Bayer v. Baxalta, Part II (Federal Circuit opinion, March 1, 2021)

A lot has happened in this matter since we last discussed it on A2C. To recap, Bayer’s expert, Dr. Addanki, had been excluded from offering his specific damages opinion, derived by employing the Nash Bargaining Theorem. In the absence of his damages value, we questioned how and with what the case might continue to trial. It turns out that the surprise is on us, because a lot remained: in fact, so much so that in November 2018, a Delaware jury awarded Bayer damages of $155,190,264 based on a reasonable royalty rate of 17.78%.

How did Bayer secure this damages sum in the absence of Dr. Addanki’s opinion?

In fact, while Dr. Addanki was not allowed to testify to a discrete damages value, he was permitted to testify to a damages-rate range, which he opined fell between 5.1% and 42.4%. As a result, the jury returned a damages figure within Dr. Addanki’s range. In turn, the Federal Circuit affirmed the damages award.

This conclusion to the Bayer v. Baxalta matter teaches us two things. First, damages experts need not opine to a specific number. Ranges on damages are just fine. In fact, very wide ranges on damages are just fine. Secondly, Daubert motions involving opinions with ranges must strategically address the essential roots of analysis, and not just a specific tainted “fruit” offered by the expert.

Bio-Rad Laboratories, University of Chicago v. 10X Genomics Inc. (Federal Circuit, August 3, 2020)

This case finally made it to the Federal Circuit. A2C awaited its outcome and the prospect of available lessons. Way back in September 2018, Judge Andrews in Delaware issued an opinion on Daubert motions. The court found defendant expert Ryan Sullivan’s analysis of comparable licenses sufficiently reliable to pass Daubert.

The court also found that James Malackowski, Bio-Rad’s expert, offered sufficient support for his comparable license analysis opinion, but not enough economic analysis to support his lost profits opinions which asserted a two-supplier market:

Judge Andrews also rejected Mr. Malackowski’s apportionment of the royalty base, which presupposed apportionment through the comparable license relied upon:

Judge Andrews allowed Bio-Rad and Mr. Malackowski to “supplement” damages opinions after having excluded Mr. Malackowski’s lost profits and reasonable royalty opinions. In this second bite of the apple, Mr. Malackowski did not offer a lost profits opinion, but rather only a reasonable royalty opinion. Mr. Malackowski relied upon the same licenses as those relied upon by the opposing expert. And in relying upon those same licenses, Mr. Malackowski’s opinions, like Dr. Sullivan’s initial opinions, were not excluded. Judge Andrews explained that Mr. Malackowski provided sufficient evidence of apportionment with what A2C views as creative analogous analysis of unpatented and unlicensed features:

Trial ensued, Plaintiffs prevailed, and they were awarded approximately $24 million. Defendant appealed the award based upon infringement, validity, willfulness and damages. The Federal Circuit affirmed the jury verdict in full and rejected the claims by defendant 10X that Mr. Malackowski failed to apportion and failed to use comparable licenses. In the first instance, while the Federal Circuit found two of three asserted patents were not infringed, because jury instructions were mute on the question of division of damages among patents, the award necessarily stood:

With regard to comparability, the Federal Circuit noted there was sufficient analysis for its assessment, and that Mr. Malackowski had met a showing of “baseline comparability.” With respect to apportionment, the Federal Circuit agreed with Judge Andrews noting:

This case is interesting to A2C, because it concerns litigation strategy and second bites at the damages apple. Had defendant’s counsel not been so successful in its first Daubert motion, would an appeal have been subsequently more successful? Additionally, would different jury instructions have afforded a new trial on damages for only the single patent?

Sprint Communications Company v. Time Warner Cable (Federal Circuit Opinion Modified March 18, 2019)

The Federal Circuit issued an opinion on March 18, 2019 that helps define what may be considered “a comparable” in a reasonable royalty analysis. In this case, a jury returned a verdict hitting Time Warner with a $140 million damages figure to be paid to Sprint. For its part, Sprint’s damages expert had offered a reasonable royalty rate analysis based, in part, on an earlier verdict from a different matter involving Sprint and Vonage.

Time Warner appealed the matter to the Federal Circuit on damages, contending that the district court improperly admitted the Sprint/Vonage verdict involving technology of disputed similarity and a different carrier. The Federal Circuit decided that the district court did not err in its admission:

The opinions of the Federal Circuit in this matter appear to broaden the scope of comparability so as to include comparable jury verdicts. But consistent with earlier Federal Circuit opinions, the differences between any comparable and the hypothetical negotiation at hand should be addressed.

In addition to the issue of comparability, the court made an interesting comment on apportionment. The court states that if the Georgia Pacific analysis is done correctly, then the analysis embodies apportionment principles. This suggests that apportionment may be done through the factor analysis, including at least via Factors 9 and 13:

We expect to see this language appear with frequency in responses to Daubert motions on apportionment.

Princeton Digital Image Corp. v. Ubisoft Entertainment SA, et al. (Opinion December 11, 2018)

We have previously written about a damages opinion having been excluded for relying on a jury verdict from an unrelated matter. Another recent effort to use a jury verdict has also been excluded in this opinion from Judge Burke in the District of Delaware.

In the opinion, Judge Burke observes that some may reasonably conclude that courts appear to have established jury-verdict reliance as per se unreliable:

In contrast to this view, however, Judge Burke suggests no such blanket rule should be presupposed to exist:

Judge Burke goes on to describe in a footnote a scenario where a jury verdict might prove relevant for damages:

It appears clear to us that the lesson remains to exercise extreme caution when entertaining use of a jury verdict in damages analysis. While the unique facts of a case may support such reliance, those occasions will likely prove exceedingly rare.

Chrimar Holding Company, LLC et al. v. ALE USA Inc. et al. (Federal Circuit – Decided May 8, 2018)

Plaintiff damages expert Robert Mills had his analysis excluded in part by Judge Love of the Eastern District of Texas. The part that was not excluded formed Mr. Mills’ testimony in trial which resulted in a damages award of $324,558 for ALE’s patent infringement.

ALE challenged the damages award at the Federal Circuit stating that, “Mr. Mills, in calculating a reasonable royalty, (1) relied on licenses not comparable to the hypothetical negotiation for the present case; (2) did not adequately separate the value of patented features from the value of standardization and the value of nonpatented features; and (3) prejudicially referred to ALE’s total net revenue and profit.” The Federal Circuit sided with Chrimar and found ALE’s arguments wanting.

With regard to the first issue, the court noted that there was not sufficient basis to exclude Mr. Mills in the Daubert motion nor in the JMOL phase of the matter, and that his license analysis satisfied the standard of “reasonable adjustments for differences in contexts.”

Regarding the second issue, the court said that Mr. Mills relied upon a standards expert, and a damages expert has right to do so for their own opinion.

Finally, for the third issue, the court explained that it was ALE itself which had first “opened the door” to introduction of defendant’s net revenue.

As in Exmark v. Briggs, the court appears perhaps to afford EMV somewhat greater latitude as part of a comparable license approach, especially where licenses make reference to a unit larger than what might otherwise be considered “the smallest saleable unit” under other analytic approaches to damages.

Greatbatch Ltd. v. AVX Corp. and AVX Filters Corp. Court Grants Motion to Set Aside Verdict (March 30, 2018)

Delaware District Court Judge Stark granted AVX’s motion to set aside the damages verdict. The reason for the set aside was solely because the jury verdict form did not separate damages for each patent accused, and thus constituted a damages award for all patents and all products.

Judge Stark himself observes this case possesses “a convoluted history.” It seems that Judge Stark sanctioned AVX because of its late production of core technical documents relevant to the infringement issues on one of the four patents in suit (i.e., the ‘715 patent). The sanctions levied involved the judge granting a motion for summary judgment that all of AVX’s Ingenio products infringed the ‘715 patent. When the case proceeded to trial concerning the three other patents in suit, the jury returned a damages award in the form of a lump sum of $37.5 million for infringement of all four patents by all of AVX’s accused products. After the jury trial, the court granted a motion for reconsideration of the sanctions. A new trial for the ‘715 patent was held and the result was that the jury found no infringement concerning some of AVX’s Ingenio products, while finding infringement with regard to a smaller subset of Ingenio products.

The reason for the jury award set-aside is attributed to the jury verdict form, which failed to request damages figures on a patent-specific and product-specific basis. Citing two Federal Circuit cases (Verizon Services Corporation v. Vonage Holdings Corp. and DDR Holdings, LLC v. Hotels.com, L.P.), the judge’s opinion seems to provide the only solution – a new trial on damages.

Important to note, AVX initially requested that the verdict form offer a damages line on a patent-by-patent basis; however, Greatbatch prevailed with its insistence for the general verdict form ultimately provided to the jury. Note in this instance, however, that because not all accused Ingenio products were found to infringe the ‘715 patent, a patent-by-patent jury verdict form might not have obviated the need for a new trial on damages under these specific circumstances.

Ericsson Inc., et al. v. TCL Communication Technology Holdings, LTD., et al. (Jury Verdict Vacated on March 8, 2018)

Judge Payne of the Eastern District of Texas vacated a jury verdict awarding Ericsson a lump sum of $75 million for infringement of its ‘510 patent by TCL Communication. Noting “for reasons stated in a contemporaneous order” (which appears not to be on the PACER docket yet), Judge Payne decided that Plaintiff expert Robert Mills’ damages analysis – partially adopted by the jury – was not acceptable as a matter of law.

From defendant’s motion for judgment as a matter of law, it appears Mr. Mills relied upon sales forecasts which included products not accused of infringing. Further, he relied upon a survey expert who did not tie analysis to the claimed invention. He is also accused of violating the entire market value rule. Finally, Mr. Mills is accused of failing properly to discount his damages analysis back to the date of the hypothetical negotiation, having instead discounted back to the date of notice (i.e., that date damages began to accrue). All but the ultimate issue regarding discounting seem sound reasons for vacating a jury award.

This last alleged transgression is most interesting, however, because it is unclear which date might make most sense for discounting a damages value. Mr. Mills appears to have derived a lump-sum damages award using a hypothetical negotiation construct. The award appears to be based upon past and future (expected) infringement. Mr. Mills then adjusted his lump sum back to the date damages should begin to accrue based upon notice. The JMOL legal argument claims, in contrast, that he should have discounted the award back to the date of the hypothetical negotiation (which was years before the notice date):

The citation above to Wang Labs, Inc. v. Toshiba Corp. appears unrelated to the specific issue of discounting. Similarly, LaserDynamics addresses an issue unrelated to discounting a royalty payment. In both those cited cases, the expert failed to assess a reasonable royalty at the time of the hypothetical negotiation, and instead used the date damages began to accrue as the hypothetical negotiation date. This appears to be different from what Mr. Mills did. Defendants are not arguing that Mills used the wrong hypothetical negotiation date: rather, they are arguing that his proffered lump-sum damages should have been discounted to the date of the hypothetical negotiation. Judge Payne’s order, however, does not elaborate on which of the defendant’s arguments he found persuasive; nor does it afford guidance with regard to the specific issue of proper discounting.

Exergen Corporation v. Kaz USA, Inc. (CAFC decided March 8, 2018)

The Court of Appeals for the Federal Circuit affirmed the damages analysis conducted by Exergen’s damages expert, Barry Sussman, but remanded the case back to the Massachusetts District court for further proceedings.

The two patents at issue involve a thermometer which takes body temperature from the forehead. The jury verdict form shows that Exergen was awarded both reasonable royalty and lost profit damages. The questions before the CAFC were: 1) whether the royalty rate, amounting to 71% of defendant-Kaz’s projected per-unit net profit, was adequately supported, and 2) whether lost profits should have included infringing sales to CVS, a store which did not sell plaintiff-Exergen’s competing thermometer.

Citing to Asetek v. CMI USA, the CAFC determined that the damages could be split between reasonable royalties and lost profits. In its affirmation, the court notes that the expert provided substantial evidence from analysis of the Georgia Pacific factors to support the reasonable royalty damages award. The court also found that the lost profits analysis showed that in a but-for world where Kaz was not selling to CVS, Exergen would have been the only other branded product available for CVS to sell, and that such information was sufficient to support the lost profits award:

Interestingly, because the court reversed the jury’s finding on infringement of claims of one of the two patents-in-suit, the parties agreed that the damages would need to be recalculated: accordingly, the CAFC remanded the damages issue for further proceedings. This is odd given plaintiff’s damages expert was allowed to offer one reasonable royalty damages figure for both patents, and did not proffer a per-patent rate.

Prisua Engineering Corp. v. Samsung Electronics Co., Ltd., et al. (Verdict February 26, 2018)

A jury in the Southern District of Florida found that Samsung infringed Prisua’s asserted patent claims and owed $4.3 million for that infringement. Prior to that verdict, Judge Moore issued opinions and orders on cross Daubert motions, as well as multiple motions in limine.

Of interest is that neither of the Daubert motions – on Prisua’s expert or Samsung’s expert – were granted; however, a critical motion in limine was.

Prisua moved to exclude Samsung’s damages expert, Mr. Lettiere, based upon his use of information that: 1) post-dated the hypothetical negotiation, 2) pertained to litigation settlements, 3) relied upon software licensing agreements, and 4) used the market-based approach. Citing Lucent and Sinclaire Refinery, Judge Moore denied #1. The judge declined to grant #2 based upon citations from Cornell, but reserved judgment. With regard to #3, Judge Moore observed, “The Court is persuaded that the ArcSoft Licensing Agreements—which directly relate to the allegedly infringing technology—are relevant and the probative value of Mr. Lettiere’s related testimony outweighs the risk of prejudice.” Finally, with regard to #4, the market-based approach was permitted to stand.

Samsung moved to exclude Mr. Leathers’ damages analysis because he relied upon Prisua’s pre-litigation offer to Samsung to license its patents. While such basis for exclusion appears compelling, Judge Moore refused to exclude via Daubert the implied pre-litigation benchmark rate.

In a motion in limine, however, Samsung successfully argued exclusion of the licensing negotiations between Prisua and Samsung:

From Samsung’s JMOL on damages, it appears that Mr. Leathers advanced the $0.09 royalty rate, but represented that the rate was derived using the ArcSoft Licensing Agreement. We will see whether Judge Moore finds this testimony (and the implied disregard of the pre-litigation offer) compelling, or whether instead he will grant the JMOL offered by Samsung.

Limelight Networks, Inc. v. XO Communications and Akamai Tech. (Opinion February 2, 2018)

On February 2, 2018, Judge Gibney re-struck the expert opinions of Stephen Prowse and Paul Meyer.

Dr. Prowse had applied the Rubenstein Bargaining Model to the hypothetical negotiation. The judge decided that such a model was properly tied neither to the facts of the case, nor to the patents in suit. Judge Gibney found Mr. Meyer’s reasonable royalty analysis based upon comparable licenses no more relevant than Dr. Prowse’s. Judge Gibney found Mr. Meyer’s comparable license analysis failed to establish that the agreements compare economically to the hypothetical license at issue.

Dr. Prowse’s analysis was excluded because it was deemed to be similar to the Nash Bargaining Model and the 25% rule of thumb, both of which have been excluded in the past. The concept of a Rubenstein Bargaining Model is that where two parties engage in a negotiation, the party with the most patience and least concern receives the benefit of that patience. In his effort to apply the model, Dr. Prowse used each company’s weighted average cost of capital as a proxy for patience.

This game theoretic model is much like other game theory models which mathematically conclude that the party with the most bargaining power or least concern enjoys the better outcome. The problem with such models is that they are irrelevant as a guide for damages experts seeking to determine the outcome of a hypothetical negotiation. In a hypothetical negotiation, there is no assessment for patience: only consideration for the value of the patented technology to those using the technology. Unlike the Rubenstein Model (and the Nash Model), the hypothetical negotiation is not iterative and the parties do not have endless time to complete the negotiation. While Judge Gibney did not delve into such specifics, he notes that:

Mr. Meyer’s exclusion based upon the use of comparables is possibly the more interesting of the two damages exclusions. Mr. Meyer had attempted to assess economic and technical comparability. In the end, however, the judge found the expert did not apportion the comparable licenses on a patent-specific basis; and when using a company acquisition price as another comparable, the expert failed to determine what portion of that acquisition was specifically for the comparable patent.