First published in LES Insights
The Federal Circuit recently vacated a $16.2 million damages award where the district court failed to exclude the value of the patent attributable to a standard. The Federal Circuit agreed with the district’s court’s methodology to the extent it relied on the parties’ actual licensing discussions. But the Federal Circuit explained that reasonable royalties for all standard-essential patents—not just those subject to a RAND commitment—must not include any value flowing to the patent from the standard’s adoption.
Patent owners seeking to monetize patents covering technologies essential to practicing a technical standard created by a standards organization (e.g. the Institute of Electrical and Electronics Engineers (IEEE)), often enter into contractual obligations to license the standard-essential patents on reasonable and nondiscriminatory (RAND) terms. In some instances, however, patent owners refuse to agree to RAND commitments. When such patents are asserted in litigation, some courts adjust the traditional analysis to account for these RAND commitments. In Commonwealth Scientific & Industrial Research Organisation v. Cisco Systems1, the Federal Circuit held that even when a patent owner doesn't agree to license the patent on RAND terms, reasonable royalty damages for infringing a standard-essential patent must not include any increase in value resulting from the standard's adoption rather than in the patent itself. The Federal Circuit also held that when attempting to determine how to properly apportion the value of a patented invention to a particular product, it is not always necessary to identify the smallest salable patent-practicing unit in an accused product. In this case, the court agreed with the district court’s methodology of focusing, instead, on earlier discussions between the parties negotiating a license rate per product for the asserted patent.
Commonwealth Scientific and Industrial Research Organisation (CSIRO) sued Cisco Systems, Inc. for infringement of its patent directed to technology included in various 802.11 WiFi standards ratified by the IEEE, including the 802.11a standard and various later iterations of 802.11. CSIRO submitted a letter of assurance to the IEEE pledging to license the patent on RAND terms for the 802.11a standard, but despite IEEE's repeated requests, CSIRO refused to make the same RAND commitments for later revisions of the 802.11 standard.
Cisco did not contest infringement or validity, so the litigation was limited to determining the appropriate damages. CSIRO contended that the difference in profit Cisco captured between accused and un-accused 802.11 products largely represented the value attributable to the asserted patent, and therefore compared market prices of these two groups of products to determine profit premiums. These comparisons led to CSIRO’s proposed premium range of $80 - $200 per accused product. Cisco proposed a damages model that was based in part on a previous license agreement between itself and CSIRO.
The district court rejected both parties' damages theories. The district court found that Cisco proposed an improperly inflated royalty rate and that the apportionment in CSIRO’s proposed damage model was arbitrary and overly broad. The district court rejected Cisco's royalty rate because the agreement was not comparable to the license that Cisco and CSIRO would have negotiated in a hypothetical negotiation. The district court also faulted Cisco's proposed apportionment for determining a royalty base, because it based royalties on prices of the wireless chips that incorporate the technology covered by the asserted patent. According to the district court, the benefit of the patent encompassed an idea that extended beyond the physical component that Cisco proposed. Finding that rampant infringement had significantly and artificially driven down chip prices, the district court rejected Cisco's low royalty base.
Instead of adopting either party's proposed damages model, the district court created its own model, calculating a royalty rate based on a form license offer developed by CSIRO in 2004, and an informal rate suggested by a Cisco executive in 2005 during licensing discussions with CSIRO. The district court noted that these two events were near the hypothetical negotiation dates of 2002-2003, and therefore provided a reasonable starting point for determining a royalty rate based on a hypothetical negotiation between the parties at that time.
The district court then considered the impact of the Georgia-Pacific factors on its starting royalty rate. The court first explained that although some courts adjust certain factors to take RAND commitments into account, such adjustment was not necessary in this case because CSIRO was only obligated to license on RAND terms for the 802.11a products, which only made up 0.03% of the accused products. Given its analysis of the Georgia-Pacific factors, the court found that CSIRO and Cisco would have been in substantially equal bargaining positions at the hypothetical negotiation. Accordingly, the court did not adjust the royalty rates it derived from CSIRO’s form license offer and informal rate suggestion.
Using CSIRO's form license offer as the starting baseline rate, the court entered judgment for CSIRO of roughly $16 million. Cisco appealed.
The Federal Circuit’s Decision
On appeal, Cisco argued that the district court (a) erred by beginning its reasonable royalty analysis using rates derived from inter-party negotiations rather than the wireless chip (the smallest salable patent-practicing unit) and (b) failed to modify the Georgia-Pacific factors to account for the value added by widespread adoption of the 802.11 standard rather than the value of the patented technology alone.
A. Calculating a Reasonable Royalty The Federal Circuit found that the district court did not err when it employed a damages model that accounted for the parties’ informal negotiations regarding the end product, specifically, CSIRO’s informal rate suggestion. Cisco contended that the district court failed to apply the smallest salable patent practicing unit in its reasonable royalty calculation. The smallest salable patent-practicing unit provides that, where a damages model apportions from a royalty base, the model should use the smallest salable patent-practicing unit as the base so that the patentee will not be improperly compensated for non-infringing components of a product and the jury will not be misled by undue emphasis on the value of the entire product. The Federal Circuit explained, however, that the smallest salable patent-practicing unit principle did not apply here, because the district court did not apportion from a royalty base at all. Rather, the district court examined the parties’ negotiations, using Cisco’s informal suggestion of $0.90 per unit as a lower bound and $1.90 per unit requested by CSIRO in its public form license offer as an upper bound. Therefore, because the parties’ discussions centered on a license rate per product for the asserted patent, the Federal Circuit explained, this starting point for the district court’s analysis already built in apportionment. Essentially, the parties had negotiated the value of the patent and nothing more, thereby necessarily discounting any value not attributable to the patent.
B. Standardization The Federal Circuit, however, agreed with Cisco that the district court failed to failed to exclude the value accruing to the asserted patent because it was essential to the 802.11 standard. The court explained that reasonable royalties for standard-essential patents generally, not just those subject to a RAND commitment, must not include any value flowing to the patent from the standard’s adoption. In analyzing the Georgia-Pacific factors, the district court improperly increased the royalty award because the asserted patent was essential to the 802.11 standard. Specifically, the district court did not consider the standard’s role in causing commercial success of the products made under the patent. The district court also did not consider that the $0.90 and $1.90 per unit rates that it used as a starting point themselves could be impacted by standardization.
Strategy and Conclusion
This case demonstrates that in a reasonable-royalty damages analysis, courts need not apply the smallest salable patent-practicing unit principle where more effective methods of estimating a patent's value may exist, but courts must exclude the value of the patent attributable to its standard-essential status in its damages methodology, regardless of whether the patent is subject to RAND commitments.