The valuation of SEPs – The Munich Regional Court’s view on how to determine a FRAND rate

In the third piece of our series offering commentary on the recent Munich Regional Court guidance order regarding FRAND cases, we delve into various valuation approaches for SEPs. We also analyse the Munich Regional Court's view on how FRAND rates should and should not be determined. Finally, we offer a forecast on how the UPC will position itself on this question. We also address the court's conception that generally cross-licenses cannot be demanded, as suggested by the Munich Regional Court. We show that this concept is questionable with respect to the FRAND declaration.

It is at the core of any SEP litigation: Who gets what amount from whom. The question of the direction of payment is straightforward in the case of unidirectional SEP licensing, i.e. from an SEP holder who often is not active in the market for the standardized product to an implementer who incorporates the standard into its products. Obviously, the implementer will pay the royalties to the SEP holder, who may or may not be a non-practicing entity (NPE). The situation becomes significantly more complex when two SEP holders who are also implementers in the relevant product markets disagree not only on the price but also on the direction of payment.

Here is what the Munich Regional Court thinks about FRAND rate determination. It is important to note that, to date, German courts have only determined whether a given offer is FRAND in the context of assessing a FRAND defence raised against the assertion of a SEP but have not actually determined a FRAND rate.

• The Munich Regional Court prefers comparable license agreements as market-based benchmarks. It emphasises that FRAND rates are best identified through real-world licensing agreements rather than attempting to value the entire standardized technology. Comparable licenses provide market-driven evidence of what parties have agreed to pay under conditions that can be scrutinised (e.g. similar scope, timeframe, product range), the court states. Real-world agreements, according to the Court’s view, reflect practical business negotiations and thus are deemed a stronger indicator of an appropriate FRAND rate. The court notes that “market economic principles are suited best” meaning free-market practices offer the most realistic benchmark for determining value.
• When relying on comparable licenses, the court looks for agreements that cover the same (or sufficiently similar) patent portfolios and product categories, signed within a reasonably recent timeframe. It also excludes outliers: extremely high or low royalty rates are discarded in the comparison. The court prefers a “FRAND range” rather than a single, rigid rate. It examines whether the proposed rate fits within the range established by comparable licenses, after factoring in the peculiarities of the specific licensing scenario (e.g. large volumes, lump-sum vs. per-unit payments).
• The court stipulates that the interval between the “most favorable” and the “least favorable” comparable agreement could be as much as threefold. Within this spectrum, a case-by-case, value-based assessment determines whether a given rate still falls into the FRAND range. Where there are numerous existing comparable licenses from the patent holder, the court insists on a minimum leeway of 15%. This ensures the patent holder is not locked into a rate forever and can adjust within at least that margin, the court states.
• The court considers the top-down approach to be “problematic,” mainly because it is extremely difficult to assess the total value of the entire standardised technology or to allocate fractional contributions accurately among numerous SEPs. The court acknowledges that a top-down calculation may serve as a “plausibility check” to corroborate licensing rates already derived from comparable licenses, but it should not be the main method for setting FRAND rates. According to the court, the top-down approach requires artificially assigning a global value to the standard and then allocating portions of that value to each patent, which is neither straightforward nor reliable.
• Regarding cross-licenses between two SEP holders who are also implementers and must license in the respective portfolio of each other, the court states that each of the mutual licenses is to be considered independently and regularly no claim can be made to get a cross-license.

Legal basis

The court does not provide a detailed legal justification for its approach, and such a justification is not necessary. It is generally accepted that the specific licensing conditions to be assessed as FRAND are to be benchmarked under the principles of antitrust law. Antitrust laws govern the conditions under which access to the standardized product market must be granted. The conditions must be fair, reasonable, and non-discriminatory (FRAND). It is evident that the non-discriminatory aspect can be most effectively evaluated by comparing license agreements with competitors of the implementer seeking a license. The implementer seeking a license must not be discriminated against the competitors who already obtained a license. This means that when the circumstances align (markets concerned, products concerned, territorial scope, sales volumes, etc.), the licensing conditions must be sufficiently similar. Any discrimination against established licensing practices in a single case must be justified.

The question of what is to be considered fair and reasonable in the given licensing context is not addressed. The Munich Regional Court has taken a different approach. It appears to view FRAND as a comprehensive concept that stands apart from its constituent parts, which are defined as "fair," "reasonable," and "non-discriminatory." This allows for a consistent evaluation of FRAND. The Munich Regional Court may deem deviations from the terms and conditions included in comparable license agreements as fair and reasonable, provided that such deviations remain within the established parameters.

The court has not provided an explanation for its position on cross-licensing. However, it has been claimed that negotiating a cross-license may present difficulties, including those related to the securities required to be provided to secure the claim of the SEP holder during ongoing negotiations.

How will the UPC assess FRAND?

Although the guidance order lacks binding effect on the UPC, it may receive attention in SEP-related UPC proceedings. This could occur if any of the parties in a UPC proceeding refer to it, or if the court decides to address it on its own initiative. So far, the UPC has addressed FRAND in two decisions Panasonic vs. Oppo and Huawei vs. Netgear. In Huawei vs. Netgear the UPC Munich Local Division refers to comparable license agreements as viable method for determining whether the offer of the SEP holder aligns with the FRAND principle (it did not apply the method, though, as it found the implementer unwilling). In the case of Panasonic vs. Oppo, the UPC Mannheim Local Division ruled that the presentation of the top-down calculation of the rate offered by the SEP holder was sufficient to trigger the implementer's obligation to make a counteroffer. In a second step, the offer was benchmarked against comparable license agreements. In the Huawei vs. Netgear case, the UPC Munich Local Division has indicated that there may be "fair" discrimination, suggesting a shared understanding of the FRAND principle with the Munich Regional Court.

Thus, we are expecting the UPC to apply a similar approach to the Regional Court Munich and put most weight on comparable license agreements when determining FRANDness. Whether the UPC will disregard other approaches to determine whether a given offer is FRAND, i.e. top-down approach, or cost-based approaches, remains to be seen. The decision between Panasonic and OPPO indicates that the UPC Mannheim Local division is receptive to the top-down approach as a valid method for determining FRAND.

What difficulties does the court face with its valuation approach

We believe that the current approach, which is to focus exclusively on comparable license agreements, is not optimal. This approach is challenging from an antitrust perspective. It is important to note that it does not fully implement the SEP holder’s commitments previously established with the Standard Setting Organization (SSO), specifically the FRAND licensing declaration. This declaration is a fundamental prerequisite for the successful adoption of the proposed technology as standard. It is further imperative to acknowledge the inherent technological interconnection between all SEPs to a particular standard. Failure to recognize this linkage poses a significant risk, as it may lead to diminished competition and elevated pricing, ultimately affecting consumers. Finally, in a cross-licensing context, there are mutual claims to get a license on FRAND terms, as specifically provided for in the FRAND licensing declaration itself.

The definition of unlawful price discrimination under the relevant antitrust law is clear. Non-discrimination does not imply that each licensee must receive the same price. Discrimination occurs when similar implementers have access to the standard with significant price differences without justification. The Munich Regional Court, in its guidance order, has granted the SEP holder significant discretion in setting prices and has indicated that it may find license fees up to three times the standard as still within the FRAND range. We have concerns about whether this view meets established antitrust law standards for market-dominant entities as SEP holders, as it lacks justification. The mere will of the SEP holder to generate higher revenue may not be sufficient to justify such high price discrimination. The same applies to the court’s view that an increase by 15% from one license case to another appears acceptable, just because FRAND rates would otherwise perpetuate forever. If inflation does not provide a sufficient rationale for such a high adaptation rate, what other factors could?

The FRAND declaration itself further refutes the Regional Court's approach. FRAND includes a three-fold definition of the conditions for access to the standard. Access conditions must not only be non-discriminatory but also fair and reasonable. The fair and reasonableness requirements are not only decorative ornaments, which just describe the non-discriminatory part. The fair and reasonable requirements have their own justification in the specific circumstances of how and what technology is adopted as standard. The adoption is only made subject to the condition that the SEP holder grants access on FRAND terms and conditions and gives a respective irrevocable FRAND commitment. Standardization diminishes inter-technology competition. The SEP holder is only permitted to contribute its technology to a comprehensive set of standardized technologies, subject to the FRAND commitment, because the standardization cuts out competition among various technologies. By getting its technology adopted as standard, the SEP holder is typically ensured of the widest spread dissemination of its technology across numerous markets, both geographically and product-wise, resulting in sales figures that would be unattainable without standardization. This competitive benefit is the result of the standardization but not of the technology contributed. The standardization avoids the very high risk of total failure and sunk costs in markets with strong network effects, here the standardization guarantees that the technology contributed gets an exploitable value at all. That is the reason why FRAND rates not only must be non-discriminatory but also must be fair and reasonable as such, i.e. with respect to the competitive position the SEP holder achieves by getting its technology adopted as standard. Focusing only on comparable licenses for determining FRAND, as the Regional Court proposes, disregards the value of the position the SEP holder achieves through the standardization, and the massive benefits standardization has for its business.

Additionally, the focus on comparable licenses overlooks the technological connection between all SEPs and a specific standard. Current industry standards, including 5G cellular technology, Wi-Fi, and video coding, specify comprehensive technology platforms to which each patented item contributes only a minimal technical function. The key point to consider is that a single item contributing to a standard cannot be exploited independently from all the other technologies adopted. All SEPs are technologically interrelated. The majority of SEPs encountered during my professional life are not comprehensible or useful without implementing thousands of other patents covering the technological basis. The participants in the standardization work who propose that their technology be adopted and give the required FRAND declaration to achieve this are aware of this fact. They are aware of all the other contributors to the standard, who have their own claim for a fair and reasonable reward on their contribution to the standard. A focus on comparable licenses, which only cover a portion of the total patent stack essential to a standard, disregards the technological link and the purpose of the FRAND declaration. The latter is intended to ensure fair and reasonable compensation for all contributors to the standard.

Contrary to the court’s opinion, it is also possible to determine the value of a standard for a particular product. Determining the value of a specific technology for a product is a daily task for all individuals involved in determining employee inventor remuneration. This task remains largely unchanged when it comes to standardized technology. And it is evident that it is being performed because all employees who invent technologies that are proposed for adoption as standard receive compensation based on the determination of the value that the invention adds to the standardized technology.

It is also possible to analyze third-party license agreements to understand the third parties' agreements regarding different SEP portfolios related to the same standard and what value these parties assigned to the complete stack of SEPs to a standard. If the relevant data are properly applied over term, sales volume, cost of capital, inflation, etc., the entire unpacking process can be reduced to two simple questions: What is the price for the total stack, i.e., the standardized technology at a given point in time, and what is the share the SEP holder has in the total stack of all SEP relevant to the standard? This suggests that the Regional Court's approach of utilizing the top-down method solely as a plausibility check may not be the appropriate strategy.

Finally, we would like to address the Court’s statement that cross licensing regularly cannot be demanded. In the given case, the parties both contributed significantly to the standards at issue, a situation that was foreseen by the standard setting organization as usually participants of a standardization project are competitors who also implement the standardized technology in their own products. Accordingly, most – if not all – FRAND declaration forms permit the licensing of SEPs, contingent on reciprocity. This means that the licensor has the right to request reciprocal licensing from the implementer who is seeking a license. The purpose is apparently to avoid giving one contributor to a standard the power to exclude other contributors to the standard from its implementation and deadlock situations where two SEPs holders mutually block each other. Such deadlock risks the success of the standard in terms of market adoption as such. Contrary to what the Munich Regional Court states, there is a claim for a cross license founded in the FRAND declaration itself and no reason to look at each mutual license individually.

For more information, contact your CMS client partner or these CMS experts: Stephan Dorn.

To read the full text order click here.