On 27 April 2023, the European Commission (the “Commission”) proposed a new regulation on the licensing of standard essential patents (the “Proposal”).[1] The objective of the Proposal is to facilitate standard essential patent (“SEP”) licensing negotiations by providing clarity on several aspects: transparency as to who owns SEPs and which SEPs are essential; transparency on fair, reasonable and non-discriminatory (“FRAND”) terms and conditions; and dispute resolution for the determination of FRAND terms.[2]
The European intellectual property landscape is already undergoing major upheaval with the launch of the Unified Patent Court (the “UPC”) on 1 June 2023.[3] If the Proposal is adopted, it could play a crucial role in shaping the future of innovation and standard setting in the European Union. The proposal is, however, controversial, and it is not certain that it will survive parliamentary scrutiny as it is.
The Proposal provides for the following significant changes:
- The European Union Intellectual Property Office (the “EUIPO”) would establish a Competence Centre to set up and maintain an EU-wide register for SEPs, including details of relevant standards, registered SEPs and SEP holders, and terms of the SEP holder’s FRAND licensing commitment. SEP holders seeking to license their SEPs for royalties and to enforce them in the EU would have to register the patents in a SEP register.
- The EUIPO would conduct non-binding checks to determine whether a patent is essential to a standard;
- Before launching a litigation, parties to a SEP licensing dispute would have to go through a mandatory conciliation procedure administered by the EUIPO. An independent conciliator would seek to help the parties reach mutually acceptable FRAND licensing terms and conditions. At the end of the process, if the parties fail to reach agreement, the conciliator will issue a non-binding report with recommendations on the FRAND rate;
- SEP holders could jointly agree on an aggregate royalty for using a standard (potentially with the help of a conciliator). Both implementers and SEP holders could request EUIPO to provide a non-binding expert opinion on the aggregate royalty. This could then serve as a basis for FRAND royalty determination for individual patent portfolios, and for the distribution of the royalty stack.
The Proposal clarifies, appropriately. that it “does not apply to royalty-free licensing”, acknowledging that, whilst licensing on FRAND terms and conditions includes licensing royalty-free, “most issues arise with royalty-bearing licensing policies”.[4]
While the Proposal aims to address a number of outstanding issues inherent in the world of FRAND licensing, there are various concerns with the Proposal that deserve further consideration during the legislative process.
In the rest of this blog post, we summarize the Proposal in more detail and provide a preliminary assessment of its main provisions.
I. Key Provisions and Critical Assessment
1. SEP Registration
The Proposal establishes a “Competence Centre” within the EUIPO (which currently focuses solely on trademarks and design rights) to set up and maintain an EU-wide register for SEPs. This register would include the following information:
- Information on relevant standards;
- Details of the registered SEP, including identification details, date of registration, and the standard version, technical specifications and specific sections for which the patent is considered essential;
- Details of the SEP holder and its legal representatives; and
- Information on licensing, including the terms of the SEP holder’s FRAND licensing commitment, existence of any public standard terms and conditions for SEP licensing to SMEs and availability for licensing through patent pools.[5]
Registration of SEPs will be mandatory for enforcement purposes.[6] SEP holders would need to request registration of their SEPs within six months following the Competence Centre’s creation of an entry in the register for the relevant standard, or the grant of the SEP by the relevant patent office.
If a SEP is not registered within the time-limit, the SEP holder will not be able to enforce it in court and will not be able to collect royalties or past damages for any use of the SEP from the time-limit until the date of registration.[7] Updates to reflect relevant changes would also have to be notified to the Competence Centre within six months from the change occurring, or the registration will be suspended until the inaccurate or incomplete information is remedied. The intention is that implementers can readily identify SEP holders so as to seek a license.
The establishment of a SEP registry risks a duplication of effort. In principle, transparency is desirable, to the extent it allows implementers to predict the licensing costs of marketing standard-compliant products. The European Telecommunications Standards Institute (the “ETSI”) has urged the Commission to rethink its Proposal, stressing that ETSI already maintains its own database, which contains not only essentiality declarations (as the envisaged EUIPO SEP register would) but also technical specifications. It should be kept in mind, however, that the ETSI database is limited to ETSI standards, and does not include SEPs claimed to read on other standards.
The consequences of non-registration might in some cases leave implementers worse off. Many patentees declare more patents as SEPs than are actually essential in order to stake a claim to a greater share of royalty stacks than they actually deserve (the perceived problem of “overdeclaration”). Many SEP holders also issue “blanket declarations”, not identifying their patents, but promising that if they have any, these will be licensed on FRAND terms. The drawback of this current system is the lack of transparency, but there is at least some promise of constraints on licensing terms for the patents included in a padded patent list or blanket declaration. The proposed system would curb these practices, and promise transparency at a cost, but may not deliver on that promise: if SEPs are not registered timely, they cannot be enforced. SEP holders will have an incentive to argue that their undeclared SEPs are not technically essential and not subject to FRAND obligations, but are still valid and infringed. There would be no constraints on injunctions or royalties for non-SEPs. Implementers in turn will argue they are standard-essential. This scenario could increase disputes, and could turn out problematic for implementers, if patents are found to be merely “commercially essential” but not SEPs, allowing injunctions and royalty charges above FRAND level.
2. Essentiality Checks
The Proposal includes essentiality checks for registered SEPs.[8] The proponents hope that this will improve transparency in the SEP landscape and help implementers to assess whether they need a license.
Essentiality checks would be carried out annually and would cover patents from each SEP holder and each specific standard in the register. SEP holders may also propose up to 100 of their SEPs for review.[9]
The results of the non-binding essentiality checks would be published in the SEP register, indicating only whether the outcome of the assessment was positive or negative.[10] Although the results of the essentiality checks are not legally binding, they can be presented as evidence before stakeholders, patent pools, public authorities, courts and arbitrators.[11]
Essentiality checks are commonplace for SEPs included in patent pools. Such pools can only include complementary (i.e., essential) patents, so as to avoid pools turning into licensing cartels and tying devices if they included substitute (i.e., non-essential patents). [12] There are some issues, however:
Essentiality checks are not comprehensive. The Proposal provides that no more than one SEP from the same patent family may be examined for essentiality[13] and that the result of the essentiality check applies to all SEPs in the same patent family.[14] However, a singular examination does not allow automatic assumption that other patents in the same family are essential to a standard or not.[15]
Assessing essentiality by samples may be unreliable. Checks will not be carried out on all patents, but only on a sample. The exact methodology has yet to be defined.[16] The use of samples is not without problems. It inevitably introduces a margin of error, as a sample size must include thousands of patents to be meaningful. [17] In the absence of a widely accepted method for assessing a large portfolio of filed patents, the results of these essentiality assessments are likely to be inaccurate.
Essentiality checks do not eliminate litigation. The essentiality finding is not binding, and can be relitigated. Even if parties accept the non-binding essentiality determination, they may still be in disagreement in relation to infringement and validity issues. The transparency is therefore limited and does not pre-empt disputes.
3. FRAND Determination Procedure
The Proposal provides for a mandatory conciliation procedure aimed at assisting parties in negotiating FRAND rates. The procedure, to be administered by the Competence Centre, would have to be completed within nine months.[18] It can be initiated at the request of the SEP holder, a potential implementer, or by both parties voluntarily to resolve FRAND disputes.[19]
Parties to a SEP dispute would be required to go through the conciliation procedure before they can initiate patent infringement proceedings or a FRAND assessment before a Member State court or the UPC.[20]
The procedure involves a “conciliator” chosen by the parties from a group of three candidates proposed by the EUIPO from a roster. If the parties fail to agree on a candidate, the Competence Centre would appoint the conciliator.[21]
The parties will be requested to make submissions and proposals.[22] The conciliator will assist them in in their endeavor to reach a FRAND rate and will make proposals. [23] At the request of a party, or if the conciliator deems it necessary, an oral hearing will be scheduled.[24] The FRAND determination will focus on terms for a global license, unless the parties agree otherwise.[25]
If, at the end of the procedure, the parties have not yet settled, the conciliator will make a final proposal on the FRAND rate, which the parties may or may not accept.[26] If the parties do not settle or if they reject the conciliator’s proposal, the conciliator will terminate the procedure and issue a report on the determination of FRAND terms and conditions The non-confidential part of this report will contain the parties’ final proposal and the methodology used by the conciliator to make the determination, and can then be used as evidence in court proceedings.[27] The purpose of the report is therefore twofold: to encourage settlement and to provide transparency on the process and recommended FRAND in cases of disagreement.[28]
Fig. 1. Stylized flowchart of the conciliation FRAND determination procedure.
Confidential nature of the EUIPO’s findings. The report on the determination of FRAND terms and conditions includes: (i) a confidential assessment of the FRAND determination; (ii) a confidential summary of the main issues of disagreement; and (iii) a methodology and an assessment of the FRAND determination. Only the latter will be publicly available.[29] However, the confidentiality of the results removes much of their potential wider benefit (transparency), as other market participants cannot use the information developed by the Competence Centre in their own FRAND negotiations.
Limited impact of non-binding FRAND determinations. The Proposal delays disputes, but does not require the parties to commit to be bound by the FRAND determination. As currently drafted, the Proposal provides that:
- Whoever triggers the FRAND determination process (SEP holder or implementer) is not required to initially state whether they accept the determination as binding;
- When responding, the other party will have to state whether “it commits to comply with [the] outcome” of the FRAND determination.[30] However, it is not clear how binding such a “commitment to comply” can be, as the Proposal does not explicitly state that the parties must enter into a license agreement on that basis;
- In any case, the outcome of the FRAND determination process will not be binding unless the party that initiated the process also agrees to be bound.
A non-binding FRAND determination may be helpful, but may not be sufficient. Our experience with mediation in FRAND determination is not promising, because parties may both think they are better off litigating – SEP holders may prefer to seek injunctions, and implementers may prefer to challenge validity. Mandatory mediation would not change this. Even if the litigation leads to review of royalty levels, national courts are free to ignore non-binding FRAND determinations. This is a particular concern in German courts, which tend to refuse to assess whether proposed license terms are FRAND, and may require a licensee to accept the SEP holder’s offer so long as it is not “absolutely unacceptable”.[31] They tend to even issue injunctions without reviewing the SEP holder’s FRAND offer at all, in violation of the principles set by the EU Court of Justice in Huawei/ZTE.[32]
Alternative approaches have been put forward, such as requiring SEP holders and implementers to enter into a “reciprocal FRAND agreement” modelled on the FRAND injunction decisions of the UK courts.[33] Under this proposal, SEP holders would be required to offer implementers a reciprocal FRAND agreement in which the SEP holder agrees to forego patent infringement remedies like injunctions, and the implementer agrees to enter into a license on FRAND terms determined in negotiations between the parties or (if no agreement is reached) set by a third party. Implementers would have until 30 days after a finding of infringement to accept the offer. If they reject it, they are subject to infringement remedies (including injunctions). A failure to comply with the third-party determination would be addressed in a contract action before a national court over FRAND terms rather than in a patent infringement proceeding.
We have in the past proposed a comparable system whereby either party could make an irrevocable offer to agree on third-party rate determination, creating a rebuttable presumption of willingness to take a license, for the purposes of whether an injunction is available.[34] Acceptable rate setting options would include:
- Arbitration, for instance, under WIPO FRAND arbitration rules; [35]
- In Germany, an offer to enter into a “rate setting” agreement (Sect 315 BGB);
- In the UK, a SEP holder request for a court to issue a “FRAND injunction”;[36]
- In The Netherlands an offer to agree on “binding advice” (Art 7:900 BW);
- A request for a declaratory judgment on the meaning of FRAND, where possible, under contract law, or a claim for specific performance of the third-party beneficiary arrangement inherent in FRAND Undertakings.
If a licensor or licensee refuses to agree to rate setting, that should be regarded as a rebuttable presumption that it is “unwilling.” The party refusing the rate determination could then rebut the presumption of unwillingness by proposing a rate that is FRAND. This solution is designed to encourage a reasonable outcome, as both parties would have an incentive to agree on a rate: (a) if the SEP holder offers third-party rate determination and the implementer refuses, an injunction is available unless the implementer has offered a FRAND rate; (b) if the implementer offers third-party rate determination and the SEP holder refuses, no injunction is issued unless the SEP holder has offered a FRAND rate and the implementer refuses; (c) if neither offers third-party rate determination, the Huawei/ZTE process is followed. The court would have the last word on the FRAND rate.
Alternative proposal
4. Determination of Aggregate Royalties
The Proposal also includes a procedure for determining an aggregate royalty (i.e., the total maximum price) for using a standard before or shortly after its publication. The objective is to determine the total royalty stack of the standard transparent, so that implementers can factor royalties into the cost of their products.
Three scenarios are considered:
- SEP holders for which FRAND commitments have been made may jointly notify the Competence Centre of the aggregate royalty which they have agreed upon, which the Competence Centre shall publish.[37]
- Where there is no agreement among SEP holders on an aggregate royalty, certain SEP holders may request the Competence Centre to appoint a conciliator to assist them. In this case, the role of the conciliator would be to facilitate the decision-making by the participating SEP holders without making any recommendation for an aggregate royalty.[38]
- Finally, SEP holders and/or implementers will be able to ask the Competence Centre for an expert opinion on an aggregate royalty. When this request is made, the Competence Centre will appoint a panel of three conciliators and manage a process in which all interested stakeholders will be invited to participate. After receiving information from all of the participants, the panel will provide a non-binding expert opinion on the aggregate royalty.[39]
SEP holders may have an incentive to declare a high royalty upfront. SEP holders may view the ex ante declaration of their individual maximum royalties, or an aggregate royalty, as an opening bid in negotiations. They may therefore be incentivized to declare a high royalty in the hope that if the price is driven down in negotiations by caps and discounts, they will end up with more than if they had declared a low royalty to begin with. In such a situation, Cournot complement problems[40] may lead to an excessive aggregate royalty, especially if SEP holders are not also active as implementers, and not all of the SEP holders participate in the exercise. Aggregate royalty calculations should therefore ideally be accompanied by an effort to form a single patent pool for each standard. Conversely, SEP holders who are also significant implementers may have various incentives. On the one hand, they may wish to generate a positive network effect for a new standard by charging low royalties and encouraging other SEP holders to do the same. On the other hand, if they can net off their royalty payments against amounts due to other large SEP holders, they may have an incentive to charge high royalties so as to lower their variable costs, and perhaps even exclude implementers with no or few SEPs from the market. It is therefore important to take into account market conditions and, where possible, to rely on hedonic pricing or contingent valuation analysis (even if only as a cross-check).
II. Conclusion
The publication of the Proposal is only a first step in the legislative process, and is controversial, especially with SEP holders. The Proposal awaits further discussion in and agreement from the European Council and the European Parliament before official adoption, with changes expected. The Commission believes that the EU-wide rules on transparency regarding SEPs and FRAND terms included in the Proposal would have a harmonizing effect within the EU, which would facilitate the work of national courts and the upcoming UPC.[41] At the same time, the Proposal deviates from approaches adopted elsewhere.[42] It remains to be seen whether the final version will retain all aspects of the current draft. The provisions on transparency of FRAND terms may be strengthened, and rate determination may become mandatory, whereas provisions on ex ante essentiality determination (and associated delays) may be weakened or abandoned as too costly and insufficiently effective.
Licensing negotiations involving sophisticated technologies are legally intricate owing to the highly complex nature of technology and the standards-setting process. Perhaps the most effective measure could be to encourage third-party rate setting proceedings that are binding on the parties, so as to encourage them to reach a negotiated agreement.
[1] The publication of the Proposal follows a call for evidence and a public consultation launched by the Commission in February 2022.
[2] The Proposal is without prejudice to national competition rules. Recital 2 and Article 1(7).
[3] The launch of the UPC on 1 June 2023 will be an important milestone for SEP enforcement. The UPC will provide a single European venue for patent infringement actions – including SEPs – that can award damages and issue injunctions in up to 17 EU Member States. See Cleary IP and Technology Insights, Contracts involving European Patents: Five Steps to Take Now in the Wake of the New EU Unitary Patent System for further details.
[4] Recital 7 and Article 1.
[5] Articles 4, 19-25.
[6] Articles 19-25.
[7] Article 24.
[8] Articles 28-33.
[9] Article 29.
[10] Article 33. The lack of an essentiality check or an ongoing essentiality check does not preclude licensing negotiations or any court or administrative procedure in relation to a registered SEP (Article 28).
[11] Article 28.
[12] See Communication from the Commission — Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, paras 250-255.
[13] Article 28(3).
[14] Article 33(1).
[15] Similar problems arise in relation to different patent claims in the same patent. Sometimes claims from multiple members of a patent family are standard-essential. However, there could also be a case where only one member of a patent family contains essential claims. If one of the patent family members that has no essential claim gets checked, the patent family as a whole will be erroneously declared non-essential.
[16] Article 29.
[17] See K. Mallinson, Essentiality Checks Might Foster SEP Licensing, But Do Not Stop Over-Declarations from Inflating Patent Counts and Making Them Unreliable Measures (16 November 2022), arguing that sample sizes of thousands of patents would be required to provide even only modest levels of precision in essential patent counts. Available at http://dx.doi.org/10.2139/ssrn.4278639.
[18] Article 37.
[19] Article 34.
[20] Article 34(1). SEP holders may seek a provisional injunction “of a financial nature” (i.e., not a sales ban but a requirement on implementers to make a deposit or post a bond to cover license fees) before the conclusion of the FRAND Determination Procedure. However, the parties are required to request the national court to suspend proceedings on the merits for the duration of the FRAND determination (Article 34(4)).
[21] Article 39.
[22] Written submissions including evidence and expert opinions can be submitted (Article 43).
[23] At least five months before the nine-month limit, the conciliator will notify to the parties a written recommendation of FRAND determination (Article 51). At the latest 45 days before the nine-month limit, the conciliator will submit a reasoned proposal for a FRAND determination (Article 55).
[24] Article 53.
[25] Article 38(6) and recital 11.
[26] Article 56.
[27] Article 57.
[28] Recital 41.
[29] Article 57.
[30] Article 38(2).
[31] See, for instance, Munich I Regional Court judgment of August 5, 2022 (21 O 8890/21) – “keepawake-message”, available at https://www.gesetze-bayern.de/Content/Document/Y-300-Z-GRURRS-B-2022-N-34498?hl=true.
[32] Court of Justice of the European Union, Case C-170/13 – Huawei v. ZTE (2015) ECLI: EU:C:2015:477.
[33] See F. Gonell (Chief licensing lawyer at Qualcomm), The European Commission can achieve its goals without putting European participation in future standards at risk. Will it?, 18 April 2023, LinkedIn. The proposal: (i) would allow the implementer to be free from the threat of an injunction by agreeing to enter into a license on FRAND terms as confirmed by a court or an agreed neutral third party; (ii) would address any concerns about the German courts’ application of the Huawei/ZTE framework resulting in injunctions against unwilling licensees without sufficient inquiry into whether the patent holder has made a FRAND offer.
[34] See https://www.clearygottlieb.com/-/media/files/the-european-commission-should-and-can-incentivise-frand-arbitration-pdf.pdf.
[35] See WIIPO Arbitration for FRAND Disputes, available at https://www.wipo.int/amc/en/center/specific-sectors/ict/frand/annex1/.
[36] See Unwired Planet v. Huawei [2020] UKSC, 26 August 2020; InterDigital v. Lenovo, [2023] EWHC 539 (Pat), 16 March 2023. For more detail, see M. Dolmans, FRAND licensing commitments – Back to first principles, Elsevier, World Patent Information 69 (2022) 102113.
[37] Article 15.
[38] Article 17.
[39] Article 18.
[40] Different owners of complementary patents tend to set the rate they demand independently, ignoring that other patentees will do the same. The resulting royalty stack may exceed the optimal amount, leading to multiple monopoly rents.
[41] See Proposal, Explanatory Memorandum, para. 2.
[42] Since the UK Supreme Court’s decision Unwired Planet v. Huawei in August 2020, SEP enforcement has spread across borders, as courts in the UK, China and other jurisdictions have signaled a willingness to resolve such disputes by imposing global licenses. Earlier this year, the High Court of Justice issued a ruling in InterDigital v. Lenovo, the country’s second global FRAND rate decision, further cementing the UK courts’ stance on SEP enforcement. In the United States, the Biden administration issued a draft policy on SEP licensing and remedies in 2021 that sought to bridge the gap between a 2019 policy that largely favored injunctions in SEP disputes and its 2013 predecessor that largely discouraged such remedies. However, last year the government withdrew the 2019 policy without adopting the 2021 draft or reinstating the 2013 version – instead opting for a case-by-case approach to enforcement. A parallel UK reform initiative, launched in 2021, is underway.