Wednesday, June 28, 2017
Judge Koh shows the way: FRAND non-compliance can be established without rate-setting exercise
Qualcomm tried hard, but unsuccessfully, to get the FTC’s antitrust lawsuit in the Northern District of California dismissed. Maybe Qualcomm hoped, more realistically, the FTC would have to amend the complaint in some important ways, possibly complicating the case to the point where the U.S. competition agency would find it hard(er) to justify using the resources required for pressing on. The reason I suspected the latter is because, based on hearsay from about seven years ago, the European Commission’s investigation of Qualcomm’s practices with a focus on Nokia (now more of a Qualcomm friend than foe), essentially got derailed by scare of conducting a resource-intensive, complex and somewhat subjective (thus more likely to be overruled) rate-setting exercise. In the FTC case here, the presently-Acting Chair of the FTC, Maureen Ohlhausen, opposed the decision authorizing the complaint, and might have been the first decision-maker to argue that the case should be dropped or settled (the latter without any useful remedies) due to litigation economics. Industry concern over such a decision by the FTC was and remains real, as an open letter to President Trump showed in April.
Fortunately (though I was sympathetic to some of the arguments in Qualcomm’s motion), the case is going forward on the basis of the FTC’s original complaint (congratulations to the FTC’s litigation team, whose partly-minimalistic approach has worked out so far) and without the hypothetical need (in a worst-case scenario) for a jury trial. And the federal judge whom Qualcomm needs to persuade at the future bench trial has taken positions on the legal issues in the case that don’t bode well for the San Diego patent holder and chipset maker. There is an important caveat here since the hurdle for denying a motion to dismiss is low, but the way Judge Koh has expressed her disagreement with Qualcomm’s various legal challenges does go beyond what is strictly needed to deny the motion. Seriously, I’ve never been happier about a decision from the Northern District of California than this one (this post continues below the document):
17-06-26 Order Denying Qualcomm Mtd by Florian Mueller on Scribd
Here’s a total non-starter to begin with:
“Qualcomm contents that FTC’s allegations of above-FRAND royalties are nonetheless contradcited by the fact that the Complaint also alleges that Qualcomm has historically collected a royalty rate of 5%, and that this rate has not changed over time. […] However, FTC’s allegation that Qualcomm has continued to collect the same 5% royalty on the total value of a handset supports, rather than contradicts, the FTC’s allegations that Qualcomm’s royalty is above FRAND. As the Complaint explains, early handsets were primarily used only to transmite voice calls. […] Accordingly, Qualcomm’s cellular communications SEPs contributed significantly to the functionality and value of a 2006 handset. […] By contrast, handsets today contain numerous features that are unrelated to cellular network connectivity, such as camereas, Wi-Fi access, and data storage. […] Thus, Qualcomm’s SEPs contribute far less to the value of a 2017 phone [than] they contributed to the value of a 2006 phone. […] Nonetheless, Qualcomm continues to collect a 5% royalty from the total value of the handset today for Qualcomm’s cellular communications SEPs, just as Qualcomm did a decade ago.”
Taken together with SEP portfolio erosion (with respect to existing standards), this point is then reinforced:
“In short, that Qualcomm collects the same 5% royalty on the total value of a 2017 smartphone as Qualcomm collected on the total value of a 2006 phone, despite the fact that both handset technology [that’s the previous point] and Qualcomm’s SEP portfolio [have] changed dramatically over the past decade, supports FTC’s allegations that Qualcomm’s SEP royalty rates are above FRAND.”
The passages quoted above can be reasonably interpreted as facts-based skepticism regarding Qualcomm’s claims of being FRAND-compliant, which shows what Qualcomm is up to here. There’s a lot more to it than merely finding that the FTC’s pleadings are sufficient. The FTC has won the single most important battle in the FRAND context that Qualcomm could possibly have lost: the royalty-base question.
Since Judge Koh believes that the FTC’s FRAND non-compliance theories are potentially sufficient to determine that Qualcomm charged supra-FRAND royalties, rate setting won’t be necessary. The case will stay focused, and some of these issues (such as the royalty base) may at the next stage be analyzed in light of whether a reasonable fact finder could ever disagree with the FTC…
The royalty base question hasn’t been resolved before. I was profoundly disappointed when Judge Robart didn’t draw a bright line in this regard (in Microsoft v. Motorola). It’s great that Judge Koh has this issue totally figured out in economic and technical terms.
The FTC-internal driving forces behind this antitrust action must be very happy and feel encouraged. Apple is litigating directly against Qualcomm in San Diego; Samsung and Intel filed informative and persuasive amicus briefs; regulators in other jurisdictions agree; and Judge Koh is now a thought leader, too. All that is missing is for Acting Chairwoman Ohlhausen (and others who may have shared her views so far) to join the mainstream. What the FTC is pursuing here is a just cause and, while most citizens won’t ever realize, is perfectly consistent with the Make America Great Again vision. Qualcomm has done and continues to do impressive research and deserves to be compensated fairly and reasonably, not overcompensated at the expense of companies that make real products and, by extension, consumers.
Another legally very important issue on which Judge Koh has taken a fairly clear position (clearer than would have been necessary at this stage of proceeding) relates to the antitrust duty to deal with competitors, i.e., Qualcomm’s obligation to honor its FRAND licensing promise vis-à-vis Intel, Samsung’s components division, and others.
In connection with (among other things) Qualcomm’s “no-license-no-chips” policy and tying, there was some argument over what the proper term for royalties changed on top of the chip price should be. The FTC consistently referred to it as a “tax” (a term that is also justified by Qualcomm’s anachronistic approach to the royalty base), and Qualcomm understandably didn’t like that. Judge Koh now calls it a “surcharge.”
The following holding by Judge Koh is another highly important point to draw attention to (and one of the things that make me wonder whether Qualcomm can defend itself at all, unless regulators in different parts of the world all got the facts totally wrong, or whether it may ultimately just have to focus on remedies since it can’t win the merits part):
“Thus, by violating its FRAND obligations twice over—by not licensing its competitors and by threatening to withhold its chips to induce OEMs to pay an above-FRAND royalty rate—Qualcomm raises the ‘all-in’ modem chip price that OEMs pay on all modem chips.”
Finally, those of use who follow Apple v. Samsung (with all the talk there about a two-horse race etc.) have or will read with great interest how Judge Koh, who is also presiding over those Apple v. Samsung cases, views the anticompetitive impact of Qualcomm’s exclusive rebate-based arrangements with Apple. In that context, she agrees with the FTC as well, though here she limits her analysis to minimum pleading standards. This does not necessarily mean that she’s less convinced. It may very well have to do with the nature of the beast: the anticompetitive effects of past exclusive deals (that foreclosed others from supplying chipsets to Apple for many years) involve certain facts regarding market share (and segmentation, possibly) and also some even harder-to-measure aspects such as Apple being a company whose decisions others like to follow