What royalty stacking is and why it matters
Royalty stacking arises when a single product requires licenses from multiple independent patent owners, and the sum of the individual royalty demands exceeds what is economically reasonable. A smartphone might incorporate hundreds of patented features from dozens of different companies — cellular communication, Wi-Fi, GPS, touch interfaces, audio codecs, display technology, and more. If each patent holder demands a percentage of the phone's price, the cumulative royalty can quickly exceed the device's profit margin. The term gained currency from Mark Lemley and Carl Shapiro's 2007 article 'Patent Holdup and Royalty Stacking,' which showed that stacking could be economically harmful even when each individual license demand appears reasonable in isolation. The problem is particularly acute for standard-essential patents (SEPs), where adoption of a standard obligates implementers to license all SEPs for that standard — making it impossible to design around any individual SEP.
Royalty stacking in standard-essential patent licensing
Standard-setting organizations (SSOs) like IEEE (Wi-Fi, 802.11), ETSI (LTE, 5G), Bluetooth SIG, and others adopt technical standards that define interoperability. SSOs require members to disclose patents they believe are essential to implementing the standard and commit to license them on FRAND (fair, reasonable, and non-discriminatory) terms. FRAND commitments limit individual royalty demands but do not automatically prevent stacking — if 500 different companies each hold SEPs for a standard and each demands 0.5% of the device price, the aggregate demand is 250%. Courts determining FRAND royalties increasingly recognize stacking as a constraint: the aggregate of all FRAND royalties for a standard must be consistent with the standard's commercial viability. Ericsson v. D-Link (Fed. Cir. 2014) held that royalty stacking is admissible evidence in FRAND cases — a defendant can introduce evidence that the cumulative burden of all SEP royalties for the asserted standard would be unreasonable.
The top-down approach to capping cumulative royalties
The top-down approach to FRAND royalty determination addresses stacking by starting from what the aggregate market can bear. Rather than independently assessing each patent's value and summing the results, top-down analysis sets a reasonable aggregate royalty cap for all SEPs in a standard and allocates portions to individual patent holders based on their proportionate share of SEPs. Judge Robart used this approach in Microsoft v. Motorola Mobility (W.D. Wash. 2013), determining reasonable aggregate royalties for the H.264 (video codec) and 802.11 (Wi-Fi) standards, then calculating Motorola's proportionate share. The top-down approach has been adopted by UK courts (Unwired Planet v. Huawei) and European courts in FRAND rate-setting. Critics note that declared SEP counts are unreliable — many declared SEPs are not actually essential — and adjustments for patent quality are subjective. Nevertheless, top-down analysis provides a practical solution to the stacking problem that bottom-up (per-patent) analysis cannot.
Patent pools as a market solution
Patent pools address royalty stacking by allowing multiple patent holders to collectively license their patents at a single negotiated aggregate rate. A patent pool administrator negotiates a pool rate with major implementers, typically lower than the sum of individual demands, and distributes royalty income among pool members based on the relative strength and number of their patents. Well-known patent pools: HEVC Advance and MPEG LA for video codecs (H.264, H.265, AV1); Via Licensing and Avanci for cellular standards (LTE, 5G); Sisvel for audio and video codecs. Pool rates serve as evidence in FRAND litigation — they represent what sophisticated parties actually agreed to pay, accounting for stacking. The limitation: patent pool rates may understate individual patent value when strong patent holders opt out of the pool to demand higher individual rates. Companies with the most valuable SEPs often prefer individual licensing for maximum leverage, leaving weaker patents in the pool.
Apportionment as a stacking mitigation tool
Apportionment requirements — limiting royalty bases to the smallest salable patent-practicing unit (SSPPU) and requiring rates to reflect only the patented technology's contribution — serve as individual-patent stacking constraints. If each SEP holder is required to base its royalty on the chip that implements the relevant standard (not the entire device), and further apportion for its patent's contribution to that chip, the individual royalty demands are naturally limited. The Federal Circuit in Ericsson v. D-Link (Fed. Cir. 2014) applied this logic: FRAND royalties for Wi-Fi SEPs should be based on the Wi-Fi chip (SSPPU), not the entire device price, with the rate reflecting the patent's contribution to the chip and the standard's contribution to the chip's value. Individual apportionment requirements do not fully solve stacking — even chip-level royalties can stack — but they significantly reduce the problem by grounding each demand in economic reality.
Royalty stacking in non-SEP technology
Outside the SEP context, royalty stacking occurs in any technology field with dense, overlapping patent coverage. Pharmaceutical combination therapies may require licenses to multiple process patents for synthesizing each drug component. Software platforms may require licenses for dozens of separately-patented features from different holders. Automotive electronics face stacking from both SEP demands (for cellular, Wi-Fi, Bluetooth) and non-SEP patents on safety systems, infotainment, and autonomous driving features. For non-SEP patents, there are no FRAND commitments to constrain individual demands — the only market-based solutions are cross-licensing (exchanging patent rights between companies with large portfolios), patent pools (voluntary collective licensing), and litigation to challenge weak patents or reduce damages. The policy debate over whether royalty stacking is a genuine market problem or merely a negotiating argument remains active, with empirical economists on both sides.
Frequently Asked Questions
What is royalty stacking?
Royalty stacking is the cumulative effect of multiple patent owners each demanding separate royalties on the same product or technology. When a product requires licenses from dozens or hundreds of different patent owners — each asserting its patents are essential — the individual royalty rates 'stack' to produce a total royalty burden that can exceed the product's profit margin or the value of the patented technology to consumers. The problem is most severe in standard-essential patent (SEP) licensing for communications standards (Wi-Fi, LTE/5G, Bluetooth, USB), where implementing a single standard may require licenses from hundreds of patent holders, each of whom has committed to license on FRAND (fair, reasonable, and non-discriminatory) terms. If each SEP holder demands 1% of the product's price, and there are 200 SEP holders, the aggregate demand is 200% — clearly impossible. The term was popularized by Mark Lemley and Carl Shapiro in their influential 2007 article 'Patent Holdup and Royalty Stacking.' Courts and policy makers have increasingly recognized royalty stacking as a factor in setting FRAND rates.
How do courts address royalty stacking in FRAND patent cases?
Courts have developed several approaches to address royalty stacking in FRAND patent cases. Aggregate royalty analysis: some courts use a 'top-down' approach that starts from a reasonable aggregate royalty for all SEPs in the standard and then allocates shares to individual patents. Judge Robart used this approach in Microsoft v. Motorola (W.D. Wash. 2013), determining a reasonable aggregate royalty for H.264 and 802.11 standards and then assessing Motorola's proportional contribution. Ericsson v. D-Link (Fed. Cir. 2014): the Federal Circuit held that royalty stacking concerns are relevant and admissible to FRAND rate determinations, but a defendant must present actual evidence of stacking — speculative stacking arguments are insufficient. The court noted that the appropriate FRAND royalty must account for 'the cumulative royalty burden' the standard creates. Patent pool rates as proxies: FRAND royalty determinations often reference patent pool rates (established by industry participants to address stacking) as evidence of reasonable aggregate royalties. Smallest salable patent-practicing unit (SSPPU): using the component level (not the device level) as the royalty base limits per-patent royalty demands and reduces stacking.
What is patent holdup and how is it related to royalty stacking?
Patent holdup and royalty stacking are related but distinct problems in SEP licensing. Patent holdup occurs when a standard is widely adopted and the implementer has made irreversible investments (manufacturing lines, product design, supply chain), giving the SEP holder leverage to demand royalties far exceeding the patent's value at the time the standard was adopted — the implementer is 'held up' because switching to a different technology would be prohibitively costly. Royalty stacking is the aggregate version of the problem: even if each individual SEP holder's demand is modest, the sum of all demands can be unreasonable. The two problems compound each other: holdup by multiple parties creates a stacking problem. FRAND commitments are designed to address both: the commitment to license on fair and reasonable terms limits individual holdup; the non-discriminatory requirement means each SEP holder must offer similar terms to all implementers, which limits the ability to stack through discriminatory pricing. Courts applying FRAND analysis consider both the individual patent's contribution and the aggregate royalty burden.
How does royalty stacking affect non-SEP patent licensing?
Royalty stacking is not limited to SEP licensing — it can occur whenever multiple patent holders hold patents that cover overlapping technology in a complex product. In the smartphone industry, a single device may be covered by thousands of patents from dozens of different companies, each asserting royalty demands. In the automotive industry, connected vehicles incorporating multiple wireless standards (LTE, Wi-Fi, Bluetooth, GPS) face cumulative royalty demands from many SEP holders. In semiconductor design, a chip may need to license patents from multiple holders covering different circuit elements. For non-SEP patents, there is no FRAND commitment limiting individual royalties, making stacking potentially more severe. Patent pools are a market-based solution: patent holders pool their patents and offer a single license at a negotiated aggregate rate, eliminating individual stacking. However, pool membership is voluntary and some patent holders prefer individual licensing for higher rates. Cross-licensing is another solution: companies with large patent portfolios often cross-license, effectively eliminating royalty demands between them — disadvantaging smaller companies that lack portfolios for cross-licensing leverage.
What is the top-down approach to FRAND royalty calculation?
The top-down approach to FRAND royalty calculation starts from an estimate of the reasonable aggregate royalty for all SEPs in a standard and then determines the individual patent owner's proportional share based on the number and relative importance of its SEPs. Steps: (1) Determine a reasonable aggregate royalty for all SEPs in the relevant standard — often benchmarked against established pool rates, historical industry royalty levels, or economic analysis; (2) Count the number of SEPs in the standard and estimate each patent holder's proportionate share — typically by counting declared-essential patents and applying quality adjustments for actual essentiality; (3) Multiply the aggregate rate by the SEP holder's proportionate share to determine its FRAND rate for the asserted patents. The top-down approach directly addresses royalty stacking by starting from what the aggregate market can bear. It was used in Microsoft v. Motorola (Judge Robart, 2013) and endorsed by UK courts. Critics argue that declared SEP counts are unreliable (many declared SEPs are not actually essential) and that the approach does not account for differences in SEP quality. The Federal Circuit has acknowledged the approach without fully endorsing it as the only permissible method.