Wednesday, January 16, 2019

The FTC has rested its likely-winning case with a final hand grenade destroying a Qualcomm mantra

Yesterday (Tuesday, January 15) was Day 6 of the FTC v. Qualcomm antitrust trial in San Jose (Northern District of California), which will continue in Judge Lucy H. Koh's court on Friday, January 18.

When FTC lead counsel Jennifer Milici said "Your Honor, the FTC rests its case," you could have heard a pin drop even if there had been thousands of people in the audience, provided that all of them would have been reasonably knowledgeable. That's because seconds before that procedural notice, former Qualcomm licensing president Marvin Blecker had been confronted, in a videotaped deposition, with an internal email in which a colleague confirmed to him that Qualcomm's chip division had actually held product shipments to a customer who had not yet accepted Qualcomm's license terms.

It had been Qualcomm's mantra all the time that they had never actually carried out the "No License-No Chips" threat against existing customers. They couldn't deny that they wouldn't accept a new customer prior to taking a license from them. And an Apple witness said that after Apple sued Qualcomm (in January 2017), Qualcomm refused to even discuss a potential 5G product partnership with Apple. But again, Qualcomm's lawyers had over and over again--you could set your watch by them--elicited testimony from current and former Qualcommm executives that the company had never carried out the threat of holding chipset shipments to an existing customer due to disagreements on licensing.

In the seconds before resting its case, the FTC made them all look like...well, I don't want to use the harsh words that I actually think would be warranted here, so let me just call them "unreliable witnesses and lawyers you better don't trust in this context."

CDRinfo, a hardware-focused website, has also reported on that testimony.

The actual legal significance of whether or not the threat ever was carried out shouldn't be overrated. It's not like robbing someone at gunpoint is allowed; illegality starts long before pulling the trigger. Also, "No License-No Chips" is just one of the FTC's four theories in this case, though it is, as the FTC's economic expert Professor Carl Shapiro (UC Berkeley) explained yesterday, one of three types of conduct that are closely interrelated.

I've previously used chess terminology in the context of this trial. With respect to No License-No Chips, the FTC has taken out what Qualcomm portrayed as its queen, even if in my opinion it was a smokescreen, at best a pawn. With respect to Qualcomm's credibility in this case, it was at least a rook.

Qualcomm's case-in-chief began yesterday with testimony that's largely irrelevant to the outcome-determinative issues. Judge Koh wasn't nearly as busy as she normally is during testimony. More about that later.

During the FTC's case-in-chief, Qualcomm's cross-examinations were largely ineffective, with only two exceptions. The first one was MediaTek's testimony, which was well-meant but which Qualcomm was able to use in part to make a small dent in the FTC's argument with respect to the licensing of rival chipset makers. The second and far more significant one reminded me of Davis Cup tennis matches where the strongest player of one team played against the weakest one of the other. Here, the unbalanced matchup pitted Cravath lawyer Gary Bornstein (for Qualcomm) against one of three FTC expert witnesses, Michael Lasinski, whose methodologies are indeed questionable in at least some respects. Maybe his report should be called the "Lazynski report" because he didn't make the necessary effort to underpin his top-down analysis more solidly, even if it might have been painstaking. The way Mr. Bornstein handled the cross-examination was a demonstration of clever legal work, measured but forceful courtroom demeanor, and raw analytical brilliance. After that experience, I wonder what I may hear from Mr. Lasinski later this year when attending Huawei v. Samsung (which I probably will), where he's actually supporting Qualcomm-like behavior on Huawei's part and, apparently, taking unreasonable positions on royalty levels.

This leads us to the FTC's third (and by far the best) expert witness, Professor Shapiro, who testified yesterday. He was relaxed, poised, and you could see his experience in teaching because he explained everything in the simplest, most understandable terms possible. The only audience that mattered here was one person: Judge Koh. (And later on, the appellate judges, who will see the transcript.) If her FTC v. Qualcomm judgment becomes a landmark ruling (like her summary judgment decision on rival chipset licensing was one of the most important pretrial decisions in the history of this industry), she'll probably be widely regarded as the world's number one technology industry judge, a position for which one obviously needs to be in Silicon Valley, but just being there doesn't suffice.

There's so much to say about Professor Shapiro's testimony that I'll do a follow-up post at some point, either based on my notes or on the transcript whenever I obtain it. In this post here, I'll just summarize what's most important with a view to how strong the FTC's case is:

  • Obviously, Professor Shapiro supported the FTC's theories.

  • He's an expert in antitrust economics, also at the intersection of antitrust and IP. According to his co-author, Stanford Professor Mark Lemley, Qualcomm has already spent millions of dollars attacking them over their papers on standard-essential patents (SEP) holdup:

  • On the basis of Qualcomm having imposed supra-FRAND royalties (he didn't say supra-FRAND yesterday, but meant it), the economic effect of the allegedly anticompetitive conduct had three components:

    1. reduction of rival chipset makers' volume/market share;

    2. reduction of device makers' margins (that's because of them absorbing part of the increased cost); and

    3. the most critical part in U.S. antitrust law: higher prices paid by consumers than otherwise (that's the part device makers passed on).

    Professor Shapiro didn't call it a vicious circle, but he did point out that the effects of these dynamics get worse because of a reduced competitiveness by other chipset makers.

  • He testified to Qualcomm having had a premium LTE baseband chip monopoly until including 2016. He acknowledged a slowly-decreasing market share, but he did not take a position on Qualcomm's position in LTE in 2017 or 2018, much less on Qualcomm's future 5G market share in 2019. To be clear, he did not say Qualcomm didn't have or wasn't going to have market power in those other years; those years were simply outside the scope of his task.

  • Professor Shapiro pointed out that the fact Qualcomm was and is innovative and made and makes great products is unrelated to the behavioral issues in this case. When Qualcomm argued that other companies just didn't have equally good chips, he said that healthy competition in a market also means that "companies who are not quite as good" have a chance to challenge the incumbent.

  • He described No License-No Chips as "a very heavy hammer that Qualcomm is bringing down on OEMs at least as a threat."

  • He basically pre-empted the testimony we were then going to hear from a true technology giant, Qualcomm's co-founder and first (and long-time) CEO, Dr. Irwin Jacobs. Qualcomm called Dr. Jacobs, and he told great stories, but mostly they had nothing to do with the issues before Judge Koh. Large parts of his testimony would have been perfect just one block away from the courthouse: for a presentation at the Tech Museum. To be clear, I really do have the greatest respect for Dr. Jacobs and I love people who successfully swim against the tide, as he and his company did with CDMA. I also appreciate the Twitter like I got from him, provided that @jacobs_irwin is not a fake account (never know). But the case is not about how Qualcomm started; it's about what it did after it succeeded. Professor Shapiro said the following:

  • Qualcomm's lead counsel Bob van Nest tried something that was just a non-starter: when cross-examining Professor Shapiro, he argued that Intel simply had enough money to compete with Qualcomm in the baseband chipset business. This was a typical point to make in front of a jury, just like he (or someone else from his firm) asked Oracle President Safra Catz (in Oracle v. Google) about the tens of millions she makes per year. With such "deep pocket" stories you can appeal to jurors' envy and/or make jurors subscribe to the fallacy that money buys everything, but you won't impress a pro like Judge Koh. Professor Shapiro calmly and with a smile pointed out that as an economist, he does not look at it that way: what matters is not whether someone has money (let me explain: in economic theory, the assumption is that money will ultimately find its way to where more money can be made, through whatever funding mechanism, be it lending or venture capital or public markets), but whether (not exactly his words) a business would be viable on its own.

  • Professor Shapiro had to suffer some more economic nonsense gladly--because of a Qualcomm expert report we'll hear about more during Qualcomm's case-in-chief. He was asked about Qualcomm's expert's claim that rival chipset makers were "freeriders" because Qualcomm doesn't proactively sue them. The FTC wanted to know how he'd respond to that, and with a smile he said he was "responding badly" because it's economically wrong. Two weeks ago I addressed the same idiocy (but without reference to any expert report, just to another pretrial filing) on this blog.

  • In one (and only one) significant respect, Mr. van Nest's cross-examination effort was successful: Professor Shapiro ultimately had no choice but to acknowledge that his economic theory of harm (ultimately also consumer harm) is based, as an essential "building block" as he conceded, on the starting point of a supra-FRAND patent royalty being charged by Qualcomm. Mr. van Nest wanted to get him to confirm that any conclusions would be "irrelevant" without that part, and I agree with Mr. van Nest in that regard (for a change).

One can sum it up like this: there was nothing Qualcomm was able to do to reduce Professor Shapiro's credibility. Nothing that Qualcomm could really do to seriously question his methodologies (unlike in the "Lazynski" case). The only thing they can try is to leverage the shortcomings of the Lasinski report and testimony against the Shapiro report--but that's not nearly as easy to achieve as it sounds:

  1. Judge Koh would have thrown out the Lasinski testimony in its entirety if it had seemed wholly unreliable to her. In other words, whatever he got to present and testify here is stuff that she may give more or less weight to, and Qualcomm made strong points against relying on that report in every respect. But realistically, some parts will still get at least a bit of weight. Qualcomm's problem is that its royalties are so out of line that even if one discounted the numbers in a report like Mr. Lasinski's, one would still find them to be supra-FRAND.

  2. Professor Shapiro made two clever moves to maximize the independence of his report from the Lasinski report:

    • He explained that using general economic bargaining theory, his report arrives at the conclusion of a supra-FRAND royalties on its own.

    • He furthermore pointed again and again to the trial testimony of device makers and rival chipset players. I believe that testimony is so incredibly strong that I'm not sure the court even needs any expert report at all to conclude that supra-FRAND royalties were imposed. It's not just that virtually the whole industry says so, but also that witnesses pointed to the disproportionate percentage of overall cellular SEP royalties paid in this industry that goes to Qualcomm (80%-90% in Huawei's case, for instance). Even Ericsson, which has interests partly aligned with Qualcomm's, undermined Qualcomm's FRAND-compliance claims.

  3. There's also the Donaldson report on "atypical" licensing terms. Qualcomm wasn't nearly as successful attacking this licensing expert with 31 years of experience at Texas Instruments as they were when attacking Mr. Lasinski.

The Lasinski report is, and will be (presumably also on appeal, where Qualcomm can reargue any Daubert points), Qualcomm's primary attack vector, but it's not like a chain that's as strong as weakest link: it's more like there are various other ways to build a chain, and Professor Shapiro has shown the way already.

By insisting on a FRAND determination, Qualcomm sought to reargue its 2017 motion to dismiss, which Judge Koh denied. The related order already took the perspective that many roads lead to Rome, to put it that way. The FTC doesn't need to have three perfect expert reports. "Two out of three ain't bad."

The FTC's litigation staff can be proud of the tremendous work they've done (and imagine the distractions resulting from the government shutdown). The case the government has rested is extremely strong, and I already have this feeling that this "Your Honor, the FTC rests its case" moment may have been one of the most important moments in worldwide antitrust history.

Nobody is perfect, and no case is perfect. Qualcomm's case-in-chief has begun, and we'll now hear testimony from partly the same witnesses (but parts of their videotaped depositions that Qualcomm views as helpful) and Qualcomm's preferred witnesses. It's just that almost all the major players have spoken already, including Qualcomm's CEO and Qualcomm's president. And yesterday we got history lessons and general talk about engineering culture (from Qualcomm's senior VP of engineering, Dr. Durga Malladi) that suggested Qualcomm doesn't even know how to kill all the trial time it has, while the FTC (which had even more testimony on the list, but then ran out of time and stopped at the best possible moment) could presumably have gone on an on for weeks with evidence buttressing its theories.

In order to turn this around, or at least to have a reasonable record for an appeal, Qualcomm will now need several hand grenades of the kind the FTC detonated seconds before resting its case. I will stay tuned. I know many of you will, too. (By the way, I wish to thank the readers who said hi at the courthouse.)

Share with other professionals via LinkedIn:

Tuesday, January 15, 2019

Qualcomm and Apple bridged the wide gap between their FRAND perspectives through complicated arrangements

Yesterday (Monday) was Day 5 of the FTC v. Qualcomm antitrust trial in San Jose (Northern District of California). In a first summary shortly after the court session, I provided an overview of how, after most of the day had gone extremely well for the FTC, Qualcomm got what athletes call a "second wind" and made tremendous headway against one of the FTC's three expert witnesses, Michael Lasinski. This is now a follow-up post with a focus on negotiation dynamics and deal structures, which was the #1 topic on Friday.

Apple's Chief Operating Officer Jeff Williams was first to testify on Monday morning. He was a member of the team that created the original iPhone, and has interacted with Qualcomm ever since. What we learned from him contributed to a better understanding of Qualcomm's dealings with other companies.

In an antitrust case pitting the U.S. government against a patent-wielding U.S. chipmaker, Apple obviously plays an important role--but many observers of this proceeding appear to miss the grand picture by focusing too narrowly on that company. Only one of the FTC's four key issues is specifically about Apple: exclusivity arrangements. Even that one is ultimately about a pattern, just that the related agreements between Apple and Qualcomm are the only manifestation of that pattern at issue in this case.

At the moment, Apple appears to be Intel's only customer. Qualcomm elicited testimony from Intel's Aicha Evans confirming that at certain points she wasn't even interested in a couple of potential customers. There may have been any number of reasons for that as Intel needs to ramp up its baseband chip business step by step, but there's no reason to assume that Intel wouldn't be interested in growing its business in this field and in having a broader customer base.

Qualcomm's practices have been a concern for the industry at large. I heard this from companies like BlackBerry and (when it was at its peak as a handset maker) Nokia, and some I can't mention due to NDAs (other than saying they're not Apple), before I first became aware of Apple's positions--which I did only in connection with the KFTC (Korea Fair Trade Commission) antitrust proceeding.

The testimony from industry players--of all sizes and from multiple countries--that the FTC has presented is extremely strong. In fact, it's so strong that the FTC's expert witnesses (we'll hear the only trained economist among them, Professor Carl Shapiro, in a few hours) are going to be much less influential in this case than they would be in others. Those industry witnesses have established so many facts relating to the question of anticompetitive harm that I don't think Qualcomm can this around by winning the "battle of the experts," which it may or may not.

As I've stated before, it's advisable not to overrate the importance of who--Apple or Qualcomm--was first to propose exclusive arrangements. If a company is made an offer that serves its anticompetitive objectives in illegal ways, it still has the obligation under the antitrust laws to refuse it. That said, we're all curious about how that type of deal came into being.

Mr. Williams nuanced Apple's proposal by differentiating beween short-term and long-term exclusivity. At the time the parties were negotiating the original (2011) Transition Agreement (which was about Apple discontinuing its relationship with Infineon and exclusively sourcing baseband chips from Qualcomm), Apple apparently had no problem with short-term exclusivity based on the market landscape. But Apple did not want to enter into an exclusive deal for the long haul, given that things can change in this industry within a relatively short period.

Qualcomm presented evidence (including Apple-internal emails, at least one of which was written by Steve Jobs) that a one-year term was to short for Apple's purposes. So the original Transition Agreement became a two-year deal (2011 and 2012). Obviously, two years still fall far short of what one would consider "long term."

A first amendment to the Transition Agreement was negotiated, and it covered the period until including 2016. Accoring to Mr. Williams, Apple did not take the initiative to propose exclusivity in that situation. And Mr. Williams mentioned a situation in which Apple had made a counterproposal to Qualcomm in which the exclusivity commitment was redlined, triggering a response by Steve Mollenkopf (now Qualcomm's CEO) that those "general design commitments" (a euphemism for exclusivity requirements with a clawback clause) were really important to Qualcomm.

It's key to view the 2011 and 2013 agreements in the wider context of Apple and Qualcomm struggling to work out mutually acceptable deal terms. Apple was by far not the only company to take issue with Qualcomm's requirement that others had to grant back a royalty-free license to all patents of their own. We heard about that from various witnesses, most recently Ericsson.

But after the enormous risks Apple had taken with the original iPhone (trying to enter an industry that was an oligopoly, with BlackBerry having been the only successful new entrant--prior to Apple--in a long time), and knowing Apple's emphasis on differentiation, it does not surprise me that Apple was particularly opposed to Qualcomm's grant-back condition.

The face-saving solution that allowed both companies to remain consistent with their principles was that Apple obtained only an indirect license--through contract manufacturers like Foxconn. The CMs were obviously unable to license Apple's patents to Qualcomm, but they were able to secure a license from Qualcomm covering the devices they manufacture(d) on Apple's behalf.

Fundamental, philosophical disagreements between Apple and Qualcomm related not only to the grant-back requirement but also to the royalty base and, directly related to it, the amount. We learned yesterday that Apple would have accepted Qualcomm's 5% royalty rate if applied to the cost of a baseband chip. Just like with the extortionate grant-back clause, it also makes sense that Apple, from the beginning, took issue with Qualcomm's royalty base (complete devices). Prior to the iPhone there had been some relatively unsuccessful personal digital assistants and early "smartphones" made by various companies such as Nokia. Also, BlackBerry added email to voice and short-message texting. But only the iPhone represented a complete paradigm shift from "mobile phones" that gradually got more and more computing functions to a "mobile computer." That's where Apple as a computer manufacturer and maker of handheld devices (at the time, the iPod was already reasonably popular) brought the two worlds together in a way that established the primacy of computing functionality over telephony. The computer absorbed the phone, not the other way round. So Apple was particularly concerned about Qualcomm seeking to be compensated for functionality its patented inventions weren't related to.

According to Mr. Williams, Apple didn't care about whether Qualcomm formally lowered its royalty or just agreed to incentive payments that had the same bottom-line effect. The royalty Apple could live with at the time (while still believing this number was not FRAND) was $7.50 per device, and that was simply because it was the average of what everyone else was paying Qualcomm. No matter what Qualcomm then claimed to be the royalty base, the net effect from Apple's point of view was that Qualcomm got compensated for the equivalent of a feature phone, not for what Apple's mobile computer offered on top.

But, as we know, Apple had to make concessions to Qualcomm. Exclusivity (under the 2011 Transition Agreement and its 2013-2016 amendment) was known before. Yesterday Mr. Williams said that under the very first agreement between Apple and Qualcomm, Apple was required to publicly renounce the WiMax standard.

The FTC is not going after Qualcomm for deals under which it paid device makers like Apple to say negative things about, and refrain from implementing, standards that might have been alternatives to standards that Qualcomm had greater benefits from (because of greater leverage over those standards, such as particularly CDMA). But I hope that at some point, in whatever jurisdiction but ideally in the United States, a regulator or a court of law or, if all else fails, a legislature will take action against such competition-distorting deals, too.

Share with other professionals via LinkedIn:

After mixed Day 5 of FTC v. Qualcomm antitrust trial, irrelevant German ruling to be handed down later

[Update] The Mannheim Regional Court indeed threw out Qualcomm's complaint for non-infringement based on claim construction (the first of three possibilities I discussed further below). [/Update]

Day 5 of the FTC v. Qualcomm antitrust trial just concluded. It was the most eventful and interesting day of the first half of this bench trial. The morning and the early afternoon were a total disaster for Qualcomm, but toward the end Qualcomm had its strongest hour to date.

I usually write about these long and tiring trial days the next morning. I'll do so again, given that we learned some interesting numbers and other relevant facts, but here's a very short summary--and then I'll provide a preview of the Qualcomm v. Apple patent infringement ruling coming up in Mannheim at 9 AM local time there (midnight Pacific Time):

  • The first witness this morning was Apple COO Jeff Williams. Some of what he said effectively nuanced what Qualcomm CEO Steve Mollenkopf said on Friday about Apple's alleged proposal to agree to exclusivity. Mr. Williams provided the court and trial watchers with interesting and credible information on various other issues. He was the first witness to be rudely interrupted by Keker van Nest & Peters' Eugene Paige, but in his low-key, friendly, unpretentious way drove all the important points home.

  • The FTC staff then read into the record some testimony by an LG Electronics IP executive that showed, among other things, how Qualcomm bullied LG when it challenged in arbitration (which wasn't even remotely as threatening to Qualcomm's business model as a challenge in a court of law would have been) Qualcomm's licensing terms.

  • Then came an FTC licensing expert, Richard Donaldson. Against his background of 31 years (!) of patent licensing work at Texas Instruments, and many years of consulting and expert testimony since he retired from TI, Mr. Donaldson explained how atypical the licensing terms Qualcomm imposes on licensees are, and he also explained that chipset-level licensing is as feasible as it is commercially viable.

    The aforementioned lawyer, Mr. Paige, cut off Mr. Donaldson's answers, and even though I think Mr. Donaldson could have given a tactically better answer to the question of whether TI sought to avoid the exhaustion of system-level (= device-level) patents through chipset licensing (he could have pointed out that this is a non-issue if a chip-level license deal involves only specified or otherwise clearly-defined chip-level patents), I think he dealt with Mr. Paige's onslaught very well. At some point Judge Lucy H. Koh intervened against those interuptions of the witness.

  • After the lunch break (which was in the middle of Mr. Donaldson's testimony) Judge Koh came back to the issue of Mr. Paige's constant interruptions and told him not to cut off a witness "after two words." She called this "improper" (I couldn't agree more--it was extremely annoying) and warned that the next party to do so would lose two minutes of its trial time.

  • The FTC played some testimony by Ericsson's licensing president Christina Petersson. While Ericsson is also very much into patent monetization, especially standard-essential patent (SEP) monetization, she said various things that seriously undermine some of Qualcomm's positions and defenses, such as with respect to Qualcomm's licensing terms being unusual (the only chipset-level outbound license deal Ericsson still has in place is with Qualcomm). Ericsson sought to position itself as a reasonable SEP licensor (I've repeatedly criticized Ericsson on this blog, but they're not the worst for sure) who understands every SEP holder has to consider he doesn't own the standard alone "in order for the system to work." In other words, Ericsson disagrees with Qualcomm's extreme demands because if everyone did that, the collective impact of royalty-stacking would be devastating. Ericsson even undermined Qualcomm's position with respect to chipset-level licensing, saying that despite the agreement it still has in place with Qualcomm, Ericsson continued to invest in research and development as well as standard-setting.

    With respect to LTE, Ericsson believes it has the strongest portfolio in the industry, with Qualcomm "definitively" weaker in Ericsson's opinion, and with some allegedly believing Nokia is also ahead of Qualcomm.

    We've now heard pretty much every significant industry player. No one is on Qualcomm's side in every respect. There are some who are 100% against Qualcomm, and some who at least disagree with Qualcomm on some key issues.

  • Then came Michael Lasinski's testimony. He's another licensing expert testifying on the FTC's behalf. His background is impressive: a former president of the U.S. & Canada chapter of the Licensing Executives Society, and a former American Bar Association IP division chair. A couple of my Twitter followers immediately vouched for him when I mentioned him.

    After an otherwise awful day for Qualcomm, we saw a world-class performance by Cravath Swaine & Moore lawyer Gary A. Bornstein. He struck just the right balance between being assertive and respectful, and he managed to highlight a number of issues relating to Mr. Lasinski's methodologies. Question by question, Mr. Bornstein cornered Mr. Lasinski and forced him to concede limitations and shortcomings, and in one context even an outright contradiction. Mr. Bornstein won this fight by a wide margin, though Mr. Lasinski made a stronger showing initially than the other expert, Mr. Donaldson, on cross-examination.

    While the positions Qualcomm takes on economic expert testimony are way too demanding (no government agency or company in the world could possibly satisfy them), there are indications that Mr. Lasinski's analysis could be the, relatively speaking, weakest link in the FTC's chain. It remains to be seen tomorrow to what extent Professor Carl Shapiro's economic analysis renders those deficiencies less relevant.

  • Apple contract manufacturer Wistron (which was spun off from Acer a long time ago) confirmed what others said about Qualcomm simply not making concessions on key licensing terms. He said they ended up agreeing on a huge upfront payment to Qualcomm, and in order to recoup that one as soon as possible, they couldn't work with other baseband chipset makers though there would been economically attractive options.

So there was a lot of shadow for Qualcomm, but also a silver lining for them with respect to Mr. Lasinski's methodologies.

After the first half of the trial (not counting the closing argument scheduled for February 1, 2019), I believe Qualcomm made the mistake of having too many cooks in the kitchen and partly the wrong cooks in certain places. The name of the game is not how many law firms you get involved. It's how effective they are, how "suitable to task" in tech lingo.

Cravath is Qualcomm's lead counsel against Apple, but the lead here in San Jose was given to Bob van Nest. Mr. van Nest himself has class and style, though he's primarily good at jury trials, and Judge Koh is the very opposite in terms of competence and professional coolness from a layperson jury. He's got a great reputation in this district, and Cravath is HQ'd in New York State.

But in retrospect I believe Qualcomm should have given Mr. Bornstein the lead here. He's been lead counsel in other high-profile antitrust cases. It seems to me that he's the smartest member of Qualcomm's trial team here, and generally all the Cravath lawyers here appear classier and more effective than the Keker van Nest lawyers apart from Mr. van Nest himself. There really is a very noticeable difference between one of the most reputable law firms in the country and a regional player good at misleading juries such as in the Oracle v. Google case, where they had a great jury strategy and benefited from a judge who made some key decisions against Oracle that the appeals court unanimously overruled in two different years. By contrast, even if Qualcomm loses (and I still think that's more likely than not to happen), I don't think anyone can blame Cravath.

The next news cycle related to Qualcomm's patents is only about six hours away. At midnight Pacific Time, or 9 AM Central European Time, the Mannheim Regional Court will announce decision, which can be a final judgment or a procedural order, on one of Qualcomm's German infringement lawsuits against Apple targeting Intel-powered iPhones. The patent-in-suit is EP2460270 on a "switch with improved biasing" ("biasing" in this context basically meaning that one voltage gets to control another).

I'm not going to stay up, or get up at midnight, for a nuisance lawsuit (which is all that this one is in practical terms), but I'm sure many will hear about the decision, so I'll quickly explain what may happen and why it's a pointless lawsuit in any event.

  • If the court clears those iPhones of infringement, it will most likely be because the court, more or less sua sponte if I understood it correctly (I watched the trial a few months ago), developed a claim construction approach based on an unasserted parallel claim.

  • The court might also stay the proceedings pending a parallel nullity action. The Swedish patent office provided an opinion according to which there's nothing inventive about that patent. And it did so on an independent basis, without any specific theory being presented to the examiner (just the prior art references).

  • Even if the court refrains from deciding the case based on its great claim construction idea and also declines to attach weight to the Swedish patent examiner's analysis, and formally enters an injunction, so what? At trial it was undisputed between the parties, and mentioned in open court, that Pegatron, one of Apple's contract manufacturers, is licensed to the patent. So there wouldn't be any bottom-line impact on the availability of any iPhone model in Germany.

Among various strong points, the weakest point Mr. Bornstein made in his cross-examination of Mr. Lasinski related to a hypothetical question they had asked him in early 2018 about whether he'd consider Qualcomm's non-SEPs more valuable if Qualcomm obtained injunctions against Apple in foreign courts. We know that some injunctions have come down: two in China (over two patents), and two in Munich, Germany (over the same patent in both Munich cases, just targeting different Apple entities). The problem with Mr. Bornstein's attempt to leverage those decisions now is that those injunctions still don't give Qualcomm any serious leverage.

The Chinese patents-in-suit have, according to Aple, been worked around by iOS 12. We'll have to wait until it becomes clearer whether the Chinese court agrees, but given that those are non-standard-essential software patents, it's perfectly plausible.

The Munich decision affects only 3% of Apple's German sales of the iPhone 7 and of the iPhone 8: direct sales to end users through its 15 retail stores and its German online store. Even those 3% of the sales of the two oldest iPhone generations on sale in a market that generally isn't huge for Apple (significant, but far from substantial) aren't really lost because some will buy other iPhone models instead and others will simply buy those iPhone models from resellers.

Anyone can see on the Internet that the "enjoined" iPhone models are still widely available in Germany.

Even the best-case outcome of the Mannheim case up for a decision in about six hours would fall short of the minimal impact of the Munich injunction: the worst-case scenario for Apple, based on what was said at trial (and in the court's preliminary opinion militated against an antitrust issue int his case), would simply be that all German iPhones (of the affected generations) would have to be manufactured by Pegatron.

Another Mannheim decision is scheduled for February 19, 2019 in a case in which the court strongly suggested to Qualcomm to stipulate to a stay given serious doubts about the validity of the asserted claim.

In between those two Mannheim decisions, the Munich I Regional Court will rule (on January 31) on a bunch of lawsuits related to a patent family that Qualcomm is asserting against Apple's Spotlight search. iOS 12 contains a workaround based on what was discussed at trial.

Share with other professionals via LinkedIn:

Sunday, January 13, 2019

Bipartisan concerns over Qualcomm's efforts to avoid competition from Intel expressed in filings with ITC

A headline starting with the word "bipartisan" reminds us of what happens in the absence of bipartisanship: the cessation of the federal government, commonly called "government shutdown." With paychecks not going out on time, it's more than remarkable that the FTC v. Qualcomm antitrust trial in the Northern District of California is set to continue irrespectively of the shutdown. I agree with what MLex's chief antitrust reporter Mike Swift wrote in the tweet below (both parts--admiration of the FTC staff's efforts and the impression that it has a strong case against Qualcomm):

When an antitrust defendant faces a strong case, and when that case tackles the core of a business model, a company may very well get desperate and try alternative routes in order to avoid a judicial decision that quite probably isn't going to be pretty from Qualcomm's perspective. Qualcomm has not yet presented its own case-in-chief. The FTC will rest its case on Tuesday (as per the current schedule), and then Qualcomm will be in the driver's seat for a few days. But Qualcomm's lawyers have cross-examined the FTC's live witnesses, and Qualcomm's lead counsel Bob van Nest made an opening statement laying out his defensive strategy. Day 2 went better for Qualcomm than the other three days, but all in all the FTC still controls the center of the chess board, with Qualcomm clinging to a last line of defense: trying to build a basis to argue that the FTC failed to prove actual anticompetitive harm. After four trial days, including testimony by Qualcomm CEO Steve Mollenkopf and Qualcomm president Cristiano Amon, it appears increasingly unlikely that Qualcomm can fundamentally challenge the FTC's allegations and theories. Qualcomm's lawyers are trying many things, and they're trying smart and hard, but it's largely just like scratching at the periphery of the issues.

Qualcomm has an army of public relations professionals and lobbyists working on this case. They've orchestrated a lot of things, some of which Apple CEO Tim Cook complained about and dismissed as falsehoods (without going into specifics) in a recent CNBC interview, and the latest prouct of those efforts is this Fox News article published on Monday morning, entitled "Trump allies warn Obama-era FTC suit against US firm giving boost to China."

I've often shared Fox News articles on social media. I admit it's my favorite TV news network by far and away, and I've had to defend it against liberals who criticized me for sharing articles from what they considered an unreliable source. One story on an antitrust case that misses the point doesn't change how I view Fox News in general. It's all too easy to see they've been used by people who in turn have been used by Qualcomm.

Let's start with the headline. As for "Obama era," I already explained in my post on FTC chairman Joseph Simons's recusal from this case (now I learned from the Fox article that it's because a law firm he was a partner of had Qualcomm for a client) that there are indeed decisions that then-outgoing President Obama made that were against American interests, such as an indirect sponsorship of suicide terrorism or a "refugee" dael with Australia when anyone with half a brain should understand that people who travel all the way from Africa or the Middle East to Australia (!) to seek asylum there, after crossing and bypassing numerous safe countries, lost their refugee status even under the Geneva Convention a long time (and many thousands of miles) ago. But the FTC's antitrust case here is not an ideological issue.

Just like Fox News doubts that my favorite president is aware of the details of this case, I guess then-outgoing President Obama was, in January 2017, more interested in his book deals, in some final retribution against Israeli prime minister Benjamin Netanyahu, in a long vacation, and a hundred or a thousand other things than in fair baseband chipset competition. Presumably he was just told that there was this investigation that had been going on for two years and the transition of power shouldn't delay filing the related lawsuit. Statistically, no president faced more Congressional slow-rolling with respect to appointments of government officials than President Trump. A lot of time could have been wasted, and at some point Qualcomm might have complained about a protracted investigation.

The most counterfactual part of the headline, however, is the idea of "giving [a] boost to China" by letting a company like Intel compete on a more level playing field and by giving a device maker like Apple more choice. Largely, the whole Fox News article just uses China as a bogeyman. Huawei's testimony was great, but merely validated and supported what U.S. companies such as Apple, Intel, or Motorola Mobility (though now owned by Lenovo, the stories go back to the times when it wasn't, and it still employs many people here), or companies from countries that are longstanding U.S. allies such as South Korea, have said. If the FTC's evidence were limited to testimony by Huawei, then I could see why some people might find this strange. But not when there's a long list of companies saying essentially the same things and even more than Huawei did. For an example, Apple and Intel provided insights that Huawei didn't and to some extent simply couldn't have.

What makes that obsession with China even more counterfactual is the fact that Qualcomm itself is known to favor China for the first 5G rollout. I just googled these two stories:

The suggestion (in the Fox News article) that playing videotaped testimony by Huawei is like different parts of the government working against each other (in light of concerns over buying Huawei infrastructure) conflates totally unrelated issues. The FTC is seeking an injunction against certain types of conduct on Qualcomm's part, none of which relate even indirectly to the procurement of Huawei products.

Neither Democrats nor Republicans have a patent on pro-competitive action, and there's no reason why Republicans should grant the Democrats one. The CCIA's Patent Progress website has published three public-interest statements filed by members of the U.S. Senate and the U.S. House of Representativesin response to the United States International Trade Commission's mid-December notice relating to the ITC's investigation of Qualcomm's first complaint against Apple (with Qualcomm seeking an import ban, which is the ITC's sole remedy).

The United States Senators and United States Representatives who filed those letters, which support Administrative Law Judge Thomas B. Pender's recommendation that no import ban be ordered given the anticompetitive effect this would have on Intel's ability to compete with Qualcomm, include members of both parties of Congress. While all members of Oregon's Congressional delegation who signed a joint letter are Democrats (well, Oregon doesn't have a Republican Senator at the moment, and only one of five U.S. Representatives from Oregon is a Republican), Congressman Andy Biggs from Arizona (member of the House Judiciary Committee as well as the Committee on Science, Space and Technology) and Congressman Darrell Issa from California are Republicans.

What's particularly interesting about Mr. Issa's letter is that he's a tech entrepreneur, a patent holder, a party to past patent litigation as a plaintiff as well as defendant, and he personally testified before the ITC in connection with a case in which Broadcom was seeking an exclusion order (U.S. import ban) against products incorporating Qualcomm baseband chips. Could a politician possibly have a more knowledgeable background to submit such a public-interest statement? Hard to imagine.

Mr. Issa would also like the district court to adjudicate the FTC's antitrust case now. Obviously, an exclusion order against Intel-powered iPhones would potentially run counter to a finding that Qualcomm illegally sought to avoid competition from Intel.

Those public-interest statements mention 5G: for the U.S. economy, it's better to have Intel compete with Qualcomm on 5G. In the end, both companies will be more innovative then.

In the past, Republican politicians such as Senator Mike Lee (Utah) have spoken out clearly against the abuse of standard-essential patents.

To combat anticompetitive conduct is not, and should not be misportrayed, as a "blue" issue, just like border security should not be a "red" concern. Qualcomm should focus on its case-in-chief, starting Tuesday, and if (as I believe is more likely than not to happen) it becomes necessary, it should adapt its business practices, which it certainly could if it wanted and had to.

Share with other professionals via LinkedIn:

Saturday, January 12, 2019

Testimony sheds light on dynamics of Qualcomm's negotiations with Apple, Samsung, VIA Telecom

On Day 4 of the FTC v. Qualcomm trial (Friday, January 11) in the Northern District of California, testimony covered the whole range of issues from market definition to anticompetitive harm to potential justifications. But the most interesting parts related to the dynamics of Qualcomm's negotiations with such industry players as Apple, Samsung, and VIA Telecom, as well as a past power struggle between Qualcomm's patent licensing and chipset divisions. When I just said "interesting," I primarily meant that it satisfies our curiosity as to what happened behind the scenes years ago and ultimately led to the mess we're learning more and more about. It's key to distinguish at all times between what's really outcome-deteminative to the #ftcqcom antitrust case and what's at best secondary, or even totally irrelevant, to the decisions Judge Lucy H. Koh will make after the ongoing bench trial. She's so much focused on the facts and the law that whitewashing and badmouthing won't help or hurt the way such tactics might affect a jury's feelings toward the parties.

Then-VIA Telecom's (now Intel's) Mark Davis described efforts to reach an agreement on IP matters with Qualcomm as more of a "dictation than negotiation," just because of Qualcomm's power and its approach to other parties. While VIA Telecom, according to the testimony, was too small to pick a fight and decided to buy the "ticket" that Qualcomm offered to its "walled garden," Apple's vice president of procurement, Tony Blevins, talked about the "watershed moment" that made Apple realize it had to formulate a strategy for independence from Qualcomm. So what had happened? Cristiano Amon, now Qualcomm's president, told Mr. Blevins, according to the testimony, that Apple didn't have a choice but to accept Qualcomm's terms (a mix of patent licensing and chip supply terms). Mr. Blevins testified he usually doesn't get involved with licensing negotiations, but in Qualcomm's case licensing and chip supply were "hopelessly entangled," so it was the only supplier with whom he also had to discuss license terms as part of the equation. On the bottom line, what Qualcomm demanded appeared unreasonable to him, and Apple, which typically wants to have (again, according to Apple's testimony) two to six suppliers for any given one of the iPhone's roughly 1,000 components, gets nervous--Mr. Blevins said--when market forces and competition no longer seem to work. Mr. Blevins stressed it's not just about pricing but also about quality, reliability, and innovation, to name but a few other aspects.

As I quoted the most famous work of Spanish literature in connection with Judge Koh's unsealing orders, the pitcher will go to the well once too often. The same apparently happened in the sense of industry players deciding at some point that "enough is enough"--and that's not just Apple, but a really long list by now, as this trial has shown and as we know from antitrust proceedings in four jurisdictions on three continents.

Without any storytelling aspect to it, Samsung's Injung Lee confirmed that negotiations with Qualcomm were unique due to the leverage it had.

In terms of outcome-determinativeness, those stories of companies feeling bullied are not irrelevant but just a puzzle piece. They serve to show that Qualcomm had leverage, was conscious of its leverage, and was prepared to bring its leverage to bear. Nothing more (other than providing a historic explanation of the current conflict). Nothing less.

Apple's Mr. Blevins told another anecdote that is directly relevant to one of Qualcomm's defensive arguments in the "No License-No Chips" context--an argument that I considered relatively weak, but that was made nonetheless. As the eighth paragraph of my most recent post on "No License-No Chips" explained, Qualcomm's witnesses claimed that Mr. Reifschneider, former head of Qualcomm's licensing business, may have threatened with cutting off chip supplies but didn't actually have the authority because it was a different division. What we heard from Apple's Mr. Blevins indicates that the licensing business was in a far more powerful position than Qualcomm's witnesses would now concede:

Mr. Blevins said he felt realy bad for Mr. Amon when, in a three-way meeting on the executive floor of Qualcomm's San Diego HQ with Mr. Amon and Mr. Reifschneider, the latter at some point interrupted Mr. Amon and (in other words) told him to shut up when Mr. Amon made a constructive and conciliatory suggestion to Apple. According to the testimony, Mr. Reifschneider said: "I run a division that accounts for two thirds of this company's profits and your division only for one third!" Mr. Blevins found this abrasive behavior in front of a customer (Apple) highly unusual.

I've seen Mr. Amon's live testimony and Mr. Reifschneider's videotaped testimony. Faced with the choice between which of the two I'd want to negotiate with, or work for, I'd clearly consider Mr. Amon someone with whom it's simply easier to work out solutions and whom one can ultimately trust, though he's tough as well. Meanwhile, Mr. Amon is president of the whole company, though in his testimony he sought to downplay his influence over the licensing business, arguing that the QTL division's internal reporting still bypasses him.

Judge Koh hears to many commercial cases that I doubt she'd have been overly impressed by the notion that a single company controlling two divisions would just let either division run totally independently of the other--especially since even Qualcomm's witnesses don't deny that the various agreements (patent licensing, chip supply, incentives for exclusivity) were always negotiated in parallel and concluded at the same time. Still, the story of QTL, not QCT, being in the driver's seat in a meeting with Apple makes it even easier to see Qualcomm's priorities and the actual command chain resulting from them.

Finally, in the last testimony of the day--by Qualcomm CEO Steve Mollenkopf--another aspect of the various Apple-Qualcomm negotiations came up: the question of who took the initiative to enter into an exclusive deal that, during the period from 2012 to 2016, made it commercially extremely hard to justify for Apple to work with any other baseband chipset supplier than Qualcomm. It's one of four types of conduct the FTC is asking the court to declare anticompetitive and to prevent from happening again.

There are three types of testimony that we must reconcile now--and we'll talk about the extremely limited relevance this has to the FTC's legal theories:

  1. Apple focused on how Qualcomm's exclusive deal prevented it from working with other baseband chip providers during the period between when it originally received chips from Infineon and when it started working together, firstly in an iPad (learning curve less steep because only data, not voice), with Intel, the acquirer of Infineon's mobile chipset division, in late 2016. As Apple explained, there was a "clawback" (which Qualcomm was forced to confirm it didn't have in any other agreement of this kind) of incentives/rebates granted to Apple in exchange for exclusivity. Qualcomm's counsel tried to focus on the fact that Apple remained free to work with others, but just like in the patent exhaustion context (where a covenant not to sue is for good reason treated analogously to a license agreement by the Federal Circuit), it makes sense to focus on actual, practical, commercial effects. A clawback clause for rebates is just a contractual penalty by any other name.

    By the way, I mistakenly wrote "callback" on Twitter--that's because I'm a programmer and we do callbacks in asynchronous programming. Also, the word "recall" came up at that point.

  2. Mr. Mollenkopf responded to his lead counsel, Bob van Nest, that it was Apple who had offered exclusivity to Qualcomm in order to get a better deal.

  3. Minutes later, Mr. Mollenkopf was asked about whether it was a good deal and he explained that it ultimately was, but initially Qualcomm was unhappy that Apple had not made a volume commitment in exchange for a $1 billion incentive under their first 2011 "Transition Agreement." Mr. Mollenkopf said he'd have preferred to have a guaranteed volume to offset the $1 billion "we offered." He literally said "we offered."

I don't think Mr. Mollenkopf contradicted himself. I'm obviously skeptical of many of Qualcomm's positions and practices as I see regulators from around the globe taking action and virtually an entire industry (though we have yet to hear the witnesses Qualcomm will present during its case-in-chief) complaining. But it would be ill-meaning to interpret "we offered" as meaning that Qualcomm took the initiative, just minutes after he had affirmatively said Apple was first to suggest.

What I believe instead is that Mr. Mollenkopf meant "we offered" in the sense of "we contractually committed to giving them" a billion dollars--as opposed to "we proposed," much less "we took the initiative to propose." So we've reconciled any potential inconsistency between statements 2 and 3.

A Twitter troll blamed me for literally quoting Mr. Mollenkopf, which was ridiculous given that I had just prior to that "we offered" quote retweeted two (!) tweets (one by MLex, one by CNET) about Qualcomm's representation that exclusivity was Apple's idea when they negotiated the 2011 contract. I appreciate PatentlyApple's pointing out that, contrary to what the trolls say, I actually have a track record of disagreeing with Apple on some major issues. There's no one on the Internet who has called more aggressively for the invalidation of certain signature iPhone patents. No one has more aggressively opposed Apple's patent damages claims. Seriously. But let's not get distracted too much. Let's reconcile any potential contradiction between statements 1 and 2.

Apple testified--and we didn't even need to hear such testimony because we could have figured--that Apple told Qualcomm the total cost (licensing plus chipset sales) was too high and Apple wanted to pay less. That's the normal course of business in procurement.

It's undisputed that Qualcomm told Apple that the effect of paying less in the first place could be achieved through "incentive payments." Here, again, let's focus on substance, not labels. The bottom-line effect was a rebate.

It's furthermore undisputed (because even a Qualcomm witness said so) that Qualcomm told Apple, in my words: if you want a better deal, you have to give us something of value.

If we assume that no one lied or misled under oath, then the sequence of events would have been this:

Apple: We want a better deal. Not those cutthroat terms. (Discount or incentive payments--just a better deal on the bottom line.)

Qualcomm: Only if you give us something of value to us. (Such as a volume commitment.)

Apple: Well, how about exclusivity?

Qualcomm: We'd prefer a volume commitment. But exclusivity can also work.

Now, what does this mean for the legal case? With a jury, it could have huge psychological impact. Blame-shifting in the eyes of some laypeople. But remember, this is a bench trial. Judge Koh is unsusceptible to any games lawyers and witnesses like to play.

I doubt that this question of who, under what circumstances, took the initiative to offer exclusivity has any weight. If this were a criminal price-fixing case, then the one who came up with the scheme might be sentenced to a couple more years than the other guy, but the latter would still go to jail, too. In price-fixing, they'd both have an anticompetitive benefit because they'd gang up on customers (the same group of customers, in fact). In the Qualcomm-Apple case, the benefit to Apple was a deal that an Apple witness basically described as "less bad" than the original deal, but still not fair in Apple's view, while Qualcomm had the benefit of raising the entrance barrier to its (Qualcomm's) competitors--with a customer the FTC argues (and Intel confirmed) is strategically extremely important for a component supplier to gain market share, build a reputation, and generate volume. By contrast, Apple didn't foreclose any market to its own competitors. And it certainly didn't leverage this deal to undercut anybody.

Also, a presumably much bigger deal was entered into between the two as of 2013. Structurally the same according to Mr. Mollenkopf, but with different numbers. And we don't know what happened in those negotiations. But again, that's not even the key point.

If it's anticompetitive, it's anticompetitive regardless of who was first to propose. If you were Intel or MediaTek, would it have made a difference to you--when Apple told you it was economically irresponsible (at least from a short-term perspective) to buy your chips and trigger Qualcomm's "clawback"--which of the two parties had come up with that idea? Obviously not. And that's why I doubt it will matter to Judge Koh (or, later on, the appeals court).

Share with other professionals via LinkedIn:

Thursday, January 10, 2019

Qualcomm offered Samsung and MediaTek a "covenant to sue last" instead of a chipset license

No one ever doubted that Qualcomm is an innovator. Any disagreements anyone has with them relate exclusively to how much leverage they can rightfully get from their patents and how they can leverage their market position in premium baseband chipsets. But Qualcomm's lawyers also seem quite innovative in their field. Only Qualcomm could come up with the idea of offering a "covenant to sue last" to companies who requested a license.

"Covenant to sue last." Not a straightforward license. Not a covenant not to sue (which at least means you'll be left alone). But a promise to sue someone only if all other alleged infringers have previously been sued. It's like saying: "We very much do reserve the right to come after you, but we'll deal with your competitors first. You may have to defend yourselves at some point, but if that happens, you'll at least be our most favored adversary timingwise. Now, in exchange for this tremendous favor, we want you to commit to all sorts of things that render you less competitive."

As I mentioned in my report on Huawei's testimony on Day 2 of the FTC v. Qualcomm antitrust trial in the Northern District of California, Qualcomm appears to be extremely afraid of patent exhaustion. Patent exhaustion--which protects the downstream from assertions of patents already licensed somewhere upstream--can result from an authorized sale (such as selling a chipset that substantially embodies certain patented inventions) as well as from a license (such as licensing a chipset maker who then sells the licensed product to device makers). Qualcomm is concerned about both exhaustion avenues, but in this post it's all about the second scenario (licensing).

As my Day 2 report also explained, the Federal Circuit held in its 2009 TransCore v. Electronic Transaction Consultants (ETC) opinion that exhaustion can be rooted not only in what is narrowly defined as a license but also by a covenant not to sue, which is a license by any other name given that a patent is, simply put, a license to sue infringers, so if you give up the right to sue, the beneficiary of such a covenant is effectively licensed.

It makes a lot of sense to focus on the actual, practical effects of a commercial agreement rather than hairsplitting and purely formalistic aspects. A lot depends on the judges. Some are more receptive to commercial arguments than others.

So Qualcomm and others realized at some point that even a covenant not to sue was not nearly as reliable an exhaustion-avoidance strategy as they once thought. But they still wanted to be able to do non-exhaustive deals in order to hamper competition from rival chipset makers through terms that came with restrictions (on whom they were allowed to sell products to) and reporting requirements (so Qualcomm would get sensitive information about a competitor's business). We hear from Lenovo on Day 1 (see the MediaTek paragraph of this post) that they had to fear that Qualcomm would enjoin MediaTek from selling chipsets to Lenovo the moment Lenovo wasn't going to have a license to Qualcomm's patent portfolio.

The name of the game for Qualcomm was and is, therefore, that it seeks to thread the needle and do convenient (and apparently anticompetitive) deals with rival chipset makers while navigating around patent exhaustion.

On Tuesday (Day 3), Andrew Hong of Samsung's chipset business and Intel's general manager Aicha Evans testified that Qualcomm simply refused to grant exhaustive patent licenses to them. In Samsung's case, it was partly about Samsung itself but the largest part of the testimony related to a contemplated joint venture, named Dragonfly, between Samsung and some Japanese companies. According to the testimony, the number one reason for which Dragonfly failed to materialize was that the various Dragonfly partners assumed NTT DoCoMo, which already had a license deal with Qualcomm, could provide the Dragonfly JV with a chipset-related license to Qualcomm's standard-essential patents. When Qualcomm made it clear that it wouldn't do this (Mr. Hong says Qualcomm told him they weren't going to support something that would have made Samsung a baseband chipset competitor within a year, given that it otherwise takes several years to enter that market), Project Dragonfly was over before it began.

Thanks to a particularly important one of Judge Koh's unsealing decisions, the FTC's pretrial brief tells us what Qualcomm actually did offer Samsung's chipset business at some point. An email by a former Qualcomm president said the following:

"[W]e were also asked for licenses by Intel and TI at a minimum, probably others (e.g., Samsung, Mediatek) as well, and we refused to enter into anything other than a non-exhaustive covenant (or covenant to sue last in the case of SS and MT)." (emphasis added)

No company that wants a real license is going to be satisfied with a "covenant to sue last." And while Qualcomm presumably told the companies it offered such a deal that in practical terms they were going to be fine since there's a huge number of companies in the world and there would always be some whom Qualcomm wouldn't sue, the problem is still that if you're general counsel of a chipset maker and your CEO asks you whether the company is reliably safe from patent assertions, you must answer: "It's not safe because Qualcomm could at some point decide to simply sue the whole world and then we'll be hit like anybody else." And, at any rate, such a covenant does not solve the problem of indemnification. Samsung's Mr. Hong said in his testimony that "in [his] experience the IP indemnification clause tends be one of the biggest items debated" in chipset supply negotiations.

Intel's Aicha Evans explained the same issue just from a different angle. She said that device makers obviously have to look at their total cost: what they pay for the chipset as well as any patent licensing costs related to it. If a chipset was licensed, they'd pay a price and that would be it. However, when there's a situation where Intel sells a chip and knows that a competitor (who'd actually like to drive Intel out of this business) is then going to collect patent royalties from the device maker that Intel doesn't even know (certainly not beforehand and, due to confidentiality clauses, usually not even afterwards), it makes it extremely hard for both Intel and the device maker to actually calculate their costs. It discourages purchases and investment.

During this trial, some Qualcomm-internal communications have been shown according to which Qualcomm's management always feared that licenses to rival chipset makers would have a devastating effect on Qualcomm's patent licensing business. Here's an interesting passage (just recently unsealed) from the FTC's trial brief:

"Qualcomm's internal documents recognize the impact that offering competitors FRAND licenses would have on Qualcomm's ability to secure elevated royalties from OEMs. In 2005, Qualcomm's Marvin Blecker explained that making a license available to a chip competitor would impair Qualcomm's ability to collect high royalties from OEM customers: 'we absolutely cannot give a chip supplier a full license to our IP with pass through rights to his customers as that would have the potential of severely impacting our subscriber licensing program.' [...] Qualcomm's views were unchanged in 2015, when it concluded that granting a FRAND license to Intel 'would destroy the whole current QTL [licensing] business.'"

The decision on Qualcomm's obligation under antitrust law (with respect to two FRAND pledges, this has already been resolved favorably under contract law) won't destroy Qualcomm's licensing business in its entirety, but it would make it considerably harder for Qualcomm to collect supra-FRAND royalties, and it would make it harder for Qualcomm to avoid competition on the merits with other chipset makers.

Four companies making chipsets have testified by now. Samsung's chipset division, Huawei (now specifically referring to the part of Mrs. Yu's testimony that related to chipset licensing), and Intel definitely supported the FTC. MediaTek tried to, but wasn't nearly as effective during the public part of the testimony. The sealed part related to some contract between Qualcomm and MediaTek, and maybe that document spoke for itself. But even if one is skeptical about MediaTek's testimony (again, not because of the intention, but because the witness wasn't as tactically shrewd as the other witnesses, most of whom are lawyers), the testimony the court heard from chipset makers appeared overwhelmingly favorable to the FTC's cause.

Share with other professionals via LinkedIn:

Seeking to defend "no license-no chips" in FTC trial, Qualcomm says it never actually cut off chip supply

When a witness testifies in FTC v. Qualcomm, questions related to different ones of the Federal Trade Commission's antitrust allegations and to different ones of Qualcomm's defenses are asked. But yesterday, on Day 3, two topics got particular attention: Qualcomm's "no license-no chips" policy (which this post is about) and its refusal to license rival chipset makers (which the next post will focus on).

While Qualcomm can't deny that it has a policy of supplying chips only to companies licensed to its patent portfolio, its lawyers have nevertheless come up with a multiplicity of attack vectors to make the case against the case against Qualcomm in this context:

They try to define the relevant market broadly enough to be able to claim they didn't have a monopoly.

They deny that the "no license-no chips" policy resulted in the acceptance of supra-FRAND patent royalties by licensees. Qualcomm argues that its standard 5% rate is FRAND anyway. The FTC will have a licensing expert testify to the opposite, and there's Huawei's testimony that Qualcomm's royalties alone account for 80%-90% of its total patent licensing cost, as well as similar testimony from other licensees. In an effort to argue that its royalty rate is accepted regardless of "no license-no chips," Qualcomm points to companies that, at different points in time, accepted that rate despite not being chipset customers at the time (in some cases because Qualcomm wasn't even selling chipsets yet).

One of the problems here is comparability, especially of very old license deal that involved foundational patents that have expired since. Foundational patents can be so valuable that even a small number of them gives a patent holder more leverage than a larger number of patents building on breakthrough innovations later (which makes them relatively narrow). And even if Qualcomm had (as it said it might have) already made FRAND pledges with respect to certain patents at the time of concluding a given license deal, the FRAND licensing obligation actually doesn't kick in until the standard has been adopted and someone requests a license in order to implement the standard.

Qualcomm's last line of defense is to claim that there was no bottom-line (rule of reason) anticompetitive harm. First, they categorically deny that there would have been anticompetitive harm even if some companies had accepted supra-FRAND royalties; but that's not really a mainstream position under the case law, no matter how badly Assistant Attorney General Makan Delrahim would like things to work that way. Second, they stress that they never actually did cut off, or would ultimately actually have cut off, chip supplies. Third, they say that whoever threatened anyone with a cutoff of chip supplies if someone didn't take a patent license (particularly, Eric Reifschneider) wouldn't have had the authority anyway to make a decision on chipset shipments. That point was stressed in particular by Qualcomm president Cristiano Amon.

At this point, I'm unconvinced of those defenses being strong enough to avoid a holding that at least some of Qualcomms behavior in the "no license-no chips" context was an antitrust violation and indeed caused anticompetitive harm (in the form of consumers ultimately paying higher prices for their wireless devices). I said "at this point" because there will be seven more trial days, and some testimony was sealed.

Let's start with the last point: the claim that people like Mr. Reifschneider didn't have the authority to carry out the threat. It's obvious that in a large organization like Qualcomm, and especially when there are formally separate entities in place (QCT for the chips and QTL for patent licensing), someone working in division A can't just make decisions for division B. However, that doesn't mean the guy from division A couldn't just call someone in division B, or take the matter to a higher level of the organization, and chipset supplies would be cut off just because someone didn't agree to Qualcomm's licensing terms.

The FTC showed a passage from a Qualcomm chip supply agreement that its trial brief (significant parts of which have been unsealed by Judge Lucy H. Koh) also mentions:

"For example, Qualcomm's supply agreement with LG Electronics provides that Qualcomm 'may terminate this Agreement if [LGE] is in default under the License' [...] Other supply agreements contain similar terms."

Footnote 4 even says that "[a] term to this effect appears in virtually all of Qualcomm's Component Supply Agreements."

Such a cross-over clause that creates a hard legal connection between a license agreement with one entity and a supply agreement with another entity makes it rather clear that those entities (QCT and QTL) did not operate totally indepenently, but very much kept the overarching interests of the parent company in mind.

Imagine you had an agreement with a telco and one with the postal service, and the postal agreement would say that if you're ever in default under your telephone contract, you might no longer get your mail. Would you then believe that those are independent entities? Obviously not. Nor would you feel safe because your account manager on the telco side isn't in charge of mail delivery.

With different witnesses having confirmed "no license-no chips" threats and even some Qualcomm-internal documents referring to such threats (also including a handwritten note such as "Eric [Reifschneider] constantly threatening [Motorola]"), Qualcomm's last-ditch defense is to argue that the threat was never actually carried out, nor would there have been a will to carry it out.

Mr. Amon, who was more forthcoming and likable than most of the other Qualcomm witnesses heard so far, explained the negative repercussions it would have had, such as that carriers relying on a particular Qualcomm-powered mobile device would have lost faith in Qualcomm if a termination of chipset supplies had happened. Also, other chipset customers might have drawn their conclusions (though they might not have had much of a choice anyway, at least in the premium segment)

I don't doubt that there would have been a very significant downside involved for Qualcomm. Mr. Amon explained that part very well. But the threats are now an established fact, and in the absence of cases in which Qualcomm would have continued chipset supplies for a long period of time despite someone refusing to take or renew a patent license, or despite someone actually challenging patent licensing terms (such as by seeking a judicial FRAND determination), the answer simply appears to be that Qualcomm never had to carry out the threat because it served its purpose.

To sum it up, Qualcomm's lawyers defend in multiple ways against the "no license-no chips" tying allegation, but no single one of those attack vectors appears strong at this stage. The parties' expert witnesses haven't been heard yet, and the economists will talk about market definition and FRAND royalty levels. What really doesn't look like it could solve the problem for Qualcomm is the notion that only because a threat was so effective that it never actually had to be carried out, or that those who made the threat lacked the authority to carry it out, or that there would have been some backlash if Qualcomm had done so.

Based on what's been said and presented so far, it's my impression that stories of licensees having accepted more or less the same royalty rates even without buying chipsets at the time (or even without Qualcomm offering any at the time) are something the FTC must counter effectively. And I believe it will be able to do so. Also, even if some people accept exorbitant royalty rates, those rates still aren't FRAND.

I particularly liked one thing the FTC said in its opening statement last Friday. The FTC's lead counsel in this case, Jennifer Milici, said that Qualcomm may well point to companies that don't respect intellectual property and refuse to take a license, but if Qualcomm's patents are as valuable as Qualcomm says they are, Qualcomm need not be afraid of proving the value of those patents through infringement litigation in cases where it may be necessary. The FTC doesn't deny that it isn't always easy to successfully enforce patents, but--the FTC argues--Qualcomm can't violate the antitrust laws only because it seeks to avoid what every other patent holder must do.

What is not at issue in this trial, but what I'd like to add nonetheless: it's absolutely key here to let patent holders prove the value of their patents through litigation. Arbitration is not a substitute for litigation unless both parties voluntarily (without undue pressure of the "no license-no chips" kind) choose arbitration, under mutually agreed-upon parameters, over litigation in Article III courts and the ITC. A few months ago I explained this in the context of Huawei v. Samsung.

Share with other professionals via LinkedIn:

Wednesday, January 9, 2019

Judge Koh enforces transparency, unseals key passages of documents in FTC v. Qualcomm

There's a flurry of activity surrounding the FTC v. Qualcomm antitrust trial in the Northern District of California. The previous two posts this morning were about Apple CEO Tim Cook's statements regarding Qualcomm in a CNBC interview and Qualcomm CEO Steve Mollenkopf's testimony, which the FTC has scheduled for Friday (January 11).

There are important developments not only in the courtroom but also on the docket of this case. Thankfully, Judge Lucy H. Koh of the United States District Court for the Northern District of California has identified a problem with Qualcomm (and in some cases maybe also third parties) demanding overredactions. As Judge Koh has already reminded counsel more than once during the trial, it is in principle a public proceeding. Confidential business information, if still competitively relevant today, must and will be protected, but public access to information is also important.

I complained about overredactions last month in connection with Qualcomm's proposed findings of fact and conclusions of law and expressed my hope that the court would not approve all of those far-reaching redactions. And indeed, the famous Don Quixote proverb ("Tanto va el cántaro a la fuente...") applied again: at trial time Judge Koh just noticed some excessive sealing requests and showed those sealing-happy lawyers that she's the gatekeeper.

Yesterday she reminded counsel of something she had unsealed, and she said: "I may unseal more and the more I'm watching..."

Today she entered an order that stresses the need to narrowly tailor any sealing requests (such as by redacting out only particular numbers and not entire passages only because of a few sensitive data points):

"If Qualcomm seeks to seal any of the exhibits to be introduced on Friday, January 11, 2019, Qualcomm shall file a narrowly tailored motion to seal today as soon as possible. Moving forward, the Court requests that Qualcomm file timely and narrowly-tailored motions to seal."

Judge Koh's orders to unseal have already proven very helpful to those monitoring the case. I'm step by step looking at what additional information has become available, and found a wealth of newly-unsealed information in the FTC's final pretrial brief, which is a whole lot easier to read and to understand after the FTC refiled it based on Judge Koh's denial of Qualcomm's excessive sealing requests (this post continues below the document):

19-01-08 FTC Trial Brief Af... by on Scribd

The effect of many passages having been unsealed is so significant that I recommend anyone who elected to read the original public version of the FTC's trial brief to re-read it now that it's far more informative, thanks to all those puzzle pieces that have been made public.

In the next two posts, I'll discuss what was said yesterday about "no license-no chips" and Qualcomm's refusal to license rival chipset makers, and in those posts I'll reference the new, less extensively-redacted version of the FTC's trial brief, as a number of facts have now entered the public domain that are key to understanding what certain witnesses said in yesterday's testimony.

Share with other professionals via LinkedIn:

Qualcomm CEO Steve Mollenkopf to testify in court on Friday (January 11): FTC v. Qualcomm

The Federal Trade Commission just submitted to the United States District Court for the Northern District of California its witness list for Day 4 of the FTC v. Qualcomm antitrust trial (Friday, January 11, 2019). In terms of what she says and how she says it, the real star witness at this trial has been Intel's Aicha Evans so far, and her unconventional but incredibly strong testimony will resume on Friday morning. With her personality she's a witness unlike any other you'll see in your life, and I'll talk about some of what she said in other posts later today. But in terms of sheer power, the superstar witness here is going to be Qualcomm CEO Steve Mollenkopf (this post continues below the image):

Mr. Mollenkopf has been with Qualcomm for more than 24 years and its CEO since 2014.

The number two executive, Qualcomm president Cristiano Amon, testified yesterday. Due to the nature of the questions asked, his testimony didn't reveal anything unexpected, though he seemed more cooperative than some other Qualcomm witnesses.

Mr. Mollenkopf's testimony on Friday will likely result in increased media attention to the trial. The court may even need to make use of the overflow room.

The issues in the FTC case will dictate the scope and focus of the questions asked, so we won't get to hear his response to Apple CEO Tim Cook's CNBC interview on Friday, according to which Apple has not had any settlement discussions with Qualcomm since the third calendar quarter of last year, contrary to representations by Qualcomm at all levels--also by Mr. Mollenkopf personally--that a settlement was near. Qualcomm's CEO may very well talk about past encounters with Mr. Cook, however, provided that Qualcomm's past relationship with Apple is on the agenda for his Friday testimony.

Share with other professionals via LinkedIn:

Will analysts stop buying Qualcomm's representations of leverage over Apple after Tim Cook's CNBC interview?

Day Three of the FTC v. Qualcomm antitrust trial was underway in San Jose (a couple of blog posts about it will follow later) when I saw a CNBC interview--Jim "Mad Money" Cramer went to California to interview Apple CEO Tim Cook--going viral. You can find the video and the complete transcript on this CNBC webpage.

With this unique style and format, Cramer made financial analysis on TV more entertaining than anybody before him. He's to Wall Street what Rush Limbaugh is to Washington politics: aggressive, outspoken, and unconventional, but none of that should discredit anyone's analysis. A hedge fund manager with decades of merger arbitrage experience called me a couple of days after IBM announced its plans to acquire Red Hat, and the first thing he said was: "Believe it or not, Florian, I watched Mad Money with Cramer last night." He was laughing, but it's a fact that even the pros pay attention to Cramer--from time to time at least.

Tim Cook explained to Jim Cramer why he believes in Apple's present and in its future, and about 17 minutes into the interview, Mr. Cramer referred to AAPL "naysayers" and said Qualcomm was telling him all the time that Apple would now have to settle the dispute with the chipmaker because of the decisions Qualcomm obtained in China and Germany. But Mr. Cook said no settlement discussions with Qualcomm had taken place since the third calendar quarter of last year, meaning that Qualcomm's symbolical victories in Germany and China simply failed to bring Apple back to the negotiating table, much less did they make Apple cave.

I know those stories about what Qualcomm tells Wall Street because I hear them from analysts all the time (and then I discuss with them how I view the state of affairs). As I recently wrote, telling investors that a settlement with Apple is near appears to be Qualcomm's #1 investor relations objective these days. Or I should say it appears to have been Qualcomm's #1 IR objective. At this point, all that Qualcomm will "accomplish" by reiterating the same fairy tales is further damage to its credibility with investors and analysts.

They weren't even consistent last year. In the summer I heard from an analyst that Qualcomm didn't have great expectations regarding its lawsuits in Munich, but thought it had an opportunity to win a case in Mannheim that would give it leverage and force Apple to settle. Then came a couple more Mannheim trials in September and October, and the way they went it can be ruled out that Qualcomm would get serious leverage there. In one of those cases the court even suggested that Qualcomm stipulate to a stay because the patent-in-suit appeared too weak.

Then, all of a sudden, Qualcomm started claiming that Munich was going to be the venue that would decide the dispute. But three weeks (minus one day) after that particular decision came down, the iPhone 7 and the iPhone 8 are still widely available in the country as I demonstrated earlier this week.

If I were a Qualcomm shareholder, I would take issue not only with the way Qualcomm communicates with me, but also with the absurdity of posting bonds over $1.5 billion (that must cost Qualcomm tens of millions of dollars per year) when there's hardly any impact other than making noise in the media and misleading Wall Street to think that Qualcomm has serious leverage, which quite apparently it does not.

For an update on the Munich situation, I emailed and called the appeals court (Oberlandesgericht München = Munich Higher Regional Court) and was told that Apple indeed appealed the pair of injunctions Qualcomm obtained and brought, as expected, a motion to stay their enforcement for the duration of the appeal. Qualcomm was ordered to respond to the motion, and the unspecified deadline for that response hasn't passed yet.

In that CNBC interview, Mr. Cook said it sould be beneath any company to do some of what Qualcomm is doing. When Mr. Cramer asked whether this referred to Qualcomm's representation of a settlement being in sight, Apple's CEO said it was part of it, but particularly stressed fake news planted by PR agencies on Qualcomm's behalf. While this wasn't mentioned, the fake "Draft Tim Cook 2020" campaign may be part of it.

I still think Qualcomm is a highly innovative and impressive company, but its credibility has suffered a lot in those two years since the FTC and, separately, Apple sued Qualcomm. The "house on fire" story is another example. It's the opposite of those settlement and leverage stories in terms of overstating the risks, but it's consistent in its untruthfulness.

Share with other professionals via LinkedIn: