[GuestPost] Opinion: Skirting FRAND requirements under the guise of promoting innovation and efficiency (Part II)

Merpel wonders where we are headed on 
the FRAND licensing level debate, and 
who is in the driver's seat? 

In courtrooms across the globe, arguments continue to rage as to the extent of an SEP owner's FRAND undertaking.  In exchange for getting their technology incorporated into a standard (meaning that, if essential to the standard, that technology has to be used by users of the standard), SEP owners have to give an undertaking - known as a FRAND undertaking.  This undertaking obliges SEP owners to be prepared to licence their patents on Fair, Reasonable and Non-Discriminatory (FRAND) terms. Otherwise, unlike normal patents where competitors do not have to use the technology, SEP owners are in a position which could allow them to extract extremely high and possibly anti-competitive royalties from their competitors or stop them from participating in the standard and market completely (which were vices the European Commission wished to be addressed with the ETSI IPR Policy).  Thus, the FRAND undertaking is a safeguard that seeks to balance users' interests with SEP owners' interests in protecting their IP. But the courtroom debates in the US, Germany, UK and China have raised numerous unanswered questions about what this means.  How wide or narrow is this FRAND undertaking?  To whom is the FRAND undertaking owed? What does FRAND even mean?  The second question was subject to the CJEU referral in the Nokia v Daimler (see previous posts here), but which has so far remained unanswered in Europe.  In the second of a two-parter opinion piece, three US patent and anti-trust litigators in the form of Mark Selwyn, Tim Syrett and Alix Pisani of WilmerHale (who have acted in some of these cases) discuss their view of what is going on and where the answer might, and should lie.  

Over to Mark, Tim and Alix:

"Part 1 of this two-part series showed how standard-essential patent (SEP) owners that now advocate against licensing component suppliers made the decision to invest in standard setting when they understood FRAND commitments to require licensing everyone and developed their large SEP portfolios with such an understanding in mind. Part 2 first explains how those SEP holders’ second argument against licensing component suppliers—that licensing at the end user level is necessary to promote efficiency—also fails to withstand examination and, instead, reflects an intent to capture unreasonable royalties based on the higher price of end devices than components. 

SEP Litigation Demonstrates Licensing Component Suppliers Would Increase Efficiency 

Some SEP owners have argued against licensing component suppliers under the pretext that it is more efficient to license end users. For example, in Qualcomm’s 2019 Ninth Circuit opening brief in its litigation against the Federal Trade Commission, Qualcomm contended its “OEM licenses are efficient[.]”1  Similarly, in an amicus brief in that case, Nokia argued that “[t]here are good reasons for SEP owners to structure their licensing programs to license end-user products,” including that it “reduces transaction costs and complexities associated with negotiating and executing licenses at multiple points in the supply chain” and that it “avoids overlapping and duplicative licensing.”2 Likewise, in a case currently on appeal in the Fifth Circuit that Continental Automotive Systems brought against Nokia and other defendants challenging their refusal to license component suppliers 3, a Nokia Technologies licensing manager explained that “in Nokia Technologies’ experience, the most efficient way to provide access to our patented technology for any value-chain is through one license agreement covering all SEPs relevant for such value chain as opposed to multiple license agreements with multiple companies in the same value chain.”4

But a review of litigations where SEPs are asserted against end-product suppliers demonstrates that negotiating with component suppliers would not decrease efficiency. To the contrary—licensing component suppliers would increase efficiency as they are frequently key to evaluating the merits of SEP assertions because they created the components on which the infringement allegations center. As the examples below demonstrate, end-product suppliers often lack knowledge about how standardized technology is incorporated into their products through components and thus it falls to the component suppliers to provide critical technical information necessary to evaluate issues such as infringement:

• In Saint Lawrence Communications LLC v. Motorola Mobility LLC, the court denied a motion to enhance damages for willfulness because, among other factors, “the accused products were alleged to infringe because of chips made by Qualcomm, which are used industry-wide, rather than technology created by Motorola. Indeed, Motorola did not even have access to the infringing source code contained on the Qualcomm chips, which militates against a finding that Motorola deliberately copied SLC’s technology.” 5

• In Core Wireless Licensing S.A.R.L. v. LG Electronics, Inc, the court observed in a case where Core Wireless asserted cellular SEPs obtained from Nokia that “[i]t is undisputed that the baseband chips referenced in Core’s infringement contentions are Qualcomm baseband chips. Accordingly, this case turns in large measure on the function of the Qualcomm baseband chips. Indeed, Core has stated that it ‘expects to rely on Qualcomm source code for most, if not all, of the thirteen standard-essential patents asserted against LG.’ 6

• When InterDigital asserted SEPs claimed essential to the LTE standard against ZTE, one of InterDigital’s experts described his task to the jury as follows: “I was asked to look at how the ZTE LTE phones process a certain message called a PDCCH message. I found that in the ZTE phones this processing was done by a Qualcomm chip. So I looked at the hardware and software source code for this Qualcomm chip.”7

• In GPNE Corp. v. Apple, Inc., GPNE alleged Apple infringed patents that it claimed were essential to the GPRS, EDGE, and/or LTE cellular standards. GPNE conceded, however, that the relevant functionality was entirely within the baseband processors Apple purchases: “Neither party contests that the patent’s contribution to the art is a signaling technique performed by the baseband processor.” 8  GPNE’s technical expert conceded that “each of the GPRS and LTE signals identified in [his] report are processed by the . . . base band processors in Apple products” and offered the same infringement theory for various accused Apple products because they all contained the same baseband processor chipsets.9  GPNE further confirmed that “the baseband processor ... enables the cellular functionality in Apple’s products.” 10  Indeed, the court held “as a matter of law that in this case, the baseband processor is the proper smallest salable patent-practicing unit” for the purpose of determining a royalty.11

• Moreover, in motions to transfer, the location of chip suppliers has received heavy weight because of the importance of chipsets to infringement analysis. 12

Similarly, in litigation where cellular network equipment has been accused of infringing SEPs, providers of network equipment have played a similar role. When Huawei asserted claimed cellular SEPs against T-Mobile, both Ericsson and Nokia argued that they should be allowed to intervene in the case because T-Mobile is a customer. Nokia contended that its intervention would “streamline and simplify” the litigation because Nokia had “[t]he technical know-how.”13  Nokia elaborated that it “also possesses a broad and deep understanding of its own technology. Accordingly, it is the most important source of information about its products and keeper of potentially key, relevant documents and the sole keeper of its source code. It would be unfair to force T-Mobile to litigate the issues that [Nokia] best understands.”14  Similarly, Ericsson argued that “documents and other tangible evidence, including source code and product specifications, likely reside with Ericsson. The technical know-how and testimony for this equipment also likely resides with Ericsson.” Echoing Nokia, Ericsson also contended that “[i]t would be unfair to force T-Mobile to litigate the issues that Ericsson best understands.” 15

As the above examples demonstrate, component suppliers bring specialized knowledge to litigation that uniquely positions them to provide information relevant to whether SEPs are infringed. They would bring the same unique insights to SEP negotiations as the parties that designed the components that provide standardized functionality. Component suppliers are also best positioned to understand whether SEPs are used and the value of any such usage, how much of the component the accused functionality represents, and what alternatives technologies are viable. In the case of cellular baseband chips, the limited number of suppliers means that they could potentially decrease the number of negotiations by obtaining licenses that would protect their customers under principles of patent exhaustion.

Financial Incentives Are The Real Driver Of SEP Owners’ Arguments

As discussed in Part 1 and above, SEP owners in the past acknowledged (and at times advocated) that a FRAND commitment requires licensing component suppliers, invested in their SEP portfolios with that understanding, and utilized the technical expertise, documents, and testimony of component suppliers in litigation. Despite all this, some SEP owners today argue against licensing to component suppliers.

The real reason for these SEP owners’ current arguments is clear: licensing at the end product level drives up royalties. In fact, as Judge Koh highlighted in her 2019 Qualcomm opinion, SEP owners have admitted as much. Judge Koh quoted the following statements made by Qualcomm employees to the IRS:

“But [] having to choose between one or the other then you’re right, obviously the handset is humongously more ... lucrative for a bunch of – a bunch of reasons.” 17

“Yeah, but if I would average royalty on all the handsets that we collect royalties on – I don’t remember what it is anymore, I used to know the number – but if – if it were ten dollars, for example, you couldn’t charge a ten-dollar royalty on a chipset that cost five dollars, or six dollars, or seven dollars.”18

That this preference for using the end device is likely to pay off in inflated royalties is further supported by studies demonstrating that “anchoring” demands to a larger royalty base inflates royalties, especially for patented features included in complex, multi-component products. When judges and jury members are exposed to revenues of a finished product—which is easy to understand—it acts as an “anchor” value from which further calculations are made. This often causes judges and jury members to find large royalties are reasonable because they are a small percentage of a large base. 19

Therefore, while SEP holders claim that licensing at the end device level promotes efficiency, the opposite is true. Determining royalties based on end devices results in outsized royalties that are not apportioned to the value of SEPs. As the Federal Circuit has explained, apportionment is critical to ensure that a reasonable royalty “does not overreach and encompass components not covered by the patent.”20  The danger of such royalties is that they undermine the incentives for innovation by misallocating rewards. SEP holders receive windfalls and are incentivized to overinvest in SEPs, while the non-standardized innovation in the end devices that is being taxed by SEPs suffers. Accordingly, royalties based on the price of a device are only allowed in the rare circumstance that “the demand for the entire product is attributable to the patented feature.”21 Such is rarely, if ever, the case when valuing a patent that reads on a complex, multi-component device that incorporates many other patented and non-patented features. “[T]he smallest salable unit approach was intended to produce a royalty base much more closely tied to the claimed invention than the entire market value of the accused products.” 22  Licensing at the component level addresses these concerns by putting the focus on the smallest salable unit that practices the standardized technology. Doing so ensures that SEP holders are compensated for the value of their SEPs and product suppliers are not disincentivized from bringing innovative products to market by paying outsized royalties that tax innovations that are separate and apart from the standard.

Overall, the evidence described above and in Part 1 demonstrates that discrimination by certain SEP owners against licensing component suppliers allows them to maximize their licensing returns by targeting more expensive products. SEP licensors’ arguments that licensing at the OEM level is consistent with their FRAND commitments and is necessary for innovation are unavailing. Historically, a FRAND commitment has included licensing at the component level, innovation for future standards does not depend on maximizing SEP licensing revenue, and it would be more—not less—efficient to license at the component level. Moreover, licensing at the component level aligns with the most appropriate methodology for valuing SEPs that claim features of complex, multi-component products.

________________________

1 Opening Brief for Appellant Qualcomm Inc. at 8, Fed. Trade Comm’n v. Qualcomm Inc., No. 19-16122 (9th Cir. Aug. 23, 2019).

2 Brief of Nokia Technologies Oy as Amicus Curiae in Support of Neither Party at 18, Fed. Trade Comm’n v. Qualcomm Inc., No. 19-16122 (9th Cir. Aug. 30, 2019).

3 Continental Automotive Sys, Inc. v. Avanci, L.L.C., et al., 20-11032.

4.  Second Declaration of Lasse Holopainen, Continental Automotive Sys, Inc. v. Avanci, L.L.C., et al., No. 19-cv-02933-M (N.D. Tex. Aug. 30, 2019), ECF 162-2. 

5 No. 2:15-CV-351-JRG, 2017 WL 6268735, at *2 (E.D. Tex. Dec. 8, 2017).

6 No. 2:14-CV-911-JRG-RSP, 2016 WL 4596108, at *1 (E.D. Tex. Sept. 3, 2016).

7 Trial Transcript at 495:7-16, InterDigital Commc’ns, Inc. v. ZTE Corp., No. 13-000009-RGA (D. Del. Oct. 21, 2014).

8 GPNE Corp. v. Apple, Inc., No. 12-CV-02885-LHK, 2014 WL 1494247, at *13 (N.D. Cal. Apr. 16, 2014).

9 Id. 

10 Id. (quoting GPNE Daubert motion).

11 Id. 

12 E.g., Adaptix, Inc. v. Apple Inc., No. 6:12-CV-124, 2013 WL 12137842, at *3-4 (E.D. Tex. Mar. 28, 2013). 

13 Movant-Intervenors Nokia Solutions and Networks US and Nokia Solutions and Networks Oy’s Partially Unopposed Motion for Leave to Intervene at 4, Huawei v. T-Mobile, No. 2:16-cv-52 (E.D. Tex. June 10, 2016) (“As a result, NSN US’s entry into the case should streamline and simplify this litigation, especially as it relates to the following network equipment that Plaintiff alleges to infringe in its complaint.”). 

14 Id. at 11. 

15  Huawei v. T-Mobile, No. 2:16-cv-52, Dkt. No. 95 at 3, 10 (E.D. Tex. Sept. 16, 2016) (citing Kolakowski Decl. ¶ 15).

16  In the third quarter of 2020, just three companies (Qualcomm, MediaTek, and HiSilicon) accounted for 81% of cellular baseband revenue.  Sravan Kundojjala, Strategy Analytics, 5G Supercharges the Baseband Industry in Q3 2020 (Feb. 2, 2021), https://www.strategyanalytics.com/strategy-analytics/blogs/components/handset-components/handset-components/2021/02/02/5g-supercharges-the-baseband-industry-in-q3-2020.

17  Fed. Trade Comm’n v. Qualcomm Inc., 411 F. Supp. 3d 658, 754 (N.D. Cal. 2019), rev’d and vacated, 969 F.3d 974 (9th Cir. 2020).

18 Id. at 757.

19 See Kattan, The Next FRAND Battle: Why the Royalty Base Matters, CPI Antirust Chronicle (2015), https://www.gibsondunn.com/wp-content/uploads/documents/publications/Kattan-Why-the-Royalty-Base-Matters-CPI-03.2015.pdf (summarizing studies).

20 VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326 (Fed. Cir. 2014) (quoting LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 70 (Fed. Cir. 2012)).  

21 LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 67-68 (Fed. Cir. 2012) (“We reaffirm that in any case involving multi-component products, patentees may not calculate damages based on sales of the entire product, as opposed to the smallest salable patent-practicing unit, without showing that the demand for the entire product is attributable to the patented feature.”).

22 VirnetX, 767 F.3d at 1327.