In recent years, the Court of Appeals for the Federal Circuit (“the Federal Circuit”) has issued several opinions related to the Entire Market Value Rule. The Federal Circuit has stressed the need to apportion the royalty base such that it is more closely tied to the patented invention for circumstances involving multi-component products in which the patented feature does not drive demand for the entire product. In LaserDynamics, the Federal Circuit stressed that in such circumstances, sales of the smallest salable patent-practicing unit (“SSPU”) should be used as opposed to sales of the entire product.1 In VirnetX, the Federal Circuit indicated that further apportionment of the royalty base beyond the SSPU may be necessary.2 In its opinion in Commonwealth Scientific and Industrial Research Organisation (“CSIRO”) v. Cisco, the Federal Circuit offered guidance regarding circumstances when apportioning the royalty base may not be necessary.
This article provides background on CSIRO v. Cisco, explains the Federal Circuit’s reasoning, and offers key takeaways from the ruling.
Background
On January 23, 1996, U.S. Patent No. 5,487,069 (“the ‘069 patent”) issued and was assigned to CSIRO. The ‘069 patent, which generally pertains to wireless local area networks, discloses techniques directed to solving issues that arise when wireless signals reflect off objects and interfere with each other. The ‘069 patent was included in the first revision of Institute of Electrical and Electronics Engineers (“IEEE”) wireless standard 802.11a and is essential to certain later iterations of the Wi-Fi standard, including 802.11g, 802.11n and 802.11ac. In a letter of assurance submitted to the IEEE, CSIRO pledged to license the patent on reasonable and non-discriminatory (“RAND”) terms with regard to the 801.11a standard, but it refused to encumber the ‘069 patent with a similar RAND commitment for the g, n and ac revisions.
In 1999, during the early days of the 802.11 standard, CSIRO licensed the technology to Radiata Inc. (“Radiata”), a company focused on the sale of wireless chips, which was founded by three individuals including a named inventor of the ‘069 patent. The Technology License Agreement (“TLA”) between CSIRO and Radiata identified an initial royalty rate of 5% of net revenue, with this rate decreasing in 1% increments, to a minimum rate of 1%, as sales volume increased.
Cisco acquired Radiata in early 2001. As part of the acquisition, the TLA was amended, largely to allow Cisco to take Radiata’s place in the TLA. The TLA was amended again in September 2003 and Cisco paid royalties under the TLA until 2007, when it stopped using Radiata-based chips in its products. The total royalties paid by Cisco under the TLA amounted to more than $900,000.
Around 2003, CSIRO began trying to license the ‘069 patent to other participants in the Wi-Fi industry. In 2004, it began offering licenses based on a “Rate Card,” but it never executed any licenses.3 The Rate Card was structured with different rate ranges depending on the number of days between the offer and the acceptance of a license. Moreover, within each range, rates decreased as sales volume increased. The lowest rates — applicable if a license was accepted within 90 days of being offered — ranged from $1.40 per unit (for sales greater than 20 million) to $1.90 per unit (for sales less than 1 million).
In 2004, while still licensing the technology under the TLA, CSIRO offered to license Cisco based on the Rate Card rates. Cisco declined the offer. The following year, Cisco’s Vice President of Intellectual Property suggested that a rate of $0.90 per unit would be more appropriate, but still the parties never agreed to a new royalty rate structured as an amount per unit. The $0.90 per unit royalty rate was not much lower than the rate Cisco was paying at the time under the TLA, but over time, Cisco’s royalties declined significantly due to rapidly decreasing chip prices.
The District Court Case
In 2011, CSIRO filed suit against Cisco in the Eastern District of Texas for infringement of the ‘069 patent. Nearly two years later, Cisco stipulated that it would not contest infringement and validity issues. The parties subsequently agreed to a bench trial to determine damages.
CSIRO’s damages expert based his damages assessment on the premise that the benefits of 802.11 products that practice the ‘069 patent over those products that do not practice the patent are attributable primarily to the ‘069 patent. As such, he compared market prices around the time of the hypothetical negotiation of 802.11 products that practiced the ‘069 patent to those that did not practice the ‘069 patent.
He determined that Cisco generated a profit premium for those products that practiced the ‘069 patent, ranging from $6.12 to $89.93 per unit for Linksys-branded products and $14 to $224 per unit for Cisco-branded products. After considering the Georgia-Pacific factors, CSIRO’s expert opined to a volume-tiered royalty rate ranging from $1.35 to $2.25 per unit, which resulted in damages in excess of $30 million.
Cisco’s damages expert used the TLA as the basis for his damages assessment. He determined that under the TLA, the royalty rate ranged from $0.04 to $0.37 per unit for Linksys-branded products and from $0.03 to $0.33 per unit for Cisco-branded products. Cisco’s expert subsequently opined to damages of just over $1 million.
On July 23, 2014, the district court issued its findings of fact and conclusions of law, in which it rejected both experts’ damages models and opinions. The district court rejected the opinion of CSIRO’s expert for various reasons, including his “performing [an] ‘arbitrary’ final apportionment and having broad profit premium ranges.”4
With regard to Cisco’s damages expert, the district court found fault with his use of the TLA, indicating that the TLA was not comparable to the license that would have resulted from the hypothetical negotiation. The district court’s criticism focused on the fact that Cisco’s damages model based its royalties on chip prices, stating that “[t]he benefit of the patent lies in the idea, not in the small amount of silicon that happens to be where that idea is physically implemented” and that “[b]asing a royalty solely on chip price is like valuing a copyrighted book based only on the costs of the binding, paper, and ink needed to actually produce the physical product. While such a calculation captures the cost of the physical product, it provides no indication of its actual value.”5
The district court identified the following four findings as reasons to support its rejection of the TLA:
- The close relationship between CSIRO and Radiata suggested that the agreed-upon terms were not necessarily reflective of an arm’s-length agreement.
- The TLA included other obligations and benefits that would not have been part of the hypothetical license.
- The timing of the agreement in 1999 hampered the TLA’s usefulness in determining a rate for the hypothetical license in 2003.
- The “primary problem with Cisco’s damages model was the fact that it based royalties on chip prices.”6
Having rejected both experts’ damages models, the district court performed its own damages calculation based on both the Rate Card and the $0.90 per unit rate mentioned by the Cisco executive. The district court reasoned that these data points were instructive because they generally coincided with the timing of the hypothetical negotiation. Using a range of $0.90 to $1.90 per unit as a starting point, the district court then considered the effect of the Georgia-Pacific factors on the rate. The district court concluded that it was not necessary to modify the Georgia-Pacific factors to account for the RAND commitment, because only 0.03% of the accused products were licensed on RAND terms. In its evaluation of the Georgia-Pacific factors, the district court concluded that factors 3, 4 and 5 suggested a downward adjustment, and factors 8, 9 and 10 suggested an upward adjustment. As a result, the district court concluded that no adjustment to the baseline rates was necessary for Cisco-branded products, but a range of $0.65 to $1.38 per unit was appropriate for Linksys-branded products, due to their lower profit margins. Based on the rates it determined, the district court awarded damages of approximately $16.2 million.
The Federal Circuit’s Opinions
Cisco argued three issues in its appeal to the Federal Circuit. First, Cisco argued that the district court erred by not considering the smallest salable patent practicing unit (“SSPU”) as part of its damages analysis. Second, Cisco argued that the district court’s analysis of the Georgia-Pacific factors was flawed because it failed to adjust the factors to account for the ‘069 patent being a standard essential patent (“SEP”). Finally, Cisco argued that the district court erred in its dismissal of the TLA evidence.
The Federal Circuit disagreed with Cisco’s first contention that the district court erred by not considering the SSPU. In its reasoning, the Court acknowledged the principle of apportionment that underlies Cisco’s argument. And while the Court also acknowledged that “the smallest salable patent-practicing unit principle states that a damages model cannot reliably apportion from a royalty base without that base being the smallest salable patent practicing unit,” the Court found that this principle did not apply in the instant matter because “the district court did not apportion from a royalty base at all.”7 Rather, the Court found that because the royalty rates the district court used in its calculation stemmed from the rates that the parties actually negotiated for the right to practice the ‘069 patent, no adjustment or apportionment was necessary to account for other factors. Furthermore, it stated that a rule that requires all damages models to begin with the SSPU is “untenable” and in conflict with prior damages models approved by the Federal Circuit based on comparable licenses.
The Federal Circuit agreed with Cisco’s second argument. It found that the district court should have modified the Georgia-Pacific factors similarly to what was done in Ericsson v. D-Link.8 While CSIRO argued that such modifications were not necessary because no obligation to license the products on RAND terms was made (i.e., the accused products did not use the 802.11a standard), the Federal Circuit found that modifications to the Georgia-Pacific factors were necessary for SEPs and not just RAND-encumbered patents. The Federal Circuit reasoned that if the Georgia-Pacific factors and analyses were not adjusted, patentees stood to receive all the benefit created from standardization rather than those benefits appropriately flowing to consumers and businesses that practice the standard. Additionally, the Federal Circuit noted that the district court failed to consider the possibility that the $0.90 to $1.90 per unit rates that it used in its analysis could have been impacted by standardization, and that some portion of those rates could have included value resulting from the standard’s adoption. In support of this notion, the Federal Circuit pointed to the fact that not even one entity had accepted a license at the Rate Card rates.
Finally, the Federal Circuit disagreed with three of the four reasons the district court identified for rejecting the TLA. The Federal Circuit pointed to the TLA amendments, noting that the close relationship between CSIRO and Radiata no longer existed by the time of the amendments (refuting the district court’s first reason) and that the timing of the amendments generally coincided with the hypothetical negotiation (refuting the district court’s third reason). With regard to the district court’s fourth reason, the Federal Circuit indicated that the district court’s argument was in conflict with the Federal Circuit’s opinion in Ericsson v. D-Link, which stated that a license may not be excluded solely because of its chosen royalty base.
In light of its findings, the Federal Circuit vacated the damages award and remanded the matter to the district court to reevaluate damages.
Key Points to Take Away
The Federal Circuit’s opinion in CSIRO v. Cisco touched on a number of issues pertinent to damages. Damages experts and attorneys should take away the following key points from the case:
- In matters involving multi-component products in which the patented feature does not drive the demand for the entire product:
- The principle of apportionment must be considered.
- As indicated by previous Federal Circuit rulings (e.g., LaserDynamics and VirnetX), the SSPU (or some portion of it) should be considered in order to produce a royalty base more closely tied to the claimed invention.
- The royalty base may not require apportionment if the royalty rate is determined by prior licenses of the patent in suit or from comparable licenses.
- Modifications to the Georgia-Pacific factors are necessary for matters involving SEPs — not just matters involving RAND-encumbered patents — to avoid the patent holder receiving the benefit resulting from standardization.
- One should consider whether rates used in determining a reasonable royalty might have been impacted by standardization.
- A potentially comparable agreement should not be excluded solely because of the type of royalty base it contains.
- LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012).
- VirnetX, Inc. v. Cisco Sys., 767 F.3d 1308 (Fed. Cir. 2014).
- The rates from the Rate Card were structured as an amount per unit sold. This differed from the percentage of revenue rate structure in the TLA.
- CSIRO v. Cisco Sys., 809 F.3d 1295, 1299 (Fed. Cir. 2015).
- CSIRO v. Cisco Sys., 809 F.3d 1295, 1300 (Fed. Cir. 2015).
- CSIRO v. Cisco Sys., 809 F.3d 1295, 1299-1300 (Fed. Cir. 2015).
- CSIRO v. Cisco Sys., 809 F.3d 1295, 1302 (Fed. Cir. 2015).
- Ericsson, Inc. v. D-Link Sys., 773 F.3d 1201 (Fed. Cir. 2014).