This is a paper – which can be found here – on the the Unwired Planet v Huawei judgment reviewed in a post of 28 April 2017. You may remember that this is primarily an IP law decision. It is nonetheless relevant to us because it is the first decision I’ve seen where an (English) court determined a FRAND rate, and because it dealt with a number of competition law issues which were relevant for such determination. It is also relevant – and topical – because the court imposed a (quasi-)worldwide license, which brings to mind the lively discussion we had at the OECD in December on extra-jurisdictional remedies following Korea’s decision to impose a global FRAND license on Qualcomm.
The paper begins by summarising the context in which SEP litigation tends to arise, before describing the specific factual background of this case. In March 2014, Unwired Planet (UP) sued Huawei, Samsung and Google in the English Patents Court alleging infringement of six of its UK patents. Five of these patents were SEPs that UP had fairly recently acquired from Ericsson. The SEPs related to various ETSI telecommunications standards (2G/GSM, 3G/UMTS and 4G/LTE). In addition to the normal invalidity and non-infringement defences which are common in these cases, the defendants claimed that UP had not offered them a licence at a FRAND rate. Various competition law defences were also raised, such as that UP had acquired patents from Ericsson through an anti-competitive agreement and that UP had abused a dominant position through its conduct in starting the litigation and requesting an injunction.
Given the complexity of the issue, the case was split into six trials, one of which was about FRAND/competition – the resulting judgment being ‘the first in which FRAND issues have been considered at full trial and in detail by an English court.’ This judgment dealt with a wide variety of topics concerning English law and FRAND, not all of which will be of interest to readers of this email. Arguably, the most interesting topics of discussion were:
- The ‘Vringo problem’ i.e. whether there are a range of possible licensing terms that might be FRAND in a given situation, or only one. The former option poses a significant problem for a judge, who must decide what consequences follow from the making of FRAND or non-FRAND offers in a particular case. What happens if the parties make different offers, both within that FRAND range? If an IP holder makes an offer in a FRAND ranges which is not accepted by the implementer, can an injunction be imposed?
Birss J held that that only a single set of FRAND terms exists. This was supported by considerations regarding the creation of persuasive judicial benchmarks for future negotiations, the promotion of certainty, and making the enforcement of FRAND conditions conceptually straightforward. The authors note that a different approach has been adopted in the Netherlands and the US, where courts accept that FRAND can fall within a bandwidth.
- The decision also considers the implications of the rules set out in the EU judgment in Huawei v ZTE. In this decision, the European court prescribed a process for negotiating a FRAND licence. It requires a SEP holder to make a concrete written offer to license on FRAND terms. If the implementer is not prepared to accept the offer, it should make a counter-offer on FRAND terms and comply with additional requirements, such as providing security to the patentee.
Huawei alleged that UP had abused a dominant position regarding its SEPs in three ways: (i) failure to follow the process set out in Huawei v ZTE; (ii) seeking excessive prices and (iii) bundling SEPs and non-SEPs. The court held that UP had a dominant position, but dismissed the claims of abuse. The judgment first holds that the Huawei v ZTE framework should be applied fluidly; deviating from it did not necessarily give rise to an abuse. Secondly, the court gave short shrift to Huawei’s claims that UP’s offers amounted to excessive pricing. These two points are connected in the reasoning. Birss J’s finding that there is only one FRAND rate means that is unlikely that the parties would be able to offer the precise (and only) FRAND rate that he identified. As such, offers above or below FRAND may be made as long as they do not ‘disrupt or prejudice the negotiations’. While UP’s offers in some cases were several times higher than the level at which Birss J determined the FRAND benchmark, he decided that offers of this nature were simply part of the process of negotiation. While the meaning of ‘disruption’ is left open, it seems that litigation is perceived as a natural part of the negotiating process, with both parties being able to continue negotiating pending litigation.
Regarding the relevance of holdup and holdout (discussed last week at length), Birss J found that both may affect negotiations. As such, he held that, in arriving at FRAND, terms the SEP holder must not engage in hold-up, and the implementer must not engage in hold-out. How this is supposed to be assessed when the parties are able to offer prices far removed from potential FRAND prices and start court proceedings is not clear, however.
Lastly, Birss J held that, although a SEP holder cannot insist on a licence which bundles SEPs and non-SEPs, it is not contrary to competition law to make an offer that combines them, particularly as UP had made clear that it was prepared to discuss licensing SEPs only.
- Regarding the calculation of the FRAND rate, Birss J remarked that it would be disproportionate to evaluate every single patent thoroughly; portfolio size provided an appropriate indication of value, subject to certain adjustments to account for issues such as over-declaration of patents in the context of SSOs and respective SEPs. Birss J assessed value on two grounds: ‘comparable’ licences and the total aggregate royalty burden applying to a particular standard.
Regarding the valuation methodology, the authors think that Birss J adoption of patent counting as ‘the only practical approach’ to establish the value of a portfolio risks exacerbating the over-declaration problem that his methodology seeks to address. The authors also note that Birss J’s holding that the SEP holder is entitled to obtain both some of the value arising from the inclusion of its technology in a standard and some of the enhanced value of the products making use of that standard is controversial, since many economists argue that a FRAND value reflects the value of patents before they are included in a standard and become essential – a position adopted by the European Commission in Motorola (and reflected in how the article reviewed below addresses criticisms against competition law enforcement in this area).
- Regarding the importance of non-discrimination, one of Huawei’s arguments was that, since Samsung had benefitted from relatively low licenses, Huawei should be entitled to similar rates. Birss J held that the non-discrimination aspect of FRAND meant that a FRAND royalty rate should not be based on the identity of a particular licensee, but on the value of the portfolio. At the same time, he held that the obligation to license at the same rate would apply only if different rates would otherwise distort competition between the two licensees – and that Huawei had not adduced enough evidence that the rates charged to it would distort competition with Samsung.
The authors note that Huawei has been granted permission to appeal Birss J’s finding that a distortion of competition is required for the non-discrimination aspect of FRAND to apply – which I suppose makes sense, since I am not aware of prior case law or academic work holding that non-discrimination obligations in FRAND should be assessed be reference to their distortionary effect on competition.
- Regarding the geographic scope of the license, Birss J held that a FRAND licence between a licensor with a global SEP portfolio and a licensee operating on a global basis should have a worldwide scope. He found this to make practical sense, and to avoid the inefficiencies of licensing country by country. He noted that the licences disclosed during the litigation were largely global portfolio licences, and went as far as to hold that it was not FRAND for Huawei to insist on a licence with a UK only scope.
The authors note that this was unexpected and actually contradicted Birss J’s previous finding that threatening to obtain a national injunction to pressure a licensee to take a global licence covering patents beyond the jurisdiction of a national judge amounted to ‘international coercion’. They observe that this may encourage more patentees, perhaps particularly non-practising entities (NPEs), to litigate in English courts.
This is a useful dissection of the main topics one needs to address when dealing with FRAND litigation – which include, but go well beyond competition issues.
I would have liked to have seen a more thorough discussion of the implications of national courts granting global licenses, particularly given the risk of conflicting judicial decisions regarding the scope and rate of the FRAND licensing obligation. However, we may soon hear more on this since I understand that Huawei has been granted permission to appeal on: (i) whether more than one set of terms can be FRAND; (ii) whether a UK-only offer is FRAND in this context; (iii) whether the court may determine FRAND terms, including rates, outside the UK; and (iv) whether it was appropriate to grant an injunction excluding Huawei from the UK market unless Huawei took a global licence.