3 Valuation from comparable contracts

Topics covered in Part 3:

• Overview of valuation from comparable contracts

• Step 1: Identifying transactions with comparable characteristics

• Step 2: “Unpacking” the rates agreed for comparable contracts

• Step 3: Adjusting unpacked rates to estimate comparable prices

• Assessing whether comparable contracts provide prices within the FRAND range

3.1 An overview of valuation from comparable contracts

Parties, courts and arbitrators often estimate reasonable terms for an SEP license (the Focal Contract) by reference to comparable contracts. In this Report, “comparable contracts” means licenses that have sufficiently similar “characteristics” – with regard to the licensed patent rights, the products implementing those rights and other circumstances that affect the commercial substance of the negotiation and, therefore, the value of the agreement. ⁽¹⁾Generally, the outcomes of arbitrations and the valuation methodologies they use are confidential. Nonetheless, in the contributors’ experience, reference to comparable contracts is common. Publicly available court judgments informed by comparables valuation are listed in Part 3.4 below.

3.1.1 The rationale

Inferring the reasonable price for an asset from transactions for similar assets is a “market-based” valuation methodology. Its central premise is that independent parties facing similar circumstances will tend to agree similar prices for similar assets.

The task is therefore to (i) identify transactions that are genuinely comparable and (ii) use the terms agreed in those transactions to estimate terms that may be reasonable for the license being valued (the Focal Contract), taking account of relevant differences in structure and circumstances that affect the terms which parties agree.

This approach is widely used in valuation in general. A familiar example is property valuation. Although each property is unique, sale prices for properties with similar relevant features – such as location, floor area and condition – can indicate a plausible range for the property being valued. Comparable properties need not be identical. They may differ in ways that affect price; where that is the case, the valuation may adjust observed prices to better reflect the specific circumstances of the property being valued.

The same basic logic can be applied to SEP licensing. The terms agreed in comparable licenses may indicate a plausible range for the Focal Contract to the extent that those licenses were negotiated at arm’s length, in sufficiently similar commercial circumstances, for sufficiently similar patent rights used in sufficiently similar products.

The practical appeal of this approach is that it relies on real-world agreements in the relevant market. In some cases, it may therefore rest on fewer, and more readily verifiable, assumptions than approaches that seek to derive a royalty directly from estimates of the value contributed by the licensed patents.

However, the use of comparable contracts in SEP valuation raises two issues that require particular attention:

• Whether contracts are genuinely comparable. A contract is informative only to the extent that the characteristics and circumstances that materially affect price are similar to those of the Focal Contract. In general, the more similar contracts are with respect to the characteristics that determine their value, the more informative the comparison is likely to be.

• Whether contracts have reasonable terms. Even where contracts have similar characteristics, the terms agreed in relation to them may not be reasonable. They too were agreed by parties without competitive outside options, and may unreasonably reflect the impact of litigation risk, bargaining power, information asymmetries or strategic considerations.

Accordingly, in SEP valuation one should not assume that a license is comparable, nor that its price is necessarily reasonable. Both propositions must be examined in practice, otherwise there is a risk that not only is the valuation inaccurate but also that the methodology can proliferate non-FRAND prices through the market, resulting in them becoming entrenched.

3.1.2 The approach

At a high level, the comparable contracts approach involves three conceptually distinct steps. They are presented in Box 3.1 in a logical sequence but, in practice, an analyst may apply them iteratively – revisiting earlier steps as additional information becomes available, depending on the specific facts of the valuation.

In principle, the risk that hold-up or hold-out influences the terms agreed in SEP licenses can limit the extent to which comparable licenses provide a reliable indication of FRAND terms for the Focal Contract. To assess the plausibility of that risk, a comparables analysis can be complemented by an assessment of whether the rates inferred from comparable contracts are reasonable in context.

Such assessment can take several forms (which are not mutually exclusive), including the following:

• Assessing the bargaining process for indications of hold-up and hold-out. This assessment considers whether features of the negotiation (for example, time pressure, financial pressure or litigation) may have affected the rate ultimately agreed. As far as imbalances in bargaining power result in non-FRAND outcomes, contracts that would be susceptible to such imbalances should be given less or no weight in the assessment.

• Assessing the breadth of the rates inferred from comparable contracts. The rates inferred from comparable contacts typically form a range, even when differences have been adjusted for to fullest extent possible. However, that range should be narrow. Otherwise, the central premise that underpins this approach would be wrong: that similar parties in similar circumstances will agree similar prices.

• Assessing the rates against an external benchmark of value. Even a narrow range of agreed terms may not converge on royalties that are genuinely FRAND. Cross-checks against external indicators of value are therefore helpful for evaluating the rates that parties agreed to in practice. Courts in several jurisdictions have considered such cross-checks when assessing the royalty rates estimated from comparable contracts.

3.2 Step 1: Identifying transactions with comparable characteristics

This section sets out an analytical framework for identifying contracts whose characteristics are sufficiently similar to those of the Focal Contract to inform the assessment of FRAND terms.

3.2.1 What does it mean to have comparable characteristics in principle?

3.2.1.1 What makes a contract comparable?

A genuinely comparable contract is one where the circumstances and characteristics that materially determine its price and value are sufficiently similar to those of the Focal Contract, such that willing parties would agree similar terms for both contracts. In practical terms, the question is whether willing parties negotiating the Focal Contract would reasonably regard the other contract as informative when negotiating FRAND terms, given their similarity.

A contract will not be comparable where differences in its commercial substance are so material that its agreed terms would not be suitable for the Focal Contract. Conversely, where two transactions are truly like-for-like in their commercial substance, materially different prices may be difficult to reconcile with the idea that similarly situated parties would agree similar terms (which may raise questions regarding the non-discrimination principle in FRAND).

Because comparability depends on context, rigid checklists can be misleading. Nonetheless, in SEP licensing, factors commonly treated as relevant (or argued to be relevant) tend to fall into four broad categories: (i) the licensed SEPs, (ii) the implementing products, (iii) the parties and (iv) the contract terms. Box 3.2 provides non-exhaustive examples.

3.2.1.2 Exact parallels are ideal, but unnecessary

Comparable contracts rarely match the Focal Contract in all respects. That is not, by itself, a reason to discard them. In many cases, independent parties negotiating the price of the Focal Contract would find informative the price agreed for a contract that is broadly similar, even if it differs in some ways that affect its price.

What matters is the nature of the differences between contracts, and how materially and predictably they affect the prices that parties agree.

• Crucial characteristics. Contracts may differ in a way that makes them not comparable, even if they are similar in all other respects. Similarity with respect to these characteristics is essential for a contract to be informative. For instance, the similarity of the licensed SEPs and products typically falls within this group.

• Material characteristics. Contracts may differ in ways that materially affect the price that parties agree on. However, if the contract is otherwise similar and the impact that a particular difference in circumstance has on price can be taken into account, then the contract may still be informative. For instance, specific features of the product market, parties and contract terms typically fall within this group.

• Trivial characteristics. The fact that certain characteristics may differ will be irrelevant to whether contracts are comparable provided that the differences are immaterial to the prices that parties agree.

The contracts shown in Box 3.3 illustrate the above points.

3.2.1.3 A structured approach to assessing comparability

In practice, it is often prudent to begin by identifying contracts that are broadly comparable (for example, licenses covering similar SEPs in similar product markets), rather than rejecting potentially informative contracts too early in the process.

Candidate contracts can then be assessed in more detail to determine:

• the extent to which the remaining differences may materially affect prices;

• whether those differences can be taken into account when valuing the Focal Contract; and

• whether some contracts constitute superior comparisons to others.

The detail of an assessment may vary depending on the facts of the negotiation and the contracts available. Nonetheless, a structured approach helps to identify areas of potential disagreement between the parties, and reduces the risk of inconsistency, selective inclusion (“cherry-picking”) and error.

We address practical and specific considerations in Section 3.2.2 below. However, in many cases, their impact can be navigated or mitigated by applying the following broad principles.

• Explicit and transparent criteria for “comparability”. First, the characteristics and circumstances that affect the value of a license and the terms that parties agree need to be identified. They are typically determined by referring to commercial experience, economic rationale and evidence on conventional industry practice.

• A clear rationale for their material impact on royalties. The factors that parties identify as relevant should be grounded in an economic rationale that explains why differences in that characteristic would affect the terms that parties agree. The stronger the rationale, the more confidence can be placed on the assessment.

• Broadly, closer comparables are more informative comparables. Generally, all other things being equal, a contract that more closely resembles the disputed contract with respect to a given characteristic will be a better comparable than contracts that are less similar. This broad observation tends to underpin how courts and arbitrators consider what it means to be “comparable.” ⁽²⁾For instance, in English law, the Court of Appeal in Optis v. Apple [2025] EWCA Civ 552, par 35, quoting Smith Kline & French Laboratories Ltd’s (Cimetidine) Patents [1990] RPC 203 (Lloyd LJ).: “The object of the comparability exercise, in this as in any other branch of the law, is to find the closest possible parallel. If there is an exact parallel, there is no point in looking any further. If there are slight differences, an allowance may be made. But once you have found your comparables, whether one or more, which enable you to arrive at the appropriate figure, it would surely be erroneous to modify that figure by reference to other cases which are not truly comparable at all, so as to bring the case into line with a predetermined range. This was, with great respect, the mistake which the hearing officer made.”

3.2.2 Practical considerations when identifying comparable contracts

Several practical considerations must be assessed when identifying comparable contracts. The main issues and best practices for addressing them are set out below:

• assessing the risk of unobserved characteristics

• scrutinizing characteristics with an ambiguous impact on price

• assessing the risk of changing circumstances

• analyzing a single price for a bundle of technologies and products

• assessing the risk of using a limited number of potentially relevant transactions

• accounting for the different nature of licensors’ comparables and licensees’ comparables

• accounting for differences in the geographic distribution of patents and products.

3.2.2.1 Assessing the risk of unobserved characteristics

A comparability assessment necessarily relies on differences in observed characteristics. This creates a risk that unobserved factors affecting price are omitted from the assessment in such a way that two contracts appear comparable when their underlying commercial substance differs.

The property analogy is helpful: even with information on location and size, sale price may reflect unobserved factors (for example, disrepair or an urgent requirement to sell). Similarly, SEP license terms may reflect factors that are not apparent from the agreement alone or from market data – for example, time pressure, strategic settlement motivations or information asymmetries.

Broadly, there are two considerations when assessing the risk of unobserved factors:

• Characteristics that commonly and materially affect contract value. In principle, the main drivers affecting the terms that parties agree on will be observable, either directly or indirectly; if they were impossible to observe, then it is less likely that parties would have been able to take them into account when agreeing prices. Therefore, the risk of unobserved differences between contracts can often be assessed and mitigated to some extent. For instance, more granular information may be available from the contract terms and supporting materials, market data on products and volumes and, where available, evidence about the negotiation context. While not all material factors will be observable, systematic scrutiny reduces the likelihood that important differences have been overlooked in the analysis.

• The number of potentially comparable contracts. Where there is a large number of comparable contracts, the risk of unobserved factors may be lower, either because it is easier to identify gaps (where seemingly comparable contracts fail to converge to a narrow range of terms) or because the impact of unobserved factors may be cancelled out to some extent – provided that unobserved factors do not skew the terms agreed for contracts in the same direction. However, the larger the number of contracts that are used in the pool of comparables, the larger their disparity of relevance is likely to become, risking the inclusion of contracts with weaker comparability. There is a trade-off between seeking to rely on a greater number of comparable contracts and seeking to rely on more closely comparable contacts.

The risk of unobserved characteristics cannot be ruled out entirely, as it is not possible to prove a negative. However, it can be mitigated. The more systematic and thorough the assessment of each of the factors that typically and materially affect the prices that parties agree for contracts, the less likely it is that material factors have been overlooked.

3.2.2.2 Scrutinizing characteristics with an ambiguous impact on price

In some markets, identifying the main features that determine prices and value is relatively straightforward. Their relevance is an accepted feature of commercial practice and the economic rationale that explains why differences affect the prices that parties agree is easy to articulate and demonstrate.

When comparing patent licenses (including SEPs) it may not be clear whether a particular characteristic materially affects the price that parties negotiate. In particular, a comparability assessment should assess the risk of the following differences:

• Superficial differences. These refer to characteristics that differ formally, but the difference has little or no impact on the economic and commercial substance of the license. For instance, in Box 3.4 the Focal Contract has the same formal scope of licensed SEPs and products as Contract A, and Contracts B and C have a narrower scope. However, in practice, the commercial substance more closely reflects Contract B. Superficial differences can be included for strategic reasons – obscuring the contract’s commercial substance to increase or decrease its apparent similarity when used as a comparable contract in subsequent negotiations.

• Uncertain differences. These refer to characteristics that differ, but the impact on price may be genuinely ambiguous. For instance, using the example in Box 3.4, the Focal Contract does offer an “option value” ⁽³⁾In this context, the option to launch new products under the same license. that Contract B lacks. If Party B, the implementer party of the Focal Contract, launched a new product, say Product Z, during the license term, those units would be covered. Whether that option value is material is unclear. If the implementer has no intention, opportunity or capability to go to market, then it is likely to be trivial. If it has all three, it could be material.

The risk and impact of superficial or uncertain differences must be assessed. This avoids excluding relevant contracts that are meaningful comparisons and including misleading ones. The specific approach may vary with context but essentially involves scrutinizing the economic rationale. For example, in Box 3.4 the wider product scope is not valuable to the Focal Contract in abstract; it depends on the materiality of the additional licensed units.

3.2.2.3 Assessing the risk of changing circumstances

In principle, the impact that a given characteristic has on contract terms may change over time. A more recent transaction may therefore be a better point of comparison than an older transaction that appears otherwise similar.

The value of a licensed technology can change for several reasons. Early transactions may reflect greater uncertainty about market development, adoption or demand for the functionality that the technology enables. Standards are also frequently revised, which can alter the functionality, performance and commercial attractiveness of compliant products. In addition, the apparent strength and scope of a licensor’s portfolio may strengthen or weaken over time, as patents are granted, invalidated, supplemented or expire. For these reasons, the price that willing parties would agree for the same rights, used in the same products, may reasonably move up or down over time.

However, in SEP settings, allowing for changes in the rates that parties agree on introduces a particular risk. Once a standard’s network effects have removed competitive alternatives, it is particularly difficult to distinguish between changes in circumstances that should reasonably affect the prices that parties agree from ex post opportunism that should not affect prices. In practice, changes in prices may be influenced more by hold-up or hold-out tactics than by changes in relevant circumstances. So, potentially, an older license could be a better comparable than a recent license when considering all factors together.

Accordingly, it is better to identify the underlying factors that directly affect contract value and that may have changed over time. For instance, rather than assessing differences in contract age, one may be able to assess changes in the indicators of portfolio strength over time, the mix of products implementing the latest generation of the standard or product volumes and price points.

Where such assessments are not feasible, contract age may in some cases provide a pragmatic filter for contracts – for example, excluding contracts that are older than a specified period.

However, although circumstances may change in principle, that does not mean they can be treated as a “fudge factor” (i.e. an arbitrary adjustment) in practice. Any potential source of changing circumstances should be specifically articulated and there should be an economic rationale for it. This means that, in principle, older licenses that are more comparable along other dimensions (provided that changes in other circumstances can be controlled for) may be a better comparable than a more recent contract.

3.2.2.4 Analyzing a single price for a bundle of technologies and products

SEP licenses often cover multiple technology standards and multiple products. For instance, a contract may cover both cellular and Wi-Fi SEPs for use in several types of products, such as smartphones, tablets, household IoT and electric vehicles.

In principle, the value of each standard to each product may vary, depending on whether:

• a particular product type implements the specific standard at all, and

• how much value that product gains from the particular standard’s functionality.

However, in practice, parties often agree a single price for the contract as a whole – without breaking down how that price relates to each component in its bundle of technologies and contracts. This can complicate comparability where contracts differ in scope or product mix.

Box 3.5 illustrates why it is important to understand how differences in product scope and mix may affect the terms that parties agree on for otherwise identical licenses. As we discuss below, in practice, these differences can be adjusted for – for instance using agreements with narrower scope for a single standard in a single product category.

3.2.2.5 Assessing the risk of using a limited number of potentially relevant transactions

When analyzing SEP licenses, relatively few comparable licenses may be available. For emerging technologies or in new product markets, there may be few (if any) prior agreements. Even mature product markets may be concentrated, meaning that there are only a handful of comparable manufacturers for a given product. Moreover, SEP licenses are typically confidential, so the pool of potentially comparable contracts is often limited to those disclosed by parties to the relevant license negotiation or dispute.

In principle, a limited number of transactions is not necessarily a barrier. A good comparable is a good comparable. However, the greater the number of observations that are available, the more they should corroborate one another and result in a clearer understanding of how the differences in their circumstances affect prices. As discussed above, with a small number of observations, the risk of exposure to unobserved factors is greater and the ability to reliably understand the significance of variation and adjustment for them may be limited. There may be a trade-off between relying on a greater number of comparables, and relying on better comparables.

3.2.2.6 Accounting for the different nature of licensors’ comparables and licensees’ comparables

Typically, each of the negotiating parties can draw on its own set of potentially comparable licenses. This can assist the valuation process by increasing the number of informative contracts available.

However, a licensor’s and a licensee’s potentially comparable contracts are different.

• The licensor may reference prior agreements for the same (or a similar) SEP portfolio licensed to various other implementers, whose products may or may not closely resemble those in the Focal Contract.

• The licensee may point to its own licenses for the same (or a similar) product portfolio but involving different SEP portfolios essential to the same standard.

Each set of contracts, therefore, requires different types of comparisons.

All other things being equal, the licensor’s comparables would be expected to provide a better starting point than the licensee’s comparables, as they refer to the same licensed portfolio as the Focal Contract ⁽⁴⁾See Optis v. Apple [2025] EWCA Civ 552, para. 95. – although, obviously, there may be important exceptions, such as if the licensor’s license or licenses are not FRAND. The challenge with the licensee’s comparables, therefore, is that comparison requires an additional layer of analysis to determine how similar they are (i.e., that captures the extent to which each of the different portfolios contributes to the standard as a whole). SEP portfolios can differ in size, essentiality, validity and technical importance so these differences need to be assessed when making comparisons across licenses for different SEP portfolios. Nonetheless, where these differences can be assessed reliably, the licensee’s comparables may contribute to the analysis.

3.2.2.7 Accounting for differences in the geographic distribution of patents and products

Because patents are territorial, the value of a portfolio can depend on (i) where the licensee sells products and (ii) where the licensor holds enforceable patent rights. For that reason, parties sometimes agree different effective royalty rates by region or apply region-specific adjustments within a license. In some cases, courts have also considered whether geographic variation affects the portfolio’s value to the particular licensee and whether this justifies an adjustment.

Three drivers of geographic variation are commonly cited:

• Differences in patent coverage. In principle, a royalty is due only when the implementer’s products are made, imported, used or sold in a jurisdiction where the licensor holds relevant, enforceable patents. Where parties negotiate a single "globa"' rate, the territorial coverage of the licensed patent portfolio may therefore be a relevant consideration. Consider a stylized example: an SEP holder’s portfolio only contains US patents. An implementer that sells only in the United States would, in principle, expect to pay a royalty on each unit sold there; an implementer that makes and sells otherwise identical devices entirely outside the United States would, in principle, owe nothing to that SEP holder; and a third implementer with worldwide sales would expect any global payment to reflect its exposure to the US market – for instance, a rate that is scaled (explicitly or implicitly) by the share of its units sold into the United States or by the value of those sales.

• Differences in legal and enforcement conditions. The practical strength of patent rights can vary across jurisdictions (for example, differences in patent examination standards, available remedies, speed of enforcement or local FRAND practice). In addition, competition law interventions may affect licensing structures in particular jurisdictions. For example, following China’s 2015 National Development and Reform Commission (NDRC) investigation, Qualcomm committed to using a royalty base set at 35 percent below the net selling price when calculating royalties for certain devices sold in China. That precedent acts as an effective reduction in royalty-bearing revenue for sales in China. ⁽⁵⁾While this discount was not codified into formal regulation, the Qualcomm decision acts as a de facto benchmark when determining royalty bases for Chinese sales. One of the reasons is that licensees may come to expect such discounts from other licensors based on the expectation that Chinese regulators could bring pressure to bear on such licensors if they do not follow the Qualcomm model. NDRC press release, February 10, 2015.

• Differences in market conditions between regions. Commercial realities may support effective rates that are lower in some markets than in others – for example, where consumers’ willingness to pay is lower or where pricing strategies are designed to promote adoption and scale (including in emerging markets). ⁽⁶⁾For example, see InterDigital v. Lenovo [2023] EWHC 539 (Pat).

The significance of these factors depends on the geographic scope of a license, which in practice ranges between the two extremes outlined below.

• Licenses with global coverage. In many product markets, parties enter global portfolio licenses covering sales across jurisdictions. Courts – notably in the United Kingdom and, in some circumstances, in China and the United States – have also determined terms on a worldwide basis, often on the premise that country-by-country licensing would be commercially inefficient. In this setting, the question is usually not whether each jurisdiction is priced independently, but whether the agreed global structure appropriately reflects material regional differences. For example, where a licensee has above-average exposure to markets with systematically lower effective rates, then the rate it pays for a global license may be lower than it would be with an average regional profile of sales. ⁽⁷⁾For instance, see Unwired Planet v. Huawei [2020] UKSC 37, para. 15, “SEP owners, which have a large portfolio of patents covering many countries, and implementers, which market their products in many countries, would as a matter of practice voluntarily negotiate worldwide licences, or at least licences from which a given territory is carved out while the rest of the world is licensed.”

• Licenses with narrow geographic scope. In some circumstances, a licensee’s sales may be concentrated in a particular jurisdiction or group of jurisdictions. In such cases, special emphasis is placed on whether it is unlikely that the patent will be able to yield the “full” typical rate in that jurisdiction – because it has no or limited patent coverage, enforcement is weak in practice or market conditions produce lower rates. In that case, the portfolio may provide less leverage and its licensing may offer less value to that particular licensee than it would to a licensee with global sales in jurisdictions where the portfolio is strongest.

In practical terms, the issue for parties, courts and arbitrators is how granular the consideration of these issues should be.

In principle, a strict patent-by-patent, country-by-country assessment may be accurate. However, in some circumstances, such an exercise may be sufficiently impractical – at least for negotiations relating to large global portfolios and worldwide sales – that willing parties would not pursue it. So, there can be a trade-off, which is possible to be assessed and mitigated in some cases. For instance, in order for FRAND to mirror the underlying patent rights, it may be possible to assess whether a portfolio is weaker across jurisdictions where the implementer sells or manufactures the licensed products by analyzing the relative number of patents in those jurisdictions.

Alternatively, some have categorized where the implementer sells or manufactures the licensed products in two broad groups: (a) major jurisdictions, where the SEP holder’s portfolio is sufficiently deep (and expected to remain sufficiently robust after validity and essentiality challenges) that the implementer faces a credible risk of effective enforcement; and (ii) minor jurisdictions, where the portfolio is thin or less enforceable, so willing parties would agree a lower rate on sales in those regions. On this view, the practical question is not whether every patent is valid and important in every territory, but whether the licensor has enough enforceable coverage in the commercially significant territories that a willing party would treat the global rate as reasonable. ⁽⁸⁾Even here, it may be contested how many patents a licensor requires for its portfolio to be considered sufficient that willing parties regard it as sufficiently enforceable that they would simply agree to the typical “major jurisdiction” rate. For instance, in InterDigital v. Lenovo [2023] EWHC 539 (Pat), pp. 180–187(paras 758 and 791) Justice Mellor considers that adjustments for weak patent coverage “is not an approach that translates necessarily to larger portfolios” as, for these patent holders, “the prospect of … losing all coverage [is] extremely remote.”

3.3 Step 2: “Unpacking” the rates agreed for comparable contracts

After having identified the relevant transactions, the identified rates need to be “unpacked.” Unpacking aims to express the value of what the parties actually agreed in the light of the circumstances they faced. Payment terms in SEP licenses are often complex and can vary substantially across agreements. Two licenses may provide the same present, cash-equivalent value of consideration to the licensor even where the structure of payment differs (for example, a lump sum paid up-front versus a running royalty contingent on future sales). For the purpose of valuing the Focal Contract, it is therefore the economic value of the consideration that matters, not the particular form the parties chose to express it.

Differences in payment structure can, however, obscure whether two licenses are similar in economic substance. The payment terms in the comparable contracts approach therefore need to be “unpacked” and expressed in a common metric that enables like-for-like comparison.

3.3.1 What it means to “unpack” prices in principle

3.3.1.1 Differences in payment structure

Licensors receive consideration in return for granting a license. Licenses define that consideration through one, or a combination, of the elements in Box 3.6.

3.3.1.2 Normalizing payment terms

Differences in payment structure can obscure the true value of the consideration that parties agree (see Box 3.7). All contracts have different payment structures. Comparing the specific payment terms does not reveal whether the prices agreed for each contract are similar or not.

To compare the value of the consideration agreed in each contract, the specific payment terms of the comparable licenses must be normalized – or “unpacked” – into common, cash-equivalent terms. There are three common approaches to unpacking, detailed below, although alternative metrics may be included in cases where that would be informative.

• Effective unit rates. This metric expresses the value of the running unit royalty that a licensor would accept instead of the payment structure it actually agreed. It states the total cash-equivalent value of the consideration in present value terms, divided by the number of licensed units expected at the time the parties agreed on the terms (accounting for the volatility risk of future volumes).

• Effective ad valorem rates. This metric expresses the value of the running ad valorem royalty that a licensor would accept instead of the payment structure it actually agreed. It states the effective unit rate as a proportion of the ASP of the licensed units. In essence, it shows the total cash-equivalent value of the consideration in present value terms, as a proportion of the licensee’s expected revenue from the licensed units (accounting for the volatility risk of future volumes and prices).

• Effective lump sum. This metric expresses the value of the single and immediately paid lump sum royalty that a licensor would accept instead of the payment structure it actually agreed. This expresses the total present value of the cash-equivalent consideration, normalized for the Focal Contract’s volume and profile of licensed units.

Box 3.8 shows the value of the payment terms for the same five contracts “unpacked” into each of the three metrics. In this illustrative example, the common metrics reveal that the payment terms in all five contracts offer the same value to the licensor; only the structure of the payment terms differs. In principle, the parties should be indifferent to the differences between them.

These metrics provide a means of expressing the economic value of the terms the parties agreed; they are not necessarily the payment structure that the parties would have selected in an alternative negotiation. For example, if the parties had agreed a running royalty rather than a lump sum, this could have involved different administrative and monitoring costs, and different risk allocation, which might have affected the rate.

3.3.2 Practical considerations when normalizing payment structures

In practice, parties and experts have often struggled to agree on unpacked royalty rates – sometimes arriving at very different results, even when analyzing the same contracts. The main sources of difference are typically (i) the valuation of non-cash consideration, (ii) discounting and risk assumptions, and (iii) assumptions about expected volumes, prices and the treatment of past sales.

However, in most cases these challenges are not so hard that they make unpacking an insurmountable barrier to a reliable and informative valuation. Generally, they can be addressed. ⁽⁹⁾See, for example, the considerations in Optis v. Apple [2025] EWCA Civ 552. Key practical considerations include:

• converting non-monetary payments into cash-equivalent terms;

• estimating discount factors;

• accounting for expected volumes (and prices);

• accounting for past sales, subsequently covered by the license;

• sensitivity testing; and

• unpacking the aggregate rate implied for all SEPs.

3.3.2.1 Converting non-monetary payments into cash-equivalent terms

Ideally, non-cash consideration should be converted into its explicit monetary value. For example:

• Cross-licenses. Where two SEP portfolios are cross-licensed, the agreement may include a “net” royalty that reflects the difference in value between the two portfolios. In principle, that net payment can be used to infer the implied “gross” value of one portfolio absent the other, provided that the scope and value of the licensed rights granted by each side can be assessed. This requires careful attention to differences in portfolio strength (including validity, essentiality, infringement and technical importance), and to whether other terms are bundled into the deal.

• Transfers of other assets or services. Patent transfers or cooperation agreements may be easier to value where there is an observable market benchmark. Where no such market exists, other valuation approaches may be required (for example, historical acquisition costs, avoided costs or projected cash flows), acknowledging the associated uncertainty.

The materiality of non-monetary consideration should be assessed, even if the value cannot be established precisely. Where it is relatively minor, it may be preferable to retain the contract as a comparable and test the sensitivity of the analysis to reasonable valuation ranges for the non-monetary element, rather than excluding the contract altogether. This is particularly important because non-monetary terms may sometimes be used strategically to make a contract appear more or less comparable than its economic substance would suggest.

3.3.2.2 Estimating discount factors

The price agreed in a license should reflect the present value of the payments – not just their nominal value. Present value depends on:

• timing: money received sooner is worth more than the same nominal amount received later; and

• risk: uncertain or contingent payments are worth less than guaranteed payments.

Estimating the present value of future cash flow requires a discount factor. In principle, discounting should use a rate that is consistent with the risk of the expected royalty cash flows. Different payment structures allocate risk differently. For example, lump sums have fixed cash payments, so they have low volatility risk. In contrast, unit royalties (i.e., royalties with a fixed cash amount per licensed unit) are contingent on volumes, and ad valorem royalties (i.e., royalties with a fixed percentage applied to a defined royalty base, usually the product’s ASP) are contingent on both volumes and prices. These differences matter when comparing the present value of cash flows with otherwise identical profiles in nominal terms.

In practice, parties may use a company’s cost of capital as a benchmark, but the appropriate rate may differ depending on how closely the royalty cash flows reflect the risks of the broader business. Where inputs are uncertain, parties should test sensitivity to plausible discount-rate ranges.

3.3.2.3 Accounting for expected volumes (and prices)

Unpacking often requires assumptions to be made about expected sales volumes (and, for ad valorem structures, expected prices) over the relevant period. In principle, the relevant expectations are those held at the time by the parties negotiating the contract. In practice, the parties’ contemporaneous forecasts may be unavailable or contested.

In practice, two alternatives can be used as proxies:

• Data on actual sales may be informative, but only to the extent that they reflect the expectations that genuinely informed the agreement when it was signed. That is often difficult to verify and, where actual sales have diverged significantly from original expectations, using these data will distort the analysis.

• Contemporary third-party commercial forecasts are often available and reliable, particularly for large companies in mature product markets. They may be less reliable, or unavailable, when considering the expectations for small implementers, niche markets or new products.

3.3.2.4 Accounting for past sales, subsequently covered by the license

As discussed in Section 1.3.4 above, it is common for SEP licenses to consider both future and past sales. This reflects the purpose of the FRAND undertaking, which is to promote adoption and widespread use of the standard, with the understanding that royalties can be negotiated and agreed on reasonable terms after the technology has already been implemented in products. For purposes of normalization, what often matters is how the total value of consideration relates to the total number of units covered by the agreement, across past and future periods. This may be expressed as a weighted average effective rate across all covered units.

However, when a license covers both past and future sales, there can be disagreement about how to unpack the underlying effective unit rate across all units.

First, the scope of relevant units may be disputed. For instance, where a “past release” clause waives royalties on past sales, the units may be interpreted as irrelevant to the average rate – in formal terms, the royalty is waived, it is not zero. Alternatively, they could be interpreted as licensing the past sales for free (i.e., at zero royalty). The distinction is important as the first interpretation would not include a past release in the rate unpacked across all licensed units. The second interpretation would, and obviously it would determine a lower rate.

Second, the rationale for setting different rates for past and future sales requires scrutiny. One consideration is that the differential may be strategic. In theory, both parties could benefit from agreeing a certain average rate, and then disproportionately allocating the consideration to future sales. It makes no difference to the commercial substance of a lump sum agreement but may affect how the license is perceived as a comparable in subsequent negotiations. Conversely, some consider that heavy discounts on past sales, or appeals to patent limitation periods, are not FRAND. The economic value enabled by the technology does not expire with age and delay may reward or incentivize hold-out tactics. As such, the extent to which particular approaches to past sales (including releases, limitation periods or discounts) are appropriate in FRAND contexts can also raise jurisdiction-specific legal questions. ⁽¹⁰⁾InterDigital v. Lenovo [2023] EWHC 539 (Pat) and InterDigital v. Lenovo [2024] EWCA Civ 743.

3.3.2.5 Sensitivity testing

Parties often unpack different rates from the same comparable contracts. ⁽¹¹⁾For example, in the UK High Court’s analysis in InterDigital v. Lenovo, the experts who were unpacking the same group of comparable licenses derived vastly different effective rates due to differences in assumptions and data. “[I]t will be seen that the experts frequently arrived at and utilised radically different figures for total consideration and total unit sales for certain PLAs [Patent Licence Agreements]. The combination of the different approaches and the use of different data sources meant that comparisons were, at best, difficult to make.” InterDigital v. Lenovo [2023] EWHC 539 (Pat), para. 294. Typically, these disparities result from differences in:

• the data used for the calculations – for example, using different data providers for expected product sales, cost of capital or patent data;

• assumptions and approach – for example, whether to differentiate between different products (such as tablets and laptops) or treat them as the same for the purposes of the analysis; or whether a license covers past sales.

Disagreement and disparity can be mitigated with transparent assumptions and careful sensitivity testing – to assess the extent to which the results depend on those assumptions. This process helps to ensure that any errors or uncertainties are identified and either corrected or appropriately accounted for, so that the prices of the purported comparables can be reliably compared – both with each other and with the Focal Contract.

3.3.2.6 Unpacking the aggregate rate implied for all SEPs

The licensor’s comparables often involve the same portfolio as the Focal Contract, whereas the licensee’s comparables involve different SEP portfolios that are essential to the same standard. The rates unpacked for different portfolios cannot be compared directly without accounting for differences in portfolio strength and contribution to the standard.

One approach is to derive, from the unpacked royalty, an implied aggregate royalty rate for the standard: an estimate of what the implementer would pay if all SEP portfolios were licensed on a consistent basis. This requires an assessment of the licensed portfolio’s relative contribution to the standard (which may depend on more than patent counts and include validity, essentiality, infringement and technical and economic importance). In principle, implied aggregate royalty rates have been used as a cross-check, but they can be sensitive to portfolio-contribution assumptions – especially in the case of small portfolios. ⁽¹²⁾For instance, if a portfolio only has a handful of SEPs then the rate it implies for all SEPs will be much more sensitive than extrapolating an implied rate from a portfolio that contains a substantial proportion of SEPs. Nonetheless, this comparison is merely relative. The sensitivity of a larger portfolio should still be considered and assessed in its own right. In principle, the metric should capture several factors, including the number of SEPs, the proportion that are valid, essential and infringed, and the technical importance of the functionality that the patents protect. In practice, some parties have focused on patent counting as a practical simplification. New metrics are being developed which may provide greater accuracy, although the proliferation of alternative sources can lead to disagreements between parties over which data providers should be used.

Box 3.9 illustrates a hypothetical example, where the implied aggregate royalty rate for the relevant standards is $40 per unit (or 8 percent of an implementing product with an ASP of $500).

3.4 Step 3: Adjusting unpacked rates to estimate comparable prices

Even the closest comparable license may differ from the Focal Contract in ways that make its unpacked rate an imprecise guide to the rate that would be reasonable for the license being valued. In such cases, adjusting the unpacked rates from comparable contracts can improve the precision of the valuation.

“Unpacking” and “adjusting” serve different purposes. As mentioned under step 2 at Section 3.3 above, unpacking aims to express the value of what the parties actually agreed, given the circumstances they actually faced. Those unpacked rates provide the empirical foundation for a comparables-based valuation, but they are only a starting point. Adjustments, by contrast, estimate the rate that might have been agreed in circumstances that more closely resemble those relevant to the Focal Contract (or, in some cases, circumstances consistent with FRAND principles).

For clarity, it is generally preferable to present unpacking and adjusting as separate steps, rather than conflating them into a single exercise of “unpacking comparable reasonable prices.”

3.4.1 What it means to adjust unpacked rates in principle

Adjustments typically have one of two aims:

1. improving comparability, or

2. improving reasonableness.

3.4.1.1 Adjustments to improve comparability of unpacked rates

These adjustments account for material differences between the Focal Contract and an otherwise comparable license, as identified in the comparability assessment.

In principle, adjustments may be made for any difference that materially affects the value exchanged and, therefore, the price that independent parties would be likely to agree. Some examples are given below:

• Adjusting for differences in the scope or mix of technologies and products. As discussed above, where a comparable license covers additional standards, patent portfolios or product categories that are not present in the Focal Contract, the unpacked rate may not be directly comparable. If sufficiently reliable information is available, the analyst may “carve out” the portion of value attributable to the additional rights or products. This typically requires either (i) an independent valuation of the incremental component or (ii) another license with a narrower scope that isolates the component to be valued.

• Adjusting for the proportion of products sold in a particular region with lower rates. In some cases, a party may have sufficient sales in a particular region that it can agree to a rate below the typical global level. The rate in a particular region maybe low for various reasons. For instance, in Unwired Planet v. Huawei sales in China were considered to be 50 percent lower than for rest of the world, so an adjustment was made to reflect the different proportion of sales in China between the comparable and Focal Contracts. ⁽¹³⁾Unwired Planet v. Huawei [2017] EWHC 711 (Pat), paras 582–583. Similarly, in InterDigital v. Lenovo the royalty rate unpacked from the closest comparable license was reduced by 27 percent, to reflect the fact that the Focal Contract had a higher proportion of sales in emerging markets, where rates are lower than in developed markets to promote growth. ⁽¹⁴⁾InterDigital v. Lenovo [2023] EWHC 539 (Pat). Licensees whose products are predominantly sold or manufactured in regions where a licensor has few patents may also achieve lower rates than a global manufacturer would agree.

• When using the licensee’s comparables, adjusting for differences in portfolio quality. The size, strength, essentiality and technical importance of SEP portfolios vary across licensors. Differences in these factors affect the terms that parties would agree to, so they should be assessed and, where feasible, appropriate adjustments made. ⁽¹⁵⁾Ensuring that the adjustments are reliable can be challenging. These issues are discussed in Section 5.3.2 below on the top-down approach.

Box 3.10 illustrates an example of adjusting the rate agreed for a particular contract, so that it more closely reflects the rate that parties would agree for the Focal Contract, which has a different product mix and proportion of sales in emerging markets. The partes agreed $1.250 per unit for Contract A, given its specific circumstances. Adjusting to reflect the Focal Contract’s circumstances, the parties would have agreed to $1.495 per unit on a like-for-like basis.

3.4.1.2 Adjustments to improve the reasonableness of unpacked rates

This second type of adjustment estimates the terms on which willing parties would have been likely to agree had negotiations not been conducted under conditions that distort bargaining outcomes, whether through hold-up or hold-out.

Unlike adjustments made to improve comparability, this type of adjustment necessarily involves a normative judgment to distinguish between:

• factors that did influence the terms agreed in practice, and

• factors that, under FRAND principles, should not influence the terms that define a reasonable outcome.

In principle, an adjustment may be considered where a feature of the negotiation environment affected the terms agreed in a comparable contract in a way that is not regarded as consistent with what willing, good-faith parties would take into account. As discussed in Section 1.3.4 above, examples may include:

• litigation costs that may be significant relative to the expected adjustment in royalties;

• the prospect of injunctions;

• the treatment of past sales (including substantial discounts or waivers), particularly where this could be perceived as incentivizing delay;

• the procedural complexity and costs associated with patent-by-patent and jurisdiction-by-jurisdiction dispute resolution.

In practice, such adjustments are often difficult. A reliable adjustment requires, first, a principled basis for concluding that the relevant factor should be excluded from the benchmark of reasonable terms and, second, sufficiently robust information to estimate its effect on the agreed consideration. Both propositions may be contested.

Box 3.11 provides an example drawn from a case before the Court of Appeal (England and Wales). In that case, the court treated discounts for past sales as a product of hold-out that suppressed the total consideration below the level that a willing licensee would agree in a FRAND-compliant license, and it adjusted the determination accordingly. That approach, however, is not accepted in all jurisdictions, and even where the underlying rationale is accepted, estimating the magnitude of the required adjustment may involve material uncertainty.

3.4.2 Practical considerations when adjusting rates

Adjustments are not without risk. The central appeal of the comparable contracts approach is that it relies on the prices that the market agreed to, not estimates. Any adjustment is a departure from what the parties actually agreed, and therefore adjustments may introduce uncertainty to various degrees. So, the more materially the valuation relies on adjustments, the further the methodology risks moving away from that initial appeal.  ⁽¹⁶⁾For instance, in Samsung v. ZTE, [2026] EWHC 999 (Pat), the judgment notes that “The experts also agreed that when considering comparable licences, the fewer adjustments needed to account for differences between the comparable licence and the licence being negotiated the better (with a caveat added by Mr Lin, which I accept, that it is important to recognise that it is not just the number of adjustments that matters; the extent of each adjustment is equally significant).”

Nonetheless, adjustments can be reliable and improve accuracy where they are grounded in evidence and applied in a disciplined manner. The following principles help to mitigate the risk of error.

3.4.2.1 A clear economic and commercial rationale

An adjustment should be grounded in a sound understanding of how the market works and why a specific difference in a given characteristic would influence the price that willing, good-faith parties would agree to.

For instance, the issue of how royalties are determined in licensing negotiations over sales in emerging markets is influenced by three distinct economic factors:

• portfolio strengths may be lower in emerging countries, which, all other things being equal, drives royalty rates down for negotiations where the licensee predominantly sells only in emerging countries;

• incomes and the effective value of IP rights may be lower in emerging markets due to local economic and policy conditions; and

• the relevance of regional variation to the terms of the license: for instance, regional variation may have little if any impact on the terms that willing parties would agree for a global license with sufficient sales in major patent jurisdictions but it may matter a great deal to the terms that parties agree for a contract where the majority of units will be sold and manufactured in regions without patent coverage.

Where the rationale for adjustment is unclear or speculative, the adjustment is unlikely to improve accuracy and may even introduce additional error.

3.4.2.2 A clear relationship between characteristics and prices

Reliable adjustments require an understanding of how specific differences in the contractual characteristics would affect the price that parties would consider reasonable, including:

• the direction of the effect (whether the differences increase or decrease the price), and

• the magnitude of the effect (by how much).

The better the understanding of this relationship between characteristics and prices, the more reliable the adjustment is likely to be.

In many cases, simplifying assumptions may be necessary and reasonable. For instance, the precise relationship may not be known but may be reasonably approximated by a linear scale – at least over small differences. In addition, data used for adjustments may be uncertain to some extent.

However, simplification is not always appropriate. The estimated price may become increasingly unreliable as its dependence on uncertain adjustments grows. At some point, it may be better to leave the rate unadjusted and address any imprecision qualitatively, rather than present an adjusted but uncertain estimate.

Box 3.12 provides an example of an adjustment that was, in those particular circumstances, considered justified in principle but unreliable in practice.

3.4.2.3 Collective and individual reliability

The reliability of adjustment should be considered collectively as well as individually. The effects of different adjustments may overlap. If two adjustments address the same underlying driver, applying both may double count the impact and reduce accuracy.

Box 3.13 provides an example of two adjustments that, in that particular circumstance, were each considered reasonable in isolation but not in combination.

3.5 Assessing whether comparable contracts provide prices within the FRAND range

The comparable contracts approach is attractive because it relies on rates that market participants have actually agreed. However, if earlier licenses were materially influenced by the parties’ relative negotiating positions or strategic conduct (for example, due to hold-up or hold-out), their use as comparable contracts risks carrying forward non-FRAND rates into subsequent negotiations.

Accordingly, any application of this method should include an assessment of whether the rates inferred from comparable contracts are themselves reasonable. In practice, three complementary approaches can be applied in parallel:

• assessing the bargaining power dynamics for indications of hold-up and hold-out;

• assessing the breadth of the rates inferred from comparable contracts;

• assessing the rates against an external benchmark of value (cross-checks).

3.5.1 Assessing the bargaining power dynamics for indications of "non-FRAND factors" (hold-up and hold-out)

This approach directly assesses the negotiation process that led to the agreed rate in order to determine whether there is evidence of tactics associated with hold-up and hold-out. Greater confidence can be placed in rates unpacked from contracts where there is little or no evidence of such conduct.

In a recent decision, the High Court (England and Wales) framed this assessment as considering the impact of “non-FRAND factors”, rather than hold-up and hold-out. It explained that: The terms “hold-out” and “hold-up” in this field are often used to imply wrongdoing, though, and to try to avoid this I think the more neutral term “non-FRAND” is useful. Further, it explained “I refer to these matters as “non-FRAND factors” deliberately. No conscious bad faith or deliberate delaying tactics on the part of [the parties] was alleged. It was the duty of their negotiators to get the best price they could…”.  ⁽¹⁷⁾[2026] EWHC 999 (Pat), paragraphs 92 and 37 respectively. 

The typical indicators of potential distortions include relatively high litigation costs compared with the expected change in royalties, the possibility of injunctive relief, fragmented litigation strategies or approaches to past sales that may reward delay. ⁽¹⁸⁾See also Section 1.3.4 above. There is a degree of subjectivity in such assessment. Different courts may reach different conclusions when assessing the potential impact of these non-FRAND factors on negotiation for the same licence.  ⁽¹⁹⁾For instance, the Munich Regional Court, Chongqing Court, and High Court (England and Wales) reach different conclusions about the extent to which non-FRAND factors affect the potentially comparable contracts available in Samsung v ZTE. See Samsung v ZTE [2026] EWHC 999 (Pat), ZTE v Samsung, The People’s Republic of China Chongqing First Intermediate Court Civil Judgment, Yu 01 Min Chu No. 905; and Munich Regional Court, 30.04.2026, 7 O 64/25 – Bestimmung der Angemessenheit von FRAND-Lizenzraten.

Depending on the evidence, a contract may be excluded as unreliable, adjusted – where the impact can be quantified with sufficient reliability ⁽²⁰⁾Regarding adjustments, see also Section 3.4.1 above. – or used only as a directional reference point (for example, as an upper or lower bound), rather than as a point estimate.

3.5.2 Assessing the breadth of the rates inferred from comparable contracts

Comparable contracts typically imply a range of rates, even after unpacking and (where appropriate) adjustments. However, the range should be reasonably narrow. If the implied rates remain widely dispersed, that may indicate that key differences have not been identified or controlled for, or that one or more licenses are not informative comparables.

One diagnostic is to compare the inferred rates across the set and assess whether they converge when expressed on a like-for-like basis. Where licensors’ and licensees’ comparables can be converted into an implied aggregate royalty rate – that is, an estimate of what an implementer would pay if all SEP portfolios were licensed – this can provide an additional check on their reasonableness, provided that a consistent approach is used to assess portfolio contributions and subject to the sensitivity of any assumptions regarding portfolio strength.

As discussed above, the implied aggregate royalty rate is sensitive to portfolio-contribution assumptions – especially for small portfolios. Imposing an “average value” assumption (that all SEPs, on average, are equally valuable) may often create a wide range of implied aggregate royalty rates, but this range will not in itself be informative in determining which licenses are reliable. So, in that context, the wide range may reveal either that rates were not similar on a like-for-like basis, or that the values of the portfolios per patent are not similar (i.e., some are genuinely more valuable than others). As a result, caution should be exercised when attempting to apply the reverse top-down approach as a “selection tool” to screen out comparables without a more detailed assessment of the transactions and portfolios in question. Box 3.14 provides a stylized example.

3.5.3 Assessing the rates against an external benchmark of value (cross-checks)

Cross-checks against external benchmarks of value can help to evaluate whether the rates inferred from comparable contracts are plausible. Courts in several jurisdictions, including the United States, United Kingdom and China, have considered such cross-checks when determining FRAND terms, comparing the implications of comparable licenses with other indicators of value. ⁽²¹⁾Part 5 discusses the various benchmarks, which include the ASP of the implementing product, statements by licensors on aggregate rates, the profit margins on chipsets and the price premium attributable to the technology. The various decisions include: in the United Kingdom, Optis v. Apple [2025] EWCA Civ 552, InterDigital v. Lenovo [2023] EWHC 539 (Pat) and Unwired Planet v. Huawei [2017] EWHC 711 (Pat); in the United States, ) TCL Communication Technology Holdings Ltd. v. Telefonaktiebolaget LM Ericsson, No. 8:14-cv-00341, (C.D. Cal. Dec. 21, 2017), amended and superseded (C.D. Cal. Sept. 14, 2018), rev’d in part, vacated in part 943 F.3d 1360 (Fed. Cir. 2019), cert. denied 141 S.Ct. 239 (2020) and In re Innovatio IP Ventures, LLC, Patent Litigation,  ⁽²²⁾921 F. Supp. 2d 903 (N.D. Ill. 2013) ⁽²³⁾ ; and, in China, the Nanjing Intermediate People’s Court in Huawei v. Conversant (2018) and the Chongqing court in Nokia v. OPPO (2021).

External benchmarks can take a variety of forms. Examples include:

• the average royalty rates that licensors announced before the standard was widely implemented;

• the price or profit margin of the implementing product; and

• the “price premium” – which indicates the proportion of a product price that is directly attributable to the licensed technology.

The merits of these benchmarks are discussed in Section 5.2 below, which covers the top-down framework.

Courts have sometimes accepted such benchmarks and at other times have rejected them where the benchmark itself was considered unreliable or insufficiently linked to the licensed technology. Nonetheless, the underlying principle is widely recognized: where a benchmark is sufficiently reliable, it can help parties, courts and arbitrators to assess whether market-observed license rates are consistent with a reasonable (including FRAND) outcome.

Benchmarks can be grouped into two broad categories.

• Bottom-up approach. This school of thought seeks to estimate the value attributable to a particular SEP portfolio (or to the functionalities it enables) using economic reasoning and evidence about portfolio quality and technological contribution, which may go beyond both patent counts and binary assessments of validity and essentiality. Basing the calculation on a hypothetical negotiation seeks to directly address the shortcomings of the comparables approach that stem from an ex post analysis – the Report will therefore addresses this method next (Part 4).

• Top-down framework. Methods belonging to this category begin with an estimate of the aggregate royalty rate for the relevant standard (the “total royalty stack”) and then allocate a share of that total to the SEP portfolio being valued, based on an assessment of its relative contribution (using proxies such as portfolio share, adjusted where possible for quality and other relevant factors). This approach provides a cross check for both the previously discussed methodologies, by ensuring that the prices agreed in practice make sense in principle, avoiding the risk of royalty stacking. Parts 4 and 5 will discuss both value-based approaches by breaking them down into methodical steps and highlighting their respective strengths and limitations.

The reliance that courts place on a cross-check depends on their assessment of each approach based on the facts in the relevant case. In that respect, the 2026 judgments relating to Samsung and ZTE are instructive. All three courts – in the UK, China, and Germany – considered evidence from comparable contracts and a “top-down” valuation. They each assessed the strength of that evidence differently. In the UK, the comparable contracts were considered to more robust and informative than the top-down valuation, which on the facts of that case was considered too sensitive to its underlying assumptions.  ⁽²⁴⁾[2026] EWHC 999 (Pat), para 49.The Chongqing Court considered both valuation approaches informative and consistent with each other – although the specific applications of those approaches differed from those in the English judgment. In contrast, the Munich Regional Court considered that none of the contracts were sufficiently comparable, and relied on a top-down assessment.  ⁽²⁵⁾ZTE v. Samsung, The People’s Republic of China Chongqing First Intermediate Court Civil Judgment, Yu 01 Min Chu No. 905; and Munich Regional Court, 30.04.2026, 7 O 64/25 – Bestimmung der Angemessenheit von FRAND-Lizenzraten.

3.6 Summary

The comparable contracts approach offers a structured way to estimate FRAND terms by reference to licenses concluded in sufficiently similar circumstances. It proceeds on the premise that agreements for similar patent rights, products, parties and contractual settings may provide useful evidence of the terms that willing parties would regard as reasonable for the Focal Contract.

As this part has shown, the approach is feasible in principle but depends heavily on evidence, judgment and the disciplined treatment of differences across contracts. This includes:

• identifying transactions that are sufficiently comparable in their economically relevant characteristics;

• “unpacking” the agreed terms into a common cash-equivalent metric, notwithstanding differences in payment structure, timing, risk allocation and non-monetary consideration; and

• adjusting unpacked rates where relevant differences remain between the comparable contract and the Focal Contract, whether to improve comparability or, in some cases, to address features of the negotiation environment that may have affected the agreed terms.

In practice, the method may be complicated by unobserved characteristics, changing circumstances over time, bundled technologies and products, limited numbers of observable transactions, and geographic variation in patent coverage, enforcement conditions and market conditions. It also requires an assessment of whether the rates inferred from comparable contracts plausibly fall within a FRAND range, including, where appropriate, by reference to external benchmarks of value.

Overall, the comparable contracts approach may provide useful market evidence for FRAND valuation, but its reliability depends on transparent criteria for comparability, careful treatment of contractual structure and appropriate cross-checks to assess reasonableness in context.