Confusing allegations of various “behaviors” are a red herring—not evidence of anything illegal, bad-faith or discriminatory—while SEP owners earnestly attempt to obtain FRAND licensing

As previously remarked in IP Finance, I recently submitted my individual comments, and commented along with other scholars of law, business and economics, among hundreds of consultation submissions in response to the US Department of Justice’s ‘Draft Policy Statement on Licensing Negotiations and Remedies for Standard-Essential Patents Subject to Voluntary F/RAND Licensing Commitments’ (”DoJ Draft Revised Statement”).

Red herring, courtesy of Wikimedia Commons

Comments on the DoJ Draft Revised Statement by academics Christian Helmers and Brian Love misleadingly imply that many measures taken by a standard-essential patent holder in seeking to obtain a FRAND license are abusive. They state there that ’we present empirical evidence of “hold-up”—i.e., evidence of opportunistic behavior by SEP enforcers that is intended to unreasonably inflate royalties.’ However, elsewhere in a supposedly supporting research paper they co-authored and cite in their comments, they admit that their analysis of these behaviors does not constitute proof of holdup: ’while these behaviors have all been associated with holdup, we stress at the outset that many are not per se unlawful and none are, standing alone, conclusive proof of holdup.’ They ’do not claim that the presence of any of these behaviors constitutes, in itself, empirical proof of holdup.’

Helmers and Love seek to maximize scope of what is deemed opportunistic to include virtually any action an SEP holder might pursue in its legitimate attempts to obtain FRAND licensing.

While the DoJ Draft Revised Statement entirely avoids using the commonly-used terms “holdup” and “holdout,” it asserts that ‘opportunistic behavior by both parties can occur’ and that ‘opportunistic conduct by either SEP holders or implementers makes the implementation of standards more costly and deters investment in future standards development, affecting all users and producers of standardized inputs and products, including small and large firms, inventors, and consumers  (emphasis added).’

The DoJ Draft Revised Statement appears aligned with commonly asserted—but disputed—existence of holdup and interpretations of the term in the context of prospective patent licensing; but it does nothing to delineate what is legitimate and possible in the real world from purported bad behaviors by SEP owners:

‘Opportunistic conduct by SEP holders to obtain, through the threat of exclusion, higher compensation for SEPs than they would have been able to negotiate prior to standardization, can deter investment in and delay introduction of standardized products, raise prices, and ultimately harm consumers and small businesses.’[1]

For example, it is impractical for royalty rates to be negotiated ex-ante to standardization, and it rarely occurs because neither licensors nor licensees seek that in practice, and it would cause antitrust concerns if it was required in standards development.

And, the DoJ has provided no supporting evidence for those alleged harms.

Another misnomer

“Opportunistic behavior” is the new euphemism for alleged holdup or any other action by patent-owners that implementers might find unwelcome. This substitute term is so vaguely defined it only sows further confusion into the already heated debate about what is and what is not Fair, Reasonable and Non-Discriminatory (FRAND) and legal when seeking Standard-Essential Patents (SEP) licensing.

No wonder one other commenter—seemingly unaligned with either the licensor or implementer camps—was so dumfounded he wrote:

'Having encountered the SEP Policy proposal today in an ad embedded in an article by the Washington Post, I urge you to revise your regulations to be intelligible to more Americans. I cannot fathom the purpose of this mission: is it to prevent Chinese manufacturers from counterfeiting American premier brands, or is it to enrich patent holders by preserving their intellectual rights?

Having graduated from the University of California with Honors in English and having taught English in public schools for 25 years, I can merely determine that these proposed regulations address a real concern. Unfortunately, I suspect that most English speakers would disregard what they cannot possibly comprehend. The likely result is a creeping cynicism about regulation and a further contempt for regulatory compliance.

I urge you to clarify the language.'[2]

Dictionary definitions for opportunistic and opportunism all cite bad intent, including:

·        Usually disapproving: using a situation to get power or an advantage.[3]

·       Taking advantage of opportunities as they arise: such as exploiting opportunities with little regard to principle or consequences.[4]

·       To quickly take advantage of a situation, usually in a way that’s just plain wrong.[5]

·       The policy or practice, as in politics, business, or one’s personal affairs, of adapting actions, decisions, etc., to expediency or effectiveness regardless of the sacrifice of ethical principles.[6]

Filling the void

Helmers and Love brazenly exploit this chaos. In absence of identification and explanation about which “opportunistic behaviors” are illegal or unacceptably bad-faith,[7] commenting on the DoJ Draft Revised Statement became a free-for-all, including by Christian Helmers and Brian Love, citing a paper they co-authored with Yassine Lefouili. They take an “everything but the kitchen sink” approach in circumscribing the wide variety of behaviors they deem unacceptably opportunistic conduct in patent holders’ attempts to obtain FRAND licensing.

In their consultation comments, Helmers and Love state that ’we present empirical evidence of “hold-up”—i.e., evidence of opportunistic behavior by SEP enforcers that is intended to unreasonably inflate royalties’ and they also assert that ‘licensees are induced to pay royalties that exceed FRAND levels.’ To the contrary, their econometric analysis is on alleged behaviors, provides no evidence of patent holdup actually occurring, and they make no assessment of what FRAND levels are or whether amounts demanded or paid are FRAND.

Buried in their cited research paper, but not in the consultation comments, the three co-authors tellingly admit that ’while these behaviors have all been associated with holdup, we stress at the outset that many are not per se unlawful and none are, standing alone, conclusive proof of holdup.’ They ’do not claim that the presence of any of these behaviors constitutes, in itself, empirical proof of holdup.’ This is because, even in combination, these behaviors say nothing about—let alone evidence of—whether patent holdup is ever occurring.

All they have done, in their empirical research, is some elaborate and lengthy econometric analysis on a variety of legal and legitimate actions by the patent holder—most of which are no more than alleged actions—that are lawful and are only to be expected in attempts to obtain FRAND licensing.

SEPs do not license themselves and they are not self-enforcing. Patent owners invariably need to take some actions if they are ever to get paid. In goods and services markets, supplies can simply be withdrawn if customers do not pay. Supply of SEP technologies cannot be withheld.

The research paper by all three authors admits they take ‘a broad view of potential holdup behavior by SEP owner (emphasis added).’ Figure 1 repeats the list and descriptions of opportunistic behaviors alleged by Hellmers, Love and Lefouili, and the table also includes my rebuttals explaining why all of these actions are legitimate and may be necessary in pursuit of FRAND licensing. Helmers, Love and Lefouili “find evidence of opportunistic behavior by the SEP enforcer in approximately 75% of patent-party level SEP assertions.”

Figure 1: Overview of measures of alleged opportunistic conduct by SEP enforcers

Opportunistic behavior (Table 1 sequence, Table 3 numbering) [8]	Description[9](emphasis added)	Rebuttal[10]
1. Any opportunistic behavior	Did the accused infringer (specifically) allege that the patent enforcer (or its predecessor) engaged in some kind of opportunistic behavior or something that might constitute a FRAND violation?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism.
3. Untimely declaration	Did the accused infringer argue that the patent enforcer (or its predecessor) did not disclose the SEP to the SSO until after the standard was adopted?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. It is normal and expected that many SEPs are not declared until standards are set.[11] SEP owners’ declarations ensure that only FRAND royalties are paid, regardless of when disclosure is made. Most new declarations are from SEP owners that have already committed to FRAND licensing for many other patents. It is unclear which patents are SEPs until the standard is set because patent claims change in prosecution and technical specifications change until the standard is completed.
4. Overdeclaration	Did the accused infringer (specifically) allege that the patent enforcer required licensees of relevant SEPs to additionally pay royalties for rights to patents that were either not essential to the relevant standardized technology generally or not relevant to the accused infringer’s specific products?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. SEP licensing offers are invariably for portfolios comprising patents that are declared by their owners to be or might become standard essential. FRAND-licensing offers reflect, probabilistically, that some patents are not or will not become standard-essential. True essentiality can only be determined by a court, which is impractical for all patents.
6. Discriminatory license	Did the accused infringer make a specific allegation that the patent enforcer adopted discriminatory or exclusionary licensing terms or practices?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of illegitimate discrimination.
2. Injunction	In the complaint (or counterclaim) did the patent enforcer expressly request an injunction?	Seeking an injunction is a fundamental patent right. Courts have acknowledged that case-specific facts might support the issuance of an injunction. Injunctions are not automatic. Only a court can decide whether and injunction is issued.
8. Parallel ITC litigation	Did the patent enforcer initiate an investigation against the accused infringer at the ITC in parallel to the district court litigation?	Seeking an ITC exclusion order is a fundamental patent right. This right is preserved if a FRAND offer has been made for licensing SEPs that the infringer will not or cannot accept.
9. Relevant litigation abroad	Did the patent enforcer seek injunctive relief against the accused infringer in related litigation filed outside the U.S. (e.g., in Germany)?	Seeking an injunction is a fundamental right in any jurisdiction in which there is patent infringement. This right is preserved if a FRAND offer has been made for licensing SEPs that the infringer will not or cannot accept.
11. EMVR vs. SSPPU (Entire Market Value Rule vs. Smallest Salable Patent-Practicing Unit)	Did the accused infringer argue that the patent enforcer improperly attempted to base the royalty owed on the price of the end product (using the EMVR), rather than the price of a component/module (the SSPPU).	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. It is not improper to base royalty offers on EMVR. This is the way FRAND royalties are most commonly derived in commercial negotiations outside of litigation. Courts have repeatedly emphasized that `[t]here is nothing inherently wrong with using the market value of the entire product' as the royalty base for a FRAND royalty (FTC v Qualcomm, 969 F. 3d 974, 999 (9th Cir. 2020) quoting Exmark Mfg. Co. v. Briggs & Stratton Power Prods. Grp., 879 F.3d 1332, 1349 (Fed. Cir. 2018)).
5. Exhaustion	Did the accused infringer (specifically) allege that the patent was already licensed (e.g., by an upstream component supplier)?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. That the infringer disagrees with the licensor on matters of law and fact is no proof or evidence of opportunism.
7. Threats to sue customers	Did the patent enforcer bring the accused infringer’s customers into the licensing dispute, either by contacting them, threatening to sue them, or actually suing them?	SEP owners have a right to exclude that is tempered by its FRAND commitments. If a FRAND offer has been made, the SEP owner is entitled seek redress for manufacture, sale and use of any infringing products that are unlicensed. That reasonably includes contacting various parties that are knowingly or unwittingly dealing with or using such products.
12. Prior licenses not comparable	Did the accused infringer argue that the patent enforcer improperly attempted to base the royalty owed on prior licenses that were not reasonably comparable (due to differences in patents, duration, geographic scope, licensee type, etc).	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. Where there is a dispute, only a court can decide whether and which licenses are comparable, including how and which characteristics in these  should be considered.
10. No disclosure	Did the accused infringer make a specific allegation that the patent enforcer simply refused to disclose the terms of prior licenses with similarly situated companies?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. The obligation of the SEP owner is to make a FRAND offer. Prior patent licensing terms are typically confidential and are subject to confidentiality agreements. Under these conditions and in absence of a court order, SEP owners are not allowed to disclose licensing terms. However, even without such restrictions, SEP owners  are under no obligation to disclose terms of other licenses, which would likely be used as starting points for negotiation, with the infringer only being willing to pay less.
13. Other	Did the accused infringer specifically allege that some other action might constitute a FRAND violation?	Mere allegations, no matter how frequent, cannot be regarded as any kind of evidence or proof of opportunism. Undefined “other” allegations merely adds the kitchen sink to the rest of the list of behaviors including mere allegations.

 As I indicate in Figure 1, for example, given that the EMVR is legitimate and the norm in most royalty rate determinations that occur outside of litigation, it is also quite in order for parties to use that royalty base in litigation.[12] It is irrelevant whether those licensors with larger than average patent portfolios are more or less inclined to stick to their guns with EMVR-based licensing than the average patent owner.

Also in FTC v. Qualcomm, the US Court of Appeals for the 9^th Circuit noted that 'the Federal Circuit rejected the premise of the district court’s determination: that the SSPPU concept is required when calculating patent damages’ and that ‘[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.’

Similarly, it is also quite usual and a the legal entitlement of any patent owner—including one subject to FRAND commitments—to pursue parallel litigation in the ITC or a foreign jurisdiction for patents granted there. There is nothing incorrect or opportunistic about that.

Helmers and Love also misunderstand standard setting, IPR policy and how FRAND rate negotiations and agreements must be pursued. For example, ETSI IPR policy requires that standards are set purely on the basis of technical criteria, and not on the basis of expected royalty costs or ex-ante negotiated royalties. The FRAND commitment was introduced to ensure balance in these circumstances. Rather than making the licensing market fairer, ex-ante licensing is impractical and would foster monopsonic and even buyers’ cartel behavior with implementers colluding to drive down royalty rates below FRAND levels (e.g., selecting the cheapest technology in a reverse auction). It is perverse for Helmers and Love to assert that absence of ex-ante licensing rate negotiations is opportunistic, since that generally does not occur, is unwelcome be either party to licensing and is disfavored by antitrust authorities.

 Devious analysis

The analysis is a big red herring. It does nothing to establish whether there actually is any patent holdup or abuse. Rather than seeking to prove the existence of actual holdup (e.g. payment of supra-FRAND royalties or suffering from lock-in and incurring switching costs to avoid paying these) or any behavior that is in fact illegal or bad faith, Helmers, Love and Lefouili cunningly duck that question. 

While the three co-authors’ regression results including statistical significance figures from their ‘linear probability model, ’and their econometric jargon will leave most readers cold, their results also beg the question: So what?

Patent holdup is neither proven nor disproven by how various legitimate behaviors—most of which are no more than allegations—correlate with characteristics that the three co-authors also choose to identify about the parties,[13] cases[14] and patents in litigation,[15] and with case outcomes in law.[16] No court found that any patent holdup occurred in any of the cases considered by the co-authors.

 Cherry picking characteristics

In econometrics, if one gets to select and reject which among numerous possible characteristics one “considers,” one can most easily support whatever point one wishes to make.

Consultation comments by Helmers and Love on the DoJ Draft Revised Statement also cite a research paper they co-authored that purports to empirically test holdout theory with evidence from litigation of SEPs. However, in this paper dated five months later than the cited paper on patent holdup, they select a different set of characteristics to the above:

'We derive empirically testable predictions from the literature supporting hold-out theory[17]—namely that hold-out should be positively associated with the size and international breadth of licensors’ SEP portfolios, but negatively associated with the “quality” of licensors’ SEPs—and we test those predictions using measures of pre- and in-litigation hold-out constructed from information disclosed in U.S. SEP cases filed 2010-2019.'

 So, the only common characteristic between their regression results and findings on holdup versus holdout is actual or claimed portfolio size. Given that the SEP enforcer behaviors in Table 1 are based on the allegations of accused infringers, it is unremarkable that SEP enforcers were found more likely to be accused of overdeclaration in both studies.

Again, the authors’ analysis merely tests the extent to which behavior (in the case of holdout) correlates with these different characteristics. This does not test or control for other closely related characteristics—such as the size of the firm—that would likely be strongly correlated with size and international breadth of patent portfolios.

Helmers and Love speculate that if holdout theory is correct, there should be more holdout with larger SEP portfolios due to overdeclaration. That does not follow: even with overdeclaration, there will on average be more valid, essential and infringed patents in large portfolios than in small portfolios.  And, there are other characteristics that may align with large portfolios, including more established licensing, and a track record of being willing to fight costly litigation that make infringers less inclined to holdout.

Inconclusive evidence

Helmers and Love claim to provide evidence of patent holdout, but they merely regurgitate many speculative allegations about the existence of holdup, “lock-in,” royalty stacking and its effects and including harms, that have been asserted by others but have never been properly supported with applicable evidence, let alone proven in court. To the contrary, these assertions have been repeatedly debunked.

Helmers and Love conduct analysis to correlate various characteristics in SEP ownership with many different SEP owner behaviors they describe as opportunistic. But these alleged behaviors are generally legal and legitimate, and so results of their analysis provides no evidence that holdup or any resulting harm is actually occurring.

It is for the courts to decide which behaviors are legitimate and which are illegal, as they have done many times. Actions such as requests for injunctions or use of EMVR licensing are not per se illegal. Empirical assessment of these behaviors, let alone on behaviors that at are mere allegations, provides no proof or evidence of patent holdup or opportunism.

---

[1] While this soups-up a legal dictionary definition with the DoJ Draft Revised Statement’s references to the threat of exclusion and hypothetical ex-ante royalty rate negotiations, it omits the explicit requirement for deception or ambush that is included in economists’ traditional definitions of holdout. ‘Oliver Williamson famously described opportunism as ‘self-interest seeking with guile.’… [Standards hold-up involves] deceiving buyers or keeping them in the dark about the terms on which a technology will be available [which] subverts the competitive process.’

[2] Comment by John Dodd,  January 20, 2022. https://www.regulations.gov/comment/ATR-2021-0001-004

[3] Cambridge Dictionary: https://dictionary.cambridge.org/dictionary/english/opportunistic

[4] Merriam-Webster: https://www.merriam-webster.com/dictionary/opportunistic

[5] Vocabluary.com https://www.vocabulary.com/dictionary/opportunistic

[6] Dictionary.com (opportunism) https://www.dictionary.com/browse/opportunism

[7] As set out in patent law and as determined in court precedents, not merely on the basis of politically-driven policy statements.

[8] As defined and alleged by Helmers, Lefouili and Love.

[9] As described by Helmers, Lefouili and Love.

[10] By Keith Mallinson, WiseHarbor.

[11] The legal position on “late disclosures” taken by multiple implementers in litigation is shown to be at odds with industry expectation and against industry practice. On the Timing of ETSI Disclosures, by Gustav Brismark, January 2021. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3988411

[12] Licensees as well as licensors have agreed to EMVR in litigation (e.g., Unwired Planet v. Huawei, and TCL v. Ericsson), where the dispute was about the royalty rate, not the royalty base.

[13] For example, whether plaintiff is a non-practicing entity, and plaintiff’s portfolio size.

[14] For example, declaratory action or motion to dismiss.

[15] For example, SEP in pool or patent reassigned.

[16] For example, plaintiff win or settlement.

[17] In absence of a citation, it is unclear which literature they are referring to for the three predictions.

Right-pricing cellular patent licensing in 4G and 5G connected vehicles

Bountiful connectivity apps in vehicles

While litigation is bogging down the licensing of cellular standard essential patents (SEPs) in vehicles with disputes about where in the production supply chain licensing may or must occur—from chip, to module, to telematic control unit (TCU), to entire vehicle—this is also delaying payment of Fair, Reasonable and Non-Discriminatory (FRAND) royalty charges in these cases and causing confusion about licensing value. This is a pity because clarity is in everyone’s urgent interest.

Well-established mobile phone licensing benchmarks conservatively imply a total value of at least around $30 per vehicle for patents essential to the 2G, 3G and 4G standards.

Great expectations for IoT hinge on cellular technology

There is strong consensus and enthusiasm in government, business and among commentators about the Internet of Things, with its multi-trillion dollar market potential. While financial and other benefits will be reaped by many vendors and users in various different industries, as well as by consumers, several generations of patented technologies developed largely by companies within the telecommunications industry over many years—up to and including newly introduced 5G—are enabling this major opportunity.

Despite cellular technologies being developed by and hitherto implemented largely among a relatively limited group of telecommunications industry OEMs producing cellular products—most significantly mobile phones as well as mobile network equipment—the variety and numbers of prospective technology implementers in IoT are far greater. While SEP licensing is well established for mobile phones and base stations—with thousands of agreements since the 1990s worth many billions of dollars every year—the industry is still in the throes of establishing the basis and pricing for use of these technologies in various different IoT applications including cars, domestic appliances, industrial robots and remote meters.

This article conservatively estimates total FRAND charges for licensing all cellular SEPs in vehicles, based on value derived therefrom and reflecting some recent court judgements on FRAND charges in other devices including smartphones. 

How to charge?

Since the early days of the 2G mobile phone industry, SEP owners most often licensed their cellular patents at royalty rates calculated as a percentage of phones’ average wholesale (i.e. unsubsidized) selling prices. One reason for this is that OEMs anticipated the subsequent downward trend in mobile phone prices, which fell dramatically following the introduction of digital cellular with 2G in the early 1990s. OEMs did not want to be locked into fixed dollar-per-unit (“DPU”) royalty charges that would increase in percentage terms as manufacturing costs were rapidly declining.

While royalties for 2G/3G/4G cellular connectivity in a mobile phone have usually continued to be charged as a percentage of the end-product selling price, the value established there—when stated as an equivalent DPU figure—is a key consideration. As average mobile phone prices increased with the widespread adoption of 3G smartphones from the late 2000s and 4G smartphones several years later, SEP licensors have, in many cases, at the behest of OEMs, “capped” percentage-based royalties to maximum DPU figures to ensure royalties paid do not exceed the value of additional features deemed less dependent on cellular connectivity. 

Similarly, DPU pricing is also applicable for other cellular-enabled “devices” including, for example, PCs and connected vehicles. There are also bountiful ways in which connectivity is exploited in these with various applications. However, a vehicle OEM, for example, would quite reasonably refuse to pay royalties for cellular SEPs that are calculated as a percentage of a vehicle’s cost or value in alloy wheels or leather seats. 

I have argued for many years against the proffered valuation methodology of basing royalties on a percentage of the sales price of a component or “smallest-saleable patent practicing unit (SSPPU)” and this approach has been soundly rejected by US and European courts.  The US Ninth Circuit Court of Appeals has ruled in Federal Trade Commission v. Qualcomm that “the district court’s analysis [relying on an SSPPU approach]is still fundamentally flawed. No court has held that the SSPPU concept is a per se rule for “reasonable royalty” calculations . . . .” Similarly, in Germany in Nokia v Daimler, the Mannheim court stated that the “royalty provided in [Daimler’s] counter-offer is not reasonable, as the reference value used in the top-down approach in the form of the average purchase price of [TCUs] is unsuitable. This reference value prevents [Nokia] from participating adequately . . . in use of the technology in the saleable end product.” (Unofficial translation.)

I am not commenting here on how aggregate royalties can or should be apportioned among SEP owners. Elsewhere, I have commented on the inaccuracies and other shortcoming in apportioning royalties based on the counts of declared-essential or judged-essential patents. 

Where to license?

While I and many others have also long argued it is also rather simpler and more efficient to license at the entire device level—as has always been the case in mobile phones—the Court of Appeals additionally ruled in the above that it is the patent licensor’s prerogative to license where it wishes. 

As discussed below, the value of cellular functionality to a connected vehicle is at least around $30 per unit, regardless of where licensing occurs in the production supply chain, and irrespective of the different formulae that could be used to calculate that figure with licensing at different stages in that supply chain. 

Valuation benchmarks

While there has never been consensus in the telecommunications industry that aggregate royalties for SEPs should be capped—with significant dissent by various licensors including Qualcomm—maximum aggregate figures proposed by some leading companies that declare many patents essential to cellular standards—when correctly interpreted and applied—provide at least some conservative valuation benchmarks. Court determinations of FRAND royalty rates for individual licensors—also as percentages of unsubsidized wholesale handset prices—in a few different cases have been based upon or cross-checked using such aggregate figures:

• 6% to 10% for LTE by Judge Selna in TCL v. Ericsson, which was subsequently overturned on appeal;
• 8.8% for LTE (i.e. 4G) by Justice Birss in Unwired Planet v. Huawei;
• 5% for 3G and 6% to 8% for 4G by the Shenzhen Intermediate Court in Huawei v. Samsung.

However, adjustments to the above are warranted because some source figures have been misinterpreted and incorrectly applied or alternative figures could have been reasonably selected as aggregate royalties in determining FRAND rates for the parties’ portfolios. 

Prior to Judge Selna’s judgement being entirely vacated on appeal, I showed he had muddled single-mode and multi-mode licensing rates in pages 5 to 7 of my critique of his “top-down” SEP royalty rate valuation analysis. As LTE was being first standardized in 2008, patent owner announcements from April that year proposed individual and aggregate single-mode LTE royalty rates. This was for like-for-like comparisons with claims of ”less onerous” licensing for rival 4G technology WiMAX at “much lower” rates and with patent pooling at a “predictable cost”. Only a couple of companies also mentioned their proposed multi-mode rates. It is only since then that Apple’s iPhones and Android-based smartphones have been multi-mode devices needing licensing of more than one generation of technology. The first of these smartphones, including even 3G, was not introduced until the second half of 2008. The aggregate rates Judge Selna used in deriving an aggregate FRAND rate of 6% to 8% (his judgement also cites a figure “not higher than 10%”), reflected only the value in LTE and not that in 2G and 3G. The correct figure for LTE handsets (i.e. multimode devices) with his methodology should, therefore, have been 11% to 15%, including an additional 5% for 3G, and conceivably more for the inclusion of 2G. 

Justice Birss also uses the “total royalty burden” in his FRAND rate determinations. He indicates, for a 4G multimode handset, “the aggregate implied by either party’s case (Huawei’s 13.3% and Unwired Planet’s 10.4%).” The average of these two figures is 11.9%. 

According to Strategy Analytics, the global wholesale average selling prices for LTE handsets (i.e. overwhelmingly multi-mode including 2G, 3G and 4G standards) were $270 in both 2018 and 2019. That equates to $29.70 to $40.50 per handset at multi-mode royalty rates of 11% and 15%, respectively.

While cellular SEP licensing revenues for Ericsson, InterDigital, Nokia and Qualcomm alone amount to many billions of dollars per year, that is overwhelmingly from mobile phone licensing with revenues understating value in cross licensing among these and other companies. For example, as Ericsson and Nokia used to have large handset device operations and still have major cellular network equipment businesses, licensing fees paid in cash among those and many other cellular industry companies significantly reflect netting off rather higher nominal charges. Major implementers—including Apple, Huawei, LG and Samsung with substantial market shares of device sales in recent years—tend to generate little or nothing in cash royalties for SEP licensing while they seek to minimize license fee outpayments through cross licensing. 

Licensing fees paid also understate value because many OEMs have remained unlicensed due to free-riding with patent “hold-out” and because some OEMs do not have licensing programs but own patents for defensive purposes. 

SEP value in vehicles versus smartphones

The value of SEP technology to vehicles is provided in various ways and applications to manufacturers, consumers and vehicle fleet operators. In some respects, this value exceeds the value that the same technology confers on a smartphones. As well as enabling in-vehicle information and entertainment systems, cellular technology:

• Connects all of a car’s occupants concurrently, while smartphones tend to be used by only one person;
• Enables remote vehicle diagnostics for maintenance, asset management tracking and route management in trucks; 
• Improves vehicle safety with C-V2X, for example, with collision avoidance alerts introduced in 4G: thus saving lives by reducing the numbers of millions dying and many more suffering from serious accidents on the roads worldwide each year; and
• Can continuously connect various third parties, including the vehicle OEM, insurance providers and fleet management service providers.

The value derived from the one-off licensing charges is also elevated in connected vehicles because these have longer working lives than smartphones.  Cars, for example, typically have 14-year lifespans before scrappage, versus seven years for mobile phones, while users in developed countries replace their phones about every 18 months.

The DPU value of cellular SEPs in vehicles is, therefore, at least comparable to that in smartphones.

Even more than a big smartphone on wheels

While there will continue to be a large proportion of costs and value in vehicles that has nothing to do with cellular capabilities, the proportion of that in information and communications technologies—significantly including cellular connectivity—is large and growing rapidly. As defined by industry analyst Markets and Markets, the global connected car market is expected to be worth $54 billion in 2020 and is projected to reach $166 billion by 2025—a compound annual growth rate of 25%. With sales of around 70 million light vehicles per year, that amounts to $600 per vehicle in 2020 rising to $2,400 per vehicle in 2025. It believes the connected truck market is also worth tens of billions of dollars per year. In addition, Markets and Markets circumscribes a separate global in-vehicle infotainment market which it projects to grow from $24.3 billion in 2019 to $54.8 billion by 2027—a compound annual growth rate of 10.7%.  Research shows that car manufacturers charge consumers from several hundred dollars to many thousands of dollars for connected car application “packages.”

In consideration of all the above and the “maximum aggregate rates” relied upon by the judges, as discussed above, an aggregate SEP value of $30 to $40 per smartphones is also reasonably applicable per connected vehicle for multimode 2G/3G/4G licensing. While DPU royalties are explicitly not derived as a percentage of a vehicle’s cost or price, it is notably that the above figures correspond to less than 0.1 % of average selling prices for cars—at $37,800 in the US and $27,400 globally— two orders of magnitude higher than for LTE smartphones at $270 over the last couple of years. 

The future in 5G

As indicated above, the connected car market is expected to quadruple in size over the next five years, with additional growth in adjacent markets that are also dependent on cellular connectivity. As well as buoying average prices and stimulating new vehicle sales volumes, connected vehicle capabilities in cars and trucks—with Markets and Markets’ market definition, or with my broader market definition—will inevitably provide among the best opportunities for vehicle OEMs to differentiate their products and bolster profit margins. For example, capabilities including C-V2X are being enhanced in 5G over what is possible in 4G, with improvements such as enhanced positioning to enable increasingly autonomous and even self-driving vehicles. While market definitions include the cost or price of tech hardware and software, utility and value to consumers will grow as autonomous capabilities—provided by C-V2X, sensors and AI—save lives while relieving occupants from driving and enabling them to work, relax or sleep. 

While cost and value to manufacturers and consumers in connected vehicles is almost entirely still in 2G, 3G and 4G today, this will increasingly be in 5G with it expecting to dominate the flow of gross additional cellular connections (a leading indicator) and account for 31 percent of all established connections worldwide by 2025. That justifies significant additional royalties for 5G in vehicles, as some cellular SEP owners are already obtaining through the licensing of 5G smartphones and other devices.

One-stop-shopping is best in IoT

While bilateral licensing is possible in IoT including connected vehicles—as it is in mobile phones—the reduced transaction costs and other benefits inherent in platform-based licensing or patent pooling is highly attractive to both licensors and licensees in IoT, as I wrote in my previous article here very recently. While all the major cellular SEP owners have preferred to license bilaterally to the relatively small number of handset OEMs, most prefer now to license these SEPs into the numerous different vertical sectors in IoT through a platform or pool. For example, while there are differences in analysis and opinion about exactly what proportion of cellular SEPs Avanci represents, there is broad agreement that it, with its 39 licensors, has most of them. Avanci licenses all its 3G and 4G SEPs for $15 per connected vehicle—the price of a car wash—regardless of how many TCUs, modules or modem chips the vehicle contains.

———————————————————————————————————————————————

A similar article to this was originally published in RCR Wireless.

Keith Mallinson is a leading industry analyst, commercial consultant and testifying expert witness. Solving business problems in wireless and mobile communications, he founded consulting firm WiseHarbor in 2007.