In-depth reporting and analytical commentary on intellectual property disputes and debates. No legal advice.

Right-sizing judicial FRAND determinations: one patent at a time? one country? worldwide? or even pools?

Context: The result of a small LinkedIn poll by ip fray is that a majority of respondents considered last week’s Panasonic v. OPPO FRAND ruling by the Mannheim Local Division (LD) (November 23, 2024 ip fray article) the most important Unified Patent Court (UPC) decision of a week that also saw other significant case law developments even at the appellate level. And next week there will be a UK appellate hearing concerning the question of whether a court could and should set a pool rate (November 15, 2024 ip fray article).

The extreme positions taken in both cases could not be further apart. In one case (UPC), a party argued that a judicial FRAND determination should be limited to only the patent-in-suit. In the other (UK), a party wants a global pool rate set in a single proceeding.

Here are some reflections on the range of options and their practical implications.

One of the most fundamental mistakes that courts can make in adjudicating standard-essential patent (SEP) cases is to believe that they could, and should, fully resolve SEP disputes through their decisions. They should content themselves with deciding cases to the extent that they can do a good job, and leave the rest of the parties.

If both parties ask a court (or an alternative dispute resolution body) to set a rate and commit to enter into a license agreement on those terms, then that can put the dispute to rest. But what if they can’t agree and one party wants to force the other into a license agreement, be it the patentee by seeking a prohibitive injunction against infringement or the implementer by seeking a specific-performance injunction to obligate a SEP owner to grant a license? In such a situation, the best starting point is to understand practical and jurisdictional limits (in geographic and substantive terms).

SEP litigation isn’t useless only because the outcome does not constitute a full substitute for a hypothetical, successful negotiation. If a subset of the issues is resolved, maybe just one issue at a time, the court may provide enough guidance, and the judgment may just build enough pressure, that the parties can handle the rest.

If anything short of global rates was useless, why would SEP holders enforce their rights in, for example, the United States? U.S. courts don’t expect implementers to take global licenses, much less global pool licenses, to avoid an injunction. There will typically be patents in other jurisdictions left to be licensed, but why would a rational implementer engage in hold-out in the rest of the world after resolving the U.S. part of the dispute? It would just mean more litigation expenses (and potentially supra-FRAND damages for past infringement) if the implementer has to expect that other jurisdictions would also find that the patentee owns valid SEPs, and that certain terms are mutually acceptable.

In Panasonic v. OPPO, the UPC’s Mannheim LD saw various alternative proposals by either party at the same time or at different stages of proceeding. OPPO apparently had all sorts of creative ideas for the geographic scope, such as that the terms for North America would be set by the UPC, and the ones for South America in China, which the court somewhat understandably didn’t want to accomodate.

Panasonic’s counsel has meanwhile also clarified that their preference was always for the UPC to rule on a global license deal, but as they have a duty to vigorously defend a client, they also raised a jurisdictional argument that one could perhaps interpret the UPC Agreement to the effect that the court has jurisdiction over FRAND counterclaims only if those relate to a single patent at a time: the patent-in-suit in a given case.

Many of the current problems in FRAND litigation are attributable to territorial overreach, which one could also call cross-jurisdictional encroachment (or. evne more pejoratively, usurpation).

In Huawei v. ZTE, the ECJ did not say that a license offer had to cover a global portfolio (much less that it would have to be an entire pool). There is nothing about scope in the decision. There is merely a reference to recognized commercial practices in the field (as a requirement for how the implementer responds to the SEP holder’s offer). What happened then is that courts in Germany, UK and other European countries

  • saw that no party could reasonably dispute that global portfolio licenses are the customary deal scope, which the courts already knew anyway (from the comparable license agreement that were presented to them) and
  • on the basis of that real-world circumstance and the “recognized commercial practices” wording in Huawei v. ZTE obligated implementers to take licenses of such scope (in order to avert sales bans).

Arguably, all that the ECJ wanted in Huawei v. ZTE was to ensure that the SEP holder would not ask for more than a FRAND rate (and would not impose non-FRAND conditions), and that the implementer’s response would be within the FRAND corridor as well. The decision was neutral about scope.

The Unwired Planet v. Huawei decision in the UK (which was affirmed by the two courts above the High Court of Justice) was not bad for implementers in every respect:

  • Only one part was deeply flawed: the judge just assumed that the SEP holder had overrated the value of its portfolio by the same factor as the implementer had underrated it. They have become way more sophisticated in UK FRAND cases ever since.
  • It was a total invitation to forum-shopping to say that no matter how tiny the percentage of global sales generated by the implementer in the UK might be, the UK should make a global FRAND determination (and, on that basis, decide whether or not to allow the enforcement of an injunction). Now that implementers are trying to engage in forum-shopping in the UK, even many SEP holders who originally welcomed the Unwired decision may not be happy about it anymore. Even before it got to that point, China started setting global rates (and ordering SEP holders to grant licenses on such terms), and SEP holders then cried foul, which was hypocritical.
  • Unlike last week’s UPC Mannheim decision, the Unwired decision focused on terms rather than behavior. Parties have recently tried in some other UK SEP cases to convince the court that the other side’s conduct was totally outrageous and, therefore, the injunction question could be decided even without a FRAND determination. They have all failed so far. The EWHC always proceeded to a FRAND determination. It’s a small sample at this point, so sooner or later it probably will happen that the FRAND determination is skipped over some utterly unreasonable behavior, but the hurdle is high. By contrast, German courts incredibly find every single time that they don’t have to make that effort. And we’re talking about a far larger sample there.
  • Arguably, Mr Justice (now Lord Justice) Birss was almost as implementer-friendly as famous Judge Richard Posner (Seventh Circuit, at the time sitting by designation on a district court) in an Apple-Motorola case. Judge Posner held that injunctions over SEPs should only be available if an implementer, after a court has set a FRAND rate and all appeals have been exhausted, refuses to pay that royalty. That is, actually, the net effect of Unwired: you don’t have to accept anything other than a court-determined rate. The difference is just that UK courts want an implementer to commit to accept that deal before they know what the FRAND rate will be, and Judge Posner didn’t about that because he didn’t even have to.

Now, what does this mean for the UPC situation and for the upcoming UK appellate hearing?

First, the UPC can’t be blamed if it expects implementers to accept global rates, given that this is the standard approach in Europe. But the UPC will then also have to deal with situations in which other jurisdictions use their greater powers to force SEP owners to grant licenses on particular terms, or in which foreign courts interfere with UPC proceedings through antisuit injunctions. The UPC, as a specialized court, may (and probably will) find itself disadvantaged vis-à-vis generalist courts. So the most sustainable outcome for the UPC would actually be if no court made decisions of extraterritorial scope. The UPC would then be one of the most attractive SEP venues, if not the most popular one, and its proceedings wouldn’t be undermined from the outside.

Second, the problem with the Mannheim LD’s decision is not the result of that particular case. They could have reached it in any number of ways. It’s that the court insists on an amalgamated (instead of strictly sequential) analysis and went too far by effectively creating additional interim steps beyond the ones laid out in Huawei v. ZTE, such as requiring the implementer to help the SEP owner to get the original infringement notice right and requiring the implementer to provide sales data to the SEP owner when there was nothing that stopped Panasonic from simply making a per-unit offer (as there was no doubt about the order of magnitude of OPPO’s sales). That rejection of sequentiality and excessive implementer-blaming creates tension now between the UPC and the European Commission until there is appellate clarification, which is not going to happen in a practically settled case like Panasonic v. OPPO though.

Third, when it comes to pool rates, the difference between court rulings and real-world results is even more fundamental. That is something to consider in the context of implementers asking courts to set pool rates and order pool administrators to grant licenses on such terms (regardless of whether the pool administrators even have the prerequisite licenses from their own licensors).

There are various commercial realities surrounding pools that must not be forgotten, among them the following:

  • Pools only serve their purpose if they are simple and efficient. They can only save transaction costs if they streamline the process. If a pool administrator has to go back to licensors every time an implementer makes a special request, the complexity of the process will become similar to, and in many cases even worse than, if the implementer negotiated bilaterally with every single licensor.
  • Pools are typically optional, and that means an implementer who believes to be in a very different situation than the ones for whom the pool terms were designed can still just talk to each and every licensor bilaterally.

The fact that courts can set bilateral rates doesn’t mean it works equally well, or makes just as much sense, for pools.

If a pool has market acceptance, the rate may just be right. And what should happen to all the license deals already in place? Those issues were already raised in a November 15, 2024 ip fray article. But there are also issues that are closely related to some of the ones addressed further above:

Only because pool licenses are often the real-world outcome doesn’t make pool rates necessarily amenable to judicial determination. If courts already find it difficult to simulate (or assess the behavior in) bilateral FRAND negotiations, how could a court possibly understand the dynamics of discussions between a pool administrator with a diversity of implementers on one side and a diversity of SEP holders on the other side? Complexity increases exponentially in that scenario.

Another challenge is that in a litigation between one implementer and the pool administrator, the vast majority of the companies who have the relevant facts are not going to be parties to the proceedings:

  • One implementer with its unique situation can’t speak for all others. For example, a FRAND rate that is fair for Tesla to pay is not necessarily the fair rate for the whole range of automakers from the KIAs of the world to Lamborghini.
  • Even if one or a couple of licensors are included in a litigation, no implementer is realistically going to sue all contributors to a large pool. And even if it sued all the existing ones, what about patent holders that haven’t joined the pool yet? Pools attract new licensors all the time. Pool administrators therefore design their terms to work in a dynamic situation.

All that this article means to advocate is that sometimes it will be better for courts to say “we give you this much guidance (or, to the plaintiff, this much leverage), but the rest you have to sort out between the parties.”

Reading too much into “recognized commercial practices in the field” (Huawei v. ZTE) has not been bottom-line positive.