Abuse of dominance and exclusive deals: Qualcomm’s chipsets are down
Kluwer Competition Law Blog
February 19, 2018
Please refer to this post as:, ‘Abuse of dominance and exclusive deals: Qualcomm’s chipsets are down’, Kluwer Competition Law Blog, February 19 2018, http://competitionlawblog.kluwercompetitionlaw.com/2018/02/19/abuse-dominance-exclusive-deals-qualcomms-chipsets/
On January 24, 2017, the European Commission announced that it had fined Qualcomm EUR 997,439,000. The Commission thinks that Qualcomm had abused its dominant position to become Apple’s sole supplier of long term evolution (“LTE“) baseband chipsets. The abuse was allegedly committed through exclusive deals, which lasted from 2011 to 2016.
Baseband chipsets are key components in smartphones and tablets, which enable connection with wireless voice and data networks. LTE (also known as 4G) is a global technical standard for high-speed mobile communication.
The official press release provides some detail on the Commission’s findings. It also raises a few questions, which should be addressed when the decision is published.
The facts according to the press release
Between 2011 and 2016, Qualcomm’s share of the LTE baseband chipsets market was often above 90%. Barriers to entry in the market were high, because baseband processor technology requires costly R&D and Qualcomm held valuable patents. Hence the finding of Qualcomm’s dominance.
Apple was an important customer, worth about one third of the market. In 2011, Apple started sourcing LTE baseband chipsets from Qualcomm. In 2013, the exclusive deal was extended to 2016. The parties agreed that Qualcomm would pay substantial rebates to Apple, provided that Apple did not use rival chipsets in iPads and iPhones. If Apple were to launch a device with rival chipsets or switch supplier, Qualcomm would stop payments and recoup most of the sums already paid.
In July 2015, the Commission started investigating the compliance of those exclusive deals with competition law. The Commission issued a statement of objections against Qualcomm in December 2015 (together with another statement of objections concerning predatory pricing for UMTS (3G) chipsets, about which there are no news). Apple finally switched part of its purchases to Intel in 2016.
The exclusive deals in the Commission’s view
In the Commission’s view, the switching costs associated with exclusivity were so high, that Apple would not deal with Qualcomm’s competitors, regardless of the quality of their chipsets. Moreover, other suppliers needed Apple in their client portfolio to win over more customers, because Apple’s premium brand devices could influence product design and procurement patterns in the industry. Thus, Qualcomm locking in Apple also foreclosed rivals from the market and stifled innovation.
To prove Apple’s lock-in and the ensuing market foreclosure, the Commission relied on several pieces of qualitative and quantitative evidence. In particular, internal documents proving that Apple had often explored switching to Intel during the contract period, but was stopped by the threat of retroactively losing rebates. The Commission also rejected a price-cost test presented by Qualcomm. The test was aimed at demonstrating that exclusivity was not anti-competitive, because competitors could beat Qualcomm’s prices.
Internal documents vs. price-cost test
In public statements made on January 24, 2017, Commissioner Vestager emphasized that Qualcomm is the first decision on exclusivity deals and abuse of dominance since the Intel judgment (Case C-413/14).
In Intel, the European Court of Justice highlighted the importance of price-cost tests, among other admissible evidence, for analyzing the “intrinsic capacity” of a practice to foreclose competitors which are at least as efficient as the dominant company, under Article 102 TFEU. Commissioner Vestager said that the Qualcomm decision complies with the Intel ruling, because it relies on a strong blend of evidence to prove foreclosure. This point will likely be subject to intense scrutiny when the final decision comes out. Much of the debate will probably revolve around the weight that should be given to Apple’s internal documents versus Qualcomm’s price-cost test.
Under one possible scenario, Apple’s calculations may show that the best offers from alternative suppliers could not beat Qualcomm’s prices, net of rebate payments and recoupment of previous rebates. This could amount to a real-life price-cost test, assuming that:
- Intel and other high-tech giants supplying chipsets are as efficient as Qualcomm, in light of their financial strength, technology and resources;
- their offers to Apple were close to, if not below, actual costs, given their likely willingness to forego substantial margins in order to win Apple and thus unlock the way to other key customers.
In this scenario, Apple’s internal papers could be even more relevant than the results of a theoretical price-cost model.
In other scenarios, a price-cost test may still be necessary to guarantee a more robust assessment of exclusionary effects. This would call for a closer scrutiny of the reasons why the Commission rejected the results of Qualcomm’s analysis.
Competition for the market?
The effects of exclusive deals should be assessed in light of their duration and breadth, and of the competitive dynamics in the interested markets. In this respect, some facts mentioned in the Commission’s press release deserve closer attention.
The dynamics of the LTE baseband chipsets market
Firstly, after five years of alleged abuse, in 2016 Apple actually switched at least part of its supplies to Intel. This could suggest that the market for LTE baseband chipsets is not so static despite Qualcomm’s high market shares. While many companies have recently left the market (e.g., Texas Instruments, Broadcom and ST-Ericsson), many of the remaining players have deep financial and technological resources (Intel, Mediatek, Spreadtrum and Marvell). Some, like Intel, are also dominant in neighboring markets (e.g., the x86 computer processing units).
On the demand side, customers like Apple, Samsung, Huawei, Xiaomi, and other producers of mobile devices, do not lack buying power. They also have the resources to develop in-house baseband processor technology. Over time, many of them have been rumored to be doing so. More generally, the natural evolution of global communications standards should tend to shake up the markets. It could also endanger existing supply relationships and thus create more opportunities to compete.
The duration of the exclusive deals
In an apparently dynamic context, Apple’s switch to Intel could suggest that there was healthy competition in the market. Qualcomm’s competitors could win long-term supply contracts with big sophisticated customers. Thus, they could continue operating despite long exclusive deals. If this were the case, one could think that the duration of Apple’s exclusive agreement with Qualcomm was not excessive. Qualcomm’s competitors could not exploit the first window of opportunity in 2013 to capture Apple. They exploited the second in 2016.
Another more simple scenario is that Qualcomm would have gone on tying Apple, had the Commission not sent its statement of objections. In this case, the Commission’s intervention was probably warranted.
The scope of the foreclosed market and the role of IP rights
Other doubts concern the scope of the foreclosed market. The press release is unclear as to how Apple’s purchasing strategy would foreclose Qualcomm’s rivals from the market. Apple only accounted for one third of the market, so the other suppliers could compete for the remaining share. In addition, Apple’s competitors should have been happy to accept better offers from Qualcomm’s competitors, regardless of Apple’s choices. The final decision should then explain how Apple’s brand leadership influenced the whole industry.
Finally, the Commission’s press release does not say much on the extent of Qualcomm’s intellectual property rights. Were they instrumental in Qualcomm’s strategy to retain customers and exclude competitors? Was there a problem with any FRAND terms provided by Qualcomm on SEP patents? It will be interesting to see the answers in the final decision. For now it seems that the press release is telling too simple a story.