually every question the appeals court asked.

The two EC staff attorneys present today, Clio Zois and Pascal Berghe, may not have been involved with SEP cases so far, but appeared very knowledgeable, particularly Mrs. Zois based on how she always nodded strongly during the most important parts of Professor Lorenz’s pleadings.

The EC’s position on the requirements for an infringement notice, as stated in the amicus brief, appeared overly strict, but the emphasis today was placed on the second and third steps of the Huawei v. ZTE analysis. The Commission left no doubt whatsoever about the fact that an implementer’s subsequent behavior cannot justify an unwillingness finding at the second step, where all that the implementer has to do is express its wish to take a license on FRAND terms. And the EC said repeatedly and consistently that after the first two steps of Huawei v. ZTE (the easy parts), the court must proceed to a determination on whether the SEP holder’s offer is FRAND.

The EC disagrees with the appeals court’s preliminary opinion: no deposit by the implementer is needed. The court must assess whether the SEP holder’s offer is FRAND, period.

The EC unequivocally rejected the notion that an implementer would have to make a deposit of an amount that could be supra-FRAND for lack of the court having made that determination.

In these respects, ip fray agrees with the Commission.

But at today’s hearing, the appeals court and the Commission may have expressed different concepts with the same terminology when they seemingly agreed that neither the SEP holder nor the implementer should have only one opportunity to make a FRAND offer. The EC’s outside counsel confirmed, but may want to nuance in the post-hearing brief, that the last offer counts. That would still mean that the Huawei v. ZTE analysis would be strictly sequential: step 1, step 2, step 3, step 4. But when does the door close? Here, the EC should have reiterated what it said in its amicus brief, which is that certain steps must be taken before infringement litigation commences.

How to deal with the FRAND range

One fundamental disagreement between the court and the EC relates to a question on which HMD’s lead counsel, very reasonably, said there was room for different views. That question is whether a SEP holder should be denied an injunction if it makes an offer at the upper end of the FRAND range and the implementer’s counteroffer is below that one, yet within the FRAND range.

Judge-rapporteur Baumann illustrated the problem very well:

• The court finds that the FRAND range is from 70K to 100K.
• The SEP owner asks for 100K.
• The implementer offers 70K.

Here, again, the EC’s oral argument was not as clear as its amicus brief. The amicus brief, particularly the flowchart that visualizes the step-by-step Huawei v. ZTE test, made it perfectly clear that even a SEP holder who makes a FRAND offer will fail to obtain injunctive relief if the implementer’s counteroffer is also FRAND.

It was very smart and fair of HMD’s lead counsel to acknowledge that this is a question on which people may have different views. It was even smarter of him to limit the amount of things he said once it was clear that the court would allow post-hearing briefs (which is unusual, given that such appeals courts in Germany very often rule from the bench and very rarely allow post-hearing briefing).

Why the Presiding Judge struggles with the notion of denying injunctive relief if both sides make a FRAND offer

Judge Meinhardt, who is more direct than the vast majority of German appellate judges (for instance, he even alluded to the unsafe-after-dark area in which the courthouse is located when he encouraged people to come to an end) was very outspoken about his reluctance to hold that a SEP holder who made a FRAND-compliant offer should be denied an injunction.

He said that patents are a property right, a patentee is entitled to an injunction to stop infringement (though he would merely want to see a deposit to view it differently), and a SEP holder whose offer was FRAND has not violated antitrust law, thus should be entitled to injunctive relief.

Again, from a policy point of view, that is a reasonable position to take as even HMD did not dispute. But it’s misplaced. It’s wrong for three independent reasons:

1. Judge Meinhardt’s thinking is static while the world around him is dynamic. He focuses on whether the SEP holder has already breached antitrust law. What the ECJ had in mind in Huawei v. ZTE, however, was to prevent any such breach in the first place. Now, again, one can question the wisdom of presenting the SEP holder with two hurdles: the SEP holder loses if its own offer is supra-FRAND (even if the implementer makes no counteroffer at all, though HMD’s counsel was unsure of that), and it also loses if, after its own offer is found to be FRAND, the implementer makes a FRAND counteroffer. But it is not an unreasonable position per se to say that if the implementer is willing to take a license on FRAND terms (albeit the lower end of the FRAND range), obtaining and, even more so, enforcing an injunction would be an abuse of market power.
2. Judge Meinhardt erroneously kept insisting on German case law that doesn’t really matter here. Why doesn’t it matter? He can flush the entire body of German case law, including every BGH decision in the history of that court, down the toilet if he wants, but he cannot disregard EU law. The fact of the matter is that he is not bound by the Federal Court of Justice’s decisions. Not in the slightest. There’s psychological pressure, group pressure if you will. But there’s no such thing as a formal legal obligation. By contrast, EU law, including ECJ case law, is indeed binding.
3. The case law he kept insisting on is what he referred to as the dolo agit affirmative defense to injunctive relief. ip fray explained that one earlier this year (March 1, 2024 ip fray article). It goes back to the Orange-Book-Standard ruling, which was similarly bad as Sisvel v. Haier, though partly excused by the fact that it came many years before Huawei v. ZTE. It’s this concept that a plaintiff will not be granted a remedy if that means an instant violation of another law. Even under the dolo agit logic (if one actually translates the full Latin sentence that starts with those two words), one could argue that a SEP holder must not obtain injunctive relief against an implementer willing to pay a FRAND (albeit the low end of FRAND) royalty. But dolo agit is not the only way under German statutory law that the Munich Higher Regional Court and other courts could do what they must do (I repeat in all caps: MUST), which is to comply with EU law. Dolo agit is just BGH case law: absolutely unimportant when the EU dictates something else. These would be the options for German courts to do the ECJ’s bidding:
   • They could point to the Treaty on the Functioning of the European Union, period. There’s nothing else that they have to do. Article 102 TFUE, period. That’s what Professor Lorenz tried to explain to the court.
   • They could invoke § 242 of the German Civil Code. Judge Meinhardt himself mentioned that statute today for scenarios in which a SEP holder would ask for a clearly supra-FRAND royalty.
   • When the BGH established the dolo agit doctrine in a SEP context, the German patent injunction statute (§ 139 of the German Patent Act) was an undifferentiated and unqualified injunction statute. That changed in August 2021 when a patent “reform” bill was signed into law and published in the Federal Law Gazette. That modified injunction statute has been totally useless in other contexts: not a single case is known, after more than three years, in which any German patent infringement decision had a different outcome because of that legislative amendment. But that’s not the issue. In the SEP context, it gives German courts a third option (which they really don’t need) to act in compliance with EU law and deny injunctions in certain circumstances.

One of the lawyers representing HMD today told the appeals court that BGH case law is not binding while ECJ case law is. That’s simply true. Germany is not a common law jurisdiction. It’s a civil law jurisdiction.

Preliminary reference standard

Judge Meinhardt said the panel was so far disinclined to grant the EC’s wish, which was reiterated today and reinforced by HMD, to make a preliminary reference to the ECJ. He said that his court, a court of second but not final instance, could exercise its discretion one way or the other, and would probably leave it to the BGH. The BGH, as he noted, would have to make a preliminary reference.

ip fray believes the Munich appeals court’s reluctance to make a preliminary reference to the ECJ is legitimate, but inappropriate. As counsel for HMD also noted, the EC is the Guardian of the (EU) Treaties. The appeals court should not disagree with the EC but decline to make a preliminary reference. It legally can, but it would be a bad idea.

In the end, it would just prolong the period of uncertainty that has begun with today’s hearing. There are now divergent opinions. As HMD’s counsel and the EC made clear, their views overlap to a large extent, but are not 100% co-extensive. Then there are VoiceAge EVS’s positions, and even VoiceAge EVS would like some more time to digest the appeals court’s new ideas. And the appeals court wants to go in a direction in which no other court has gone.

There is hope that the appellate panel will realize that this is a tricky question. It has conflicting goals. The judges want to stay faithful to the BGH, but they must comply with EU law. They’re psychologically captives of the BGH, but legally they are little more than minions of the ECJ. The judges may also have draw some conclusions from the well-reasoned criticism that VoiceAge EVS, HMD and the EC leveled at the appeals court’s new idea. As Judge Meinhardt quipped, when both sides are unhappy, the court may be right. Politicians often say the same thing. But he wasn’t being totally serious.

ip fray agrees with all of HMD’s and the EC’s criticism of the appeals court’s idea, and with key parts of VoiceAge EVS’s criticism. The appeals court came up with that idea on the eve of the hearing. It doesn’t really solve any problem and just creates new ones. In light of all of that, the appeals court may ultimately conclude that instead of writing a lengthy decision that stands little chance of being adopted as EU-wide guidance, they should just refer the key questions to the ECJ at the earliest opportunity.

Both SEP owners and implementers need a reasonable degree of legal uncertainty. Today is the day that Sisvel v. Haier died. It’s dead case law walking. There may still be some decisions under that one, but there may not even be a single more decision on that basis as the judges sitting on the lower courts in Germany may not want to take any risks now that the EC has made it clear that German case law on SEPs is wrong. It stated it clearly, even if in the cautious language of diplomacy.

The sooner this matter is resolved in Luxembourg, the better for the ecosystem.

Judge Meinhardt, too, may have been diplomatic when he said the BGH didn’t intend Sisvel v. Haier to have the effect it has had, which is that virtually every implementer is deemed an unwilling licensee, often on ridiculous grounds. He may know, or otherwise someone should tell him, that what came out of Sisvel v. Haier is the fully intended effect of an extremist approach to patent law. We’ll probably never know whether all five judges who signed the judgment voted in favor. It’s possible that some lacked the backbone and others the expertise at the time to realize what they were doing. But those who wanted to go in that direction absolutely wanted to shift the focus from the FRANDliness of a SEP holder’s offer to other considerations. The Sisvel v. Haier ruling was clear about that, and the way Retired Presiding Judge Professor Meier-Beck defended it at various in-person, online and hybrid events left no doubt about it either.

Many implementers are indeed unwilling licensees. But not all of them.

Many SEP holders are reasonable licensors. But there are bad actors among them, too.

A jurisdiction that will always gravitate to one extreme or the other must grow up. The EC’s intervention has clearly had an effect. Even though the Munich appeals court doesn’t agree with the EC, it has realized the importance of the matter and it has figured out that something is rotten in the Federal Republic of Germany.

It’s symptomatic that the appeals court didn’t expect the large crowds of lawyers (relatively speaking) who wanted to attend today’s hearing. Many of them had to stand because there wasn’t room for enough chairs. It wouldn’t have been hard to hold that particular hearing in a larger courtroom. of which there are several in Munich. The court just wasn’t aware to the full extent of how big an issue this is.

Now they know. They will get post-hearing pleadings. hopefully also from the EC. They can think about it again. And if they think about the magnitude of the challenge and the relevance of the issue, the best choice they can make is a referral to the ECJ, after giving the parties and the EC the opportunity to propose, and comment on, the specific questions to be put before Europe’s top court.

Having said that, Happy Halloween!