FRAND Royalty

A FRAND (fair, reasonable, and non-discriminatory) commitment is a contractual obligation that patent holders make when declaring patents essential to a technical standard: ORIGIN: when companies participate in standard-setting organizations (SSOs) like IEEE (Wi-Fi), 3GPP (LTE/5G), ETSI (GSM, 4G), JEDEC (memory), or ISO/IEC, they often contribute patented technology to the standard; SSO POLICIES: to prevent patent hold-up after the standard is adopted and implementers are locked in, SSOs require members to disclose potentially essential patents and commit to license on FRAND terms; THE DISCLOSURE OBLIGATION: if a member's patents may be essential to implementing the standard, the member must disclose them so the SSO can consider alternatives or require a FRAND commitment; FRAND COMMITMENT: the patent holder commits to license the declared standard-essential patents (SEPs) on terms that are: (a) FAIR: generally a market-rate royalty reflecting the patent's contribution to the standard; (b) REASONABLE: not exploitative of the lock-in that standard adoption creates; (c) NON-DISCRIMINATORY: similarly situated licensees get similar terms; WHO IS BOUND: the FRAND commitment runs with the patent — if the SEP is sold to a patent assertion entity, the FRAND obligation transfers; the new owner must honor the commitment; ROYALTY-FREE OPTION: some SSOs (particularly IEEE post-2015) also allow royalty-free (RF) licensing commitments; RF removes the licensing dispute entirely but reduces patent monetization incentives; WHAT FRAND DOES NOT REQUIRE: FRAND does not specify a particular rate; it defines a floor and ceiling by context; the parties must negotiate (or litigate) the specific rate; CONSEQUENCES OF REFUSAL: if an SEP holder refuses to license on FRAND terms after committing, they may: (a) lose the right to injunctive relief (eBay analysis + FRAND breach); (b) face antitrust liability; (c) be required by courts to license at a court-determined FRAND rate.

FRAND rate determination is one of the most contested areas in patent law — courts have developed several methodologies: METHODOLOGY 1 — COMPARABLE LICENSES: the most widely used approach; find real licenses for the same or comparable SEPs; adjust for differences in timing, scope, geography, and essentiality; ERICSSON v. D-LINK (Fed. Cir. 2014): 'comparable licenses' are the most probative evidence of FRAND rates; the court emphasized that licenses to the same patents under similar terms are the starting point; COMPARABILITY REQUIREMENTS: licenses must involve patents of similar technical value; cannot use licenses made under threat of litigation or before the FRAND commitment was made; METHODOLOGY 2 — TOP-DOWN ANALYSIS: start with the royalty-bearing value of the end product (or the smallest salable patent-practicing unit); determine the total reasonable royalty for all SEPs in the relevant standard; allocate a proportionate share to the specific patentee's SEPs based on their share of essential claims; TCL COMMUNICATION v. ERICSSON (C.D. Cal. 2018): Judge Selna applied a modified top-down approach; METHODOLOGY 3 — MODIFIED GEORGIA-PACIFIC: standard Georgia-Pacific factors adjusted for the FRAND context; critical adjustment: the hypothetical negotiation must occur BEFORE standard adoption (to avoid hold-up) even though the negotiation is being reconstructed after the fact; Ericsson v. D-Link: the hypothetical negotiation should be 'purified' of the hold-up value that results from standard adoption; SMALLEST SALABLE PATENT-PRACTICING UNIT (SSPPU): some courts require apportionment to the SSPPU (e.g., the modem chip that implements the standard) rather than the end device price; ROYALTY STACKING CONSIDERATION: the cumulative royalty from all SEP holders in the standard is relevant; a rate that would be reasonable in isolation may be unreasonable if the total stack is prohibitive.

The non-discrimination (ND) prong of FRAND requires that the SEP holder offer comparable terms to similarly situated licensees: ABSOLUTE vs. RELATIVE NON-DISCRIMINATION: ABSOLUTE: every licensee gets the same rate regardless of their competitive position or market power; RELATIVE: similarly situated licensees get comparable terms; most modern SSO policies use relative non-discrimination; WHAT 'SIMILARLY SITUATED' MEANS: factors considered include: size of licensee; volume of licensed products; geographic scope of license; whether cross-license offsets are involved; whether the licensee is a direct competitor of the SEP holder; TCL v. ERICSSON (Fed. Cir. 2020): the court examined whether Ericsson's licensing offers to different smartphone manufacturers were comparably non-discriminatory; DISCRIMINATION IN PATENT POOL CONTEXT: pool royalty rates are offered on identical terms to all licensees — this is a natural implementation of ND; individual licensing must be checked against pool rates and other known licenses; MOST-FAVORED LICENSEE CLAUSES: some SEP licenses include MFL provisions — if the SEP holder grants a lower rate to another licensee, the original licensee gets the lower rate retroactively; DISCRIMINATORY LICENSING PATTERNS: offering lower rates to joint-venture partners or affiliated companies than to independent competitors; refusing to license at any rate vs. offering licenses to competitors at disadvantageous terms; INTERNATIONAL VARIATION: ETSI's FRAND policy (Article 6.1) requires members to grant irrevocable licenses on fair, reasonable, and non-discriminatory terms; IEEE's 2015 patent policy reform changed the royalty base definition (smallest salable patent-practicing unit) and limited injunctions for SEPs; different SSOs have different ND interpretations; INJUNCTIONS AND NON-DISCRIMINATION: courts are reluctant to grant injunctions for FRAND-committed SEPs because injunctions give the SEP holder hold-up leverage inconsistent with the FRAND commitment.

Several major FRAND litigations have shaped how courts handle SEP disputes: ERICSSON v. D-LINK SYSTEMS (Fed. Cir. 2014): overturned district court FRAND jury instructions that used unadjusted Georgia-Pacific factors; held that FRAND-committed SEPs require special jury instructions recognizing: the ex ante nature of the hypothetical negotiation (before standard adoption); the need to avoid capturing hold-up value; the relevance of royalty stacking; established that comparable licenses are the most probative evidence; MICROSOFT v. MOTOROLA (W.D. Wash. 2013): Judge Robart determined FRAND rates for Motorola's 802.11 and H.264 SEPs; Motorola demanded 2.25% of end product price; court set rate at $0.555–$1.84 per unit — far below Motorola's demand; method: comparable licenses + apportionment; HUAWEI v. ZTE (CJEU 2015): EU Court of Justice addressed FRAND in antitrust context; if SEP holder refuses to negotiate in good faith, SEP holder may have abused dominant position (TFEU Article 102); CJEU required SEP holders to: make written FRAND offer before seeking injunction; set a specific royalty rate; licensee must respond diligently; TCL COMMUNICATION v. ERICSSON (C.D. Cal. 2018 / Fed. Cir. 2020): Judge Selna set FRAND royalty for Ericsson's 4G LTE SEPs; used top-down approach; Fed. Cir. reversed on procedural grounds but confirmed the top-down methodology is valid; APPLE v. QUALCOMM (S.D. Cal. settled 2019): $4.5B settlement after Apple challenged Qualcomm's no-license-no-chips policy as FRAND violation; terms undisclosed but Qualcomm resumed chip supply; CONTINENTAL AUTOMOTIVE v. AVANCI (N.D. Tex.): automotive industry challenging whether FRAND applies to component manufacturers (Tier 1 suppliers) or only to end-product manufacturers (OEMs).

FRAND policy attempts to balance two opposing risks in SEP licensing — hold-up by SEP holders and hold-out by implementers: HOLD-UP: occurs when the SEP holder demands royalties far exceeding the ex ante value of the patented technology (the value the technology would have had before becoming standard-essential); MECHANISM: once a standard is adopted and implementers have built products around it (investing in design-in, manufacturing, distribution), switching to an alternative is extremely costly; the SEP holder can exploit this lock-in to demand higher royalties than the technology would have commanded in a competitive market; FRAND's RESPONSE TO HOLD-UP: FRAND commits the SEP holder to the ex ante value; the hypothetical negotiation in FRAND rate-setting is conducted as if it occurred before the standard was adopted; courts adjust royalty bases (SSPPU vs. end product) to avoid capturing hold-up value; HOLD-OUT: the mirror problem — implementers use the threat of protracted litigation to delay licensing and obtain below-FRAND rates; MECHANISM: SEP holders cannot easily obtain injunctions for FRAND-committed patents (eBay analysis + FRAND waiver of injunction right); this reduces the SEP holder's leverage; implementers 'hold out' by litigating in hope the SEP holder settles at below-FRAND rates to avoid the litigation burden; MECHANISM FOR HOLD-OUT: implementer refuses to negotiate in good faith; uses litigation to delay; counts on SEP holder's desire for a settlement over extended PTAB and district court proceedings; BALANCING: Huawei v. ZTE (CJEU 2015) addresses hold-out by requiring implementers to respond promptly and in good faith to FRAND offers or lose the FRAND defense; US courts have not fully adopted a comparable framework; POLICY TENSION: courts disagree on which risk (hold-up or hold-out) is greater and how to address it — this drives much of the ongoing FRAND litigation.