It has been two months since closing arguments were presented in the FTC versus Qualcomm Trial, and Judge Koh could issue a verdict at any moment. In these past few months, I have been analyzing what a ruling might mean. I’m not a lawyer, but have seen a lot of antitrust action and was considered a “key witness” in the AMD and Intel spat, and in my industry analyst role, I watch key decisions carefully.
During my process of reviewing this FTC case and trial, I couldn't help but notice just how bizarre it is in totality. As I said here, the case started as a clown show, and the trial didn't disappoint, as it continued to be a clown show until the end. While I'm not a big top 10 list guy, I wanted to share with you my Top 10 FTC Trial oddities that stuck out to me.
1/ Earlier global rulings FTC cited have been overturned, weakened or resolved
In its filing, the FTC cited global agencies that have issued antitrust rulings against Qualcomm. Where are those decisions now?
- Korea: After 10 (yes ten) years of investigations and on-going appeals, South Korea's Supreme Court ruled that the Korea Fair Trade Commission’s (KFTC) fine against Qualcomm was incorrect. I couldn’t help but notice Roger Kay’s article about Korea and the scandals surrounding individuals in the vicinity of the KFTC Qualcomm case.
- Japan: After 9 (yes nine) years of thorough investigation, Japan revoked its cease and desist order and deemed Qualcomm's licensing program "lawful."
- Taiwan: The Taiwan Fair Trade Commission (TFTC) reduced Qualcomm's $740M fine to the $93M the company had already paid. Qualcomm also agreed to collaborate with the Taiwanese ecosystem on 5G and to invest $700M in the country. To me, this was Taiwan's way to show it had made a mistake but had to save face by making Qualcomm spend $800M.
- China: China’s NDRC fined Qualcomm $975M, but accepted the company’s rectification plan for addressing its concerns; it affirmed the value of Qualcomm’s IP portfolio, confirmed the licensing model as lawful, and blessed the reduced Chinese SEP licensing rates as fair. Subsequently, most Chinese handset OEMs inked licensing deals with Qualcomm.
- EU: In a 10-year-old case unrelated to its licensing business, the EC fined Qualcomm $1.2B for alleged anti-competitive pricing and blamed it for the failure of European chipset vendor Icera (even though Icera was bought by NVIDIA, and later shut down because its architecture of software-upgradable modems wasn’t competitive); as you would expect, Qualcomm appealed the decision.
From my vantage point, the premise of the FTC citing that what is good for other countries is also good for the U.S. is very rapidly losing its teeth. Most importantly, these bodies are afirming Qualcomm's licensing practices.
2/ Lots of 4G and 5G modem competition
One of the key points the FTC levied against Qualcomm was that lack of 4G modem competition resulted from alleged anti-competitive actions of Qualcomm. The slight problem with this point, and what came out in testimonies, was that Apple decreased its percentage of Qualcomm modems even during the arbitrary period defined by the FTC, and now uses 100% Intel modems in its new devices; also, both Huawei and Samsung increased the percentage of homegrown modems in their phones from 0 to 50%.
So if the #1, #2, and #3 handset makers decreased their purchase of Qualcomm modems to less than 50% (and Qualcomm’s overall modem market share declined significantly), how could the FTC’s allegation be true? 11 different companies testified for the FTC, but what struck me most was Apple's. What I did hear from Apple testimony and evidence, was that competing 4G solutions were late or slow and that Apple didn’t adopt them until they could meet its requirements.
But how about 5G, as this is what the FTC is worried about, and what Judge Koh will be ruling on? As I personally walked the halls of the Mobile World Congress conference in Barcelona, I saw evidence of 5G modem development and announced modems. That’s five 5G modems from Intel, Huawei (HiSilicon), Samsung, MediaTek, and UNISoC. Moor Insights & Strategy analyst Anshel Sag documented many of these modems here.
While announcements don't equal similar performance, compatibility, and announcement dates, there’s a heck of a lot more competition in wireless modems than there is in X86 processors (three vendors, two in practice), PC GPUs (three vendors, two with discrete), PC operating systems (three vendors), smartphone operating systems (two vendors) and smartphone app stores (three vendors). Seriously, who is the FTC kidding here?
3/ No smartphone handset maker harm
To demonstrate Qualcomm’s abuse of its monopoly power, the FTC had to provide evidence that harm had occurred, not just a theory of what could happen. So what about the harm to handset makers? Apple, which I believe was the primary instigator of the FTC action (based on its common interest agreement with the FTC-- an unusual arrangement between a government agency and a company), has an astounding 90% of the industry profits during the period under scrutiny. The FTC provided no evidence that Qualcomm’s licensing practices had harmed Apple (obviously). Similarly, the FTC didn’t show any evidence of harm to Huawei, Samsung, or any other handset vendor.
I also heard no handset maker under oath talking about how it was harmed by Qualcomm’s licensing practices, and I'm not seeing any harm today. What I heard and am seeing today, is that Qualcomm enabled trillions of dollars in other companies’ valuations and a large ecosystem of diverse smartphone vendors.
4/ No ecosystem and, by extension, consumer harm
While not central to its case, did the FTC or its expert witnesses demonstrate broader ecosystem (and consumer) harm? FTC expert Carl Shapiro theorized but presented no evidence that prices increased or that innovation was hampered. The ideas that prices didn’t decrease as much as they could have, or innovation didn’t flourish as much as it might have, shouldn’t have been justification for an antitrust case.
Think back ten years ago, of how much you paid per megabit of data to AT&T. Likely a 10X lower price per magabit, right? Moreover, think of how the price of smartphones with equal capabilities has declined: a phone that you could buy for $1,000 five years ago might cost around $200 today. It turned out in trial testimony, that I believe much of Apple's beef with Qualcomm was over ~$7.50 in license fees. If you ask me, ~$7.50 seems like a bargain for the vast amount of technology and on-going design required to make all cellular systems work, so that your little smartphone can connect outside your home. Ya think? Apple says its beef is with "no license, no chips" is the issue, which I address below, and other jurisdictions having issues, which I address in #1.
5/ Zero testimony for "no license, no chips" leverage
The FTC claimed that Qualcomm’s "no license, no chips" policy gave it leverage to charge excessive royalties because it could withhold chips. I expected the FTC to roll out an array of testimony citing examples of Qualcomm withholding 4G LTE production chips when smartphone vendors didn’t renew their licenses, and vendors impacted by reduced sales and profits. It never happened. Literally, not one single time. Not one single witness testified that Qualcomm withheld modems for commercial phones from a vendor renegotiating a license. There were some emails suggesting it, but no one said it actually happened.
This lack of real-world examples despite hundreds of patent negotiations underscored a flaw in Professor Shapiro’s theory. He assumed that only one side had bargaining leverage in negotiations (e.g. Qualcomm, with patents to license, and chips to sell) but didn’t account for the leverage of the other party (e.g. OEM with money to acquire a patent license, chips, patents to cross-license). In the real world, both sides have varying degrees of leverage, as evidenced by the Apple example- Apple’s incredible buying power explains how it managed to negotiate discounts, exclusive deals, and special terms, as it always did (and I believe continues to do) with its suppliers.
Qualcomm needed to sell its chips as much as smartphone vendors needed its patents- with the notable difference being, that OEMs could still sell phones and make money for years without licensing Qualcomm patents (exhibit A, Apple), whereas Qualcomm couldn’t make money if it didn’t sell chips to OEMs that refused to license its patents.
Apple reiterates it intends to pay a fair price for any chips purchased from Qualcomm. I'm interested to see the San Diego Trial evidence as it will lay out exactly what Qualcomm offered and what Apple countered with and negotiations back and forth. We will all get to see what "fair" means which will be so interesting as a court ruled that three of Qualcomm's 130,000 had $1.40 in value. In FRAND law, companies who are deemed to negotiate in bad faith are considered "unwilling licensees" and can be charged above FRAND prices.
6/ FTC's case based on theory without hard facts
As I said above, the FTC never actually presented any empirical evidence that harm was done to competing chipmakers, handset vendors, the ecosystem, or consumers. I have followed many antitrust cases, and I am used to seeing economists showing hard historical data to show that when competition left the market, prices would rise or innovation stalled, and when competition entered, prices declined and innovation soared. The FTC showed no quantification, actual data, or causality... which blew my mind.
The FTC did provide theories of what "could" or "might" happen; it’s almost like the FTC invented the Pre-Crimes unit in the Minority Report movie. What also blew my mind, was that the FTC’s lead economist, Carl Shapiro, was the same person criticized by Judge Leon in the TWC-AT&T merger trial for once again making overly theoretical arguments based on the same flawed theory.
7/ Huawei as FTC “star witness”
The first FTC witness called to testify against Qualcomm was Huawei, which, by the way, would benefit greatly if Qualcomm were weakened by a negative ruling. Bloomberg’s Ian King and Kartikay Mehrotra went so far as to call Huawei the “US's star witness” in their story, “The U.S.’s Star Witness in the Qualcomm Antitrust Suit: China’s Huawei.” While I might debate that Apple was really the star witness, the fact that Huawei could even be considered a key witness for a U.S. Government agency was shocking. As in any trial, the motivations of witnesses need to be weighed against their testimony. I am not saying that Huawei wasn’t honest in its testimony; I’m just pointing out that the company has much to gain and can move into the 5G driver’s seat if Qualcomm is weakened.
Also, considering all of the consternation in Washington, D.C. over Huawei, why on earth did the FTC lead with Huawei? Just last week I attended the LeadershIP Forum in Washington, D.C., right across from the FTC Building. At the event, DHS Cybersecurity and Infrastructure Security Agency (CISA) Assistant Director Jeanette Manfra gave chilling responses to questions on the national security risks of having 5G led by "countries and companies that pose a threat to the nation". She never used the words "China" or “Huawei”, but she didn't have to. Everyone in the audience knew what she was talking about. The FTC’s choice to lead witness testimony with Huawei given the current environment was truly bizarre.
8/ FTC case initiated under cover of darkness during a presidential transition
As I wrote here, it looked like the FTC case was filed under questionable circumstances and timing. The FTC’s filing was:
- a “midnight filing” right after the chairwoman announced her resignation
- under the nose of the new incoming Trump administration
- with only three of five commissioners at the post
- with a non-unanimous decision and a very strong dissenting opinion from the interim chairwoman. You can find the dissenting statement here. It's ugly.
Could the suit have been filed under any shadier circumstances?
9/ FTC weakening Qualcomm will actually reduce competition
Perhaps the FTC thought it was protecting competition by filing this suit, but it could ironically end up harming it through its actions. Qualcomm is not only the inventor of key technology underpinnings of 5G, but I believe it is also the U.S.’s only hope for 5G leadership, and also the main technology provider and funder for the entire Android smartphone ecosystem. What do you think would happen if the Judge in the FTC trial issued a ruling that limited Qualcomm’s ability to continue investing in 5G R&D and in the development of Android smartphone platforms?
It’s easy for me to imagine that most Android phones wouldn’t be as competitive with Apple iPhones as they are today, and we likely would see reduced competition and innovation within the Android ecosystem, and across the smartphone industry as a whole. Samsung and Huawei have deep pockets and would probably be OK, as would Apple, but other Android brands like Sony, LG, Motorola, Lenovo, Oppo, OnePlus, HTC, ZTE, Nokia, and Xiaomi, to name the major ones, would likely suffer. Hence, reducing competition.
10/ Trump Administration duality
The White House was clear about Qualcomm's value to the country when it called it a "national treasure," one too important to be purchased and stripped by formerly Singapore-based Broadcom Limited. The notion was that the technology Qualcomm provided was critical to national security and needed to be developed in the U.S.
Since that order, President Trump has declared that he wants the U.S. to be first in 5G rollouts and even 6G. So my question is, how does the U.S. lead (and continue to lead) in 5G if the FTC strips Qualcomm of its ability to do advanced R&D ahead of the industry, or lead the ecosystem? Who can fill the gap?
Don't be confused about the difference between fundamental R&D and commercial product development-- they are very different. Huawei and Qualcomm are the two biggest investors in wireless research by a large margin. So what happens if the FTC weakens Qualcomm? I strongly believe Huawei and China will define the future of 5G and 6G. If the U.S. is to maintain its leadership in wireless and have some control over the evolution of technology that underpins its national security, the White House, other government agencies, and the FTC urgently should get on the same page about 5G and what is best for the country... as their actions appear to lead to different outcomes.
Wrapping up
I have followed many antitrust cases and was considered a "key witness" in the spats between AMD, Intel, and global regulatory bodies. This doesn't make me a lawyer- I am not, but it does give me the experience to know what to look for, the impacts and questions to ask. I am one of a handful of analysts that have worked for device makers and chipmakers, I believe, giving me a unique, empathetic view, understanding both sides of the argument.
The FTC's case against Qualcomm was and is still bizarre to me. I believe it was initiated under questionable circumstances, is based on a flawed theory, isn’t substantiated with empirical evidence of harm, benefits massive and powerful corporate instigators, could paradoxically reduce competition, and may lead to serious national security issues.
Note: Moor Insights & Strategy writers and editors may have contributed to this article.
Disclosure: Moor Insights & Strategy, like all research and analyst firms, provides or has provided paid research, analysis, advising, or consulting to many high-tech companies in the industry, including Advanced Micro Devices, Apstra, ARM Holdings, Bitfusion, Cisco Systems, Dell EMC, Diablo Technologies, Echelon, Ericcson, Frame, Gen Z Consortium, Glue Networks, GlobalFoundries, Google (Nest), HP Inc. HewlettPackard Enterprise, Huawei Technologies, IBM, Jabil Circuit, Intel, Interdigital, Konica Minolta, Lenovo, Linux Foundation, MACOM (Applied Micro), MapBox, Mavenir, Mesosphere, Microsoft, National Instruments, NOKIA (Alcatel Lucent), Nortek, NVIDIA, ONUG, OpenStack Foundation, Peraso, Portworx, Protequus, Pure Storage, Qualcomm, Rackspace, Rambus, Red Hat, Samsung Technologies, Silver Peak, SONY, Springpath, Sprint, Stratus Technologies, TensTorrent, Tobii Technology, Synaptics, Verizon Communications, Vidyo, Wellsmith, Xilinx, Zebra, which may be cited in this article.
