Oracle Corporation et al v. SAP AG et al

MOTION for Judgment as a Matter of Law and, MOTION for New Trial

N.D. Cal.February 23, 2011

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Robert A. Mittelstaedt (SBN 060359) Jason McDonell (SBN 115084) Elaine Wallace (SBN 197882) JONES DAY 555 California Street, 26th Floor San Francisco, CA 94104 Telephone: (415) 626-3939 Facsimile: (415) 875-5700 ramittelstaedt@jonesday.com jmcdonell@jonesday.com ewallace@jonesday.com Tharan Gregory Lanier (SBN 138784) Jane L. Froyd (SBN 220776) JONES DAY 1755 Embarcadero Road Palo Alto, CA 94303 Telephone: (650) 739-3939 Facsimile: (650) 739-3900 tglanier@jonesday.com jfroyd@jonesday.com Scott W. Cowan (Admitted Pro Hac Vice) Joshua L. Fuchs (Admitted Pro Hac Vice) JONES DAY 717 Texas, Suite 3300 Houston, TX 77002 Telephone: (832) 239-3939 Facsimile: (832) 239-3600 swcowan@jonesday.com jlfuchs@jonesday.com Attorneys for Defendants SAP AG, SAP AMERICA, INC., and TOMORROWNOW, INC. UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA OAKLAND DIVISION ORACLE USA, INC., et al., Plaintiffs, v. SAP AG, et al., Defendants. Case No. 07-CV-1658 PJH (EDL) DEFENDANTS’ RENEWED MOTION FOR JUDGMENT AS A MATTER OF LAW AND NEW TRIAL MOTION Date: July 13, 2011 Time: 9:00 a.m. Courtroom: 3, Third Floor Judge: Hon. Phyllis J. Hamilton Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page1 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TABLE OF CONTENTS Page(s) SVI-90062v1 - i - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) I. INTRODUCTION .............................................................................................................. 2 II. SUMMARY OF ARGUMENT .......................................................................................... 2 III. SUMMARY OF RECORD EVIDENCE ............................................................................ 5 A. Oracle’s Actual Harm from Infringement Was Minimal ........................................ 5 B. Oracle Does Not License the Copyrighted Works to Provide Third Party Support and Presented No Objective Evidence on the Value of a License ............ 5 C. Oracle Invented Factors to Price the PeopleSoft/JDE and Siebel Licenses ............ 6 1. The Amount Oracle Claims It Would Have Charged for a License ........... 6 2. The Value of the Infringed Intellectual Property as a Whole ..................... 7 3. Liability Evidence for Stipulated Claims .................................................... 8 4. Meyer’s Factors for Determining the Result of a “Hypothetical Negotiation” to Use the PeopleSoft, JDE, and Siebel Works ..................... 8 (a) Meyer Relied on Little More than a Handful of SAP Documents and Testimony to Conduct His Factor-Based Analysis ........................................................................................... 9 (b) Meyer’s Calculations Rested on Baseless Assumptions ............... 11 5. Oracle’s Counsel Invited the Jury to Speculate ........................................ 11 D. Oracle Relied Solely on the Self-Interested Testimony of a Single Oracle Executive to Estimate the Database Software Hypothetical License .................... 12 IV. LEGAL STANDARD ....................................................................................................... 13 V. ARGUMENT: RENEWED MOTION FOR JMOL ......................................................... 14 A. Oracle Is Not Entitled to Hypothetical License Fees as Actual Damages because the Undisputed Evidence Showed that It Did Not Lose License Fees ....................................................................................................................... 14 1. “Actual Damages” under the Copyright Act ............................................. 15 2. The Standards for Permitting a “Hypothetical License” Fee as Actual Damages ........................................................................................ 16 3. Oracle Did Not Lose a License Fee from Any Third Party ...................... 18 4. Oracle Did Not Lose a License Fee from Defendants .............................. 18 (a) Oracle Witnesses Unequivocally Testified that Oracle Would Not Have Granted a License to Defendants ...................... 19 (b) Direct Competition between the Parties and the Lack of Any Benchmarks Further Confirm that a Hypothetical License Fee Is Inapplicable Here ............................................................... 20 B. Oracle Failed to Offer Legally Sufficient Evidence to Value the Hypothetical License ............................................................................................. 21 1. Hypothetical License Awards May Not Be Based on Undue Speculation ................................................................................................ 22 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page2 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TABLE OF CONTENTS (continued) Page SVI-90062v1 - ii - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) 2. Oracle Failed to Present Objective Evidence of Benchmark Licenses ..................................................................................................... 23 3. Oracle’s Hypothetical License Claim for the PeopleSoft, JDE, and Siebel Works Is Unreasonable and Unduly Speculative ........................... 24 (a) Factors Emphasized by Oracle’s Counsel Are Insufficient to Support a Non-Speculative License .............................................. 24 (b) Meyer’s “Background” Factors Are Irrelevant to Pricing a Hypothetical License ..................................................................... 27 (c) Meyer’s Primary Factors Were Insufficient to Support a Non-Speculative License Calculation ........................................... 29 (i) Evidence of Goals and Expectations Cannot Independently Establish Market Price .............................. 29 (ii) Meyer’s Evidence Does Not Provide a Legally Sufficient Basis to Calculate a Hypothetical License Price .................................................................................. 30 (iii) Meyer’s License Calculations Rested on Unjustified Assumptions ...................................................................... 33 (iv) The Wide Range of License Values Generated by Meyer’s Analysis Forced the Jury to Speculate ................ 34 (d) Oracle’s Counsel Improperly Encouraged Speculation ................ 35 4. Oracle’s Database Damages Claim Suffers From Similar Deficiencies ............................................................................................... 36 (a) Oracle Presented No Objective Evidence of the Database License Price or Structure ............................................................. 36 (b) Oracle Presented Only Subjective Testimony of What It Claims It Would Have Charged for a Database License............... 37 (c) Oracle’s Database License Calculation Similarly Was Based on Unjustified Assumptions .......................................................... 38 VI. ARGUMENT: NEW TRIAL MOTION ........................................................................... 39 A. Oracle’s Improper Hypothetical License Theory Was Based upon Undue Speculation ............................................................................................................ 39 B. The Award Far Exceeds Oracle’s Actual Harm .................................................... 43 C. The Award Was Infected by Oracle’s Presentation of Irrelevant and Prejudicial Liability Evidence ............................................................................... 45 D. The Court Should Remit Damages to $28 Million or No More Than $408.7 Million ................................................................................................................... 47 VII. CONCLUSION ................................................................................................................. 48 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page3 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TABLE OF AUTHORITIES Page(s) SVI-90062v1 - iii - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Cases Air-Sea Forwarders, Inc. v. Air Asia Co., Ltd., 880 F.2d 176 (9th Cir. 1989)...................................................................................................... 13 Alvarado v. Federal Express Corp., No. C 04-0098 SI, 2008 U.S. Dist. LEXIS 21238 (N.D. Cal. Mar. 18, 2008) .......................... 43 Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (1986) ................................................................................................................... 13 Anglo-American Gen. Agents v. Jackson Nat’l Life Ins. Co., 83 F.R.D. 41 (N.D. Cal. 1979) ............................................................................................. 14, 47 Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435 (9th Cir. 1994)...................................................................................................... 25 Baker v. Urban Outfitters, Inc., 254 F. Supp. 2d 346 (S.D.N.Y. 2003) .................................................................................. 15, 19 Barrera v. Brooklyn Music, Ltd., 346 F. Supp. 2d 400 (S.D.N.Y. 2004) ........................................................................................ 28 Bi-Rite v. Button Master, 578 F. Supp. 59 (S.D.N.Y. 1983) ............................................................................................... 23 Bruce v. Weekly World News, Inc., 310 F.3d 25 (1st Cir. 2002) ........................................................................................................ 37 Business Trends Analysts, Inc. v. Freedonia Group, Inc., 887 F.2d 399 (2d Cir. 1989) ................................................................................................ passim Childress v. Taylor, 798 F. Supp. 983 (S.D.N.Y. 1992) ................................................................................. 33, 38, 40 Country Road Music, Inc. v. MP3.com, Inc., 279 F. Supp. 2d 325 (S.D.N.Y. 2003) .................................................................................. 22, 26 Cream Records, Inc. v. Jos. Schlitz Brewing Co., 754 F.2d 826 (9th Cir. 1985).......................................................................................... 15, 16, 18 D & S Redi-Mix v. Sierra Redi-Mix & Contracting Co., 692 F.2d 1245 (9th Cir. 1982).................................................................................................... 47 DaimlerChrysler Servs. v. Summit Nat’l, No. 02-71871, 2006 WL 208787 (E.D. Mich. Jan. 26, 2006) ................................................... 27 Drew v. Equifax Info. Servs., LLC, No. C 07-00726 SI, 2010 WL 5022466 (N.D. Cal. Dec. 3, 2010)............................................. 14 Encyclopedia Brown Prods., Ltd. v. Home Box Office, Inc., 25 F. Supp. 2d 395 (S.D.N.Y. 1998) .............................................................................. 16, 19, 21 Faulkner v. Nat’l Geographic Soc’y, 576 F. Supp. 2d 609 (S.D.N.Y. 2008) ........................................................................................ 28 Frank Music Corp. v. Metro-Goldwyn-Mayer, Inc., 772 F.2d 505 (9th Cir. 1985)............................................................................................... passim Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp 1116 (S.D.N.Y. 1970) ............................................................................................ 41 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page4 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TABLE OF AUTHORITIES (continued) Page(s) SVI-90062v1 - iv - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Herrington v. County of Sonoma, 883 F.2d 739 (9th Cir. 1989)...................................................................................................... 14 Honda Motor Co. v. Oberg, 512 U.S. 415 (1994) ................................................................................................................... 48 In re First Alliance Mortg. Co., 471 F.3d 977 (9th Cir. 2006)...................................................................................................... 13 Informatica Corp. v. Business Objects Data Integration, Inc., No. C 02-03378 EDL, 2007 WL 2344962 (N.D. Cal. Aug. 16, 2007) ...................................... 43 Interplan Architects, Inc. v. C.I. Thomas, Inc., No. 4:08-cv-03181, 2010 U.S. Dist. LEXIS 114306 (S.D. Tex. Oct. 27, 2010) ........... 23, 24, 37 Jamison Bus. Sys., Inc. v. Unique Software Support Corp., No. CV 02-4887 (ETB), 2005 U.S. Dist. LEXIS 45480 (E.D.N.Y. May 26, 2005) ................. 35 Jarvis v. K2, Inc., 486 F.3d 526 (9th Cir. 2007)............................................................................................... passim Lakeside-Scott v. Multnomah County, 556 F.3d 797 (9th Cir. 2009)...................................................................................................... 13 Leland Med. Ctrs., Inc. v. Weiss, No. 4:07cv67, 2007 WL 2900599 (E.D. Tex. Sept. 28, 2007) ........................................... passim Locklin v. Switzer Bros., Inc., 235 F. Supp. 904 (N.D. Cal. 1964) ............................................................................................ 41 Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed. Cir. 2009) .................................................................................................. 41 Mackie v. Rieser, 296 F.3d 909 (9th Cir. 2002)............................................................................................... passim Miller v. Glenn Miller Prods., Inc., 454 F.3d 975 (9th Cir. 2006)...................................................................................................... 13 Molski v. M.J. Cable, Inc., 481 F.3d 724 (9th Cir. 2007)................................................................................................ 13, 39 National Conference of Bar Examiners v. Multistate Legal Studies, Inc., 458 F. Supp. 2d 252 (E.D. Pa. 2006) ............................................................................. 17, 18, 20 Oboler v. Goldin, 714 F.2d 211 (2d Cir. 1983) ........................................................................................................ 46 On Davis v. The Gap, Inc., 246 F.3d 152 (2d Cir. 2001) ................................................................................................ passim Polar Bear Prods., Inc. v. Timex Corp., 384 F.3d 700 (9th Cir. 2004)............................................................................................... passim Powell v. Penhollow, 260 Fed. App’x 683 (5th Cir. 2007) ........................................................................................... 26 Propet USA, Inc. v. Shugart, No. C06-0186-MAT, 2007 U.S. Dist. LEXIS 69222 (W.D. Wash. Sept. 19, 2007) ................. 26 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page5 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TABLE OF AUTHORITIES (continued) Page(s) SVI-90062v1 - v - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) R.S.E., Inc. v. Pennsy Supply, Inc., 523 F. Supp. 954 (D.C. Pa. 1981) .............................................................................................. 35 ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860 (Fed. Cir. 2010) .................................................................................................... 41 Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390 (1940) ................................................................................................................... 25 Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157 (9th Cir. 1977).............................................................................................. 18, 22 Sinclair Ref. Co. v. Jenkins Petroleum Process Co., 289 U.S. 689 (1933) ................................................................................................................... 41 Smith v. Rush, No. C04-2280Z, 2006 U.S. Dist. LEXIS 27412 (W.D. Wash. Apr. 7, 2006)................ 23, 24, 37 Stehrenberger v. R.J. Reynolds Tobacco Holdings, Inc., 335 F. Supp. 2d 466 (S.D.N.Y. 2004) .................................................................................. 28, 46 Technologies, S.A. v. Cyrano, Inc., 460 F. Supp. 2d 197 (D. Mass. 2006) .................................................................................. 23, 34 Trans-World Mfg. Co. v. Al Nyman & Sons, Inc., 750 F.2d 1552 (Fed. Cir. 1984) .................................................................................................. 41 Uniloc USA, Inc. v. Microsoft Corp., Nos. 2010-1035, 2010-1055, 2011 U.S. App. LEXIS 11 (Fed. Cir. Jan. 4, 2011) .............. 41, 42 Wall Data, Inc. v. Los Angeles County Sheriff’s Dept., 447 F.3d 769 (9th Cir. 2006)............................................................................................... passim Wordtech Sys., Inc. v. Integrated Network Solutions, Inc., 609 F.3d 1308 (Fed. Cir. 2010) .................................................................................................. 41 Statutes 17 U.S.C. § 102 .............................................................................................................................. 25 17 U.S.C. § 504 ....................................................................................................................... passim Rules Fed. R. Civ. P. 50 .................................................................................................................... passim Fed. R. Civ. P. 59 ............................................................................................................. 1, 4, 13, 39 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page6 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 1 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) NOTICE OF MOTIONS PLEASE TAKE NOTICE THAT on July 13, 2011, at 9:00 a.m., in the above-titled Court, located at 1301 Clay Street, Oakland, California, in Courtroom 3 before the Honorable Phyllis J. Hamilton, Defendants SAP AG, SAP America, Inc. (together, “SAP”) and TomorrowNow, Inc. (“TN,” and with SAP, “Defendants”) will bring this renewed motion for judgment as a matter of law and new trial motion pursuant to 17 U.S.C. § 504, Rules 50(b) and 59 of the Federal Rules of Civil Procedure (“Rule 50(b)” and “Rule 59”), and Civil Local Rules 7-2, 7-4, and 7-5, against Plaintiff Oracle International Corporation (“Oracle”). The motions are based on the Memorandum of Points and Authorities herein and the Declaration of Tharan Gregory Lanier (“Lanier Decl.”) and exhibits attached thereto. RELIEF REQUESTED Defendants respectfully request that the Court: (1) grant the renewed motion for judgment as a matter of law that Oracle is not entitled to actual damages for copyright infringement in the form of a hypothetical license because Oracle did not establish that, but for infringement, it would have licensed the asserted copyrighted works for the use at issue and, alternatively, because Oracle did not present a sufficient evidentiary basis for its hypothetical license claim, resulting in an award based on undue speculation; and (2) grant the new trial motion because the jury’s verdict is grossly excessive, against the weight of the evidence, and resulted in a miscarriage of justice. Under both motions, Defendants request that the Court remit the verdict to $28 million (or no more than $408.7 million) or order a new trial to determine damages based on lost profits and infringer’s profits. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page7 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 21 22 23 24 25 26 27 28 20 SVI-90062v1 - 2 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) I. INTRODUCTION “[A]warding the copyright owner the lost license fee can risk abuse. Once the defendant has infringed, the owner may claim unreasonable amounts as the license fee . . . .” On Davis v. The Gap, Inc., 246 F.3d 152, 166 (2d Cir. 2001). “Making the plaintiff whole is plainly different from punishing the infringer by charging the highest possible rate for the infringement.” Jarvis v. K2, Inc., 486 F.3d 526, 535 (9th Cir. 2007). The jury’s $1.3 billion “hypothetical license” fee award was an unreasonable amount that went far beyond what was needed to make Oracle whole. The enormous fee did not—indeed, could not—represent Oracle’s actual damages because Oracle never lost a license fee from Defendants or anyone else. The Copyright Act does not permit such awards that are un-tethered to a plaintiff’s actual harm, and the hypothetical license claim should not have gone to the jury. Even were the hypothetical license approach permissible in this case, the jury’s $1.3 billion damages award improperly was based on undue speculation. Oracle took “billions” of dollars in aspirations, corporate acquisitions, research and development costs, total intellectual property value, and its own self-interested, post hoc demands, mixed those huge numbers into a pot, and told the jury that its “actual damages” are “probably somewhere in there.” Where even Oracle’s expert estimated the harm at $408.7 million, only unwarranted speculation, combined with inflammatory and inappropriate liability evidence, could generate such an excessive award. The jury’s $1.3 billion verdict is thus in conflict with copyright law and founded on sheer speculation. It cannot stand. II. SUMMARY OF ARGUMENT Copyright law permits an award of actual damages only to compensate for actual harm to the market value of a copyrighted work caused by infringement. In contrast to the “reasonable royalty” of patent law—a statutorily mandated floor below which a patent damages award may not fall—there is no separate statutory basis for recovery of a “hypothetical license” fee under the Copyright Act. A hypothetical license is available only if a plaintiff proves that, but for the infringement, the parties would have agreed to such a license or that it would have licensed a third party for the same use and was prevented from doing so by the infringement. The Ninth Circuit has never approved a challenged award of hypothetical license fees absent such proof. Permitting Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page8 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 3 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Instead, unhindered by real-world transactions—because there were none—Oracle relied at trial on speculation about counterfactual license agreements and negotiations that never could or would have existed and used that speculation to inflate its damages claim. Asked to imagine a hypothetical negotiation that they conceded could never have taken place, Oracle executives self- servingly testified that they would have demanded a license fee with “a ‘B’ on it in billions.” Oracle also offered evidence placing the full value of all of its intellectual property in various billion-dollar amounts, rather than focusing, as the law requires, on the value of Defendants’ infringing use of the copyrighted works. Not even Oracle’s damages expert Paul Meyer could offer a particular value of the alleged lost license, instead “opining” that the hypothetical license recovery of hypothetical license fees when none were lost is the antithesis of awarding “actual” damages and necessarily leads to reliance on speculative and counterfactual evidence regarding the amount to which imaginary parties might have agreed for a purely fictional license. In denying Defendants’ first motion for partial summary judgment, the Court cautioned that Oracle could claim hypothetical license fees only if it “present[ed] evidence sufficient to allow the jury to assess fair market value without ‘undue speculation.’” ECF No. 628 (Order) at 4. But Oracle did not present non-speculative evidence of the fact of a lost license fee or the valuation of such a hypothetical lost license. Rather, the trial evidence conclusively demonstrated that Oracle never would have obtained license fees absent Defendants’ infringement —not under any condition, not in any amount, and not from any party. Oracle and SAP are fierce competitors in applications software, the field in which TN provided maintenance service. Oracle executives uniformly testified that Oracle never would have granted SAP a license because Oracle was not interested in handing over its share of the applications market to its biggest competitor. In fact, Oracle testified that it would have been “terrified” to do so. SAP executives likewise confirmed that SAP never would have entered into such a license with Oracle. And Oracle did not contend that it was deprived of the opportunity to license the works to a third party. Indeed, Oracle offered no evidence of benchmark licenses or transactions, or any prior licensing history of the subject works, to prove that a license between the parties, or between Oracle and a third party, ever existed or was possible. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page9 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 4 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) The jury was asked to determine Oracle’s “actual damages suffered as a result of the infringement.” See ECF No. 1004 (Special Verdict Form); ECF No. 1005 (Jury Instructions) at 7. There is no legal theory under which a verdict so wildly out of proportion to Oracle’s actual damages can be allowed to stand. Accordingly, pursuant to Rule 50(b), the Court should, grant judgment as a matter of law that Oracle cannot recover damages in the form of hypothetically lost license fees, and remit the damages award to the maximum actual damages supported by the law and evidence—the sum of Oracle’s properly proven lost profits and Defendants’ disgorged profits due to infringement. At the very least, the Court can and should exercise its discretion to award a new trial pursuant to Rule 59 because the $1.3 billion award is excessive, speculative, and unreasonable, and goes far beyond what is needed to “[m]ak[e] the plaintiff whole.” damages were “at least” $1.66 billion. In closing, Oracle’s counsel asked the jury to pick a number in the $1.66 to $3 billion range, stating that “the answer is probably somewhere in there.” Thus, the jury was invited—indeed, forced—to speculate to reach its verdict. Allowing the hypothetical license damages theory to go to the jury on these facts necessarily resulted in this improper speculation. The evidence did establish, however, that Oracle had a non-speculative remedy—lost profits plus non-duplicative infringer’s profits. In fact, both sides’ experts calculated Oracle’s lost profits and infringer’s profits based on the customers each determined would not have left Oracle to purchase TN support or SAP products but for the infringement. The rigorous customer analysis performed by Defendants’ damages expert, Stephen Clarke, yielded $28 million in combined lost profits and infringer’s profits, while Oracle’s expert, Meyer, offered $408.7 million as his calculation. The $1.3 billion award is thus grossly excessive on its face, far surpassing Oracle’s actual harm as determined by both experts. The award is also against the clear weight of the evidence, which compelled a much lower amount, even under Oracle’s damages theory. Ultimately, the verdict flowed not from objective evidence of Oracle’s actual limited loss, but from subjective and speculative evidence that purportedly related to pricing a hypothetical license—a problem compounded by Oracle’s presentation of irrelevant liability evidence, the sole function of which was to inflame the jury and extract impermissible punitive damages. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page10 of 54 1 2 3 13 14 15 28 4 5 6 7 8 9 10 11 12 16 17 18 19 20 21 22 23 24 25 26 27 SVI-90062v1 - 5 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) III. SUMMARY OF RECORD EVIDENCE A. Oracle’s Actual Harm from Infringement Was Minimal. Throughout its existence, TN had but 358 customers, only 86 of which ever purchased anything from SAP while a TN customer. Lanier Decl. ¶¶ 1, 3, 5, 9, Exs. 1, 36.1 Many of these customers would have left Oracle even absent TN’s infringement, as confirmed by undisputed evidence that TN played no part in their decisions to purchase SAP products. Lanier Decl. ¶¶ 6- 8, 20-21, Exs. 1, 6-8. Consistent with these actual results and based on a customer-by-customer analysis, Clarke calculated lost profits of $19.3 million and infringer’s profits of $8.7 million ($28 million total). Lanier Decl. ¶¶ 2, 22, Exs. 1, 15. Oracle presented no evidence from even a single customer stating that it purchased SAP products because of TN. Nonetheless, based on a customer analysis purportedly similar to Clarke’s, Meyer calculated $120.7 million in lost profits and $288 million in infringer’s profits ($408.7 million total). Lanier Decl. ¶¶ 2, 11, Exs. 1, 15. B. Oracle Does Not License the Copyrighted Works to Provide Third Party Support and Presented No Objective Evidence on the Value of a License. Although there was competent evidence of damages from claimed actual lost customers, Oracle did not ask the jury to award damages on that basis. Instead, Oracle claimed it suffered “actual damages” in the form of hypothetical license fees for the PeopleSoft, J.D. Edwards (“JDE”), Siebel, and Oracle database software. Lanier Decl. ¶ 32, Ex. 1. Oracle did not, however, offer any objective evidence to support a hypothetical license price, such as benchmark licenses for comparable use of comparable works. This is because, as Oracle co-President Safra Catz testified, there are no “comparable” licenses to the hypothetical license at issue here. Lanier Decl. ¶ 28, Ex. 1; see also D.I. 256 (Decl. of Colleen A. Kelly) ¶¶ 3-4. Oracle’s damages expert, Paul Meyer, similarly testified that Oracle has never “in the history of Oracle” granted a license for a competitor to use its database software to compete for maintenance customers, that such a license would be “unique,” and that there are no comparable licenses. Lanier Decl. ¶ 33, Ex. 1. SAP’s damages expert, Stephen Clarke testified the “ideal thing we are looking for” is an established benchmark existing in the marketplace, but that he did not find any. Lanier Decl. ¶ 34, Ex. 1. 1 Defendants present the specific record citations in the Lanier Declaration. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page11 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 25 26 23 24 27 28 SVI-90062v1 - 6 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Both sides’ witnesses confirmed that such licenses never would have existed between SAP and Oracle. Former Oracle co-President Charles Phillips testified that Oracle would have been “terrified” to provide a license to SAP for TN’s use, and that in January 2005, such a license would have been “unthinkable.” Lanier Decl. ¶ 25, Ex. 1. Oracle Senior Vice President for Global Practices Richard Allison, testifying about licensing practices, admitted that it would not “make business sense” to provide a license to a competitor “to use your own IP to take your customers.” Lanier Decl. ¶¶ 26-27, Ex. 1. Indeed, Allison testified that had TN requested a license for use of the database, “I wouldn’t have given them a license to do what they’re doing.” Id. And Catz testified that no licenses allow a company to use Oracle’s intellectual property to compete for customers; such use would be “unprecedented.” Lanier Decl. ¶ 28, Ex. 1. On Defendants’ side, SAP’s CFO, Werner Brandt, testified that SAP would not have entered into a license with Oracle when it purchased TN; SAP would have either asked TN to make further changes to its procedures or “would have considered not to buy TN.” Lanier Decl. ¶ 30, Ex. 1. This testimony comes as no surprise, given SAP and Oracle’s status as direct competitors. Oracle’s Phillips and SAP Executive Board member Shai Agassi each testified that they consider the rival company to be the “largest” and “chief” competitor, while Ellison and SAP CEO Bill McDermott testified that the companies are “fierce” and “formidable” competitors. Lanier Decl. ¶¶ 35-36, 38-39, Exs. 1, 5. And Catz described this fierce competition between the parties for customers as a “zero sum game”—every PeopleSoft customer SAP acquires is one Oracle no longer has. Lanier Decl. ¶ 37, Ex. 1. C. Oracle Invented Factors to Price the PeopleSoft/JDE and Siebel Licenses. Lacking evidence of benchmarks, and in the face of undisputed evidence that no license would have been agreed upon by the parties, Oracle posited several factors it argued supported “at least” a $1.66 billion claim for the “hypothetical” PeopleSoft/JDE and Siebel licenses. 1. The Amount Oracle Claims It Would Have Charged for a License. Despite claiming that it would not have granted its biggest competitor, SAP, a license permitting TN to use the copyrighted works to compete in providing support, see Lanier Decl. ¶¶ 25-28, Ex. 1, Oracle presented testimony from its executives that the license price it would have Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page12 of 54 1 2 3 4 5 6 7 8 9 23 24 25 26 27 28 10 11 12 13 14 15 16 17 18 19 20 21 22 SVI-90062v1 - 7 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) charged at the time of the hypothetical negotiation would have been billions of dollars. Phillips testified that he would have charged $3-4 billion for a license to use the PeopleSoft software and “billions” for a Siebel license—$4-5 billion total; Oracle’s CEO Larry Ellison testified that he would have charged $3.3 billion for the PeopleSoft software and $600 million for Siebel, totaling $4 billion; and Catz testified that any license fee would be “in the billions.” Lanier Decl. ¶¶ 42- 44, Exs. 1, 40. In closing, Oracle’s counsel encouraged the jury to consider the executives’ demands in calculating a hypothetical license award. Lanier Decl. ¶ 45, Exs. 1, 40. 2. The Value of the Infringed Intellectual Property as a Whole. Although Oracle executives acknowledged that Oracle never lost the ability to license the works at issue, Oracle argued that its hypothetical license should be based on the full value of the intellectual property that TN “took” (as distinct from the value of its use). Lanier Decl. ¶¶ 52, 56, 63, 100, Ex. 1. To set this value, Oracle offered evidence of its research and development (“R&D”) costs and repeatedly referenced the prices paid to acquire PeopleSoft (~$11 billion) and Siebel (~$6 billion). Lanier Decl. ¶¶ 48-51, 53-55, 57-60, Exs. 1, 37. To support its R&D costs measure, Oracle relied on evidence of R&D costs for all of its “intellectual property,” including for dates beyond the relevant time period and products not at issue. For example, in opening, Oracle purported to depict the increasing “Billions of Dollars” that it spends on R&D over time. Lanier Decl. ¶ 48, Exs. 1, 37. Oracle employee Edward Screven then testified that Oracle’s overall R&D budget in 2010 was $4 billion and described the costs and difficulties he encountered in developing Oracle’s E-Business Suite—a product not at issue. Lanier Decl. ¶ 49, Ex. 1. And Ellison testified that Oracle has spent $65 billion over its lifetime to develop and acquire all of its intellectual property assets. Lanier Decl. ¶ 50, Ex. 1. Further, despite acknowledging that an acquisition price encompasses many assets unrelated to intellectual property, Oracle executives testified that they would have based a hypothetical license on the billions spent to acquire PeopleSoft and Siebel. Lanier Decl. ¶¶ 51-55, 57-59, Ex. 1. Phillips testified that the PeopleSoft/JDE license price “better be a big number” because Oracle paid $11 billion for PeopleSoft; likewise, for Siebel, he testified that “the opening number for me would have a ‘B’ on it [as] in billions if I just paid 6 billion.” Lanier Decl. ¶ 54, Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page13 of 54 1 2 3 4 5 6 7 15 16 17 18 19 20 21 22 23 24 25 26 27 28 8 9 10 11 12 13 14 SVI-90062v1 - 8 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Ex. 1. Similarly, Catz testified that the license fee would be “definitely in the billions” as Oracle had paid $11 billion for PeopleSoft and “this whole business is a business about billions.” Lanier Decl. ¶ 57, Ex. 1. Meyer and Oracle’s counsel reinforced this testimony by arguing that Oracle would need billions of dollars to recoup the billion-dollar “investments” in the companies’ intellectual property and to motivate it to give up those investments. Lanier Decl. ¶¶ 62, 64, Ex. 1. 3. Liability Evidence for Stipulated Claims. Over Defendants’ objections, Oracle submitted evidence of liability for contributory copyright infringement, to which SAP had stipulated, arguing that it provided “context” and was relevant to Meyer’s hypothetical license analysis as evidence of “Risk Acceptance.” Lanier Decl. ¶¶ 75-76, Ex. 1. According to Oracle, SAP’s decision to acquire TN despite the risk of lawsuit required a more expensive license. Lanier Decl. ¶¶ 73, 76, Ex. 1. The Court allowed some evidence “that [SAP] willingly took the risk of a lawsuit” for damages and “for context,” but stated that “at some point, [liability evidence] becomes cumulative particularly given that liability’s no longer at issue in this case.” Lanier Decl. ¶ 77, Ex. 1. Yet day-after-day, Oracle presented cumulative liability evidence on contributory infringement, including evidence purporting to show that SAP knew of TN’s infringement and hid behind a so-called “liability shield.” Lanier Decl. ¶¶ 65-71, Exs. 1, 3, 19, 24, 28-29. Oracle also presented evidence on liability for the computer access, trespass, interference, and breach of contract claims, for which it no longer sought damages. Oracle employees Buffy Ransom and Edward Screven testified that TN “corrupted” Oracle’s customer support data and used fake login- names and an automated downloading tool. Oracle’s computer forensics expert, Kevin Mandia, testified on TN’s cross-use of website log-in credentials and violations of website terms of use. And Oracle offered testimony by former TN employee John Ritchie on an automated downloading tool and his opinion that TN’s activities slowed and crashed Oracle’s website (a claim for which Oracle had previously agreed not to seek any damages). Lanier Decl. ¶¶ 78-81, Exs. 1, 9. 4. Meyer’s Factors for Determining the Result of a “Hypothetical Negotiation” to Use the PeopleSoft, JDE, and Siebel Works. Oracle centered its damages case on Meyer’s testimony purporting to calculate the price Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page14 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 9 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) According to Meyer, each side hoped and expected to exploit the works for billions, thus dictating a billion-plus-dollar license price. Lanier Decl. ¶¶ 135-39, Exs. 1, 20, 38. To support his conclusion that Oracle expected billions in losses from licensing the infringed works, Meyer relied only on SAP—not Oracle—documents and testimony. Lanier Decl. ¶ 150, Exs. 1, 38. This is because, as Ellison agreed and Meyer confirmed, “there is not a single public or private document, internal or external e-mail, PowerPoint, speech, slide, scribble on a napkin” reflecting that Oracle expected to lose thousands of customers and billions in revenue due to TN. Lanier Decl. ¶¶ 146, 150, Exs. 1, 38. In fact, contemporaneous documents show that Oracle expected TN to have little to no impact—certainly, not impact in the billions—and Oracle documents post- hypothetical negotiation concede the minimal effect TN had or could have. Lanier Decl. ¶¶ 141- 45, Exs. 10-14. Indeed, Oracle executive Juan Jones described the idea that customers would switch to SAP applications based on the TN offer as the “silliest argument I’ve ever heard.” Lanier Decl. ¶ 145, Ex. 12. Oracle viewed TN as no threat at all because it understood that switching ERP software vendors is costly and that customers do not switch absent compelling business justifications. Lanier Decl. ¶¶ 145, 147-49, Exs. 1, 2, 12. As Jones testified, it takes “a of the hypothetical licenses based on the supposed result of a hypothetical negotiation between Oracle and SAP to use the PeopleSoft, JDE, and Siebel works. Meyer told the jury to consider factors he invented and particularly focused on “Goals/Business Plans Related to the Copyrighted Works” and “Expected Financial Benefits/Impacts” (with Oracle’s counsel describing Meyer’s other factors as mere “background”). Lanier Decl. ¶¶ 94, 96, 102, Exs. 1, 38. Meyer then broke these two factors into three issues that the parties purportedly would have considered in the negotiation. Lanier Decl. ¶ 97, Ex. 1. On Oracle’s side, Meyer claimed to consider: (1) “Oracle’s Goals for PeopleSoft Acquisition,” (2) “Risk to Oracle’s Investment in PeopleSoft [and Siebel],” and (3) “Oracle’s Expected Financial Impacts.” Lanier Decl. ¶ 99, Ex. 1. On SAP’s, he claimed to consider: (1) “SAP’s Goals for New Offering,” (2) “SAP’s Expected Impact on Oracle,” and (3) “SAP’s Expected Financial Gains.” Lanier Decl. ¶ 98, Ex. 1. (a) Meyer Relied on Little More than a Handful of SAP Documents and Testimony to Conduct His Factor-Based Analysis. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page15 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 10 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) lot more than just some—some savings on support” for a customer to migrate; it is a “significant undertaking.” Lanier Decl. ¶ 149, Ex. 2. But even the handful of SAP documents and testimony on which Meyer did rely did not provide an objective basis upon which to value a hypothetical license in the billions. For the PeopleSoft/JDE license, the documents and testimony show only what SAP purportedly aspired to achieve in terms of sales of its own products and services through its “Safe Passage” marketing program as a whole—not the value added to those sales as a result of the optional TN offering. Meyer relied almost entirely on one SAP document—“A Roadmap for PSFT Customers to SAP,” also known as the “Ziemen document”—to calculate both sides’ “expected” financial gains or losses from the PeopleSoft/JDE license. Lanier Decl. ¶¶ 107-08, 133, Exs. 1, 20. Meyer testified that it proved that SAP expected to obtain 3,000 new software customers from PeopleSoft as a result of the TN support offering. Lanier Decl. ¶ 108, Ex. 1. In reality, the document relates not to TN’s projected impact, but to the “Safe Passage initiative,” a marketing program in which the sale of SAP products and services were the main objective and technical support for Oracle products was only an optional component. Lanier Decl. ¶¶ 107, 109, Exs. 1, 20. The document does not even specifically concern TN, but contemplates that support might be provided by TN “or other vendors” that SAP was investigating when the document was drafted. Lanier Decl. ¶ 107, Ex. 20 at SAP-OR00253280. Further, Meyer admitted that the customer numbers in this document were merely “assumptions.” Lanier Decl. ¶ 110, Ex. 1. For similar reasons, the other documents to which Meyer referred the jury did not establish the value, if any, that TN’s use of the copyrighted works lent to Safe Passage.2 For the Siebel hypothetical license, Meyer again relied solely on documents containing only speculative aspirations.3 But like the Ziemen document, these documents provided no 2 The other evidence on which Meyer relied includes: three versions of John Zepecki’s “PeopleSoft 1-2-3” memo, two versions of “Safe Passage: Winning Customers and Markets from Oracle-PeopleSoft-J.D. Edwards,” a “TomorrowNow Integration Meeting” document, Executive Board minutes, an analyst-phone-call transcript in which SAP announced the TN acquisition and Safe Passage launch, and excerpts of Shai Agassi testimony. Lanier Decl. ¶¶ 105-06, 112-15, 117-19, 121-22, Exs. 1, 16-19, 21-23, 38. 3 These documents included the October 17, 2005 Business Case, the “Siebel Safe Passage Program Playbook,” and an email from TN’s CEO Andrew Nelson. Lanier Decl. ¶¶ 124-25, 127- 31, Exs. 1, 25, 33-34, 38. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page16 of 54 1 2 3 16 17 18 19 20 21 22 23 24 25 26 4 5 6 7 8 9 10 11 12 13 14 15 27 28 SVI-90062v1 - 11 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) insight into the number of customers SAP reasonably could expect to gain by virtue of TN. (b) Meyer’s Calculations Rested on Baseless Assumptions. Ultimately, Meyer transformed these documents into “expected” converted customers and financial impacts. Lanier Decl. ¶¶ 134-35, Exs. 1, 20, 38. For the PeopleSoft/JDE license, Meyer concluded that SAP and Oracle expected 1,375 to 5,592 customers to leave PeopleSoft for SAP and settled on a wide range of 1,375 to 3,000 for his calculations. Lanier Decl. ¶ 134, Exs. 1, 38. To determine the parties’ “expected” financial impacts, he multiplied this range by a revenue-per- customer figure he estimated from the revenue figures for “Up-Switch,” “Cross-Sell,” and “Maintenance” in the Ziemen document, without any information on the source of those numbers. Lanier Decl. ¶ 135, Exs. 1, 20. Meyer testified that he relied on the Ziemen document for his per- customer valuation because it shows “SAP’s own projections” for the potential revenues to be derived from TN customers. See id. But Meyer’s conclusion that this document relates to TN’s impact was pure guesswork as he knew neither the bases for the numbers nor how they could translate into the value of TN’s use of the works as opposed to the Safe Passage program as whole. Lanier Decl. ¶ 110, Ex. 1. The jury could only speculate to make that determination. Using these assumptions about customers and potential revenues per customer, Meyer calculated SAP’s expected gains from a PeopleSoft/JDE license as ranging from $881 million to $2.69 billion, a variance of over 300%; he calculated Oracle’s expected financial impact as ranging from $1.36 billion to $2.46 billion, a variance of over a billion dollars. Lanier Decl. ¶¶ 136-37, Exs. 1, 38. Based on this enormous range of supposed “impacts,” Meyer concluded, without explanation, that the value of the PeopleSoft/JDE license would be “at least $1.5 billion.” Lanier Decl. ¶ 138, Exs. 1, 38. Similarly, Meyer proposed a value for the Siebel license ranging from $97 million to $247 million, pricing it “at least $100 million.” Lanier Decl. ¶ 139, Exs. 1, 38. Combined, these licenses totaled “at least” $1.6 billion. Lanier Decl. ¶ 140, Ex. 1. 5. Oracle’s Counsel Invited the Jury to Speculate. During closing argument, Oracle’s counsel purported to summarize the manifold “valuations” of Oracle’s hypothetical license claim presented by its various fact and expert witnesses, which ranged from $897 million to $5 billion. Lanier Decl. ¶¶ 151-52, Exs. 1, 40. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page17 of 54 1 2 3 4 5 6 7 8 9 22 23 24 25 26 27 28 10 11 12 13 14 15 16 17 18 19 20 21 SVI-90062v1 - 12 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Faced with this broad assortment of numbers, Oracle’s counsel effectively conceded that even Meyer had been unable to accurately quantify the license amount and that many of the valuations offered during the trial were not supported by the evidence. Lanier Decl. ¶ 153, Ex. 1. Rather than offering objective evidence to assist the jury with calculating the hypothetical license fee, Oracle’s counsel invited the jury simply to pick some number between his $1.66 billion valuation and $3 billion, stating “the answer is probably somewhere in there.” Id. D. Oracle Relied Solely on the Self-Interested Testimony of a Single Oracle Executive to Estimate the Database Software Hypothetical License. Oracle used a different approach to calculate the database software hypothetical license. Oracle sells an Enterprise Edition license for database software and prices it by the number of processors on the computer used to run the software. Lanier Decl. ¶¶ 154, 156-57, Exs. 1, 32, 35. During the relevant time period, the price for an Enterprise Edition license was $40,000 per processor, plus $8,800 per processor per year for support. Lanier Decl. ¶¶ 157, 168, Ex. 1. The TN servers on which Oracle database software ran had 27 processors. Lanier Decl. ¶ 169, Ex. 1. Rather than following Oracle’s standard pricing structure, Meyer used a “per-customer” pricing structure devised solely for purposes of this case by Oracle employee Richard Allison, which required TN to pay for a database license for each customer who “benefited” from TN’s use of the database software, and to pay a 6-processor price ($292,000) for each license. Lanier Decl. ¶¶ 158-60, 162-65, Exs. 1, 38. Allison admitted that there was no “real world” transaction in which any customer had actually agreed to license the database software on this basis, other than his unsupported assertion that outsourcers pay per-customer. Lanier Decl. ¶ 161, Ex. 1. Meyer accepted Allison’s pricing structure and assumptions wholesale (admitting that had Allison suggested that twice the price was correct, Meyer would have accepted that assertion) and concluded that TN would have required 172 per-customer licenses, one for each customer who purportedly “benefited” from TN’s use of the database software. To determine the number of customers who “benefited,” Meyer “assume[d] that those customers [running human resources software] took—took benefits from the fixes and updates” developed on the Oracle database software TN used, whether or not that customer’s environment ran on an Oracle database. Lanier Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page18 of 54 1 2 3 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 7 SVI-90062v1 - 13 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Decl. ¶¶ 163-67, Exs. 1, 38. Meyer never explained this assumption. And the documents he cited as sources in his demonstrative slide show only that TN supported some customers using Oracle databases, not that non-Oracle database customers “benefited” from the Oracle database software. Lanier Decl. ¶¶ 155, 164, Exs. 1, 30, 31, 38. Using these assumptions, Meyer priced the database license at $55.6 million, which was close to the most generous calculation of TN’s total revenues during its entire operation ($59 million). Lanier Decl. ¶¶ 166, 170, Exs. 1, 39. IV. LEGAL STANDARD Judgment as a matter of law is warranted if there is no “legally sufficient evidentiary basis” for a reasonable jury to find for the non-movant. Fed. R. Civ. P. 50(a)-(b). Although the court views “the evidence in the light most favorable to the party in whose favor the jury returned a verdict” and draws all reasonable inferences in that party’s favor, “a reasonable inference cannot be supported by only threadbare conclusory statements instead of significant probative evidence.” Lakeside-Scott v. Multnomah County, 556 F.3d 797, 802 (9th Cir. 2009) (internal citation omitted). To defeat a motion for judgment as a matter of law, “the non-moving party must come forward with more than . . . a scintilla of evidence.” Miller v. Glenn Miller Prods., Inc., 454 F.3d 975, 988 (9th Cir. 2006) (citing Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252 (1986)). “Consequently, [judgment as a matter of law] is appropriate when the jury could have relied only on speculation to reach its verdict.” Lakeside-Scott, 556 F.3d at 802-03. A new trial may be granted “for any of the reasons for which new trials have heretofore been granted in actions at law in the courts of the United States,” including: the verdict is against the weight of the evidence, the damages are excessive, to prevent a miscarriage of justice, or, for other reasons, the trial was not fair to the moving party. Fed. R. Civ. P. 59(a); see also Molski v. M.J. Cable, Inc., 481 F.3d 724, 729 (9th Cir. 2007). In considering a motion for new trial, “[t]he judge can weigh the evidence and assess the credibility of witnesses, and need not view the evidence from the perspective most favorable to the prevailing party.” Air-Sea Forwarders, Inc. v. Air Asia Co., Ltd., 880 F.2d 176, 190 (9th Cir. 1989). A jury award cannot be upheld where “it is clearly not supported by the evidence or only based on speculation or guesswork.” In re First Alliance Mortg. Co., 471 F.3d 977, 1001 (9th Cir. 2006). And a damages award must be Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page19 of 54 1 2 3 4 5 6 7 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 8 SVI-90062v1 - 14 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) vacated where it is “so grossly excessive that it shock[s] the conscience.” Herrington v. County of Sonoma, 883 F.2d 739, 741 (9th Cir. 1989). In determining whether damages are excessive, “the [c]ourt must focus on evidence of the qualitative harm suffered by plaintiff.” Drew v. Equifax Info. Servs., LLC, No. C 07-00726 SI, 2010 WL 5022466, at *4 (N.D. Cal. Dec. 3, 2010). Where a damages award is excessive, the court “has the power to order a remittitur . . . and, if plaintiff does not agree to the remittitur, a new trial.” Anglo-American Gen. Agents v. Jackson Nat’l Life Ins. Co., 83 F.R.D. 41, 45 (N.D. Cal. 1979). V. ARGUMENT: RENEWED MOTION FOR JMOL Granting Defendants’ renewed motion for judgment as a matter of law that Oracle is not entitled to actual damages in the form of a hypothetical license fee is consistent with and logically flows from this Court’s prior rulings on the availability of such damages. The Court made clear that Oracle could claim hypothetical lost license fees only if it “present[ed] evidence sufficient to allow the jury to assess fair market value without ‘undue speculation.’” ECF No. 628 (Order) at 4. The Court left open the possibility that Oracle could present objective evidence at trial to show that Oracle incurred actual damages in the form of lost license fees. Ultimately, however, the trial evidence, which went well beyond what was available at summary judgment, left no room for doubt: Oracle did not lose a license fee because of the infringement, and thus is not entitled to actual damages in the form of a hypothetical license. Based on the now complete record, the Court can and should grant judgment as a matter of law that Oracle cannot recover such damages. Moreover, even if Oracle was entitled to seek hypothetical license damages, it failed to present a sufficient evidentiary basis for its hypothetical license claim, resulting in an award based on undue speculation. Because no objective evidence exists to ever establish a non-speculative license amount, the Court should grant judgment as a matter of law that this theory was legally improper and remit damages or order a new trial limited to lost profits and infringer’s profits. A. Oracle Is Not Entitled to Hypothetical License Fees as Actual Damages because the Undisputed Evidence Showed that It Did Not Lose License Fees. Under the Copyright Act, Oracle had a choice between pursuing statutory damages or actual damages and non-duplicative infringers’ profits. 17 U.S.C. § 504(a)-(c). Oracle elected to Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page20 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 15 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) seek recovery of actual damages and infringer’s profits. ECF No. 745 (Joint Pretrial Statement) at 9. As a result, and as explained below, Oracle needed to prove that it suffered actual damages— whether in the form of lost profits or lost license fees—as a result of Defendants’ infringement. Oracle did not (and could not) meet this burden of proving lost license fees at trial, resulting in the award of an impermissible, wildly speculative license fee, not “actual damages.” All of the evidence—including the unequivocal testimony of Oracle and SAP’s executives and damages experts, the parties’ fierce competition, and the absence of any comparable Oracle licenses or other benchmarks—compels this conclusion. 1. “Actual Damages” under the Copyright Act. Actual damages are calculated “by the loss in the fair market value of the copyright,” determined by either “the profits lost due to the infringement” or the “value of the use of the copyrighted work to the infringer.” Polar Bear Prods., Inc. v. Timex Corp., 384 F.3d 700, 708 (9th Cir. 2004); see also Mackie v. Rieser, 296 F.3d 909, 917 (9th Cir. 2002); Frank Music Corp. v. Metro-Goldwyn-Mayer, Inc., 772 F.2d 505, 512 (9th Cir. 1985). The most common and well- accepted way to measure “actual damages” is to prove a plaintiff’s lost profits. Baker v. Urban Outfitters, Inc., 254 F. Supp. 2d 346, 356 (S.D.N.Y. 2003) (“Typically, a plaintiff establishes impairment of market value by demonstrating that he lost sales or other profits that he would have obtained from the sale or license of the infringed work ‘but for’ the defendant’s infringement.”). However, where the plaintiff’s injury consists of lost or reduced license fees from the defendant or third parties, courts have in appropriate cases permitted recovery in the form of a hypothetical license fee, which is intended to compensate for a license fee that was actually lost. See Jarvis, 486 F.3d at 533; Cream Records, Inc. v. Jos. Schlitz Brewing Co., 754 F.2d 826 (9th Cir. 1985). But before attempting to quantify actual damages on a license or any other theory, a trier of fact must consider the threshold inquiry: Did the plaintiff actually suffer damages? More specifically, would plaintiff have made a sale (lost profits) or licensed the use (hypothetical license) of the copyrighted work, absent the intervening infringement? See Polar Bear, 384 F.3d at 711; Jarvis, 486 F.3d at 533; Frank Music, 772 F.2d at 514. If the answer is “no,” then the plaintiff may not recover such damages. Allowing recovery of purely hypothetical license fees Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page21 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 16 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) where the plaintiff never would have granted a license and thus lost no actual license fees contradicts both the plain language of the statute and Congressional intent to limit “actual damages” to those designed to redress harm “suffered by him or her as a result of the infringement.” 17 U.S.C. § 504(b); see also Business Trends Analysts, Inc. v. Freedonia Group, Inc., 887 F.2d 399, 405-06 (2d Cir. 1989). 2. The Standards for Permitting a “Hypothetical License” Fee as Actual Damages. Based on well-understood principles applying to all claims for actual damages, a plaintiff seeking actual damages in the form of a hypothetical license fee must show that the infringement caused a loss of or reduction in license fees—just as a plaintiff seeking actual damages in the form of lost profits must prove that the infringement caused lost sales. As a result, the Ninth Circuit has never upheld a hypothetical license award absent proof that the plaintiff would have licensed the infringed work to the defendant or a third party for the specific use at issue, and that the infringement caused the loss of that opportunity. In Polar Bear, the defendant had a license which then expired, providing a basis to determine if a license fee was lost. 384 F.3d at 704, 709. In Jarvis, the Ninth Circuit affirmed a hypothetical license award for infringement of copyrighted photographs where the plaintiff had licensed photographs to the defendant in the past. 486 F.3d at 528, 533-34. In Mackie, the Ninth Circuit upheld a hypothetical license award where the parties were not competitors and the plaintiff “had previously given permission for others to use” the copyrighted work in the same way. 296 F.3d at 913, 917. And in Cream Records, the Ninth Circuit upheld recovery of actual damages in the form of a lost license fee where unauthorized use of a copyrighted song in defendant’s commercial deterred potential licensees from licensing the song for their commercials. 754 F.2d at 827-28. Other courts are in accord. In Encyclopedia Brown Prods., Ltd. v. Home Box Office, Inc., the court upheld a hypothetical license award where the defendant previously had a license to use the copyrighted video, but continued to use it after the license had expired. 25 F. Supp. 2d 395, 401-02 (S.D.N.Y. 1998). In On Davis, the Second Circuit allowed recovery of a hypothetical license fee for the defendant’s unauthorized use of copyrighted eyewear in defendant’s Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page22 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 17 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) advertising campaign where the parties were not competitors and the plaintiff had obtained a $50 royalty for use of the sunglasses in a different magazine feature. 246 F.3d at 161-62. Conversely, where there is no evidence that the parties actually would have agreed to a license and the plaintiff has not otherwise lost a license fee, courts have held that actual damages under the statute may not be measured by a hypothetical license fee. In Business Trends, for example, the parties directly competed in the publication of economic analyses and forecasts, and both produced a robotics industry study. 877 F.2d at 401. Upon finding that the defendant infringed the plaintiff’s study, the district court awarded a hypothetical license fee. See id. at 404. The Second Circuit reversed, holding that the “proof in the instant case is inadequate to support such an award” because the parties were direct competitors and there was no evidence that they had any inclination to agree to a license for the infringing activities. Id. at 407. Similarly, in National Conference of Bar Examiners v. Multistate Legal Studies, Inc., the court held that the plaintiffs could not recover hypothetical license damages because the evidence established that “there could not have been such a contract” between the parties. 458 F. Supp. 2d 252, 261 (E.D. Pa. 2006). There, the plaintiff developed copyrighted questions for use in the Multistate Bar Examination (“MBE”), which the defendant copied without authorization for use in its bar examination preparation course. See id. at 256. In declining to award lost licensing fees as actual damages for infringement of the copyrighted exam questions, the court considered evidence that such a license would not have made business sense for the parties. See id. at 261. Specifically, noting that plaintiff “goes to great lengths to maintain the secrecy of those questions” so that they may be reused, the court found that the plaintiff would not have agreed to license those questions to defendant “because to release current MBE questions is to undermine the validity of the entire examination.” Id. The court also held that “there is also no evidence to suggest that defendants would have licensed released questions, because such questions do not provide the crucial information defendants sought—previews of upcoming tests.” Id. “Since plaintiff lost no hypothetical royalties,” the court concluded that it could not award actual damages in the form of hypothetical license royalties. Id. Finally, in Frank Music, the district court declined to award actual damages for the Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page23 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 18 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) unauthorized use of a portion of a musical in a Las Vegas show, finding no evidence that the infringement had diminished the market value of the work. 722 F.2d at 513-14. The Ninth Circuit affirmed, stating that the infringement did not reduce the plaintiff’s ability to license the musical to other Las Vegas hotels and thus it was not entitled to a hypothetical license award. Id. 3. Oracle Did Not Lose a License Fee from Any Third Party. Oracle has never argued that it is entitled to a hypothetical license because it lost the opportunity to license the works to third parties for the same use as was made by TN. By its own admission, Oracle has never given any entity a license to “copy Oracle’s application software and support materials in order to create their own fixes, patches or updates for customers.” D.I. 256 (Decl. of Colleen A. Kelly) ¶¶ 3-4. This is further evidenced by the absence of any benchmark licenses or trial testimony regarding the same. See infra III.B.3. Because Oracle is not in the business of licensing its copyrighted works to third-party support providers, it cannot argue that Defendants’ infringement diminished the licensing value of the infringed works. Cf. Frank Music, 772 F.2d at 513-14; Cream Records, 754 F.2d at 827-28; Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp., 562 F.2d 1157, 1162, 1179 (9th Cir. 1977). Accordingly, Oracle’s only avenue for proving its entitlement to hypothetical lost license fees was to show that, but for infringement, the parties would have agreed to a license. It did not do so at trial, and it could not have done so. 4. Oracle Did Not Lose a License Fee from Defendants. Taken together, the cases permitting and prohibiting a hypothetical license claim demonstrate that there are several factors to consider when determining whether the parties would have agreed to a license, including whether: (1) the parties are direct competitors, (2) the parties sell goods in the same market, (3) the parties previously entered into a similar license, (4) the plaintiff has granted a similar license, and (5) the license makes business sense to the parties. See Polar Bear, 384 F.3d at 711 (parties would have agreed to license because they had licensed in the past); Business Trends, 887 F.2d at 405-06 (direct competitors would not have agreed to license); Nat’l Conference of Bar Examiners, 458 F. Supp. 2d at 261 (parties would not have agreed to license because it would not have made business sense); On Davis, 246 F.3d at 161-62 Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page24 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 19 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Oracle’s witnesses uniformly and unequivocally testified that Oracle never would have granted a license to Defendants. Oracle’s executives confirmed that, immediately following the PeopleSoft acquisition, it would not have made business sense for Oracle to give its biggest competitor a license to permit TN to use Oracle’s intellectual property to compete against Oracle in providing support services for the very PeopleSoft customers it just acquired. Lanier Decl. ¶¶ 25-28, Ex. 1. In fact, it was “terrif[ying],” “unprecedented,” and “unthinkable.” See id. And (parties would have agreed to license because they did not compete and plaintiff had licensed work in the past); Encyclopedia Brown, 25 F. Supp. 2d at 401-02 (same); Baker v. Urban Outfitters, 254 F. Supp. 2d at 357-58 (same). Evidence of direct competition, by itself, strongly supports a conclusion that a hypothetical license is not a proper measure of actual damages because direct competitors are unlikely to license their intellectual property to one another. See Business Trends, 887 F.2d at 405-06; Encyclopedia Brown, 25 F. Supp. 2d at 401-02. And the absence of benchmark licenses demonstrates that there is no viable licensing market from which the plaintiff would have derived licensing fees and thus that there has been no harm to the fair market value of the work in the form of lost license fees. Accordingly, cases in which the Ninth Circuit affirmed a hypothetical license fee all involved non-competitors. See Wall Data, Inc. v. LA County Sheriff’s Dept., 447 F.3d 769 (9th Cir. 2006) (computer software developer vs. municipal customer); Polar Bear, 384 F.3d 700 (watch company vs. film production company); Frank Music, 772 F.3d 505 (music production company vs. hotel); Jarvis, 486 F.3d 526 (photographer vs. advertiser); Mackie, 296 F.3d 909 (artist vs. symphony). And the Ninth Circuit has never affirmed a hypothetical license award where no comparable transactions establishing the likelihood of a license between the parties exist. Jarvis, 486 F.3d at 534 (upholding license award where parties had previous license for infringed works); Polar Bear, 384 F.3d at 710-11 (same); Mackie, 296 F.3d at 917 (affirming award where plaintiff previously gave “permission to others to use” work). Oracle’s claim fails on every factor, based on the trial evidence. (a) Oracle Witnesses Unequivocally Testified that Oracle Would Not Have Granted a License to Defendants. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page25 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 20 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) As in Business Trends, the parties’ direct competition is a critical and undisputed fact, demonstrating that a hypothetical license fee is not a proper damages measure. 887 F.2d at 405- 06; Jarvis, 486 F.3d at 533. Confirming the logic of this conclusion, Phillips testified it was “unthinkable” that Oracle would have ever granted SAP a license in light of the competitive relationship. Lanier Decl. ¶ 25, Ex. 1. (“I would have been terrified if my largest competitor now Defendants’ side of the purely hypothetical (and counterfactual) negotiating table was equally unwilling to agree to a license. Brandt testified that SAP would have asked TN to further change its procedures or “would have considered not to buy TN,” rather than entering into a license with Oracle. Lanier Decl. ¶ 30, Ex. 1. Just as Oracle never would have granted such a license, SAP would not have entered into one either. As in National Conference of Bar Examiners, the undisputed fact that it would not have made business sense for these parties to agree to such an unprecedented license supports the conclusion that Oracle may not recover actual damages in the form of a hypothetical license. Nat’l Conference of Bar Examiners, 458 F. Supp. 2d at 261; cf. Jarvis, 486 F.3d at 533; Business Trends, 887 F.2d at 405-06. Oracle executives’ unambiguous pronouncements that Oracle never would have granted a license to SAP only underscore Oracle’s failure to show the causal connection required to recover damages in the form of hypothetical license fees. (b) Direct Competition between the Parties and the Lack of Any Benchmarks Further Confirm that a Hypothetical License Fee Is Inapplicable Here. The trial testimony about SAP and Oracle’s status as direct competitors conclusively demonstrates that Oracle never would have given such a license to Defendants. There is no question that each considers the other its chief competitor in the applications business. Lanier Decl. ¶¶ 35-36, 38-39, Exs. 1, 5 (Phillips, Ellison, Aggasi and McDermott describing fierce competition between the two companies). Both damages experts confirmed this competitive relationship. Lanier Decl. ¶¶ 40-41, Ex. 1. And Catz described the fierce competition between the parties for customers as a “zero sum game,” such that every PeopleSoft customer SAP acquires is one Oracle no longer has. Lanier Decl. ¶ 37, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page26 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 21 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) had the exact same product I had and this competitor is three times my size and can go in and sell the same thing, support the exact same customers. . . . So it would have been -- yeah, it would have just been unthinkable at the time.”). There is also no evidence—such as comparable Oracle licenses—to suggest that Oracle would have even negotiated, let alone granted, a license to Defendants. See Jarvis, 486 F.3d at 528, 533-34 (allowing hypothetical license where parties were not direct competitors and had previously entered into copyright licenses for work at issue); Encyclopedia Brown, 25 F. Supp. 2d at 401 (same). As its own executives and damages expert testified, Oracle has never given any entity a license that would cover the stipulated infringing activities here. Allison testified that Oracle has never granted a license for a competitor to use Oracle’s intellectual property the way TN used it, and that such a license would be “unique.” Lanier Decl. ¶¶ 26-27, Ex. 1. Similarly, Catz admitted that there are no “comparable” licenses to the one Oracle seeks from SAP; for example, none of Oracle’s licenses with its partners or resellers allow them to use Oracle’s intellectual property to compete for customers. Lanier Decl. ¶ 28, Ex. 1. And Meyer testified that Oracle has never “in the history of Oracle” granted a license for a competitor to use its database software to compete for maintenance customers. Lanier Decl. ¶ 33, Ex. 1. Ultimately, there was no evidence of licenses for comparable use of software applications. Brandt testified that he knows of no situation in which SAP tried to license a competitor’s software to provide maintenance services. Lanier Decl. ¶ 29, Ex. 1. Similarly, McDermott testified that he is not aware of any licenses in which intellectual property is used to compete for customers. Lanier Decl. ¶ 31, Ex. 1. And Clarke testified that he did not find any established royalty or benchmark existing in the marketplace. Lanier Decl. ¶ 34, Ex. 1. That Oracle has never granted the type of copyright license it now insists it would have required from Defendants, and could not muster a single benchmark license with anyone, much less a competitor, confirms that Oracle cannot recover actual damages in the form of a hypothetical license fee. B. Oracle Failed to Offer Legally Sufficient Evidence to Value the Hypothetical License. Even were Oracle entitled to seek actual damages in the form of a hypothetical license Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page27 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 22 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) because it had proven it lost a license fee, JMOL is still (and independently) appropriate. The claimed amounts Oracle presented to the jury were unreasonable and lacked an objective basis, supported only by improper factors, speculative evidence, and unfounded assumptions, all of which were insufficient to determine the value of TN’s actual use of the copyrighted works. 1. Hypothetical License Awards May Not Be Based on Undue Speculation. The amount of a hypothetical license—reached if and only if a plaintiff meets the threshold inquiry of proving license fees were lost—is calculated by considering “‘what a willing buyer would have been reasonably required to pay to a willing seller’ for plaintiffs’ work . . . at the hypothetical time of sale.” Jarvis, 486 F.3d at 533-34 (quoting Sid & Marty Krofft, 562 F.2d at 1174); see also Mackie, 296 F.3d at 917. Courts “recognize that awarding the copyright owner the lost license fee can risk abuse” because “[o]nce the defendant has infringed, the owner may claim unreasonable amounts as the license fee.” On Davis, 246 F.3d at 166. Thus, determining the hypothetical license price requires an “objective, not a subjective” analysis, and “[e]xcessively speculative” claims must be rejected. Jarvis, 486 F.3d at 534; see also Mackie, 296 F.3d at 917; Polar Bear, 384 F.3d at 709; On Davis, 246 F.3d at 166. Moreover, the amount of the hypothetical license must be based on the actual use defendant made of the work, not simply “the highest use for which plaintiff might license.” On Davis, 246 F.3d at 166 n.5; see also Jarvis, 486 F.3d at 535 (“Making the plaintiff whole is plainly different from punishing the infringer by charging the highest possible rate for the infringement.”); Wall Data, 447 F.3d at 786; Country Road Music, Inc. v. MP3.com, Inc., 279 F. Supp. 2d 325, 330-31 (S.D.N.Y. 2003). In applying the willing buyer/willing seller test, courts focus on objective evidence of benchmark transactions, such as licenses previously negotiated for comparable use of the infringed work and benchmark licenses for comparable uses of comparable works. Jarvis, 486 F.3d at 533 (affirming award based on evidence showing what defendant typically paid to license photographs, prior dealings with plaintiff, and what plaintiff typically charged to license photographs); Polar Bear, 384 F.3d at 709 (affirming award based on parties’ existing license); Mackie, 296 F.3d at 913, 917 (affirming award based on evidence of past practice). Objective evidence is necessary to price a hypothetical license because the hypothetical license is simply a Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page28 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 23 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) construct designed to help calculate “actual damages suffered as a result of the infringement.” ECF No. 1005 (Final Jury Instructions) at 7; 17 U.S.C. § 504(b). Hypothetical license claims that are un-tethered to real world damages and far afield of “actual damages” suffered are too speculative and cannot stand. See Mackie, 296 F.3d at 917 (rejecting $85,000 hypothetical license and affirming $1,000 award where evidence showed plaintiff had granted permission for others to use for free); On Davis, 246 F.3d at 161 (rejecting “wildly inflated” $2.5 million hypothetical license and affirming $50 award based on past licensing history).4 Subjective evidence alone cannot establish a non-speculative hypothetical license. Interplan Architects, Inc. v. C.I. Thomas, Inc., No. 4:08-cv-03181, 2010 U.S. Dist. LEXIS 114306, at *34-37 (S.D. Tex. Oct. 27, 2010) (dismissing hypothetical license based solely on plaintiff’s declaration stating price he would have charged); Smith v. Rush, No. C04-2280Z, 2006 U.S. Dist. LEXIS 27412, at *2-3 (W.D. Wash. Apr. 7, 2006) (same); Leland Med. Ctrs., Inc. v. Weiss, No. 4:07cv67, 2007 WL 2900599, at *6-8 (E.D. Tex. Sept. 28, 2007) (granting motion to strike expert testimony pricing hypothetical license based on what parties “would hope to earn” from hypothetical license). Accordingly, courts routinely preclude recovery of hypothetical license damages in the absence of objective, non-speculative evidence to quantify the fee. See Bi-Rite v. Button Master, 578 F. Supp. 59, 60 (S.D.N.Y. 1983) (holding hypothetical license not appropriate where sole evidence offered was non-comparable benchmarks); Interplan Architects, 2010 U.S. Dist. LEXIS 114306, at *34-37; Smith, 2006 U.S. Dist. LEXIS 27412, at *2-3; Technologies, S.A. v. Cyrano, Inc., 460 F. Supp. 2d 197, 200-03 (D. Mass. 2006) (holding that unreliable evidence of projections was too speculative to support hypothetical license). 2. Oracle Failed to Present Objective Evidence of Benchmark Licenses. There is no dispute that the objective factors on which courts rely to calculate the price of a hypothetical license—previously negotiated licenses for the work at issue or other benchmark licenses—are absent here. Oracle executives Allison and Catz testified not only that Oracle has 4 Notably, hypothetical license fees affirmed on appeal have been in modest amounts. In the Ninth Circuit, they range from $37,500 (Polar Bear, 384 F.3d at 709) to $210,000 (Wall Data, 447 F.3d at 787). Courts in this Circuit have never allowed the hypothetical license theory to be a “runaway train,” justifying enormous amounts of imagined damages. The disparity between the affirmed awards and the $1.3 billion here signals just how amiss this verdict is. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page29 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 24 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) never granted a comparable license that would permit a competitor to use Oracle software to compete for Oracle’s customers, but also that such a license would be “unique” and “unprecedented.” Lanier Decl. ¶¶ 26-28, Ex. 1. Further, SAP executives Brandt and McDermott were similarly unaware of analogous situations in which a company has licensed software to or from a competitor to provide support services. Lanier Decl. ¶¶ 29, 31, Ex. 1. And both sides’ damages experts agreed that no benchmark licenses exist. Lanier Decl. ¶¶ 33-34, Ex. 1. 3. Oracle’s Hypothetical License Claim for the PeopleSoft, JDE, and Siebel Works Is Unreasonable and Unduly Speculative. Unable to ground its license claim in objective, real-world licensing history and practices, Oracle instead invited the jury to rely on other “factors”—including the price Oracle claimed it would have demanded, the value of its intellectual property as a whole, and fictitious “negotiation” factors—to support its demand for “at least” $1.6 billion in damages. But none of the factors or evidence that Oracle offered, alone or together, suffice to support a reasonable, non- speculative award of actual damages. (a) Factors Emphasized by Oracle’s Counsel Are Insufficient to Support a Non-Speculative License. Throughout its case-in-chief and closing, Oracle’s counsel urged the jury to consider what Oracle would have demanded to license the works and the purported value of the infringed intellectual property as whole. Case law does not support reliance on either, and even if it did, the evidence offered in support is too speculative to justify the damages awarded here. First, it is well established that the price a plaintiff claims it would have demanded to license its work cannot support a hypothetical license claim because it is too speculative. Jarvis, 486 F.3d at 534 (“The question is not what the owner would have charged, but rather what is the fair market value.”) (citing On Davis, 246 F.3d at 166); Interplan Architects, 2010 U.S. Dist. LEXIS 114306, at *34-37 (dismissing license claim based solely on plaintiff’s declaration stating price he would have charged to use copyrighted work); Smith, 2006 U.S. Dist. LEXIS 27412, at *2-3 (same). Where, as here, Oracle lacked objective factors to establish a non-speculative license price, one-sided, subjective testimony by Oracle’s executives that they would have Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page30 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 25 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) demanded “billions” is inadequate to support Oracle’s license claim. See id.; Lanier Decl. ¶¶ 42- 44, Exs. 1, 40. Moreover, speculation by Ellison, Catz, and Phillips concerning what they would have charged, years ago, for purely hypothetical and counterfactual licenses is insufficient to support an objective hypothetical license calculation. See id. Second, calculating actual damages for copyright infringement based on the value of intellectual property as a whole is improper as a matter of law. Copyright law permits the award of damages attributable to infringement. 17 U.S.C. § 504(b). It does not permit a plaintiff to recover value that resides in patented technology, trade secrets, trademarks, or any of the myriad un-copyrightable elements of a computer program, such as ideas, processes and systems, functional elements, or parts of the program that make it work more efficiently. See 17 U.S.C. § 102(b); Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390, 405-06 (1940) (“[t]hat would be not to do equity but to inflict an unauthorized penalty”); Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1439 (9th Cir. 1994); U.S. Copyright Office Circular 61, “Copyright Registration for Computer Programs.” Lanier Decl. ¶ 212, Ex. 42. Here, Oracle did not even purport to engage in the necessary exercise of identifying and valuing the use of only the copyright-protectable elements of the infringed works. Instead, Oracle’s inflated hypothetical license claim was based on the value of the intellectual property as a whole (including all value attributable to non-protectable functional elements and other intellectual property rights). As a result, Oracle’s damages approach—and the jury’s verdict—compensated Oracle for infringement of rights that were not at issue during trial. Moreover, by suggesting that Oracle’s hypothetical license should be based on the value of the intellectual property that TN “took”—measured by Oracle’s R&D5 and acquisition costs— rather than on TN’s actual use of the copyrightable elements of the software, Oracle essentially sought compensation for the full cost to replace software, instead of the value of TN’s use of that software. The Ninth Circuit and other courts are clear that a hypothetical license is measured by what a willing buyer would have been reasonably required to pay a willing seller at the time of 5 Evidence of R&D costs also runs afoul of this Court’s Order on summary judgment precluding Oracle from recovering “research and development costs to plaintiffs as actual damages for infringement.” See ECF No. 762 (8/17/10 Order) at 22-23. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page31 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 26 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Even were it appropriate to consider intellectual property value as a whole, the evidence on which Oracle relied to calculate this value was impermissibly speculative. In support of its R&D costs measure, Oracle relied on evidence of R&D costs for all of its intellectual property and/or maintenance products, including for dates beyond the relevant time period, products not at issue, and rights not alleged to have been infringed (patents, trade secrets, and trademarks). Lanier Decl. ¶¶ 48-50, Exs. 1, 37. This evidence cannot support a non-speculative calculation of the R&D costs for the infringed works, let alone a calculation of the value of those works or a limited license for TN to use the copyrighted elements therein. Oracle never even provided an licensing for the actual use made of the work. See Wall Data, 447 F.3d at 786; On Davis, 246 F.3d at 166 n.5; Country Road Music, 279 F. Supp. 2d at 330-31. Courts expressly reject attempts to recover hypothetical license fees for more than the actual use made of plaintiff’s work. See id.; Powell v. Penhollow, 260 Fed. App’x 683, 690-91 (5th Cir. 2007) (affirming no hypothetical license for completed architectural plans because defendant infringed only incomplete plans); Propet USA, Inc. v. Shugart, No. C06-0186-MAT, 2007 U.S. Dist. LEXIS 69222, at *4-5 (W.D. Wash. Sept. 19, 2007) (holding recovery proper for “only those images shown by [plaintiff] to have been used in violation of the asserted license terms . . .”). Recovering the value of intellectual property as though it were stolen rather than the copyrightable elements improperly used is particularly inappropriate here because Oracle has, at all times, retained its rights to use, distribute, license, and profit from its software and support materials. Despite Oracle’s misleading suggestions that TN’s use of Oracle software to service customers amounted to a taking of “real property,” Lanier Decl. ¶ 61, Ex. 1, Oracle witnesses admitted that Oracle still possesses and can license the copyrighted software programs at issue. Lanier Decl. ¶¶ 52, 56, Ex. 1. This Court has held that where, as here, “plaintiffs retained their right to use, distribute, license, and profit from the software and support materials. It would not be equitable, logical, or legally permissible to award plaintiffs the full replacement value of property that they never lost or gave away.” ECF No. 762 (8/17/10 Order) at 19. Ultimately, considering the full value of the intellectual property, rather than actual use of the work, is inappropriate as a matter of law in calculating a hypothetical license price. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page32 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 27 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Evidence of SAP’s Purported “Need for the Works” and its “Competitive Relationship” with Oracle: These factors are both insufficient to price a hypothetical license. With respect to SAP’s purported need for the infringed works, at least one court found that the subjective consideration of whether a defendant “needed” access to a copyright owner’s software is insufficient to establish the fair market value of a hypothetical license. See DaimlerChrysler Servs. v. Summit Nat’l, No. 02-71871, 2006 WL 208787, at *1-2 (E.D. Mich. Jan. 26, 2006) explanation as to how its R&D costs relate to the total value of its intellectual property, except to suggest that such costs might discourage Oracle from granting a license that would deprive it of the maintenance stream that funds its R&D efforts. Lanier Decl. ¶ 60, Ex. 1. But that connection is too attenuated to establish an objective, non-speculative license price. Similarly, evidence of the price that Oracle paid to acquire PeopleSoft and Siebel is too speculative a basis on which to value Oracle’s intellectual property, let alone value TN’s use. Oracle witnesses testified that they would price a hypothetical license based on the billions spent to acquire PeopleSoft and Siebel (e.g., by multiplying the purchase price of the acquired company by the percentage of customers likely to be lost as a result of the license). Lanier Decl. ¶¶ 51, 53- 55, 57-59, Ex. 1. But the acquisition price of an entire company reflects the value of all of the company’s assets, including real estate, office equipment, bank accounts, and customer relationships, as well as all of the intellectual property owned by the company (patents, trademarks, etc.). Lanier Decl. ¶ 53, Ex. 1. As a result, the rights Oracle acquired when it purchased PeopleSoft and Siebel (i.e., full ownership rights) include far more than just intellectual property, and certainly far more than the price of a limited license to use the copyrighted elements of the works at issue here. (b) Meyer’s “Background” Factors Are Irrelevant to Pricing a Hypothetical License. Meyer’s “background” factors for Oracle’s claimed hypothetical license fees—such as SAP’s alleged “need for the works,” its “competitive relationship” with Oracle, and its “risk acceptance,” all of which are founded solely on speculation—do not make his testimony or the jury award any less speculative. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page33 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 28 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) (rejecting valuation of hypothetical license based on plaintiff’s “subjective estimate as to the price it would have charged” had it known that the infringed work was in fact “critical” to defendant’s business). Further, as discussed above, evidence of the parties’ competitive relationship counsels against the award of a hypothetical license. See Business Trends, 887 F.2d at 405-06; Leland Med. Ctrs., 2007 WL 2900599, at *8 (holding that evidence that parties were competitors that would not have agreed to a license “practically strips [plaintiff] of any causation claim” in support of a hypothetical license). Even if evidence of SAP’s purported “need for the works” and its “competitive relationship” with Oracle were relevant to pricing a license, these factors cannot establish a license price; at most, as Meyer confirmed, such facts can only apply upward or downward pressure to it. Lanier Decl. ¶¶ 101, 138, Exs. 1, 38. SAP’s “Risk Acceptance”: SAP’s alleged “risk acceptance” that TN was infringing is factually (by Meyer’s own admission) and legally (by case law) irrelevant to the license price. As a threshold matter, it is improper to factor risk of litigation into a hypothetical license analysis, since the analysis necessarily assumes no infringement will occur.6 Lanier Decl. ¶ 74, Ex. 1, Barrera v. Brooklyn Music, Ltd., 346 F. Supp. 2d 400, 410-11 (S.D.N.Y. 2004) (holding that evidence regarding party’s willingness to infringe or fear of impending litigation “cannot logically represent part of the fair market value of a license authorizing such use”). Courts also recognize that evidence of willful infringement is not admissible to prove hypothetical license damages because it is inconsistent with the limits on damages under the Copyright Act and imports a legally impermissible punitive component into the analysis. See Stehrenberger v. R.J. Reynolds Tobacco Holdings, Inc., 335 F. Supp. 2d 466, 468 (S.D.N.Y. 2004) (holding that evidence of willful infringement “does not define a fair and reasonable license fee, but represents concepts of punishment for infringement,” and thus “form[s] no part of ‘actual damages’ under the statute”); Faulkner v. Nat’l Geographic Soc’y, 576 F. Supp. 2d 609, 613 (S.D.N.Y. 2008) (holding that because “[c]opyright infringement is a strict liability wrong,” it is inappropriate to increase a purely compensatory measure of damages with evidence that would support an award 6 Despite suggesting that SAP’s so-called “accepted risk of litigation” relates to a hypothetical license, Meyer acknowledged that in a hypothetical negotiation, one assumes that the license, not the infringement, would have occurred. Lanier Decl. ¶ 73, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page34 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 29 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Noting that “the inquiry is an objective one into the resultant fair market value after negotiation between a willing buyer and seller,” the court excluded the expert’s opinion. Id. at *7-8. Addressing the claimed incentives needed to motivate the plaintiff to license and the of prohibited punitive or foregone statutory damages). (c) Meyer’s Primary Factors Were Insufficient to Support a Non- Speculative License Calculation. Meyer’s analysis of “primary factors”—SAP and Oracle’s purported goals and expectations in a hypothetical negotiation for the PeopleSoft/JDE and Siebel licenses—did not provide the jury sufficient, non-speculative grounds on which to base a hypothetical license award. Instead, his approach was purely subjective, focusing on guesses and inferences regarding the parties’ hopes and goals for exploiting the copyrighted works and lacking any empirical, objective bases to support a hypothetical license fee. (i) Evidence of Goals and Expectations Cannot Independently Establish Market Price. As a matter of law, evidence of the parties’ subjective goals (i.e., what a party hopes or aims to occur), or even their expectations (i.e., what a party believes actually will occur), cannot establish the objective market price of a hypothetical license. See Leland Med. Ctrs., 2007 WL 2900599, at *6-8. In Leland Medical Centers, the court excluded as “not reliable and merely speculative” expert testimony that purported to calculate the price of a hypothetical license by relying on subjective evidence of the parties’ goals and expectations. Id. at *8. There, the plaintiff sought hypothetical license damages for its proprietary hospital floor plan design. Id. at *1-2. Reasoning that the plaintiff would not have licensed its floor plan design without “secur[ing] other benefits which were potentially more lucrative,” including equity in and fees from the licensee’s hospital business, the plaintiff’s expert considered four factors to calculate the hypothetical license price: “the financial incentives necessary to entice [plaintiff] to dispose of his interest in the West Texas hospital [whose design allegedly infringed plaintiff’s copyright]; the anticipated fees [plaintiff] would have earned on proposed projects; the efficiency of the design; and the anticipated profits to be earned by the Foundation defendants.” Id. at *6-7. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page35 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 30 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Despite evidence of “expectations” being insufficient to establish an objective market price of a hypothetical license, Meyer relies on documents and testimony that he claimed establish SAP and Oracle’s purported expected financial impact of TN’s support offering (and thus, the price of the PS/JDE and Siebel licenses). Lanier Decl. ¶¶ 105-06, Exs. 1, 38. However, these documents contain only subjective “goals,” “scenarios,” and “assumptions,” regarding the sale of SAP software through the Safe Passage marketing program, for which the offer of TN support was an optional component. Lanier Decl. ¶ 109, Ex. 1. Thus, they do not even purport to provide an “expected” value of TN’s contribution to these sales, let alone provide an objective basis upon which to value TN’s use of the copyrighted works. The drafters and recipients of these documents never testified that these documents projected the value of TN’s contribution to Safe Passage. And self-serving testimony by Oracle witnesses interpreting these documents, is completely speculative. See, e.g., Lanier Decl. ¶ 111, Ex 1. plaintiff’s expectations as to the financial impact of such a license, the court held that “[w]hat [the plaintiff] would hope to earn” in connection with such a license “bears no relationship to the market value diminution or injury to his single page schematic of a hospital.” Id. at *7. Similarly, the court found the expert’s consideration of the defendant’s expected profits, based on projections contained in budgeting documents, “merely speculative,” particularly as the expert “ha[d] not demonstrated how [defendant’s] profits translate into his actual damage number.” Id. While noting that “efficiency of the design” might bear on its market value, the court concluded that the expert’s analysis was too unreliable and speculative to present at trial. Id. Here, just as in Leland Medical Center, evidence of the factors Meyer claims Oracle would consider in a hypothetical negotiation does not provide an objective, fair market value of such a license. Lanier Decl. ¶ 96, Ex. 1. Likewise, evidence of SAP’s purported goals or expected gains in licensing the software is too speculative to itself support a license calculation. Like the expert in Leland Medical Center, Meyer never showed how SAP’s purported goals or expectations relate to the actual harm Oracle suffered by virtue of infringement. (ii) Meyer’s Evidence Does Not Provide a Legally Sufficient Basis to Calculate a Hypothetical License Price. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page36 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 31 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) The “Ziemen” Document: For the PeopleSoft/JDE license, Meyer primarily relied on the “Ziemen document” to calculate both sides’ “expected” financial gains or losses from the PeopleSoft/JDE license. Lanier Decl. ¶¶ 108, 135, Exs. 1, 20, 38. But this document concerns the “Safe Passage” marketing program as a whole, not SAP’s expectations on the success of an optional support offering (TN or otherwise). Lanier Decl. ¶¶ 107, 109, Exs. 1, 20. Oracle’s own expert conceded that any customer numbers therein were simply “assumptions.” Lanier Decl. ¶ 110, Ex. 1. Oracle offered no credible, objective evidence that SAP, or a reasonable willing buyer in SAP’s position, would have based its willingness to pay for a license on the assumptions in this document. Indeed, Oracle offered no corroborating evidence—only testimony from its own executives purporting to interpret the SAP documents as reflecting SAP’s expectations, despite Meyer’s concession that he had no idea how the analysis was done. Lanier Decl. ¶ 111, Ex. 1. None of the other evidence on which Meyer relied offers any better support for his conclusions. The “PeopleSoft 1-2-3” Memorandum: Meyer claimed that Zepecki’s “PeopleSoft 1-2-3” Memorandum identifies TN as the most important step in SAP’s strategy to convert thousands of customers. Lanier Decl. ¶¶ 113-14, Exs. 1, 17, 19, 21. In reality, it proposes a marketing strategy —later to become “Safe Passage”—that could include optional third-party support for PeopleSoft customers, with no indication of how big a part TN might play.7 See id. Zepecki, after meeting with TN, foresaw, at best, a “niche” market, resulting in a possible 5% of PeopleSoft’s customers—not the 20% or 30% assumed by Meyer. Lanier Decl. ¶¶ 115-16, Exs. 1, 18. The “Safe Passage: Winning Customers and Markets from Oracle-PeopleSoft-J.D. Edwards” Document: Meyer claimed that this document shows that SAP expected to convert 450 customers in the first 30 days of Safe Passage, and 50% or a “majority” thereafter. Lanier Decl. ¶¶ 117-18, Exs. 1, 23, 26. The reference to 450 customers in 30 days, however, refers to the number of customers to whom SAP hoped to “[r]each out”; similarly, converting a “majority” of customers was SAP’s “goal,” not an expectation. Lanier Decl. ¶ 117, Ex. 26 at SAP-OR00092050. This document never values the impact of TN’s use of Oracle software to 7 Similarly, the December 15, 2004 Executive Board Minutes show only that SAP approved a strategy to offer maintenance, with no discussion of TN. Lanier Decl. ¶ 112, Ex. 16. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page37 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 32 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Documents Used to Support the Siebel Hypothetical License: Similarly, for the Siebel hypothetical license, Meyer relied on documents containing only subjective and speculative aspirations. He testified that the October 17, 2005 Business Case and the “Siebel Safe Passage Program Playbook” showed that SAP expected to gain 200-300 Siebel customers through TN. Lanier Decl. ¶¶ 124-25, 127-28, Exs. 1, 33-34, 38. But Oracle offered no data to support this the success of the Safe Passage program. Id. The “TN Integration Meeting” Document: Meyer claimed that this document shows that SAP believed TN could convert 2,000 to 4,000 customers, described as “conservative” and “upbeat” estimates, respectively. Lanier Decl. ¶¶ 67, 119, Exs. 1, 28. But SAP AG executive Gerhard Oswald confirmed what is obvious on the document’s face: Those numbers are only potential “scenarios” (i.e., what could happen), not estimates of what SAP believed would happen. Id. Lanier Decl. ¶ 120, Ex. 1. The Analyst Call Transcript: With regard to the analyst call announcing the TN acquisition and Safe Passage launch, Meyer claimed that the transcript shows SAP expected to convert 100% of shared SAP/PeopleSoft customers—4,000 of the approximately 10,000 PeopleSoft customers. Lanier Decl. ¶¶ 121-22, Exs. 1, 22, 38. The document shows no such thing. Indeed, when asked on the call to project the number of shared customers that would migrate to SAP via Safe Passage, Agassi declined to speculate and deferred to the analysts to draw their own conclusions. Lanier Decl. ¶ 121, Ex. 2 at SAP-OR00329587-88. Meyer conceded that the SAP executives never stated the number of the customers they thought might switch to SAP through Safe Passage. Lanier Decl. ¶ 123, Exs. 1, 22. Excerpts from Agassi Testimony: Meyer also relied on a few excerpts from Agassi’s trial testimony to conclude that SAP expected to capture 60% of PeopleSoft customers via TN, which he calculated as 5,952 customers. Lanier Decl. ¶ 131, Ex. 1. But Agassi testified that “60%” referred to PeopleSoft customers who “might be subject to conversion” at the time of acquisition, not the percentage he expected would convert. Lanier Decl. ¶ 132, Ex. 5. Meyer admitted that Agassi viewed converting the majority of PeopleSoft customers as “aspirational,” not an expectation. Lanier Decl. ¶ 133, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page38 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 33 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Further, Meyer’s calculation of Oracle’s expected financial gains, based on the same SAP documents used to calculate SAP’s purported expectations, requires even more speculation. Calculating Oracle’s expected financial gains by way of SAP’s internal documents necessarily comprises pure speculation. As both Ellison and Meyer admitted, there is not a shred of evidence that confirms Meyer’s conclusion that Oracle expected losses of between $1.4 and $2.7 billion for estimate or confirm that it reflected SAP’s reasonable expectations of actual results, and Meyer acknowledged that TN only ever had “10 or so” Siebel customers. Lanier Decl. ¶ 126, Ex. 1. Meyer also relied on an email from TN’s CEO Andrew Nelson, which noted the truism that every $1 of TN stand-alone revenue represented $18 of lost Oracle revenue, as establishing that SAP expected to impact Oracle in that amount. Lanier Decl. ¶¶ 129-30, Exs. 1, 25, 38. This proposition—advanced by TN, not SAP—simply illustrates that TN’s pricing was half of Oracle’s and, if one expected customers to stay on support for 9 years, then $1 of TN revenue in year one equals $18 lost to Oracle over 9 years. It provides no insight into the number of customers that SAP could reasonably expect to gain by virtue of TN that it would not have obtained otherwise. (iii) Meyer’s License Calculations Rested on Unjustified Assumptions. One must accept a series of unjustified assumptions to arrive at Meyer’s range of SAP’s expected financial gains, calculated based on SAP’s purported range of “expected” customers and revenues per customer, including that the documents on which he relied actually establish SAP’s expectations (they did not) and that they provide an objective basis upon which to value TN’s effect on the success of Safe Passage (they do not). Such assumptions are inappropriate to calculate a hypothetical license price. It is one thing for a company to make assumptions about future sales, establish goals, or even set forth expectations. But it is too big a leap to say that those assumptions and goals establish what a company would be willing to pay for a license. See Childress v. Taylor, 798 F. Supp. 983, 991-92 (S.D.N.Y. 1992) (rejecting as unduly speculative a hypothetical license claim where the plaintiff’s proposed royalty rate was “unsupported by the evidence,” as it required several “tenuous” assumptions, including that the infringed work (a play) would have been highly successful with another producer). Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page39 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 34 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Meyer’s methodology generated a huge range of supposed values for the PeopleSoft/JDE and Siebel licenses and improperly forced the jury to speculate on a hypothetical license amount. In selecting a fair market value for the PeopleSoft license from a range of $881 million to $2.69 billion, Meyer could not commit to a value; instead, he simply concluded that the value of the PeopleSoft license would be “at least $1.5 billion,” without any reasoned explanation. Lanier Decl. ¶¶ 136-38, Exs. 1, 38. Similarly, in choosing a fair market value for the Siebel license from a range of $97 million to $247 million, Meyer equivocated, arbitrarily pricing it “at least $100 the PeopleSoft/JDE license and $164 million for the Siebel license. Lanier Decl. ¶¶ 146, 150, Exs. 1, 38. Indeed, evidence contemporaneous with and post-dating the hypothetical negotiation shows that Oracle expected TN would have little to no negative impact on Oracle and, instead, that Oracle ridiculed the idea that TN could have significant impact. Lanier Decl. ¶¶ 141-45, Exs. 10-14. That expectation was reasonable because it is undisputed that customers do not switch products without a compelling business reason. Lanier Decl. ¶¶ 145, 147-49, Exs. 1, 2, 12. In the end, that both SAP and Oracle’s purported goals as to TN’s financial impact (i.e., 1,375 to 5,592 PeopleSoft customers and 200 Siebel customers), Lanier Decl. ¶¶ 134, 139, Exs. 1, 38) so exceeded TN’s actual impact (a total of 358 customers) exposes these “projections” as unreliable and unreasonable and itself justifies exclusion of Oracle’s hypothetical license claim. See Technologies, S.A., 460 F. Supp. 2d at 200-03 (holding that actual damages, whether in the form of a hypothetical license or lost profits, could not be based on evidence of projections shown to be “not realistic” and unreliable because, inter alia, the projections far outstripped the defendant’s actual profits and would have resulted in a “windfall” to the plaintiff). Here, even could the SAP documents credibly be described as containing projections, the Court should preclude Oracle’s recovery of a hypothetical license based on those “projections” because they were not projections of the value of TN’s use of the copyrighted works, because they have otherwise been proven to be unrealistic and speculative, and because recovery would result in a more than $1 billion windfall. See Leland Med. Ctrs., 2007 WL 2900599, at *6-8. (iv) The Wide Range of License Values Generated by Meyer’s Analysis Forced the Jury to Speculate. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page40 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 35 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) million.” Lanier Decl. ¶ 139, Exs. 1, 38. Courts do not hesitate to bar recovery of hypothetical licenses where a plaintiff cannot adequately explain the basis for its claim. See, e.g., Jamison Bus. Sys., Inc. v. Unique Software Support Corp., No. CV 02-4887 (ETB), 2005 U.S. Dist. LEXIS 45480, at *57-61 (E.D.N.Y. May 26, 2005) (precluding hypothetical license claim for $75,000 because plaintiff “did not explain how he arrived at this calculation”). As one court previously admonished Meyer, “plus or minus a guess is, after all, still a guess,” and such guesswork is insufficient to establish a non-speculative hypothetical license amount. See Lanier Decl. ¶ 103, Ex. 1. Given the great uncertainty underlying Meyer’s damages calculations, the Court should exclude Oracle’s hypothetical license claim. Cf. Polar Bear, 384 F.3d at 709 (hypothetical license fee appropriate “provided the amount is not based on undue speculation”). (d) Oracle’s Counsel Improperly Encouraged Speculation. Illustrative of the speculative nature of its claimed award, Oracle ultimately presented the jury with a spate of potential license valuations that had an almost six-fold range, from $897 million to $5 billion. Lanier Decl. ¶¶ 151-53, Exs. 1, 40. Rather than offering objective evidence to assist the jury in determining a fair market value for the license that even Oracle admitted its expert could not accurately quantify, Oracle’s counsel invited the jury to engage in guesswork and simply pick a number between $1.66 billion and $3 billion. See id. Such an invitation to speculate was inappropriate and warrants exclusion of Oracle’s damages claim. See R.S.E., Inc. v. Pennsy Supply, Inc., 523 F. Supp. 954, 970 (D.C. Pa. 1981). In R.S.E., the defendant moved under Rule 50(b), arguing that the plaintiff improperly “encouraged the jury to speculate rather than arrive at a reasoned decision regarding damages.” Id. at 970. Rather than offering a formula prepared by an expert for the jury to consider, plaintiff’s counsel instructed the jury: “if you think the profit percentages are too high, cut them down.” Id. In granting the motion, the court admonished that “[g]uesswork based upon the studies of an expert cannot form the foundation of a damage award” and found that “plaintiff failed to provide sufficient evidence to the jury that would provide it with the raw data it would need to arrive at a damage award without resorting to guesswork or speculation.” Id. at 970-71. As in R.S.E., Oracle failed to give the jury the objective tools it needed to determine a Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page41 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 36 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) hypothetical license, instead encouraging it to arbitrarily adjust Meyer’s damages figures. The most certainty Oracle could offer was that the appropriate damages amount “probably” fell within a more than billion dollar range. Lanier Decl. ¶ 153, Ex. 1. Oracle’s request not only underscored its haphazard approach to calculating damages, it flaunted it. 4. Oracle’s Database Damages Claim Suffers From Similar Deficiencies. The Court’s order on summary judgment made clear that to calculate a hypothetical license, “[t]he question is not what Oracle would have charged for a license, ‘but what is the fair market value.’” ECF No. 628 at 5. Ignoring the Court’s guidance, Oracle presented only one- sided testimony of what it “would have charged” for a license to use Oracle database software. The license calculation was not based on objective evidence—Oracle’s witnesses admitted that no evidence of a similar license exists—but rather on the unsupported speculation of a single Oracle employee, Richard Allison, parroted back as “expert” opinion by Meyer. (a) Oracle Presented No Objective Evidence of the Database License Price or Structure. Once again, Oracle offered no objective evidence, such as benchmark licenses, to calculate a reasonable price for the database software hypothetical license. Lanier Decl. ¶¶ 33, 157, Ex. 1. Absent such evidence, Oracle’s claim that it would have charged on a “per-customer” basis had no objective foundation and must be rejected as unduly speculative. Oracle offered no credible evidence that a database license has ever been priced on a “per- customer” basis, only Allison’s uncorroborated statements about outsourcers. Lanier Decl. ¶ 161, Ex. 1. Oracle does not structure its standard database license in this manner (Lanier Decl. ¶¶ 156, 160, Ex. 1), and there was no evidence that any party ever agreed to pay the price that Allison said he would have charged. The only evidence that the license should be priced per customer is Allison’s testimony that Oracle would have liked to charge the price resulting from that structure. Lanier Decl. ¶ 160, Ex. 1. (“Q. Now, what you then did is you made the assumption that Oracle would like to charge for this license on a per-customer basis; is that right? A. Correct.”). And Allison’s testimony confirmed that the per-customer structure is not necessary; specifically, Allison admitted that it is possible to support multiple customers on multiple databases using a Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page42 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 37 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) single instance of the database software, without the need for TN to license one copy of the software per customer. See id. Ultimately then, the structure and pricing of the hypothetical database license is too speculative to support a hypothetical license damages award. (b) Oracle Presented Only Subjective Testimony of What It Claims It Would Have Charged for a Database License. Oracle’s database hypothetical license calculation is insufficient because it is based entirely on Oracle’s post hoc, subjective view of the price it claims it would have charged, rather than on a fair market value. Jarvis, 486 F.3d at 534; Bruce v. Weekly World News, Inc., 310 F.3d 25, 29-30 (1st Cir. 2002) (rejecting per-use hypothetical license and expert’s contention that license “could be whatever we feel is fair”); Interplan Architects, 2010 U.S. Dist. LEXIS 114306, at *34-37 (dismissing claim based solely on plaintiff’s statement regarding price he would have charged to use work); Smith, 2006 U.S. Dist. LEXIS 27412, at *2-3 (same). Even Oracle acknowledged in motion practice that “any purportedly unreasonable or unsupported opinion of any Oracle executive . . . about what the license amount should be ‘has no place in this calculus.’” ECF No. 483 (Opp’n to Mot. for Summ. J.) at 5. Without objective evidence to ground Oracle’s self-serving calculation in reality, Allison’s subjective testimony is insufficient to permit a hypothetical license for TN’s use of the database software. See Jarvis, 486 F.3d at 534; Interplan, 2010 U.S. Dist. LEXIS 114306, at *34-37; Smith, 2006 U.S. Dist. LEXIS 27412, at *2-3. Moreover, this evidence provides, at best, only one half of the analysis necessary to price a hypothetical license. As discussed supra, the hypothetical license inquiry is an objective one— the result of a negotiation between a willing buyer and a willing seller. See Jarvis, 486 F.3d at 534; see also Polar Bear, 384 F.3d at 709; Mackie, 296 F.3d at 917. Oracle failed to analyze, much less substantiate, what TN, as a willing buyer, would have agreed to pay. Focusing only on what the willing seller would have demanded is particularly problematic here, where Oracle’s calculation (1) deviated significantly from the standard pricing structure, (2) did not arise from comparable benchmark licenses, and (3) ultimately resulted in an estimated licensing fee almost equivalent to TN’s revenues for its entire 7-year history. Lanier Decl. at ¶¶ 166, 170, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page43 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 38 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) In sum, the trial evidence demonstrates that the enormous “hypothetical license” fees that Oracle obtained from the jury were not “actual” damages. Instead, that award represented an unduly speculative, counterfactual, and punitive award that wholly failed to measure Oracle’s (c) Oracle’s Database License Calculation Similarly Was Based on Unjustified Assumptions. Finally, Oracle’s calculation of a database license is based entirely on subjective and speculative assumptions not tied to reality; thus, the Court should reject it as unduly speculative. See Jarvis, 486 F.3d at 534 (“[e]xcessively speculative” damage claims are to be rejected); Polar Bear, 384 F.3d at 709 (hypothetical lost license fee may be awarded “provided the amount is not based on undue speculation”); Mackie, 296 F.3d at 917; Childress, 798 F. Supp. at 991-92. First, the per-customer structure is based only on an unjustified assumption that Oracle would have charged on a per-customer basis, without any corroborating evidence. This structure is not even supported by logic, as it is analogous to demanding that a user buy a separate copy of Microsoft’s Excel software every time that user wanted to make an Excel spreadsheet for a different client. Second, Meyer’s use of 172 customers in his calculation was based on a speculative estimate of customers who “benefited” from the license, premised on an unjustified assumption that all TN HRMS customers benefited from the fixes and updates developed on TN’s versions of the database software, regardless of whether a customer’s environment ran on an Oracle database. Lanier Decl. ¶ 164, Exs. 1, 38. Meyer used this assumption to add one-third more customers to his per-customer calculus, significantly inflating the ultimate price. Third, Meyer assumed that Oracle would price each of the 172 per-customer licenses on the “average” size server TN used—a six-processor server. Lanier Decl. ¶¶ 162, 165, Ex. 1. But that assumption, based only on Allison’s say-so, leads to the absurd conclusion that TN would have purchased a license for 1,032 processors, when TN supported all of its customers, even those running other database software, with 27 processors. Lanier Decl. ¶ 169, Ex. 1. This extreme deviation from reality demonstrates the speculative nature of the license Oracle presented and belies a claim that any reasonable, willing buyer would have agreed to such an inflated fee. * * * Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page44 of 54 1 2 3 4 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 5 SVI-90062v1 - 39 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) actual damages—which, in fact, were in the form of lost profits, not lost license fees. Accordingly, this Court should grant Defendants’ motion for judgment as a matter of law that Oracle may not recover actual damages in the form of a hypothetical license, both because no license fees were lost and because Oracle’s claim was entirely, and inappropriately, speculative. VI. ARGUMENT: NEW TRIAL MOTION The reasons that justify granting judgment as a matter of law under Rule 50(b) also support Defendants’ motion for a new trial under Rule 59(a). To grant a new trial, however, the Court need only weigh the evidence for itself and determine that the verdict is against the weight of the evidence, that the damages are excessive, or that, all things considered, the verdict represents a miscarriage of justice. Fed. R. Civ. P. 59(a); see also Molski, 481 F.3d at 729. All of these conditions are present here. The $1.3 billion verdict wildly exceeds any actual harm to Oracle and is contrary to the clear weight of the evidence. This grossly excessive award inevitably resulted from Oracle presenting speculative, irrelevant, and prejudicial evidence. Mountains of liability evidence when liability was not in dispute improperly endorsed punishment through compensatory damages. Irrelevant evidence regarding the purported value of intellectual property as a whole (as opposed to the value of its use) allowed Oracle to present figure after figure in the billions, confusing and desensitizing the jury to the staggering damages claim. Wholly speculative evidence was given the imprimatur of expert opinion by Oracle’s disclosed and undisclosed “experts.” And in violation of this Court’s express admonition and well-established rules, Oracle’s closing argument not only described the conduct at issue as stealing but did so in a way that pandered to the juror employed by Best Buy. To avoid a miscarriage of justice, the Court should remit damages or grant a new trial. A. Oracle’s Improper Hypothetical License Theory Was Based upon Undue Speculation. As discussed above, Oracle’s claim to lost hypothetical license fees was based upon undue speculation and on “factors” and evidence that were simply insufficient to determine the value of TN’s actual use of the copyrighted works. Even were a hypothetical license fee recoverable as a Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page45 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 40 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) In closing, Oracle’s counsel also argued that in valuing the licenses, the jury should “not take facts” about what happened “in 2006, 2007, 2008 or 2009 and try to import them back into January [2005].” Lanier Decl. ¶ 24, Ex. 1. This was a rank invitation to speculate instead of ground its verdict in known, real-world facts, for such facts would not have supported Oracle’s massive damages claim. Oracle’s counsel’s directive also contradicted the Court’s instruction form of actual damages here, those damages still must reflect the actual use made by Defendants of the copyrighted aspects of the works, not simply “the highest use for which plaintiff might license.” On Davis, 246 F.3d at 166 n.5; see also Wall Data, 447 F.3d at 786. Defendants’ approach properly captured the harm to Oracle resulting from TN’s actual use of the copyrighted works by directly identifying which customers’ purchases from TN and SAP may have been the result of TN’s services. By contrast, Oracle’s approach deliberately ignored evidence of the value of use made by Defendants, as reflected in TN’s actual impact upon Oracle. Oracle never attempted to isolate the value of TN’s use of the works, much less the copyrightable elements of those works, from the assumed or hoped-for goals attributed to a broader Safe Passage marketing program. Instead, Oracle asked the jury, time after time, to overlook the fact that TN actually used the copyrighted works on behalf of only 358 customers, claiming that fact to be irrelevant. Lanier Decl. ¶¶ 4, 10, 24, Ex. 1. And, despite having testified in prior cases that damages experts should “be aware of what happens in the future to make sure that you get the proper result at the time of the hypothetical,” Meyer steadfastly refused to give any consideration or weight to the fact that TN only acquired 358 customers. Lanier Decl. ¶¶ 10, 18, Ex. 1. Rather, Meyer speculated that TN’s use of the copyrighted works would have garnered TN 3,000 or more customers based on SAP’s aspirations for its Safe Passage marketing program as a whole. Lanier Decl. ¶¶ 107-09, 135, Exs. 1, 20. He attributed the fact that TN only got 358 customers—instead of the speculative 3,000—entirely to SAP’s failure to execute on its business plans, a proposition for which there was no support whatsoever. Lanier Decl. ¶ 19, Ex. 1; see Childress, 798 F. Supp. at 991-92 (rejecting as speculative plaintiff’s request for higher royalty rate based on claim that commercial failure of infringing play was not due to play’s value, but to defendant’s failure to successfully execute business plans). Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page46 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 41 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) that the jury could consider facts after the hypothetical negotiation, see ECF No. 1005 at 9, as well as case law establishing that evidence of actual financial performance relating to sale of an infringing product is admissible as probative of the infringer’s anticipated profits; indeed, the Supreme Court described such evidence as “a book of wisdom that courts may not neglect.” Sinclair Ref. Co. v. Jenkins Petroleum Process Co., 289 U.S. 689, 698 (1933); see also Trans- World Mfg. Co. v. Al Nyman & Sons, Inc., 750 F.2d 1552, 1568 (Fed. Cir. 1984) (“Evidence of the infringer’s actual profits generally is admissible as probative of [the alleged infringer’s] anticipated profits”); Locklin v. Switzer Bros., Inc., 235 F. Supp. 904, 906-07 (N.D. Cal. 1964). Instead of focusing on the value of Defendants’ actual use of copyrighted works, Oracle and Meyer confused the jury with the fictitious and speculative “negotiation factors” approach that was inconsistent with both copyright and patent law and that grossly inflated Oracle’s damages claim. Despite describing his factors as “[c]onsistent with Georgia-Pacific Corp. v. U.S. Plywood Corp.” (Lanier Decl. ¶ 95, Exs. 1, 38), very few of Meyer’s factors resemble those used in Georgia-Pacific.8 See Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp 1116, 1119- 20 (S.D.N.Y. 1970). Rather, Meyer crafted his own factors to suit the speculative evidence he found in SAP’s documents regarding SAP’s purported hopes for Safe Passage.9 But none of the figures offered in support of these factors—all in the billions—measured a reasonable, factually- 8 Notwithstanding his purported reliance on Georgia-Pacific, Meyer’s approach plainly disregards recent Federal Circuit guidance making clear its intent to prevent abuse by scrutinizing hypothetical license awards to ensure they are supported by substantial evidence. Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1324-36 (Fed. Cir. 2009) (rejecting hypothetical license award as “speculation or guesswork” based on failure to present benchmark licenses or to show that actual use of infringing device supported a substantial lump-sum license fee), cert. denied, 130 S. Ct. 3324 (2010); ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 868 (Fed. Cir. 2010) (rejecting hypothetical license award as based on “speculative and unreliable evidence divorced from proof of economic harm linked to claimed invention . . .” and faulting expert for relying on irrelevant licenses to drive up hypothetical license fee); Wordtech Sys., Inc. v. Integrated Network Solutions, Inc., 609 F.3d 1308, 1318-22 (Fed. Cir. 2010) (rejecting hypothetical license award as “speculation” and “guesswork,” based on inapplicable license agreements and unsupported profit projections); Uniloc USA, Inc. v. Microsoft Corp., Nos. 2010-1035, 2010-1055, 2011 U.S. App. LEXIS 11, at * 59-66 (Fed. Cir. Jan. 4, 2011) (reiterating reasoning in Lucent and progeny that hypothetical license awards must be supported by evidence “tied to the relevant facts and circumstances of the particular case at issue . . . .” and that unrelated evidence “does not support compensation for infringement but punishes beyond the reach of the statute”). 9 This is not the first time Meyer has given a biased and unreliable opinion on the value of a hypothetical license; at trial, Meyer admitted that another court found that he had exaggerated a value-of-use claim and that it was too speculative to stand. Lanier Decl. ¶ 102, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page47 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 42 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) grounded value of TN’s actual use of the copyrighted works. See supra at V.B. In conjunction with Meyer’s flawed analysis, Oracle offered evidence of R&D costs and the price to acquire PeopleSoft and Siebel to support a “value of intellectual property” theory of measuring a hypothetical license that is contrary to the law that confused and prejudiced the jury. Oracle offered this theory through unreliable opinion testimony of Meyer and speculative, undisclosed “expert” testimony of executives, including Ellison, Catz, and Phillips.10 Using this legally improper theory, Oracle bombarded the jury with evidence of the $11.1 billion price it paid for PeopleSoft and the $6.1 billion price it paid for Siebel, as well with as the “billions” of dollars it invests in R&D. Oracle’s approach created confusion because such evidence conflated the overall value of acquired corporations (and their massive acquisition prices) with the value of TN’s actual use of intellectual property (the only thing that should have been at issue in the “hypothetical license” analysis). For example, Phillips testified: If I paid for something one day for 11 billion and my competitor wants it the next day, which -- they would have to pay close to what I paid for it. I had to pay for it up front 11 billion. They should have to pay billions to have access to it as well. Lanier Decl. ¶ 55, Ex. 1. Neither this simplistic analysis nor any other offered by Oracle values TN’s actual use. By focusing on the “something” that cost $11 billion—the entire PeopleSoft acquisition—instead of valuing the actual use of the protectable elements of the works infringed by TN, Oracle misled the jury as to how damages ought to be measured and desensitized it to Oracle’s billion-dollar claims. When Oracle’s Catz testified that “this whole business is a business about billions,” she epitomized Oracle’s effort to make this dispute an abstract and speculative exercise, instead of an objective valuation of TN’s impermissible, but limited, use of the copyrighted aspects of certain software works. Lanier Decl. ¶ 58, Ex. 1. The “whole business” of software may well be about billions, but the only business the jury had was to value Oracle’s actual damages from Defendants’ infringement—a different question entirely, and one whose proper answer would be a much more modest number than $1.3 billion. Cf. Uniloc, 2011 U.S. App. LEXIS 11, at *72-74 10 Defendants objected to this testimony at trial, in their Rule 702 motion to exclude Meyer, and motion in limine to exclude testimony from undisclosed experts. Lanier Decl. ¶¶ 46- 47, Ex. 1; ECF No. 798 (Mot. to Exclude Meyer); ECF No. 728 (Defs.’ Mots. in Limine) at 10-11. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page48 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 43 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) (upholding the lower court’s grant of new trial based on improper presentation of billion dollar figures that were irrelevant to the royalty calculations). In short, improper speculation, not damages evidence grounded in reality, caused this excessive verdict. B. The Award Far Exceeds Oracle’s Actual Harm. The undue speculation reflected by the jury’s verdict is confirmed by the fact that the $1.3 billion verdict far exceeds the actual harm Oracle suffered. Thus, on its face, that excessive verdict warrants a new trial or remittitur. See Alvarado v. Federal Express Corp., No. C 04-0098 SI, 2008 U.S. Dist. LEXIS 21238, at *10-12 (N.D. Cal. Mar. 18, 2008); Informatica Corp. v. Business Objects Data Integration, Inc., No. C 02-03378 EDL, 2007 WL 2344962, at *4 (N.D. Cal. Aug. 16, 2007). The jury’s award was based on Oracle’s speculation purporting to show that SAP “expected” that a majority of the 14,000 PeopleSoft and Siebel customers would leave Oracle for TN support, and then replace their Oracle software with SAP software. Relying on SAP documents setting forth mere “goals” and “assumptions” relating to a broader Safe Passage program, and ignoring evidence of what actually occurred, Meyer assumed that SAP believed that as a result of TN alone it would convert thousands of PeopleSoft and Siebel customers and obtain revenues up to billions of dollars. Lanier Decl. ¶¶ 105-06, 134-36, Exs. 1, 20, 38. This speculation was the foundation for Meyer’s more than billion-dollar-plus license calculations. In reality, TN’s modest success in attracting customers away from Oracle, not grand speculation about what SAP might have hoped to accomplish by acquiring TN or its broader Safe Passage program, reflects Oracle’s actual harm. Lanier Decl. ¶¶ 3, 5, 9, Ex. 1. Indeed, after three and a half years of litigation, with massive discovery, Oracle could not present any evidence that even one of these 358 customers purchased SAP software because of TN’s infringement. Defendants, by contrast, presented evidence that TN customers would have left Oracle anyway and that TN played little or no part in customers’ decisions to buy SAP products. Lanier Decl. ¶¶ 6-8, 20-21, Exs. 1, 6-8. Applying this reality-based approach, SAP expert Clarke calculated lost profits and infringer’s profits totaling $28 million. Lanier Decl. ¶¶ 2, 22, Exs. 1, 15. Even Oracle expert Meyer’s lost profits and infringer’s profits calculation totaled a maximum of $408.7 million—less than one third of the jury’s $1.3 billion award. Lanier Decl. ¶¶ 2, 11, Exs. 1, 15. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page49 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 44 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) And Meyer’s calculation was necessarily inflated—he performed almost no analysis to determine whether these lost profits and infringer’s profits were actually caused by TN’s use of the works.11 The $1.3 billion award shocks the conscience, not only because it far exceeds actual harm —as calculated by both experts in the form of lost profits and infringer’s profits—but also because it is against the clear weight of the evidence. The award purportedly represents what a prudent willing buyer and seller would have agreed was a reasonable fee to use the copyrighted works as part of a marketing plan to persuade Oracle customers to migrate to SAP. Meyer based his billion-dollar-plus license calculations on his conclusion that both SAP and Oracle expected TN to impact their businesses on the order of billions of dollars, such that the parties reasonably would have agreed to a license price of at least $1.6 billion prior to any infringement. In the end, not only did Oracle fail to present evidence actually establishing SAP’s valuation of TN’s impact, but the evidence showed, overwhelmingly, that Oracle itself did not expect customers to migrate to SAP because of TN. Lanier Decl. ¶¶ 141-49, Exs. 1, 2, 10-14. Oracle’s sole response to this argument—that it would not have dismissed TN’s potential impact had it known about the infringement—is nonsensical. The very premise of a hypothetical license negotiation is that no infringement has yet occurred and that the parties are in good faith attempting to agree upon reasonable compensation for the permitted use of the copyrighted work going forward. Lanier Decl. ¶¶ 73-74, Ex. 1. Whether Oracle knew of TN’s infringement is irrelevant to the hypothetical negotiation. In any event, Meyer admitted that Oracle knew (1) that TN was owned by SAP, (2) that TN was offering as good or better service than Oracle at half price, and (3) the impact TN was having in the marketplace. Lanier Decl. ¶ 17, Ex. 1. Yet, even with that knowledge, Oracle did not perceive TN as a significant threat. Knowledge that TN was using Oracle’s software would not have made the threat greater, as customers still would have received the same opportunity to obtain service at half price.12 11 Even these amounts exceed actual harm because they include profits Defendants gained as a result of infringement. Infringer’s profits do not represent actual damages, but the copyright law provides for their disgorgement as a disincentive to infringe. Polar Bear, 384 F.3d at 708. 12 Clarke testified that whether or not Oracle knew how TN was providing service, Oracle knew what TN was doing in the marketplace, so that would not have “affected what they thought about the—the success of the program.” Lanier Decl. ¶ 23, Ex. 1. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page50 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 45 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) In sum, even had Oracle suffered harm in the form of lost license fees, the jury’s award greatly surpassed any rational quantification of fees for a non-exclusive, non-transferable, license of limited purpose and duration to use the copyrightable elements of the software. Instead, as addressed supra, the award improperly included the value of non-protectable elements of the infringed works, the price to acquire whole companies, and the cost to develop unrelated products. The award was speculative and excessive and should be set aside. C. The Award Was Infected by Oracle’s Presentation of Irrelevant and Prejudicial Liability Evidence. The jury’s grossly excessive verdict in large part is attributable to Oracle’s continuous presentation of irrelevant and prejudicial liability evidence that served only to inflame the jury. A new trial, or remittitur, is necessary to redress the resulting unfairness. Evidence regarding SAP’s liability for stipulated contributory copyright infringement and TN’s liability for other stipulated claims had no place at trial. Although Oracle led the Court to believe that liability evidence for stipulated claims supported Oracle’s license calculation, such evidence does not, as a matter of law, properly factor into a hypothetical license analysis. See supra at V.B.3.b. Moreover, the daily barrage of liability evidence and argument that Oracle offered went far beyond what could legitimately be considered “context,” particularly in light of the Court’s guidance that “at some point, it becomes cumulative particularly given that liability’s no longer at issue in this case.” Lanier Decl. ¶ 77, Ex. 1. With regard to the stipulated contributory infringement claim, Oracle presented extensive documentary and testimonial evidence purporting to show that SAP knew of TN’s infringing activities from the outset and hid behind a so-called “liability shield.” Lanier Decl. ¶¶ 65-71, Exs. 1, 3, 19, 24, 28-29. Bombarding the jury with evidence of SAP’s supposed knowledge and risk acceptance was unnecessary for context to resolve the sole issue in this case—appropriate compensation for the harm Oracle incurred as a result of TN’s actual use of the copyrighted works. With regard to the stipulated claims against TN for which Oracle no longer sought damages, Oracle prominently featured in its case testimony from various witnesses about supposed corruption of data, purported fraudulent access to Oracle websites, and alleged crashes Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page51 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 46 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) Defendants also suffered prejudice from Oracle’s violations of the Court’s prohibition on the use of words with criminal overtones, such as “theft” and “steal,” to describe the copyright infringement. Lanier Decl. ¶ 90, Ex. 1. Although the Court sustained Defendants’ objection, Oracle flouted that ruling by presenting evidence and argument regarding stealing and theft, including a reference to stealing from one juror’s employer—Best Buy—which served only to to Oracle’s computers. Lanier Decl. ¶¶ 78-81, Exs. 1, 9. While both sides agreed that the parties could present evidence relating to these stipulated claims as “background or context,” it strains credibility to describe Oracle’s onslaught of liability evidence as merely contextual; in reality, such evidence served as the foundation for Oracle’s exaggerated damages claim. That the parties agreed not to object at trial to otherwise relevant evidence solely because it related to the stipulated claims did not give Oracle free license to bolster its speculative hypothetical license claim with excessive, inflammatory, and irrelevant liability evidence. See id. By abusing the opportunity to present liability evidence when only damages were at issue, Oracle unfairly skewed the verdict. This liability evidence was confusing because it conflated issues of liability and concepts of punishment with calculating damages under copyright law, which does not permit punitive damages. See Stehrenberger, 335 F. Supp. 2d at 468 (holding willful infringement evidence not admissible to price hypothetical license). Further, this evidence was prejudicial because, having been offered evidence of Defendants’ alleged willfulness, the jury’s view of Defendants was inevitably tainted without Oracle having established objective, fair market value. This prejudice was only exacerbated when Defendants’ witnesses were prevented from testifying about steps that SAP took to mitigate risk of infringement. Lanier Decl. ¶¶ 82-86, Ex. 1. The exclusion of mitigating evidence—and Oracle’s subsequent argument to the jury that it should draw a negative inference from the absence of such evidence—further inflamed the jury and placed Defendants at an unfair disadvantage. Lanier Decl. ¶ 87, Ex. 1. The result of this onslaught of liability evidence was a verdict that so far outstripped the record evidence of actual harm to Oracle that it can only be described as a punitive damages award, which cannot stand under copyright law. See Oboler v. Goldin, 714 F.2d 211, 213 (2d Cir. 1983) (“[P]unitive damages are not available under the Copyright Act of 1976.”); Stehrenberger, 335 F. Supp. 2d at 468. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page52 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 SVI-90062v1 - 47 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) further inflame the jury. Lanier Decl. ¶¶ 91-93, Ex. 1. This evidence included Ms. Catz’s analogies to stealing a $2,000 watch and selling it for $20 and to using a $15 crowbar to break into and “clean out” the $10 million contents of a house. Lanier Decl. ¶¶ 88-89, Ex. 1. D. The Court Should Remit Damages to $28 Million or No More Than $408.7 Million. The maximum amount of lost profits and infringer’s profits supported by the evidence—to which the Court should remit damages—is Clarke’s $28 million. In no event, however, could remitted damages be higher than Meyer’s $408.7 million. When a court determines that a damages award is excessive, it may order a remittitur, reducing the damages award to “the maximum amount sustainable by the proof.” D & S Redi-Mix v. Sierra Redi-Mix & Contracting Co., 692 F.2d 1245, 1249 (9th Cir. 1982) (affirming remittitur); see also Anglo-American Gen. Agents v. Jackson Nat’l Life Ins. Co., 83 F.R.D. 41, 45 (N.D. Cal. 1979). Both Meyer and Clarke offered opinions on Oracle’s actual lost support profits and Defendants’ infringer’s profits resulting from TN’s use of the copyrighted works. Meyer calculated lost profits of $120.7 million and infringer’s profits of $288 million, totaling $408.7 million. Lanier Decl. ¶¶ 2, 11, Exs. 1, 15. He also presented alternative calculations of $36 million of lost profits and $236 million of infringer’s profits, totaling $272 million. Lanier Decl. ¶¶ 14-15, Ex. 1. Clarke calculated Oracle’s lost profits of $19.3 million and SAP’s infringer’s profits of $8.7 million, totaling $28 million. Lanier Decl. ¶¶ 2, 22, Exs. 1, 15. The two main differences between Meyer’s and Clarke’s calculations show why Clarke’s calculation represents the maximum amount sustained by the evidence. First, Meyer’s $120 million projected lost profits through to 2015 purportedly “to reflect the ongoing impact.” Lanier Decl. ¶ 13, Ex. 1. By contrast, Clarke’s approach appropriately cut off damages as of TN’s cessation of operations (and hence the end of infringement) in late 2008. See id.; see also Frank Music, 772 F.2d at 512, 513 n.6 (noting that hypothetical license award must “approximate what a reasonable market price would have been at the time of the infringement”) (emphasis added); Jarvis, 486 F.3d at 534 (same). In his alternative $36 million lost profits calculation, Meyer also cut off damages as of TN’s wind-down. Lanier Decl. ¶ 14, Ex. 1. Second, relying on ample Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page53 of 54 1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 4 SVI-90062v1 - 48 - DEFS.’ RENEWED MOTION FOR JMOL AND NEW TRIAL MOTION Case No. 07-CV-1658 PJH (EDL) customer-specific evidence, Clarke properly excluded more customers on causation grounds, based on his findings that the customers would have left Oracle or would have purchased from SAP even without TN’s infringing activities. Lanier Decl. ¶¶ 16, 20-21, Ex. 1. VII. CONCLUSION It is the duty of this Court to intervene when a jury awards “damages so high as to require correction.” Honda Motor Co. v. Oberg, 512 U.S. 415, 424 (1994). This jury did so here. Accordingly, this Court should grant Defendants’ motion for judgment as a matter of law that Oracle may not recover actual damages in the form of a hypothetical license fee; grant Defendants’ new trial motion that jury’s verdict is grossly excessive, against the weight of the evidence, and resulted in a miscarriage of justice; and order a new trial or remittitur. Dated: February 23, 2011 JONES DAY By: /s/ Tharan Gregory Lanier Tharan Gregory Lanier Counsel for Defendants SAP AG, SAP AMERICA, INC., and TOMORROWNOW, INC. Case4:07-cv-01658-PJH Document1044 Filed02/23/11 Page54 of 54