Context: In 2021, Germany adopted an amendment to its national patent law that, besides purely technical changes, had two key elements. The most hotly-debated topic was an amendment to the patent injunction statute, which theoretically comes with a proportionality requirement but, as of now (almost three years after its entry into force), has not made an impact on a single case. It is evidently ineffectual. The second-most important item was a rule that is deficient in other ways and has made an impact on a number of cases, but as understood early by experts, is inherently unreliable as it merely suggests to the Federal Patent Court that it would be desirable to obtain a preliminary opinion on a nullity complaint within a six-month window (and that opinion does not even have to be useful, though in practice it typically is).
What’s new: As announced on Wednesday (May 22, 2024 ip fray article), further research was necessary to verify that last week’s Atlas Global v. TP-Link standard-essential patent (SEP) in Munich is the latest example of a failure of the 2021 German patent “reform” bill to achieve its desired effects. ip fray has now obtained enough information to be able to say with certainty that the statutory six-month target was not met and the injunction issued without the court having, or electing to await, the benefit of a preliminary opinion by the Federal Patent Court.
Direct impact: It is not clear whether the Federal Patent Court will or will not voice doubts about the validity of the patent-in-suit. In case it does, the injunction would not have issued if the opinion had been provided within the six-month window.
Wider ramifications: The failure of the 2021 German patent “reform” bill is instructive with a view to the current debate over the EU SEP Regulation. One cannot blame policy makers in the executive branch, and much less lawmakers, for their lack of understanding. But it would be their duty to obtain reliable expert advice, which is admittedly more difficult in such a highly specialized field than in some other areas of policy making. In the German patent reform debate, there were essentially two camps: one that wanted to dilute it (and therefore had no motivation to highlight certain shortcomings) and one that just wanted some kind of result, even if of no or limited value.
As a spokeswoman for the Munich I Regional Court confirmed to ip fray, the judgment in case no. 7 O 4832/23 over EP3295571 (title: “pilot transmission and reception for orthogonal frequency division multiple access”) was announced on May 16, 2024.
The infringement action was filed with the Munich court on April 17, 2023. The defendant brought its nullity complaint on September 8, 2023 in accordance with statutory requirements (in time for its answer to the infringement complaint). According to the patent register of the German Patent and Trademark Office, the nullity complaint was deemed to be served on the defendant as per November 7, 2023.
It is unclear why service of process within Germany (given that the nullity complaint was served on the patent attorneys representing the patent holder) took two months. There may have been court-internal reasons, but it’s also possible that the patentee played a delay game, which happens quite often.
At any rate, the six-month period started on November 7, 2023, so the preliminary opinion should have been rendered by May 7, 2023, which would have been between the trial and the judgment in the infringement case.
Let’s now look at what the statute says and discuss how an intelligently-designed statute could have prevented this situation. First, the German-language version of the relevant passage of the statute (§ 83 of the German Patent Act), followed by an unofficial translation:
“(1) In dem Verfahren wegen Erklärung der Nichtigkeit des Patents oder des ergänzenden Schutzzertifikats weist das Patentgericht die Parteien so früh wie möglich auf Gesichtspunkte hin, die für die Entscheidung voraussichtlich von besonderer Bedeutung sein werden oder der Konzentration der Verhandlung auf die für die Entscheidung wesentlichen Fragen dienlich sind. Dieser Hinweis soll innerhalb von sechs Monaten nach Zustellung der Klage erfolgen. …”
Unofficial translation:
“(1) In the nullity action concerning the patent or supplemental protection certificate, the Federal Patent Court informs the parties as early as possible of the aspects that are foreseeably of particular relevance to the decision or useful for the purpose of focusing the trial on the outcome-determinative questions. That notice shall be given within six months of service of the [nullity] complaint.”
What was not the issue in this case, but constitutes another deficiency, is that the statute only applies to proceedings before the Federal Patent Court, but not to nullity actions before the European Patent Office (which wouldn’t be bound by a German law anyway) or the German Patent and Trademark Office.
What could have become a widespread issue (but isn’t yet) is that the first sentence merely requires the Federal Patent Court to provide an outline of the issues in the case. The Federal Patent Court could do that without the slightest indication of whether a patent is likely valid or invalid, in which case the infringement proceedings would not be impacted and the so-called “injunction gap” would not be addressed at all. But the judges of the Federal Patent Court tend to communicate their views reasonably specifically, beyond what the statute strictly requires them to do.
The biggest problem is the one that has now deprived TP-Link of a potential opportunity to obtain a stay of the infringement case: the word “soll” (which is typically translated as “shall”).
Interestingly, “shall” is one of the most-litigated terms in English-language jurisdictions, whether in a statutory or contractual context. The question is often whether it means “would you please?” or “you must!” But at least there can be a dispute. Not so with “soll”: that word is clearly understood in German law never to mean “you must!”
In practice, “soll” means “please try to make it work.” It’s a target, not a deadline.
What could the German legislature have done to avoid this problem?
The starting point of the legislative debate was the injunction gap: infringement actions typically conclude in Germany long before there is a first-instance nullity decision. And before that revised statute, the Federal Patent Court provided its preliminary opinions only late in the process, much closer to the nullity trial date than to the filing of the complaint.
But the real issue was not even that the Federal Patent Court needed more time. The cause of the injunction gap problem is that German patent case law, in no small part with meanwhile-retired Judge Dr. Thomas Kuehnen as the thought leader, set an unreasonably high hurdle for stays of infringement actions pending the resolution of a nullity action (or opposition proceeding in a patent office).
The most logical solution would have been to change the legal standard for stays. One organization named Patentverein (“Patent Club”), whose members are mostly small and medium-sized enterprises, has publicly stated that a stay should be mandatory. That would go too far. But the 2021 patent “reform” bill could have established a more reasonable standard than the one under the case law without requiring automatic stays.
The second-most logical solution would have been to make a stay mandatory until the Federal Patent Court (or a patent office where the patent is challenged) renders a preliminary opinion on validity.
Instead, the German ministry of justice and the legislature opted for an unprofessional approach that computer programmers call a “race condition”: they just assumed that with a six-month target, those opinions would come down in time before an injunction would issue. In this context, “race condition” means to rely upon the timely conclusion of a parallel process instead of actually ensuring that the other process has indeed done its job, such as by waiting for a message.
That unprofessional statutory structure invites gamesmanship for the purpose of delaying the start of the six-month window. And combined with the fact that it’s not a hard requirement but just a wishful-thinking type of target (“oh, would you be so kind as to get it done in six months?”), it doesn’t work reliably.
Since the six-month target kicked in (which was in 2022), many German patent infringement cases have been stayed that probably wouldn’t have been stayed without it. So unlike the botched injunction statute, it does produce an effect, and even quite often. But not reliably as this case here shows (and it’s far from the only one based on what ip fray has heard from the patent litigation community).