Retired UK judge: ETSI FRAND pledge requires component-level licensing of cellular standard-essential patents

Today's IPKat/LSE Joint Event was entitled "The CJEU's billion-dollar questions -- who gets a SEP license and when should an injunction be granted?" One of Europe's most famous patent judges, recently-retired Lord Justice Sir Christopher Floyd, gave a clear answer to the first question: in his interpretation, ETSI's standard-essential patent (SEP) licensing pledge entitles every maker of equipment, including suppliers of components, to a license on FRAND terms.

That conclusion didn't surprise me. The ETSI agreement must be interpreted under French law, and at my Brussels conference on component-level SEP licensing in November 2019, French law professor Philippe Stoffel-Munck took the same position. What made the judge's position today particularly noteworthy is that he previously criticized the ETSI FRAND pledge for containing only about half the clarity that he'd like to see in it. He provide one example of such a shortcoming: the pledge doesn't specify in what forum any disputes over licensing terms should be resolved.

While some major cellular SEP holders--such as InterDigital, whose licensing chief Eeva Hakoranta also spoke today--argue that licensing at the end-product level is the standard in their industry, two industry representatives at today's webinar--though it's important to note they all expressed only their personal opinions--explained why component-level licensing is key to the ability of standardization to serve its purpose. Intel's IP policy chief Dr. Rebekka Porath mentioned that Intel, a member of approximately 300 standard-setting organizations, does grant SEP licenses at the component level. Last summer, a component-level SEP license deal between Huawei and Sharp became known (neither Huawei nor Sharp spoke today). Automotive supplier Continental's IP chief Dr. Roman Bonn explained the supply chain for connected cars, where cellular standards are implemented in the baseband chipset. What corroborates this view is what WilmerHale's patent and antitrust attorney Tim Syrett explained: he's litigated various SEP cases in the U.S. involving SEPs, and the infringement analysis always focused on the source code of the baseband chip. (This is a structural difference between SEP litigation in the U.S. and Germany; in the latter country, infringement allegations are typically based on the specification of a standard, not on what the accused products actually do.)

With respect to industry practice, SEP litigants, particularly Nokia, frequently point to the Avanci patent pool, which licenses end-product makers (and to the extent it has anything to offer to tier 1 suppliers, i.e. car makers' direct suppliers, that's not a full and exhaustive license). In today's IPKat/LSE webinar, Mrs. Hakaronta from InterDigital mentioned that 20 automotive brands had taken an Avanci license (while complaining in no uncertain terms about the attitude of other auto makers to SEP licensing, with a particular emphasis on Daimler).

The number of 20 (Avanci-licensee brands) is an overstatement because some of them have licenses that don't cover 4G (just up to 3G). Volkswagen's chief patent counsel Uwe Wiesner was among the speakers of yesterday's patent pools webinar, organized by the European Commission's DG GROW. His presentation quoted an October 2020 paper according to which "[t]he automakers that have taken a license represented approximately 12% of the total worldwide vehicle production in 2019." Volkswagen's high-volume brands apparently don't have a 4G license from Avanci.

Mr. Wiesner's Avanci "case study" painted anything but a rosy picture. He criticized Avanci's rule of licensing only car makers for not meeting the needs of potential customers. To put it differently, Mr. Wiesner didn't quite sound like a perfectly happy customer: at least it's fair to say he sees a lot of room for improvement, and isn't fully sold yet on Avanci's terms, despite being generally sympathetic to pools.

The Dusseldorf Regional Court's preliminary reference of access to component-level licenses to the European Court of Justice won't be decided in webinars. But such events shed light on the underlying facts and on industry realities. Plus, today a fomer high-profile judge took a crystal clear position on access to component-level SEP licenses based on the ETSI FRAND pledge.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Patent royalties may exceed $120 per smartphone, undermine industry profitability: working paper

A working paper by an Intel in-house counsel and two WilmerHale lawyers, The Smartphone Royalty Stack: Surveying Royalty Demands for the Components Within Modern Smartphones, has just been published (direct link to PDF). Intel Vice President and Associate General Counsel Ann Armstrong and WilmerHale's Joseph Mueller and Timothy Syrett have made an invaluable contribution to the debate over reasonable royalties and incentives for innovation in this field.

This first-rate paper finally answers the billion-dollar question everyone with an interest in smartphone patents has been asking for some time: the total licensing cost per device. The authors have thoroughly researched the licensing environment and highlight various key facts that should give policymakers, regulators and courts pause. They note that royalty stacking, "in which the cumulative demands of patent holders across the relevant technology or the device threaten to make it economically unviable to offer the product, [...] is not merely a theoretical concern" (as, by the way, the likes of Qualcomm allege). Based on publicly-available data, these competent authors "estimate potential patent royalties in excess of $120 on a hypothetical $400 smartphone--which is almost equal to the cost of [the] device's components" (estimated to be $120 to $150 in total based on figures published by Nomura Securities in reliance on Gartner data). They conclude that "those costs may be undermining industry profitability--and, in turn, diminishing incentives to invest and compete". I also believe that smartphone-related patent licensing costs, relating to standard-essential as well as non-standard-essential patents, must come down. Policymakers, antitrust enforcers and judges -- Judge Posner certainly did his best in this regard -- will hopefully bring those fees down in the years ahead.

The paper does properly distinguish between royalty demands and actual royalty payments. Patent holders frequently have to lower their demands during the course of negotiation. Cross-licenses and "patent exhaustion arising from licensed sales by component suppliers" can also make a major difference, but the terms on which companies actually agree are usually kept confidential. Royalty demands sometimes surface in litigation.

The authors based their study entirely on public documents. They (especially the WilmerHale lawyers, who, among other things, defend Apple against Samsung's counterclaims) have obviously seen some confidential license agreements, but couldn't make use of any of that information for their working paper. They also don't speak for any particular company or firm. Apple just demanded a "reasonable royalty" of $40 per device from Samsung at the recent California trial, for five software patents. Now a paper authored in part by lawyers representing Apple against Samsung (with a defensive focus, but still) says that $120 per device for everyone's patents, -- hardware and software patents, standard-essential and non-standard-essential patents -- may be "diminishing incentives to invest and compete". This shows independent thinking and writing. I would not be surprised to see Samsung's lawyers quote certain key findings of this study in their U.S. litigations with Apple. The paper appears slightly Apple-friendly to me in the context of the design patents-related part of Apple v. Samsung, but within reason (I agree in principle with what it says about that). The study also notes that UI patents can typically be worked around, and "[a] truly distinctive and innovative user interface--as distinct from a copied or derivative design--may result in minimal or no royalty exposure".

One key characteristic of the study is that it analyzes licensing costs on a component-by-component (including software components) basis: cellular baseband chip, random access memory (different kinds), flash memory (different kinds), WiFi, Bluetooth, GPS, NFC, battery, power management, audio (different subcategories such as MP3), camera/video (non-standards-based as well as standards-based formats like JPEG and H.264), applications processor, operating system, other pre-installed software, SMS, MMS, email, W3C (royalty-free standards), UPnP (royalty-free), digital media sharing, USB, user interface, outer design (also an area in which the study notes that infringement can be easily avoided).

The study has a much broader focus than my own litigation monitoring in recent years. Its findings appear plausible to me, except that I believe the "operating system" part of the royalty stack is underestimated. No operating system patent holder ever told me what their demands or actual deal terms were, but a couple of years ago I downloaded a litigation-related document that was publicly accessible for less than a day on the ITC document system that mentioned a major operating system patent holder's royalty demands. Against that background I think the study published today is very conservative (to say the least) with respect to operating system patent licensing costs -- but this, if anything, reinforces the overall message.

This is not a policy paper per se, but it does raise and stress policy concerns, particularly about non-practicing entities (NPEs), colloquially often referred to as "patent trolls", and the growing problem of "privateering" (patent transfers from major operating companies to NPEs in order to hide behind others that will assert patents aggressively against the original patent holder's competitors). Certain patent holders' demands, tactics and positions are discussed as examples of factors that exacerbate the royalty-stacking problem. Those patent holders include Ericsson and, to a far greater extent, Nokia, a company that has sold patents to a number of NPEs in recent years and is itself increasinly turning into a patent assertion entity.

This paper is recommended reading for everyone with an interest in smartphone IP issues from a legal and/or economic point of view. It's particularly recommended reading for all those who could, through their actions and decisions, address at least parts of the problem this paper describes. I believe it will be quoted a lot in court documents and academic writings in the years ahead.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|