Designing Patent ADR: The PMAC Rules Through the Lens of FRAND

Arbitration as a Distinct Procedural Product

Arbitration is more ambitious than mediation, and that ambition has a price. It promises a private, specialized, and enforceable decision rendered by arbitrators selected for their technical or economic expertise. It is sometimes presented as high-end justice—an expression worth weighing carefully. UPC judges need no instruction in sophistication; the UPC is itself a sophisticated institution. But arbitration offers a different product: flexibility, confidentiality, party autonomy, technical specialization, and potentially universal enforceability. That difference is one of kind, not degree, and it is the starting point for any serious analysis of the role of the Patent Mediation and Arbitration Centre (PMAC).

The most recent illustration is the Court of Appeal's judgment in Acer/ASUS v. Nokia.1The Court held that Nokia's adjustable licence offers—under which final RAND terms would be determined by ICC arbitration—were offers of licences on RAND terms capable of acceptance. The implementers' refusal closed off access to the English court's declaratory jurisdiction to determine RAND terms. The Court also recognized the attraction of arbitration for global RAND disputes: a single tribunal with a global mandate, whose award would be enforceable under the New York Convention.2

That judgment does not bind the UPC, and it would be wrong to transpose it mechanically: the UPC operates under EU law, the UPCA, its own Rules of Procedure, and the Huawei v. ZTE framework.3But the question is now unavoidable—whether the UPC could, in an appropriate FRAND case, give procedural significance to a serious PMAC arbitration proposal. The answer should be: possibly, but carefully.

A mere invitation to arbitrate should not automatically make one party "willing" and the other "unwilling"—that would be too crude. But where a proposal is concrete, balanced, institutionally reliable, cost-conscious, and adapted to the dispute, refusal may become relevant. The Court of Appeal noted carefully in Nokia that the implementers had made no substantiated objection or counterproposal to the arbitration offered. That nuance is essential.

For the PMAC, this suggests a path. A FRAND arbitration proposal should not be a procedural trap. It should include, at a minimum, a neutral appointment mechanism, a credible confidentiality structure for comparable licences, a timetable, cost controls, rules on the use of evidence already exchanged before the UPC, and some form of anonymized publication or reasoned summary where possible.

The point is not merely to say "let us arbitrate." It is to make an offer sufficiently precise that the other party—and, if necessary, a court—can see what procedure is being offered, what issues are being removed from litigation, what remains before the UPC, and how confidentiality, timing, and costs will be controlled.

The PMAC Arbitration Rules already move in that direction. They contain confidentiality provisions but also allow publication of anonymized awards or decisions unless a party objects within sixty days.4This is not a minor detail. FRAND arbitration cannot become a black hole: if all significant licensing disputes disappeared into confidentiality, the law would not develop; if everything were public, parties would not come. The balance lies somewhere between secrecy and judgment.

The recent Samsung/ZTE developments point in the same direction. The UPC Mannheim Local Division issued a settlement proposal in the global SEP dispute and suggested PMAC mediation if the parties did not accept the proposed alternatives.5That is not arbitration, and it is not binding. But it shows a willingness to integrate the PMAC into the UPC's procedural imagination—which may be the most important lesson from Ljubljana: the Centre is not being launched as a decorative annex, but as a real procedural instrument within the UPC ecosystem.`

The Procedural Mechanics: PMAC as a UPC-Facing Process

The PMAC Arbitration Rules are not a set of general institutional arbitration rules transplanted into patent law.6They are written for disputes that often have a pending or potential UPC counterpart: patent infringement and validity litigation, FRAND licensing disputes, portfolio disputes, related commercial questions, and issues that may be better separated from the main judicial action.7

The Rules first create multiple gateways into the Centre.8A party may initiate an ADR information session; parties already before the UPC may be informed about ADR options and may elect arbitration; and the interaction between the Centre and the UPC is expressly organized where proceedings are pending and both parties jointly request a stay. This is not compulsory arbitration by another name—it is a consent-based mechanism capable of procedural recognition by the court when both parties engage with it seriously.

The Request and Response are unusually practical instruments.9The Request is not merely a notice of arbitration. It must identify the parties, the relevant IP rights and technology, the amount in dispute, any UPC case number, the arbitration agreement or consent, the relief sought, the proposed tribunal structure, the seat, language, and applicable law—and, in quantum, damages, or royalty disputes, the economic rationale for the claim. The Response must address jurisdictional objections, language, facts, relief, tribunal composition, seat, and applicable law within thirty days. That architecture is valuable in FRAND practice precisely because it forces the parties to define the dispute early rather than sheltering behind abstract willingness language.

The tribunal design is part of the product.10The Rules build in independence, impartiality, disclosure obligations, and challenge mechanisms, and authorize the Centre to manage appointments where the parties cannot agree. That matters in patent and FRAND disputes, where neutrality is not merely formal: the parties need arbitrators who understand patent validity, infringement, essentiality, licensing practice, economics, and confidentiality rings—without being captured by either side of the SEP ecosystem.

The case management provisions are deliberately flexible.11The seat may be fixed by the parties or, failing agreement, by the tribunal. Language may be agreed or determined. The tribunal must conduct proceedings fairly and efficiently, avoiding unnecessary delay and expense, and the initial case management conference should produce a procedural order covering submissions, witnesses, experts, document production, hearings, preliminary issues, interim measures, and sequencing. Expedited proceedings target a final award within six months from transmission of the file to the tribunal; ordinary proceedings require the tribunal to work within a structured timetable.

The evidentiary and confidentiality architecture is central.12The tribunal determines the admissibility, relevance, materiality, and weight of evidence; may invite additional evidence; holds hearings in camera unless otherwise agreed; and, under Article 47, has tools to protect confidential licence agreements and other sensitive material, including limited-access regimes and confidentiality undertakings. That is precisely where FRAND disputes require more than ordinary litigation templates.

Urgent relief is not neglected.13The Rules authorize interim and protective measures, including preservation of evidence and measures to prevent serious harm, and provide for an emergency arbitrator where urgent measures cannot await constitution of the tribunal. Critically, a request to a competent court for interim measures is not treated as a waiver of the arbitration agreement—a provision essential in patent practice, where interim relief, evidence preservation, and parallel court steps may remain necessary.

The most commercially significant provisions are probably Articles 48 to 50.14They transform FRAND arbitration from a slogan into a procedural toolkit. The parties and tribunal may define the scope of the FRAND dispute; address representative sampling, essentiality, selected licence terms, temporary terms, royalty base, royalty range, territorial scope, and methodology; use expert determination for essentiality or technical issues; and coordinate with parallel UPC, national, competition, or regulatory proceedings. In practice, a PMAC arbitration can be designed as a full global FRAND determination, but equally as a narrower valuation, royalty-methodology, or portfolio-sampling exercise.

Article 36 is another important bridge.15Where arbitration leads to a consent award, or where settlement terms need judicial effect, the parties may request confirmation by the UPC under Rules 365 and 11.2 of the Rules of Procedure. The distinction matters: the arbitral tribunal does not itself revoke or limit a patent erga omnes, but the UPC framework can give procedural and enforcement effect to settlements or consent awards in a way that connects private resolution with the public court system.

That is precisely where counsel can add value. A PMAC clause or PMAC offer should not simply name the Centre. It should identify the patent or portfolio perimeter, the FRAND questions to be decided, the status of any UPC action, whether parallel actions will be stayed, the evidence protocol for comparable licences, the confidentiality ring, the timetable, the costs architecture, the form of award, and the route—if any—to UPC confirmation. PMAC strategy is therefore not merely about choosing ADR: it is about designing the procedural interface between arbitration, mediation, expert determination, and UPC litigation.

The Cost Question

UPC litigation is expensive. From January 1, 2026, infringement actions and declarations of non-infringement carry fixed court fees of EUR 14,600, with additional value-based fees in many cases; revocation actions are set at EUR 26,500. High-value disputes can trigger very significant court fees—one estimate placing first-instance UPC fees for disputes exceeding EUR 50 million at EUR 444,900.16

The PMAC is not free either.17But its cost structure must be compared with the cost of full-scale UPC litigation, not with an abstract ideal of costless justice. In the right procedural design, PMAC arbitration can be materially more cost-competitive. The registration fee is EUR 3,000, EUR 6,000, or EUR 8,000 depending on the amount in dispute. Arbitrators' fees generally range from EUR 350 to EUR 750 per hour, while expedited arbitration uses an hourly rate of EUR 350 with a EUR 40,000 cap, save in exceptional circumstances.18

This direct comparison matters. A high-value UPC dispute may generate substantial court fees before representation, experts, translations, and internal litigation management are factored in. A narrowly defined PMAC arbitration—particularly an expedited, sole-arbitrator, or online procedure—can keep decision-making costs materially lower. That is especially true where the question is not "who wins the patent war?" but a specific issue: royalty methodology, quantum, a sample-based essentiality question, or the terms of a licence.

The qualification is important. A three-arbitrator, high-value, fully contested global SEP arbitration can itself become expensive; the PMAC Schedule expressly increases administrative fees for three-arbitrator cases as the amount in dispute rises.19The cost advantage therefore lies less in the label "arbitration" than in procedural architecture: narrowing issues, using a sole arbitrator where appropriate, agreeing to online hearings, using expedited arbitration, exploiting early UPC referral reductions, and deploying expert determination for technical or valuation questions.

Mediation carries a lower price point, but it too has a cost: the mediation fee schedule provides for a EUR 1,000 registration fee per party, administrative fees of EUR 4,000, EUR 15,000, or EUR 25,000 depending on the amount in dispute, and reductions for certain cases including early UPC referrals and small or micro-enterprises.20The point is not that ADR is free—it is that it can be proportionate when the process is designed around the real commercial obstacle.

This raises an access-to-justice concern. If the PMAC becomes only a forum for large SEP portfolios, multinational implementers, and high-value pharmaceutical disputes, it may still succeed, but without realizing its full potential. The question is not whether sophisticated parties can afford PMAC arbitration—they can. The question is whether PMAC can also become useful where full UPC litigation is disproportionate to the issue that actually needs to be decided. The Centre should therefore develop cost-sensitive routes: expedited quantum procedures, capped expert determinations, model protocols for SMEs, early neutral evaluation, and partial referrals limited to specific issues.21

Mediation and expert determination may be decisive here. The PMAC should not only be the premium vehicle for patent ADR—it can also be the practical workshop in which parties isolate the technical, economic, or licensing problem that the court does not need to resolve in full.

Privatization of Justice? Framing the Debate

There is a deeper concern: the privatization of justice. It should not be overstated. Arbitration and mediation do not undermine public justice merely because they are private. They can reduce costs, preserve business relationships, protect confidential information, and resolve disputes that courts cannot easily handle in a commercially complete way.

Nor should the PMAC be assimilated too quickly to ordinary private arbitration.22It was created by the UPCA and operates within the broader institutional architecture of the UPC. Its Rules expressly refer to UPC litigation, the CCMS, Rules 11 and 365 of the Rules of Procedure, and the possibility of judicial confirmation of settlements or consent awards. That embeddedness is a major strength: the Centre is not outside the UPC system—it is one of the ways in which the system can manage technically complex, commercially dense disputes without forcing every issue into a full judicial track.

The legal architecture itself reflects that balance. Article 35 of the UPCA gives the Centre facilities for mediation and arbitration but draws a firm line: a patent may not be revoked or limited in mediation or arbitration proceedings. At the same time, the UPC Rules of Procedure allow the Court to confirm settlements and consent awards, including terms by which a patent owner undertakes to limit, surrender, agree to revocation, or not assert a patent. That is not a contradiction—it is the distinction between private inter partes resolution and public effects that require a court-facing mechanism.

But the concern should not be ignored.23Patent law develops through public decisions. The UPC is still young, and its legitimacy depends not only on efficiency but on the creation of a coherent body of case law. If too many high-value or strategically important disputes disappear into confidential ADR, public law may lose some of its raw material.

The better framing is not "public justice versus private justice," but rather: which issues require public adjudication, and which require a technically reliable, confidential, and enforceable inter partes mechanism? The answer is not to oppose the PMAC, but to design the right balance. Validity and infringement questions of general importance should remain with the courts where appropriate; inter partes economic questions may often be better resolved through ADR. A finding that affects the patent as a public right—or that shapes UPC doctrine on validity, infringement, proportionality, or remedies—is categorically different from an inter partes determination of royalty rate, payment structure, portfolio sampling, or the treatment of confidential comparable licences.

Anonymized awards, reasoned summaries, and judicial confirmation of settlements can help preserve some transparency. Because the PMAC Rules already contemplate anonymized publication absent a party objection, confidentiality need not mean complete invisibility. The task is to create a membrane between private settlement and public law: enough confidentiality for parties to use the Centre, enough transparency for the European patent system to learn from what happens there. The PMAC should be a complement to public justice, not an ideological substitute—and it is precisely if that balance is achieved that it will not weaken the development of European patent law, but may strengthen it.

Conclusion: From Symbolism to Practice

The practical conclusion is simple. The PMAC will succeed if lawyers do not use it as a slogan. It requires careful procedural work: a precise clause, a delimited referral, a workable information protocol, a confidentiality structure, cost modelling, and a plan for how any award or settlement will interact with pending UPC proceedings. In that sense, the PMAC is not merely another institution. It is a new procedural instrument for European patent strategy.

• 1

  Acer Inc. v. Nokia Techs. Oy, [2026] EWCA Civ ___ (May 2026); Matthieu Dhenne, Coming Together? Acer v Nokia and the Contractual Turn of FRAND Arbitration, Kluwer Patent Blog (May 13, 2026).

• 2Convention on the Recognition and Enforcement of Foreign Arbitral Awards arts. III, V, June 10, 1958, 21 U.S.T. 2517, 330 U.N.T.S. 38.
• 3Case C-170/13, Huawei Techs. Co. v. ZTE Corp., ECLI:EU:C:2015:477 (July 16, 2015).
• 4PMAC Arbitration Rules art. 47(7) (providing for publication of anonymized awards or decisions absent a party objection within sixty days).
• 5Cleary Gottlieb, Continued FRAND Divergence: UPC Issues First-Ever Settlement Proposal and Munich Court Grants Injunction in Samsung-ZTE Cellular SEP Dispute (2026), https://www.clearygottlieb.com.
• 6Agreement on a Unified Patent Court art. 35, Feb. 19, 2013, O.J. C 175/1 [hereinafter UPCA]; PMAC, Patent Mediation and Arbitration Centre, https://www.pmac-upc.org (describing the Centre as an integral part of the UPC with embedded governance and financing).
• 7PMAC Arbitration Rules arts. 2, 4–8, Admin. Comm. Decision D-AC/07/04112025_rev.2 (Apr. 24, 2026).
• 8UPC Rules of Procedure rr. 11, 365; UPCA arts. 35, 79, 82.
• 9PMAC Arbitration Rules arts. 7–8.
• 10PMAC Arbitration Rules arts. 9–12.
• 11PMAC Arbitration Rules arts. 15–21, 40–41.
• 12PMAC Arbitration Rules arts. 23, 26–28, 47.
• 13PMAC Arbitration Rules arts. 24–25.
• 14PMAC Arbitration Rules arts. 47–50.
• 15PMAC Arbitration Rules art. 36; UPC Rules of Procedure rr. 11.2, 365.
• 16

  UPC Rules of Procedure r. 370 and Table of Fees

• 17PMAC Schedule of Arbitration Fees pts. 2–5, Admin. Comm. Decision D-AC/10/04112025_rev.2 (Apr. 24, 2026).
• 18PMAC Arbitration Rules arts. 39, 51–54; PMAC Schedule of Arbitration Fees pts. 2–5.
• 19 
• 20PMAC Schedule of Mediation Fees, https://www.pmac-upc.org.
• 21PMAC Expert Determination Rules; PMAC Schedule of Expert Determination Fees, https://www.pmac-upc.org.
• 22UPCA art. 35(2) (the Centre may provide mediation and arbitration facilities, but a patent may not be revoked or limited in those proceedings); UPCA arts. 79, 82; UPC Rules of Procedure rr. 11, 365.
• 23

  See generally Megan M. La Belle, Patent Law as Public Law, 20 Geo. Mason L. Rev. 41 (2012).