RICHARD G. ANDREWS

Richard K. Herrmann, Morris James LLP, Wilmington, DE; David Boies, Roseanne C. Baxter, Boies, Schiller & Flexner LLP, Armonk, NY; D. Michael Underhill, Richard S. Meyer, Edward H. Takashima, Patrick M. Lafferty, Boies, Schiller & Flexner LLP, Washington, DC; Matthew C. Lapple, Lapple IP Law, P.C., Irvine, CA, for Plaintiff AVM Technologies, LLC. Richard L. Horowitz, David E. Moore, Potter Anderson & Corroon LLP, Wilmington, DE; William F. Lee, Lauren B. Fletcher, Wilmer Cutler Pickering Hale and Dorr LLP, Boston, MA; David C. Marcus, Wilmer Cutler Pickering Hale and Dorr LLP, Los Angeles, CA., for Defendant Intel Corporation.

MEMORANDUM OPINION

RICHARD G. ANDREWS, District Judge.

Presently before the Court are a number of motions, including Intel's Daubert Motion to Exclude the Testimony of Larry Evans (D.I.191), Intel's Motion for Summary Judgment of No Damages (D.I.161), and Intel's Motion in Limine (D.I. 240, Ex. 15, Motion in Limine # 2, p. 3) to exclude testimony of Joseph Tran, an inventor of the '547 patent. Evans' proposed testimony—that a reasonable royalty can be established by one litigation settlement agreement involving a different patent—is not reliable. AVM's proposed inventor testimony concerning damages fares no better; it is mostly late-disclosed, unreliable expert testimony. Thus, Intel's Daubert motion will be granted; Intel's Motion in Limine will also be granted; but the Court will hold the Motion for Summary Judgment of No Damages in abeyance pending a further conference with the attorneys.

To better understand the instant motions, a brief description of the procedural history is helpful. On May 15, 2012, AVM submitted an expert report from Larry Evans relating to damages. In that report, Evans opined that AVM is entitled to reasonable royalty damages of $150 to $300 million “or more.” (D.I. 163 at Ex. 14 ¶ 70). Intel filed its motion for summary judgment of no damages on July 24, 2012. (D.I.161). The Court held oral argument on the summary judgment motion on September 28, 2012.

On October 31, 2012, Intel filed a Daubert motion to exclude Evans' testimony. (D.I.191). The motion was fully briefed. (D.I.192, 206, 212). On January 4, 2013, the Court issued an opinion, stating that the Court believed Evans' testimony should be excluded entirely but that its decision would be better informed after hearing live testimony from Evans, subject to cross-examination. (D.I. 230 at 7). The Daubert hearing was scheduled for February 1, 2013.

In connection with the pretrial order filed on January 18, 2013, Intel filed three motions in limine, including one motion partly seeking to preclude Joseph Tran from offering any expert testimony (D.I. 240 at Ex. 15, Motion in Limine # 2 at 3), and a second motion seeking to exclude evidence and testimony regarding (i) Intel's total accused revenues or Intel's overall corporate revenues and profits; (ii) the four Intel settlement agreements on which Evans relied; and (iii) opinions or documents not discussed in Evans' report. (D.I. 240 at Ex. 15, Motion in Limine # 3). The pretrial conference was held on January 25, 2013. During the pretrial conference, AVM argued that, whether the Court excluded Evans' testimony, AVM should be permitted to present testimony from Tran concerning damages.

On January 30, 2013, AVM filed a bench memorandum (D.I.255) and an offer of proof (D.I.257) regarding Tran's proposed testimony about damages. Intel responded the next day. (D.I.258).

On February 1, 2013, the Court held the Daubert hearing, at which Evans testified and was cross-examined. AVM disclosed at the hearing that, in light of the Court's January 4 opinion, its damages theory would rely solely on a single 2009 settlement agreement with the Wisconsin Alumni Research Foundation (“WARF”) concerning U.S. Patent 5,781,752 (“the '752 patent”). (D.I. 268 at 114). The parties also presented argument as to whether Tran should be permitted to testify about damages. The parties filed additional submissions after the February 1 hearing. On February 4, 2013, AVM filed a bench memorandum about Evans' damages testimony (D.I.262) as well as a letter providing further argument as to Tran's proposed testimony. (D.I.265). Intel filed a letter addressing the 2009 WARF agreement and the underlying facts of the litigation that led to that agreement. (D.I.264). On February 5, 2013, the parties filed dueling letters concerning the litigation that resulted in the 2009 WARF agreement. (D.I.266, 267).

For the reasons that follow, the Court grants Intel's Daubert motion. The Court also concludes that Tran should not be permitted to testify concerning damages beyond facts and information within his personal knowledge. Thus, Tran would not be permitted to testify about anything speculative or hypothetical, including what he would have done in hypothetical negotiations. With specific reference to AVM's amended offer of proof concerning Tran's testimony (D.I.257), Tran would be permitted to testify about the benefits of the '547 patent, the legal assignments of the '547 patent, and his actual negotiations with Intel in 2006. (D.I. 257 at 1 through second full paragraph on 3 and § V). For the reasons discussed below, however, the Court concludes that most of Tran's testimony must be excluded because it is untimely disclosed and unreliable expert testimony. Accordingly, the Court also grants Intel's motion in limine.

A. Evans' Testimony

At the Daubert hearing, AVM disclosed that Evans' expert opinion would rely solely on the 2009 WARF agreement and that it would not present a damages theory based on Intel's revenues or the three other license agreements (MicroUnity, Intergraph, and Transmeta) discussed in Evans' report, about which the Court had previously indicated its intent to exclude. ( See D.I. 230). Thus, with respect to Evans' testimony, this opinion focuses on whether Evans' reliance on the 2009 WARF agreement is sufficient to support AVM's damages theory and can be presented to the jury.

Evans' expert report states that “Intel's patent licensing history is principally the result of patent infringement litigation settlements.” (D.I. 163 at Ex. 14 ¶ 41). The report provides seven examples of such license agreements. As one example, Evans describes Intel's October 16, 2009 agreement with WARF, which settled a patent infringement action brought by WARF with respect to the '752 patent. ( Id. at ¶ 41(e)). The report states that the '752 patent “disclosed and claimed technology that improved the speed and efficiency of Intel microprocessors.” ( Id.). Intel paid $110,000,000 to settle the litigation and obtain the license to the '752 patent. ( Id.). Evans' report also describes an April 28, 2003 license agreement between Intel and WARF for three patents, for which Intel paid a lump sum of $6.2 million. ( Id. at ¶ 41(f)). The report further states that Evans did not know anything about any of the seven license agreements discussed other than their express terms and information from press releases with respect to the settlements and resulting licenses. ( Id. at ¶ 42). Evans concludes that “Intel has paid lump sum royalties of $110 to $675 million for licenses to practice the claims of microprocessor-related patents comparable to the '547 Patent.” ( Id. at ¶ 42). Evans does not explain why the $6.2 million WARF agreement is not included in determining this range.

With respect to the 2009 WARF agreement, Evans states that “the technology disclosed and claimed in [the '752 Patent] was not as important to Intel's commercial interests as the '547 Patent and that the royalty base for it was far less than the royalty base for the '547 Patent.” ( Id. at ¶ 59). Evans concludes, therefore, that “the lump sum to be paid by Intel for a paid-up license under the '547 Patent should exceed that for WARF's '752 Patent.”( Id.).

Although Evans' report discloses his opinion concerning the 2009 WARF agreement on which he now solely relies, that opinion is not supported by any methodology that explains why the 2009 WARF agreement by itself could be the basis for an accurate conclusion about the hypothetical negotiation over the '547 patent. To say that one litigation settlement agreement relating to a different patent and executed five years after the hypothetical negotiation would have taken place—even assuming the '752 patent is comparable to the '547 patent—is the basis for an opinion is completely speculative without, at a minimum, some analysis of the litigation that led to the settlement. Without analysis of the litigation, the conclusion cannot be based on “sound economic and factual predicates.” See Riles v. Shell Exploration & Production Co., 298 F.3d 1302, 1311 (Fed.Cir.2002).

Evans' report simply contains no analysis of factors that might affect the value of the settlement. For example, what was the amount of damages ultimately sought in the litigation? Would the issue of willfulness have been tried, with the possibility of treble damages? Or, as in LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed.Cir.2012), had sanctions been imposed? Id. at 58 (discussing the BenQ settlement agreement, which was entered into after two years of litigation within two weeks of the anticipated trial and after BenQ had been sanctioned repeatedly by the court for discovery misconduct and misrepresentation). Evans also conceded at the February 1 hearing that he did not focus on the provision in the WARF agreement in which WARF agreed to dispute resolution before filing litigation on any of its 2000 other patents (D.I. 268 at 80–81) against Intel or its customers and did not compare that provision to other license agreements. ( Id. at 81–83).

Whereas multiple settlement agreements might show a pattern, a single settlement agreement on a different patent without any analysis of the settlement context is not a reliable method for calculating damages. See LaserDynamics, 694 F.3d at 77–79 (noting that “[t]he propriety of using prior settlement agreements to prove the amount of a reasonable royally is questionable” and further stating that “[t]he notion that license fees are tainted by the coercive environment of patent litigation are unsuitable to prove a reasonable royalty is a logical extension of Georgia–Pacific, the premise of which assumes a voluntary agreement will be reached between a willing licensee, with validity and infringement of the patent not being disputed”).

Moreover, neither Evans nor Joseph McAlexander, AVM's technical expert, address why the 2003 license agreement for three WARF patents mentioned in paragraph 41(f) of Evans' expert report is irrelevant. An analysis that relies on a single license agreement but does not take into account why other licenses are not comparable cannot be a reliable analysis. Evans has not explained why his analysis did not consider the 2003 WARF agreement. In fact, Evans testified at the February 1 hearing that he never reviewed the three patents that were the subject of the 2003 WARF agreement and conceded that it was not one of the agreements on which he relied in his original expert report. (D.I. 268 at 78–79).

AVM asserts that Federal Circuit precedent permits damages claims to be based on a single license agreement, or a single settlement agreement for the patent-in-suit, or a single license agreement for a comparable technology. (D.I. 262 at 1) (citing Dow Chemical Co. v. Mee Industries, Inc., 341 F.3d 1370 (Fed.Cir.2003); Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed.Cir.2009); ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860 (Fed.Cir.2010) (per curiam)). AVM further asserts that “[t]here is no logical reason why a settlement agreement for a comparable technology should be treated differently.” (D.I. 262 at 1). AVM's attempt to extend the logic of the Federal Circuit precedent it cites is a “bridge too far.” AVM does not cite to a single case where any court permitted a damages claim to be based on a single settlement agreement for a comparable technology. The only case to which AVM cites in support of this proposition is a discovery decision. See Tyco Healthcare Group LP v. E–Z–EM, Inc., 2010 WL 774878, at *1 (E.D.Tex. Mar. 2, 2010). The court in Tyco addressed whether settlement communications that resulted in a license agreement were discoverable. Id. The court reasoned that “ResQNet suggests that the underlying negotiations are relevant to the calculation of a reasonable royalty using the hypothetical negotiation damages model.” Id. at *2 (emphasis added). The court did not address specifically whether damages claims could be based solely on a single settlement agreement involving comparable technology. And, in any event, what is missing here is any analysis of the underlying negotiations between WARF and Intel or the context in which the settlement agreement was reached. Even assuming that a single settlement agreement on a comparable technology could be the basis for a reliable conclusion, which the Court doubts, Evans' analysis falls far short of what would be necessary for such a conclusion.

The parties' dueling February 4 and February 5 letters are further evidence of the type of analysis required before the 2009 WARF agreement could conceivably have been the basis for a reliable conclusion. ( See D.I. 264, 266, 267). For example, in its February 4 letter, Intel states that the WARF court “rejected Intel's invalidity defense just before trial.” (D.I. 264 at 3). AVM disputes this characterization and argues that the court “merely denied Intel's motion ‘on one of its theories of invalidity.’ ” (D.I.266). Intel responds in its February 5 letter that the court “rejected Intel's anticipation defense as inconsistent with the court's claim construction.” (D.I.267). At an absolute minimum, some analysis of the defenses available to Intel at trial or the strength of those defenses would need to have been done before relying solely on the 2009 WARF agreement.

Thus, for the reasons discussed, the Court concludes that Evans' testimony must be excluded entirely.

B. Tran's Testimony

AVM proffered five areas in which it proposed to have Tran testify. (D.I.257). They are: (1) about the benefits of the patent in suit (to which Intel does not object); (2) the legal assignments of the patent in suit (to which Intel does not object); (3) Tran's negotiations with Intel in 2006; (4) Tran's conduct in a hypothetical negotiation; and (5) Tran's knowledge of 1999 negotiations with Intel over other patents (to which Intel does not appear to object, but which is of little apparent relevance). Intel objects to most of the third area and all of the fourth area.

AVM asserts that Tran's testimony is not expert and that, as the co-inventor of the '547 patent and former CEO of AVM's predecessor, he is “uniquely qualified” to offer testimony concerning the Georgia–Pacific factors. (D.I. 255 at 3). Intel objects to Tran testifying to damages on three grounds: (1) Tran's proposed testimony would be improper expert opinion testimony; (2) even if Tran's testimony is not improper expert opinion, it was disclosedtoo late; and (3) even if Tran is permitted to testify, his testimony is not sufficient to establish AVM's damages. (D.I. 258 at 1–3).

The dispute over Tran's testimony arises against a backdrop in which damages are presumed if infringement is proven. See35 U.S.C. § 284. Reasonable royalty damages must be awarded if infringement is proven. See Dow Chemical Co. v. Mee Industries, Inc., 341 F.3d 1370, 1381–82 (Fed.Cir.2003) (holding that district court erred in concluding that “no damages could be awarded, in light of the presumption of damages when infringement is proven”); Lindemann Maschinenfabrik GmbH v. Am. Hoist & Derrick Co., 895 F.2d 1403, 1406–08 (Fed.Cir.1990); cf. Devex Corp. v. General Motors Corp., 667 F.2d 347, 363 (3d Cir.1981) (“The statute requires the award of a reasonable royalty, but to argue that this requirement exists even in the absence of any evidence from which a court may derive a reasonable royalty goes beyond the possible meaning of the statute.”). If there were a trial, and Intel's products were found to be infringing, there is admissible evidence from which a reasonable royalty could be awarded, although it would likely be a very small fraction of what AVM is seeking through the testimony that is the subject of these motions. See infra at footnote 13 (describing admissible evidence that might bear upon a calculation of reasonable royalty damages). That reasonable royalty damages must be awarded if infringement is found, however, does not mean that the rules of evidence do not apply to proposed testimony.

Much of Tran's proffered testimony would be improper expert opinion, including, for example, an analysis of the estimated costs that “Intel necessarily would have incurred to avoid the '547 patent” (D.I. 257 at 8); the calculation of a “per design technology use” fee that Intel would pay for each different product design using the '547 patent ( id. at 3–4); a multiplication factor to apply to the per design technology use fee ( id. at 4); and a final damages number based on the application of the adjusted and discounted per design technology use fee to the number of product designs ( id. at 12). These calculations are the province of expert analysis. See, e.g., Veritas Operating Corp. v. Microsoft Corp., 2008 WL 657936, at *33 (W.D.Wash. Jan. 17, 2007) (lay witness may not offer an opinion on ultimate patent damages, “including determining a reasonable royalty”).

Tran's proposed testimony regarding events that “would have” occurred is also inadmissible. Although a lay witness may testify to facts within his personal knowledge, a lay witness may not offer testimony as to events that “would have occurred.” See Donlin v. Philips Lighting N. Am. Corp., 581 F.3d 73, 82–84 (3d Cir.2009) (vacating district court judgment where damages testimony from lay witness “went beyond those easily verifiable facts with their personal knowledge and instead required forward-looking speculation for which she lacked necessary training”). Tran's testimony as to what would have happened in a hypothetical negotiation would not be based on his personal knowledge and, therefore, is not admissible. See Wilburn v. Maritrans GP, Inc., 139 F.3d 350, 356 (3d Cir.1998) (expert, not lay, witness may address hypothetical questions).

Although much of Tran's proffered testimony is improper expert opinion or improper speculative or hypothetical testimony, his proffered testimony as to facts within his personal knowledge is admissible. Tran would be permitted to testify to the benefits of the '547 patent (D.I. 257 at § I), the legal assignments of the '547 patent ( id. at § II), his actual negotiations with Intel in 2006 ( id. at § III through second full paragraph on page 3), and his 1999 negotiations. ( Id. at § V). This testimony is based on fact and within Tran's personal knowledge.

The proposed expert and hypothetical testimony is also disclosed far too late, on the eve of trial. AVM never identified Tran as having knowledge of damages. During Tran's deposition, Intel questioned Tran about damages, including asking him the amount that he would have accepted for a license to the '547 patent. He testified that he had “no idea[ ]” what he would have accepted for a license. (D.I. 259 at Ex. 1 at 174, 175, 186). Tran further testified that he did not have any number in mind as to what AVM should seek in damages from Intel. ( Id. at Ex. 1 at 68–69). Intel would be prejudiced by AVM's late disclosure of Tran's damages analysis. Intel was denied any opportunity to seek discovery into Tran's damages theory or develop its own response and, therefore, would be prejudiced by Tran's proposed expert and hypothetical testimony. See e.g., ePlus, Inc. v. Lawson Software, Inc., 700 F.3d 509, 523 (Fed.Cir.2012) (holding that the district court's concerns that the defendant would be prejudiced by a changing of the damage calculation on the eve of trial provided the district court “with sufficient basis to preclude ePlus from presenting any evidence of damages at trial”); MicroStrategy, Inc. v. Business Objects, S.A., 429 F.3d 1344, 1357 (Fed.Cir.2005) (exclusion of non-expert damages opinion is proper where “new theories surprised [Defendant] on the eve of trial and prejudiced any response”).

Finally, Tran's expert and hypothetical testimony, even if timely disclosed, is insufficient to meet AVM's burden of proving the damages Tran asserts. Before presenting a damages number to the jury, the patentee must establish that its calculation is based on reliable methodology and concrete facts. See, e.g., Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1315 (Fed.Cir.2011) (citing Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 591, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)). Tran's proffered analysis is to some extent a rehash of the proposed Evans' testimony, which the Court has concluded today and in its earlier opinion must be excluded. To the extent it is not a rehash, Tran's qualification to offer licensing testimony of the sort that is proffered is doubtful, and the methodology has not been shown to be reliable. As a matter of law, AVM cannot introduce a damages theory to the jury without establishing its reliability. See Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1324, 1335 (Fed.Cir.2009) (vacating damages award where the “jury's damages award [was] not supported by substantial evidence, but is based mainly on speculation or guesswork”).

AVM relies on CPG Products Corp. v. Pegasus Luggage, Inc., 776 F.2d 1007 (Fed.Cir.1985), for the proposition that Tran's testimony is legally sufficient to support AVM's damages claim. This case is essentially irrelevant to the issue at hand. Although the Federal Circuit did uphold the trial court's award of damages, the Court was focused on the issue of whether the defendant could have offered testimony challenging the reasonable royalty rate. Id. at 1010–11. The defendant did not present such testimony and the Federal Circuit held that the defendant could not be heard to complain after the fact. Id. The trial court merely found that a fact witness could testify concerning the fact that there was a 20% difference in the price of the plaintiff's product and the competing product sold by the defendant to support a claim for a 20% royalty. See CPG Products Corp. v. Pegasus Luggage, Inc., 1983 WL 248, at *12, *20 (S.D.Fla.1983) (district court opinion). CPG is consistent with the later-decided Lindemann Maschinenfabrik and Dow Chemical cases. CPG did not conclude that a fact witness may perform a damages calculation or testify as to what would have occurred in a hypothetical negotiation.

Thus, for the reasons discussed, the Court concludes that, although Tran would be permitted to testify to facts within his personal knowledge, the Tran expert and hypothetical testimony proffered in D.I. 257 at 3–13 should be excluded because it is untimely disclosed expert and otherwise inadmissible testimony.

Because Evans' testimony and much of Tran's proffered testimony concerning damages have been excluded, and because the Court understood that AVM did not want to proceed to trial without at least one of the evidentiary bases for a theory of damages in excess of $100,000,000, the Court vacated the trial date (February 11, 2013). In view of Federal Circuit precedent, however, the Court cannot at this time grant Intel's motion for summary judgment of no damages. The Court does grant the motion to exclude Evans' testimony and the motion in limine with regard to Tran's preferred expert and hypothetical testimony.

An appropriate order will follow.

ORDER

For the reasons set forth in the accompanying Memorandum Opinion, it is hereby ORDERED that:

1. Defendant Intel Corporation's Daubert Motion to Exclude the Testimony of Larry Evans (D.I.191) IS GRANTED.

2. Defendant Intel Corporation's Motion in Limine # 2 (D.I.240, Ex. 15) IS GRANTED IN PART with respect to the portion to seeking to preclude Joseph Tran from offering expert opinion.

3. The February 7, 2013 Oral Order granting Intel Corporation's Motion for Summary Judgment of No Damages IS VACATED and will be held in abeyance pending a further conference with counsel.