FTC Case Against Qualcomm FTC Case Against Qualcomm Doesn't Hold Up to Reality
Kevin Krewell Jan., 21 2019
Last Tuesday morning I spent the day at the U.S Court House in San Jose, California to hear the FTC wrap up its case against Qualcomm. The day started with key FTC expert witness Professor Carl Shapiro. Shapiro is industrial organization economics professor from the Haas School of Business at the University of California, Berkeley, and had previously held an antitrust position in Justice Dept. He generated a report on Qualcomm market power (monopoly) in CDMA and premium LTE modems. His analysis focused on the time between 2006-2016. He claimed that three Qualcomm policies gave it the market power to demand “Super FRAND” royalties (FRAND stands for fair, reasonable, and non-discriminatory). A super-FRAND license implies that Qualcomm made too much money. Those policies were: a “No-license-no chips” policy, incentive payments to phone OEMs (specifically 2013 Apple agreement), and a refusal to license standard essential patents (SEPs) to rival chipmakers.
The no-license-no chip policy was designed to make sure that new handset OEMs take appropriate patent licenses before receiving chips to build phones. Without this policy OEMs could build smartphones with Qualcomm’s chips, then simply refuse to pay Qualcomm for all additional IP that applied to these phones. Qualcomm’s only recourse would be to sue OEMs for patent infringement, a costly process that could take years in each case. In the meantime, these OEMs could continue to sell phones and profit without paying Qualcomm’s IP license. As you might image, handset OEMs sometimes stall paying for the required licenses as long as possible. Delays sometimes happen when OEMs renegotiate expiring license agreements, however, Qualcomm claimed it continues to sell chips to OEMs as long as they continue to negotiate in good faith, even if their license agreement lapses. This would seem to refute the FTC’s argument that Qualcomm uses chip supply as leverage to extract higher royalties from OEMs. The only case that the FTC could dig up was a delay in supplying new sample chips to LG while LG had not yet signed for the new WCDMA license.
In the case of the 2013 Apple agreement, Apple was to receive a large incentive payment (~$640M) if it used Qualcomm modems exclusively. Evidence was presented that Intel (using the modem it bought from Infineon) attempted to win back Apple business starting with iPad (for data only modems). The FTC hypothesized that Apple’s loss of (some of) this incentive payment for breaking the exclusivity was a barrier to entry for Intel. The iPad opportunity had too little volume to compensate for the loss of incentive payments. However, Intel modems at that time were not capable of replacing Qualcomm modems in handsets, because they lacked key features and performance, and didn’t meet Apple’s requirements. But, then Apple sacrificed the Qualcomm payments in 2016 when it launched iPhones with Intel modems. So that begs the question: how big of a barrier did the agreement pose? It seemed to neither prevent Intel from investing the required capital to develop a product that met Apple’s requirements, nor did it dissuade Apple from using Intel modems when they were eventually ready. The economic test for Shapiro was, could Intel buy out the penalty payment and still make a profit. But the iPad design volume was too small and Intel’s modem was inferior, so is that a real test of a monopoly?
While professor Shapiro pressed the case that his analysis showed Qualcomm had enjoyed monopolistic power from 2006 to 2016, he failed so show that there was any quantifiable cost to consumers. Nor that there were any real competitive solution from other vendors. In addition, Professor Shapiro's own report showed that Qualcomm's share has declined before 2016, and continued to do so in 2017 and 2018, in both CDMA and premium LTE. Clearly Qualcomm’s market share was not due to monopoly power, but to lack of suitable alternatives. So that begs another question: why did the FTC take Qualcomm to court?
In a video deposition, Marv Blecker (a former Qualcomm executive) was interviewed and he recounted how the company rejected “exhaustive” licenses to competitors, in particular, he remembered VIA and Samsung asking for licenses. Qualcomm’s policy was to not provide exhaustive licenses to competitors (chip makers), because the royalty for all applicable patents was already accounted for and paid by handset OEMs. It’s hard to understand the FTC’s contention, that not licensing competing chipmakers was a barrier to entry, when they could already use Qualcomm’s technology IP for free (because handset OEMs paid the royalty). How would paying royalties to Qualcomm for cellular SEPs help Intel or Mediatek make reduce their costs and improve their margins?
Later in the day, Qualcomm called its first two witnesses: company co-founder Dr. Irwin Jacobs and Senior Vice President of 4G and 5G Engineering Durga Malladi. Jacobs gave a detailed history of the company and its original mission. He also told of the risks Qualcomm took to build the CDMA (code division multiple access) business – a digital 2G technology that many thought would be too complex and costly to ever work for consumer markets. The superiority of CDMA over frequency division multiple access (FDMA) and time division multiple access (TDMA) led to its adoption by a number of carriers, despite the initial limited number of chip suppliers (Qualcomm) of CDMA modems. With CDMA, more subscribers could be supported in the same bandwidth. With 3G, evolutions of CDMA became the basis for all networks globally as the widely-deployed European standard, WCDMA (Wideband-CDMA), was built on CDMA.
The cross examination of Irwin did focus on a specific case where Qualcomm supposedly withheld development chips from LG for a time while they were still negotiating a patent license for a new 3G WCDMA standard. The issue was eventually resolved, and Qualcomm never withheld commercial chips from LG. This was also a digression for the FTC, as the example involved WCDMA chips, which appears outside the scope of the FTC’s case involving CDMA and premium LTE modems.
Durga Malladi, SVP & GM of 4G/5G at Qualcomm, has 400 US Patents to his name; and in his testimony he stressed how the company is a leader not only in the market, but also in standards, and develops from the perspective of holistic end-to-end system solutions.
Based on today's testimony, I don't see that the FTC has a compelling case. Especially since Qualcomm’s market share declined significantly when Mediatek entered the CDMA market and Apple migrated to Intel modems. Whatever advantage Qualcomm had by being first to market with new technologies has been repeatedly mitigated over time by competitors such as Intel, MediaTek, Samsung, and HiSilicon.
I will be attending future court dates as my schedule allows to see how the case proceeds. And remember, the burden of proof is on the FTC that Qualcomm stifled competition and caused harm to consumers (through higher prices).
Kevin Krewell
Principal Analyst, TIRIAS Research
Twitter: @Krewell
The author and members of the TIRIAS Research staff do not hold equity positions in any of the companies mentioned. TIRIAS Research tracks and consults for companies throughout the electronics ecosystem from semiconductors to systems and sensors to the cloud.
https://www.forbes.com/sites/tiriasresearch/2019/01/21/ftc-case-against-qualcomm-doesnt-hold-up-to-reality/#74bfd8284e33