In various industries including pharmaceuticals, where a product is typically protected by only a few patents, if intellectual property is licensed at all, licensing tends to be atomized: company-by company for individual patents or small portfolios, and nation-by-nation rather than globally in many cases.
In marked contrast, patent pooling is increasingly attractive for licensing cellular technologies with emerging IoT including 5G because it can provide greater transparency, predictability, and various efficiencies such as lower transaction costs at scale in standard-essential patent (SEP) licensing with multiple dimensions and complexities including:
- Scores of patent jurisdictions, but with a handful of these most significant by far;¹
- Hundreds of patent owners, while most SEPs are owned by a very small proportion of these;²
- Thousands of implementers, with the vast majority of these in IoT, outside of the cellular industry vertical;³ and
- Hundreds of thousands of patents declared essential to the cellular standards, while large proportions of these are not actually essential.⁴
No more than a dozen or so patent owners have significantly or profitably operated their own licensing programs to monetize SEPs in cellular⁵, and even fewer have in other technologies such as video codecs where patent pooling has predominated for many years.⁶
While the number of jurisdictions and patent owners has not increased substantially for cellular SEPs in recent years, the number of prospective implementers is increasing enormously, as is the number of SEPs. Patent pooling enables those new implementers to be most effectively licensed by all SEP owners.
One size does not fit all implementers
Whereas cellular technologies were once all implemented by a relatively small number of OEMs and in only a few different types of device, including cellphones and PCs, in IoT there are a very wide and expanding array of applications from sensors and humble products such as meters to very complex and costly apparatus such as self-driving cars, surgical instruments and industrial equipment. Market verticals are also numerous including manufacturing, mining, agriculture, smart grids, smart cities and smart homes. Licensing across all this diversity requires wide distribution and range in licensing packages, segmented to suit different implementations, including various licensing price points reflecting the significantly different values derived from the SEPs by implementers and their customers in some applications versus others.
Benefits for all licensors
The inherently larger scale in patent pooling than in bilateral licensing makes this all more cost effective for both licensors and licensees in IoT.
Some major cellular SEP owners— including Ericsson, InterDigital, Nokia and Qualcomm—significantly monetize their patents by licensing cellular device OEMs directly. Others who are also major device implementers—including Huawei, LG and Samsung—significantly reduce or eliminate their device licensing costs through cross-licensing, also directly, with other cellular device OEMs. Many more cellular SEP owners have too few SEPs to profitably support their own cash-royalty generating licensing programs. Instead, while some license their video-codec SEPs through patent pooling, others’ cellular SEPs have largely remained dormant, for “defensive” purposes.
For decades now, cellular patent pools—including those for 3G, administered by Via Licensing and SISVEL subsidiary 3G Licensing SA, and for 4G, administered by SISVEL and Via Licensing—all failed to make any significant impact versus bilateral licensing, and, as I predicted in 2010, had no prospect of doing so.
However, the outlook for patent pooling in cellular technologies improved dramatically when Avanci entered the market in 2016 with its 2G/3G/4G licensing platform focused exclusively on IoT, with licensing and charges based on end-devices and the SEP value conferred to them. While none of the major cellular SEP owners—including all the companies named above—joined any pools that were seeking to license all types of device including mobile phones, all four of those significant SEP monetizers named above agreed to pool their cellular SEPs for IoT licensing through Avanci along with 35 other licensors. While various assessors disagreed about the relative positions of individual SEP owners, they broadly agreed that those four companies collectively accounted for most SEPs.
Patent pooling enables those companies with existing programs also to reach a far larger range of licensees than previously with their own licensing programs, and, for SEP owners without licensing programs, it enables them to be able to monetize their patents at all. The smaller licensors particularly benefit from various economies of scale and cost efficiencies, including distribution (i.e. in licensing “sales” and marketing), in operations and administration because they have lower licensing revenue potential. For example, assessing patent essentiality and relative value among SEP owners is expensive and can be contentious. Costly litigation is more likely required to ensure payment of royalties outside of pooling arrangements.
Removing roadblocks to efficient licensing
Pooling is logical, efficient, beneficial to SEP owners and implementers, and widely favored in some instances (i.e. for video codecs generally and for cellular in IoT). A recent business review letter (BRL) on the matter of Avanci’s proposed new 5G licensing platform for IoT, from the U.S. Department of Justice’s Antitrust Division, notes that the Department had ‘long recognized that patent pools can “provide procompetitive benefits by integrating complementary technologies, reducing transaction costs, clearing blocking positions, and avoiding costly infringement litigation.”’
Nevertheless, there remain impediments to SEP licensing in general that threaten to prevent patent owners reaping their just rewards and that impair potential advantages in patent pooling. While the following contentions are not all fully resolved to final appeal in all major jurisdictions, obstructions to well-established licensing and valuation methods are being removed and some trends that are hostile to SEP owners are reversing:
- End-device licensing. While SEP technologies are significantly implemented in baseband modem chips, patented cellular capabilities reach extensively beyond these components into modules and across entire devices. By licensing at the end-device level, all patents can be included in a single agreement. Nevertheless, there is significant dispute between many licensors who insist on licensing end-devices and some implementers who want to license at the chip or module level. Challenges to well-established and extensive end-product licensing practices have been made under antitrust laws, as well as in contract and patent laws. In FTC v. Qualcomm, the US Ninth Circuit Court of Appeals has recently rejected all antitrust-based claims including exclusive dealing: “Qualcomm is under no antitrust duty to license rival chip suppliers.” The aforementioned BRL also recognizes considerable efficiencies in licensing vehicles and notes that the Department of Justice’s Antitrust Guidelines state that field-of-use restrictions, such as licensing at the end-product level, can be procompetitive.⁷
- End-device-based royalty rate determination. Relatedly to the above, there has also been significant litigation to undermine the well-established and predominant method of basing royalty charges on the value in end products, and, instead, determine royalties based on the, so-called, smallest-saleable patent practicing unit. However, the Ninth Circuit Court of Appeals has also ruled in FTC v. Qualcomm that that there is no obligation for patent owners to calculate royalties on that alternative basis, even if a chip can be deemed to be the SSPPU in any implementation. ‘No court has held that the SSPPU concept is a per se rule for “reasonable royalty” calculations; instead, the concept is used as a tool in jury cases to minimize potential jury confusion when the jury is weighing complex expert testimony about patent damages.’ No jury was ever involved in FTC v. Qualcomm litigation, nor is one ever involved in determining royalties in normal business operations outside of court. The German Mannheim court in Nokia v Daimler has found that Nokia’s [end-product-based] automotive licensing offer was fair, but that neither Daimler nor its supplier Continental “were seriously willing or prepared to conclude a license agreement with the applicant on [fair and reasonable]terms.” The Department of Justice also notes in its BRL of Avanci’s proposed 5G IoT licensing program that there are a variety of ways to value the patented technology and stated that it believes parties should be given flexibility to license in a manner that best rewards and encourages innovation.⁸
The value delivered by these cellular SEP technologies substantially exceeds chip implementer design and fabrication costs in many cases. Licensing at the chip level, and calculating royalties on this basis would tend to drive royalties down to values in the most basic implementations (e.g. a “connected” lightbulb) and undervalue patents in high value end-devices (e.g. connected cars with cellular V2X capabilities). The result would be economically inefficient and dysfunctional with reduced incentives for SEP owners to contribute their technologies to the standards. Chip-based licensing would result in more uniform charging which could be too costly for low-value applications or under-compensate for inclusion of SEPs in high-value applications.
- Global licensing. While numerous technology products, including those containing cellular SEPs, are global with manufacturing and sales including many nations, and with thousands of global patent licenses agreed and underpinned by many billions of dollars of royalty payments made over many years, some implementers have held out from taking such licenses by arguing that national courts do not have extra-territorial jurisdiction. However, The UK Supreme Court’s recent appeal judgment in Unwired Planet v Huawei and Conversant v Huawei and ZTE rules that the English Court has power to grant an injunction in respect of UK national patents unless the implementer enters a global license of a multinational patent portfolio, and to determine the terms of that license. The UKSC recognizes that national courts determine validity and infringement of national patents but the ETSI IPR policy—as is applicable to cellular SEP licensing— empowers a national court to determine FRAND rates and terms. There was no UK lower-court determination of validity or infringement for foreign patents. Implementers remain free to challenge patents at the national level and seek a reduction in royalties should that be successful for significant technologies. Worldwide licensing is the norm, is supported by the ETSI IPR FRAND agreement, and is not precluded by national courts. There is no international forum for appeal, and so it is only national courts that can deal with disputes about worldwide licensing.
Extending SEP licensing reach for IoT through patent pooling
The ways in which cellular SEPs are licensed are expanding as these technologies are increasingly being implemented in many more devices than mobile handsets, tablets and PCs. While bilateral licensing is not going away any time soon, and will always remain an option due to antitrust concerns that would arise if one or even several patent pools were the only means of licensing, patent pooling or similar platform-based licensing is increasingly attractive in IoT including 5G. This is because the numbers of implementers and applications is increasing dramatically outside the cellular industry vertical, and because the value that IoT can provide significantly varies from case to case, accordingly. Patent pools have the scale to be able to offer one-stop-shopping to implementers with most or all of the SEPs they need to license, and with royalty charges that are proportionate to the value they derive from the sophisticated cellular technologies they use, such as upcoming ultra-reliable and low-latency communications in 5G.
While bilateral licensing is most likely to continue to predominate among the technology developers and implementers within the cellular industry, the Avanci licensing platform —initially focusing on connected vehicles and smart meters with charges (e.g. up to $15 per vehicle for 3G & 4G) based on value in those devices and applications—is beating a new path in cellular technology licensing. This is attractive in IoT, including 5G, in a way that has already proven successful for all devices in video codec standard technology licensing.
Endnotes
¹While it is not cost effective or necessary to patent in all jurisdictions, effective IP protection for products manufactured or sold in several or many nations also requires patenting in multiple jurisdictions including US, China, Europe and Japan.
²Net of duplicates, I estimated 265 different companies had declared patents as essential to 3GPP standards in the ETSI IPR database by May 2019. While patent counting is not an accurate or reliable of method of comparing SEPs among different patent owners, as illustrated by wide disparities in various studies’ patent essentiality assessment results, these studies far less controversially all show that most SEPs are owned by fewer than 10 companies.
³While, according to Strategy Analytics, 37 OEMs accounted for 83% of cellular handset shipments in 2018, the GSM Suppliers Association had identified, already by August 2020 (less than 18 months from 5G’s inception), 93 vendors who had announced 401 available or forthcoming 5G devices including 18 different “form factors” among these. While GSA’s figures include various end-devices, they also include intermediate products including modules and vehicle onboard units which are increasingly being incorporated in plethora of additional end-devices (e.g. cars) by many different OEMs serving their respective vertical markets. For example, licensing platform Avanci, with 39 licensors, has licensed 14 vehicle manufacturers. There are also many OEMs in other verticals, including healthcare, industrial automation and asset tracking. For example, 36,000 Bluetooth Special Interest Group members including manufacturers and others in numerous vertical sectors all cross-licence their SEPs for the Bluetooth standard under a royalty-free pooling arrangement.
⁴While 332,511 patents had been declared essential to 3GPP standards in the ETSI IPR database by September 2020, studies claim that many of these are not actually essential.
⁵A total of 56 companies, including patent pools and patent-assertion entities, were identified as prospectively generating cellular SEP royalties, with 5 “Leaders” and 16 other public companies accounting for more than 90% of these, in A New Dataset on Mobile Phone Patent License Royalties, by Haber, Galetovic and Zaretzki, 2H 2017.
⁶The vast majority of video codec SEPs are licensed to thousands of licensees through only a few patent pools including those administered by MPEG LA, HEVC Advance and Velos Media.
⁷The BRL states at page 18: ‘Here, the efficiencies from the proposed field of use appear to be considerable and are likely to outweigh the potential competitive harm caused by limiting the scope of the Standard [Patent Licensing Agreement] to connected vehicles. The Antitrust Guidelines for the Licensing of Intellectual Property make clear that field-of-use restrictions can be procompetitive because they allow the licensor “to exploit [its patents]as efficiently and effectively as possible” and that they may “increase the licensor’s incentive to license.”’ (Citations omitted.)
⁸The BRL states at page 20: ‘There are “a variety of ways” parties might value patented technology, including setting royalties based on end-product revenue. The essential cellular SEPs licensed here are subject to FRAND commitments. Avanci represents that its current rates for the 4G Platform are FRAND and reflect input from both licensors and licensees, and that Avanci intends its 5G rates also to be FRAND. There is no single correct way to calculate a reasonable royalty in the FRAND context. Each standards-essential patent holder will have to decide whether the Avanci Platform comports with its own FRAND commitments. Standards implementors can enforce the commitments in contract proceedings if there are disputes. The Department believes parties should be given flexibility to license in a manner, consistent with these commitments, that best rewards and encourages innovation.’ (Citations omitted.)
