How courts set FRAND rates: what the EPO’s new study means for implementers

FRAND rate determination by courts now follows a recognisable pattern: judges rely mainly on comparable licences and use the top-down method mostly as a cross-check. That is the central finding of “Methodologies for FRAND determination”, a study by the European Patent Office (EPO) Observatory presented on 24 June 2026, which reviews 65 court decisions from seven jurisdictions between 2013 and 2025. For SMEs building connected devices for Europe or Latin America, it shows how a court would test the royalty they are being asked to pay for standard essential patents (SEPs).

Key takeaways

  • The study covers 65 decisions from the US, UK, Japan, India, China, Germany and the Netherlands, prepared by the EPO with the consultancy BRELA.
  • Of 19 rate-setting decisions with a classifiable method, 13 relied mainly on comparable licences, five on top-down and one on both.
  • The recurring aggregate royalty figures for 3G and 4G persist largely because courts cite earlier decisions, which the study says raises questions about their robustness.
  • Patent counting is central to both methods but contested, especially the gap between declared and truly essential patents.
  • Germany is the main forum for assessing whether licence offers are FRAND, joined recently by the Unified Patent Court (UPC).

What did the EPO study find?

FRAND (fair, reasonable and non-discriminatory) is the commitment that standard-setting organisations require from owners of patents essential to standards such as 5G, Wi-Fi or video codecs. It does not fix a rate, so disputes arise. The EPO announced the study on 24 June 2026 as the most comprehensive global corpus of such court decisions to date.

According to the executive summary, the corpus has three parts: 20 cases and 33 decisions in which a court set the rate (intended to be exhaustive up to March 2026), 19 decisions assessing whether a given offer was FRAND, and 13 decisions on whether a method is admissible. The study is descriptive: it does not recommend a method. Seventeen national offices, including Spain’s and Portugal’s, took part. The Spanish Patent and Trademark Office (OEPM) reported it the same day.

How do courts set FRAND rates?

Comparable licences Top-down
How it works Derives the rate from real licences between similar parties Fixes an aggregate royalty for all SEPs in the standard, then gives the patent owner a share
Use in the case law Primary method in most decisions Mostly a cross-check; primary where comparables were missing or unreliable
Main difficulties Choosing licences, “unpacking” lump sums and cross-licences, adjusting for different portfolios Setting the aggregate rate and apportioning it, usually by patent counts
By jurisdiction UK and German courts rely on it, using top-down only as a cross-check US and Chinese courts have used it as a primary method

Courts are offered far more licences than they accept: the number proposed in the surveyed cases ranged from 2 to 54. Bottom-up, cost-based and theoretical approaches such as the Shapley value have gained no traction, and some courts rejected them outright.

Why patent counts and essentiality matter

Both methods need a measure of how strong one portfolio is against the rest. Courts have generally preferred counting patents over counting contributions to the standard, but have warned that “mere patent counting and dividing is not enough”. The open issues the study identifies are the gap between patents declared essential and those that truly are, the treatment of pending and expired patents, and the lack of an accepted way to factor in validity.

For implementers, this is where negotiation leverage lies: a portfolio that looks large on declarations may be much smaller once essentiality and validity are tested.

What this means for your business

  • Ask the SEP holder for claim charts and evidence of essentiality, not only declaration lists.
  • Ask which comparable licences support the rate and how lump sums were converted; request them under a confidentiality agreement.
  • Test the aggregate royalty the offer assumes, and the share attributed to the licensor’s portfolio.
  • Keep a dated record of offers, counter-offers and your willingness to take a licence: in European injunction cases, conduct during negotiation weighs heavily.
  • Consider confidential forums: the EPO itself points to the new UPC Patent Mediation and Arbitration Centre for licensing disputes.

If your product implements cellular, Wi-Fi or video standards and you are entering several markets, our patent and SEP licensing team can review the offer before you sign.

Where implementers get this wrong

  • Ignoring the first letter. Delay can be read as unwillingness, which weakens your position in European courts.
  • Accepting a headline rate without asking how it was built, which comparables support it and how many patents are truly essential.
  • Negotiating country by country when the licence is global, and paying twice for the same portfolio.
  • Assuming the case law applies everywhere. The study covers no Latin American or African court, so local law and practice must be checked separately.

Our IP licensing and disputes team coordinates SEP negotiations and, if needed, litigation or mediation, with local counsel in each country.

Frequently asked questions

What is FRAND rate determination?

It is the process by which a court, arbitrator or the parties fix the royalty for a licence of standard essential patents on fair, reasonable and non-discriminatory terms. Because FRAND commitments do not set a figure, courts rely on methods such as comparable licences or a top-down calculation to reach one.

Which method do courts prefer for setting FRAND rates?

According to the EPO study, comparable licences are the main method: they were the primary approach in 13 of the 19 rate-setting decisions where the method could be classified. The top-down approach was primary in five, mostly where suitable comparables were unavailable, and both were used in parallel in one case.

Does the EPO study cover Latin America?

No. The study reviews decisions from the United States, the United Kingdom, Japan, India, China, Germany and the Netherlands. Its findings help any negotiation, but companies selling in Latin America or Africa need to check how local courts and competition authorities treat SEP disputes before relying on them.

Can IP Global Guard help with a SEP licence negotiation?

Yes. We review the offer, the portfolio and the comparables, coordinate technical essentiality checks with European patent attorneys, and support the negotiation or any mediation or litigation. We coordinate qualified local counsel in Europe, Latin America and Africa, with one point of contact.

How IP Global Guard can support your SEP negotiations

The EPO study gives implementers a clear map of how courts test FRAND offers, and the questions to ask before paying. IP Global Guard, the IP services line of META Channel Corporation Limited, handles patents and licensing across more than 25 jurisdictions in Europe, Latin America and Africa, with one strategy and one billing relationship.

Share the licence offer you have received, the standards your product uses and the markets you sell in. We will tell you which points to test and how to structure your response. Talk to our licensing team.

This article is general information, not legal advice, and reflects the position on the date of publication.

Sources