Patent licenses arising out of litigation may play a larger role in reasonable royalty determinations for patent litigation damages after the Federal Circuit’s recent opinion in ResQNet.com v. Lansa.24 Generally, experts apply the fifteen Georgia-Pacific factors to arrive at a reasonable royalty for patent infringement damages.25 The analysis is an inexact science, built around a hypothetical negotiation between the patentee and the infringer just before the infringement began.26 While this basic framework has not been altered by the decision, ResQNet and the cases following it may have increased the importance of litigation licenses for the Georgia-Pacific analysis and thus also the likelihood of admissibility of these licenses at trial, though the context of litigation still leaves them subject to greater scrutiny than nonlitigation licenses. Under these cases, settlement negotiations leading to these licenses may also be discoverable, as they can be relevant to the calculation of a reasonable royalty.
Courts have generally not considered litigation licenses valid sources of data for reasonable royalty determinations in patent cases. As the Supreme Court stated in Rude v. Westcott, “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,” because motives and considerations beyond the perceived value of the patents may influence the payment amount.27
Following ResQNet, litigation licenses may be more important to the damages analysis in patent cases. In ResQNet, the accused infringer appealed the district court’s damages award for infringement of a patent relating to screen-recognition software, arguing that the district court erred in relying on the faulty analysis of ResQNet’s damages expert.28 At issue was a set of seven licenses.29 Five of the seven licenses did not mention the patents in suit and had no link to the relevant technology, yet conveniently had “extremely high rates.”30 The Federal Circuit found that ResQNet’s expert applied faulty methodology in relying on these licenses to raise the royalty rate and that the district court erred in relying on such “speculative and unreliable evidence.”31 In contrast, the Federal Circuit found that the remaining two licenses, which arose out of litigation over the patents in suit, were the “most reliable” of the licenses on the record and had a much lower royalty rate than the five “unrelated licenses.”32
On remand, the Federal Circuit instructed the district court not to rely on unrelated licenses for the reasonable royalty analysis, and implicitly permitted the consideration of litigation licenses.33 That said, the court also recognized potential concerns with litigation licenses, such as the problem of license fees negotiated under the threat of high litigation costs; artificial depression of licenses by widespread infringement; and other pricing influences based on the “fact that an infringer had to be ordered by a court to pay damages.”34
A string of cases in the Eastern District of Texas have interpreted the ResQNet decision, falling on both sides of the divide. Two cases have relied on ResQNet for the proposition that litigation licenses are admissible as representative licenses for a royalty determination.35 Subsequently, another judge in the Eastern District of Texas declined to extend the ResQNet holding to reach this conclusion, essentially maintaining the status quo.36 A recently decided Federal Circuit case, however, suggests that litigation licenses, even if admissible, must be analyzed carefully or they can skew the results of the reasonable royalty analysis.
Relying on ResQNet¸ one Eastern District of Texas judge held in Datatreasury Corp. v. Wells Fargo & Co. that litigation licenses are admissible at trial.37 The court denied the defendants’ motion, in limine, to prevent the plaintiff from offering evidence of litigation licenses for calculation of a reasonable royalty, consideration of nonobviousness, and determination of commercial success.38 The court stated that any concerns regarding their reliability should be addressed by adjusting their weight in an analysis rather than by excluding them outright.39 Recognizing that ResQNet involved a bench trial, the court suggested that any concerns about potential jury prejudice may be addressed through a jury instruction that gives the jury guidance on applying litigation-related licenses.40
Another judge in the Eastern District of Texas stated that litigation licenses may be “central to the fact-finder’s determination of damages.”41 In Tyco Healthcare Group LP v. E-Z-EM, Inc., the plaintiffs sought to compel the defendants to produce documents relating to settlement negotiations that culminated in a patent license, which the defendants had withheld on the basis of a settlement negotiation privilege.42 The court determined that this privilege does not apply in light of ResQNet.43 This is because under ResQNet, litigation licenses are potentially admissible and important and therefore parties are entitled to discovery relating to the settlement agreements.44 Although Tyco did not involve a determination of whether litigation licenses are admissible at trial for damages purposes, the case suggests that litigation licenses are potentially admissible to show damages and that the corresponding license negotiations are discoverable for that reason.
In Fenner Investments, Ltd. v. Hewlett-Packard Co., another judge took a more conservative view of the ResQNet decision, finding that settlement licenses entered into as a result of prior litigation with third parties would not be admissible at trial.45 The Fenner court found that the ResQNet decision did not alter the admissibility of agreements entered into under the threat of litigation, stating that the litigation-related licenses were part of the record and that their admissibility was not before the ResQNet court.46 The Fenner court further noted that district courts routinely exclude litigation licenses because of potential prejudice and jury confusion, citing extensively to case law from a number of districts.47 Two differences helped distinguish Fenner from ResQNet: the litigation licenses were being used to analyze different Georgia-Pacific factors; and there was a potential concern of jury confusion in Fenner, but not in ResQNet, because the litigation licenses had been admitted during a bench trial.48 Based in part on this reasoning, the court granted the plaintiff’s motion to exclude litigation licenses from being introduced at trial under Fed. R. Evid. 403, finding that the potential for prejudice and jury confusion substantially outweighed any probative value the litigation licenses may have.49
Most recently, the Federal Circuit’s opinion in Wordtech Systems, Inc. v. Integrated Network Solutions, Inc. suggests that parties must still use caution when relying on litigation licenses for royalty determinations.50 In Wordtech, the defendants challenged the excessiveness of the jury’s award for patent infringement as unsupported by the evidence at trial.51 The Federal Circuit agreed and remanded for a trial on damages.52 At trial, plaintiff Wordtech had offered no expert on damages, but through its president had offered damages evidence including 13 patent license agreements that Wordtech had granted to third parties for rights to some or all of the patents in suit.53 The Federal Circuit agreed with the defendants that these patent licenses did not support the verdict.54 The jury had awarded a lump-sum royalty, but for the two lump-sum licenses out of the 13 license agreements, Wordtech had failed to provide the jury sufficient evidence of how these sums related to the volume of sales or projected sales.55 The court also found that the remaining 11 licenses, which used running royalties (rates as a percentage of sales), failed to support the verdict because the running royalty rates were far lower than the 26.3 percent rate that the jury had effectively awarded.56 The Federal Circuit also noted that several of the running royalty licenses arose out of litigation and quoted ResQNet for the proposition that “litigation itself can skew the results of the hypothetical negotiation.”57 Thus, the court seemed to suggest that litigation licenses are less reliable than other licenses for determining a reasonable royalty.
Based on the ResQNet case and its progeny, parties in patent suits should be aware that litigation licenses may be admissible, even in jury trials, and that settlement agreements and negotiations that result in patent licenses may be discoverable. At the same time, looking to the cautionary guidance of Fenner and Wordtech, parties cannot necessarily rely on the admissibility or reliability of litigation licenses as a strategy for establishing a reasonable royalty. Finally, parties considering settlement should also be aware that the settlement and negotiation process could play a role in determining a reasonable royalty in later litigation.