United States District Court, N.D. California, San Jose Division
January 27, 2015
SENTIUS INTERNATIONAL, LLC, Plaintiff,
MICROSOFT CORPORATION, Defendant.
ORDER GRANTING-IN-PART MOTION TO EXCLUDE (Re: Docket No. 131)
PAUL S. GREWAL, Magistrate Judge.
In the aftermath of ResQNet.com, Inc. v. Lansa, Inc.,  settlement agreements are now indisputably fair game in assessing patent damages. Despite the suggestions of its own precedent and that of the Supreme Court,  the Federal Circuit recognized that "the most reliable license in the record [may be one] that arose out of litigation." But such licenses are hardly now per se admissible; the Federal Circuit itself has cautioned as much. This case illustrates why.
Robert Mills is a damages expert. Plaintiff Sentius International, LLC has tendered Mills for his opinions regarding the proper measure of damages adequate to compensate Sentius for the alleged infringement by Defendant Microsoft Corporation. In calculating patent damages, Mills not only relied on a settlement agreement, but also a decision in an earlier case regarding a motion for judgment as a matter of law. Mills also relies on the so-called "income approach" theory as well as the overall profitability of the accused products.
Microsoft now brings a Daubert motion seeking to exclude Mills' testimony. Because Mills' methodology does not entirely pass muster under Daubert and Rule 403, the court GRANTS Microsoft's motion in part, excluding Mills' testimony to the extent that it relies on the Lucent JMOL and the Arendi settlement agreement. Because Mills' income approach theory is methodologically sound, the court DENIES Microsoft's motions to the extent that it seeks to exclude Mills' testimony relating to this approach. Finally, the court GRANTS Microsoft's motion to exclude Mills' testimony relating Microsoft's and Microsoft Office's overall profitability, such that Mills may not refer to Microsoft or Office's profitability but may refer to Microsoft's profit margins.
Expert testimony may only be admitted in a manner consistent with the Federal Rules of Evidence, Daubert,  Kumho  and more recent appellate court progeny. Federal Rule of Evidence 702 allows admission of "scientific, technical, or other specialized knowledge" by a qualified expert if it will help "the trier of fact to understand the evidence or to determine a fact in issue." When considering expert testimony, the trial court serves "as a gatekeeper' to exclude junk science that does not meet Federal Rule of Evidence 702's reliability standards."
An expert witness may provide opinion testimony if: (1) "the testimony is based upon sufficient facts or data;" (2) "the testimony is the product of reliable principles and methods; and" (3) "the expert has reliably applied the principles and methods to the facts of the case." "Under Daubert, the district judge is a gatekeeper, not a fact finder.' When an expert meets the threshold established by Rule 702 as explained in Daubert, the expert may testify and the jury decides how much weight to give that testimony." The inquiry into the admissibility of an expert opinion is a "flexible one" where shaky "but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion."
A trial court must be sure that its review of expert testimony focuses "solely on principles and methodology, not on the conclusions that they generate." "A judge must be cautious not to overstep its gatekeeping role and weigh facts, evaluate the correctness of conclusions, impose its own preferred methodology, or judge credibility, including the credibility of one expert over another. These tasks are solely reserved for the fact finder." The Federal Circuit recently clarified that this limitation of the gatekeeping role of the judge to the exclusion of "testimony based on unreliable principles and methods is particularly essential in the context of patent damages." This is because "questions regarding which facts are most relevant or reliable to calculating a reasonable royalty are for the jury.'"
35 U.S.C. § 284 provides that upon "finding for the claimant, the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer, together with interest and costs as fixed by the court." The goal of the damages award is not to punish the infringer, but rather to make the patentee whole by ascertaining what the patent holder would have made had the infringer not infringed. Damages may take the form of "reasonable royalty [the patentee] would have received through arms-length bargaining" in a hypothetical negotiation. The hypothetical negotiation approach "attempts to ascertain the royalty upon which the parties would have agreed had they successfully negotiated an agreement just before infringement began." No matter the form, "[t]he burden of proving damages falls on the patentee."
A patent holder may use settlement agreements to establish reasonable royalty damages "under certain limited circumstances." This is because, as the Supreme Court observed over one hundred years ago, settlement agreements have little probative value to establish patent damages:
It is clear that a payment of any sum in settlement of a claim for an alleged infringement cannot be taken as a standard to measure the value of the improvements patented in determining the damages sustained by the owners of the patent in other cases of infringement. Many considerations other than the value of the improvements patented may induce the payment in such cases. The avoidance of the risk and expense of litigation will always be a potential motive for a settlement.
That said, a settlement agreement may be admissible when "the most reliable license in [the] record [is one that] arose out of litigation." Accordingly, to rely on a settlement agreement to establish patent damages, a patent holder "must only show that [its expert's] consideration of [the] patent litigation settlement is sufficiently reliable to be admissible under Daubert. "
Sentius asked Mills to estimate the damages caused by Microsoft's alleged infringement of its United States Reissue Patent Nos. RE 40, 731 and RE 43, 633 based on certain features of applications within the Microsoft Office suite products. Mills used the Georgia-Pacific factors to determine a reasonable royalty to which the parties would have agreed in a hypothetical negotiation for a license to Sentius' patents. To determine damages for Microsoft's use of background grammar and spell checkers, Mills used two main approaches that are at issue here.
First, Mills looked to an earlier patent infringement dispute between Microsoft and Alcatel-Lucent. In that dispute, Lucent alleged that "certain features of [Microsoft's products], when used, practice the methods of [Lucent's patent]." After a jury found infringement and awarded Lucent damages against Microsoft, the Federal Circuit vacated the jury's damages award and remanded the case for a new trial on damages. On retrial, the jury awarded Lucent damages of $70 million and Microsoft moved for judgment as a matter of law. Granting-in-part Microsoft's motion, Judge Huff found that "as a matter of law a reasonable jury could not have returned a verdict in excess of $26.3 million based on the evidence of record." Ultimately, the parties resolved their dispute, entering into a settlement agreement in which Microsoft agreed to pay Lucent [REDACTED\].
Mills determined that he could look to the Lucent dispute for guidance because Sentius' technical expert Vijay Madisetti concluded that the technology at issue in Lucent was comparable to the Sentius reissue patents at issue here. Mills claimed that his report explained how he "adjust[ed] the [REDACTED\] Paid by Microsoft to Lucent" to provide "insight into a reasonable royalty" for Microsoft's use of the asserted patents for the accused spell and grammar check features. However, rather than using the [REDACTED\] figure Microsoft agreed to pay in the settlement agreement as the starting point for this analysis, Mills noted that in the Lucent JMOL, Judge Huff awarded Lucent $26.3 million in damages. Mills used Judge Huffs analysis to conclude that $24.6 million of those damages were attributable to Microsoft Outlook. Dividing that $24.6 million figure by [REDACTED\] - the approximate number of Outlook sales made through the life of the asserted patent - Mills concluded that "Microsoft effectively paid Alcatel-Lucent no less than $0.22 per copy of Outlook sold in the United States."
Mills determined the value consumers place on the accused spelling and grammar check features at issue here is "considerably greater" than the value consumers place on the technology at issue in Lucent.  To account for this additional value, Mills said that he needed to make a "significant upward adjustment" to the per-copy amount Microsoft effectively paid Lucent for Outlook when using this account to determine damages for Sentius. He then increased the percopy amount Microsoft effectively paid by a factor of two for accused products that include both the accused spell and grammar check features. Ultimately, even though his calculations were based on the damages Judge Huff calculated in the Lucent JMOL, Mills opined that the "financial terms of the settlement agreement" between Microsoft and Lucent indicated that "an effective percopy royalty of not less than $0.44 is reasonable" for Microsoft's use of the asserted patents here.
Second, Mills used an alternative theory which he called the "income approach" to value damages in relation to the accused spell and grammar check features. To support this theory, Mills relied on the results of a survey conducted by another Sentius expert, William Wecker. Wecker's survey purportedly established that approximately 11.2% of Microsoft consumers who purchased and used the accused products and features would not have purchased the accused products if they had not included the accused spell and grammar check features. Based on this 11.2% figure, Mills estimated Microsoft would have lost [REDACTED\] if it had removed the accused features. Mills divided the profit Microsoft made on that revenue by the number of accused units sold during the damages period to yield an average at-risk profit per copy of $1.99. Analyzing the relative bargaining power of the parties, Mills determined that Sentius would have been in a position to negotiate for "more than 22% percent and perhaps as much as 50% of the atrisk profit." He determined that an effective per-copy royalty of $0.44 was a "conservative outcome" because such a royalty would only award Sentius about 22% of the $1.99 average at-risk profit per copy.
Ultimately, Mills concluded that based on the guidance provided by the Lucent dispute and by his income approach theory, an effective per-copy royalty of not less than $0.44 was "reasonable" for accused products that included both the accused spell and grammar check features. For accused products that included only the accused spell checker, Mills opined that "an effective per-copy royalty of not less than $0.22 is reasonable."
To calculate a reasonable royalty for Microsoft's use of the accused smart tags feature, Mills looked to a litigation settlement agreement between Microsoft and Arendi for guidance. In actions filed in the United States District Court for the District of Delaware and the District Court in The Hague, Arendi S.A.R.L alleged that Microsoft's inclusion of smart tags in its products infringed Arendi's United States and European patents. Before trial, the parties entered into a settlement agreement in which Microsoft promised pay to Arendi approximately [REDACTED\]. Mills used this [REDACTED\] calculate a $0.03 per-copy royalty rate for Microsoft's use of the accused smart tags feature.
This court has jurisdiction under 28 U.S.C. §§ 1331 and 1338. The parties further consented to the jurisdiction of the undersigned magistrate judge under 28 U.S.C. § 636(c) and Fed.R.Civ.P. 72(a).
At issue is whether Mills' survey and opinion should be excluded from trial. The court agrees that Mills' reliance on the Lucent and Arendi disputes is improper and excludes Mills' testimony to the extent that he relies on these disputes. The court also excludes reference to Microsoft's overall profitability and the profitability of Office as unfairly prejudicial to Microsoft but allows Mills to refer to Microsoft's profit margin to the extent necessary to support his income approach theory. The court finds that Mills' income approach theory is sufficient to get to the jury and that Mills may apply a royalty to each accused product without showing that someone has used the accused features in in each product sold.
First, Mills' reliance on the Lucent JMOL and the Arendi settlement is erroneous. Sentius creatively argues that Mills primarily relied on the [REDACTED\] Microsoft agreed to pay to Lucent in settlement and only used the JMOL for the "limited purpose" of determining what portion of that the amount was attributable to a license for Microsoft Outlook. But although Sentius mentions the [REDACTED\] in expert report,  both the report and his deposition show that he derived his reasonable royalty rate from $24.6 million damages award Judge Huff granted in the Lucent JMOL. Sentius contends that the court should not fault Mills for his "conservative" decision to base damages on the amount Judge Huff awarded in the Lucent JMOL. However, the fact that Mills' royalty rate would have been higher if he based his calculations on the [REDACTED\] settlement agreement figure does not insulate Mills' use of the Lucent JMOL from Daubert scrutiny. A damages theory that stems from an erroneous methodology is not admissible even if it results in a low ultimate damages figure.
Sentius has not met its burden to show that Mills' use of the Lucent JMOL passes muster under Daubert for several reasons. Neither party has cited to a case in which a court addressed whether a patent holder may use a JMOL to derive damages for patent infringement. However, it is clear that use of JMOLs to establish damages based on a reasonable royalty theory presents concerns similar to the issues that make courts wary about using settlement agreements to establish patent damages.
Like settlements agreements, JMOLs occur in a different context than a hypothetical negotiation. As stated above, courts have long been reluctant to allow use of settlement agreements to establish reasonable royalty damages in part because these agreements are made in a different context than the situation in which parties face in a hypothetical negotiation. Similarly, the difference between the procedural posture of a JMOL and the context of a hypothetical negotiation cautions against use of JMOLs to establish damages based on a hypothetical negotiation theory. The Lucent JMOL did not, as Sentius claims, address "the exact same question that Mr. Mills needs to answer about the outcome of a hypothetical negotiation." In the JMOL Judge Huff did not independently assess, as the parties would in a hypothetical negotiation, what royalty would be reasonable. Rather, Judge Huff analyzed the more limited question of whether Lucent had offered sufficient evidence at trial to support the jury's $70 million damages award. Because the Lucent JMOL does not address the "exact same question" Mills faced, it has limited probative value on the royalty to which the parties would have agreed in a hypothetical negotiation.
The disparate nature of the questions Judge Huff and Mills sought to resolve is not the only way in which the Lucent JMOL differs from the hypothetical negotiation at issue here. The Lucent JMOL was a judicial decision from another court regarding a plaintiff who is not involved in this case and a patent that Sentius does not assert here. Further, the Lucent JMOL concerned a hypothetical negotiation that did not take place at the same time as the hypothetical negotiation at issue here. Sentius has not cited to any legal, academic or other authority that says that parties would have considered a JMOL decision at all when determining reasonable royalties, let alone a JMOL that occurred in such a different context than the hypothetical negotiation at issue. To the contrary, courts have found that settlement agreements have limited probative value when they, like the Lucent JMOL, occurred in a different time frame than the hypothetical negotiation at issue. As a result, these vast differences, coupled with the lack of any authority supporting Mills' use of the Lucent JMOL, establish that Sentius has not met its burden to show that Mills' analysis of the Lucent JMOL is sufficiently reliable to be admissible under Daubert. Accordingly, Mills' damages testimony is excluded to the extent that it relies on the Lucent JMOL.
Further, even if the court found that Mills' analysis relating to the Lucent JMOL passed muster under Daubert, it would still exclude the analysis as unfairly prejudicial to Microsoft under Rule 403. As stated above, the Lucent JMOL has little probative value, and it is clear that its admission will "cause unfair prejudice, confuse the issues, and waste time." In particular, admission of the Lucent JMOL will almost certainly cause Microsoft to waste time explaining and re-litigating the issues and circumstances involved in the Lucent dispute, confuse the jury about the facts and circumstances of that dispute and prejudice the jury by suggesting that Microsoft is a serial infringer that frequently is found liable for damages for patent infringement. Given that the Lucent JMOL is of limited usefulness on the issue of Sentius' patent damages, these dangers substantially outweigh the Lucent JMOL's probative value.
The story is similar with the settlement agreement in Arendi. Because the Federal Circuit "recognize[s] that settlement agreements can be pertinent to the issue of reasonable royalties, " the Arendi agreement is not inadmissible merely because it is a settlement agreement. But that does not end the inquiry. The Federal Circuit also recognizes that "in attempting to establish a reasonable royalty, the "licenses relied on by the patentee in proving damages [must be] sufficiently comparable to the hypothetical license at issue in suit." The "use of past patent licenses under [ Georgia-Pacific ] factors 1 and 2 must account for differences in the technologies and economic circumstances of the contracting parties." A damages expert also may not rely on licenses that are from "vastly different situation[s]."
Mills has not adequately accounted for the vast differences between the "economic circumstances" of the parties in the Arendi settlement agreement and the parties in the hypothetical negotiation at issue here. In the Arendi settlement, Arendi alleged that Microsoft had infringed United States and European patents, whereas here Sentius alleges that Microsoft has infringed only United States patents. Further, in settlement, Arendi granted Microsoft [REDACTED\] while in this hypothetical negotiation Sentius would have granted Microsoft rights only to the two reissued patents it asserts. Mills recognized that these differences impacted the Arendi settlement because he noted that" [REDACTED\]." Yet after recognizing this, Mills simply concluded that "no further adjustment [was] necessary to account for this issue" because he calculated a "conservatively low effective royalty rate" for the Arendi settlement." Mills then opined that Microsoft would likely claim that a "downward adjustment is warranted" to the Arendi royalty rate because Microsoft would only receive a United Stated license to Sentius' two asserted patents [REDACTED\]. But a simple "downward adjustment" is not sufficient in light of Mills' failure to quantify the value that [REDACTED\] contributed to the Arendi settlement. And, like the allegedly "conservative" nature of Mills' use of the Lucent JMOL, Mills' supposed arrival at a "conservative" final number for the effective royalty rate of the Arendi agreement does not cure his failure to use a reliable methodology to arrive at that number.
Sentius contends that Mills' failure to account for Arendi's European infringement claim and [REDACTED\] is justified in part because Microsoft's Rule 30(b)(6) witness did not know how "any specific facts" impacted negotiation of the Arendi agreement. Even if Sentius is correct that Microsoft did not provide insight on the impact of these facts, this does not excuse Mills from his duty to account for the differences between the Arendi settlement and the hypothetical negotiation here. Mills could still have looked to his own knowledge and experience to account for how these factors influenced the Arendi settlement. Without an accounting of the effect of the European infringement claims and value of the patents that Arendi did not assert, Mills could not meet his duty to "account for differences in the... economic circumstances of the contracting parties." Accordingly, Mills' analysis is excluded to the extent it relies on the Arendi settlement.
Second, Mills' income approach damages theory is sufficient to get to the jury. Microsoft claims that this theory must be excluded because it is "wholly derivative" of Mills' damages analysis based on the Lucent JMOL. However, to calculate damages based on his income approach, Mills did not begin with the Lucent JMOL or settlement agreement. Rather, Mills used Wecker's survey to calculate the revenue and profit that Microsoft would have allegedly lost if it had not included the accused spell and grammar check features in its accused products. Mills then divided this "at risk" profit by the number of infringing units to obtain an average at-risk profit per copy of $1.99. Ultimately, Mills opined given the "relative bargaining strengths" of the parties, an effective $0.44 per-copy royalty is a "conservative outcome" in part because it gives Sentius approximately 22% of Microsoft's average at-risk profit per copy.
Mills' income approach theory is connected to his Lucent damages analysis in the sense that Mills ultimately arrived upon the same effective per-copy royalty rate of not less than $0.44 for use of the accused spell and grammar check feature that Mills had previously concluded was "reasonable." But Mills' income approach theory did not rely on figures from Lucent to yield the $0.44 per copy royalty rate. To the contrary, Mills used the income approach to independently arrive at the effective per-copy $0.44 royalty rate by using Wecker's survey results and Microsoft's revenue and profit to determine Microsoft's at-risk profit per copy and then analyzing how the parties would divide this at-risk profit based on their relative bargaining positions. Because it stands on its own ground, Mills' income approach theory is therefore not derivative of his flawed analysis relating to the Lucent dispute.
Microsoft argues that even if Mills' analysis is not derivative of his Lucent damages theory, it is still fundamentally flawed because Mills analysis of the parties' relative bargaining strengths is "hand-waving at best." However, Mills did not, as Sentius claims, follow an unprincipled "rule of thumb" approach similar to the theories that the Federal Circuit rejected in VirnetX and Uniloc to reach his conclusions about the relative bargaining power of parties. Unlike in VirnetX and Uniloc, where the experts relied on "rules of thumb" without establishing that those rules applied to the hypothetical negotiation at issue, here Mills analyzed how the various factors impacted the parties' bargaining strengths. To accomplish this, Mills' considered a non-exhaustive list of "[p]rinciple factors" relating to the hypothetical negotiation. Mills then determined that some of the factors favored Microsoft while others favored Sentius. Ultimately, Mills concluded that Sentius would be in a position to negotiate "more than 22 percent and perhaps as much as 50 percent of the at-risk profit."
Despite Mills' focus on the particularities of the parties' situation, Microsoft argued that Mills' analysis was insufficient because his "bargaining considerations" were either "generic, " favored Microsoft or did not support "any quantitative value." However, these complaints do not warrant exclusion of Mills' income approach theory because they either mischaracterize Mills' analysis or go to weight, not admissibility. Microsoft's assertion that Mills' analysis is flawed because some of Mills' factors "directionally would favor Microsoft" fails because Mills recognized that some of his factors would decrease Sentius' bargaining power. And Microsoft's complaint that Mills' analysis used "generic" factors or was not sufficiently quantitative goes to weight, not admissibility. Microsoft cites to no case in which a court excluded testimony because an expert did not precisely quantify his analysis of parties' relative bargaining positions. While the court agrees with Microsoft that Mills could have looked to other factors to determine the parties' relative bargaining power or examined his chosen factors in more depth, this is an issue that Microsoft can address at trial through cross-examination.
The court also rejects Microsoft's claim that it must exclude Mills' income approach theory under the entire market value rule. The Federal Circuit recognizes the entire market value rule as a "narrow exception" to the general rule that "royalties be based not on the entire product, but instead on the smallest salable patent-practicing unit.'" "A patentee may assess damages based on the entire market value of the patented technology only if the patentee can show that "the patented feature creates the basis for customer demand or substantially creates the value of the component parts." Mills may not estimate damages based on the entire market value of the accused products here because Mills acknowledges that claimed technology does not drive demand for accused products.
Microsoft claims that despites this acknowledgement, Mills nevertheless derived damages from the entire market value of the accused products. Microsoft contends that Mills violated the entire market value because he calculated the profit Microsoft would have allegedly lost if it had not included the accused features in its products and then concluded that Microsoft would have agreed to pay Sentius 22% of that lost profit. Put differently, according to Microsoft, Mills applied a running royalty rate to the entire portion of the revenue and profit Microsoft would have allegedly lost if it had not included the patented technology in its accused products. At oral argument, Microsoft said that this approach would grant Sentius 2.5% of Microsoft's profit on the entire Office suite or entire standalone application.
This interpretation not only distorts Mills' analysis, but would require exclusion of any damages analysis that is premised on the income approach theory. One can characterize any damages figure calculated under the income approach theory as a percentage of the defendant's total lost profits or revenue. Mills' income approach theory is not a hidden attempt to avoid the entire market value rule because Mills did not derive damages using Microsoft's revenue and profit from all sales of the accused products. Instead, Mills bases damages only upon Microsoft's profits and revenue that is associated with sales of the accused products that Sentius contents Microsoft would have lost without the inclusion of the accused features in its products.
Third, Mills may apply a royalty to each accused product without showing that someone has used the accused features in in each product sold. It is well-established that a defendant must perform the claimed method in the United States in order to infringe a method claim. Further, a patent holder may not receive damages for "sales of devices merely capable of infringing" without "proof" that those devices are "programmed for and actually executed the claimed method." But a patent holder is not "required to demonstrate a one-to-one correspondence between units sold and directly infringing customers" and "[p]roof of inducing infringement or direct infringement may be shown by circumstantial evidence."
Mills opined that his royalty rate for each accused feature should be applied to a royalty base of the total number of copies of the accused Office products and applications that contain that feature sold in the United States during the relevant time frame without trying to determine the extent to which consumers actually used each accused feature. Microsoft claimed that this was improper because Mills had reason to believe that that someone did not use each accused feature in each Office product sold in the United States. To support this claim, Microsoft notes that it had provided in this case [REDACTED\].
Microsoft is correct that Sentius is required to show "proof" that the accused devices "actually executed the claimed method." But under Chiuminatta, a patent holder may show such "proof" through "circumstantial evidence." Here, the record includes such evidence of circumstantial evidence of demand and use. As Sentius is not required to show a "one-to-one correspondence between units sold and directly infringing costumers, " this "circumstantial evidence of demand and use" provides sufficient proof to support Mills' use of a royalty base of all accused products that are capable of infringement.
Finally, Mills may not refer to the Microsoft's profitability as a whole and to the profitability of Microsoft's Office business at trial but may refer to Microsoft's profit margin to the extent necessary to supports its income approach theory. Microsoft's profitability is relevant to Mills' damages analysis because Microsoft's profitability indicates the importance of Office to Microsoft and the accused spell and grammar check features are components of Office. Mills also uses the profitability of Microsoft Office to calculate his income approach theory. However, presenting such large sums - [REDACTED\] respectively - would undoubtedly "skew the damages horizon for the jury."
Mills may not mention Microsoft's profitability at trial. He may use Office's profitability and revenue to the extent necessary to support his income approach theory but must refer only to Microsoft's profit margin and not to Office's overall profitability and revenue at trial. The court appreciates that jurors may be able to derive Microsoft's profits and revenue from its profit margin. On balance, however, the court concludes that this method will allow Mills to present his damages theory without unfairly prejudicing the jury against Microsoft.
The court GRANTS Microsoft's motion in part, excluding Mills' testimony to extent that it relies on the Lucent JMOL and Arendi settlement agreement. The court finds that Mills' income approach theory is sufficient and DENIES Microsoft's motion to the extent that it seeks to exclude Mills' testimony relating to this approach. Finally, the court GRANTS Microsoft's motion to exclude Mills' testimony relating to Microsoft's and Microsoft Office's profitability in part, holding that Mills may not refer to Microsoft or Office's profitability but may refer to Microsoft's profit margins to the extent necessary to support his income approach theory.