Why evidence remains king when it comes to pricing SEPs

Standardised technologies are used in devices across a wide range of consumer and industrial markets. Wireless communications are deployed in smartphones, home appliances and now automobiles, while video coding has expanded beyond streaming media to play a vital role in videoconferencing and telepresence applications. As next-generation technical standards build upon legacy standards (eg, AVC/H.264 through H.266/VVC, WiFi 3 through WiFi 6 and 3G through 5G), SEP licensors and prospective licensees (many of whom are new to SEPs and their licensing idiosyncrasies) are developing and evaluating SEP-valuation frameworks. 

Two, in particular, have come to dominate the licensing landscape:

• the top-down framework, which focuses on a standard’s aggregate value and apportions that value to individual patents essential to that standard; and
• the bottom-up framework, which focuses on individual SEPs and values them independently of other essential patents. 

Top-down methodologies have been widely adopted around the world in recent years. The European Commission’s 2017 FRAND guidelines endorsed it, while US courts have applied it in 4G LTE disputes. However, the bottom-up framework is still relevant, having been used in US court decisions, such as the foundational Microsoft v Motorola. The Japanese Patent Office's 2022 Guide to Licensing SEPs has noted both top-down and bottom-up approaches and some US courts have drawn from both methodologies as appropriate in a given case, for example, In re Innovatio IP Ventures.

The top-down approach

The top-down framework starts by assessing the collective value of a standard, rather than individual value of specific patents. First, an aggregate royalty is determined, attributable to a standard’s features as a whole. Then, a royalty is allocated to a respective SEP owner based on the proportional contributions of that SEP to the standard. For example, if a standard contributes 10% of a product’s value, a patent owner controlling 20% of all SEPs to that standard would be allocated 2% of the value of each such product. 

While this may appear conceptually straightforward, it depends on several hotly contested parameters, such as:

• total aggregate royalty;
• royalty base; and
• total number of actually essential SEPs.

With regard to the aggregate royalty, a standard’s value may vary with a target product. For example, cellular functionality may be relevant to some products but ancillary to others. The relevant royalty base (eg, the smallest saleable patent practicing unit) may vary from a chip or module that provides the standard’s functionality, to the entirety of the product. The total number of SEPs may also vary with differing certainty depending on the standard. 

Both the International Telecommunication Union, which oversees video coding standards (eg, H.266/VVC), and the Institute for Electrical and Electronics Engineers, which oversees wireless local area network standards (eg, 802.11), accept blanket declarations of essentiality that do not identify specific patents. In contrast, the European Telecommunications Standards Institute (ETSI), which oversees cellular standards (eg, 5G NR), does require patent-specific declarations of essentiality. However, even the reliability of ETSI’s patent-specific declarations may be uncertain in view of possible over-declaration of patents that may not actually be essential. 

While various courts have applied top-down calculations in individual cases, they have often differed in their conclusions. In the United Kingdom in 2017, Justice Birss determined in Unwired Planet v Huawei that there were 800 patent families essential to 4G LTE and relevant to handsets. In 2018, the Court of Shenzhen in China applied the approach in Huawei v Samsung to determine whether the parties’ initial offers were FRAND. 

Interestingly, in 2019, Judge Selna, a US federal judge, declined to adopt Justice Birss’ findings or follow the Shenzhen Court’s methodology in TCL v Ericsson. Selna instead determined that there were 1,498 patent families essential to 4G LTE. So, while the top-down framework has logical appeal, its ultimate accuracy depends directly on the certainty of its inputs.

The bottom-up approach

This starts with an assessment of individual SEPs within a patent portfolio, often based on the modified Georgia-Pacific factors set forth by Judge Robart in Microsoft v Motorola:

• FRAND licensing rates from comparable licences;
• the value of the patented technology apart from the standard;
• the patent’s contribution to the standard; and
• the value of the patented technology to a product without reference to the value of the standard.

This approach builds on the familiar Georgia-Pacific framework, which is used to assess patent damages and focuses on facts that are available to both parties through discovery. 

However, the bottom-up approach is not immune from criticism. It also depends on commonly contested facts and extrapolations. For example, while comparable licences are often viewed as the best indications of patent value, the specific context of each list (eg, the relevant industry and negotiation) must be carefully analysed. The comparability of licences may also be heavily disputed, for example, where small dollar amounts and small volume licences are extrapolated to large dollar amounts and large volumes, and vice versa. Comparable licences may be difficult to determine in the context of new generations of standards that include both legacy and new features. 

The fact-specific analysis of patent value under the modified Georgia-Pacific factors may differentiate value of patents directed to new technology versus legacy technology. However, it might fail to address royalty stacking, which arises from patents of multiple adjacent and overlapping contributors to the new technology, and legacy patents that cover the new technology’s scope. In royalty stacking, the cumulative royalty (eg, via a bottom-up approach) for all SEPs for a standard is greater than the aggregate value of the standard to the product, and may even exceed the product’s value. For example, if each SEP for a standard is worth 2% of the product value, but there are more than 50 SEPs, the cumulative royalty would be greater than 100% of the product value. 

The road ahead

Ultimately, the quantification of SEP value will continue to be a fiercely contested matter as new generations of standards are developed that incorporate combinations of legacy technology and new technology. As those standardised technologies are deployed across an increasingly diverse set of products and markets, they each come with different expectations. Moving forward, parties may continue to use top-down, bottom-up or hybrid models to develop mutually acceptable resolutions for licensing. However, it is crucial that licensors and licensees adapt their licensing (and expectations concerning the appropriate royalty framework) to the best evidence available in a given licensing context.