FTC vs. Qualcomm: My Analysis of the Case Up To This Point

You don’t have to look hard, or far, to find the most high-profile legal battle going on at the moment. The case with the most industry implications and perhaps one of the most important wireless/telecom related legal dispute of our time is the FTC bringing anti-trust allegations against Qualcomm.

There is a vast array of fascinating dynamics about this case which has already had extremely high-level executive testimony thus far. I’ve never attended a court case, and since this one is in my backyard, a San Jose district court with the Honorable Judge Judy Koh presiding, I wanted to see the action for myself.

The legal chess match is something I find interesting at an intellectual level. I greatly enjoyed this part of what I witnessed on Friday. Part of this is their legal team’s strategy to bring out specific points or counterpoints in evidence, including witness testimony, and some of it is what each sides legal team fights to have sealed or unsealed in evidence, which means, what the public can see vs. what only the Judge can see. Overall, I think I fully understand what the FTC has built their case around and how Qualcomm intends to defend themselves against that narrative. While the case is still a few weeks away from being concluded, I’ll cover what I think is happening up to this point.

What the FTC Wants To Prove
The FTC has a few bullets in their gun. They firstly continued to seek to prove that Qualcomm’s business model is unique to Qualcomm. They essentially brought evidence and witnesses in an attempt to demonstrate that Qualcomm’s business is unique in the industry. The business model in question is not licensing of IP or selling of chips, and it is what their legal team calls “no license, no chips.” Essentially, and this is true, Qualcomm does not sell chips, meaning Snapdragon SoC chips, modems, and other core computing bits of silicon to any vendor who does not also sign a broader license with them. It is important to note there are chips Qualcomm sells that do not require a license but the most important ones that are also key innovations from Qualcomm are the chips which require a license.

While all these license agreements are 100% negotiable, the FTC lawyers went to lengths to prove in evidence and witness testimony that only Qualcomm has a no license, no chips business model. This is tactic number one for the FTC.

The second strategy of the FTC is to prove that the business model unique to Qualcomm leads to higher costs due to licensing, royalties, etc., and the effect is less competition. This is particularly important when it comes to licensing technology to a rival chip maker like Intel, Samsung, HiSilicon, etc. Qualcomm owns nearly all the patents for CDMA; therefore any competing modem maker would need to acquire those patents to make a CDMA modem and sell it to the broader market. I find this particular dynamic interesting because Qualcomm started as a licensing company and essentially created the standard, and underlying technology for CDMA. Those licensing standards fall under FRAND (Fair Reasonable and Non-Discriminatory) pricing. But the irony is, at least from the case of the FTC, that competitors to Qualcomm are complaining about unfair business practices from Qualcomm when in fact they would not be able to compete at all in the market if it was not for Qualcomm’s technology IP around CDMA.

As a result, the FTC expert witnesses are chiming in about Qualcomm’s monopoly power around CDMA due to the massive investment in IP and technology around CDMA Qualcomm spent billions in R&D to achieve. Qualcomm had the dominant position from an IP and technology standpoint concerning CDMA/LTE and possibly even 5G. But the FTC needs to prove that dominant position leads to not just monopoly pricing, meaning outside of FRAND, but also that it led to harm in the industry and for competition.

The Burden of Proof
With the fundamentals tactics of the FTC clear, the question is are they making strides in proving not just an abuse of market pricing but also the said practice led to market harm. So let’s explore a few of the points.

Qualcomm’s unique business model. While the FTC did a good job proving Qualcomm’s business model is mostly unique to them, there was some scrutiny of this point. When Apple’s head of procurement Jim Blevins was asked if any other chip maker required a license to buy chips he noted that only NXP came back to him with such a requirement. But he went on to say that when he balked at that request NXP quickly reorganized the terms and dropped that requirement.

What the FTC did a good job of proving was that Qualcomm’s business model, while not entirely exclusive, is much more demanding or rigid. Testimonies from key witnesses outlined how not only would Qualcomm not drop the requirement for a license to buy chips, but that the license from Qualcomm included a cross license of said companies IP. This was one of the areas that Apple, rightly so, had a huge problem with. The FTC proved that Qualcomm not willing to budge on key terms, as NXP did, meant that they felt they had more leverage and thus the power to dictate terms.

Interestingly, I learned Apple never signed a license with Qualcomm despite Qualcomm’s push for them to sign a license. Apple got around this because their suppliers had signed a license with Qualcomm and thus the two companies put together a series of agreements to bridge some of the gaps. Two key parts of this agreement that came out were Apple’s agreeing to exclusivity with Qualcomm for a period three years. During that agreement, if Apple remained exclusive to Qualcomm, Apple would receive incentives to the tune of one billion dollars. Interestingly, the FTC wanted these incentive payments to come across as Qualcomm incentivizing Apple to not buy from competitors, namely Intel. But the Qualcomm defense, with Qualcomm CEO Steve Mollenkopf, brought out that it was Apple that demanded the one billion from Qualcomm partly because of the engineering costs associated with a modem change. Mollenkopf said that such an ask from a customer is not uncommon, but the amount Apple asked for was more substantial than any customer asks previously. The conclusion here is while those incentive payments were worth it to both parties, in the end, the FTC’s attempt to make it look like these payments forced Apple not to choose a competitor was not a strong part of this particular tactic.

One other point worth mentioning is the FTC spent a few days, and are still hitting this angle with expert witnesses today, that Qualcomm’s threat of not providing chips any longer to any licensee, not in good standing is an enormous threat. However, when Qualcomm’s lawyers pressed Steve Mollenkopf, on this issue, he acknowledged that while they have the right to stop providing chips they have never exercised this right. Two cases in point. One was a similar situation to Apple with Sony where Sony raised a dispute and stopped selling paying Qualcomm. Qualcomm continued to supply chips to Sony while they worked out the issues. Second, since Apple’s suppliers have stopped paying Qualcomm, Qualcomm has continued to provide chips to Apple for devices like the iPhone 7 during the time that royalties are not being paid to Qualcomm. Here again, the FTC’s attempt to prove this as monopoly leverage fell short because there was not an example where Qualcomm actually enforced this issue and thus pulled the rug out from a customer.

Note, there is some confusion in the media over this point with comments from Apple’s COO Jeff Williams yesterday that they were interested in dual-sourcing Qualcomm and Intel in new model iPhones, but Qualcomm refused to provide them chips. This is specific to new devices going forward, where the point about Qualcomm not providing chips only matters about older devices. Qualcomm, in this case, choose to honor their agreement with Apple and Apple supplies for older models already in market but not provide chips for new models. So these are entirely different things.

Lastly, on the point of market harm. This I think is still the hardest thing for the FTC to prove and I say that for a few specific reasons. First, as a part of the FTC’s attempt to prove monopoly tactics via Qualcomm’s business model they tried to establish that Qualcomm had no competition in premium chipsets. Premium defined here as devices costing over $400 which is where Qualcomm’s royalty cap is set. An interesting tidbit came out today that before 2014 Qualcomm had 100% share of premium LTE/CDMA chipsets and that declined to 63% in 2016, and obviously quite a bit lower today given Apple is 100% Intel modems. What has to be established here, however, is that Qualcomm’s business model made it so that others could not compete in the premium tier. Proving lack of competition in the premium tier is the only angle since there are, and have been, many competitors not in the premium tier of modem chipsets.

From what I heard on Friday, I heard more evidence supporting that Qualcomm had superior technology than I did that their business model made it, so companies like MediaTek, Intel, HiSilicon, or Samsung were prohibited from making compelling solutions for the premium tier. This was made clear by Apple, LG, Motorola, and a few other suppliers when they essentially confirmed there was no compelling alternative to Qualcomm in premium and their decision to choose Qualcomm was based on the superior performance. My read on that was Qualcomm was the best option, not the only option. It seems reasonable that you could argue the possibility that said competitors simply weren’t good at making superior modems because that is Qualcomm’s bread and butter and that is the reason, not their tactics, their modems were not competitive.

Lastly, and I’m sorry this is long but this is a meaty issue, with the clear but pricing details which were revealed in the case, I’m more convinced that establishing Qualcomm’s pricing is outside of FRAND will be a challenge. While the FTC is focusing on discriminatory pricing, I’m not sure they can argue it isn’t far. Case in point, the price Apple has paid for access to Qualcomm’s patent portfolio, and their chipsets in their hardware is less than $20 broadly. Some specific clarity showed the price they pay for the modem, and Qualcomm’s is benchmarkable better than Intel’s, is $7.50 after incentive payments. Given how important cellular connectivity is to the device itself and the spur of innovation that has led to in software development, which has benefitted Apple as much as any company, it is hard to argue costs less than $20 and specifically less than $10 is not fair for the overall value added to the smartphone.

The next week or so will be Qualcomm presenting their defense. I’d love to have another summary, and I hope to attend the closing sessions as well to get some more informed perspective. While this analysis is not exhaustive, I tried to include the key nuances that stood out to me this far in the case — lots more to come.

Published by

Ben Bajarin

Ben Bajarin is a Principal Analyst and the head of primary research at Creative Strategies, Inc - An industry analysis, market intelligence and research firm located in Silicon Valley. His primary focus is consumer technology and market trend research and he is responsible for studying over 30 countries. Full Bio

Leave a Reply

Your email address will not be published. Required fields are marked *