Chinese ZTE v. Samsung FRAND judgment: third way, but much closer to Munich than London; English machine translation and analysis

English machine translation

Jared Cho’s analysis: Chongqing Court parts ways with EWHC and revisits its own prior position in OPPO v. Nokia

Executive summary

Key findings:

• The Chongqing Court applied a combination of top-down and comparable licensing approaches to reach a final FRAND rate of $731M, covering 2G–5G cross-licensing fees from January 1, 2024, to December 31, 2029, and 5G cross-licensing fees from January 1, 2019 to December 31, 2023.
• Under the comparable licensing approach, the Court used ZTE–Samsung (2021) as the comparable for 2G–4G SEPs, and Samsung–Nokia for 5G SEPs.
• ZTE’s proposed comparables — ZTE–Samsung (2021; 2G–4G), Samsung–Nokia, Samsung–Ericsson, and Samsung–InterDigital (NEI) — were mostly accepted. ZTE did not advance any VOX (Vivo, Oppo, Xiaomi) agreements, unlike in the parallel UK proceeding.
• Samsung’s proposed comparables — ZTE–Samsung (2021), ZTE–Apple (2020), and five others with stated reference value (Samsung–Docomo, Samsung–Huawei, Samsung–Datang, Samsung–NEC, and one anonymous counterparty) — were all rejected.
• Under the top-down approach, the Court concluded that ZTE’s 5G multimode royalty rate is approximately 0.271%–0.294% for 2024–2029, based on inputs drawn almost entirely from ZTE’s economic expert report: ZTE 5G portfolio strength of 7.7%; a 5G aggregate royalty burden of 7.8%–8.5% (2024–2029) and 4.341%–5.273% (2019–2023); generational value distribution of 5G/4G/3G/2G = 70:24:3:3 (2024–2029) and 50:40:5:5 (2019–2023); and a blended regional discount (figures redacted).
• On ZTE–Samsung (2021), the Court held — applying California contract law as the governing law — that the prior license agreement does not cover 5G SEPs, that the Covenant Not to Sue (CNS) carries no independent licensing consideration (treating it instead as a deferred obligation to be consolidated into any renewal), and that a past-sales-release discount applies given ZTE’s financial distress at the time of signing.
• On parallel litigation and jurisdictional scope, the Court stated that if a foreign court sets a rate for “other countries’ patents” within the licensed scope, the Chongqing Court will give respect accordingly — a diplomatically framed but firm assertion that foreign courts may only set rates for their own territory’s patents, while Chinese courts may set a global rate.

International ramifications: The Chongqing decision, unsurprisingly, denied nearly every material finding of the EWHC in the parallel UK case — throwing into sharp relief the unresolved tension over judicial sovereignty and the legitimate interests of Chinese entities in global SEP disputes. The collision is not merely doctrinal. Under the China State Council’s new regulation (April 13, 2026 ip fray article), any attempt by Samsung to seek enforcement of the EWHC decision in China risks triggering an anti-enforcement order, and potentially exposes the company to civil, administrative, or even criminal liability.

I. The negotiation history: eight meetings, rounds of offers, no deal

The full text reveals that Samsung and ZTE held at least eight business meetings (likely in-person) and exchanged at least four rounds of offers before reaching an impasse. Litigation followed across six jurisdictions — China, the UK, Germany, the UPC, and Brazil — a multi-front engagement that this author has previously described as the telecommunications industry’s own “Crusade”.

During the proceedings, Chongqing Court attempted mediation. ZTE expressed cooperation; Samsung’s response, described as more complicated, is redacted. Mediation failed.

II. Prisoner’s Dilemma: Samsung’s carefully worded response on compliance

Among the more striking procedural features of the Chongqing judgment is an explicit inquiry directed at both parties regarding their willingness to perform any FRAND rate the Court might set — including whether they would withdraw injunction claims elsewhere. The three questions put to the parties were:

1. whether they accept the licensing conditions as determined by the Chongqing Court;
2. whether they will voluntarily perform such conditions upon an effective judgment; and
3. whether they will withdraw patent infringement proceedings, in particular injunction claims, in other jurisdictions.

Samsung’s answer was nuanced. It asserted that it abides by laws and judgments in every jurisdiction and would respond to an effective Chinese court judgment in accordance with the law. It added, however, a significant carve-out: because Samsung had already committed to the UK court that it would be unconditionally bound by the EWHC decision, it would perform the Chongqing judgment only insofar as that judgment “does not contradict Samsung’s preexisting obligations before the UK court”. Samsung also conditioned withdrawal of injunctive proceedings on ZTE doing the same globally.

ZTE, by contrast, expressed unqualified willingness to perform the Chongqing judgment and to withdraw all other litigation if Samsung reciprocated.

The context matters. Under China’s State Council regulations on countering unjustified extraterritorial jurisdiction, any public pushback by Samsung against the Chongqing Court’s jurisdictional authority could expose the company to civil, administrative, or criminal liability in China — a market in which Samsung has significant presence. The conditioning language was, accordingly, the narrowest defensible response available.

III. The ZTE–Samsung (2021) agreement: Chongqing and London read the same contract differently

The divergence between the Chongqing Court and the EWHC begins at the foundational level: what did the ZTE–Samsung deal of 2021 actually cover?

Applying California Civil Code Articles 1638 and 1639 based on expert submissions, the Chongqing Court reached three conclusions that run mostly contrary to the EWHC’s findings:

• First, ZTE–Samsung (2021) does not cover 5G SEP licensing — based on the literal text of the agreement, its preamble, and the negotiation history.
• Second, the CNTS consideration is not separately reflected in the licensing fee. The CNTS, in the Court’s view, is merely an undertaking to refrain from filing infringement litigation during a defined period; the economic value of that undertaking is deferred and consolidated into any renewal rather than reflected in the 2021 fee.
• Third, a past-sales-release discount applies to the 2021 fee because, as a first-ever bilateral agreement concluded while ZTE was managing the financial aftermath of US export control penalties, the royalty rate is very likely to have been discounted — making it an unreliable starting point for market-rate benchmarking without adjustment.

IV. Selection of comparables: an asymmetric and unprecedented outcome

The Court’s approach to comparable selection is the most analytically striking — and arguably the most legally contentious — feature of the judgment.

ZTE proposed ZTE–Samsung (2021; 2G–4G only), Samsung–Nokia, Samsung–Ericsson, and Samsung–InterDigital (NEI).

Samsung proposed ZTE–Samsung (2021; 5G covered), ZTE–Apple (2020), and five agreements described as having reference value: Samsung–NEC, Samsung–Docomo, Samsung–Huawei, Samsung–Datang, and one anonymous agreement (“NDHD”). The Court rejected all of them.

The specific rejections are notable:

• ZTE–Apple (2020): This one was rejected on the basis that ZTE’s financial distress in that period “distorted” the licensing environment, rendering the agreement unable to reflect the portfolio’s market value. The implicit consequence of this reasoning — left unstated by the Court — is that all ZTE agreements concluded around 2019–2020 are similarly tainted.
• Samsung–Ericsson and Samsung–InterDigital: Considered less comparable than Samsung–Nokia. The Court’s reasoning: Samsung–Nokia was reached through negotiation untainted by litigation (unlike Samsung–Ericsson) or arbitration (unlike Samsung–InterDigital), and Nokia and Ericsson are more similarly positioned to ZTE in business scope and scale than InterDigital is.

The result is unprecedented in Chinese SEP jurisprudence: an agreement covering the same licensed portfolio (ZTE–Apple 2020, covering ZTE’s portfolio) is rejected as non-comparable, while an agreement covering a different portfolio (Samsung–Nokia, covering Nokia’s portfolio) is accepted. In prior decisions — Huawei v. Conversant, OPPO v. Nokia, —identity was treated as the primary factor conferring comparability. The Chongqing Court has now subordinated portfolio identity to the circumstances of negotiation.

There is a further wrinkle. At the close of the decision, the Court notes that Samsung submitted a new cross-license offer in October 2025 — the specific figure is redacted — that was substantially larger than the rate Samsung’s own unpacking of its proposed comparables would have implied. Samsung explained the gap as reflecting “genuine good faith in resolving the dispute”. The Court declined to accept that explanation, treating the inconsistency instead as evidence that Samsung’s proposed comparables were not reliable — a conclusion that penalizes Samsung for having made a settlement-oriented offer during litigation. A decision that Samsung shot itself in the foot with — or rather, heaved a boulder onto itself.

V. The top-down approach: updating OPPO v. Nokia with a two-stage model

Chongqing Court substantially followed ZTE’s economic expert and applied a top-down methodology consistent with — and explicitly designed to reconcile with — its 2023 decision in Oppo v. Nokia. To achieve internal consistency across the two judgments, the Court introduced a two-stage structure: a “5G early stage” covering 2019–2023, and a “5G mature stage” covering 2024–2029, with different aggregate royalty burdens and generational value weightings applied to each.

The calculation formula adopted by the Court is:

• 5G single-mode rate = 5G ARB × ZTE’s 5G portfolio strength × regional discount
• 5G multimode rate = [5G single-mode rate × 5G generational value distribution %] + [4G multimode rate × (2G–4G) generational value distribution %]

Where the 4G multimode rate is derived from ZTE–Samsung (2021).

Applying inputs drawn almost entirely from ZTE’s submissions, including

• a 7.7% portfolio strength figure supported by ten data sources;
• aggregate royalty burdens of 7.8%–8.5% (2024–2029) and 4.341%–5.273% (2019–2023) derived from a hedonic regression model; and
• generational distributions of 70:24:3:3 and 50:40:5:5 respectively;
• the regional discount figure remains redacted.

The Court concluded a ZTE 5G multimode royalty rate of approximately 0.271%–0.294% for the 2024–2029 period.

VI. Jurisdictional stance: a quiet but firm assertion of global rate-setting authority

The decision’s treatment of parallel foreign proceedings deserves particular attention. The Chongqing Court acknowledges the existence of simultaneous litigation in multiple jurisdictions and frames its jurisdictional stance in measured language: since “rate-setting disputes aim to promote the parties to reach a licensing agreement regarding the relevant SEPs”, foreign courts that set a rate for their countries’ patents within the licensed scope will be accorded respect by the Chongqing Court.

The formulation is polite. The implication is unambiguous: Chinese courts may set a global rate; foreign courts may set rates for their own territory’s patents. This is a direct, if tacit, repudiation of the EWHC’s global rate-setting jurisdiction — and, read alongside the China State Council regulations on countering foreign judgments, it carries enforcement teeth. Any effort by Samsung to enforce the EWHC judgment in China risks triggering anti-enforcement order proceedings and potential liabilities.

VII. What next?

Samsung’s Korean parent company has 30 days to appeal to Supreme People’s Court and it is almost certain to do so — not only because of the potential impact of an appeal on substance, but because filing preserves the status quo and delays the date on which the judgment becomes effective and enforceable.

For ZTE, the decision is an unqualified victory on every material point. It may consider petitioning for partial enforcement before the appeal concludes — a mechanism available under Chinese civil procedure — or simply wait for the final judgment to take full effect.