Judge Davis would make a good damages expert (CSIRO v Cisco)

On July 23, 2014, Judge Davis of the Eastern District of Texas issued findings of fact and conclusions of law regarding damages figures in CSIRO v Cisco, Case No. 6:11-cv-00343-LED. Infringement and validity were stipulated, and the parties agreed to a bench trial on damages, which lasted four days. CSIRO's '069 patent related to IEEE 802.11 wireless standards (excluding 802.11b).

There was an allegation that CSIRO had made a RAND commitment to the IEEE regarding one of the 802.11 standards (802.11a), but Judge Davis put very little emphasis on this point, as CSIRO had volunteered to license the '069 patent anyway (even for 802.11 standards where there were no RAND commitments) and the number of accused products involving 802.11a was less than 0.1% of the total accused products.

The bulk of Judge Davis' 30+ page opinion is in three parts - articulating flaws in the plaintiff's damages model, articulating flaws in the defendant's damages model, and fashioning a more appropriate and defensible damages model to come to a final damages figure. At the time of the hypothetical negotiation, the parties were in actual negotiations, with both sides specifying potential per-unit royalties (some in legal offers, some in offhand remarks). But rather than use these actual numbers from the actual parties to bookend the range of royalties, both experts unnecessarily complicated what could have and should have been a (relatively) straightforward damages analysis.

CSIRO's damages analysis was based on profit differential between 802.11b products (which were not accused) and later 802.11 products (which were). Plaintiff's expert opined that this profit differential was primarily attributable to the '069 patent. The Court identified several problems with this analysis, however. First, the pricing data used to determine the profit differential was based on "a single product from each category of products compared, from a single retailer, at a single point in time" and thus constituted an impermissibly small sample size. Next, the ranges of profit premium ranged from $6.12 to $89.93 for Linksys products and $14.27 to $224.25 for Cisco products, and the "broad disparities indicate the inherent unreliability of this method. With such broad ranges, it is impossible to reliably determine where the value of the patented technology lies." Further, the Court identified several advantages of the later 802.11 standards not related to the '069 patent, advantages wholly ignored by plaintiff's expert. Finally, although CSIRO had a voluntary licensing program that topped out at $1.90 per product, the plaintiff's damages model had a top range of $2.25, which the Court found to be "illogical."

The Court was not any more persuaded by Cisco's analysis. Cisco's damages model was largely premised on an agreement with Radiata from several years before the hypothetical negotiation date, and - critically - several years before the 802.11 standard and accused products rose to prominence. Also, CSIRO's agreement with Radiata was not "purely disinterested business negotiations" since Radiata was a business partner with CSIRO and was but one small part of a much larger overall business agreement. The Court was also unpersuaded by Cisco's damages model being based on chip prices during the damages period, since the "chip prices in the damages period result[ed] from rampant infringement which occurred in the wireless industry." While not explicitly stated, the impression given by the opinion was that the chip prices would undoubtedly have been higher had any of the players in the industry actually paid CSIRO for the '069 patent, presumably by adding that license fee into the price of the chips. For example, if the chips cost $5 but that figure failed to include any license to CSIRO, those chips would have cost $7 had there been a license to the '069 patent, so use of the $5 figure was off by a significant percentage. The Court gave an analogy to this situation involving copyright:

Prior to 2008, outside of the Radiata TLA, no company in the industry sought a license from CSIRO to the '069 Patent and CSIRO received no royalties whatsoever for that technology. 02/03/2014 PM Trial Tr. at 38:21-24; 02/03/2014 AM Trial Tr. at 46:14-21. It is simply illogical to attempt to value the contributions of the '069 Patent based on wireless chip prices that were artificially deflated because of pervasive infringement. Basing 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.

After finding that neither party had provided helpful guidance from the parties, the Court evaluated the evidence itself to come to a proper damages figure. While ignoring the royalty rates of later licenses (which both parties, and the Court, agreed were entitled to little weight), the Court did use the later licenses as a basis for the structure of the hypothetical negotiation, since all of the licenses were per-unit royalties with volume discounts. The Court then used CSIRO's voluntary licensing program as the upper bound of the royalty range, and a suggestion from Cisco to CSIRO (not a formal offer) as the lower bound. Between these, the range was $0.90 to $1.90 for Cisco products, and 27.4% lower (or $0.65 to $1.38) for Linksys products, since the Linksys products had a 27.4% lower profit margin. The Court then went through the various Georgia-Pacific factors to determine how, if at all, the opening range should be adjusted. All told, the Court reached a royalty model that decreased from $1.90 (or $1.38) for minimal volume to $0.90 (or $0.65) for sales of more than 20 million units.