As I mentioned this morning, the Federal Circuit affirmed in part and reversed in part the $1.5 billion judgment against Marvell for the infringement of two patents relating to hard-disk drives (opinion here). The basic facts are that Marvell allegedly used the patented technology during a sales cycle that occurred in the United States, and that the sales cycle allegedly resulted in a "design win", i.e,, resulted in customers buying Marvell's semiconductor chips. Carnegie Mellon sought reasonable royalties for Marvell's sales of those chips, and enhanced damages for Marvell's alleged willful infringement. The district court entered judgment for Carnegie Mellon in the amount of $1.54 billion, which included $287 million in enhanced damages for willful infringement. I'll focus on the damages issues. (The Federal Circuit affirmed on validity and infringement, and also on the district court's rejection of Marvell's laches defense.)
First, on willful infringement, the Federal Circuit reversed, holding that Marvell's invalidity defense was objectively reasonable (p.28):
The court also relied on the proposition that it mattered whether Marvell developed its invalidity defense when undertaking its infringing activity. It said: “[I]n order for Marvell to have a ‘reasonable defense’ to infringement for the time period of 2001–2009, there needs to be some proof that the basis for such invalidity defense was known to the infringers or even the person having ordinary skill in the art.” Id. at 630. The court stated that “Marvell proffered no evidence that anyone at Marvell knew of the [allegedly anticipating] Worstell Patent from 2001 until this litigation began in 2009,” adding: “Even if the Court concluded that Marvell has now put forth a reasonable defense to infringement that has been developed during litigation, such a determination would not be dispositive.” Id. But our precedent is to the contrary. “The state of mind of the accused infringer is not relevant to th[e] objective inquiry” into the risk of liability to the defendant necessary for a finding of recklessness. Seagate, 497 F.3d at 1371. On that basis we have repeatedly assessed objective reasonableness of a defense without requiring that the infringer had the defense in mind before the litigation. See Halo, 769 F.3d at 1381–83; Bard Peripheral Vascular, Inc. v. W.L. Gore & Assocs., Inc., 682 F.3d 1003, 1008 (Fed. Cir. 2012); iLOR, LLC v. Google, Inc., 631 F.3d 1372, 1377 (Fed. Cir. 2011); DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc., 567 F.3d 1314, 1336 (Fed. Cir. 2009).
As I've suggested in the past, from an economic perspective I would award enhanced damages only when there is reason to believe that compensatory damages will not suffice to deter, and we can be reasonably confident we will not be overdeterring lawful conduct on the part of persons accused of infringement. Requiring a showing of objectively unreasonableness reduces the risk of overdeterrence and therefore probably makes sense, even when (as the district court may have thought was the case here) the objectively reasonable defense was not apparent to the defendant during the time it was infringing. People who think that enhanced damages (always or sometimes) should be conditioned on the defendant's subjective state of mind alone will disagree with me here.
The Federal Circuit also affirmed the 50 cent per chip royalty rate and rejected Marvell's argument that the district court should have given more weight to Marvell's proffered comparable licenses (pp. 33-35).
Now to the heart of the matter. According to Carnegie Mellon, Marvell stimulated demand for its semiconductor chips as a result of infringing conduct that occurred in the United States during Marvell's sales cycle. For any such chips that were made or sold in the U.S., or were imported into the U.S., Carnegie Mellon is entitled to a 50 cent royalty. But what about the chips that were made and sold abroad and not imported into the U.S.? Does Carnegie Mellon get a 50 cent royalty on these, because they were caused by Marvell's infringing conduct in the United States during the sales cycle? Or not, because this would undermine the rule that U.S. patent law is not applicable extraterritorially? In Power Integrations Inc. v. Fairchild Semiconductor Int’l, Inc., 711 F.3d 1348 (Fed. Cir. 2013), the Federal Circuit held that a U.S. patent owner is not entitled to lost profits for sales made abroad even if those sales were caused by infringing conduct that occurred in the United States. But maybe reasonable royalties are different, because (as the court says at p.32) "A key inquiry in the analysis is what it would have been worth to the defendant, as it saw things at the time, to obtain the authority to use the patented technology,considering the benefits it would expect to receive from using the technology and the alternatives it might have pursued. See AstraZeneca AB v. Apotex Corp., 782 F.3d 1324, 1334–35 (Fed. Cir. 2015)." If Carnegie Mellon and Marvell had sat down to negotiate a license before the alleged infringement began, wouldn't they have taken into account the value Marvell expected to derive from the use of the patent during the sales cycle, even if some of that value derived from foreign sales? Or would that undermine the rule against extraterritoriality?
I've been conflicted on this one all along, and even now I'm not entirely sure which side has the better of the policy argument. But the legal issue now appears to be settled, with the Federal Circuit concluding that the portion of the royalty based on the products made and sold abroad and not imported into the U.S. could not stand (pp. 37-38):
For the present context, we think that § 271(a) provides the basis for drawing the needed line. It states a clear definition of what conduct Congress intended to reach—making or using or selling in the United States or importing into the United States, even if one or more of those activities also occur abroad. . . . Where a physical product is being employed to measure damages for the infringing use of patented methods, we conclude, territoriality is satisfied when and only when any one of those domestic actions for that unit (e.g., sale) is proved to be present, even if others of the listed activities for that unit (e.g., making, using) take place abroad. Significantly, once one extends the extraterritoriality principle to confining how damages are calculated, it makes no sense to insist that the action respecting the product being used for measurement itself be an infringing action. Thus, here the claim is a method claim, but the damages measuring product practices the method in its normal intended use, cf. Quanta Computer, Inc. v. LG Electronics, Inc., 553 U.S. 617 (2008) (applying exhaustion to sale of unit that sufficiently embodies a method claim); and the hypothetical negotiation would have employed the number of units sold to measure the value of the method’s domestic use (before production and after), as discussed above. In these circumstances, the inquiry is whether any of the § 271(a)-listed activities with respect to that product occur domestically. . . .
There are significant conceptual differences between different measures of monetary compensation for infringement—including between what agreement the parties would have reached to value a defendant’s use of the patentee’s technology (reasonable royalty) and what amount of otherwise-made profits, based on sales at certain prices, the patentee lost as a result of the defendant’s use of the patentee’s technology (lost profits). See AstraZeneca, 782 F.3d at 1334–35; Warsaw Orthopedic, Inc. v. NuVasive, Inc., 778 F.3d 1365, 1377 (Fed. Cir. 2015). But in the respect that is crucial here, we think that there is a related constraint. In the lost-profits context, this court indicated in Power Integrations that, where the direct measure of damages was foreign activity (i.e., making, using, selling outside § 271(a)), it was not enough, given the required strength of the presumption against extraterritoriality, that the damages-measuring foreign activity have been factually caused, in the ordinary sense, by domestic activity constituting infringement under § 271(a). 711 F.3d at 1371–72. We think that the presumption against extraterritoriality, to be given its due, requires something similar in the present royalty setting. Although all of Marvell’s sales are strongly enough tied to its domestic infringement as a causation matter to have been part of the hypothetical-negotiation agreement, that conclusion is not enough to use the sales as a direct measure of the royalty except as to sales that are domestic (where there is no domestic making or using and no importing). As a practical matter, given the ease of finding cross-border causal connections, anything less would make too little of the presumption against extraterritoriality that must inform our application of the patent laws to damages.
The court then remanded for recalculation of the compensatory damages, to exclude any royalty for chips that were made and sold abroad and not imported into the United States.