Wednesday, May 23, 2012

The German approach to FRAND: let's err on the side of injunctions

German court rulings aren't routinely published on the Internet, and the courts never make complaints and other pleadings available to the public. But U.S. proceedings are more transparent, and German rulings sometimes show up there. Earlier this week, Microsoft and Motorola Mobility, which is now a wholly-owned subsidiary of Google, provided to the United States District Court for the Western District of Washington certified translations of two decisions the Mannheim Regional Court simultaneously announced on May 2, 2012. Those decisions relate to two different patents, both of which were deemed essential to the H.264 video codec standard and infringed by certain Microsoft products, particularly including Windows 7 (along with the Internet Explorer 9 and Windows Media Player) and the Xbox 360. For another six months or so, Motorola cannot enforce those rulings due to a preliminary injunction ordered by the aforementioned U.S. district court.

I have uploaded the two certified translations to Scribd:

  • The first document contains both a certified translation and, thereafter, the original German text of one such ruling.

  • The second document contains only a translation of a ruling related to another H.264 patent.

The FRAND-related parts of both rulings are consistent. However, the first translation was provided by Motorola, and the second one by Microsoft. That's why the English versions use different terminology.

The Mannheim court's position on the FRAND issues raised by Motorola's enforcement of apparently H.264-essential patents is symptomatic of the patent enforcement extremism that attracts so many litigants, including some flagrant antitrust offenders, to this jurisdiction. I see that attitude all the time, and I've talked about it before. But thanks to the certified English translations of those German rulings, many more people now have an opportunity to experience this FRAND-hostile environment page by page, paragraph by paragraph.

In the following, I'll sum up how Judge Dr. Kircher and his colleagues arrived at the decision to order a Germany-wide injunction against Windows 7, the Internet Explorer 9, the Windows Media Player and the Xbox 360 gaming console (to the extent that those technologies implement the H.264 standard).

After identifying an infringement (Section A of the reasoning), the ruling addresses the question of whether a license agreement, or at least an agreement that precludes Motorola from seeking injunctive relief, exists (Section B), followed by an analysis of antitrust considerations (Section C). The remaining sections then explain some decisions that result from the infringement finding in combination with the dismissal of all defenses.

At that high level, there's nothing special about this structure. FRAND-related defenses are usually a combination of contractual and antitrust claims (for example, this also applies to Apple's FRAND-related defenses and counterclaims against Samsung in the Northern District of California). The only Germany-specific aspect of this structure is that an invalidity argument can only result in a stay of an infringement proceeding but not serve as a defense in its own rights. That's due to bifurcation, a major structural deficiency of German patent litigation that I explained in a recent post.

The section on contract-based defenses is consistent with findings by other jurisdictions to the extent that it doesn't consider a FRAND pledge to constitute a license agreement per se, but it doesn't stop there: the decision furthermore denies that the FRAND pledge Motorola made to the ITU has any contractual significance, and holds that it "cannot be understood as a binding license offer to any number of third parties unknown to the Plaintiff, requiring only acceptance by a third party, but as a request to license seekers to submit their own FRAND offers". The last part of that is, in my view, a highly questionable application of the German Orange-Book-Standard logic to those pledges. It's highly questionable because the Orange-Book-Standard ruling, as this Mannheim decision notes, "referred to a de facto standard" as opposed to a standard set by a consortium that requires participants to make a FRAND promise.

The only kind of hard and fast legal significance that the Mannheim ruling attaches to Motorola's FRAND promise is that "due to the voluntary obligations assumed by the declarer to provide a license free of discrimination and abuse, it renders the antitrust review as to the patent owner's market-dominating position obsolete". Frankly, I don't see how that can be correct if everything else the Mannheim court says about FRAND declarations (meaning that they don't grant contractual rights to third parties) is true. It doesn't make sense to me. Those FRAND declarations don't say "I have a dominant position in a relevant market under antitrust law". They just say something like "I am prepared to grant licenses on FRAND terms". A dominant position creates a FRAND obligation without pledge, but one can also make a FRAND promise without having an obligation due to a dominant position. If I made a promise on this blog that I will grant everyone a FRAND license to my posts here, I still wouldn't have a dominant market position -- and I still wouldn't have an obligation under antitrust law as long as I don't have a dominant market position. The bottom line is that certain German judges go out of their way to deny that FRAND promises are binding, and they try to apply Orange-Book-Standard to more than it originally related to.

In the contract-related part, the decision also says it's "implausible" that a FRAND pledge constitutes a "waiver by the patent holder of injunctive relief as a means of enforcing its patent claims against an unknown number of potential patent infringers" because "[a] patent holder who submits a Patent Statement and Licensing Declaration Form merely offers to waive its exclusivity rights under the patent by establishing a license agreement--and not unconditionally". But it isn't really "implausible". If someone promises to grant everyone a license on FRAND terms, he does give up on the notion of exclusivity -- and as a result, there shouldn't be any injunctions since all the parties have to talk about is money. That is fundamentally different from a patent that a patent holder can retain exclusively, or license only to a limited number of parties.

A passage in the section on antitrust-related considerations makes it quite clear why certain German courts don't think outside of the box in connection with industry standards:

"If the seeker of the license were in a position to successfully defend against claims for an injunction by the patent owner by arguing that the latter was obligated to grant a license anyhow, on its own volition, the patent owner would be at the mercy of any dishonest licensee, for whom there would be no more incentive to enter into licensing negotiations [reference to literature]."

This is the problem with certain German courts: they think that licensees can only be kept honest at the threat of an injunction. The claim of "no more incentive" is wrong. Even though German patent law doesn't provide for a tripling of damages for willful infringement, it still wouldn't be advisable for anyone to just keep infringing only because there won't be an injunction. Patent damages in Germany can be based on the equivalent of a license fee, on the lost profits of the right holder, or lead to a disgorgement of infringer's profits. As a result, damages for infringement can far exceed the cost of licensing. Moreover, Germany has a "loser pays" system under which an infringer who loses a lawsuit would end up paying all court fees as well as the other party's attorneys' fees (not actual fees, but an amount that is relative to the amount in dispute, and those costs are also quite significant). Also, patent infringement can be a crime in Germany: not the average inadvertent infringement, but if someone kept infringing after a final non-appealable court ruling, there would be a clear disincentive that is stronger than anything civil law can do.

It's this kind of thinking -- the value of a patent depends on the entitlement to injunctive relief -- that makes certain German judges think that it's better to grant 1,000 unjustified injunctions than to deny injunctive relief in only one case in which it would be warranted. This approach ignores economic realities. It's dogmatic. It's old-fashioned. But until lawmakers take action, it might stay this way.

The attitude I just described is also reflected by the antitrust-related part (Section C of the reasoning) of the decision. The 2.25% royalty demand made by Motorola is just restated. It's neither endorsed nor criticized. The discussion then focuses on whether Microsoft's counterproposal is so good that, under Orange-Book-Standard, Motorola can't refuse it without committing an antitrust violation of the kind that represents an obstacle to an injunction. Motorola's various objections to Microsoft's counterproposal are given significant weight even though Microsoft actually multiplied the base figure, which was directly derived from the terms of the MPEG LA pool, by a factor of 10 -- such a generous multiple that a court could easily, and in my view should, conclude that there's plenty of margin of error to offset any of Motorola's relatively minor objections. For example, if Motorola claims that its patents are more valuable than the ones in the MPEG LA pool, that's one thing -- but ten times more?

The ruling mentions that Microsoft provided an expert report according to which its offer was far more than a FRAND royalty. Instead of appointing a court expert to look into the reasonableness of these terms, the Mannheim court decided to just rush to an injunction:

"In the judicial division's view, this review can be limited to the question of whether the rejection of the offer constituted an obvious violation of antitrust law. As the BGH made clear in its Orange-Book-Standard decision, the infringement proceedings are to be relieved to an avoidable extent of the difficult and time-consuming task of determining the exact amount of a non-obstructive or non-discriminatory licensing fee (BGH, supra, margin no. 39). In the case of a license offer subject to Section 315 BGB, the infringement court should thus be able to limit itself to finding that the patent owner is obligated to accept the offer of a license agreement and to a determination based on discretion if the prospective licensee has deposited an amount that in any event is sufficient and if the other requirements of a 'compulsory license objection' are also present (BGH, supra, margin no. 40). In connection with the question of whether the deposited amount is sufficient for sure (in any event), a summary review by the court is enough (see Kühnen, supra, margin no. 1296)."

In that passage, two things are particularly noteworthy. The first sentence emphasizes that only an "obvious" (in the other certified translation: "blatant") violation of antitrust law is a basis for denying an injunction. This means that German judges have no qualms about giving antitrust violators the immense leverage that an injunction constitutes -- they should just ensure that their violations aren't "obvious", or "blatant". The other word I'd like to draw your attention to its "summary" (before "review"): instead of a diligent analysis of whether a defendant's licensing offer is too good to refuse without violating antitrust law, the defendant has to offer so much more than the actually reasonable rate that it's a no-brainer for the court. But even Microsoft's factor of 10 (as compared to the MPEG LA pool rate) wasn't enough to make this a no-brainer. So what will defendants have to do? Offer 100 times the FRAND rate just so the court doesn't have to think about it? I'm sure there are certain German judges who would grant an injunction even if a factor of 100 was applied.

Further above I said that this ruling reflects patent enforcement extremism. You've now seen that this means it's fine for patent holders to get far higher royalties than the ones they're entitled to (or, if a would-be licensee doesn't specify a counterproposal and agrees on a FRAND rate subject to judicial review, he has to post a bond or make a deposit far in excess of the appropriate amount). And more than anything else, it means that in case of doubt (even if that doubt is hardly reasonable), an injunction over a stanard-essential patent will issue.

It's noteworthy that the Mannheim court granted Motorola this injunction even after the European Commission opened two formal antitrust investigations of Motorola's related conduct.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: