The Federal Circuit has hit the ground running in the early weeks of 2022, overturning two sizable damages verdicts on February 4: the California Institute of Technology’s (Caltech’s) $1.1B award against Apple and Broadcom and Quarterhill Inc. subsidiary Wi-LAN Inc.’s (WiLAN’s) $85.2M award against Apple. Now, recent activity suggests that at least the latter ruling could potentially have a ripple effect on other large damages awards currently on appeal. The ensuing weeks have seen the defendant-appellants in those cases cite the WiLAN opinion, which faulted a district court for allowing a flawed methodology that was “untethered to the facts of this case”, as relevant authority justifying similar reversals.

The WiLAN Litigation: Two Large Awards Overturned Due to Flawed Damages Methodology

February’s WiLAN ruling was not the first time that a damages verdict was overturned in that case, which first went to trial in July 2018—culminating in a verdict of infringement and a $145.1M lump-sum damages award. Northern District of California Judge Dana M. Sabraw granted Apple’s post-trial motion challenging the damages verdict in early January 2019, rejecting WiLAN’s reliance on a “direct valuation” approach for its apportionment theory and faulting its experts for their use of the Voice over LTE (VoLTE) standard, not the patented technology, as that theory’s starting point. As a result, Judge Sabraw granted Apple’s motion for a conditional remittitur of $10M, which WiLAN rejected in favor of a new trial on damages.

WiLAN then suffered a second setback in the runup to the damages retrial. In October 2019, the court trimmed the plaintiff’s damages case in a ruling that largely granted a Daubert motion filed by Apple, excluding several theories offered by WiLAN’s damages expert. These included a valuation theory based on the alleged benefits of the asserted claims—finding that WiLAN had erred in relying on the testing of products that practiced another standard not at issue (LTE), and by again relying on an unreliable direct valuation approach. The court also took issue with WiLAN’s “alternative cost” theory, which was based on Apple’s alleged cost to improve cellular infrastructure as a design-around; as well as another theory based on licenses with third parties that the court deemed insufficiently comparable. The case then proceeded to trial in January 2020, with WiLAN asserting a damages theory that relied on a WiLAN expert’s analysis of the licenses not excluded in the aforementioned order. Later that month, the jury returned a damages award of $85.2M, based on a per-unit royalty amount of $0.45 multiplied by 189.4M infringing units. The final judgment, including interest, was $109M.

Ruling on appeal in February, the Federal Circuit agreed with Apple that one of the same WiLAN damages experts it used in the last trial, David Kennedy, had used a flawed methodology. During the retrial, Kennedy calculated reasonable royalty damages using a hypothetical negotiation approach based on three allegedly comparable agreements. Kennedy attempted to adjust for differences between those licenses and the hypothetical negotiation in this case, which would have only involved the two tried patents, by arguing that they were also “key patents” in the other agreements, and by citing the fact that Apple did not switch to a noninfringing alternative after the start of litigation.

This approach, the Federal Circuit found, was “untethered to the facts of this case”. For one, the court distinguished the three licenses as having been obtained prior to litigation, so “unlike Apple, the licensees did not continue to use the technology after being found to infringe”. Even more important was that two of the three licenses “treated the asserted patents as chaff, not wheat”, with no evidence suggesting otherwise. While a third license covered one of the patents-in-suit, the Federal Circuit found that the expert failed to account for how other patents in the license impacted the royalty rate—rendering his opinion unreliable. As a result, the court vacated the damages award and remanded for another trial.

For more on that decision, see “WiLAN’s $85.2M Damages Verdict From Second Apple Trial Cannot Stand, Rules Federal Circuit” (February 2022).

Cisco Cites WiLAN Decision in Centripetal Networks Appeal

One of the damage awards potentially impacted by the WiLAN ruling came in litigation filed by Centripetal Networks against Cisco. In October 2020, following a 22-day bench trial conducted entirely “by Zoom”, Eastern District of Virginia Judge Henry C. Morgan Jr. awarded nearly $2B in enhanced past damages, together with a royalty to run on certain products in stages for the following several years. The ruling was based in part on Judge Morgan’s consideration of an allegedly comparable license between Centripetal and Keysight Technologies. (Background regarding that ruling can be found at “Virginia Judge Issues $1.9B Judgment in Network Security Bench Trial” (October 2020).) Cisco has appealed that judgment (2021-1888), arguing to the Federal Circuit, among other things, that the court nowhere found that Cisco “combin[ed] multiple separately-sold Cisco products in [the] specific ways” purportedly required to infringe the asserted claims (and support Centripetal’s damages case) and that Judge Morgan, “after revealing a late-discovered disqualifying financial interest . . . nonetheless held onto the case for months, issuing two lengthy posttrial opinions and entering a blockbuster damages judgment—despite a bright-line statutory obligation to recuse himself”.

On February 11, 2022, Cisco filed a notice of supplemental authority citing the WiLAN ruling in its appeal. The filing recounted certain key reasons behind the Federal Circuit’s rejection of that plaintiff’s damages case as “untethered” to the facts, including its expert’s failure to show for two licenses “that the asserted patents were important to the licenses’ overall value (e.g., the patents were specifically discussed in the negotiations)” and, for the third license, his failure “to adequately explain why a 25% royalty rate reduction was sufficient to account for the patents not at issue in the litigation”.

Those same “critiques apply even more forcefully here”, Cisco now argues. In particular, the company underscores an allegedly similar failure by Centripetal’s expert to provide “evidence accounting for the different economic circumstances between the hypothetical negotiation and the Keysight license, including that the Keysight license involved far more patents than the four held infringed”. Also similarly problematic, asserts Cisco, is the court’s failure to adjust the royalty rate downward to account for the greater number of patents in the comparable license at issue: “While the district court considered the number of patents in the Keysight license as a ‘factor … in favor of a royalty rate reduction,’ it adopted the license’s highest royalty rate (10%) without any downward adjustment or explanation for why that rate was ‘a meaningful proxy for the royalty rate of’ the asserted patents”.

For this reason, Cisco concludes, the Federal Circuit should conclude that Centripetal’s damages case was inadmissible under Daubert and overturn the resulting judgment.

Apple Touts WiLAN Ruling in Challenging PanOptis and VirnetX Judgments

Apple, for its part, has argued that the Federal Circuit’s WiLAN ruling also undercuts judgments against it in litigation from two NPE plaintiffs.

            - PanOptis v. Apple

One of those cases, a standard essential patent (SEP) lawsuit filed against the company by several subsidiaries of PanOptis Holdings, LLC (collectively, “Optis” or “PanOptis”), also saw two damages verdicts. The first, a $506.2M damages award issued by a Texas jury in August 2020, was partially overturned in April 2021 by District Judge Rodney Gilstrap. The reason was that the plaintiff had requested a separate bench trial for issues related to its fair, reasonable, and non-discriminatory (FRAND) licensing obligations, a plan that the defendant had not challenged. As a result, Judge Gilstrap determined that the trial’s omission of FRAND issues led the jury to arrive at an unacceptably ambiguous verdict and ordered a second trial on damages. That second trial, which explicitly asked the jury to determine a FRAND royalty, ended in August 2021 with a verdict awarding $300M in damages.

Apple’s February 9 notice of supplemental authority citing WiLAN comes as the court considers its motion for judgment as a matter of law on damages, which challenges a damages opinion offered by David Kennedy—the very same expert whose testimony proved fatal in WiLAN. That motion highlights a variety of factual and legal issues undercutting Kennedy’s damages theory, some possibly overlapping those rejected in WiLAN—including its reliance on a license with Qualcomm that, per Apple, Kennedy failed to show was sufficiently comparable to form the basis of a $300M royalty. “The Federal Circuit’s reasoning squarely supports Apple’s challenge to Mr. Kennedy’s opinions in this case, including his improper reliance on the Qualcomm agreements”, Apple now asserts.

            - VirnetX v. Apple

Apple has also cited WiLAN in an appeal stemming from one of two long-running cases filed against it by VirnetX Inc., both of which have seen large jury verdicts make their way to the Federal Circuit and back.

The first of the two cases (6:10-cv-00417) was filed in April 2010 and went to trial in November 2012, leading to a $368M verdict that was partially reversed on appeal as to both infringement and damages. That case was subsequently consolidated with a second VirnetX case (6:12-cv-00855), with that combined case then seeing a trial of its own that ended in a $625M verdict in February 2016. However, that same July, District Judge Robert W. Schroeder III granted Apple’s motion for a new trial and unconsolidated the cases, accepting the company’s argument that the similarity between the two consolidated cases had resulted in undue prejudice and confusion.

A second, September 2016 trial in the -417 case then brought about another verdict that included a $302M award, later enhanced to $439M. That judgment was affirmed on appeal by the Federal Circuit, with the Supreme Court denying review in February 2020. The -855 case was the next one to see another trial, in April 2018, culminating in two verdicts entered later that month: one finding infringement of all asserted claims and awarding $502.6M in damages, and another finding willfulness for those claims. However, the Federal Circuit partly overturned the infringement verdict and remanded for further proceedings on damages, since that part of the verdict did not allocate damages by asserted patent. An additional jury issued a new, $502.8M damages verdict in November 2020, in a trial that was among the first to see an in-person jury trial since the start of the COVID-19 pandemic.

It is the appeal of that last verdict in which Apple has invoked WiLAN, in support of its argument that “VirnetX’s damages methodology was inherently unreliable”. Apple argues in its February 10 notice of supplemental authority that VirnetX’s expert made the same mistake as WiLAN’s of failing to adjust the proposed royalty rate to account for differences in the number of patents from a comparable license:

Even VirnetX admits that its expert did not adjust the royalty rate to account for the fact that the purportedly comparable licenses included far more patents than the two at issue in the underlying litigation. . . . Rather, its expert opined that the royalty rate should remain the same regardless of (1) whether one or two infringing features was at issue and (2) the number of unasserted patents included in the comparable licenses.

Through this alleged error, and because the VirnetX expert also did not substantiate his stated understanding that the “patents-in-suit drove [prior] deal[s]” as did David Kennedy in WiLAN, Apple asserts that this case “presents the same ‘methodological and factual error[]’ that the [WiLAN] Court rejected”.

Further details about prior VirnetX litigation, including the most recent retrial, can be found here.