Sunday, March 19, 2023

OPPO, others stand to benefit from Lenovo's resounding UK FRAND victory over InterDigital: experts' selectivity, excessive discounts, conduct-based arguments fail to persuade judge

This is not a "rapid response" piece by any measure. It's been three days since Justice Edward James Mellor rendered his InterDigital v. Lenovo FRAND judgment. Many others have reported and commented since, and in a first step I pointed--via LinkedIn--to a great summary by Michael Burdon of Simmons & Simmons. Some others had to--or have yet to--correct the conclusions to which they originally jumped.

It's a landmark ruling. Probably the worst that ever happened to InterDigital in any litigation, and arguably one of the most significant setbacks for standard-essential patent (SEP) licensors in general (though nowhere near as bad for them as Microsoft v. Motorola was back in the day). In the short term, a major implementer with a much better licensing track record than Lenovo stands to benefit hugely: OPPO. This may also help Apple against the Optis/Unwired group, where a UK ruling is likely imminent and some of the same players are involved, though the fact patterns may be very different. In the mid term, it very much depends on what the appeals court will decide, but it will be hard for InterDigital to get Justice Mellor's determinations on the credibility of experts--where InterDigital's experts performed poorly--overturned. The UK is now a less attractive venue for SEP enforcement. This InterDigital v. Lenovo decision actually invites a lot more SEP enforcement litigation, but in other jurisdictions, as I'll discuss further below.

It's not all bad for InterDigital, a company that has recently been very good at renewing major agreements (or at least agreeing on a path toward renewal) and has a lot of potential, but the bad far outweighs the good here:

  • The $138.7M lump-sum payment for an absurdly long period (2007-2023) comes down to a per-unit royalty of $0.175, which is only marginally more than Lenovo's best offer of $0.16, and just a little over a third of the $0.498 per-unit royalty InterDigital was seeking.

    Without a doubt, this is the key measure, and there is no explaining away InterDigital's (and its experts') defeat. InterDigital itself conceded defeat by announcing its appeal.

  • The court totally rejected InterDigital's headline rates ("program rates") as comparables because discounts of up to 85% (Samsung) made those numbers pretty meaningless.

  • The court found neither party in compliance with its FRAND licensing obligations: InterDigital was categorized as an unwilling licensor (for consistently making supra-FRAND demands) and Lenovo an unwilling licensee (which is unsurprising). The ruling notes that both parties can be unwilling at the same time. The problem for InterDigital is that this is unprofitable: it can stick that finding of Lenovo's unwillingness to its office walls, but it's not getting any compensation for it. InterDigital had some pretty good conduct-based arguments in this case. But if you have a strong case based on the implementer's behavior, you have to enforce in jurisdictions like Germany (Sisvel v. Haier) where conduct is given a lot of weight. Justice Mellor simply declined to let InterDigital's criticism of Lenovo's behavior result in an increase of what would be seen as the objective market value of the patent license in question.

  • The part that SEP holders will, of course, appreciate is neither new nor spectacular: a willing licensee can't just hide behind statutes of limitation (for the recovery of damages for past infringement) but will make a release payment to fairly compensate for all past use (which in this case may also entail a post-judgment award of interest). I'm not aware of any case law to the contrary in any of the jurisdictions that focus on the willingness of licensees. But it is useful clarification that will be cited elsewhere.

  • Where InterDigital's expert witness on comparables clearly lost all credibility with the court was when he pointed to 20 license agreements outside the top 20 most important ones. The license agreements Lenovo focused on accounted for roughly 98% of all units, and the ones InterDigital's expert focused on amount to roughly 2% of the business. Such selectivity is ridiculous.

    In a jurisdiction where a court mostly delegates these factual findings to an expert witness, the decision might have been different. An economic expert might have independently developed a middle-ground scenario. Justice Mellor notes that InterDigital--which didn't want to undermine the extreme positions its experts were taking--failed to present a more conservative set of numbers. A court-appointed expert might have come up with some new analysis. But when there's a battle between experts and a judge has to make the decision, it's risky to take extreme positions that are easily rejected because they're irrationally selective.

What I consider to be the most important passage is the final sentence of para. 516:

"However, it is important to keep in mind that these volume discounts were ‘applied’ to InterDigital’s ‘program rates’ which were paid only by the smallest and least sophisticated licensees."

Justice Mellor goes on to clarify that he's not faulting InterDigital for being pragmatic and business-oriented in the sense of offering huge incentives to the Samsungs of the world in order to strike deals with them rather than having to engage in protracted multi-jurisdictional litigation, during the course of which InterDigital's cash flow would be adversely impacted and shareholders would get nervous. This is not a moral question. It is, however, relevant to the "ND" part of FRAND.

I would describe Justice Mellor's position as follows:

  • You can't build comparables by squeezing the weak

  • if at the same time you cave to the big hold-outs.

That is not fair, and it ends up distorting competition. It makes the rich richer at the expense of the poor.

InterDigital has only one revenue source: licensing. Apple and Samsung each account for roughly one quarter of InterDigital's income. It's easy to see why InterDigital would rather litigate against someone like Lenovo, and avoid litigation against Apple and Samsung through huge concessions. They're within their rights to make that choice, but is that still going to be the right strategy going forward? If more courts adopt Justice Mellor's position that InterDigital's "program rates" have nothing to do with market realities and fair valuation, then it just won't work as even the little guys will end up paying relatively low per-unit royalties.

Should a long-term hold-out like Lenovo benefit from rates that were agreed upon with willing licensees (companies that renewed their agreements in time, or shortly after expiration of the previous one)? That is a very valid question. But an injunction is a prospective remedy, and that's the context in which those UK FRAND determinations are made, so it becomes just a numbers game. There are no signs of the UK adopting the behavior-centric Sisvel v. Haier approach. If you want that one, you have to go to Germany. The current problem there is that the Federal Patent Court puts out those premature opinions after six months, but the solution is simply to assert enough patents that something will stick.

Courts rarely intend to invite more litigation, and Justice Mellor is no exception. But if one thinks it through, what should a SEP holder like InterDigital do when facing such situations as with Lenovo?

The UK now looks rather unattractive. It took about a year since trial, and now InterDigital needs to appeal, which will take a lot of time and is not too likely to move the dial. I'm not blaming Justice Mellor, who clearly says in his decision that he also had other matters to adjudicate during that period. But is it really a good idea to hold a multi-week trial with expert witnesses and then have a judge write a book (225 pages) about the case? In the meantime, it could have obtained injunctions in such jurisdictions as Germany, Brazil, or Colombia--and pretty soon, the Unified Patent Court, of course. The amount of collateral required to enforce German injunctions can be high, especially against large-scale implementers, but after two rounds of litigation (regional court and higher regional court), that problem goes away.

The idea of a "one-stop shop" solution--a jurisdiction that sets a global FRAND rate and will enjoin the infringer unless that one is accepted--used to be appealing. After InterDigital v. Lenovo, it's clear that this can backfire. The other jurisdiction that sets global FRAND rates even without the consent of both parties is China, and the judges there will definitely take note--as they absolutely should--of the UK findings. Short of a successful appeal, InterDigital can hardly expect to get more than 17 cents per unit from any Chinese device maker seeking a FRAND determination in its jurisdiction.

InterDigital is going to lose against OPPO. It won't get leverage in Germany because OPPO left the market; it is facing an uphill battle in the UK; and in the meantime, the decision may just be made in China.

OPPO may also benefit from this in its UK litigation with Nokia, but just like in Optis v. Apple, the facts may be very different, so it's too early to tell. I believe Nokia is a more courageous SEP holder than InterDigital, so I doubt that they grant discounts on the order of 85% to licensees like Samsung and Apple. They're probably a lot more consistent. But OPPO is also a much more sophisticated and licensing-oriented adversary than Lenovo.

The UK judiciary was eager to attract SEP injunction cases. Now that they get more of them, they start to realize that those complex cases place an enormous burden on the judges (and, by the way, with very little benefit to UK taxpayers). By taking a long time to decide those cases, and then setting relatively low per-unit royalties, they are now actively discouraging such filings. I'm not saying that it's their strategy, but it's the net effect of what's happening.