les Nouvelles March 2017 Article of the Month
An Experience-Based Look At The Licensing Practices That Drive The Cellular Communications Industry: Whole Portfolio/Whole Device Licensing

Marvin Blecker

Retired-Formerly Qualcomm, Inc.
President Qualcomm Technology Licensing
San Diego, CA, USA

Tom Sanchez

Kelly & Krause, L.P.
Partner
Dallas, TX., USA

Eric Stasik

Avvika AB
Director
Stockholm, Sweden

I. Introduction

A. A Success Story

Cellular communications technologies have undergone dramatic changes during the last 25 years, as major technological improvements revolutionized the way the world communicates, and both consumes and creates digital content. Significant investments in research and development (“R&D”), combined with efficient standardization and patent licensing practices, brought about these changes and enabled successive generations of cellular communication standards to be implemented quickly and widely. The high data rates and lower prices achieved through third generation (3G) standards ushered in many of these changes, and created innumerable benefits at all levels of the industry, as mobile devices became viable platforms for internet browsing, sharing photographs, social networking, and even watching videos. The broad deployment of advanced devices and high-speed cellular communications networks has also supported the development of entirely new businesses—from Uber to ApplePay to mobile streaming video from Netflix, Amazon, DirecTV, CBS, NBC and others—based entirely on high-speed cellular communications infrastructure. With even faster data rates and more efficient use of wireless spectrum, the fourth generation (4G) ecosystem promises to be larger and even more successful than the 3G standard it is replacing. Those improvements will undoubtedly continue as fifth generation (5G) standards are developed, which will further increase performance and efficiency of cellular communications for the benefit of all.

Not only have the performance and functionality of cellular technology rapidly improved, but also prices of equipment and services have sharply decreased. Cellular phones—particularly smartphones—and other wireless devices (e.g., tablets) are now available at prices billions of users around the world can afford. The World Bank estimates that 96 percent of the world’s population is covered by mobile radio services and globally there are 93 mobile cellular subscriptions per 100 people.¹ As users acquire more and more devices, many companies have availed themselves of the industry’s standardized technologies, and entered the market for cellular communications equipment by focusing on developing sleek and sophisticated handsets. Many of these companies were not substantively involved in creating the foundational communications technology. Additional advances in cellular communications standards and technologies are likely to further reduce prices, enabling even more people around the world to acquire advanced wireless devices and more companies to enter the competitive handset market.

B. Calls for Change

The technological revolution briefly summarized above has taken place in the context of a set of long-standing industry standard licensing practices, which we discuss in some detail in this paper. Today, some industry participants, along with some voices in academia and within government agencies, are demanding fundamental changes to those licensing mechanisms. In particular, some influential voices have challenged the notion that whole-portfolio licenses covering and protecting whole devices (“whole-portfolio/ whole-device” licensing) are efficient, fair, reasonable, sensible or desirable. Recently, this viewpoint achieved a success in modifying the IEEE’s intellectual property rights policy in a manner intended to force changes in the licensing practices of innovators who contribute their patented technologies to standards, including by defining “fair and reasonable” in a new way that may preclude portfolio owners from licensing on a whole-portfolio/whole-device basis.² Similar changes are being advocated within other standards-development organizations (SDOs), and to regulators and courts around the world.

We are concerned that such changes are being proposed and debated in a historical vacuum. The changes that some are advocating would be fundamental departures from the licensing practices that have been widely used in the cellular communications industry from its inception to the present. Despite the importance of licensing to creating a healthy ecosystem and continuing innovation, discussion about changes to those longstanding licensing practices in SDOs, courts and government agencies is being largely conducted without consideration of how and why the challenged practices developed, received widespread acceptance within the cellular industry, have persisted for at least 20 years, and have contributed to the widely acknowledged successful growth and innovation of the wireless industry. We believe that making policy in this intellectual vacuum is unwise, and potentially highly detrimental to the industry and ultimately to consumers.

C. Historical and Market Experience

Each of the authors of this paper has participated intensively in portfolio licensing in the cellular industry for at least 20 years—that is, since the very early days of the industry. Specifically, in 1990, Tom Sanchez was put in charge of the licensing program at Motorola Cellular Subscriber, the predecessor to Motorola Mobility, and was responsible for licensing out Motorola’s significant patent portfolio, which grew to include numerous SEPs for the GSM standard. While at Motorola, Sanchez developed a template for what would later be termed FRAND licensing, which generally included a license to Motorola’s entire patent portfolio using the device price as the royalty base. In 2001, Sanchez joined Research in Motion (RIM), now known as Blackberry. When he joined RIM/Blackberry, Sanchez was primarily responsible for patent licensing and litigation, and later took on responsibility for all patent activities until he left in 2012 as a Senior Vice President of Patents and Licensing. Marvin Blecker joined Qualcomm in 1992 and immediately became involved in developing Qualcomm’s licensing program, including key early agreements with Panasonic, LG, Samsung and Hyundai. He continued negotiating out-licenses for Qualcomm as its licensing business rapidly expanded, and served as Qualcomm Technology Licensing (QTL) Division President until he “semi-retired” in 2008. Thereafter, he continued supporting the activities of QTL part-time until his full retirement in 2014. Eric Stasik began as a Senior Patent Engineer in 1992 for Ericsson-GE Mobile Communications responsible for obtaining patents on D-AMPS (IS-54) and GSM handsets and negotiating SEP in-licenses. From 1995 he was Manager of IPR activities for the UMTS and GSM Systems division of Ericsson Radio Systems responsible for patenting and licensing within Ericsson’s GSM and UMTS infrastructure business. From 2000 Stasik was Director of IPR & Licensing for Telefonaktiebolaget LM Ericsson, the corporate entity for all of Ericsson where he was primarily responsible for negotiating in-licenses. In 2002, Stasik left Ericsson and became an independent consultant advising and representing chipset vendors, handset manufactures, infrastructure manufacturers, and network operators in the mobile communications industry negotiating both in- and out-licenses. Since 2011 Stasik has also been the managing director for MKA Gruppen KB which acts as the exclusive broker for the sale of patents owned by the Swedish-Finnish Telecoms operator Telia Sonera AB.

Collectively, we have been involved in all manner of licensing negotiations from the early days of the industry through the present. Indeed, in years past we have even sat across the table from one another in hard-fought negotiations. Based on our extensive experience, we provide real-world perspectives on the history and continued utility of the longstanding practice of whole-portfolio/whole-device licensing. In our view, the historical background behind whole-portfolio/whole-device licensing, and the importance of this practice to the success of the industry is essential to understanding the legal and policy issues being debated today.

In this article we focus on licensing of major portfolios by companies that have made significant investments in high-risk R&D.³ Most of the major technological developments underpinning the success of the wireless telecommunications industry have come from a relatively few, innovative companies that have made very large investments in R&D and, as a result, have vast portfolios of relevant patents. This continues to be true even for the latest development of the 4G LTE standards. According to one estimate based on IP declarations made to ETSI as of November 2012, nine companies (Ericsson, Huawei, InterDigital, LG, Motorola, Nokia, Qualcomm, Samsung and ZTE) accounted for 71.4 percent of all patents declared essential to the LTE or LTE-Advanced standards.⁴

Confidentiality obligations preclude us from pointing to as many specific, real-world examples in the course of our discussions as we would prefer. However, given the extent, depth, and diversity of our respective experiences in licensing, where the three of us are in agreement as to the facts concerning or reasons for historical industry practice, this gives us high confidence that our consensus understanding is accurate.

In this article we: (i) provide an overview of standards setting in the cellular communications industry and how license negotiations are typically conducted, (ii) look at the prevalence of whole-portfolio or freedom-of-operation licenses; (iii) discuss the standard practice of using the price of a whole device as the royalty base; and (iv) consider the extent to which technical advances in wireless devices, and additional features found in present-day smartphones, have (or have not) impacted licensing practices.

In our view, the widespread availability of high-performance cellular handsets and numerous new businesses that have sprung up around this technology, testify to the tremendous dynamism and vigorous competition in the cellular industry. Although licensing all of the patented technology that goes into modern communication devices is necessarily complex, the whole-portfolio/whole-device licensing model has functioned well, simplified the process for both the licensee and licensor, and enabled the cellular communications industry to become a historic success story and a triumph of technological development, standardization, and dissemination through licensing. We write to explain how and why the practice of whole-portfolio/whole-device licensing developed, which we hope and believe can usefully inform current debates about licensing policy for the future of the cellular industry.

II. Standards Setting, Patent Portfolios and Licensing Negotiations in the Cellular Communications Industry

The cellular communications industry is built around standards and has flourished in part because newcomers can implement standardized technologies, and make and sell competitive devices that are interoperable with existing networks, without significant investments in R&D. The efficient and high-performance standards used in the industry must overcome numerous challenges, and major innovators devote significant resources to developing the innovations necessary to solve those engineering problems. Some of those innovators have developed portfolios with tens of thousands of patents, many of which are essential to the standard or otherwise practiced by cellular equipment. Below, we provide a brief high-level overview of standards development in the cellular communications industry and the patent portfolios held by major innovators, before giving a brief practical overview of the mechanics of licensing negotiations.

A. Standards Setting in the Cellular Communications Industry

Telecommunication standards are remarkably complex engineering creations. The modern cellular phone is undoubtedly by far the most complex consumer device that has ever existed. The paper documentation of the 4G LTE standard occupies thousands of pages, describing in great detail everything from overall system architecture through every detail of interaction between a device and a network, down to what each bit in each block of data means.

Standards projects typically start out with defined performance goals. For example, the goal for the 3G standard was a 10-times increase in capacity over second generation (2G) cellular.⁵ Once the goals are established, the process of determining requirements and developing technologies to achieve those requirements begins.⁶ The necessary inventions do not come easily or inevitably; participating innovators pour billions of dollars into R&D and contribute the resulting technical solutions to the standard-development efforts. Once participants reach agreement that they have a fully functional, fully detailed set of specifications, the standard is frozen and published.⁷

The difficult problems that need to be solved, and the level of detail contained in standards so that all devices are interoperable, makes this a laborious process. As an example, more than 200 companies from 13 countries spent approximately 886,000 person hours over 15 years developing 2G-related standard releases.⁸ Standards for subsequent generations have been released more quickly, but have taken even greater investments of time and resources by more companies. Developing 4G LTE releases is ongoing and has already taken more than nine years, and more than a million person hours from participants from more than 320 companies.⁹ These figures reflect only the time spent in actual SDO meetings. They do not account for the time engineers spent outside SDO meetings actually developing the innovations necessary to meet the goals and requirements set for the standard. That engineering time, for the major innovators, amounts to many multiples of the SDO meeting time.

Many of the innovations incorporated into standards have been patented, and must be licensed by anyone wishing to build fully standards compliant equipment. SDOs place great emphasis on developing the best possible technological solutions. At the same time, participants are prohibited from discussing licensing terms and conditions for the patented technologies used in standards.¹⁰ Instead, most SDOs request that participants disclose any patents they have for technologies included in the standard and agree to license those patents on fair, reasonable and non-discriminatory (FRAND) terms.¹¹ This procedure limits the ability of patent holders to exercise their statutory right to exclude or prevent other entities from implementing the standard, and avoids bogging down SDOs with what would otherwise surely be long and drawn out discussions over licensing terms.

Deferring licensing discussions is central to the efficient process of quickly designing new standards because licensing negotiations are complicated and often absorb significant resources, both in terms of time and money. Most SDOs in the cellular communications industry intentionally put licensing matters to the side so they can focus on developing new standards. For example, ETSI members specifically do not have a duty to disclose commercial terms, which are “a matter for discussion between the IPR holder and the potential licensee, outside of ETSI.”¹² If those commercial licensing terms had to be resolved during the course of standardization, standardization—and further technological innovation in the entire industry—would be greatly slowed. Once a standard is adopted (perhaps with some bilateral licenses completed and others still pending), wireless operators and implementers can start to develop their plans for implementing equipment, devices and services that utilize the standard. If they had to wait for final negotiation of license terms with all patent holders, that could result in drawing out the process by additional years and delaying the implementation of new technologies.

B. Major Patent Portfolios in the Cellular Industry

Major R&D contributors in the cellular communications industry invest heavily in high-risk R&D that may or may not succeed and, even if successful, may or may not be adopted by the industry, whether through standardization or in practice. One study estimates that, worldwide, companies invested over $1.8 trillion in cellular R&D between 2009 and 2013.¹³ As a result of those investments, major innovators typically have patent portfolios with large numbers of declared SEPs for cellular and other standards used by handsets, and significant numbers of non-SEPs as well. Their declared SEPs and non-SEPs cover inventions used in 3G and 4G air interface standards, WiFi, GPS, video codec technologies, radio chips, baseband processors, software and user interfaces, among other technologies. Major portfolios include patents on inventions used in products ranging from small components, to complete handsets, to base stations, to protocols governing the interactions among handsets and base stations in a network.

In an effort to develop the most efficient and highest performing cellular communications equipment, SDOs develop standards that incorporate many different inventions and, as a result, thousands of declared SEPs. For example, nearly 6,000 patents or patent families have been declared as essential to the 4G LTE and LTE-Advanced standards.¹⁴

C. Licensing Negotiations in the Cellular Communications Industry

To operate with cellular networks, handsets need to comply with the relevant communications standards, which means that handset manufacturers must use technologies covered by SEPs for those standards. SEPs or “standard essential patents” are patents on technologies that must be practiced in order to comply with a standard.¹⁵ They need licenses, at a minimum, to all of those SEPs, or else they will be liable for infringement. Generally, handsets also practice patented technologies that are non-SEPs, but are as a practical matter “commercially necessary” because they cover functionalities that customers typically demand, such as operating system functionalities, cameras, video codecs, WiFi or GPS. So, handset manufacturers need licenses not only to SEPs, but also to many non-SEPs.

Because of the potentially large number of patents involved, the transaction costs associated with licensing patent-by-patent, country-by-country, product-by-product would be astronomical. In the cellular communications industry, patent licensing—whether for SEPs, non-SEPs or both—generally follows the same course. With the exception of a few major innovators, patent holders will generally have to convince implementers that their patents are valid and interesting to (or, more commonly, are infringed by) the implementer. Implementers may already be using this technology without paying royalties, and will have to be convinced to pay for a license. These are often serious discussions that take place over numerous meetings among patent attorneys and technical experts, who analyze the patent claims and products in laborious detail. It can take months or years to negotiate a patent license and, as anyone who has sat through a long afternoon discussing claim charts knows, parties cannot discuss thousands or even hundreds of patents in detail if they hope to reach an agreement within a reasonable amount of time. As such, the parties generally agree to discuss a subset of representative patents from the whole portfolio. Depending on the size of the portfolio and patent holder, this may be anywhere from 20 to 100 individual patents.

Once the licensee is convinced that a license is necessary, commercial negotiations begin. In most cases, the parties reach an amicable agreement for a whole-portfolio license, which necessarily includes a license to all of the patent holder’s SEPs and often to the patent holder’s whole-portfolio, including all of the patent holder’s SEPs and many, if not all, of its non-SEPs. As discussed below, in our experience this type of broad “freedom of action” license to the patent holder’s entire portfolio is desired by both parties and is the most efficient, and most common, result.

While the process of negotiating a patent license has always been arduous, in the past, parties were generally able to negotiate license agreements, and it was relatively rare to reach an impasse and seek court intervention. Recent developments in law and policy have somewhat altered the landscape. Although the European Court of Justice recently confirmed, in Huawei v. ZTE,¹⁶ that SEP holders can seek and obtain injunctions, as a practical matter, that right has been constrained. Infringers of SEPs now have greater incentives to try their luck with litigation by challenging the validity of patents and, if unsuccessful, arguing over damages.

While weakening IP rights may be beneficial for implementers in the short-term, we believe that this trend poses serious risks for the future health of the mobile communications industry as a whole. The royalty fees generated by licensing patents are important to the entire industry because they compensate innovators for their successful investments in high-risk R&D that enable the industry’s progress and encourage continued investments. If IP rights are devalued, (i) new market entrants will get a “free ride” by being able to use such IP “on the cheap” without making any comparable investment in standardized technologies, thereby creating an inequitable market, and (ii) innovators will be less likely to make significant investments in high-risk R&D and less inclined to contribute valuable patented technology for use in standards, slowing technological progress in the industry. Courts, government agencies and others interested in the competitive health of the industry should be wary of attacks on patents and licensing practices that seek to further weaken IP rights and encourage use of proprietary technology without fair and reasonable compensation to the owner of the technology.

III. In the Wireless Communications Industry, Whole-Portfolio Licenses are Efficient and
Desirable

Whole-portfolio licenses, which give access to a licensor’s entire relevant portfolio of patents,¹⁷ are efficient and beneficial to all players in light of the broad patent portfolios held by many licensors in the cellular communications industry. We believe that it is for this reason that they are the historical norm throughout this industry. From the earliest years of the industry, major licensors, negotiated licenses to their entire patent portfolios, rather than on a patent-by-patent basis. In a publicly available 1995 agreement, Ericsson granted to NEC Corporation a portfolio patent license to all of Ericsson’s patents—a “whole-portfolio” license.¹⁸ We are aware that this was the routine out-licensing practice of Ericsson and also Alcatel, Bosch, Motorola, Nokia, Qualcomm, and others by that date or earlier. While almost all such licenses are non-public, we can say from our knowledge and experience that almost all patent licenses in the industry (at least those granted by industry participants) have been whole-portfolio licenses, and that this remains true up to the present.

Certain trends in the law and in the industry are encouraging some actors to challenge this model and instead attempt to conduct licensing on a patent-by-patent, country-by-country, and device-by-device basis. However, there are strong reasons why whole-portfolio licensing has been the durable model used by both innovators and implementers to license large portfolios of SEPs (and non-SEPs) for more than two decades. We review these reasons below.

A. Whole-Portfolio Licenses are the Most Efficient and Practical Way to License Major Portfolios

Given the breadth of major portfolios and the complexity of the devices being produced, whole-portfolio licenses are efficient, generally desirable to both licensors and licensees, and promote competition and innovation. Whole-portfolio licensing is consistent with the goals of licensing and standardization programs— to encourage the development and efficient licensing of the best possible technology, and seeing to it that the technological innovations incorporated into standards are implemented widely and quickly. In particular, whole-portfolio licenses promote competition (including introduction of new market entrants) and innovation by minimizing transaction costs, avoiding litigation costs, and providing parties with predictability about revenues and costs.

1. Minimizing Transaction Costs

Whole-portfolio licensing is practical and efficient because it minimizes the transaction costs of negotiating and enforcing license agreements for patents in major portfolios. As a practical matter, licensing on a patent-by-patent, country-by-country, and device-by-device basis is completely unworkable for the complex technologies that power the telecommunications industry, and the large portfolios held by major innovators.

For licensees, obtaining a whole-portfolio license avoids burdens and risks that would otherwise be associated with attempting to identify, in advance, the particular patents that read on a device, or component thereof, and then negotiating licenses specific to that device and those patents. It is not unusual for major portfolios to contain tens of thousands of patents in jurisdictions around the world, and the sheer number of patents makes it a daunting task for licensees to obtain à-la-carte licenses to all of the numerous patents they need. Not only is the size of the portfolios a significant issue for licensees to grapple with, but this is an extremely dynamic industry. Device makers’ product offerings are continually changing and standards are being amended and updated with new revisions and technologies. Negotiating license agreements on a device-by-device, patent-by-patent, and revision-by-revision basis would be a never-ending and impractical task that would expose handset manufacturers to continuous risks, with expensive consequences whenever they fail to obtain all of the necessary licenses. With relatively short product life cycles (one to two years), the resulting licensing uncertainty and risk would keep total device cost uncertain, potentially throughout its life cycle.

For licensors, licensing on a whole-portfolio basis avoids not only the same drawn-out, impractical negotiations, but also costly and burdensome efforts to monitor and manage compliance with a partial license. If a licensee did not have a whole-portfolio license, the licensor would need to expend resources scrutinizing the licensee’s products to determine whether they read on any patents that are not licensed, and then negotiate new agreements each time it found an unlicensed patent being practiced. Given the complexity of both the patents and the devices, the pace at which new devices are released, and the number of patents at issue, this would be a gargantuan task.

2. Avoiding Litigation Costs

Whole-portfolio licenses permit freedom of action and help parties to avoid litigation costs. Whenever a licensee fails to obtain a whole-portfolio license, it leaves its business exposed to an infringement suit for any or all of the unlicensed patents and constrains its design freedom. It is costly and risky for a licensee to attempt to analyze, in advance, which of the thousands of patents in a major portfolio read on a particular device. Leaving potentially applicable patents unlicensed would inevitably lead to more disputes and, likely, litigation.

This is hardly a hypothetical concern. HTC’s dealings with Nokia provide a real-world example. HTC was founded in 1997, and entered into a license agreement with Nokia. But it licensed only Nokia’s wireless SEPs, not Nokia’s entire portfolio. Predictably enough, Nokia then sued for infringement of some of the unlicensed patents in the U.S. International Trade Commission, a U.S. federal court, Germany, the United Kingdom, Italy, the Netherlands, France and Japan.¹⁹ Nokia obtained injunctions in the United Kingdom and Germany, and HTC ultimately agreed to settle all suits and to take a license to Nokia’s non-SEPs for an additional royalty on top of the royalty for the SEPs. If HTC had taken a license to Nokia’s entire portfolio in the first place, it would have avoided litigation, and the attendant costs and business disruption, as well as uncertainty surrounding HTC’s device business. A similar situation led to litigation between Nokia and Apple, with the same result: Apple took a license to additional Nokia patents and made a substantial payment to Nokia.²⁰ In our view, those lawsuits, ultimately resulting in whole-portfolio licenses, were a foregone conclusion from the moment that the licensees agreed to license only some of Nokia’s patents. Clearly, whole-portfolio licenses from the outset would have been more efficient for all parties.

3. Predictable Revenues and Costs

Whole-portfolio licenses have the additional benefits of providing certainty about license costs and predictability for forecasting financial results. Licensees know the royalties that they have to pay and do not have to worry about negotiating additional licenses, at additional costs, for rights to use certain patents that they did not initially license. Similarly, licensors can better predict the license revenue that they will receive because they know the royalties in advance and do not have to speculate about additional revenues that they might (or might not) receive through further negotiations or litigation. In our experience, in order to better manage their overall businesses, all parties to license negotiations value the certainty and predictability that come with whole-portfolio licenses.

B. Licensors and Licensees Generally Prefer Whole-Portfolio Licenses

Because whole-portfolio licenses are efficient and give manufacturers freedom to focus on designing and building the best and most efficient devices without concern of infringing any of the licensor’s patents, they are generally desired by both licensors and licensees. Licensees making fully functional devices that comply with a particular standard will clearly want (and need) a license to all of a licensor’s SEPs for that standard. Under the IP policy for the SDO that created the standard, licensors generally have an affirmative obligation to offer to grant licenses to all of their SEPs on FRAND terms and conditions. If a device manufacturer is to design, make or have made, and/or sell a fully standard-compliant device, it must by definition use all of the licensor’s SEPs. To the extent that a licensee thinks some of the patents a licensor claims are essential are invalid or not infringed—in other words, not SEPs—it can certainly raise those issues during negotiations to attempt to secure more favorable terms. However, even if the licensee believes that some of those patents are, for whatever reason, invalid or not infringed, it would be foolhardy to license only some SEPs. Taking a license to only some of the SEPs in a major portfolio would be a ticket to litigation and additional licensing costs.

For many of the same reasons, in practice, licensees find it efficient to license not only all SEPs, but also all non-SEPs, to guarantee freedom of operation. Major portfolios often include many non-SEPs that are important and widely practiced in various aspects of cellular communications equipment. Even though the non- SEPs may compete with other technologies protected by patents held by other companies, any benefit that might be obtained by carving out such patents is vastly outweighed by the increased transaction costs to do so, and the risk of litigation if, despite the manufacturer’s belief, unlicensed patents are nevertheless infringed. Consequently, manufacturers will generally want to secure a whole-portfolio license to guarantee that they have freedom of operation to make devices without concern of infringing a licensor’s patent rights.

In fact, in some instances where patent holders have attempted to offer partial portfolio licenses, licensees have resisted. They were concerned about ending up in litigation with the licensor, and having to make additional royalty payments, as a result of a partial license. Consequently, licensees insisted on whole-portfolio licenses. Qualcomm consistently heard such concerns from licensees; Qualcomm did offer whole-portfolio licenses and no more than a handful of licensees (out of approximately 250) ever requested a SEP-only license from Qualcomm. Instead, licensees generally wanted whole-portfolio licenses and even objected when Qualcomm attempted to withhold some non-SEPs, raising similar concerns about potentially needing to make additional royalty payments. Ericsson’s experience as licensor and licensee was similar.

Our experiences negotiating in-licenses are consistent. In our experience, it has been consistently true over the last 20 years that major innovators negotiating cross-licenses overwhelmingly want a license to their counter party’s entire portfolio, and not a partial license, so that they have freedom of action to make and sell devices without concern for potential infringement claims. Indeed, Qualcomm generally sought whole-portfolio licenses in its in-licensing. It was also Ericsson’s policy to obtain freedom of action by negotiating broad, whole-portfolio licenses including all SEPs and non-SEPs. For both companies, the only licenses taken to individual patents were in cases where the patent owner had only a small number of individual, relevant patents to license.

As a simple matter of commercial reality, licensees are motivated to negotiate whole-portfolio licenses instead of SEP-only licenses. In the same way that it costs much less to buy a whole car than to assemble one from parts purchased separately, the best way to obtain a license to the large number of patents that innovators need is to negotiate a license for all of them at the same time. Moreover, for most licensees, paying more to take a voluntary license to non-SEPs and gaining freedom of action is a better value than paying less for a SEP-only license and proceeding at risk of suit for infringement of unlicensed non-essential patents.

In light of the serious downsides of a partial license, and based on our experience in the industry, licensees almost always demand whole portfolio licenses. Generally, licensees who say that they do not want a whole-portfolio license do so as a negotiation position in an attempt to drive down the price of a license or to drive up the patent holder’s costs (in terms of extended negotiations and repeated patent-by-patent and country-by-country litigation until the whole portfolio of patents is tested in litigation). This shortsighted approach, if followed through, will likely lead to costly, inefficient and repeated disputes. In our experience, good-faith negotiators do not engage in this tactic, but we recognize that some licensees seek to exploit these tactics in an effort to secure better commercial terms in license agreements. Nevertheless, we believe that a portfolio-wide license consistent with the needs of all parties has historically proven to be be, and remains, the most appropriate, efficient, and common license structure in the industry.

C. Interference with Whole-Portfolio Licensing May Lead Patent Holders To Sell Patents, Making Licensing More Complex and Costly

Patent holders generally make significant, high-risk investments in research and development, and need to be able to earn a return on those investments. In the cellular communications industry, not only do patent holders face the usual risk of failure associated with development of new technologies, but they also face the additional risks (a) that their technologies will not be incorporated into standards or used in products and (b) that even if incorporated in a standard, that standard will not be widely adopted within the industry. Therefore, despite significant investments and effort, the number of “successful” R&D projects is low. One way to earn a return on the few successful investments is by licensing the patents. Historically, Ericsson, Motorola, Nokia, Qualcomm and other significant innovators in the cellular communications industry generally kept the patents for their inventions, and also acquired patents from others. They then licensed their entire portfolios to companies manufacturing cellular devices, as well as other equipment. The industry has flourished under that system, introducing highly innovative products and services, and new device manufacturers have repeatedly been able to quickly enter the market and gain market share under this licensing system while paying their fair share of the cost of prior development of standardized technologies through their payment of fair and reasonable royalties to patent holders.²¹ If regulators were to interfere with the common practice of negotiating whole-portfolio licenses—for example by encouraging piecemeal licensing—they would risk doing harm to the entire industry by stifling innovation and disrupting the fair, competitive balance in the market.

Consider the following unintended consequence of such regulatory interference. If additional restrictions were placed on the ability to license entire portfolios, large patent holders might instead find it more advantageous to sell off pieces of their portfolios to other companies. In fact, this has already been happening in recent years. For example, Nokia sold off small portions of its enormous portfolio of SEPs and non-SEPs to non-practicing entities including Sisvel,²² Vringo,²³ Conversant,²⁴ and others. Nokia has also transferred tens of thousands of patents, including (presumably) both SEPs and non-SEPs, into a Delaware trust called the “2011 Intellectual Property Asset Trust.”²⁵ The trust records further assignments to Sisvel, Core Wireless s.a.r.l., and others. Nokia is not alone. In January 2013, Ericsson transferred to Unwired Planet 2,185 U.S. and international patents and patent applications related to 2G, 3G and 4G LTE technologies.²⁶ And, LG, Panasonic, and Ericsson (by way of Highbridge Principle Strategies LLC) appear to have provided Optis Wireless Technology, LLC and PanOptis Patent Management LLC with a large number of patents.²⁷

Those sales are already affecting the efficient operation of the industry. Not so long ago, implementers could obtain a license to all of the SEPs and non-SEPs that resulted from Nokia’s large R&D investments by entering into a single, whole-portfolio license with Nokia. Licensing what had been Nokia’s entire portfolio will now require several license negotiations with completely separate entities. Policies that make it harder for innovators to reap rewards for their innovation through licensing will encourage such fragmentation.

The resulting inefficiencies can already be seen in publicly available information regarding two different patent litigations. For example, in 2011, Apple concluded a license agreement with Nokia “in settlement of all patent litigation between the companies”²⁸ for a one-time payment and on-going royalties from Apple to Nokia. However, because Nokia had transferred a portion of its patent portfolio to an entity not covered by the license agreement, Apple did not receive a license to all patents that originated with Nokia as part of the settlement. As noted above, Nokia had transferred a portion of its SEP and non-SEP portfolio (roughly 2,000 patents in total in approximately 400 patent families, 100 of which were declared essential to 2G, 3G, and 4G standards) to Core Wireless Licensing s.a.r.l. Core Wireless then sued Apple for infringement of “eight patents related to wireless communications in multiple versions of its iPhone and iPad.”²⁹ Less than a year after the settlement of all patent litigation between Apple and Nokia, Apple found itself again being sued for the infringement of Nokia patents (or more precisely, patents that originated with Nokia).

ZTE faced a similar situation after it signed a global cross-licensing agreement with Ericsson in 2012 that resolved all pending litigation between the companies.³⁰ However, as noted above, Ericsson had sold off part of its portfolio to Optis Cellular, which then sued ZTE for infringement of some of the patents it acquired from Ericsson. This serial litigation illustrates precisely why licensees demand whole portfolio licenses, and the challenges that result when large patent owners like Nokia and Ericsson fragment their portfolios or do not provide whole-portfolio licenses.

Not only does fragmentation increase transaction costs and cause greater uncertainty about whether devices are properly licensed or are infringing, but also it will likely increase the total cost of obtaining licenses. Implementers already complain about an alleged “royalty stack” as they need to obtain licenses from many different SEP holders. Whatever the actual “stack” paid by implementers today may be (this is a controversial question about which little hard information is publicly available),³¹ dividing up portfolios may add layers to that stack. And it may add more non-practicing entities (NPEs, which do not make or sell products implementing a standard) to the stack, because they are likely candidates to purchase such patents. Unlike the original patent owner, which may need to obtain a reciprocal license, NPEs are in a position to convert the need for a reciprocal license into cash compensation, which may drive up the cash component of royalty compensation.

Selling off patents to NPEs or other innovators is, of course, perfectly permissible and common in all industries. Nevertheless, it presents an already growing challenge for the cellular communications industry, and policies and regulations that discourage whole-portfolio licensing, particularly the licensing of whole portfolios of SEPs, threaten to exacerbate the situation. In our view, this would substantially increase patent licensing costs in the industry with little—if any—corresponding benefit.

We are now in the fourth generation of standards-based digital cellular networks and services in less than 20 years. This could not have been accomplished without an efficient system of standard development and patent licensing within the industry. The result has been less costly devices and services, using an astounding number of innovative technologies that deliver an amazing array of cellular telephone and digital services delivered efficiently to the consumer.

IV. Whole-Device Licensing is an Efficient and Universally Accepted Norm

The longstanding practice from the first licensing programs in the cellular communications industry has been to license complete handsets, not the components of which they are constructed. This is the logical and natural thing to do. Many of the same reasons for utilizing whole-portfolio licensing apply equally to licensing the whole devices rather than trying to determine which patents read on each of the numerous and changing components within, or subassemblies of, devices. Indeed, typical licenses in the cellular industry provide that any complete handset made by the licensee that complies with a specified set of standards (e.g., 3G or 4G LTE) is licensed under the licensor’s entire patent portfolio—a highly efficient approach that obviates all the potential device-by-device and patent-by-patent wrangling referred to above.

Nevertheless, some are currently advocating that SEP owners should be required to grant licenses to component manufacturers, rather than only to the manufacturers of complete devices. The motivation for this is multiplication: Because components are necessarily cheaper than complete devices, the total royalty payments would be reduced if the royalty rate stayed the same but was multiplied by the cost of the components. Of course, a change in royalty base would not mean that the value of the licensed intellectual property in the handset has been reduced, but clearly these advocates expect that a requirement to grant licenses at the component level will result in overall lower royalty payments to innovators.

However, putting to one side the question of whether a reduction in total return to innovators would be healthy for the industry, based on our experience, component-level licensing would be a significant change, and it would make licensing far more burdensome and inefficient than the existing practice of licensing at the device level.

A. Device Price Has Traditionally Been Used as the Royalty Base

From the start of the cellular communications industry, the device (handset) price has been the widely used royalty base in license agreements. All of the early innovators in the wireless communications industry that made significant investments to develop the fundamental technologies that support wireless communications—including Alcatel, Ericsson, Lucent, Nokia, Qualcomm, and Siemens—used the “box price” or “net selling price,”³² as the royalty base in their early license agreements. For convenience, this article uses the term “device price” to refer to either the “box price” or the “net selling price.” Regardless of the specific contract terms, the important point is that nearly all agreements are based on the price received by the manufacturer from the sale of a fully-functional, end-user device (e.g., handset), and not the individual components, subassemblies or combinations of components contained in it.

In some of the earliest license agreements in this industry, Motorola used the “box price” as the royalty base, which included the cost of the handset as well as other items, such as the charger. Later Motorola agreements continued to use the device price as the royalty base, but subtracted the cost of certain accessories included in the “box.” For example, a 1992 license between Motorola and Ericsson set the royalty rate at 2.25 percent “of the price (or greater Fair Market Value, FMV)” of the “subscriber equipment,” excluding “externally connected power sources and externally connected devices having other substantial non-radio uses (e.g., facsimile and computer peripherals).”³³ Qualcomm’s earliest bi-lateral license agreements with Motorola and with ATT in 1990 followed this model using the “net selling price” (or greater FMV).

Other companies followed Motorola’s lead, and used the device price as the royalty base when establishing their licensing programs and setting royalty rates.³⁴ In our experience, all significant patent holders in the cellular communications industry use the device price as the royalty base in license agreements. Based on a review of publicly available disclosures, including regulatory filings, documents produced in litigation and press releases, the following 29 major licensor companies (listed in order of the number of declared SEPs) continue to license their 3G and 4G SEPs at the device level and use the device price as the royalty base: Qualcomm, InterDigital, Nokia, Ericsson, Motorola, Samsung, Huawei, Panasonic, NEC, Sharp, NTT, Siemens, ZTE, Apple, Alcatel-Lucent, Nortel Networks, Mitsubishi Electric, Dolby Laboratories, Fujitsu, Toshiba, Infineon, Wi-Lan, Thomson Licensing, AT&T, Deutsche Telekom, Vringo, Hewlett-Packard, SK Telecom, Telecom Italia. Additionally, the Sipro, Sisvel and Viva Licensing patent pools—which include patents for a number of other innovators—also license at the device level and use the device price as the royalty base.³⁵

B. Benefits of Using the Device Price as the Royalty Base

Using the device price as the royalty base has many advantages for licensors and licensees in the cellular communications industry. In particular, based on our experience, using the device price as the royalty base is appropriate because it (i) best accounts for the diverse patents in major portfolios, (ii) best reflects the value created by the licensed patents, (iii) provides freedom of operation to licensees to develop new devices and change component suppliers, and (iv) simplifies relationships between licensors and licensees, helping to avoid disputes.

1. The Device Price Accounts for the Diverse Patents in Major Portfolios.
Major portfolios typically have a broad array of patents that read on (i) various components alone, (ii) various components in combination, (iii) complete handsets alone, and (iv) complete handsets in networks. This is because, as noted above, cellular communication standards cover everything from the general architecture of the network, to the operational features of fully functional infrastructure equipment working alone or together with other devices, to algorithms carried out by baseband processors (chipsets), to the meaning of one bit in one block of information, and a myriad of other issues in between. Those operations and features of the standard often are the result of a patented invention contributed by one of the participants to development of the standard. In our collective experience, major innovators in the cellular communications industry frequently hold patents on numerous different kinds of technologies used in cellular communications. They are likely to have patents covering chipsets, handsets, network infrastructure, protocols, and entire networks. The practical and efficient way to license such a broad portfolio is to license, and determine the royalty based upon the price of, the handset—which uses all those features.

Indeed, major portfolios will generally have patents that read on an entire device, and for those patents, the device price is the only natural and appropriate royalty base. Because any such patents should be licensed using the device price as the royalty base, it is practical and efficient to also use the device price as the royalty base for other patents that are licensed from the same portfolio. Even if a major portfolio contained no patents that read on a whole device—an exceedingly unlikely possibility in our experience—the device price would nevertheless still be the most appropriate royalty base for portfolios that include a diverse array of patents. The portfolio will invariably include patents with claims that read on various components and combinations of components. It would not be practical to determine a separate royalty base and rate for each of those patents and the price of the handset is the only logical and appropriate royalty base.³⁶

Because many patents have claims reading on entire handsets and combinations of components, if patents were instead licensed separately to the manufacturers of various components, handset manufacturers would still need a license to many patents in most portfolios. This would create an additional layer of complexity and additional transaction costs without providing any clear benefit to the handset manufacturer, component supplier or patent holder. Of course, the negotiated composite royalty rate takes into account the value of the licensed intellectual property and the fact that the various patents in the portfolio have varying scopes.

2.The Device Price Best Reflects the Value Created by the Licensed Patents.

The device price is the appropriate royalty base for licensing a major portfolio because it most accurately represents the value of the licensed patent rights. Even if one could determine the specific components of a device to which particular patents apply and the cost of those components, such cost does not necessarily represent the significant value that high-speed cellular connectivity brings to those devices. For example, consider the cost of an iPhone compared to an iPod Touch. The iPhone 5S 16GB and iPod Touch 6th Generation 16GB both possess front- and rear-facing cameras, the same display with touch screen functionalities and have the same video and audio playback capabilities. The main difference between them is that the iPhone has cellular connectivity and the iPod Touch has only WiFi. The iPhone sells for $450 (unsubsidized, as sold without service contract), while the iPod touch sells for $199—less than half the price. Clearly, the additional value generated by adding a cellular modem to an iPhone Touch cannot be measured by just the cost of the additional electronics, which are considerably less than the $250 price differential.³⁷ The price of the device sold on the open market represents the value to consumers of that device, and largely reflects the value of the cellular communications technology.

3. Using the Device Price as the Royalty Base Simplifies Relationships and Avoids Disputes.

Using the device price as the royalty base simplifies relationships and avoids potential disputes by permitting straightforward and auditable royalty calculations based on the price at which the licensee actually sells the product. Third party market research firms provide data on device sales, which licensors can use to confirm royalty reports and verify that licensees are complying with their obligations to pay royalties. In contrast, licensors generally will not have access to accurate and verifiable prices for various components used in a device (particularly because different device manufacturers may receive different pricing from the component supplier dependent upon many factors, including payment terms, warranty, quantities, lead times, and other commercial details). The lack of independent information would surely lead to negotiations that are more contentious, require complicated monitoring, and invite gamesmanship and disputes over royalty payments. These added costs would need to be considered in setting royalty rates and prices charged by the licensee for its devices. Those inefficiencies are all avoided when the device price is used as the royalty base, which benefits the entire industry and consumers.

4. Licensing the Whole Device Is Appropriate for Major Portfolios and Promotes Freedom of Operation.

Licensing the whole device has the added benefit of providing licensees with the freedom of operation that they desire. Licensees want freedom to make and sell devices without concerns about infringing a licensor’s patents, and the best way to achieve that is simply to license the whole device under all the patents. Naturally, when the whole device is licensed, the royalty base should be the price of that whole device.

Freedom of action is particularly important for two reasons. First, there is the completely practical matter that the component suppliers can, and frequently do, change from generation to generation of a device maker’s products. Baseband chip supplier A may win a socket in a phone for the 2015 product lineup, but may lose out to supplier B in the 2016 lineup.³⁸ Thus, with component level licensing, new crops of companies would likely need to be licensed for successive generations of devices, creating unnecessary delays in design and manufacturing of new products.

Second, the nature of the components included in devices changes over time. To take just one example, GPS and WiFi capabilities were not included in the earliest handsets, but are standard now. In their earliest incarnations, GPS and WiFi were implemented in cellular phones by unique components; now they are generally integrated into chipsets that perform other functions as well (e.g., cellular baseband components). As those new capabilities and components are added to the handset and/or to other components in the handset, IP holders would have to negotiate licenses— potentially with new companies—to cover those additional features. To be certain that they are making and selling non-infringing products with respect to each licensor’s patent portfolio, handset manufacturers would need to investigate the license status of each of their component suppliers on an ongoing basis—an arduous task.

Each of these variables multiplies the complexities for both licensors and licensees in making sure that all the technologies used in a device are licensed for that device. The whole-device licensing approach solves this problem completely, while providing freedom of operation, simplifying relationships and reducing the potential for disputes.

C. SDO Members Adopted Standards Knowing That the Device Price Would Be the Royalty Base

It should be noted that the 3G and 4G cellular communications standards that are widely used today were adopted with all players knowing that the key SEP holders licensed their patents using the device price as the royalty base. No one was surprised by this arrangement. It was part of the basis of the bargain made among innovators and implementers during the development of the standards.

The relevant SDOs are open to all who wish to participate. Their members include innovators, implementers, and hybrids between the two. Many of the members have been involved through prior generations of cellular communications technologies. Consequently, the relevant standards were adopted with full knowledge of the licensing landscape, including the fact that many SEPs were owned by companies that license them on a whole-device basis, using the device price as the royalty base. Other players who chose not to participate in the work of standard development were (or certainly could, and should, have been) aware of that fact.

There were extensive discussions before the 3G UMTS standard was adopted. That standard was developed with full knowledge that the device price would be used as the royalty base, at least for licenses from Ericsson, Motorola, Nokia, and Qualcomm.

The same holds true for the 4G LTE standard. In that round of standards development, an industry organization called Next Generation Mobile Networks (NGMN) set up an IP Policy Committee to explore potential licensing issues. Largely instigated by those meetings, many of the expected significant licensors voluntarily and publicly stated, prior to adoption of the standard, their anticipated license terms on their websites. Significant SEP holders, including Alcatel, Ericsson, Huawei, Motorola, Nokia, Nortel, Qualcomm, and ZTE, all stated that they would use the device price as the royalty base in their license agreements.³⁹ Numerous agreements were then negotiated and signed using the device price as the royalty base.

Whole-device licensing was well known. It was part of the bargain that all participants relied on.

V. Whole-Portfolio/Whole-Device Licensing has Proven Itself to be, and Remains, the Best Way to License IP Rights for Present-Day Smartphones

We have been involved with this industry from the beginning, and have witnessed the transformation of cellular handsets from large, clunky analog boxes that could barely manage voice communications in limited areas to sleek, sophisticated, multi-media smartphones (digital devices more powerful than computers were just a few years ago) that work around the world. As smartphones grow increasingly advanced and integrate features like high-resolution screens and advanced video processing, critics have argued that the longstanding practice of licensing whole portfolios with the device price as the royalty base no longer makes sense. We strongly disagree. In fact, as phones grow more complex and incorporate additional features that rely on the core cellular communications technology, whole-portfolio licensing at the device level is increasingly important.

The additional complexity in present-day smartphones makes it more complicated to determine, in advance, the patents that read on (or potentially read on) devices, subassemblies or components thereof. The number of technologies (and patents) incorporated into smartphones continues to grow. So too, the number of patents held by the major innovators has expanded. As noted above, any attempt to identify all the SSPPUs for all the patents in a major portfolio would be unreasonably difficult and costly. The increasing complexities of technologies and patents only make that task harder to complete and less realistic to achieve. Correspondingly, it is all the more important for manufacturers to have broad protection from claims of infringement for their increasingly complex devices, and all the more attractive for patent holders to license on a whole-portfolio and whole-device basis.

VI. Conclusion

Whole-portfolio/whole-device licenses have been the standard practice through four generations of wireless communications technologies. There are excellent reasons for this. For major patent portfolios, whole-portfolio licenses are efficient, desirable and promote competition and innovation. Similarly, licensing fully-functional, whole devices, using the device price as the royalty base, best addresses the value created by the cellular communications technology being licensed, and is the most efficient and practical approach for both licensors and licensees. Whole-portfolio/whole-device licensing has been the foundation of the bargain between innovators and implementers since the beginning of the industry, and it has served the industry and consumers spectacularly well, as demonstrated by the unprecedented growth and dynamism of cellular communications. Any government agency or other organization asked to limit the flexibility to form these types of agreements should consider the background behind and reasons for current licensing practices in the cellular communications industry, and should proceed cautiously so as not to harm the industry by restricting efficient and appropriate licensing. Ill-advised policymaking at the behest of the implementer segment of the industry risks (i) elevating the cost of license maintenance and enforcement, (ii) generating unnecessary additional legal disputes, (iii) disincentivizing investment in innovation, and (iv) providing unfair advantages to market participants that have not invested in industry development.

Available at Social Science Research Network (SSRN): https://ssrn.com/abstract=2855078

---

1. The World Bank, “World Development Indicators: Power and Communications,” http://wdi.worldbank.org/table/5.11. Of course, statistics on the number of subscriptions per capita do not account for people who have multiple phones, or a phone plus a tablet, eReader, or PC with a cellular connection.
2. Press Release, IEEE Statement Regarding Updating of its Standards-Related Patent Policy (Feb. 8, 2015), available at https://www.ieee.org/about/news/2015/8_february_2015.html. This new policy has been roundly criticized as harmful to innovation and intellectual property rights, including by former Commissioner of the U.S. Federal Trade Commission Joshua Wright. FTC Commissioner Wright Criticizes DOJ IEEE Letter, ABA Intellectual Property Committee tidBITS, available at http://www.americanbar.org/content/dam/aba/publications/antitrust_law/at315000_tidbits_20150315.authcheckdam.pdf.
3. This paper does not discuss small-scale licensing by non-practicing entities and other entities that may own only a small number of patents, and represent a limited portion of the patent holdings in the industry.
4. Cyber Creative Institute, “Evaluation of LTE Essential Patents Declared to ETSI,” at 9, available at http://cybersoken.com/ research/pdf/lte03EN.pdf.
5. Boston Consulting Group, “The Mobile Revolution,” at 29 (Jan. 2015); Kirti Gupta, “Technology Standards and Competition in the Mobile Wireless Industry,” 22 Geo. Mason L. Rev. 865, 871-72 (2015).
6. Boston Consulting Group, “The Mobile Revolution,” at 29- 30 (Jan. 2015).
7. Boston Consulting Group, “The Mobile Revolution,” at 30 (Jan. 2015).
8. Boston Consulting Group, “The Mobile Revolution,” at 30 (Jan. 2015).
9. Boston Consulting Group, “The Mobile Revolution,” at 30 (Jan. 2015).
10. ETSI’s guidelines state: “Specific licensing terms and negotiations are commercial issues between the companies and shall not be addressed within ETSI. Technical Bodies are not the appropriate place to discuss IPR Issues. Technical Bodies do not have the competence to deal with commercial issues. Members attending ETSI Technical Bodies are often technical experts who do not have legal or business responsibilities with regard to licensing issues. Discussion on licensing issues among competitors in a standards making process can significantly complicate, delay or derail this process.” ETSI Guide on Intellectual Property Rights § 4.1 (Adopted by Board #94 on September 19, 2013), available at http://www.etsi.org/images/files/IPR/etsi-guide-on-ipr.pdf.
11. Some SDOs use the term RAND for “reasonable and non-discriminatory” instead of FRAND. Within the industry, FRAND and RAND are generally understood to have the same meaning.
12. ETSI Guide on Intellectual Property Rights § 2.2 (Adopted by Board #94 on September 19, 2013), available at http:// www.etsi.org/images/files/IPR/etsi-guide-on-ipr.pdf.
13. Boston Consulting Group, “The Mobile Revolution,” at 5 (Jan. 2015).
14. Cyber Creative Institute, “Evaluation of LTE Essential Patents Declared to ETSI,” at 4, available at http://cybersoken.com/ research/pdf/lte03EN.pdf.
15. For example, the European Telecommunications Standards Institute (“ETSI”), explains a patent is essential if it is “not possible on technical (but not commercial) grounds, taking into account normal technical practice and the state of the art generally available at the time of standardization, to make, sell, lease, otherwise dispose of, repair, use or operate EQUIPMENT or METHODS which comply with a STANDARD without infringing that IPR [intellectual property right]. For the avoidance of doubt in exceptional cases where a STANDARD can only be implemented by technical solutions, all of which are infringements of IPRs, all such IPRs shall be considered ESSENTIAL.” ETSI Rules of Procedure, Nov. 19, 2014, ¶ 15.6, http://www. etsi.org/images/files/IPR/etsi-ipr-policy.pdf. Non-SEPs are patents that are not SEPs.
16. Huawei Technologies Co. Ltd. v. ZTE Corp., Case No. C-170/13 (Court of Justice of the European Union, July 16, 2015), available at http://curia.europa.eu/juris/document/ document.jsf?doclang=EN&text=&pageIndex=1&part=1&mode=lst&docid=165911&occ=first&dir=&cid=3858.
17. Throughout this paper, we use the terms “entire portfolio of patents” or “whole portfolio” to mean a licensor’s entire (or nearly entire) portfolio of patents in existence at the time of the agreement. Many whole-portfolio licenses also include future patents, at least for a period of time. Some licensors may withhold a small subset of patents that are not essential (non-SEPs) to implementation of the wireless/cellular standard from a license.
18. This agreement was filed as Plaintiff’s Exhibit 74 in Harris Corp. v. Ericsson, Inc., No. 98 Civ. 2903 (N.D. Tex.).
19. See Michael Calia, “Nokia and HTC Settle Patent Litigation,” Wall St. J., (Feb. 7, 2014), http://www.wsj.com/articles/SB 10001424052702304680904579369141759210688.
20. See Kevin J. O’Brien, “Nokia Settles 2-Year Fight With Apple on Patents,” N.Y. Times, (June 14, 2011), http://www.nytimes.com/2011/06/15/technology/15nokia.html?_r=0.
21. For example, numerous companies have entered the market for cellular handsets, and have rapidly gained market share from more established rivals, which had made substantial contributions to developing the patented technologies used in the industry and controlled numerous SEPs. In particular, companies like Apple, HTC, Huawei, LG, RIM/Blackberry, Samsung, Xiomei, and ZTE all entered the market and rapidly gained meaningful market share prior to developing or owning any meaningful SEPs for cellular communications standards.
22. Sisvel International, “Sisvel Acquires Over 450 Nokia Patents, Including Over 350 Patents Essential to Wireless Standards”(Jan. 12, 2012), http://www.sisvel.com/index.php/sisvel-news/257-sisvel-acquires-over-450-nokia-patents-including-over-350-patents-essential-to-wireless-standards2.
23. Business Wire, “Vringo and Nokia Execute Patent Purcahse Agreement” (Aug. 9, 2012), http://www.businesswire.com/news/home/20120809005600/en/Vringo-Nokia-Execute-Patent-Purchase-Agreement#.VgpmrrRt7UI.
24. Conversant Intellectual Property Management, “Core Wireless Licensing, http://www.conversantip.com/patent-category/core-wireless/.
25. PlainSite, “2011 Intellectual Property Asset Trust,” http://www.plainsite.org/flashlight/2011-intellectual-property-asset-trust/.
26. Eric Savitz, “Litigation Ahead! Unwired Planet Buys 2,185 Ericsson Patents,” Forbes (Jan. 14, 2013), http://www.forbes. com/sites/ericsavitz/2013/01/14/litigation-ahead-unwired-planet-buys-2185-ericsson-patents/.
27. PlainSite, “Optis Cellular Technology,” LLC, http://www. plainsite.org/flashlight/optis-cellular-technology-llc/page-2/table-patents.
28. Nokia, “Nokia Enters Into Patent License Agreement with Apple” (June 14, 2011), http://company.nokia.com/en/news/ press-releases/2011/06/14/nokia-enters-into-patent-license-agreement-with-apple.
29. John Letzing, “Apple Sued by Firm in Patent Deal with Microsoft,” Wall St. Journl, March 5, 2012, available at http:// www.wsj.com/articles/SB100014240529702033706045772637 70072976042.
30. Fierce Wireless Europe, “Ericsson, ZTE Settle Global Patent Dispute” (Jan. 20, 2012), http://www.fiercewireless.com/europe/story/ericsson-zte-settle-global-patent-dispute/2012-01-20.
31. An analysis of royalty revenues by major licensors suggests that total mobile-SEP royalties on handsets are about five percent. Keith Mallinson, Cumulative Mobile-SEP Royalty Payments No More Than Around five percent of Mobile Handset Revenues, IP Finance (Aug. 19, 2015), http://www.wiseharbor.com/pdfs/Mallinson%20on%20cumulative%20mobile%20SEP%20royalties%20for%20IP%20Finance%202015Aug19.pdf
32. The “net selling price” is usually specified in the relevant license agreement to be the box price less specified deductions for ancillary costs such as taxes, shipping, insurance and included accessories. The “box price” is usually specified as including the price of everything in the box with the device or handset.
33. This agreement was filed as Plaintiff’s Exhibit 72 in Harris Corp. v. Ericsson, Inc., No. 98 Civ. 2903 (N.D. Tex.).
34. Although Qualcomm may have initially “followed Motorola’s lead,” Qualcomm soon came to understand that licensing complete end user devices on the basis of the “net selling price” of those devices is the most efficient and pragmatic way to license its broad patent portfolio.
35. In the case of cellular handsets, the device price is generally the price of a handset, less certain specified deductions. In other contexts, for example, when a car is equipped with cellular connectivity, it may be appropriate for the royalty base to be defined to be a cellular communications module.

1. 
   
2. 36. For similar reasons, the Smallest Saleable Patent Practicing Unit (SSPPU) theory espoused by certain advocates would not be feasible in the context of negotiating a license agreement. And, even if it were applied, when considering a major portfolio, the SSPPU would undoubtedly be the device, which is the smallest unit that implements all licensed patents in the portfolio.
   37. Based on estimates of the bill of materials in the iPhone 5C, the cost of the cellular connectivity components was around $32 at the time it was released. See https://technology.ihs. com/451425/groundbreaking-iphone-5s-carries-199-bom-and-manufacturing-cost-ihs-teardown-reveals.
   38. For example, as companies such as Samsung and Apple prepare to release new handsets, the baseband and application processers that will be used in those new models, and potential shifts to different chip suppliers, are widely reported. See Jungah Lee, “Samsung’s Galaxy S6 Isn’t About the Phone, It’s About the Chips,” Bloomberg (Feb. 26, 2015), available at http:// www.bloomberg.com/news/articles/2015-02-26/samsung-looks-beyond-apple-to-challenge-qualcomm-with-next-phone; “Intel Inside the iPhone: LTE Modem and Camera?,” Seeking Alpha (Mar. 16, 2015), http://seekingalpha.com/article/3003076-intel-inside-the-iphone-lte-modem-and-camera. Changing chip suppliers is likely to become increasingly prevalent as Huawei, Intel, Mediatek, Qualcomm, Samsung, Spreadtrum, Via and other manufacturers all compete vigorously for market share.
   39. Erik Stasik, “Royalty Rates And Licensing Strategies For Essential Patents on LTE (4G) Telecommunication Standards,” les Nouvelles, Sept. 2010 at 114, 115-16.