“An accurate depiction of the debate over the appropriate royalty base for cellular SEPs must account for the many reasons that ordinarily make using the SSPPU the most accurate and proper methodological approach.”
A recent series of five articles on IPWatchdog address various aspects of licensing cellular standard essential patents (SEPs) on fair, reasonable and non-discriminatory (FRAND) terms by examining statements from entities involved with licensing. The authors also provide their commentary on the statements and cite various authorities that they suggest are consistent or inconsistent with principles advocated in the statements. The articles lean heavily in favor of the positions of a few companies that derive significant revenue from SEP licensing. For this reason, they fail to present a balanced view. Indeed, to read the series, one might conclude that the major priority for SEP licensing should be to extract excessive revenues for SEP patent owners. Quite the contrary, a key priority should be applying FRAND safeguards against outsized, windfall profits resulting from abuse of SEPs to the detriment of innovative companies that engage in research and development and supply products to the marketplace. Those safeguards include applying well-established principles of patent law to SEPs, including when it comes to patent valuation and patent litigation, where a patent holder is rewarded with fair royalties that reflect the incremental value of any infringed and valid SEP.
This two-part series responds to certain issues addressed in those articles to fill the gaps on FRAND licensing and offer a perspective that seeks to vindicate the true meaning and role of FRAND. Part one starts by addressing the third article in the series. In that article, the authors address what licensing statements have said about the appropriate royalty base for licensing cellular SEPs on FRAND terms. The authors incorrectly paint a position that FRAND royalties based on the smallest saleable patent practicing unit (SSPPU) are out of step with judicial authority and industry practice. This could not be further from the truth. The authors overlook black letter patent authority, industry practice, and changes in technology that support using the SSPPU as the royalty base for cellular SEPs, just as when valuing any patent that reads on a component in a complex multi-component product.
Considerable Legal Authority Supports the Use of the SSPPU
For more than a century, U.S. patent law has required that a patent owner “must in every case give evidence tending to separate or apportion the defendant’s profits and the patentee’s damages between the patented feature and the unpatented features . . . or he must show . . . that the profits and damages are to be calculated on the whole machine, for the reason that the entire value of the whole machine, as a marketable article, is properly and legally attributable to the patented feature.” Garretson v. Clark, 111 U.S. 120, 121 (1884). More recently, the Federal Circuit has described that “[n]o matter what the form of the royalty, a patentee must take care to seek only those damages attributable to the infringing features.” VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326 (Fed. Cir. 2014). Apportionment is critical to ensure “that a reasonable royalty ‘does not overreach and encompass components not covered by the patent.’” Id. (quoting LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 70 (Fed. Cir. 2012). It is “particularly true” that a SEP holder “should only be compensated for the approximate incremental benefit derived from his invention,” and “not any value added by the standard’s adoption of the patented technology.” Ericsson, Inc. v. D-Link Sys., Inc., 773 F.3d 1201, 1232-33 (Fed. Cir. 2014).
The Federal Circuit has further recognized that the SSPPU is an appropriate starting point for a royalty base to achieve apportionment:
[T]he requirement that a patentee identify damages associated with the smallest salable patent-practicing unit is simply a step toward meeting the requirement of apportionment. Where the smallest salable unit is, in fact, a multi-component product containing several non-infringing features with no relation to the patented feature . . ., the patentee must do more to estimate what portion of the value of that product is attributable to the patented technology.
VirnetX, 767 F.3d at 1327; see also Fed. Trade Comm’n, The Evolving IP Marketplace Aligning Patent Notice and Remedies with Competition 25 (Mar. 2011) (“Courts should identify as the appropriate base that which the parties would have chosen in the hypothetical negotiation as best suited for accurately valuing the invention. This may often be the smallest priceable component containing the invention.”).
Trial courts have found that using the SSPPU as a royalty base is well suited to ensuring royalties for SEPs are reasonable and non-discriminatory. For example, in In re Innovatio IP Ventures Patent Litigation, the court held that there are “several significant advantages” to an approach that uses the “the profit margin on the sale of a chip for a chip manufacturer as the maximum potential royalty,” including that “it accounts for both the principle of non-discrimination and royalty stacking concerns in RAND licensing.” No. 11-cv-09308, 2013 WL 5593609, at *38 (N.D. Ill. Oct. 3, 2013). In particular, the court observed that “[c]onsidering the profit of the chip manufacturer on the chip, rather than the profit margins of the Manufacturers on the accused products, is appropriate because a RAND licensor such as Innovatio cannot discriminate between licensees on the basis of their position in the market.” Id. Further, the court concluded that “[i]f the royalty is excessive in comparison to a chip manufacturer’s profit margin on a chip, therefore, the royalty is too high.” Id. The court determined that this approach to determining royalties was beneficial because it “allows the court to base its RAND rate on objective considerations and sound hypotheses, rather than on mere speculation.” Id. at 39. Similarly, in GPNE Corp. v. Apple, Inc., the court held “as a matter of law” that for the asserted cellular SEPs, “the baseband processor is the proper smallest salable patent-practicing unit.” No. 12-CV- 02885-LHK, 2014 WL 1494247, at *13 (N.D. Cal. Apr. 16, 2014). As in Innovatio, the GPNE court concluded setting royalties based on chip price is “objective [and] based on quantifiable inputs.” Id. at *8.
The authors cite authorities that recognize the use of the SSPPU to derive royalties is not compulsory in every case, but those decisions in no way foreclose use of the SSPPU. In HTC Corp. v. Telefonaktiebolaget LM Ericsson, while the court held that a FRAND commitment to the European Telecommunications Standards Institute (ETSI) does not require, as a matter of French law, a SEP owner to license based on the SSPPU, it also held the inverse is true: “To be clear, the ETSI IPR policy neither requires nor precludes a license with a royalty based on the SSPPU.” No. 6:18-CV-00243-JRG, 2019 WL 126980, at *6 (E.D. Tex. Jan. 7, 2019) (emphasis supplied). That conclusion is consistent with an ETSI statement that it leaves SEP holders and licensees the choice about what royalty base to use: “It is reiterated that specific licensing terms and negotiations are commercial matters between the companies and shall not be addressed within ETSI. The basic principle of the ETSI IPR regime remains FRAND with no specific preference for any licensing model.” Likewise, in Federal Trade Commission v. Qualcomm Inc., the court simply observed that no court has held there is a “per se rule” that reasonable royalties must be calculated using the SSPPU. 969 F.3d 974, 998-99 (9th Cir. 2020). It is equally true that no court has held that royalties must be derived from the end device price. And, indeed, unlike with the SSPPU, courts carefully limit the circumstances when the end device can be used to the rare circumstance that the patentee “can prove that the patented invention drives demand for the accused end product.” Commonwealth Sci. & Indus. Rsch. Org. v. Cisco Sys., Inc., 809 F.3d 1295, 1302 (Fed. Cir. 2015).
Industry Practice Supports the Use of the SSPPU for FRAND Royalties
The third article cites a number of statements advocating the use of the end product as a royalty base and only two, from the Fair Standards Alliance (FSA) and Apple, that support the use of the SSPPU. In so doing, the article downplays significant industry recognition that the SSPPU is the appropriate royalty base for determining FRAND royalties for cellular SEPs.
As an initial matter, it is worth noting that the FSA position represents the views of a broad membership, which ranges from small and medium-sized to large companies spanning a range of industries, including car companies, computer and software companies, and cellular network operators. The FSA counts as members the following companies, among others: AirTies, BMW, Bullitt, Bury, Cisco, Continental, Daimler, Dell, Denso, Ford, Google, Harman, Hitachi, Honda, HP, Hyundai, Lenovo, Microsoft, Molex, Nordic Semiconductor, Sagemcom, Sierra Wireless, T-Mobile, Tesla, Titan, Toyota, Valeo, Visteon, and Volkswagen.
Beyond the FSA and Apple, other prominent companies—that both own their own and license others’ SEPs—have advocated the use of the SSPPU for FRAND royalties for cellular SEPs. Samsung, for example, in a 2015 submission to the European Commission, advocated that “[n]otwithstanding the unique circumstances of each negotiation and the case-by-case nature of the FRAND determination, the following elements ought to guide the process,” including the following:
The value that the functionality of the claimed invention in a given SEP contributes to the value of the relevant functionality of the smallest saleable unit that practices the claims of the SEP. Where small elements of multi-component products relate to the claim invention embodied in the SEP at hand, calculating a royalty on the entire end product carries a risk that the patentee will be improperly compensated for non-infringing components of that product[.]
Samsung, Response to DG Enterprise Questionnaire Patents and Standards 10-11 (Feb. 13, 2015). Likewise, Intel’s website includes a statement on Standards and Intel Business observing that a “FRAND royalty should reflect a number of factors including a royalty based on the smallest salable unit that practices the standard, the technical value of the patented technology compared to alternatives available during the standard-setting process, and the overall royalty that could reasonably be charged for all patents essential to the standard.”
As the authors acknowledge in the fifth article, an agreement on FRAND licensing principles, issued by a European Committee for Standardization Workshop in June 2019 and supported by more than 50 companies from a broad range of industries, observes that the SSPPU will often be a more suitable royalty base than the end product: “SEPs should be valued based on their own technical merits and scope, not based on downstream values or uses. In many cases this will involve focusing on the smallest component that directly or indirectly infringes the SEP, not the end product incorporating additional technologies.” Core Principles and Approaches for Licensing of SEPs, CEN/CENELEC Workshop Agreement 95000 10, 35 (June 2019).
In addition, while the article describes the current advocacy of certain SEP licensors for end product licensing of cellular SEPs, it omits that certain of these companies have previously advocated that cellular SEPs must be licensed to component suppliers. For example, while the article cites Ericsson and Qualcomm as examples of companies advocating licensing at the end product level, the trial testimony in the Federal Trade Commission’s case against Qualcomm confirmed that Qualcomm has a chip-level SEP license from Ericsson. Fed. Trade Comm’n v. Qualcomm Inc., 411 F. Supp. 3d 658, 753 (N.D. Cal. 2019), rev’d and vacated, 969 F.3d 974 (9th Cir. 2020) (FTC). Indeed, Ericsson was one of a number of companies that caused the European Commission to open an investigation in 2007 regarding whether Qualcomm was harming competition by breaching its FRAND commitments and failing to license competing baseband chip suppliers. Nokia, another company cited in the article as supporting end product licensing, was part of that group and advocated to the European Commission that Qualcomm’s termination of a license with baseband chip supplier Texas Instruments “would breach Qualcomm’s duty to license on FRAND terms, after having induced SSOs to base the CDMA and UMTS/W-CDMA standards on Qualcomm’s technology. Qualcomm actively led the ITU and ETSI, through its unequivocal IPR commitments, to develop a standard based on its patents.” FTC, ECF 893-2 at p. 135. As a Qualcomm executive explained to the Internal Revenue Service, Nokia and Ericsson switched positions and followed Qualcomm’s approach of focusing on the end product to make more money: “[S]o they also – following our lead I might say – you know, decided hey, we can license these patents and make money by doing and we can make more money licensing this than licensing the chip. So like they licensed the cell phone, not the chip.” FTC, 411 F. Supp. 3d at 755 (quoting CX-6786-R at 42:17-21).
Technological Developments Require Focus on Component Licensing
As described above, there is a track record of component licensing of cellular SEPs that some SEP licensors have tried to replace with the more lucrative practice of end product licensing that seeks to tax innovations in those products that have nothing to do with SEPs. As more and more technology has been introduced in smartphones and as cellular functionality has been introduced into a plethora of new products (such as cars), it has become clearer that SSPPU is the most balanced, fair, and proper mechanism for royalty valuation.
For example, as early as 2006, Nokia recognized that cellular functionality represented a diminishing portion of technology in a cellular phone: “To avoid doubt, there is no objection to charging a royalty based on the price of the handset, provided that the rate is modified to take into account the fact that Qualcomm’s patents cover functionalities that represent a small and decreasing proportion of that price.” FTC, ECF 893-2 at p. 135 (emphasis in original). Likewise, as Judge Koh recounted, Qualcomm’s own documents demonstrate its recognition that the user experience of a phone long ago became much more important than the cellular functionality provided by a chip. Thus, a 2008 Qualcomm strategic plan described the diminishing role of the chip or modem as follows:
FTC, 411 F. Supp. 3d at 781 (quoting CX7559-018) (emphasis in original).
The unreasonableness of levying a royalty based on the full device price is illustrated by considering the impact of adding memory to a smartphone. Samsung, for example, sells the Galaxy S21 Ultra 5G with 128 GB of memory for $499.99 and the same phone with 256 GB of memory for $549.99. Further, Samsung is introducing an A series of 5G smartphones for under $300, with a lower quality camera, lower quality display technology, and other lesser features. This variability in pricing of Samsung 5G phones demonstrates that the device price reflects a variety of factors and is not a reliable proxy for determining the value of 5G patents. InterDigital’s FRAND statement advocates that “[l]icenses should be made available at the point of the value chain”—i.e., the end device—“that reflects the value of the innovation and efficient industry practice.” But it is clear that adding more memory or more non-cellular features to a device has nothing to do with cellular innovation. Accordingly, there is no justification for a SEP licensor to obtain a higher rate simply because a device offers more features, be it more memory, a better screen, or a desirable user experience. Rather, the royalties for SEPs should, like any reasonable royalty, reflect just the value of the patented inventions.
The SSPPU is an Accurate and Proper Royalty Base
An accurate depiction of the debate over the appropriate royalty base for cellular SEPs must account for the many reasons that ordinarily make using the SSPPU the most accurate and proper methodological approach, including significant judicial authority, industry practice, and the changing nature of technology in modern devices such that cellular functionality represents a diminishing portion of the overall functionality of the end product. In this way, setting a FRAND royalty does not differ from determining any reasonable royalty, where the focus should properly be on simply gauging the incremental value of the patent, not inflating the amount by unfairly taxing unrelated innovation. Moreover, proper recognition of SSPPU principles promote the core economic goals of the FRAND safeguard—and protect against abuse of alleged SEPs, a topic taken up in the second article.
Image Source: Deposit Photos
Image ID: 13449947