FRAND Licensing

FRAND stands for Fair, Reasonable, and Non-Discriminatory — it is the licensing commitment that patent owners make to standards organizations as a condition of having their patents incorporated into technology standards: WHY FRAND COMMITMENTS EXIST: when a technology standard (e.g., 4G LTE, Wi-Fi, Bluetooth, USB-C) is developed, standards bodies like 3GPP, IEEE, ETSI, and ITU select technical specifications from among multiple competing approaches; the selected specifications may be covered by patents held by participating companies; without FRAND commitments, the patent owners would have complete hold-up power over any company that wanted to implement the standard — they could demand whatever royalty they chose because the implementer has no alternative (the standard mandates the technology); FRAND commitments prevent hold-up: the patent owner agrees in advance to license on FRAND terms to anyone who wants to implement the standard; in exchange, the standard adopts the technology (which may be more valuable than non-patented alternatives); WHO MAKES FRAND COMMITMENTS: companies that participate in standards bodies and hold patents essential to a standard; declarations are submitted to the standards body (e.g., ETSI's IPRD database); the commitment typically runs with the patent — binds the patent owner and successors; WHAT FRAND REQUIRES: fair (appropriate to the value of the contribution, not exploitative of the standardization lock-in); reasonable (in line with comparable licenses, industry norms); non-discriminatory (similarly situated licensees get similar terms); FRAND commitments do NOT specify a particular rate; they require the rate to be FRAND — leaving what is FRAND for negotiation or courts to determine; WHAT FRAND DOES NOT DO: does not require royalty-free licensing; does not eliminate the patent owner's right to sue for infringement; does not specify a specific royalty rate; HOLD-UP PROBLEM: even with FRAND commitments, implementers argue that SEP holders use litigation and injunction threats to demand above-FRAND rates; this 'SEP hold-up' is a major policy debate.

US and UK courts have developed specific methodologies for determining FRAND-compliant royalty rates: US APPROACH — TOP-DOWN METHODOLOGY (TCL v. Ericsson, C.D. Cal. 2017): identify the aggregate royalty that all SEP holders for the standard could collectively charge (the 'royalty stack' ceiling); determine the patent owner's proportionate share of all SEPs in the standard; apply that proportionate share to the aggregate royalty ceiling to get the per-patent royalty; EXAMPLE: if all 3G SEP holders could collectively charge 8% on handsets; Ericsson owns 8% of all 3G SEPs; Ericsson's FRAND rate = 8% × 8% = 0.64% of handset selling price; US APPROACH — COMPARABLE LICENSES: most probative method when available; look at licenses the SEP holder has actually given to similarly situated licensees; Ericsson v. D-Link (Fed. Cir. 2014): FRAND royalties must be apportioned to the SEP's contribution to the standard; cannot use entire market value of the device; UK APPROACH (Unwired Planet v. Huawei, UK Supreme Court 2020): UK courts can set a GLOBAL FRAND rate; a licensee who refuses the UK FRAND license faces an injunction; this is controversial because it allows UK courts to set worldwide licensing terms; EU APPROACH: CJEU Huawei v. ZTE (2015): SEP holders must follow a specific protocol before suing for infringement: (1) alert the implementer of infringement; (2) make a FRAND licensing offer with rate justification; (3) implementer must respond diligently; (4) if implementer makes a FRAND counteroffer, no injunction allowed; if implementer not diligent, injunction may be granted; ROYALTY STACKING: a single product (e.g., a smartphone) may implement dozens of standards (3G, 4G, 5G, Wi-Fi, Bluetooth, USB) each covered by multiple SEP holders; if each SEP holder claimed 1-2% royalty, aggregate royalties could exceed 20-30% of device price; courts consider royalty stacking when assessing FRAND-ness of a rate.

Standards-essential patents (SEPs) and non-essential (implementation) patents play very different roles in standardization: ESSENTIAL PATENT DEFINITION: a patent is standards-essential if implementing the standard technically necessarily requires practicing the patent; if there is no technically feasible way to implement the standard without infringing the patent, it is essential; SELF-DECLARED vs. TECHNICALLY ESSENTIAL: most standards bodies accept self-declarations (a company declares its patents are essential); over-declaration is common — many declared SEPs are actually NOT technically essential; studies show that only 20-50% of declared SEPs are actually technically essential when reviewed; IMPLICATIONS OF OVER-DECLARATION: companies may declare more SEPs than they have to appear more valuable in cross-licensing negotiations; in FRAND rate disputes, over-declaration matters because the SEP holder's proportionate share of truly essential patents determines FRAND rate; expert analysis is used to identify which declared patents are truly essential; ESSENTIALITY DETERMINATION: done through: (a) CLAIM CHARTS: map the patent claims to specific paragraphs of the standard specification; (b) TECHNICAL EXPERTS: a patent expert determines whether the claim could be avoided by implementing the standard differently; (c) STANDARDS BODY REVIEWS: some bodies (ETSI) have patchy review processes; others (MPEG LA) do more rigorous reviews; CONVERSE — NON-ESSENTIAL IMPLEMENTATION PATENTS: implementation patents cover specific technical choices for how to implement a standard, not the standard itself; not all phones implementing 5G infringe the same non-essential patents — different manufacturers may use different implementations; PATENT POOLS: organizations like MPEG LA (HEVC, H.264, MPEG-2), Avanci (cellular), Via LA collect SEPs from multiple holders and offer bundle licenses; implementers can license all pooled SEPs with one agreement; pools reduce transaction costs but create collective pricing power concerns.

FRAND litigation has produced landmark decisions across multiple jurisdictions: TCL COMMUNICATION v. ERICSSON (C.D. Cal. 2017, aff'd in part Fed. Cir. 2020): US court set a global FRAND rate for Ericsson's 2G/3G/4G SEPs; court used top-down methodology + comparable licenses; established precedent for US courts adjudicating FRAND rates; UNWIRED PLANET v. HUAWEI (UK Supreme Court 2020): UK courts CAN set a global FRAND license; Huawei must accept global license or face UK injunction; established UK as a major FRAND forum; HUAWEI v. ZTE (CJEU 2015): EU framework for SEP holder obligations before seeking injunction (notice → FRAND offer → diligent negotiation); QUALCOMM ANTITRUST CASES: FTC v. QUALCOMM (N.D. Cal. 2019): Qualcomm's 'no license, no chips' policy violated antitrust law; 9th Circuit (2020) REVERSED — Qualcomm's licensing practices are not antitrust violations; APPLE v. QUALCOMM (settled 2019): billion-dollar settlement; Apple agreed to multi-year license to Qualcomm's SEP portfolio; illustrates value of large-scale SEP licensing; IN RE ERICSSON (IPR, 2020+): multiple IPR petitions filed against Ericsson SEPs; PTAB decisions affecting SEP portfolio validity are increasingly important; SAMSUNG v. APPLE (ITC 2013): ITC exclusion order on Apple products based on Samsung SEPs; USTR vetoed the exclusion order (first veto in 26 years) — indicating US policy that SEP holders should not get exclusion orders; PATENT POOL CASES: MPEG LA v. MOTOROLA MOBILITY: court upheld MPEG LA pool structure; demonstrates that patent pools can satisfy FRAND obligations when pool rates are themselves FRAND; CURRENT FRONTIER: 5G SEP disputes; Ericsson, Nokia, Qualcomm, Huawei all hold large 5G SEP portfolios; new round of global litigation expected as 5G deployment continues.

Receiving a FRAND licensing demand requires careful navigation — the implementer has obligations too: INITIAL RESPONSE REQUIREMENTS: under Huawei v. ZTE (CJEU 2015), an implementer who refuses to negotiate or delays unreasonably loses injunction protection; the implementer must respond diligently to any FRAND offer and make a FRAND counteroffer if dissatisfied with the terms; STEP 1 — ESSENTIALITY ANALYSIS: retain a technical expert to assess whether the asserted patents are truly essential to the standard; many declared SEPs are not actually technically essential; non-essential patents need not be licensed on FRAND terms; STEP 2 — COMPARABLE LICENSE ANALYSIS: obtain or analyze comparable license agreements executed by the SEP holder with similarly situated licensees; under non-discrimination, similarly situated licensees should get similar rates; if the SEP holder has given better rates to a competitor, demand comparable terms; STEP 3 — PORTFOLIO VALUATION: how many truly essential patents does the licensor actually have? What is their proportionate share of all SEPs for the standard? Top-down methodology can cap the aggregate royalty; STEP 4 — MAKE A FRAND COUNTEROFFER: respond to the SEP holder's demand with your own FRAND offer based on your analysis; document the offer carefully (it is evidence in subsequent litigation); STEP 5 — NEGOTIATE: most FRAND disputes settle without litigation; having technical experts and comparable license evidence gives the implementer negotiating leverage; STEP 6 — LITIGATION STRATEGY IF NO DEAL: (a) seek a FRAND rate determination in court; (b) challenge essential patents in IPR (validity); (c) argue non-infringement (are the claims actually practiced?); ITC vs. DISTRICT COURT: if the SEP holder files an ITC complaint, the USTR veto precedent (Samsung v. Apple 2013) suggests exclusion orders for SEPs are disfavored but not prohibited; district court is typically more favorable to implementers.