The licensing of SEPs for cellular mobile technology is a global matter — standards, technologies, and the scope of many licensees’ operations are global or at least multi-national. Unfortunately, there is no unified global adjudicatory system that would enforce the rights of IP owners when licensing negotiations fail. Patent owners must enforce their patent rights by litigating over specific patents in specific jurisdictions. This raises the costs of enforcement and favors hold-out strategies by implementers. The Unwired Planet series of judgments upholding UK courts’ ability to determine FRAND terms for global licenses, offers some hope that courts could provide something close to the unitary global adjudicatory mechanism that matches the global nature of the technology market. However, some implementers have reacted to this development by submitting themselves to injunctions on their relatively modest UK sales in order to avoid taking a license on court determined FRAND terms, and have argued that they are then at best liable for damages on past sales in the UK only. If the courts agree that damages should be metered only on UK past sales and at a rate close to the FRAND rate for global licenses, this will give implementers strong incentives to engage in such strategies.
By Kalyan Dasgupta & David J. Teece1
I. INTRODUCTION
For more than two decades, the system of licensing standards-essential technology pertaining to mobile cellular devices has skirted around the awkward fact that the business and the technology are global in scope, but the only way for IP rights-holders to enforce their rights is through national courts. A typical pattern observed in the industry is that licensees2 continue to use an individual licensor’s portfolio of standards-essential patents without payment, until the licensor puts them on notice. Ultimately, the threat of litigation is a critical cudgel that licensors can use to guide licensees towards a negotiated settlement. However, putative licensees can potentially pursue a “divide and conquer” strategy in litigation — i.e. use the limitations of national patent law to restrict the scope of the payments they make to licensees by attempting to tie court awards to national patents. If successful, licensees can force licensors into pursuing numerous litigations in numerous jurisdictions, with each litigation involving significant costs and risks to the licensor (which is the party on which the burden of pursuing enforcement falls).
This is, of course, a highly inefficient outcome. Where licenses are willingly agreed, in good faith, they are almost inevitably global or at least multi-territorial.3 Both licensee and licensor will prefer global licenses, as these are better for both parties. Put differently, attempts to goad national courts into viewing licensing issues — which include awards for past use as well as a license for ongoing use — through the prism of patents in suit in a particular country are purely the products of litigation strategy. One would never expect to see such licenses arise organically. Issues such as licenses for single territories, or licenses limited to just patents-in-suit, or damages awards for past infringement based only on sales made in a particular jurisdiction only arise because no multi-jurisdictional enforcement institution exists that reflects the inextricably global nature of the technology market.
Recent decisions in the UK courts4 appear to offer something of a solution. The UK courts have recognized that the FRAND commitment that guides the licensing of mobile SEPs is an internationally effective one. Thus, UK courts can adjudicate the terms of a global license pursuant to the FRAND commitment by virtue of the fact that UK patents are among those included in the license. The UK courts have also taken, in our view, a highly balanced position on the determination of FRAND terms. These decisions offer hope that the UK courts can serve as an avenue through which consistent, balanced decisions on global FRAND licenses can be reached. We believe that this is important to maintaining the credibility of FRAND licensing and to protecting the generation and adoption of new technologies.
However, the UK’s status as a venue for FRAND litigation suffers from the fact that the UK is not necessarily a large market for many licensees. In particular, recent cases show that it may be attractive for many licensees to submit to an injunction in the UK and forego future UK sales. In this event, licensees have argued that they are only liable for damages on past UK sales. If courts permit this strategy, then it creates an obvious attraction for licensees to evade the UK courts’ power to adjudicate global FRAND licenses — the attraction is measured by the potentially enormous difference between paying up for a global license (including many years of past use on a global basis) and paying only for past sales in the UK. Thus, the proper scope and basis of damages — whether these should be awarded on UK sales or global sales, or (alternatively) whether they should be at all constrained by FRAND — is a critical one for courts to think through. Two recent summary judgment rulings in the UK courts appear to point in potentially different directions on this issue.
From an economic viewpoint, it would be unfortunate if licensees were allowed to drastically reduce what they pay for their past use of a SEP portfolio based primarily on whether their forward-looking UK sales were low enough that they could afford to take a UK injunction instead of a global license. This would resurrect the inefficiency created by the “divide and conquer” tactics discussed above. We also believe that the frictions, uncertainties and costs of the enforcement process — which we think fall asymmetrically on licensors who cannot exclude licensees from using their patents absent the threat of enforcement action — should be more fully factored into the analysis of comparable licenses. Observed rates in such licenses may reflect the bargaining power that licensees have via the threat of raising the costs and risks of enforcement by licensors, including via exploiting the global fragmentation of enforcement.
II. A WELCOME STEP TOWARDS ALIGNING THE JURISPRUDENCE WITH ECONOMIC THINKING
In October 2020, the UK Supreme Court affirmed5 the decision of the Court of Appeals to affirm the decision of Mr. Justice Birss in Huawei versus Unwired Planet. (“Unwired Planet.”)6 After all of this lengthy litigation, Birss J’s original judgment7 has stood up remarkably well, with all of its fundamental aspects surviving prolonged review by the higher courts. The most commented-on aspect of Birrs’ judgement was that an English court could set rates for a global FRAND license. On the face of it, this is a significant step in aligning the jurisprudence with economic reality and the economic policy-making inherent in the FRAND commitment. As Birss J recognized in Unwired Planet, licensing of SEPs is inherently global in its nature, even though the mechanisms for enforcing patent rights are resolutely national.8
A. The Importance of Recognizing the Reality of Global Licensing
Birss J and the justices involved with the subsequent appeals have recognized the enormous inefficiencies in patent-by-patent, country-by-country licensing. They have also recognized that no “market” for such patent-by-patent or country-by-country license exists. As we pointed out in our introduction to this article, the series of judgments by the British courts correct an important distortion that exists in the current global landscape for licensing SEP patents. In an idealized world, responsible licensees would seek to negotiate licenses with at least those licensors whose portfolios were well-established (e.g. because they had been litigated or because they had already consummated several existing portfolio licenses with important industry actors) and which would almost certainly contain at least some valid and infringed patents. Further, broad-scope licenses (e.g. global or significantly multi-territorial and multi-standard in their nature) carry obvious efficiencies for licensor and licensee alike, not just in terms of reducing the transactions costs of licensing, but also in providing freedom to operate for the licensee.
In the real world, however, licensees know that patents are not self-enforcing and might try to place the burden of enforcing patent rights upon the licensor. The process of enforcement is itself costly and may create asymmetric risks for the licensor. Forcing licensors to litigate on a country-by-country (and patent-by-patent) basis is an obvious way to raise the patent owners’ enforcement costs. In turn, the additional costs and risks created by this much more arduous enforcement process will reduce the attractiveness of litigation for the licensor and might reduce its bargaining power in negotiations with licensees.9 The potential availability of “one-stop shopping” for a global license thus corrects this imbalance in bargaining power and reduces the potential inefficiencies and rent-seeking opportunities created by the threat that courts will limit themselves to country-specific licenses. The UK judgments in the wake of Unwired Planet are an important step in recognizing that disputes around the value of SEP portfolios are disputes about globally-relevant portfolios of patents, not disputes about specific national patents (even if these specific national patents are what are “in suit.”)
B. The Importance of Recognizing that FRAND Addresses Hold-up and Hold-out
There are other aspects of the Unwired Planet suite of decisions that might be seen as “pro-licensor.” These are, in our view, pro-licensor only to the extent that they mitigate imbalances that had become embedded in conventional perspectives on SEP licensing (which to a substantial degree stem from the failure to recognize that a major source of bargaining power for licensees of SEPs is the difficulty and effort of enforcing patent rights in the first place).10 The UK court decisions appear to pay about as much attention to the dangers of hold-out or reverse hold-up of licensors by licensees as they do to the possibility of hold-up of licensees by licensors.11 FRAND rates are (implicitly) defined in reference to the objectives of avoiding both hold-up and hold-out. This appears to comport with the objectives set out in ETSI Intellectual Property Rights (“IPR”) policy, which emphasizes that the objectives of the IPR policy are to ensure a balance of interests between downstream users and upstream implementers. 12 ETSI was rightfully focused on advancing a robust telecommunications ecosystem that supported both the generation and utilization of advanced mobile technologies.
The Unwired Planet judgments also introduce the concept of a benchmark FRAND rate.13 In economic terms, this benchmark FRAND rate appears to correspond to the incremental surplus that the licensor’s technology creates for the licensee, including value that is created by virtue of its inclusion in the standard. The Court of Appeal and Supreme Court judgments allow for the possibility that there are potentially many divisions of this incremental surplus that are consistent with a “FRAND”-like balance of interests between licensors and licensees, i.e. FRAND is in a range. The FRAND benchmark rate appears to be the upper bound of the FRAND range. The licensor has an obligation to make licenses available on FRAND terms, but this obligation is met so long as the licensor is prepared to offer at least the FRAND benchmark rate to the licensee. What the judgments collectively make clear, however, is that the licensor has no obligation to offer a “best price” to all licensees, and differences in pricing and other terms between different licensees are only discriminatory if they have the effect of distorting competition between those different licensees (i.e. the “ND” prong of FRAND is interpreted in a fashion similar to “discrimination” in competition cases, i.e. differences in terms offered to “similarly situated” firms are of concern if and only if they are capable of creating competitive distortions in the downstream market).14
This approach has significant practical merit given that the analysis of comparable real-world licenses is the primary basis that the courts use for identifying FRAND rates. In the real world, licensors might be prepared to use the FRAND benchmark rate as a focal point, but also offer discounts to cooperative licensees in order to achieve a license agreement in a timely manner. These willing licensee discounts can take many forms — for example, licensors can agree to meter the licensee’s past sales at a lower rate than ongoing sales; or they can be similar to volume discounts wherein the efficiencies associated with receiving large cash injections in the form of large lump-sum payments are recognized. These discounts made to willing licensees can be seen as providing incentives for reasonable and cooperative behavior. However, the FRAND benchmark rate approach recognizes that fidelity to FRAND principles does not require extending these discounts to licensees who do not take a license in a timely manner and whose actions trigger litigation,15 and who thus may not have acted in ways consistent with those expected of a “willing licensee.” In fact, creating a wedge between court-awarded rates and lower rates achieved in willingly negotiated licenses avoids the risk that recalcitrant or plainly unwilling licensees will be put on the same footing as willing licensees. This wedge is consistent with a broader policy objective of ensuring that participants in the mobile ecosystem are not unduly incentivized to litigate instead of negotiating; and that potential licensees are not rewarded for dilatory conduct. Put differently, it helps keep license negotiations out of the courthouse and in the marketplace.
In summary, there are important ways in which the series of judgments handed down by British courts align case law with good global public policy. The mobile wireless ecosystem is global in nature; licenses are global in nature; the sales and manufacturing activities of many implementers are global or at least multi-territorial in nature; and global licenses offer significant benefits in terms of “freedom to operate” for licensees.
Besides offering licensors the chance to litigate in a sophisticated jurisdiction that offers “one-stop shopping” for a global license, the Unwired Planet rulings also make other important contributions, e.g. the FRAND benchmark rate concept and the clarity offered on the “ND” (non-discrimination) prong of FRAND. These contributions also help align the case law with the policy objectives inherent in standardization (and reflected in the ETSI IPR Policy). Much as courts may be reluctant to act as policy makers, the interpretation of FRAND invariably requires some consideration of the policy objectives of standards organizations such as ETSI.
C. Licensees Can Still Exploit the Global Fragmentation of Patent Enforcement
The above indicates that the UK will emerge as a natural venue for FRAND litigation, especially as many will interpret the stance of the UK courts thus far as amenable to the open innovation paradigm that has supported technology development and diffusion in mobile telecommunications. In our view, this would be a quite welcome development. However, it is also our view that the positive aspects of these decisions risk are significantly limited by (a) the fact that the UK’s standing as a venue for FRAND jurisdiction is not matched by its commercial importance to licensees; and (b) there is still uncertainty about how the now-accepted concepts of global licenses will interact with the narrower frame of UK damages law. We explain our concerns below.
The UK may not be a commercially significant market for many implementers. Further, many implementers use standards-essential technology for many years before the licensor pursues enforcement action, of which litigation is the last stage. In our experience, the licensee may sometimes have as much as a decade’s worth of infringing sales before a trial on FRAND terms commences. The upshot is that by the time UK courts are adjudicating the award of a FRAND license, future sales may be relatively small compared to past sales. Further, future sales in the UK may be particularly small compared to global past sales, and past UK sales may only ever have comprised a relatively small fraction of overall sales.
Furthermore, the relief offered by UK courts takes the form of what some call a “FRAND injunction,” in which the alternatives are for the licensee to take a license on terms that the Court considers FRAND, or to submit to an injunction. The FRAND license may be accompanied by an award for past damages, just as a willingly negotiated commercial license will contain payment for past use. This, naturally, would be based on a global FRAND rate metered against global sales.
Faced with these facts and these alternatives, some licensees may find it attractive to accept an injunction on their future UK sales rather than accept a FRAND award on all their global past and future sales. Licensees’ ability to exercise this option also diminishes any bargaining power that the availability of injunctive relief confers on the licensee in negotiations that may continue in parallel to litigation proceedings. In this circumstance, the only relief that is available to the licensor is damages on past sales. This raises three questions of importance to economics and public policy:
- Injunctions and damages awards are linked to infringement of UK patents. So, should damages awards by UK courts be based solely on sales in the UK?
- What is the relationship between the FRAND rate that might have been considered in a global license and the damages rate applied to UK only sales?
- More fundamentally, once a licensor has elected for an injunction rather than taking a FRAND license, is FRAND still an operative issue in the determination of damages?
III. NATIONAL PATENT LAW AND THE INTERNATIONAL FRAND AGREEMENT: THE IMPORTANCE OF CONSISTENCY
A. The Economic Perspective: Why Are Courts Involved at All?
For economists, the questions raised above are policy questions, and the “right” answers to these questions are determined by reference to effects on the licensing system and thereby on innovation in both upstream and downstream markets. From an economic perspective, desirable outcomes would involve a FRAND-compatible division of surplus between licensor and licensee, which ideally would be determined in negotiations between willing licensors and willing licensees. Where such negotiations break down — and bargaining theory suggests that they sometimes will do — there would ideally be an adjudicatory mechanism that replicates the economically desirable outcome (i.e. one that promotes innovation, adoption and competition).
The adjudicatory mechanism — if it is aligned to broader economically policy objectives — would recognize that good policy includes providing the right behavioral signals to participants in the licensing market. Unfortunately, no unitary adjudicatory mechanism exists, as the relevant patent rights are national in scope and licensing disputes are necessarily resolved though the complicated mechanism of proving validity and infringement, albeit on a small subset of patents in a given national jurisdiction. The mechanism forces even licensors with significant portfolios that have been licensed several times before to survive validity and infringement challenges on a small and specific subset of patents — even though the probability that the portfolio as a whole contains several valid and infringed patents may be close to one.16 The prospect of the licensor having to litigate in multiple jurisdictions only adds to the inefficiencies, uncertainties and transactions costs associated with the lack of a unitary enforcement mechanism for what are really globally relevant patent portfolios, as opposed to individual national patents.
From an economic perspective, then, the UK court decisions are highly welcome in that they offer something of a substitute to an idealized enforcement mechanism.17 However, the UK courts have not yet articulated what is an essential complement — at least from the economic policy perspective that we adopt in this article — to their mandate to adjudicate the terms of global FRAND licenses.
B. Damages Awards and License Awards Should be Made on a Consistent Basis
From an economic efficiency perspective, it would be unfortunate if courts gave licensees the option to evade global licenses (and payments for past infringement on a global basis) by electing to suffer a UK injunction and paying damages only on past UK sales. Doing so would invite infringement and hold-out — after all, the worst outcome that a licensee would face would be to pay damages on a small fraction of its global past sales. The licensor would then have to adopt the difficult course of seeking out litigation in multiple jurisdictions—each of which would be time-consuming, would carry risks in terms of re-establishing the validity of patents, and risks in inconsistent approaches to FRAND by different courts in different jurisdictions. In this context, hold-out would always be the most profitable strategy for some licensees. Of course, some licensees who make very substantial sales in the UK would not find it attractive to elect for an injunction — but this would create a wedge between licensees with high exposure to the UK market and licensees with low enough exposure that the strategic calculus would be to elect for an injunction instead of a license. Further, licensees could choose to limit their exposure to the UK precisely for the purpose of limiting their royalty burden. Thus, UK consumers might suffer a more limited choice of devices as a result of licensees having a relatively economically attractive option of submitting to a UK injunction.
From an economic policy perspective, damages and license awards are inherently linked. Indeed, payment for past use — which is economically similar to reasonable royalty damages — might be seen as a component of the overall license award. In fact, in the many negotiated licenses that involve only the payment of a single lump-sum by the licensee, past use and ongoing royalties are inseparable. Both play an important role in ensuring that the balancing of incentives between upstream innovators and downstream users is maintained. From this perspective, creating a wedge between the geographic scope of “pure” damages awards (when all the UK use is in the past) and license awards is problematic. For example, consider a licensee that has already made sales of roughly 40m infringing units globally, and 5m infringing units in the UK. Why should the compensation for this past use depend so radically — by 35m units in this example — on the level of the licensee’s future UK sales?18
C. Limits to the Applicability of FRAND
The discussion above assumes that damages for UK-only damages will be awarded at the same FRAND rate that would be used to determine a global license. However, as the second and third questions we pose above imply, one could also consider whether FRAND principles should apply at all when a licensee opts for an injunction for the specific purpose of putting barriers in the licensor’s pursuit of a global license on FRAND terms. Indeed, we would argue that the very act of accepting an injunction instead of submitting to a court-adjudicated global license on FRAND terms19 suggests that the party is not willing to pay for a global license on FRAND terms and should not benefit from the availability of the FRAND commitment.
The English courts have framed FRAND in balanced terms, and not as a one-sided commitment designed to favor licensees. In this vein, allowing licensees to effectively select the royalty base that they are liable for — which would be the effective result of making the option of submitting to an injunction economically attractive — would go against the balanced reasoning of the UK courts thus far. It would also go against the policy objectives inherent in the ETSI IPR policy. From a policy perspective, at least, it would seem appropriate to prevent licensees from undermining the global FRAND licensing regime by preventing licensees from invoking it for the express purpose of avoiding taking a FRAND license.
In this circumstance, the Courts can restore the FRAND balance by either allowing the licensor to claim damages on global sales, or by considering damages from the perspective of a commercial negotiation. In the first case, one can justify this from the perspective of a FRAND license that willing licensor and willing licensee would have agreed ex-ante, just before the licensee began making use of the licensor’s SEP portfolio. The license agreement forged in such a hypothetical negotiation would almost certainly have been a global or multi-territorial one. Alternatively, one can also consider the terms of a hypothetical territorially-specific license, recognizing that a licensor (at least) would never willingly enter into a negotiation on a territorially-specific license constrained by the global FRAND rate that it would have offered in the typical global FRAND license. In fact, if the licensor was prepared to have made a FRAND license available on global terms, but the licensee insisted on negotiating on a territorially-specific basis, then the licensor would not be violating its FRAND commitment if it offered only a global FRAND license but was prepared to negotiate a territorially-specific license on different terms, unconstrained by FRAND.
In this latter case, the licensor would have been able to enjoin the licensee’s entire future flow of profits from UK sales, and thus bargaining would occur in the shadow of this threat point. Logically, one would expect the licensor to offer and also achieve rates significantly higher than the rate it would have achieved in a global license. While these uplifted rates might seem “supra-FRAND,” they would in fact be consistent with striking a FRAND balance between licensor and licensee, as agreeing to such a territorially specific license would be a very inefficient alternative to a global license. Of course, it seems very unlikely that a willing licensee would ever, in the real world, ask for such a license. If courts want to entertain the notion of a hypothetical negotiation over a UK-only license, then, the appropriate negotiation to model would be one that is not negotiated under the FRAND constraint. One can expect such a negotiation to yield a damages rate that is much higher (perhaps several multiples of) the rate that would have been agreed in a global FRAND license.20
IV. THE CURRENT LEGAL RULINGS IN THE UK ON DAMAGES FOR SEPS
The issues we address above have been raised in actual UK litigation around SEP licensing. In Philips v. AsusTek, Marcus Smith J considered a request for a summary ruling by ASUS that in the event of ASUS submitting to an injunction in the UK, it should only be liable for damages on past UK sales, at the rate that would apply in a global FRAND license. Marcus Smith J declined ASUS’ request, as a matter that was too complex for summary judgment purposes.21 In IPCom v. HTC, however, Birss J agreed to strike out a portion of IPCom’s case that claimed damages for past infringement on a worldwide basis. He limited the scope of damages to only infringing devices that were imported into the UK. In the latter case, HTC had already submitted to an injunction on UK sales and IPCom’s claim was only for damages, not for a FRAND license determined by the court. Birss J ruled that damages must be caused by the infringing actions of the user (HTC). In his view, the infringement of UK patents did not cause the sales of HTC devices in other parts of the world, and so did not fall within the scope of a UK award of damages.22 IPCom has been given permission to appeal the ruling.
Birss J also stated that back damages metered at a FRAND rate and awarded in the context of a FRAND license award might appropriately be global in scope. He suggests that this is because the issue of a global FRAND license (in which payment for past use is bound up) inherently involves consideration of the interaction between the internationally effective FRAND undertaking. It is not completely clear as to whether Birss J has eliminated the possibility that past damages could be assessed on a global basis in the circumstance in which the licensor had asked the court to adjudicate on the terms of a FRAND license, but the licensee had instead opted to simply stop selling phones in the UK to avoid a license.23,24
Regardless, the reality is that HTC will (if Birss J’s judgment is upheld by the higher courts) end up paying perhaps 1/100th of what it would have paid by way of payment for past use alone, simply because its current commercial position allows it to pull out of making sales in the UK. Regardless of the legal logic, there is no economically valid reason for rewarding any licensee in this way. More broadly, given the relatively high levels of exit and entry into the mobile devices business over a ten-year period, one might find that many firms who have made significant past sales on a global basis, without taking licenses, may be in a position where they are able to pull out of individual national markets simply in order to minimize their liability for past infringement.25
V. CONCLUSION
While we might not rise to the level of Mr. Bumble in Oliver Twist, who famously protested that the law was an idiot and indeed an “ass” (as the term is used in Britain) for supposing that his wife acted under his direction, national laws are still clearly a very imperfect instrument for upholding a robust system of licensing mobile SEPs. There are a great many licensing agreements involving mobile SEPs that have been commercially agreed between implementers and SEP owners, and these licenses most often pertain to the portfolio as a whole. These licenses, forged in the shadow of the FRAND commitment, contain compensation for past use, which also pertains to the portfolio as a whole. The Unwired Planet judgments were salient precisely because they paid due attention to the realities of portfolio licensing. The judgments at least implicitly recognized that proving the validity and infringement of the UK patents-in-suit was but a device to establish that the portfolio as a whole was enforceable. It ought to follow from this that in a commercial negotiation, a licensor that had proven validity and infringement of patents in the UK — which would typically have counterpart patents pertaining to the same technology in other jurisdictions — would be able to use these as a means to obtain a global portfolio license on FRAND terms. A hypothetical commercial negotiation at the time of first infringement is, of course, precisely the counterfactual on which damages awards should be based. Regardless of the merits of the legal reasoning, it would be economically unfortunate if the logic and reality of the global portfolio licensing system were reflected in license determinations, but then tossed aside for purposes of deciding damages awards.
Alternatively, one could also take the view that where licensees “elect” to take an injunction in lieu of a court-determined global FRAND license, the benefit of the FRAND obligation is no longer available to the licensee. The right damages counterfactual may then be a commercial negotiation over a UK license, attended by the threat of excluding the licensee from the UK market, which may yield a damages rate (applied to a UK sales base) that is much greater than the global FRAND license rate.
At the very minimum, scholars and indeed judges should recognize the very critical role that the frictions of enforcement coupled with the limitations of national patent law play in limiting licensor’s ability to efficiently collect payment for their IP. The decisions of implementers such as HTC and ASUS to eschew or threaten to eschew UK sales for the express purpose of evading a court-awarded global license indicate that these limitations of the law are actually an important threat that confers significant bargaining power on licensees. This is an important point to appreciate when evaluating license agreements as part of a “comparable” license analysis used in determining FRAND rates, as these licenses are forged in the context of highly imperfect enforcement mechanisms that generally favor the licensee.26
1 The authors are affiliated, respectively, with the University of California-Berkeley and Berkeley Research Group LLC, and Berkeley Research Group, LLC. The authors have been retained to provide independent expert analysis by IPCom GmbH and Koninklijke Philips N.V. in recent UK litigation, including the Asustek and HTC matters referenced herein. The views expressed in this article are, however, those of the authors alone. The authors thank Sophie Lawrance and James Boon at Bristows LLP for helpful comments.
2 The term “licensees” is used to encompass both implementers who actually take a license and implementers who are negotiating over license terms or litigating over license terms. Likewise, the term “licensors” encompasses potential licensors as well as actual licensors.
3 Smaller licenses might be agreed in cases where the geographic scope of the licensee’s activity is limited and is likely to stay that way. Many licensees may value the freedom to operate that is inherent in a global license, however.
4 The term “UK courts” is used in this paper, although of course, the initial decisions in Unwired Planet were made by English courts and then upheld by the UK Supreme Court.
5 [2020] UKSC 37.
6 [2018] EWCA Civ 2344.
7 [2017]] EWHC 711 (Pat).
8 The Courts have not, however, ruled out the possibility that single-territory or national licenses could be FRAND in some circumstances.
9 These costs and risks are inherent in the non-self-enforcing nature of patent rights. In the SEP arena, we particularly note that licensors monetized their SEP portfolios through multiple license agreements with multiple licensees. In this context, there is substantial scope for “externalities” across different licensing situations. For example, different courts can and do come to different verdicts on the validity and essentiality of the same patent. Thus, an adverse ruling on validity (or indeed even on FRAND) might result in significant problems licensing a portfolio to future licensees. Conversely, a positive verdict on validity may create some positive momentum for a licensing programme, but it is always subject to subsequent challenge and it does not ensure that subsequent putative licensees will take licenses.
10 See Melamed, A. Douglas and Carl Shapiro (2018), “How Antitrust Law Can Make FRAND Commitments More Effective”, Yale Law Journal, Volume 127, Number 7, pp. 2110-2119., which expresses the once-conventional view emphasizing the “hold up” of licensees by SEP holders.
11 See, for example, the UKSC decision at paragraph 10 and paragraph 59.
12 See ETSI IPR Policy, Section 3.1., which states:
It is ETSI’s objective to create Standards and Technical Specifications that are based on solutions which best meet the technical objectives of the European telecommunications sector. … In achieving this objective, the ETSI IPR Policy seeks a balance between the needs of standardization for public use in the field of telecommunications and the rights of the owners of IPRs.
13 See, for example, the discussion in the Court of Appeal’s decision upholding the original judgment at paragraphs 195-207.
14 See the Court of Appeals decision at paragraph 197, endorsing an “effects-based” approach to discrimination in this case. The Court stated “Once the hold-up effect is dealt with by ensuring that the license is available at a rate which does not exceed that which is fair and reasonable, it is difficult to see any purpose in preventing the patentee from charging less than the license is worth if it chooses to do so.”
15 We should recognize, given the costs and risks involved, that licensors will not typically want to use litigation as a first resort. In Europe and the UK, the framework around the availability of injunctive relief for SEPs outlined by the European Court in Huawei v. ZTE, further limits the attraction of litigation as a first resort of licensors.
16 We would expect that the licensee’s subjective assessment of this probability will be higher for portfolios with a significant licensing history or portfolios whose member patents have been tested in litigation.
17 Of course, other jurisdictions such as Germany and the U.S. do provide alternative litigation venues. However, the UK courts have more clearly articulated the issues around global-level licensing than have others, and thus have held up hope that licensors would not have to litigate in multiple jurisdictions (with all the potential inconsistencies, risks and uncertainties that entails) in order to secure a global license.
18 The licensor can, in theory, replace its “one stop” UK litigation with litigation in multiple other jurisdictions. But this would recreate the inefficiencies and uncertainties that we discussed above. One can readily imagine that the threat to submit to a UK injunction would be a potent source of bargaining power for licensees and would likely result in a relatively favorable settlement of the dispute for them.
19 It is important to note that the choice the licensee is making is not between submitting to an injunction or accepting a rate that has been dictated by the licensor, but submitting to an injunction instead of taking a license on terms that have been determined to be FRAND by a scrupulous and neutral arbiter. Further, we note that in the case of licensees that have not made serious FRAND-compatible counteroffers or have not otherwise demonstrated serious commitment to signing a license in a timely fashion, a FRAND license would perhaps be too generous a remedy. One could argue for higher-than-FRAND rates to apply to court-awarded licenses where the licensee has not behaved in a cooperative fashion.
20 The hypothetical negotiation should, of course, be based on what would be paid assuming that there are at least some valid and infringed patents in the licensor’s portfolio.
21 Philips v. Asustek [2020] EWHC 29 (Ch). It is worth noting two concerns that Marcus Smith J raised in connection with Philips’ contention that damages should be based on the hypothetical license that would have been agreed at the time of first infringement, i.e., a global FRAND license. He suggested that this would allow Philips to collect damages on patents in foreign countries that had not been proven valid and infringed (see the reference to “enabling the [H]older to recover royalties for sales in those other jurisdictions which, had the Holder litigated there, would not have been recovered” at paragraph 47.2 (a)); and he further suggested that Philips was not precluded from collecting damages in other countries, even after it had collected “global” damages in the U.K (paragraph 47.2.(c)). However, it is worth pointing out that UK courts already make global FRAND license adjudications which can be based only on proving the validity and infringement of a small number of UK patents, without requiring evidence of infringement of patents in foreign countries. Second, from an economic perspective, it is unlikely that licensors would seek to litigate in multiple jurisdictions after a global award has been made in the U.K — indeed the very appeal of bringing suit in the UK is the hope that litigation in multiple jurisdictions can be avoided or at least minimized. Additionally, licensors cannot act as opportunistically as the judgment suggests, as they are involved in negotiation and potentially litigation with multiple licensees and opportunistic behavior can turn out to have negative consequences for future licensing and litigation. Again, economic incentives temper legal possibilities.
22 IPCom v. HTC [2020] EWHC 2914 (Pat).
23 There were unique factual circumstances involved in the IPCom v. HTC proceeding, as the need for a FRAND inquiry had been dismissed. However, it is not entirely clear to us whether if, in the course of a FRAND inquiry, a licensee announced its intention to submit to an injunction, the licensee’s conscious choice to submit to an injunction in order to avoid taking a global license would be accounted for in determining the scope of damages.
24 The case originally commenced in 2011. HTC claimed to have implemented a “work around” which meant that it was no longer using the IPCom patent that was in suit and found infringed. IPCom subsequently found that HTC had imported at least some devices into the UK that nonetheless used IPCom’s patent-in-suit. Notwithstanding the work around, there would still have been some use of IPCom’s patent-in-suit for a period of several years before the workaround was implemented, as well as possible use of other technologies in IPCom’s portfolio.
25 It appears that this strategy of submitting to an injunction was first used by ZyXEL in TQ Delta v. ZyXEL,[2019] EWCA Civ 1277, in a dispute involving DSL patents. In that particular case, the change in the licensee’s economic incentives with respect to taking a license arose because the patent-in-suit that had survived the technical trial was set to expire in three months, which limited the value of seeking a license.
26 This stems simply from the fact that the licensee cannot be excluded from using the licensor’s IP until such time as the licensor is able to prevail in court on issues of validity and infringement. There are also other important factors — such as the need for a licensor to interact with multiple licensees over the same portfolio — that also create asymmetries that favor the licensee. These asymmetries cut against the notion that licensors derive the power to hold-up licensees simply because they have declared-essential patents. In most other contexts (involving tangible goods instead of ideas) monopoly power or hold-up power does not require going to the courts to effectuate.
