While I was away earlier this month, R.J. Reynolds Vapor Company filed a petition for certiorari presenting the following two questions:
1. Does the Federal Circuit’s judicially created “built-in apportionment” exception, which allows patent owners to use unrelated prior licenses to prove damages without providing “evidence tending to separate or apportion” the patent’s contribution, violate Garretson v. Clark’s requirement that apportionment “must in every case” be shown?
2. Alternatively, in light of the Federal Circuit’s intervening en banc decision in EcoFactor, Inc. v. Google LLC, 137 F.4th 1333, 1339-40, 1346 (2025)—which reiterated that expert damages opinion evidence “that is connected to existing data only by the ipse dixit of the expert” is inadmissible under Daubert and Federal Rule of Evidence 702 and rejected damages expert testimony regarding a royalty rate purportedly used in reaching lump-sum license agreements that did not support such a rate—should the Court grant the petition, vacate the judgment, and remand for application of that precedent, as is typical practice when an intervening development reasonably shows that the lower court’s decision rests on a premise that it would reject if given the opportunity for further consideration?
The cert. petition has already been discussed on Bloomberg, Law360, and (in greater detail) Patently-O, so I’ll just add a few thoughts of my own here. I previously noted the underlying non-precedential decision, in which the Federal Circuit affirmed the $95 million damages award, here.
First, as question # 1 above states, the petition squarely aims at the Federal Circuit’s use of the “built-in apportionment” concept. In the typical case applying this concept, the patent owner’s damages expert asserts that the express or implicit royalty rate set forth in a purportedly comparable portfolio license which included, among many other patents and/or other IP rights, the patent in suit (or, as here, a supposedly comparable patent or patents) implicitly or effectively apportioned the value of the patent in suit, because (for example) the patent in suit was the most valuable patent in the portfolio, or for some reason is said to be as or more valuable than the patents licensed in the portfolio, etc. Commentators--including William Lee and Mark Lemley, in an article that I noted here and which is cited throughout the petition—have argued that the courts have wrongly permitted use of the built-in apportionment concept in cases in which the basis for doing so has been slight to none, and that this has resulted in inflated jury awards. I think they are basically right about that, and that this case may be an appropriate vehicle for the Supreme Court to step in with a needed correction. The Federal Circuit itself did not address the issue in its recent EcoFactor decision (see here), though as suggested by question # 2 above that decision’s focus on the reliability of expert testimony should be relevant in this context as well.
Second, it’s interesting that the petition cites (as it should) not only old Supreme Court patent cases such as Garretson v. Clark, 111 U.S. 120 (1884), which addressed a type of apportionment, but also a copyright case and a trademark case that each stressed the need for proper apportionment. See Petition p.16, citing Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390 (1940), and Dewberry Group, Inc. v. Dewberry Engineers Inc., 145 S. Ct. 681 (2025). The implication, I think, is that the current patent damages jurisprudence on apportionment is something of an outlier, for no good reason. That may be so, although the petition notes that the latter case (Dewberry) stands for the proposition that an award of the defendant’s profits for trademark infringement must be “limited to those ‘properly attributable’ to the named defendant, not to other, unnamed affiliates” (Brief at 16). Subject to that caveat, however, U.S. trademark and copyright law actually follow different rules; in a trademark case, once the decision has been made to award profits, the plaintiff is entitled to all of the defendant’s net profits—though to avoid windfalls, disgorgement is permissible only if the unauthorized use of a trademark is a substantial factor in the defendant’s profits. See Pamela Samuelson, John M. Golden & Mark P. Gergen, Recalibrating the Disgorgement Remedy in Intellectual Property Cases, 100 B.U. L. Rev. 1999, 2009-11 (2020). (In my view, this rule was more defensible back when disgorgement was permitted only in cases of willful infringement. Unfortunately, however, the Supreme Court jettisoned that limitation on disgorgement in Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), previously noted on this blog here.) And, of course, U.S. patent law doesn’t allow disgorgement at all except in design patent cases, where (irrationally) the plaintiff is entitled to the entire profit attributable to the infringing article of manufacture. Indeed, one of many odd things about IP damages law in the U.S. is that, as the preceding examples show--and in contrast to the law as it exists in many other countries--the rules can vary quite a bit from one body of IP law to another. But there should at least be a plausible reason for a different rule, and I think the petitioners are right to imply that, at least with regard to the apportionment principle, patent law doesn’t need to deviate from copyright by applying the built-in apportionment concept.
Third, as Dennis Crouch points out in his post, Garretson v. Clark involved a different sort of apportionment issue, namely the need to ensure that an award reflects the value of the patented invention and not other features of the end product. As Professor Crouch also notes, however, the “basic idea” is the same, namely that the award should reflect only the patent in suit and not other patents or other IP rights or other features of the accused product. To be sure, that basic idea is not always easy to apply, and reasonable minds may differ about what it means in a given situation—for example, when a plaintiff can show that, but for the infringement of a single patent, a prospective purchaser of a complex end product incorporating many other features would have purchased that product from the plaintiff rather than from the defendant. See, e.g., Mentor Graphics Corp. v. EVE-USA, Inc., 851 F.3d 1275 (Fed. Cir. 2017), where the Federal Circuit held—properly in my opinion—that in such a case the plaintiff could recover its entire lost profit without further apportionment. Moreover, as I point out in my forthcoming book Remedies in Intellectual Property Law, apportionment can mean slightly different, albeit related, things in different contexts, even if the “basic idea” is in some sense the same: that the royalty should reflect the value the IP in suit contributes to profitability or cost savings, as distinct from the value of other IP not in suit, or other features of the end products into which the IP is incorporated (the issue in Garretson); that the value to the defendant of the specific IP right at issue should be isolated from the aggregate royalty rate implicit in a purportedly comparable portfolio license (this case); and that the royalty base should be one which does not present an undue risk that the trier of fact will inflate (or deflate?) the appropriate award (as in the many cases involving the applicability of the SSPPU rule). The overarching idea, in other words, is that we should aspire to compensate the patentee for the profit or royalty it lost as a consequence of the infringement, but not to put it in a better position than it otherwise would have occupied (unless, of course, the conditions for awarding enhanced damages are present). The difficulty lies in carrying out that principle, both conceptually and in a cost-effective manner. In this regard, advocates of built-in apportionment might argue that the concept economizes on information/administrative costs, and that whatever loss of accuracy arises as a result is worth it-- but in my view, that case has yet to be persuasively made.
No comments:
Post a Comment