Indian SEP ruling penalizes textbook holdout primarily through fee-shifting: Ericsson v. Lava

In order to save you time, ip fray has read and digested the FRAND part of the decision (approximately 120 pages) and would like to highlight just a few key observations:

F/R distinction: On page 353, the judgment seeks to explain what “fair” means versus what “reasonable” contributes to the term FRAND. It is, however, hard to see how a licensing proposal could be unfair and reasonable, or fair but unreasonable. In the United States, the concept was traditionally called RAND, but as more and more cases reached U.S. courts that involved ETSI standards, FRAND became common. In the end, the “F” merely makes the acronym more pleasing on the ear. In contracts, terms like “commercially reasonable efforts” are common, and no one needs to add “fair” to give them meaning. One could even have a debate over whether a discriminatory set of terms can be fair and/or reasonable, but the ND part of the term comes in handy in a scenario where someone offers a royalty rate that may be acceptable per se, but does not charge (or charges a lot less) to others.

Lava’s extreme holdout: So far, the SEP ruling that illustrated hold-out behavior more than any other was Sisvel v. Haier. But “you ain’t seen nothing yet” until you see some of the most absurd aspects of Ericsson v. Lava. It took well over a year before they finally signed an NDA, and actually the only change that had to be made to the text of the NDA during a long period was to change the forum (presumably the one for potential disputes arising under the NDA) from Sweden to Singapore. Worse still, Lava was unable to show any license agreement the company had concluded with any other holder of SEPs reading on the same standard, a sign of free-riding as Judge Amit Bansal noted. Lava rejected all of Ericsson’s offers, even if the terms had been accepted by local competitors like Micromax, as being allegedly supra-FRAND, but never once cared to specify what would be FRAND. That (and more) led the court to note that such behavior “poses a challenge to the integrity of the FRAND framework, affecting the overall ecosystem of SEPs.”

Legal standard: Aligned with the bulk of global case law (the ruling refers to decisions in various jurisdictions), the judgment rejects the notion that the smallest saleable patent-practicing unit (SSPPU) should be the royalty base, as well as the position that only the eight patents-in-suit (one of which was held invalid) should be licensed. It has to be a portfolio license, and interestingly Lava brought a suit of its own against Ericsson (even before the infringement action began) that also related to all of Ericsson’s cellular SEPs. The Delhi HC also disagreed with Lava’s insistence on a top-down approach to valuation.

Testimony on foreign law by general fact witness: What’s a bit unusual is that a fact witness was asked about whether there was global case law to a certain effect, and denied. First, it’s hard to testify on the absence of something because there may have been some sealed decision somewhere in the world that the witness might not be aware of. Second, it might make sense to leave the discussion of global case law to the parties’ submissions and to ask only such witnesses about it who can give an opinion on foreign laws.

“Split rates”: The ruling notes that “Ericsson offered split rates to Lava, i.e., separate rates for India and the rest of the world.” A Chinese FRAND determination in a case brought by OPPO against Nokia also distinguished between target markets. But there are SEP holders and other stakeholders who have strong reservations concerning such distinctions. It’s a controversial topic.

Too late for injunction: It was apparently too late to order an injunction over any of the patents-in-suit. It took too long. That was also the case by the time Sisvel v. Haier was decided by the Federal Court of Justice of Germany, but in that case there was a need to determine whether injunctive relief should have been available at an earlier stage. One can nevertheless compare the two cases in the hypothetical: what if the question of injunctive relief had still played a role in Ericsson v. Lava and if the Huawei v. ZTE framework by the European Court of Justice had governed the case? It would have been an easy one to resolve in Ericsson’s favor as Ericsson didn’t ask Lava for more than its comparable competitors were already paying. Lava would then have failed at the next step of the Huawei v. ZTE analysis because it failed to make a counteroffer. Admittedly, there was some other conduct that suggested Lava was just an unwilling licensee, such as the disproportionate delay prior to signing an NDA. The overall pattern might have warranted a holding that Lava failed to make its first contribution under Huawei v. ZTE: concluding a reasonably acceptable NDA without undue delay could be reasonably considered a requirement. The problem with the German Sisvel v. Haier “retro” approach is just that various aspects of an implementer’s conduct that relate to conduct further down the road are then imported into the analysis of the initial declaration of a willingness to take a license.

Costs resulting from bad-faith conduct: Assuming that the judgment is affirmed in the event of a potential appeal, the outcome will have been costlier for Lava than just taking a license on the same terms as comparable companies. Annual interest of 5% accumulates over the years if they appeal instead of paying. Also, the court decided to “release the amount deposited by Lava before this Court in Fixed Deposit along with the accrued interest, after deduction of [Tax Deducted at Source], in favour of Ericsson.” That amount is smaller than the damages award against which it will be accounted, so it depends on Lava’s financial situation over the years whether this is net-negative for them or neutral (they presumably would have preferred to use the money instead of making a fixed deposit). And Lava’s hold-out conduct was key to the court’s conclusion that fee-shifting was warranted:

833.
As noted earlier, Lava did not negotiate with Ericsson in good faith, which resulted in the present litigation. Prior to institution of the suits as well as during the pendency of the suits, Ericsson made several offers to Lava to settle the case, which was not accepted by Lava and neither any counter-offer was made by Lava.
834.
Taking into account the aforesaid position, I am of the view that in the present facts and circumstances, Ericsson is entitled to recover actual costs from Lava.

Is that enough of a deterrent? There are two angles from which to answer that question (again, just assuming for now that the ruling will not be overturned on appeal):

• With the benefit of 20/20 hindsight, Lava will have regrets as far as only the deal with Ericsson is concerned. This ended up costing additional money and must have been a distraction.
• But at the outset Lava did this to get Ericsson to back down and lower its rates (there was no formal counteroffer, but it’s always possible that such parties have conversations in which one party, without making an official offer, indicates what it would accept). Ericsson remained principled.
• Maybe other companies granted Lava cheaper licenses in the meantime as they saw that it’s a difficult target: not a huge amount of money to be made from them, and prepared to see a lengthy case through in court. In that regard, Lava may have saved more money on other deals than it lost with respect to Ericsson.

That takes us to the final sentence of the judgment:

“Given the significant market for standardized products, it is hoped that in times to come India will become a leading neutral venue for global SEP resolution.”

With the greatest respect, that decision does not pave the way for that. Apart from political dynamics that may further weaken that jurisdiction, the only thing that would have helped to get Lava to act more constructively would have been the threat of an injunction. But those Indian court cases take forever, for a variety of reasons, and the Ericsson v. Lava judgment did not reach (though a dictum would always be possible) the question of whether an injunction should have issued.

In a scenario where the defendant is not a local player like Lava but a company selling into all of the world’s major markets, there would have been a list of jurisdictions in which Ericsson could have gained leverage long before this ruling by the Delhi HC. So far, the High Court of Justice in London has not had a case where an injunction appeared warranted against an implementer willing to commit to take a license at a FRAND rate to be determined by the court. Lava’s fact pattern, however, might even have persuaded a London judge that after this kind of conduct, they didn’t deserve a FRAND trial. In Germany, it would have been an extremely clear case. In the United States, it might still have been difficult to obtain an injunction in district court, but a U.S. import ban from the ITC would have been a possibility. There are emerging markets such as Brazil and Colombia. And it would really have been interesting to see this case adjudicated in China.

So if the question is whether India should weaken local SEP enforcement (as some implementers argue) or whether it actually needs to strengthen it, Ericsson v. Lava is yet another case that fails to prove any allegations of SEP overleveraging (and, as a result, overcompensation) in India.