Summary
holding that " jury could ... easily conclude that a reasonable investor would ... view" a struggling company's CEO's "misstatements/omissions regarding cash flow as significantly altering the ‘total mix’ of information available"
Summary of this case from Rex & Roberta Ling Living Tr. v. B Commc'ns Ltd.Opinion
No. 02 Civ. 05571 (RJH) (HBP).
February 17, 2011.
MEMORANDUM OPINION AND ORDER
TABLE OF CONTENTS
Morrison v. National Australia Morrison Morrison
BACKGROUND ........................................................................................ 2 DISCUSSION ........................................................................................ 9 I. Impact of Bank on Plaintiffs' Claims ................ 9 A. Overview of ...................................................................... 9 B. Applied ......................................................................... 13 C. Plaintiffs' Motion to Conform the Pleadings .............................................. 25 II. Vivendi's Motion for Judgment as a Matter of Law Pursuant to Rule 50 ................ 25 A. Standard of Review ....................................................................... 26 B. Material Misstatements and Omissions ..................................................... 28 C. Scienter ................................................................................. 42 1. Statements Not Specifically Attributable to Messier or Hannezo ....................... 43 2. Impact of the Messier and Hannezo Verdicts ........................................... 48 3. Whether Vivendi is Entitled to a New Trial Based on the Alleged Inconsistency in the Verdict .............................................................. 55 (a) Vivendi Waived Its Right to Object to the Verdict on Inconsistency Grounds ................................................................................. 56 (b) The Verdict is Not Inconsistent ................................................... 58 4. Loss Causation ....................................................................... 63 (a) Connection between the Fraud and the Events ....................................... 65 (b) Connection between the Events and Share Price Declines ............................ 73 (c) Whether the Misstatements Caused Inflation ........................................ 75 5. Effect of Jury's Finding of Zero Inflation on Certain Dates during the Class Period .................................................................................... 79 6. Forward-Looking Statements ............................................................. 86 (a) The Challeged Statements Do Not Fall Within the PSLRA'S "Safe Harbor" .................................................................................. 87 (b) Waiver ............................................................................... 91 7. Puffery .............................................................................. 94 III. Vivendi's Motion for a New Trial Pursuant to Rule 59 ................................ 96 A. Standard of Review ....................................................................... 97 B. Compromise Verdict ....................................................................... 98 C. Alleged Failure to Identify Misstatements until the Close of Evidence ................... 104 D. Verdict Form ............................................................................ 108 E. Plaintiffs' Summation ................................................................... 111 IV. Class Plaintiffs' Motion for Entry of Final Judgment ...................................... 114 CONCLUSION ...................................................................................... 121This is a securities fraud class action brought on behalf of shareholders of a French company, Vivendi Universal, S.A. ("Vivendi") against Vivendi and its former Chief Executive Officer, Jean-Marie Messier and its former Chief Financial Officer, Guillaume Hannezo (collectively, "defendants"). The action was tried before a jury from October 2009 to January 2010. At the close of plaintiffs' case, all three defendants moved for judgment as a matter of law pursuant to Federal Rule of Civil Procedure 50(a). The Court reserved decision on most aspects of defendants' Rule 50(a) motions and the case was submitted to the jury. On January 29, 2010, the jury returned its verdict. The jury found that Vivendi had violated Section 10(b) of the Securities Exchange Act of 1934 (the "Exchange Act") and the Securities and Exchange Commission's Rule 10b-5 (collectively, "Section 10(b)"), but that neither Messier nor Hannezo had committed a primary or secondary violation of Section 10(b) or Section 20(a) of that Act. No judgment has yet been entered on the verdict.
The Court granted Hannezo's Rule 50(a) motion in part, finding that in light of the evidence adduced at trial, Hannezo could not be found primarily liable for statements made by Vivendi that were not publicly attributed to him. Wright v. Ernst Young LLP, 152 F.3d 169, 174-75 (2d Cir. 1998); see also Lattanzio v. Deloitte Touche LLP, 476 F.3d 147, 155 (2d Cir. 2007) (holding that an accounting firm could not be held liable for reviewing another company's financial statements and reaffirming Wright, 152 F.3d 169). But see Pacific Inv. Mgmt. Co. LLC v. Mayer Brown LLP, 603 F.3d 144, 154-55, 158 n. 6 (2d Cir. 2010) (leaving open the question of whether attribution was required for primary corporate actors); In re Scholastic Corp. Sec. Litig., 252 F.3d 63, 75-76 (2d Cir. 2001). That ruling has not been challenged posttrial, and is not at issue in the pending motions. In all other respects, the Court reserved decision on defendants' motions for judgment as a matter of law.
Vivendi now renews its motion for judgment as a matter of law pursuant to Federal Rule of Civil Procedure 50(b), or, in the alternative, moves for a new trial pursuant to Federal Rule of Civil Procedure 59. Plaintiffs move for the entry of judgment, for an award of pre-judgment interest, and for approval of their proposal for post-verdict class notice and claims administration. This opinion sets forth the Court's ruling on these motions. It also addresses the impact of the Supreme Court's recent decision in Morrison v. National Australia Bank Ltd., 130 S.Ct. 2869 (2010) on the action and modifies its class certification in light of that decision.
BACKGROUND
This action was originally brought in 2002 by U.S. and foreign shareholders of Vivendi who alleged that they purchased ordinary shares, or American Depository Receipts that represent those shares (hereinafter, "ADRs"), at artificially inflated prices as a result of defendants' material misrepresentations and omissions between October 30, 2000 and August 14, 2002, inclusive (the "Class Period"), in violation of §§ 10(b) and 20(a) of the Exchange Act. 15 U.S.C. §§ 78j(b) and 78t(a). The ordinary shares in question traded primarily on the Paris Bourse, and did not trade on any U.S exchange. The ADRs were listed and traded on the New York Stock Exchange ("NYSE"). After the initial class action complaint was filed, a large number of related actions were filed and were consolidated by the Court into a single action, and a consolidated class action complaint was filed.
An American Depository Share is a security that represents an ownership interest in a specified number of a company's ordinary shares. Pinker v. Roche Holdings Ltd., 292 F.3d 361 (3d Cir. 2002); Exchange Act Release No. 29226, 1991 SEC LEXIS 936 (hereinafter "SEC Release"), at *3-4 n. 5 (May 23, 2001). An American Depository Receipt ("ADR") is a physical certificate that evidences ADSs (in much that same way that a stock certificate evidences shares of stock). Exchange Act Release No. 29226, 1991 SEC LEXIS 936, at *3 n. 5) For the sake of simplicity, this opinion uses the term "ADR" to refer to either the physical certificate or to the security evidenced by such certificate. ADRs are generally issued by a United States bank, and the underlying ordinary shares are typically ordinary shares of a foreign issuer, which are deposited with a foreign bank or other custodian and segregated from the other ordinary shares issued by the foreign issuer. An ADR holder can generally exchange his or her ADRs for the underlying shares at any time. Id. at *3-4; Greene, et al., U.S. REGULATION OF THE INTERNATIONAL SECURITIES AND DERIVATIVES MARKET § 2.02[1] at 2-19 (9th ed. 2009).
In February 2003, defendants moved to dismiss on various grounds. Of particular relevance here, defendants argued that this Court lacked subject matter jurisdiction over any claims brought by "foreign-cubed" class members — i.e., foreign shareholders who purchased their shares of Vivendi, a foreign company, on foreign exchanges. The Court, in a decision by Judge Baer, rejected that argument. In re Vivendi Universal, S.A. Sec. Litig., 381 F. Supp. 2d 158, 169-70 (S.D.N.Y. 2003) (" Vivendi I"). Judge Baer held that the Court had subject matter jurisdiction over those claims under the "conduct test" which was one of two tests — the conduct and effects tests — long in use in the Second Circuit to determine whether a court could exercise subject matter jurisdiction over foreign securities transactions. Id., see also Morrison v. Nat'l Austl. Bank Ltd., 547 F.3d 171 (2d Cir. 2008). Under the "conduct test," subject matter jurisdiction existed "if activities in this country were more than merely preparatory to a fraud and culpable acts or omissions occurring here directly caused losses to investors abroad." Morrison, 547 F.3d at 171 (citations omitted). The determination as to whether American activities "directly" caused losses to foreigners was fact-specific, and "depend[ed] on what and how much was done in the United States and on what and how much was done abroad." Id. (citation omitted).
The "effects test" — which has never been in play in this case — asked "whether the wrongful conduct had a substantial effect in the United States or upon United States citizens." See Morrison, 547 F.3d at 170. As discussed below, both of these tests were abrogated by the Supreme Court in Morrison, 130 S.Ct. 2869.
Applying that test, Judge Baer concluded that the Court had subject matter jurisdiction over claims by foreign plaintiffs who purchased Vivendi shares on foreign exchanges, in large part due to the fact that Messrs. Messier and Hannezo had engaged in significant conduct in the United States related to the alleged fraud — in particular, they moved their headquarters to New York and split their time between the U.S. and France during the crucial time period in which investors claimed to have been misled. Vivendi I, 81 F. Supp. 2d at 169-70. Judge Baer's decision on subject matter jurisdiction was affirmed by this Court on reconsideration. In re Vivendi Universal, S.A. Sec. Litig., No. 02 Civ. 5571 (RJH), 2004 WL 2375830, at *3-7 (S.D.N.Y. Oct. 22, 2004) (" Vivendi II").
After defendants' motion to dismiss was denied, and after some discovery had occurred, plaintiffs moved to certify a class consisting of Vivendi shareholders from the United States and various European countries. Defendants raised numerous objections to class certification, including, most prominently, an objection to the inclusion of foreign shareholders in the class. This Court considered the requirements for class certification set forth in Federal Rule of Civil Procedure 23(b)(3) and on May 21, 2007, certified a single class consisting of "all persons from the United States, France, England, and the Netherlands who purchased or otherwise acquired ordinary shares or American Depositary Shares of Vivendi Universal, S.A. between October 30, 2000 and August 14, 2002." In re Vivendi Universal, S.A. Sec. Litig., 242 F.R.D. 76, 109 (S.D.N.Y. 2007) (" Vivendi III"). Defendants filed a motion for partial reconsideration of the class certification opinion, which was denied on March 31, 2009. In re Vivendi Universal, S.A. Sec. Litig., No. 02 Civ. 5571 (RJH), 2009 WL 855799 (S.D.N.Y. Mar. 31, 2009) (" Vivendi IV").
In the wake of the Court's class certification opinion, defendants have traditionally distinguished four categories of plaintiffs (1) the plaintiffs who were included in the class certified by this Court on May 21, 2007 ("Class Plaintiffs"); (2) Liberty Media Corp., LMC Capital LLC, Liberty Programming Co. LLC, LMC USA VI, Inc., LMC USA VII, Inc., LMC USA VIII, Inc., LMC USA X, Inc., Liberty HSN LLC Holdings, Inc., and Liberty Media International, Inc. (collectively "Liberty Media"); (3) GAMCO Investors, Inc. ("GAMCO"); and (4) those plaintiffs who brought separate actions after being excluded from the class (the "Individual Plaintiffs"). This opinion relates solely to the claims of Class Plaintiffs. For the sake of convenience, this opinion will use the terms "plaintiffs" and "Class Plaintiffs" interchangeably to refer to Class Plaintiffs.
Meanwhile, the parties had been conducting fact and expert discovery. In August 2008, defendants filed various motions for full or partial summary judgment on numerous grounds. A principle contention was that plaintiffs allegedly had failed to prove loss causation, a required element of plaintiffs' claims. The Court carefully considered defendants' arguments, including those relating to loss causation, and denied defendants' motion. In re Vivendi Universal S.A. Sec. Litig., 634 F. Supp. 2d 352 (S.D.N.Y. 2009) (" Vivendi V"). Thereafter, Class Plaintiffs' claims were set down for trial in the fall of 2009.
On October 5, 2009, a jury trial began in this Court on Class Plaintiffs' claims that Vivendi, Mr. Messier, and Mr. Hannezo violated Section 10(b), and that Mr. Messier and Mr. Hannezo also violated Section 20(a). During plaintiffs' direct case, the jury saw videotaped deposition testimony from over twenty fact witnesses, including employees and senior executives at Vivendi and its subsidiaries, former members of Vivendi's Board, and employees of the rating agencies whose responsibility it was to cover Vivendi. Four fact witnesses also testified live: Anne Brassens, a former member of Vivendi's finance department who left the company in May 2002; Marie-Josée Kravis, a former member of Vivendi's Board, who served on the Board throughout the Class Period; Hannezo; and Gerard Morel, who was then serving as a class representative. Three experts testified for plaintiffs: Andrew Mintzer testified regarding generally accepted accounting principles ("GAAP") in the United States; Xavier Oustalniol testified regarding French GAAP, and Dr. Blaine Nye testified regarding loss causation and damages. Plaintiffs also introduced over five hundred documents into the record as part of their direct case. Plaintiffs rested their case on November 20, 2009.
In an opinion dated November 19, 2009, this Court relieved Mr. Morel of his responsibilities as class representative in the interests of comity after Vivendi brought a lawsuit in France seeking to enjoin Mr. Morel from participating in this action. In re Vivendi Universal, S.A. Sec. Litig., No. 02 Civ. 5571 (RJH), 2009 WL 3859066, at *7 (S.D.N.Y. Nov. 19, 2009).
Defendants then presented their case to the jury. Three fact witnesses testified for defendants: Messier; Hubert Dupont-L'Hôtelain, Vivendi's Treasurer during the Class Period and through the time of the trial; and Pierre Trotot, the Senior Executive Vice President and Chief Financial Officer of Cegetel, a Vivendi affiliate, during the Class Period. Six expert witnesses testified for defendants: Christine Hammer testified regarding Vivendi's cash balances during the Class Period; Andrew Fleming testified regarding Vivendi's liquidity situation during the Class Period; James Parrish testified regarding the work of rating agencies in general and regarding communication flow between Vivendi and various rating agencies; James Milner testified regarding French and U.S. GAAP; Professor Ronald Gilson testified regarding whether the risks alleged by plaintiffs to have been concealed by defendants were known to the market during the Class Period; and Dr. William Silber testified regarding potential damages. Defendants introduced over 250 additional documents into the record and rested on December 17, 2009.
Plaintiffs put on a brief rebuttal case, after which the jury was given a recess for the holiday season while counsel and the Court finalized the jury charges and the verdict form.
In early January 2010 the jury returned to hear extended closing arguments and begin its deliberations. The Court provided the jury with a copy of its instructions to review as needed during deliberations and a Verdict Form to complete. To aid in their deliberations, the jury was provided with a list prepared jointly by all counsel which identified all the documents admitted at trial (the "Exhibit List"). The Exhibit List also
identified which defendants each document was admitted against, and stated whether the document was subject to any limiting instructions
The seventy-two page Verdict Form identified fifty-seven sets of statements alleged by plaintiffs to have violated Section 10(b), each of which were set forth in Table A. Certain of those statements were alleged to be Section 10(b) violations by Vivendi only, and others were alleged to be Section 10(b) violations by Vivendi and Messier and/or Hannezo. The Verdict Form asked the jury to determine whether plaintiffs had proven the elements of their Section 10(b) claim with respect to each of the fifty-seven statements for each defendant against whom that false statement was alleged. For any statement as to which the jury found a Section 10(b) violation, the Verdict Form instructed to jury to determine whether the defendant(s) who had committed the violation had acted knowingly or recklessly. The Verdict Form also instructed the jury that if they found Section 10(b) liability as to any defendant with respect to any (or all) of the fifty-seven statements, they should identify the daily inflation amount (in euros/dollars per share), if any, in the price of Vivendi's ordinary shares and ADRs that they found to have been caused by the Section 10(b) violations. Consistent with 15 U.S.C. § 78u-4(f) of the Private Securities Litigation Reform Act ("PSLRA"), the Verdict Form also asked the jury to determine the percentage of responsibility to assign to each defendant whom the jury found to have violated Section 10(b), to the extent the jury found that any of the defendants had acted recklessly with respect to any of the fifty-seven statements. Lastly, the Verdict Form asked the jury to determine whether Messier and Hannezo had violated Section 20(a) of the Exchange Act.
The statements on Table A that were alleged to be Section 10(b) violations by Vivendi and Messier and/or Hannezo were those statements that were publicly attributed to Messier and/or Hannezo, such that Messier and/or Hannezo could be primarily liable for them. See cases cited in footnote 1, supra.
After fourteen days of deliberations, the jury reached its verdict. The jury found that Vivendi had violated Section 10(b) as to all fifty-seven alleged misstatements, and that it acted recklessly with respect to each statement. The jury found that Messier and Hannezo had not violated Section 10(b) or Section 20(a), thereby absolving Messier and Hannezo of liability. The jury found the daily inflation in the price of Vivendi's ordinary shares and ADRs' to be approximately half of the daily inflation amount that Dr. Nye had calculated on most days in the Class Period. However, the jury found that the inflation in Vivendi's stock prices was zero from September 11, 2001 to September 28, 2001, and also on those days between November 2001 and August 14, 2002 on which Vivendi's ordinary shares traded but Vivendi's ADRs' did not (or vice versa). Consistent with their finding that Messier and Hannezo had not violated the securities laws, the jury found Vivendi 100% responsible for plaintiffs' losses. Once the verdict had been read, the jurors were polled in open court, and they each affirmed the verdict
As noted, Vivendi's ordinary shares were traded in France and its ADRs' were traded in the United States. Because France and the United States have different public holidays, on certain days in the Class Period the stock markets were closed in France but not the United States (or vice versa).
Immediately after the Court had discharged the jury, Vivendi orally renewed its motion for judgment as a matter of law and also moved for a new trial. The Court approved an extended briefing schedule under which briefing on Vivendi's post-trial motions would be completed by mid-June 2010. Thereafter, plaintiffs moved for the entry of judgment, for an award of prejudgment interest, and for approval of their proposed class notice and claims administration procedures.
On June 24, 2010, the Supreme Court issued its opinion in Morrison v. National Australia Bank, 130 S.Ct. 2869 (2010), holding that Section 10(b) does not apply extraterritorially. In the wake of Morrison, the Court asked Vivendi and Class Plaintiffs to submit supplemental briefs addressing the impact of Morrison on the pending motions, and seeking such other relief as might be appropriate in light of Morrison. The parties submitted simultaneous briefs on this issue on July 16, 2010. Oral argument was held on all pending motions on July 26, 2010.
DISCUSSION
I. Impact of Morrison v. National Australia Bank on Plaintiffs' Claims
Vivendi and plaintiffs disagree over the impact of the Supreme Court's recent decision in Morrison on this action. Plaintiffs contend that Morrison has no impact on this case because all of Vivendi's ordinary shares are listed on the New York Stock Exchange in connection with its ADR program and, therefore, ordinary share purchasers satisfy Morrison's bright line test that limits Section 10(b) claims to "securities listed on domestic exchanges, and domestic transactions in other securities. . . ." Morrison, 130 S.Ct. at 2884. Vivendi points out that while some ordinary shares were listed on the NYSE they were not listed for trading purposes and served only as backup to the ADRs that were traded domestically. Moreover, actual transactions in Vivendi ordinary shares only took place on foreign exchanges, such as the Bourse, on which the shares were listed for trading. If anything is clear, Vivendi argues, it is that Morrison excludes from Section 10(b) "transactions conducted upon foreign exchanges," id. at 2882 (emphasis deleted), because the Exchange Act was not "intended to `regulate' foreign securities exchanges," id. at 2884 (emphasis deleted).
A. Overview of Morrison
Morrison was a case brought by Australian citizens who purchased ordinary shares of an Australian bank, National Australian Bank ("NAB") on foreign exchanges. NAB's ordinary shares were traded on the Australian stock exchange and other foreign exchanges, but not on any exchange in the United States. 130 S.Ct. at 2875. However, NAB did list ADRs, which represent the right to receive a specific number of NAB ordinary shares, on the New York Stock Exchange. Id. The plaintiffs in Morrison sought to bring claims against NAB in the United States under Section 10(b), alleging that a Florida-based subsidiary of NAB had falsified financial data, which was then forwarded to and disseminated by NAB as part of its public filings.
The defendants moved to dismiss for lack of subject matter jurisdiction and for failure to state a claim under Rule 12(b)(6). The district court granted the motion to dismiss for lack of subject matter jurisdiction, In re National Australia Bank Sec. Litig., No. 03 Civ. 6537 (BSJ), 2006 WL 3844465, at *8 (S.D.N.Y. Oct. 25, 2006), and the Second Circuit affirmed, Morrison, 547 F.3d at 176. The Second Circuit reasoned that under the circuit's "conduct" test for assessing subject matter jurisdiction, the court lacked jurisdiction over the claims in question because the actions taken at NAB in Australia were more central to the fraud than the alleged manipulation of numbers at the Florida subsidiary, there was no allegation that the fraud had any impact on America or Americans, and the chain of causation between the Florida subsidiary's actions and the statements that reached investors was lengthy. Morrison, 547 F.3d at 176-77.
On appeal, the majority of the Supreme Court, in an opinion written by Justice Scalia, framed the question before it as follows: "[W]hether § 10(b) of the Securities Exchange Act of 1934 provides a cause of action to foreign plaintiffs suing foreign and American defendants for misconduct in connection with securities traded on foreign exchanges;" that is, whether the "F-cubed" claims could be asserted under the securities laws. Morrison, 130 S.Ct. at 2874. However, the Supreme Court — unlike the Second Circuit — did not consider this to be a question of subject matter jurisdiction. Id. at 2877. Rather, the Supreme Court explained that "to ask what conduct § 10(b) reaches is to ask what conduct § 10(b) prohibits, which is a merits question." Id.
Turning to that merits question, the majority in Morrison concluded that Section 10(b) does not apply extraterritorially, applying a presumption that "[w]hen a statute gives no clear indication of an extraterritorial application, it has none." Id. at 2878. In so holding, the Supreme Court rejected the Second Circuit's longstanding "conduct and effects" tests, which focused on whether the wrongful conduct had a substantial effect on United States markets or citizens or occurred in the United States. Id. at 2881; Cornwell v. Credit Suisse Group, 729 F. Supp. 2d 620, 622 (S.D.N.Y. 2010) ("In Morrison, the Supreme Court roundly (and derisively) buried the venerable `conduct or effect' test the Second Circuit devised and for years had employed to determine whether the protections and remedies contained in § 10(b) of the Exchange Act apply extraterritorially to reach fraudulent securities transactions abroad. . . .") The majority found that the "conduct and effects" tests lacked any textual basis, were "not easy to administer," and yielded inconsistent and unpredictable results. Id. at 2879-80.
Having roundly rejected the "conduct and effects" tests, and having concluded that Section 10(b) does not apply extraterritorially, the Supreme Court went on to consider whether to dismiss the claims of the plaintiffs before it. Many essentially foreign transactions have some domestic aspect so the issue remained where to draw the line in particular cases. Id. 2884. The plaintiffs in Morrison argued that they were seeking only domestic application of Section 10(b) because Florida was where NAB's subsidiary had engaged in the challenged financial manipulation. But the Supreme Court rejected plaintiffs' suggestion that the deceptive conduct in Florida brought plaintiffs' claims within the ambit of Section 10(b). Id. at 2883-84. In reaching that conclusion, the Supreme Court did not confine its discussion to the particular fact pattern before it. See id. at 2884-88. Instead, it "went out of its way to fashion a new rule designed to correct the enumerated flaws the Court found in the Second Circuit's tests" and made clear that it sought to replace the Second Circuit's unpredictable jurisprudence with a new, bright-line rule that would yield consistent, certain results. Cornwell, 729 F. Supp. 2d at 625.
That new rule is as follows: "[I]t is in our view only transactions in securities listed on domestic exchanges, and domestic transactions in other securities, to which § 10(b) applies." Morrison, 130 S.Ct. at 2884. Justice Scalia repeated that rule using different, but presumably equivalent language, at the end of his majority opinion: "Section 10(b) reaches the use of a manipulative or deceptive device or contrivance only in connection with the purchase or sale of a security listed on an American stock exchange, and the purchase or sale of any other security in the United States." Id. at 2888. The Court reasoned that in addition to bringing clarity and certainty to the securities law field, the rule it was announcing would avoid conflicts with foreign securities laws. "The transactional test we have adopted — whether the purchase or sale is made in the United States, or involves a security listed on a domestic exchange," the Supreme Court explained would avoid "the problem of interference with foreign laws that application of § 10(b) abroad would produce." Id. at 2886. Applying its newly enunciated rule to the facts before it, the Supreme Court held that plaintiffs failed to state a claim upon which relief could be granted under Section 10(b) because the purchase or
sale of NAB ordinary shares in Australia "involve[d] no securities listed on a domestic exchange, and all aspects of the purchases complained of by those petitioners who still have live claims occurred outside the United States." Id. at 2888. It should be noted that the Court was never presented with and did not consider the arguments plaintiffs make here, that the listing of NAB's ADRs on the NYSE required the simultaneous listing of its ordinary shares (albeit not for trading purposes) and, therefore, that NAB's ordinary shares actually met the test enunciated.
The Court's reference to "petitioners who still have live claims" was intended to exclude the named petitioner, Robert Morrison, an American investor who had purchased NAB ADRs, whose claims had been dismissed by the district court for failure to allege damages — a decision that had not been challenged. Morrison, 130 S.Ct. at 2876 n. 1.
B. Morrison Applied
Though the Supreme Court purported to lay out a bright-line rule regarding the extraterritorial application of Section 10(b), Morrison's impact on this case is far from clear. The parties agree that Morrison has no impact on the claims of ADR purchasers since Vivendi's ADRs were listed and traded on the NYSE. However, the parties disagree over Morrison's impact on the claims of foreign and American purchasers of ordinary shares, transactions that necessarily took place on foreign exchanges.
As noted, plaintiffs contend that Vivendi's ordinary shares were, in fact, "listed on a domestic exchange," such that Section 10(b) claims by ordinary share purchasers (whether foreign of American) meet the second prong of the Morrison test. Plaintiffs' line of argument in this regard is straightforward yet complex. It begins with the undisputed fact that Vivendi listed and sold ADRs on the NYSE. Vivendi's ADRs were sold in the United States as part of a U.S. public offering in 2000, making them "level 3" ADRs — indicating the highest level of issuer involvement. Because Vivendi chose to sell its ADRs in the U.S. through a public offering, Vivendi was required to register under the `33 Act a corresponding number of its ordinary shares with the SEC. See SEC Release, 1991 SEC LEXIS 936, at *35 ("[W]hen there is a public offering of securities in ADR form, both the ADRs and the deposited securities must be registered."); id. at n. 47 ("ADRs are registered on Form F-6 and the deposited securities are usually registered on Form F-1, F-2, F-3 or F-4."). Consistent with those requirements, Vivendi deposited the ordinary shares underlying its ADR offering with a French bank, registered those shares with the SEC on Form F-4, and also listed them on the NYSE — albeit not for trading purposes. Relying on 17 C.F.R. § 240.12d1-1(a), plaintiffs then argue that the registration of the ordinary shares underlying Vivendi's ADR issuance caused the entire class of Vivendi's ordinary shares (including those ordinary shares that did not underlie any ADRs') to be registered with the SEC. Vivendi also was required to register its
ADR facilities may be either "unsponsored" or "sponsored." Roche Holdings, 292 F.3d at 367; SEC Release, 1991 SEC LEXIS 936, at *7. A depositary may establish an unsponsored facility without the participation of the issuer of the deposited securities. In contrast, a "sponsored" ADR facility is established jointly by an issuer and a depositary. "Sponsored" ADR facilities fall into three categories. Level 1 facilities are those in which the ADRs trade over-the-counter, for example, on the "pink sheets." Level 2 refers to ADRs quoted on the NASDAQ or listed on a national securities exchange when the ADRs have not been offered in a U.S. public offering. Level 3 denotes ADRs quoted on the NASDAQ or listed on a national securities exchange after a U.S. public offering of ADRs. Levels 1, 2, and 3 generally indicate lower to higher degrees of issuer involvement with the facility, and lower to higher amounts of information made available by the issuer to the public. SEC Release, 1991 SEC LEXIS 936 at *8, 10-11 n. 21.
See SEC Form F-4, dated October 30, 2000 (Margolies Decl. Ex. 6) (registering Vivendi Universal ordinary shares); SEC Form F-6, dated November 3, 2000 (Margolies Decl. Ex. 7) (registering Vivendi Universal ADRs).
17 C.F.R. § 240.12d1-1(a) provides as follows: "Registration effective as to class or series. (a) An application filed pursuant to section 12(b) and (c) of the act for registration of a security on a national securities exchange shall be deemed to apply for registration of the entire class of such security. Registration shall become effective, as provided in section 12(d) of the act, (1) as to the shares or amounts of such class then issued, and (2), without further application for registration, upon issuance as to additional shares of amounts of such class then or thereafter authorized." (emphasis added). To the Court's knowledge, this regulation has only been cited by two prior judicial decisions, Ellerin v. Mass. Mutual Life Ins. Co., 270 F.2d 259 (2d Cir. 1959) and American Standard Inc. v. Crane Co., No. 68 Civ. 2461, 1972 WL 333 (S.D.N.Y. July 6, 1972), neither of which has any obvious relevance to the argument plaintiffs now make.
ordinary shares pursuant to Section 12(b) of the Exchange Act and did so. Consequently, plaintiffs' contend, all of Vivendi's ordinary shares were registered or listed — plaintiffs claim the terms are interchangeable — on a U.S. exchange. Thus, plaintiffs contend, all purchasers of ordinary shares (whether foreign or American) can bring Section 10(b) claims under the test announced in Morrison even though their shares were traded aboard not in the U.S. Simply put, Justice Scalia stated that Section 10(b) applies to "securities listed on domestic exchanges" and, plaintiffs contend, Vivendi's ordinary shares meet that test. P.Supp.Br. at 10; Morrison, 130 S.Ct. at 2884, 2888. This argument is not unmoored from all policy considerations. When a foreign issuer decides to access U.S. capital markets by listing and trading ADRs, it subjects itself to SEC reporting requirements, and it would not be illogical to subject that company to the antifraud provisions of the Exchange Act at least where there is a sufficient nexus to the United States. Indeed, that premise underlies both the conduct and effects tests and the Morrison bright line test. Although these standards diverge on the issue of extraterritoriality, as Justice Scalia noted, transnational transactions have both domestic and foreign aspects and the issue becomes one of line-drawing under either test.
See SEC Form 8-A (Margolies Decl. Ex. 8) (application for registration of Vivendi ordinary shares, represented by ADRs on the NYSE); SEC Forms 20-F for the fiscal years ended December 31, 2000, 2001 and 2002, respectively (Margolies Decl. Ex. 9) (each noting that Vivendi's ordinary shares are "listed" on the NYSE, "not for quotation or trading purposes, but only in connection with the registration of the American Depositary Shares pursuant to the requirements of the Securities and Exchange Commission").
Plaintiffs argue that under Morrison, the terms "registered" and "listed" are essentially equivalent since any share that is registered on an exchange is also listed on an exchange. To this extent, plaintiffs are correct. While Section 10(b) refers to the purchase or sale of securities "registered" on a national securities exchange, 15 U.S.C. § 78j(b), in today's common parlance at least, shares are "listed" on exchanges, see Morrison, 130 S.Ct. at 2884-85 ("The [Exchange] Act's registration requirements apply only to securities listed on national securities exchanges.") (citing 15 U.S.C. § 781(a)).
That being said, there appears to be a technical flaw in plaintiffs' argument. It is true that the registration of any shares under Section 12 of the Exchange Act extends registration to the entire class of securities. And when a foreign company registers ADRs with the SEC, it must also register the underlying ordinary shares, necessarily resulting in the registration with the SEC of all ordinary shares. But registration with the SEC is not the same as listing (registering) on an exchange. The sample NYSE listing application provided to the Court at argument indicates that only a discrete number of ordinary shares are listed; this being the number of ordinary shares needed to back-up the ADRs being listed. Thus while all ordinary shares of a foreign issuer are deemed to be registered with the SEC, a lesser fixed amount of shares are actually listed with the Exchange. And ordinary shares that are not listed on an exchange (for any purpose) would fall outside plaintiffs' literalist reading of the Morrison bright-line test as well as the underlying language of Section 10(b).
While the record is sketchy, it appears that Vivendi separately registered with the SEC approximately 500 million ADRs and a roughly equivalent number of ordinary shares in connection with the three-way merger of Vivendi, Seagram and Canal Plus in November, 2000. (Margolis Decl. in Supp. of Pls.' Suppl. Br., Exs. 6, 7.) Seagram shareholders had the right to receive approximately 400 million Vivendi ADRs in exchange for their Seagram stock. ( Id.) Vivendi's listing application with the NYSE is not before the Court, but it would likely have listed the 500 million ADRs that had just been registered (with the SEC), as well as the ordinary shares registered (with the SEC) to back up the newly-issued ADRs.
Some number of ordinary shares may have already been listed in connection with an existing market for Vivendi ADRs.
At this point, the fact that approximately 500 million ordinary shares of Vivendi were listed on the NYSE (not for trading purposes) would not appear to affect the analysis since an equal number of ADRs represented those ordinary shares. The ADRs were both listed and traded on the NYSE, and thereby fall within any reading of Morrison. This would render moot the issue of whether the simultaneous listing of the underlying ordinary shares (not for trading purposes) would itself meet the Morrison test. But there appears to be a wrinkle. According to Vivendi's Form 20-F filed on July 2, 2001, there were only 122 million outstanding ADRs as of December 31, 2000. This appears to reflect a significant migration of the ADRs issued in the November 2000 merger back to European markets. Unless Vivendi amended its listing agreement with the NYSE, up to 378 million ordinary shares would still be listed with the NYSE despite the fact that many ADRs appear to have been exchanged by the Depository for ordinary shares and, therefore, are no longer outstanding.
Assuming, arguendo, that there were ordinary shares of Vivendi that remained "listed" but were un-tethered to any ADRs, would a purchase of such shares by a foreign purchaser trading on a foreign exchange satisfy the Morrison test, as plaintiffs claim? That is, do "foreign cubed" transactions actually survive Morrison where ordinary shares are listed but not traded on a domestic exchange as a result of a foreign issuer's ADR program. All the courts who have directly or indirectly addressed this issue have dismissed the argument as a technical one that is contrary to the "spirit" of Morrison. In re Royal Bank of Scotland Group PLC Securities Litigation, No. 09 Civ. 300 (DAB), 2011 WL 167749, at *7 (Jan. 11, 2011); In re Alstom SA Securities Litigation, No. 03 Civ. 6595 (VM), 2010 WL 3718863, at *2-3 (S.D.N.Y. Sept. 14, 2010); Sqalambo v. McKenzie, No. 09 Civ. 10087 (SAS), 2010 WL 3119349, at *17 (S.D.N.Y. Aug. 6, 2010). These decisions focus, as is appropriate, on the Morrison opinion as a whole, which can be read as adopting a test that turns on the territorial location of the subject transaction. E.g., In re Royal Bank, 2011 WL 16779, at *8. This focus is consonant with the express view of the Supreme Court, which "reject[ed] the notion that the Exchange Act reaches conduct in this country affecting exchanges or transactions abroad. . . ." Morrison, 130 S.Ct. at 2885. As the Court observed "no one . . . thought that the Act was intended to `regulate' foreign securities exchanges," id. at 2884, and furthermore, there is no "national public interest" in "transactions conducted upon foreign exchanges and markets," id. at 2882.
Since ordinary share are fungible, it is not obvious how they could be traced after an ADR is exchanged for an ordinary share which then trades in Europe.
Read in this context, perhaps Justice Scalia simply made a mistake. He stated the test as being whether the alleged fraud concerned the purchase or sale of a security "listed on an American stock exchange," id. at 2888, when he really meant to say a security "listed and traded" on a domestic exchange. Perhaps so, but then there is the question of the actual language of the statute, as well as Justice Scalia's interpretation thereof in footnote 10 of Morrison.
By its terms, Section 10(b) applies to the "purchase or sale of any security registered [listed] on a national securities exchange or any security not so registered [listed]." Justice Scalia points out in footnote 10 that the second phrase regarding unregistered securities is exclusively focused on domestic transactions in securities (as opposed, one presumes, to both domestic and foreign). This is so, Justice Scalia explains, because if the phrase referred to "all" purchases and sales of unregistered securities (domestic and foreign, presumably), the phrase would have been at best surplusage:
[T]he only alternative to that reading [that only domestic transactions in unregistered securities are proscribed] makes nonsense of the phrase, causing it to cover all purchases and sales of registered securities, and all purchases and sales of nonregistered securities — a thought which, if intended, would surely have been expressed by the simpler phrase "all purchases and sales of securities."Id. at 2885 n. 10. If this is so, the first phrase referring to the "purchase or sale of any security registered on a national securities exchange" logically refers to both domestic and foreign transactions in registered securities, (that is, all purchases and sales of registered securities as opposed to only domestic purchases and sales of unregistered securities). Morrison's footnote 10 gives this Court pause, but ultimately the Court concludes that it cannot carry the freight that plaintiffs ask it to bear. There is no indication that the Morrison majority read Section 10(b) as applying to securities that may be cross-listed on domestic and foreign exchanges, but where the purchase and sale does not arise from the domestic listing, particularly where (as here) the domestic listing is not even for trading purposes. Indeed, even under the old conduct and effect tests similar cross-listing scenarios would not by themselves support a Section 10(b) claim. It is unlikely that the Morrison court in any way intended to broaden Section 10(b)'s reach when it replaced the conduct and effects tests with its new transactional standard. This is not to say that this Court's reading of Morrison is free from doubt, or that Morrison's reading of Section 10(b) is free of potential inconsistency, only that resolution of these issues is fairly the province of the Supreme Court or Congress.
This Court respectfully demurs from the reading of Section 10(b) proffered in Morrison's footnote 10. Accepting the Supreme Court's determination that Congress did not intend Section 10(b) to have extraterritorial application, one could simply read the section as referring to `all domestic purchases and sales of registered or unregistered securities.' While it could have been more simply expressed, the more laborious form chosen is not necessarily `nonsense' but more likely the result of legislators simply tacking on "or any security not so registered" to the original version while it was working its way to passage. See Steve Thel, The Original Conception of Section 10(b) of the Securities Exchange Act, 42 STAN. L. REV. 385, 443-44 n. 263.
The next issue is whether Morrison also requires the Court to dismiss the claims of American purchasers of Vivendi's ordinary shares. A threshold question is whether Vivendi's request that the Court dismiss the claims of American ordinary shareholders is timely. Plaintiffs contend that Vivendi waived any right to challenge the claims of American purchasers of ordinary shares by failing to argue at an earlier stage of this litigation that the Court lacked subject matter jurisdiction over these claims. Plaintiffs are correct that Vivendi never previously challenged the claims of American, as opposed to foreign, ordinary share purchasers for lack of subject matter jurisdiction. However, one reason for that failure was that until the Supreme Court's recent decision in Morrison, it was well-settled under Second Circuit precedent that American purchasers of a foreign company's shares could bring Section 10(b) claims. Indeed, Morrison repudiated the Second Circuit's longstanding jurisprudence to a degree "that would surprise . . . generations of American investors — and . . . the Congress that passed the Exchange Act." Id. 130 S.Ct. at 2895 (Stevens' concurrence.)
Vivendi seeks either entry of judgment as a matter of law dismissing these claims pursuant to Rule 50, or dismissal for failure to state a claim pursuant to Rule 12(h)(2). The Court believes that Vivendi's motion is properly construed as a motion pursuant to Rule 12(h)(2), and therefore analyzes Vivendi's claim as such.
In this context, Vivendi did not waive its right to seek dismissal of the claims of American purchasers of ordinary shares for failure to state a claim under Rule 12(h)(2), a defense that arose from intervening Supreme Court authority. See Holzager v. Valley Hosp., 646 F.2d 792, 796 (2d Cir. 1981) ("In any event a party cannot be deemed to have waived objections or defenses which were not known to be available at the time they could first have been made, especially when it does raise the objections as soon as their cognizability is made apparent.") (citing Curtis Publishing Co. v. Butts, 388 U.S. 130, 143, 145 (1967)). The Circuits have recognized that "[a]n exception to normal law of the
case and waiver rules is recognized when an intervening decision from a superior court changes the controlling law." Beazer E. v. Mead Corp., 525 F.3d 255, 263 (3d Cir. 2008). In such contexts, the Circuits will allow parties, for example, to raise arguments that they did not raise in their opening briefs. E.g., West v. Ortiz, 2007 WL 706924, at *5 (10th Cir. Mar. 9, 2007) (allowing an appellant to raise an issue made viable by intervening Supreme Court case law for the first time on reply); DSC Commc'ns Corp. v. Next Level Commc'ns, 107 F.3d 322, 326 n. 2 (5th Cir. 1997) (party that waived an issue by failing to include it in its opening brief could raise the issue in a supplemental brief based on an intervening change of law). It would be curious indeed if Vivendi would be allowed to raise an argument on appeal that it could not assert before this Court. While the language of Rule 12(h)(2) provides that a failure to state a claim defense may be raised "in any pleading allowed or ordered under Rule 7(a); (B) by a motion under Rule 12(c) [for judgment on the pleadings]; or (C) at trial," and does not expressly contemplate motions made after trial but before entry of judgment, the Court concludes that under the circumstances here — in which judgment has yet to be entered and will not be entered for quite some time since the damages phase of this case has yet to occur — it is appropriate to permit Vivendi to bring a motion pursuant to Rule 12(h)(2).
While a number of cases have stated that motions to dismiss for failure to state a claim under Rule 12(h)(2) cannot be brought post-trial, see, e.g., Arbaugh v. Y H Corp., 546 U.S. 500, 507 (2006); Kontrick v. Ryan, 540 U.S. 443, 459 (2004); Snead v. Dep't of Soc. Servs. of City of N.Y., 409 F.Supp. 995, 1000 (S.D.N.Y. 1975), the Court does not find these cases to be dispositive. None of these cases addressed a situation in which the party seeking dismissal for failure to state a claim brought its motion after a trial on liability, but before judgment had been entered and before the damages phase of a case had begun. Second, none of these cases involved situations in which the justification for the moving party's failure to raise the defense at or before trial was that intervening Supreme Court authority made available a defense that had not previously been available.
Turning to the substance of the issue, the Court finds that American ordinary share purchasers cannot bring Section 10(b) in the wake of Morrison. In reaching this conclusion, the Court joins other lower courts that have rejected the argument that a transaction qualifies as a "domestic transaction" under Morrison whenever the purchaser or seller resides in the United States, even if the transaction itself takes place entirely over a foreign exchange. See Cornwell, 729 F. Supp. 2d at 627; Harry Stackhouse v. Toyota Motor Co., et al., No. 10 Civ. 0922 (DSF), 2010 WL 3377409, at *1 (C.D. Cal. July 16, 2010; In re Royal Bank, 2011 WL 167749, *7-8.
Though the Supreme Court in Morrison did not explicitly define the phrase "domestic transactions," there can be little doubt that the phrase was intended to be a reference to the location of the transaction, not to the location of the purchaser and that the Supreme Court clearly sought to bar claims based on purchases and sales of foreign securities on foreign exchanges, even though the purchasers were American.
As Judge Marrero has pointed out, reading Morrision to permit Section 10(b) claims "based strictly on the American connection of the purchaser or seller . . . simply amounts to a restoration of the core element of the effects test." Cornwall, 729 F. Supp. 2d at 624. The Morrison Court made clear that it did not believe that the American citizenship of the plaintiff was itself sufficient to give rise to Section 10(b) claims when it analogized the citizenship of securities purchasers to the American plaintiff in EEOC v. Arabian American Oil Co., 499 U.S. 244 (1991) (hereinafter " Aramco"), who sought to assert discrimination claims arising out of his employment abroad:
[T]he presumption against extraterritorial application would be a craven watchdog indeed if it retreated to its kennel whenever some domestic activity is involved in the case. The concurrence seems to imagine just such a timid sentinel, but our cases are to the contrary. In Aramco, for example, the Title VII plaintiff had been hired in Houston, and was an American citizen. The Court concluded, however, that neither that territorial event nor that relationship was the `focus' of congressional concern, but rather domestic employment.
Applying the same model of analysis here, we think that the focus of the Exchange Act is not upon the place where the deception originated, but upon purchases and sales of securities in the United States.Morrison, 130 S.Ct. at 2884 (citing Aramco, 499 U.S. at 255). Just as the plaintiff in Aramco's American citizenship was not enough to render his employment "domestic employment" subject to Title VII, the American citizenship of a person who purchase a foreign company's shares on a foreign exchange does not render that a "domestic transaction."
Finally, it simply blinks reality to ignore Justice Stevens' concurring opinion in Morrison which underscored the stark impact of the majority's opinion on American investors who purchased shares abroad and fell victim to securities fraud:
Imagine, for example, an American investor who buys shares in a company listed only on an overseas exchange. That company has a major American subsidiary with executives based in New York City; and it was in New York City that the executives masterminded and implemented a massive deception which artificially inflated the stock price — and which will, upon disclosure, cause the price to plummet. Or imagine that those same executives go knocking on doors in Manhattan and convince an unsophisticated retiree, on the basis of material misrepresentations, to invest her life savings in the company's doomed securities. Both of these investors would, under the Court's new test, be barred from seeking relief under § 10(b).Id. at 2895 (Stevens, J., concurring). The purchases described in Justice Stevens' hypothetical would easily qualify as "domestic transactions" under plaintiffs' definition. If Justice Stevens has misinterpreted the Supreme Court's holding, one might have expected the majority opinion to address that misunderstanding. The majority's silence in this regard speaks volumes.
For all of these reasons, the Court finds that the Section 10(b) claims of Americans who purchased Vivendi's ordinary shares, like the claims of foreigners who purchased Vivendi's ordinary shares, do not survive Morrison. Accordingly, these claims are dismissed. Furthermore, the Court hereby amends the class definition in this case to exclude purchasers of ordinary shares. See Fed.R.Civ.P. 23(c)(1)(C) ("An order that grants or denies class certification may be altered or amended before final judgment."); Boucher v. Syracuse Univ., 164 F.3d 113, 118 (2d Cir. 1999) ("Courts are required to reassess their class rulings as the case develops.") (internal quotations omitted). The class going forward shall consist of all persons from the United States, France, England and the Netherlands who purchased or otherwise acquired Vivendi ADRs between October 30, 2000 and August 14, 2002.
Vivendi also asks the Court to limit the class to include only those who purchased Vivendi ADRs, thereby excluding those who "otherwise acquired" ADRs during the Class Period. Vivendi's attempt to raise this issue at this juncture is perplexing. Vivendi never objected at the class certification stage to the inclusion in the class of persons who "otherwise acquired" shares of Vivendi. Now, years after the Court certified this class action, Vivendi seeks to challenge the class definition on this ground, citing Morrison as the only apparent authority for its new argument. But Morrison addressed the extra-territorial application of Section 10(b); it did not purport to say anything about whether persons who acquired their Vivendi stock through means other than direct purchases can bring Section 10(b) claims. The Court, therefore, declines to modify the class definition to exclude those who "otherwise acquired" Vivendi ADRs during the Class Period. Whether any particular claimant who acquired Vivendi shares through means other than a purchase is eligible for damages will be addressed, if the issue properly presents itself, during the claims administration procedure.
C. Plaintiffs' Motion to Conform the Pleadings
In a last ditch attempt to avoid the impact of Morrison, plaintiffs argue that if the Court dismisses the Section 10(b) claims of any class members, it should permit such class members to recover under New York common law fraud (even though plaintiffs never pled New York common law fraud), either by conforming the pleadings under Federal Rule of Civil Procedure 15(b)(2) or by entering judgment in favor of plaintiffs under Federal Rule of Civil Procedure 54(c). Plaintiffs contend that the elements of common law fraud under New York law are substantially identical to the elements of Section 10(b) liability and that, accordingly, the proof at trial that established Vivendi's liability under Section 10(b) also established Vivendi's liability under the doctrine of common law fraud. However, plaintiffs ignore a critical difference between common law fraud and Section 10(b) liability: whereas Section 10(b) liability must be proven by a preponderance of the evidence, Herman MacLean v. Huddleston, 459 U.S. 375, 390 (1983), common law fraud under New York law must be proven by clear and convincing evidence, Gaidon v. Guardian Life Ins. Co. of Am., 725 N.E.2d 598, 607 (N.Y. 1999). The jury's finding that Vivendi was liable for securities fraud under the lower preponderance of the evidence standard applicable to Section 10(b) actions cannot support the entry of judgment against Vivendi on common law fraud claims which are subject to a higher burden of proof.
II. Vivendi's Motion for Judgment as a Matter of Law Pursuant to Rule 50
Vivendi seeks full or partial judgment as a matter of law on the following grounds: (1) plaintiffs failed to establish that Vivendi omitted material information from any statements or made any misrepresentations; (2) plaintiffs failed to establish that anyone at Vivendi acted with scienter; (3) plaintiffs failed to prove loss causation; (4) the Court must find that September 11, 2001 is the last day for which any class member can recover damages because the jury found that inflation was zero for an approximately three week period beginning on September 11, 2001; (5) certain of the fifty-seven statements that the jury found to have violated Section 10(b) were inactionable forward-looking statements protected by the PSLRA's "safe harbor" provision; and (6) certain of the fifty-seven statements that the jury found to have violated Section 10(b) were inactionable puffery. The Court finds that Vivendi is not entitled to full or partial judgment as a matter of law on any of these grounds and denies Vivendi's Rule 50 motion, with one minor exception discussed below.
The Court here addresses the arguments made in Vivendi's Rule 50(a) motion and in its renewed motion for judgment as a matter of law pursuant to Rule 50(b) since the latter motion incorporated all arguments previously made.
A. Standard of Review
A defendant is entitled to judgment as a matter of law if, after a party has been fully heard on an issue during trial, the Court finds that "a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue. . . ." Fed.R.Civ.P 50(a)(1). A party may move for judgment as a matter of law "at any time before the case is submitted to the jury." Fed.R.Civ.P. 50(a)(2). In ruling upon a Rule 50(a) motion,
[T]he court must draw all reasonable inferences in favor of the nonmoving party, and it may not make credibility determinations or weigh the evidence. Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge. Thus, although the court should review the record as a whole, it must disregard all evidence favorable to the moving party that the jury is not required to believe.Reeves v. Sanderson Plumbing, 530 U.S. 133, 150-51 (2000) (internal quotation marks and citations omitted). "[A] court may grant a motion for judgment as a matter of law `only if it can conclude that, with credibility assessments made against the moving party and all inferences drawn against the moving party, a reasonable juror would have been compelled to accept the view of the moving party.'" Zellner v. Summerlin, 494 F.3d 344, 370-71 (2d Cir. 2007) (quoting Piesco v. Koch, 12 F.3d 332, 343 (2d Cir. 1993)).
If the court does not grant a motion for judgment as a matter of law under Rule 50(a), the court is considered to have submitted the action to the jury subject to the Court's later decision of the issues raised by the motion, and the moving party may renew its motion after trial pursuant to Rule 50(b). The standard for granting a motion under Rule 50(b) is the same as the standard for granting a motion under Rule 50(a). Raspente v. Nat'l R.R. Passenger Corp., 111 F.3d 239, 241 n. 3 (2d Cir. 1997).
The substantive law underlying plaintiffs' claims and Vivendi's Rule 50 motion is well established. To establish a violation of Section 10(b), plaintiffs were required to prove by a preponderance of the evidence that, in connection with the purchase or sale of a security: (1) the defendant made a untrue statement of material fact, or omitted to state a material fact which made what was said, under the circumstances, misleading; (2) the defendant acted with scienter; (3) plaintiffs justifiably relied on the misstatement or omission; and (4) plaintiffs suffered an economic loss as a result of the misstatement or omission. See Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 341-42 (2005); In re Salomon Analyst Metromedia Litig., 544 F.3d 474, 478 n. 1 (2d Cir. 2008). There was more than sufficient evidence at trial on each element of plaintiffs' claims to support the jury's finding of liability against Vivendi.
B. Material Misstatements and Omissions
In the late 1990s, Vivendi (then known as Vivendi, S.A.) acquired a number of companies and took on significant debt as part of a strategy to turn the company into a global media and communications leader. A key element of that strategy was a three-way merger between Vivendi, The Seagram Company Limited (a Canadian beverage company owned by the Bronfman family, which also owned Universal, the Hollywood film and music company), and Canal Plus, S.A. (a French cable company). The three-way merger (the "Merger") was announced on October 30, 2000 through Vivendi's filing of a Registration Statement and Prospectus on Form F-4 ("F-4") with the SEC. In that F-4, Vivendi announced an objective "to grow pro forma adjusted EBITDA [Earnings Before Interest, Taxes, Depreciation and Amortization] at an approximate 35% compound annual growth rate through 2002." The Merger was approved by the shareholders of all three companies and took effect in December 2000.
In the wake of the Merger, Vivendi continued to pursue various acquisitions and transactions, and continued to take on additional debt. In February 2001, Vivendi acquired a 35% stake in Moroccan telephone company, Maroc Telecom, for €2.3 billion and entered into a written agreement with the Kingdom of Morocco to purchase an additional 16% stake in Maroc Telecom by February 28, 2002 for a price of €1.1 billion. And in December of 2001, Vivendi entered into a transaction acquiring USA networks for $10.3 billion, and to acquire a 10% stake in another American television company, EchoStar, for $1.5 billion.
Vivendi's numerous acquisitions translated into debt levels that began to test that company's liquidity. According to plaintiffs, Vivendi concealed its growing liquidity risk from the shareholders by making misleading statements that touted the company's financial health and performance while failing to disclose its true liquidity position. Plaintiff also alleged that Vivendi further concealed its liquidity risks (1) by failing to make full disclosure of the true nature of its trumpeted 35% EBITA growth rate; (2) by engaging in earnings management; (3) by failing to disclose a €1.1 commitment to buy additional stock in Maroc Telecom; (4) by failing to disclose that a €1.3 billion advance from a related company, Cegetel, was subject to immediate repayment; and (5) by entering into a secret "portage" agreement to avoid having to consolidate earnings with another affiliate, Telco. According to plaintiffs, Vivendi's undisclosed liquidity risk began to materialize in the first half of 2002 leading the company to the brink of bankruptcy and causing a sharp decline in Vivendi's share price.
Vivendi, of course, vigorously contested plaintiffs' version of the facts with respect to every issue identified above. Vivendi witnesses testified that there was no hidden liquidity risk, that the company always had a healthy cash flow, and that its disclosures on EBIDTA growth, earnings, Maroc Telecom, Cegetel and Telco were accurate and did not further conceal a (non-existent) liquidity problem. The jury could have accepted Vivendi's version of the facts but it chose not to. And plaintiffs certainly produced enough evidence on which a reasonable jury could have found that Vivendi had a grave liquidity risk that was ignored by management in their numerous, unconditional statements of the company's financial health and masked by the company's inadequate disclosure of specific transactions.
Before the Court delves into that evidence, a word is warranted about the evidence that the jury was not required to believe in this case. First, a jury is not required to believe the testimony of interested witnesses. Kerman v. City of New York, 374 F.3d 93, 123 (2d Cir. 2004). Messier and Hannezo were clearly interested witnesses since they were defendants in the case. Thus, the jury was not required to believe any of their testimony to the extent the jury found it to be self-serving or not worthy of belief in light of the other evidence at trial, including the many documents collected from Vivendi's files that were written during the Class Period.
Second, much of the testimony in this case was expert testimony. On almost all issues, there was conflicting expert testimony presented by both sides. It was for the jury to decide which experts to believe, and which portions of their testimony to believe. See Knapp v. Leandro, 46 F.3d 170, 179 (2d Cir. 1995) (jury was entitled to decide to believe one expert's testimony over the conflicting testimony of another expert); United States v. Artuso, 618 F.2d 192, 195 (2d Cir. 1980) ("The jury, as factfinder, was entitled to credit the testimony of the Government's psychiatric expert over that of the defense expert."); Jury Charge No. 14 (instructing the jury to give the expert testimony "whatever weight, if any, you find it deserves in light of all of the other evidence in this case.") At the end of the day, the jury could reasonably have decided not to credit the testimony of the defense experts.
With that in mind, the Court turns to a sampling of the evidence that could have supported a finding that Vivendi made material misstatements and omissions regarding its true liquidity condition throughout the Class Period. One of the striking things about this case was the large number of highly damaging documents from Vivendi's files that were presented to the jury throughout the trial. One does not often see so many "bad" documents in a single case. The clearest example was the compilation of documents known as the "Book of Warnings." (PX-1, PX-2.) This compilation, put together by Hannezo, contained a series of internal memoranda that Hannezo had written and sent primarily to Messier, in which he raised serious questions about the company's financial health. Many of the memoranda therein used highly colorful language, and on their face, they could easily have been read as suggesting blatant wrongdoing. It is a rare case indeed in which one of the most senior officers at a company not only writes such memoranda in the first instance, but also gathers them together and entitles his compilation "the Book of Warnings."
At trial, plaintiffs' contrasted Hannezo's internal statements in the "Book of Warnings" with Vivendi's public statements regarding its economic performance. For example, plaintiffs presented evidence that on December 5, 2000, Messier made a speech to shareholders urging them to approve the three-way Merger, in which he stated that the company had a "very sound financial footing" and stating that, "[t]hanks to [its] free net cash flow and the opportunities to dispose of some holdings, such as our stake in BSkyB, [it] would have an additional war chest of 10 billion euros for 2001-2002 before the first euro of debt." (PX-720 at 7.) A few weeks after Messier gave that speech, Hannezo wrote him a memo dated January 10, 2001, which stated: "I believe it is wrong to reason . . . in terms of free cash flow ( there won't be any this year)." (PX-168) (emphasis added). Just two days later, on January 12, 2000, Vivendi filed Messier's December 5, 2000 speech with the SEC (PX-720). Moreover, Messier continued to tout the strength of Vivendi's cash flow throughout much of 2001. ( See, e.g., PX-474; PX-701; PX-415.) While the existence and degree of Vivendi's cash flow problems was disputed, there was more than sufficient evidence from which a jury could conclude that Vivendi's public statements regarding its "free cash flow" were false. A jury could have easily concluded that a reasonable investor would have viewed the misstatements/omissions regarding cash flow as significantly altering the "total mix" of information available. See TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449 (1976) (a fact is material if there is a "substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the `total mix' of information made available").
To provide another example, on June 26, 2001, Vivendi filed a 6-K ("Report of Foreign Private Issuer") stating that the foundations of its communications-related businesses were "particularly healthy and strong" because of, inter alia, "a healthy balance sheet," "a pro forma net debt that is practically non-existent," "record-high net income," and "the strength of our cash flow." (PX-474 at 3-4.) However, plaintiffs showed the jury videotaped deposition testimony from another Vivendi employee, Mr. Brisard, who testified that Vivendi's financial condition around in June 2001 was far from healthy. Brisard testified that beginning in June 2001, Vivendi's treasurer "expressed concerns over the cash situation, the liquidity situation of Vivendi Universal . . . He clearly raised the issue of a cash problem inside Vivendi Universal." (Brisard Dep. 33:24-34:06.) Brisard further testified that he attended meetings of the Finance Department between June and October 2001 in which Hannezo suggested on several occasions that Vivendi was "running out of cash" and would be facing "bankruptcy" if it continued on its current path. ( Id. 34:16-26:3.) A member of Vivendi's finance department, Anne Brassens, also testified that the company's liquidity situation was "tense" throughout much of the Class Period and became "dangerous" by late 2001. (Tr. at 376:11-14).
Vivendi argued to the jury that neither Brisard nor Brassens knew what they were talking about and offered testimony from Messier and Hannezo that Vivendi's upbeat public statements regarding its earnings and results were not misleading. The jury, of course, was not required to find this testimony credible. In addition, Vivendi argued that the company always had cash available and always paid its bills on time, but the fact that Vivendi never actually reached the point of insolvency does not prove that it never had liquidity problems. It was well within the province of the jury to find, relying on the Brassens and Brisard testimony as well as Hannezo's Book of Warnings, that Vivendi's June 26, 2001 statement, which touted Vivendi's financial health and strong results, omitted material information regarding Vivendi's true liquidity condition. The jury could have reached similar conclusions with respect to other statements in the second half of 2001 which painted a positive picture about Vivendi's financial results without disclosing the "dangerous" liquidity situation facing the company.
There was ample evidence to support such a conclusion. In addition to the Brisard and Brassens testimony quoted above, Brassens testified that the USA Networks Transaction "would create havoc with the debt level of Vivendi" and that "in December 2001, which is when these deals took place, the cash situation at Vivendi was extremely tense," ( see Brassens De. 32:10-32:25). Hannezo also warned Messier in an August 2001 memo that Vivendi needed to "reduce the debt that is becoming a real problem" (PX-148) (emphasis added), and in a November 27, 2001 memo that Vivendi was "already at the limit (beyond the limit"), a somewhat ambiguous reference that could reasonably be construed as suggesting that Vivendi's liquidity situation continued to deteriorate in late 2001. (PX. 145.)
One of the more prominent documents in the Book of Warnings was a memorandum which was referred to colloquially throughout the trial as the "death seat" memo. In that memo, which Hannezo wrote in December 2001 after Vivendi announced the USA Networks Transaction and narrowly avoided being downgraded by the rating agencies, Hannezo told Messier:
I told you that I would talk to you about the personal consequences that I'm drawing from my painful and humiliating meetings with the ratings agencies . . . For the first time, I felt the wind pass by from the cannonball of something that, from a personal point of view, I do not want to put up with a downgrade, which would have led to a liquidity crisis; the jeers of all those who have waited for us at the pivotal occasion, desperate for such a long time to see us stumble . . . and the unpleasant feeling of being in a car whose driver is accelerating in a sharp turn while I'm the one in the death seat . . . The only thing that I am asking is that it doesn't all end in shame.
(PX-136.) Just days after Hannezo sent this memo, Vivendi reported on a conference call that the USA Networks Transaction was "not putting pressure on Vivendi Universal," and that "[a]s far as the global debt ratio of the group is concerned, our target is to have in 02 a debit EB[IT]DA ratio well below three times and especially we are focusing to reach that target ahead of the end of the first half of 2002 which means that Vivendi Universal will end up its program of selling its non-core assets in the first half of 2002, it will give us a very comfortable BBB credit rating target that we are comfortable with." (PX-499 at 7, 11.) The jury, juxtaposing these statements, could reasonably have concluded that it was materially misleading for Vivendi to state the USA Networks Transaction was "not putting pressure" on the company, and that it believed that its credit rating would be "very comfortable" by the end of the first half of 2002, without mentioning that the company had narrowly avoided a ratings downgrade, that a ratings downgrade would have led to a liquidity crisis, and that the only way the company had avoided a downgrade was by promising the rating agencies that it would sell a large amount of assets.
Defendants argued that plaintiffs were taking the "death seat" memo and most of the other documents from the Book of Warnings out of context. For example, Hannezo testified that he believed Vivendi's public statements that the transaction was not putting pressure on the company and that Vivendi expected to have a comfortable credit rating were accurate. ( See id. at 1040:17-1041.) More generally, both he and Messier testified that they acted at all times in good faith, and that they believed that all of Vivendi's public statements were accurate. But Hannezo and Messier were interested witnesses. The jury was not required to credit their testimony if they found it to be self-serving, implausible or inconsistent with the damaging documents written at the time the events were unfolding, long before this lawsuit was initiated. See Kerman, 374 F.3d at 123.
Vivendi also made extremely damaging remarks in its sworn court filings in a separate arbitration and litigation between Vivendi and Messier (the "Arbitration Documents"), regarding Messier's claim for a multi-million dollar severance payment. In the portions of these documents that were admitted as admissions by Vivendi (only), Vivendi stated repeatedly that it faced a "serious liquidity crisis threatening its very existence" in June/July 2002 ( see, e.g., PX-1288 at 11), and also stated that Messier "engineered over $100 billion in shareholder losses; [and] took one of France's largest and most respected companies to the brink of insolvency." (PX-1297 at 2.)
Vivendi attempted to minimize the significance of these admissions, arguing that the attorneys who wrote these documents exaggerated the "liquidity crisis." Vivendi tried to convince the jury that the purported "liquidity crisis" was simply a short-lived phenomenon in July 2002 that was easily overcome. Vivendi further argued that any admission about a liquidity crisis in the summer of 2002 — and there were other such
admissions in addition to the Arbitration Documents — revealed nothing about the company's liquidity situation at earlier points in the Class Period, and that the evidence showed that Vivendi did not face a serious liquidity risk in December 2001 (when plaintiffs alleged that the undisclosed liquidity risk was at its peak), much less an undisclosed one. But the jury could easily have found this argument to be implausible. It was certainly reasonable for the jury to infer that a company that admitted that it was on the verge of bankruptcy in the summer of 2002 had serious, undisclosed liquidity problems long before the summer of 2002 — an inference that, as noted above, found support in other evidence at trial.
See, e.g., PX-942 at 2 (memo written by Espinasse, listing factors that caused Vivendi's "treasury crisis"); PX-1211 (Fortou describing Vivendi as "not far from insolvency").
See footnote 33, supra, for some of the evidence that could have supported an inference that Vivendi faced a tense liquidity situation in December 2001.
There was also sufficient evidence for the jury to conclude that Vivendi made material misstatements or omissions with respect to five specific subjects that related directly or indirectly to Vivendi's evolving liquidity crisis: the impact of purchase accounting; earnings management; the Maroc Telecom acquisition, the Cegetel current account; and the Telco portage. For instance, plaintiffs argued at trial that Vivendi's repeated statements that it was meeting its aggressive EBIDTA growth rate targets of 35% and was an attractive investment because it had "the highest growth rates in the industry" were materially misleading because they failed to disclose that much of that EBITDA growth would be due to one-time "purchase accounting benefits" in the wake of the Merger, and not due to improved operational results.
Plaintiffs introduced considerable evidence that tended to support such an inference. Among other documents on this topic, plaintiffs introduced an internal company document in which a major subsidiary, Universal Music Group ("UMG"), appeared to track the impact of purchase accounting on its 2001 EBITDA figures and to show what those figures would be absent purchase accounting. ( See PX-902; see also Tr. 2686:16-2687:1 (Mintzer testimony).) The jury was shown an email from Nick Henny, the CFO of UMG, dated October 11, 2001, in which he stated that "[o]n a year to date basis, we would have shown an EBITDA decline of 13%, but for the purchase accounting benefit." (PX-626.) Similarly, plaintiffs introduced an email from a member of Vivendi's Board, Edgar Bronfman, Jr., to various UMG executives, dated January 11, 2001, in which he made the following statement with respect to UMG's 2001 results: "EBITDA achievement occurred with huge purchase accounting benefits. Therefore, what really happened is that revenues were down .8% — but EBITDA was down approx 15%, meaning that our margins are deteriorating dramatically." (PX-709.) Vivendi's public statements, however, never stated that its margins were deteriorating dramatically, or that the only way it was able to achieve EBITDA growth was due to purchase accounting benefits. Plaintiffs' accounting expert testified that in his opinion, approximately half of the 35% EBITDA growth that was projected as an objective in October 2000, and then reported as results actually achieved in 2001 and 2002, was not due to operational success, but instead to one-time purchase accounting benefits. ( See Tr. 2234:1-2236:1; PX-1287.) The jury could have concluded based on this testimony and evidence that Vivendi's statements regarding its EBITDA growth rate were materially misleading:
See also PX-704 ("We are up about 17 percent on EBITDA on flat sales. About half of the profit increase over the prior year is the result of purchase accounting benefits and FX gains.")
Vivendi, in rebuttal, offered testimony to the effect that the whole market knew that Vivendi was using purchase accounting in the wake of the Merger and, indeed, that it was required by accounting rules to use purchase accounting. But the second point was never really in dispute at trial. And plaintiffs' argument about Vivendi's failure to disclose the extent of purchase accounting benefits did not rise or fall on whether it violated GAAP as a technical matter. Though plaintiffs did argue that this non-disclosure was a GAAP violation ( see, e.g., Tr. 2155-56 (Mintzer identifies failure to disclose purchase accounting benefits as one of Vivendi's violations of the accounting rules)), plaintiffs' broader point was simply that it was misleading for the company to trumpet its + 35% EBITDA growth rate — "the highest growth rates in the industry" — without disclosing that a significant portion of the growth would be due to one-time purchase accounting effects of the Merger, rather than operational growth. (Hannezo referred to this internally as "accounting magic." (PX-145).)
Vivendi also pointed to market analyst reports acknowledging that purchase accounting would enhance Vivendi's EBITDA ( see, e.g., DX-957, PX-1088) and argued that these documents proved that the market fully understood the alleged impact of the "purchase accounting benefits." But Hannezo himself had written prior to the merger that "the analysts will not have it easy to track purchase accounting benefits." (PX-654.) The jury could therefore have concluded that even if analysts may have been aware that purchase accounting would have some positive impact on Vivendi's EBITDA, the market was not aware of how much of an impact purchase accounting would have, making the 35% projection materially misleading to the extent that it failed to give any sense of the degree of the purchase accounting impact. Vivendi further argued that any impact of purchase accounting on its EBITDA had no impact on its liquidity because EBITDA is not cash. Though plaintiffs' expert, Mr. Mintzer, conceded on cross-examination that "EBITDA does not mean cash" (Tr. 2588:9-10), he testified that the failure to disclose the impact of purchase accounting benefits was related to Vivendi's liquidity situation because it "directly impacted financial information fundamental, or basic, to the assessment of Vivendi's liquidity." ( See Tr. 2156:23-2157:4.) In the end, it was up to the jury to weigh this evidence. It was certainly reasonable for them to find that Vivendi's touting of its EBITDA growth rate was materially misleading and concealed Vivendi's true liquidity condition.
Plaintiffs launched a further attack on Vivendi's earnings disclosures and claimed that the company engaged in blatant earnings management to boost reported earnings. Internal memoranda were put in evidence which could be read as showing that Messier applied pressure on his managers to "stretch" the numbers (PX-345), to meet targeted growth rates and thereby avoid depressing stock price. (PX-414) (" . . . update 3 was a total floor. We have to stay at it and cannot accept any negative variance . . ."). Other exhibits indicated that managers bent to this pressure. (PX-226.) ("[W]e had inflated Q2 2001 Ebitda by 70 M€ in order to show a growth in H1 2001 over H1 2000. . . ."). And, Hannezo, in another colorful note to Messier, complained of the pressure to improve the numbers: "I beg for mercy. We cannot, after everyone has met his stretched targets for Q2 and 24 hours before we send the information to the audit committee, re-stretch the targets and remake everyone's closing." Again, Vivendi presented testimony to attempt to put a good face on the documents; yet the jury was entitled to consider the conflicting evidence and draw its own inferences.
With respect to the Maroc Telecom acquisition, the issue for the jury was whether Vivendi had a secret commitment to purchase an additional interest in this Moroccan telecommunication company for €1.1 billion. Vivendi conceded that there was an undisclosed written commitment but argued that it did not have to be disclosed during 2001 as Messier and the King of Morocco had an oral side agreement that the terms of the €1.1 billion would be modified. Vivendi and the King of Morocco in fact announced a revised agreement in 2002. A jury, however, could have concluded there was no modification of the commitment until well into 2002 and that Vivendi did not wish to disclose the existence or amount of its original commitment throughout 2001 because it would have highlighted for the rating agencies the company's persistent liquidity problems. Again, the basis for such an inference is found in another (handwritten) note from Hannezo to Messier written in early 2002: "After today's conference call, it is time to sound the alarm. The facts: We are committed to buy 15% of Maroc Telecom before the end of February for €1.1 billion. This commitment is found in the same document that justifies consolidation. This commitment was not disclosed in the H1 financial statements." (PX 284) (emphasis in original.)
There are similar discrepancies between internal documents and the testimony of Vivendi witnesses regarding the acquisition of a Polish company, Telco. Plaintiffs contended that Vivendi did not want to consolidate its income statement with Telco because the rating agencies were pressing Vivendi to sell its interest in Telco to decrease debt. (PX-1039; PX-539). Plaintiff's introduced expert testimony to the effect that Vivendi owed 51% of Telco but concealed its control by parking 2% of the stock with another entity through a so-called "portage agreement." Vivendi witnesses testified that there was no portage agreement, that Vivendi controlled only 48% of Telco and, therefore, that consolidation was not proper. But again there were internal documents which pointed to control (PX-755; PX-922; PX-526) ("Following the closing, we will control 51% of Telco.")); which underscored Vivendi's intent to avoid consolidation (PX-539) (e-mail from Hannezo to Luczycki and Gibert, dated Jan. 30, 2002: "CONSO ELECTRIM [TELCO] NOOOOO NEVER. DG please advise. URGENT."); and which acknowledged that consolidation should have been disclosed (PX-634) (e-mail from Luczycki to Gibert) ("Since you now have a 51% interest in the holding company, with control, Electrim must be consolidated in U.S. GAAP.)
Finally, plaintiffs' contend that Vivendi further concealed its liquidity risk by the way it reported over €600 million in current account loans from its majority-owned subsidiary, Cegetel. Cegetel's minority shareholders had an effective right to call the loan at any time and plaintiffs argued that this fact was not properly disclosed. As Vivendi's liquidity problems increased in June, 2002, a dispute between Vivendi and the minority shareholders erupted resulting in the repayment of the intercompany loan which exacerbated the financial crisis. (PX-1036)
The foregoing recitation merely scratches the surface of the large body of evidence that was introduced over the course of the trial that could reasonably support inferences that Vivendi made material misstatements and omissions throughout the Class Period. The Vivendi defendants collectively put on a spirited defense, as evidenced by the verdict for individual defendants' Messier and Hannezo, but the Court finds little difficulty in concluding that the jury was well within its prerogatives in rejecting the company's defenses and finding that material misstatements and omissions were made by Vivendi.
C. Scienter
Vivendi also seeks judgment as a matter of law on the grounds that plaintiffs failed to introduce sufficient evidence to enable a reasonable juror to conclude that Vivendi acted with scienter. Scienter is "a mental state embracing intent to deceive, manipulate, or defraud." S. Cherry St. LLC v. Hennessee Group LLC, 573 F.3d 98, 108 (2d Cir. 2009) (quoting Tellabs v. Makor Issues Rights, Ltd., 551 U.S. 308, 318 (2007)). The scienter element can be satisfied by proving recklessness, but not by showing that the defendant acted accidentally, mistakenly, or negligently. Id. at 109. When the defendant is a corporate entity, the law imputes the state of mind of the employees or agents who made the statement(s) to the corporation. See Teamsters Local 445 Freight Div. Pension Fund v. Dynex Capital Inc., 531 F.3d 190, 195 (2d Cir. 2008) ("To prove liability against a corporation, of course, a plaintiff must prove that an agent of the corporation committed a culpable act with the requisite scienter, and that the act (and accompanying mental state) are attributable to the corporation.").
A defendant's conduct is reckless if it is "highly unreasonable and . . . represents an extreme departure from the standards of ordinary care to the extent that the danger was either known to the defendant or so obvious that the defendant must have been aware of it." Kalnit v. Eichler, 264 F.3d 131, 142 (2d Cir. 2001) (quoting In re Carter-Wallace, Inc. Sec. Litig., 220 F.3d 36, 39 (2d Cir. 2000)).
Vivendi argued in its Rule 50(a) motion that there was insufficient evidence to support a finding of scienter against Messier or Hannezo, or derivatively against Vivendi. Vivendi now renews its argument relying heavily at this juncture on the jury's finding that Messier and Hannezo were not liable for securities fraud. In essence, Vivendi argues that because the jury exonerated Messier and Hannezo, it could not have found that Vivendi acted with scienter. In making this argument, Vivendi distinguishes between two classes of statements: (1) the eighteen statements on Table A that were not attributed to Messier or Hannezo and as to which the jury was asked to rule only upon Vivendi's liability; and (2) the thirty-nine statements on Table A of the Verdict Form that were publicly attributed to either Messier or Hannezo (or both), and as to which the jury was asked to rule upon the liability of Vivendi and Messier and/or Hannezo. As to the eighteen statements not publicly attributed to a specific person at Vivendi, Vivendi contends that it cannot be liable for these statements because plaintiffs failed to identify the speaker who made the statements or to introduce any evidence regarding the state of mind of any speaker other than Messier or Hannezo. As to the remaining thirty-nine statements, Vivendi begins by assuming that the jury must have found that Messier and Hannezo did not act with scienter (since, according to Vivendi, scienter is the only element upon which the jury could have based its differing verdicts), and then argues that because Vivendi's scienter necessarily depends on the state of mind of the individuals who made the statements, a jury finding that Messier and Hannezo did not act with scienter precludes a finding of scienter against the company.
1. Statements Not Specifically Attributable to Messier or Hannezo
By way of background, seventeen of the statements not specifically attributable to Messier or Hannezo were from Vivendi press releases relating to financial matters. The other statement — Statement No. 55 — was a quotation that was attributed to "Vivendi Universal spokesperson Antoine Lefort" in an article published by Bloomberg on June 20, 2002 regarding Vivendi's sale of shares in its water business to Deutsche Bank. (PX-1154.)
The Court finds that Vivendi is entitled to judgment as a matter of law with respect to Statement No. 55. This statement was made by one Antoine Lefort, but plaintiffs did not introduce any evidence whatsoever regarding Lefort's knowledge or state of mind. Nor did they introduce any evidence that anyone at Vivendi endorsed or authorized this statement, or that any actor at Vivendi who had scienter directed Lefort to make this statement. There was no evidence that Messier or Hannezo had any involvement in Lefort's provision of this statement to the journalist. It is not even clear from the record who Lefort is, what his role at Vivendi was, and whether he would have had any reason to believe that his statement was false. The record simply reflects that a journalist from Bloomberg obtained a quote from Lefort and included it in an article about Vivendi. Plaintiffs have failed to prove scienter with respect to this statement. See Teamsters Local 445, 531 F.3d at 195; Southland Sec. Corp. v. INSpire Ins. Solutions Inc., 365 F.3d 353, 366 (5th Cir. 2004) ("[W]here, as in fraud, an essentially subjective state of mind is an element of a cause of action also involving some sort of conduct, such as a misrepresentation, the required state of mind must actually exist in the individual making (or being a cause of the making of) the misrepresentation. . . .").
However, the Court's ruling in Vivendi's favor with respect to Statement No. 55 does not require the Court to upset the jury's daily inflation figures. In this case, the jury calculated the daily inflation in Vivendi's stock prices by endorsing Dr. Nye's overall approach to damages, but reducing his figures across the board. Dr. Nye's analysis did not link changes in inflation to each misstatement on Table A since his opinion was based on a theory that Vivendi had concealed its liquidity risk throughout the Class Period. ( See, e.g., Tr. 3862:16-19). Under Dr. Nye's analysis, whether Statement No. 55 was materially misleading or not was of little relevance, and would not alter his conclusions regarding the inflation in Vivendi's share price — indeed, Statement No. 55 was made on a date upon which inflation remained constant in Dr. Nye's analysis. (PX-1486.) Since the jury followed Dr. Nye's overall approach, there is nothing to suggest that the jury would have modified its inflation figures if it had found Statement No. 55 to be inactionable. Therefore, the Court will not overturn the jury's daily inflation findings on this basis.
The Court now turns to the seventeen statements on Table A which came from Vivendi press releases and were not publicly attributed to a specific actor at Vivendi. Vivendi argues that plaintiffs failed to identify any speaker for these statements, and that this precludes a finding that Vivendi violated Section 10(b) by making it impossible to determine which Vivendi agent allegedly acted with scienter. However, Vivendi's contention that plaintiffs failed to identify a speaker for these statements is premised on an unduly narrow notion of what it means to identify the speaker of a particular statement. Many (perhaps most) press releases by corporations are drafted and reviewed by numerous individuals before they are publicly released. Many press releases are unsigned. Statements in such press releases cannot necessarily be pegged to a single "speaker." But that certainly does not mean that nobody made the statement, and cannot preclude Section 10(b) liability against the corporation if it can be shown that one or more of the persons who played a meaningful role in drafting or reviewing the press release knew it was false and did not correct it before it was issued. See Southland, 365 F.3d at 366 ("For purposes of determining whether a statement made by the corporation was made by it with the requisite Rule 10(b) scienter we believe it appropriate to look to the state of mind of the individual corporate official or officials who make or issue the statement (or order or approve it or its making or issuance, or who furnish information or language for inclusion therein, or the like). . . .") (emphasis added); id. at 365 ("[C]orporate documents that have no stated author or statements within documents not attributed to any individual may be charged to one or more corporate officers provided specific factual allegations link the individual to the statement at issue. Such specific facts tying a corporate officer to a statement would include a signature on the document or particular factual allegations explaining the individual's involvement in the formulation of either the entire document, or that specific portion of the document, containing the statement.").
Vivendi would ask this Court to adopt a rule that would require a statement to be publicly attributed to a particular agent of the corporation in order for that agent to be the "speaker" of the statement. But that would eviscerate Section 10(b). It would enable companies to avoid Section 10(b) liability by issuing all press releases unsigned, since any statements therein (other than quotations) would have no speaker. Alternatively, companies could ensure that all press releases go out under the name of a secretary in the public relations department who has no idea whether the substance of the press release is correct or not. The secretary would then be the "speaker" but would lack scienter, making the statement inactionable. That is certainly not the law.
In this case, plaintiffs presented evidence from which the jury could have concluded that Messier and/or Hannezo actively participated in making, reviewing and/or authorizing the statements in Vivendi's press releases. There was evidence in the record that both Messier and Hannezo participated extensively in drafting press releases relating to financial matters. See, e.g., PX-807 (email from Messier to Hannezo dated Sept. 21, 2001 discussing wording of press release); PX-112 (Messier handwritten edits to press release); PX-107 (Hannezo handwritten edits to press release); PX-1434 (email noting that that Vivendi employees were "awaiting feedback" from Hannezo on the 2001Q2 earnings release). Hannezo and Messier both testified that the communications department at Vivendi would coordinate the drafting of press releases by soliciting information from different business units and departments at Vivendi, and that numerous senior officials at Vivendi — including Messier, Hannezo, Vivendi's counsel, and often its external auditors — would review, comment on, and revise press releases before they were issued. ( See, e.g., Tr. 1475:6-1478:6; Tr. 1483:1-9; Tr. 4447:2-4448:22.)
The evidence of Messier's lead role in the issuance of all press releases was substantial. Hannezo testified that Messier had the authority to issue press releases on behalf of Vivendi, "sometimes after approval of the board." (Tr. 1541:9-11.) Messier testified that the Board had the final say over earnings press releases and press releases relating to major transactions (Tr. 4452:21-24), but his testimony, viewed as a whole, could easily have been interpreted as showing that he endorsed each of Vivendi's press releases in their final form. For example, Messier testified that after the Board had reviewed the press releases, he would inform Catherine Gros or Anita Larsen at Vivendi to "take [the Board's comments] into account and then issue the press release[s]," (Tr. 4453:2-5.), and repeatedly testified about the press releases in a manner that suggested that he was in charge of them and that they fully reflected his views. ( See, e.g. Tr. 4251:1-2; Tr. 4299:7-10; 4392:20-4393:6).
The fact that the Board had final say over certain press releases — those relating to earnings and major acquisitions — does not undermine a conclusion that Messier and/or Hannezo could have supplied Vivendi's scienter with respect to misleading statements in press releases that relate to financial matters. Plaintiffs argued at trial that Messier and Hannezo concealed information from the Board. (Tr. 7364:6-10.) If the jury believed that information was concealed from the Board, and believed that Messier and/or Hannezo acted with scienter in connection with the issuance of these press releases, then the Board's approval of certain press releases would not negate the existence of Messier and/or Hannezo's scienter.
Hannezo testified that he personally had the opportunity to read, verify, and comment on the information in press releases "when they were dealing with financial matters." (Tr. 1541:19-22.) Hannezo emphasized that he did not have the final say over what the press releases said, (Tr. 1541:23-1543:12); however, Hannezo did not identify or allude to any instance in which a press release on financial matters was modified after he had given his comments in such a way that the release no longer accurately presented the financial information that he had sought to convey. The jury could reasonably have concluded that Hannezo endorsed the financial aspects of all of Vivendi's press releases. Based on all of this evidence, the jury could have concluded that Messier and/or Hannezo made approved or authorized the challenged statements in Vivendi's press releases even if the statements were not publicly attributed to them in the form of a quotation, thereby supplying Vivendi's scienter for those statements. That, of course, raises the question of whether there was sufficient evidence that Messier and/or Hannezo in fact acted with scienter in making the statements on Table A (other than Statement No. 55), an issue the Court turns to next.
2. Impact of the Messier and Hannezo Verdicts
In focusing on the verdicts in favor of Messier and Hannezo, rather than the trial evidence, Vivendi misunderstands the nature of a motion for judgment as a matter of law under Rule 50. The Court's task in evaluating a Rule 50 motion is to look at the trial evidence and assess whether that evidence was sufficient to support the verdict. See Fed.R.Civ.P. 50(a) ("If a party has been fully heard on an issue during a jury trial and the court finds that a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue, the court may . . . grant a motion for judgment as a matter of law. . . ."); Zellner, 494 F.3d at 370-71. The Court's task on a Rule 50 motion is not to examine different aspects of the jury's verdict to determine whether they can be logically reconciled with one another. In this case there was more than enough evidence from which a reasonable jury could have found that Vivendi, acting through Messier and Hannezo, acted with scienter. Indeed, the evidence in this case could reasonably have supported a finding of scienter against each of the defendants. The fact that the jury absolved Messier and Hannezo of liability does not negate the fact that there was sufficient evidence in the record in the first instance to enable a reasonable jury to find against all three defendants on the issue of scienter, thereby foreclosing judgment as a matter of law in Vivendi's favor. See Fed.R.Civ.P. 50(a).
The evidence in the record that supported a finding of scienter overlaps to a considerable extent with the evidence discussed in the section on material misstatements or omissions. The same evidence of the stark contrast between Vivendi's internal documents and its external statements could equally have supported an inference that Vivendi (acting through Hannezo, Messier, and others) acted with scienter at the time it made the public statements on Table A. By way of example, internal memoranda that acknowledge the use of purchase account to achieve "the highest growth rates in the industry" while actually masking a decline in EBITDA in certain operations could support the inference that (i) Messier and Hannezo were aware the public would have difficulty discerning the impact of purchase accounting on Vivendi's EBITDA; and (ii) that they therefore knew that it would be misleading to report EBITDA targets and results without informing the public of the extent of the impact purchase accounting was having on those figures; and (iii) that they nonetheless signed off, whether recklessly or deliberately, on the F-4 and numerous other statements reporting on Vivendi's achievement of EBITDA targets which omitted such disclosure.
The Book of Warnings also contained several other documents that could have supported an inference of scienter against Messier and Hannezo when juxtaposed against Vivendi's contemporaneous public statements. Compare PX-159 (February 7, 2001 memo from Hannezo to Messier) ("Our banks are at their limit. They are beginning to get worried . . .") with PX-680 (February 14, 2001 press release) ("Vivendi Universal enters its first full year of operations with strong growth prospects and a very strong balance sheet . . ."); compare PX-128 (e-mail from Hannezo to Messier dated March 1, 2002) ("Debt is supposed to be repaid after all. What do we do when we generate negative cash flows at the central level that are barely offset by inaccessible cash flows at the minority level, yet we continue to make acquisitions because, like a collector, we don't know how not to look, and selling is torture?) (emphasis added) with PX-295 (Vivendi press release dated March 5, 2001) ("Vivendi Universal has reached of exceeded all of its operational targets in 2001 . . . Operating Free Cash Flow of 2.026 billion euros, ahead of guidance (1.2 to 1.5 billion euros) and up 2 billion euros over 2000.").
Since Messier received Hannezo's February 7, 2001 memo and since it was admitted against him, the juxtaposition of PX-159 and PX-680 could also have led the jury to conclude that Messier had scienter.
Similarly, a jury could have concluded that both Hannezo and Messier knowingly or recklessly participated in concealing financial commitments to Maroc Telecom and thereby misled the public as to Vivendi's growing liquidity crisis. It was undisputed at trial that in February 2001, Vivendi had entered into an agreement with the Kingdom of Morocco to purchase an additional 16% stake in Maroc Telecom before February 2002 for €1.1 billion. Plaintiffs introduced evidence that could have supported an inference that Messier and Hannezo deliberately or recklessly concealed this from the public. For instance, plaintiffs introduced a memo dated February 17, 2001 written by Hannezo to Messier which discussed Vivendi's agreement to purchase an additional 16% in Maroc Telecom and acknowledged that Vivendi was planning to keep the agreement a secret. See PX-661 ("We have absolutely committed ourselves not to mention, even indirectly . . . the existence of this agreement . . .") (emphasis in original). Just weeks later in a press release announcing Vivendi's 2000 earnings, Vivendi announced its acquisition of a 35% stake in Maroc Telecom for €2.3 billion, but omitted to disclose the agreement to purchase the additional 16% for €1.1 billion. (PX-342). Vivendi continued to make statements about Maroc Telecom throughout 2001 that omitted to disclose the €1.1 billion commitment. ( See, e.g., PX-701; PX-280). The jury could easily infer from this evidence that Messier and/or Hannezo acted with scienter by omitting to disclose the €1.1 billion commitment, thereby concealing Vivendi's true liquidity condition.
As noted, Vivendi contends that the "unrefuted evidence" shows that the Kingdom of Morocco orally promised Vivendi it would not have to pay the €1.1 billion under the 2001 agreement. ( See, e.g., Tr. 1513:9-12 (Hannezo testimony); Tr. 4465:10-4466:25 (Messier testimony).). But Messier testified that the alleged oral promise was made in June 2001, and his testimony does not therefore explain the failure to disclose the commitment in public statements about the agreement in March 2001. ( See PX-342.) And Hannezo himself had written to Messier in November 2001 that "we still have the pay the Morocco stage 2," (PX-145), which appeared to contradict his testimony that from about September 2001 onwards, he believed that Kingdom of Morocco had orally promised that Vivendi would not have to repay the debt. ( See Tr. 1513:9-12.) The jury could easily have concluded that both Hannezo and Messier acted with scienter in connection with Vivendi's statements surrounding Maroc Telecom.
Vivendi argues that it was the Kingdom of Morocco, not Vivendi, that insisted that the agreement to purchase the additional 16% not be disclosed, as if to suggest that this fact, if true, would negate a finding of scienter. But if Hannezo and Messier knew that it would be materially misleading not to disclose the agreement and went along with a cover-up anyway, or acted recklessly in making statements that failed to disclose that agreement, the fact that they did so at somebody else's bequest would not negate their own scienter.
There was also other evidence that could have supported an inference that Messier acted with an intent to defraud by ordering the heads of various business units within Vivendi and Hannezo to inflate their financial results so that Vivendi would meet its stated targets. For example, in early 2001, Messier wrote an email to Hannezo in which he stated,
If we want to depress our share price by 20% we do that without asking ourselves further questions! Update 3 was a total floor. We have to stay at it and cannot accept any negative variances . . . You cannot accept any change/update 3 . . . which could jeopardize those targets.
I am sorry but at this stage we need to go back to work and interact again with all [Business Units] and perhaps decide not to take some decisions in 2000 even if it would have been better for the future. But with those figures and the current volatility of the market, March 9th would be a [illegible] disaster.
We clearly need to regain 200M€ at operating income and net income level and 350M€ at EBITDA and current income level.
(PX-415) (underlining in original); see also PX-227 (noting that "nothing would be more detrimental" than to miss a the target). Similarly, on July 17, 2001, while Vivendi was closing its second quarter 2001 results, Messier wrote a memo to Hannezo and John Luczycki stating the following: "Target EBITDA growth Q2 has to remain in the 30+ — 35% range need of 20 to 70m€ more." (PX-57). In both instances, Vivendi announced shortly after Messier wrote these documents that it had met its stated targets for that reporting period. The jury could have interpreted these documents as showing that Messier was asking his subordinates to inflate their earnings to prevent Vivendi from missing its targets. Messier testified that in writing these emails, he was merely challenging the heads of the business units to "reassess their business judgment as many
times as needed to get to the proper account." (Tr. 4391:10-11; 4391:14-20.) But the jury was not required to believe him. In context of the entire case, the documents themselves were sufficient to enable a rational jury to conclude that Messier knowingly directed his subordinates to inflate Vivendi's earnings to make Vivendi's liquidity position look stronger than it actually was.
A thorough review of the trial transcripts reveals numerous additional examples. In short, there was an abundance of evidence from which a reasonable jury could infer that more likely than not, Messier, Hannezo, and Vivendi acted with scienter by knowingly or recklessly making, authorizing, or participating in the making of misleading public statements that suggested that the company was in good financial health and failed to disclose its deteriorating liquidity condition.
Vivendi's position appears to be that all of this evidence of scienter is nullified by the fact that Messier and Hannezo both testified throughout the trial that they acted in good faith. But it is hard to imagine a case ever making it to trial in which the defendants did not profess to have acted in good faith. Both Hannezo and Messier were interested witnesses, and the jury could have rejected their testimony as self-serving in light of all of the evidence in the record, particularly the large number of contemporaneous documents that on their face appeared to suggest impropriety. See Kerman, 374 F.3d at 123.
Vivendi attempts to cast aside all of the compelling evidence that could have supported an inference of scienter against all three defendants, arguing that "[t]he verdict [absolving Messier and Hannezo of liability] confirms . . . that Vivendi correctly argued that there was no proof that it acted with scienter." (Def. 50(b) Reply at 1) (emphasis added). But the verdict did no such thing. The verdict simply demonstrates that the jury concluded that plaintiffs did not prove by a preponderance of the evidence that Messier and Hannezo violated Section 10(b) based on the evidence that was admissible against Messier and Hannezo. The verdict does not (and could not) establish that there was "no proof" that Vivendi acted with scienter. Having reviewed the actual evidence in the record, the Court finds that the evidence was plainly sufficient to support a finding that Vivendi, through Messier and/or Hannezo, acted with scienter.
It appears that Vivendi's real argument with respect to scienter is that the verdict was inconsistent to the extent that the jury found Vivendi liable, but exonerated Hannezo and Messier. Vivendi disputes this characterization of its argument by employing an overly narrow definition of what constitutes an inconsistent verdict. Vivendi contends that in order to qualify as an inconsistent verdict, the jury's finding on one claim must negate an essential element of another claim against the same defendant, relying on a statement by the Second Circuit in Kosmynka v. Polaris Indus., Inc., 462 F.3d 74, 86 (2d Cir. 2006) that "[u]nder New York law, a verdict is inconsistent if a jury's finding on one claim necessarily negates an element of another cause of action." (internal quotation marks and citation omitted). To be sure, a verdict in which a jury's finding on one claim negates an element of another cause of action against the same defendant is one form of inconsistent verdict.
But courts routinely analyze other types of arguments as inconsistency challenges where the basic nature of the argument is that the jury's factual findings and/or legal conclusions are logically irreconcilable with each other, even if the jury's findings on one cause of action do not necessarily negate an essential element of another claim against the same defendant. Of particular relevance here, courts regularly analyze challenges to split verdicts among multiple defendants as inconsistency challenges. See, e.g., Zhang v. Am. Gem Seafoods, Inc., 339 F.3d 1020, 1032 (9th Cir. 2003) (analyzing challenge to jury's finding that a corporation was liable for discrimination but that its agent was not as an inconsistency challenge); Doe By and Through Doe v. Washington County, 150 F.3d 920, 923-24 (8th Cir. 1988) (analyzing challenge to jury verdict that county was liable on § 1983 claim but county sheriff was not as an inconsistency challenge); de Feliciano v. de Jesus, 873 F.2d 447, 452 (1st Cir. 1989) (noting that situations "where a jury returns verdicts in favor of an employee defendant, but against an employer whose liability was derivative of the employee's liability" raise "`inconsistency' problems"); Int'l Longshoremen's Union v. Hawaiian Pineapple Co., 226 F.2d 875, 881 (9th Cir. 1955) (analyzing challenge to jury's finding exonerating individual union leaders but finding the defendant unions liable as an inconsistency challenge). Thus, Kosmynka does not set forth the exclusive test for determining whether a verdict is inconsistent, as Vivendi suggests; instead, Kosmynka simply provides an example of a verdict that it inconsistent because it contains findings, whether legal or factual, that cannot be reconciled with other findings therein. In this case, the challenge that Vivendi brings to the jury's scienter findings is fairly characterized as an argument that the verdict was inconsistent to the extent that the jury found Vivendi liable but exonerated Messier and Hannezo.
3. Whether Vivendi is Entitled to a New Trial Based on the Alleged Inconsistency in the Verdict
An inconsistent verdict is a possible ground for a new trial, but not for entry of judgment as a matter of law. See Kosmynka., 462 F.3d at 87; Tolbert, 242 F.3d at 74. Here, however, the Court finds that Vivendi is not entitled to a new trial based on the alleged inconsistency in the verdict because Vivendi waived its right to bring such an objection by waiting until its post-trial motion to do so and, more importantly, finds that the verdict is not actually inconsistent.
(a) Vivendi Waived Its Right to Object to the Verdict on Inconsistency Grounds
"It is well established that a party waives its objection to any inconsistency in a jury verdict if it fails to object to the verdict prior to the excusing of the jury." Kozmynka, 462 F.3d at 83; Jarvis v. Ford Motor Co., 283 F.3d 33, 56-57 (2d Cir. 2002) (party waived objection to inconsistency between two general verdicts by failing to object to the jury instructions or verdict form before the jury retired). In this case, Vivendi is arguing that the jury in effect made inconsistent scienter findings with respect to the three defendants, yet Vivendi never objected to the jury charges or the Verdict Form on the ground that they could produce potentially inconsistent verdicts or scienter findings ( see Dkt. 994 (Defendants' Revised Objections to the Court's Proposed Jury Instructions (hereinafter "Def. Revised Objections")).
The Court here cites to the version of Vivendi's objections to the Court's jury charges that it considers to be the operative version for purposes of the record in this case. In Vivendi's motion papers, it cites a prior version of its objections entitled "Defendants' Objections and Responses to Judge Holwell's Proposed Charge to the Jury," which Vivendi filed at approximately 5 a.m. on January 4, 2010 (hereinafter "the Jan. 4 Objections"). ( See Dixon Decl. Ex. 26 at 20-86; see also Dkt. 989). The Court notes, for the sake of clarity of the appeal record, that the Court did not accept that filing from Vivendi because it raised a number of objections that Vivendi had not properly brought to the Court's attention pursuant to the Court's procedure for soliciting comments on the Court's proposed jury charges. ( See Tr. 6747:2-24.) That procedure was as follows: Prior to the trial, to Court solicited proposed jury charges from all parties and objections to the proposed jury charges. Near the close of the trial, the Court asked the parties to submit supplemental proposed jury charges. The parties submitted their supplemental jury charges on December 14, 2009. The Court reviewed the parties' pre-trial and December 14 submissions, drafted its own proposed jury charges, and circulated the first draft of its proposed jury charges to all counsel on the evening of December 21, 2009. On the evening of December 21, 2009, the Court also received an unsolicited submission from defendants containing defendants' objections to plaintiffs' December 14, 2009 proposed supplemental jury charges. The Court held an informal (off-the-record) charging conference on December 22, in which all counsel were asked to identify their objections to the Court's proposed jury charges (which had been circulated the previous evening). After the first charging conference, the Court submitted revised jury charges to the parties and a proposed verdict form. Over the next two weeks, the Court's proposed jury charges and verdict form went through a number of revisions in response to objections raised by the parties at additional informal charging conferences on December 23 and 29 and in related correspondence. Early on Monday, January 4, 2010, Vivendi filed the Jan. 4 Objections, which purported to be Vivendi's final objections to the Court's jury charges. The Court reviewed the Jan. 4 Objections and found that the document contained objections to the Court's jury charges that had not been raised at any point since the Court first circulated its proposed draft jury charges to counsel on December 21, 2009. The Court, therefore, declined to accept the Jan. 4 Objections and asked defendants to "resubmit their objections and identify those that are new and those that were made during the three charging conferences that we had." (Tr. 6747:20-22.) Defendants heeded the Court's request and re-filed their objections late on the night of January 6, 2010 (and submitted hard copies to the Court on the morning of January 7, 2010.) ( See Dkt. 994 (Defendants' Revised Objections to the Court's Proposed Jury Instructions). Defendants' Revised Objections contained a sentence on the first page stating that defendants "hereby incorporate by reference and are not waiving all previous objections in correspondence and in pleading." ( Id. at 1) On January 11, 2010, the Court rejected aspects of defendants' Revised Objections. (Tr. 7432:18) In doing so, the Court noted that "to the extent that [the above-quoted sentence] is an attempt to preserve objections to the court's proposed charges that were neither raised during the three charging conferences, on December 22, 23 and 29th, nor in correspondence addressed to the court's proposed charges, the court considers such objections to be untimely and they have been waived." ( Id. at 7432:12-17 (emphasis added).)
The one objection that Vivendi did make was to the Verdict Form's use of one question on liability for each of the fifty-seven challenged statements, followed by a second question on the element of scienter. Vivendi's specific objection to this part of the Verdict Form was that a single "up or down" question on liability for each statement "unnecessarily aggregates four distinct findings the jury has to make into one, taking the jury's focus off the essential elements Plaintiffs must prove and placing the jury's focus instead upon ultimate liability." (Dixon Decl. Ex. 26 (1/4/10 Saunders Ltr.) at 3-4; see also 12/23/09 Saunders Ltr. at 1-2 (objecting to the absence of any specific questions regarding loss causation).) Vivendi also objected that an overall question on liability followed by a specific question on scienter would "confuse and mislead the jury," but did not argue that this could lead to inconsistent verdicts. ( Id.) Having failed to argue that the Verdict Form might produce inconsistent scienter findings, Vivendi cannot now argue that it preserved such an objection by making a different objection to the same part of the Verdict Form. See Fed.R.Civ.P. 51 (any objection must "stat[e] distinctly the matter
The reason the Court included a separate question on scienter after its general "up or down" question on liability was to comply with the PSLRA, 15 U.S.C. § 78u-4(f)(3), which requires the Court to instruct the jury to answer special interrogatories regarding, inter alia, whether any defendant found to have violated Section 10(b) committed a knowing violation. This requirement is present because a defendant can only be jointly and severally liable for an entire jury award in a multiple defendant case if they are found to have acted knowingly (rather than recklessly). See id. at § 78u-4(f)(2)(A).
objected to and the grounds of the objection"); Jarvis, 283 F.3d at 54 (finding waiver of inconsistency challenge on the ground that "[defendant's] pre-trial statement that the court should charge either negligence or strict liability, but not both, failed to alert the court to the precise nature of [defendant's] objection and its legal grounding").
Vivendi also failed to object to the alleged inconsistency after the verdict was read and before the jury was discharged, which further supports the Court's finding of waiver in this case. Kosmynka, 462 F.3d at 83; DiBella v. Hopkins, 403 F.3d 102, 117 (2d Cir. 2005) (plaintiff waived objection that verdicts were inconsistent by failing to object while the jury was still empanelled instead objecting for the first time in his posttrial motion).
(b) The Verdict is Not Inconsistent
Even if Vivendi had not waived its inconsistency challenge, it would fail on the merits. It is well-settled that "[w]hen confronted with a potentially inconsistent jury verdict, the court must adopt a view of the case, if there is one, that resolves any seeming inconsistency." Turley v. Police Dep't of the City of New York, 167 F.3d 757, 760 (2d Cir. 1999) (internal quotation marks and citation omitted); see also Gallick v. Baltimore Ohio R.R. Co., 372 U.S. 108, 119 (1963); Vichare v. AMBAC, 106 F.3d 457, 464-65 (2d Cir. 1996) (collecting cases). Thus, an inconsistency challenge to a verdict will succeed only if the Court is unable to determine any reasonable way to reconcile the jury's findings. See Turley, 167 F.3d at 760.
Here, it is possible to reconcile the jury's finding that Vivendi was liable but Messier and Hannezo were not based on the different evidence admissible against each of the three defendants. Over half of the documents at trial were admitted against Vivendi but not against Messier or Hannezo, or were admitted against Vivendi for all purposes but against Messier and Hannezo only for limited purposes. ( See Exhibit List.) The jury reasonably could have concluded that plaintiffs had not proven the material misstatement or omission or scienter elements of their Section 10(b) claim against Messier and Hannezo based on the more limited pool of documents admitted against them, but that when the additional evidence that was admitted against Vivendi was added to the mix, plaintiffs had proven their Section 10(b) claims against Vivendi and that Messier and/or Hannezo supplied Vivendi's scienter.
Vivendi attempts to refute this proposition by arguing that none of the specific documents admitted against Vivendi but not against Messier and Hannezo can justify the jury's different verdicts. In making this argument, Vivendi starts from the premise that "scienter is the only element of Section 10(b) upon which the jury could have reached differing verdicts as to Vivendi and Messier and Hannezo," and then argues that none of documents admitted against Vivendi but not against Messier and Hannezo prove anything about Vivendi's scienter at the time the statements in question were made. (See Def. 50(b) Reply at 4-5.) But the notion that scienter is the only element upon which the jury could have based its differing verdicts is a fallacy. Vivendi contends that
Because the jury found Vivendi liable for each of the 57 statements . . . in Table A, but did not find Mr. Messier or Mr. Hannezo liable for any, it follows that the jury believed that neither Mr. Messier nor Mr. Hannezo had the requisite scienter. There can be no distinction between Vivendi and the Individual Defendants with respect to whether a statement was false or misleading. A statement is false or misleading or it is not. Either Defendants were able to rebut the presumption of reliance, or they were not. Likewise, Plaintiffs either suffered a loss as a result of the fraud or they did not. Only scienter remains as a basis for distinguishing among Defendants as to liability.
(Def. 50(b) Br. at 9 n. 4.) But it does not necessarily follow from the fact that the jury found Messier and Hannezo not liable that the jury concluded that neither Messier nor Hannezo acted with scienter. Nor is it true that there can be no distinction between Vivendi and the individual defendants with respect to whether a statement was false or misleading, since different evidence was admitted against Vivendi and the individual defendants which bore on the question of whether the statements in question were false.
Regarding scienter, significant evidence that was admitted against Vivendi, but not against Hannezo and/or Messier, could have led the jury to find that plaintiffs proved that Vivendi violated Section 10(b) based on the scienter of Hannezo and Messier, even if the jury was unable to conclude that plaintiffs had met their burden of proof against Hannezo and Messier based on the more limited pool of documents admitted against them. The clearest examples of such evidence were Vivendi's admissions in the Arbitration Documents that at the end of June 2002, "Vivendi was in the midst of an immediate and critical financial crisis threatening its very existence" (PX-1288 at 5), that Vivendi faced a "severe liquidity crisis" ( id.), and that Messier had taken Vivendi to the "brink of bankruptcy" (PX-1291 at 2). These documents were admitted against Vivendi, but not against Hannezo and Messier for any purpose — a distinction that was made abundantly clear to the jury at trial.
Other documents containing similar admissions of the liquidity crisis were admitted against Vivendi but not against Messier and Hannezo. See, e.g., PX-942 (memo from Espinasse, discussing factors that caused Vivendi's "treasury crisis"); PX-1218 (April 2003 address by Fortou to Vivendi shareholders stating that Vivendi "will never recuperate the maximum values that existed before the crisis" and that Vivendi in July 2002 was "not far from insolvency").
Vivendi contends that the admissions in the Arbitration Documents that the company faced a "severe liquidity crisis" in the summer of 2002 cannot prove that Vivendi had or concealed liquidity problems before that time, that is, during the Class Period when most of the alleged misstatements and omissions were made. This argument is flawed in two respects: it ignores the fact that the jury's task was to consider the evidence as a whole and to assess its cumulative weight, and it ignores the fact that statements made at a later point, while certainly not dispositive, may be highly relevant to establishing facts at an earlier time. As discussed above, there was significant evidence that suggested that Vivendi had liquidity problems throughout the Class Period and that Messier and/or Hannezo knowingly or recklessly concealed those risks from the public.
Defendants naturally sought to counter that inference by introducing evidence, for example, that Vivendi always paid its bills on time. After considering all the evidence that was admitted against all three defendants on this issue, the jury may have been on the fence about whether Vivendi had liquidity problems in 2001 or whether a party had acted with scienter. It would come as no surprise that the Arbitration Documents could have tipped the scales in plaintiffs' favor. While admissions that Vivendi faced a "severe liquidity crisis" in June/July 2002, do not necessarily prove that Vivendi also had liquidity problems before the summer of 2002, it is rational to infer that a liquidity crisis severe enough to threaten Vivendi's very existence did not develop overnight. Thus, the company's own admissions may have been the "final straw" that persuaded the jury that Vivendi did have undisclosed liquidity problems throughout the Class Period, despite defendants' protestations to the contrary.
Apart from the Arbitration Documents, there was plenty of other evidence that was admitted against Vivendi, but that was either inadmissible or less than fully admissible against one of both of the individual defendants, that might explain the jury's differing verdicts. For instance, highly probative documents relating to earnings management was admitted against Vivendi, but not against Hannezo or Messier. See, e.g., PX-926T (May 2, 2002 memo from Pierre Trotot to Philippe Germond ("we had inflated 2001 Q2 by nearly €70 M out of the need to present 2001 half-year earnings up from 2000 earnings (. . .!) and this operation will have to be renewed again this year."); PX-188 (Note from Pierre Trotot to John Luczycki) ("For Q2 and H1, Cegetel will be able to stretch for an additional 15 million euros.").
In a similar vein, some of the more damaging evidence relating to the impact of purchase accounting on UMG's reported EBITDA was admitted against Vivendi only. See, e.g., PX-709 ("EBITDA achievement occurred with huge purchase accounting benefits. Therefore, what really happened is that revenues were down .8% — but EBITDA was down approx 15%, meaning that our margins are deteriorating dramatically."); PX-626 ("On a year to date basis, we would have shown an EBITDA decline of 13%, but for the purchase accounting benefit.") The Court does not know precisely what logic led the jury to reach the verdict that it reached; though it does appear from the fact that the jury found Vivendi liable that the jury gave more weight to the documents admitted against it than to Messier and Hannezo's testimony. In this context, it would be bold indeed for this Court to conclude that there is no "reasonable" way to explain the differing verdicts in this case, given that over half of the documents at trial were not admitted or less than fully so, against the individual defendants. See Turley, 167 F.3d at 760. The jury could easily have found that the cumulative weight of the evidence admitted against Vivendi was sufficient to enable them to find that Vivendi violated Section 10(b), but that plaintiffs had not proven their case against the individual defendants based on the more limited pool of documents admitted against each individual defendant. Because the verdict is not inconsistent, a new trial is not warranted. Id.
Even if the Court had found an inconsistency between the jury's general verdicts that Vivendi was liable but Messier and Hannezo were not, that would not necessarily require the Court to grant a new trial. In criminal cases, it is clear that there is no prohibition against inconsistent verdicts. See United States v. Powell, 469 U.S. 57 (1984). Though the Supreme Court has not squarely addressed whether inconsistent general verdicts should be allowed to stand in civil cases, see City of Los Angeles v. Heller, 475 U.S. 796, 805-06 (1986) (Stevens, J., dissenting), many courts have held that a district court has the discretion to let inconsistent general verdicts stand in civil cases. See Zhang, 339 F.3d at 1035 ("We have found no Supreme Court or Ninth Circuit cases in which an appellate court has directed the trial court to grant a new trial due to inconsistencies between general verdicts, and Ninth Circuit precedent dictates that we cannot do so.") (collecting cases allowing apparently inconsistent verdicts to stand); Merchant v. Ruhle, 740 F.2d 86, 91 (1st Cir. 1984) (expressing a "substantial reluctance to consider inconsistency in civil jury verdicts a basis for new trials"); Jayne v. Mason Dixon Lines, 124 F.2d 317, 319 (2d Cir. 1941) (stating in dicta that it would not have been fatal to the verdict if "no rational reconciliation of the verdict was possible"); Malm v. U.S. Lines Co., 269 F.Supp. 731, 731-32 (S.D.N.Y. 1967) ("Inconsistent jury verdicts upon different counts or claims are not an anomaly in the law, which at times recognizes a jury's right to an idiosyncratic position, provided the challenged verdict is based upon the evidence and the law."). But see Diamond Shamrock Corp. v. Zinke Trumbo, Ltd., 791 F.2d 1416, 1425 (10th Cir. 1986); Will v. Comprehensive Accounting Corp., 776 F.2d 665, 677 (7th Cir. 1985); Shaun P. Martin, Rationalizing the Irrational: The Treatment of Untenable Federal Civil Jury Verdicts, 28 Creighton L. Rev 683, 708 (1995).
4. Loss Causation
Vivendi also contends that plaintiffs failed to show loss causation at trial. The legal standards governing loss causation and plaintiff's theory of loss causation were discussed in detail in this Court's summary judgment opinion. Vivendi V, 634 F. Supp. 2d 352. As set forth therein, loss causation is the required "causal link" between the alleged fraud and the subsequent decline in value of the stock when the fraud comes to light. Id. at 360. It is typically shown by the reaction of the market to a "corrective disclosure" which reveals a prior misleading statement. Id. at 363. However, loss causation may also be shown by the "materialization of risk" method, whereby a concealed risk — here, a liquidity crisis — comes to light in a series of revealing events that negatively affect stock price over time. Unlike corrective disclosures, these events do not identify prior company statements as misleading, but they must reveal new information previously concealed and fall within the "zone of risk" concealed so that the events were foreseeable consequences of the fraud. Id. at 363-64. In addition, a plaintiff must show that the loss was caused by materializations of the concealed risk and not other factors. Id. at 364.
Vivendi now makes three arguments with respect to loss causation: (a) that plaintiffs did not show a connection between the alleged fraud and the nine events that plaintiffs' expert, Dr. Nye, opined were materializations of Vivendi's undisclosed liquidity risk, (b) that plaintiffs failed to connect the nine events that Dr. Nye opined were materializations of Vivendi's undisclosed liquidity risk with any share price declines, and (c) that plaintiffs failed to prove that the fifty-seven misstatements on Table A caused inflation in Vivendi's share price. Many of these arguments are duplicative of arguments made at summary judgment and in defendants' motion in limine to exclude Dr. Nye's testimony. This opinion addresses only those arguments that have not been addressed by this Court's prior rulings, or that warrant further discussion in light of the current posture of the case and recent Second Circuit authority.
The Court refers the reader to its summary judgment opinion, Vivendi V, 634 F. Supp. 2d 352, and to its ruling on Vivendi's motion in limine to exclude Dr. Nye's testimony (Dkt. 929 at ¶ 7) for the Court's previous analysis of these issues.
(a) Connection between the Fraud and the Events
Vivendi argues that plaintiffs failed to prove the requisite connection between the nine events in 2002 that Dr. Nye opined were materializations of Vivendi's undisclosed liquidity risk and the alleged fraud. Vivendi contends that plaintiffs did not satisfy this burden because their theory of the case rested on a conception of the "zone of risk" that is too broad to comport with Second Circuit authority on the subject. Vivendi makes two specific arguments in this regard. First, Vivendi argues that for an event to fall within the zone of risk concealed by the alleged fraud, a reasonable investor who believed the fraud must have perceived each of the events in question as "remote or highly unlikely." (Def. 50(b) Br. at 32-33, 34-36). Second, Vivendi argues that the nine events did not reveal anything undisclosed about the "specific misrepresentations" alleged by plaintiff, and therefore cannot be said to fall within the zone of risk concealed by the fraud under the Second Circuit's recent decision in Omnicom Group, Inc. Sec. Litig., 597 F.3d 501 (2d Cir. 2010). (Def. 50(b) Br. at 33-34, 36-37.)
As to the first point, Vivendi misreads Castellano v. Young Rubican, 257 F.3d 171, 188 (2d Cir. 2001) which analyzed forseeability by asking whether one who believed in the fraud would perceive a "zone or risk" to be highly unlikely, not whether a specific event within the zone of risk was unlikely. Here plaintiffs offered substantial evidence that the zone of risk — a liquidity crisis — would have been thought unlikely by shareholders who believed Vivendi's repeated assurances about its financial health. Moreover, even if one were to focus on specific events, such as the multiple rating downgrades or the sudden sale of treasury shares, the evidence could have led a reasonable jury to conclude that an investor would have considered the events on the nine days identified by Dr. Nye to be remote or highly unlikely in light of Vivendi's earlier fraudulent statements.
With respect to ratings downgrades, Vivendi argues that the market knew that the company's credit rating might be downgraded in 2002 because it had received a "possible downgrade" rating by Moody's in June 2000 (months before the Class Period began), and had been placed on CreditWatch Negative by S P on December 17, 2000. But in assessing the likelihood of the downgrades that actually took place in May, July, and August of 2002, reasonable investors would consider the total mix of information available at that time. Here, that evidence included many statements by Vivendi touting its strong financial results.
In addition, the December 17, 2001 S P press release in which Vivendi was placed on CreditWatch stated that "Vivendi Universal's long- and short-term ratings will be affirmed, with a stable outlook, upon the closing of the wine and spirits division for about $8 billion, expected by year-end 2001." ( See DX-1150). After the wine and spirits transaction closed, reasonable investors could have thought Vivendi's rating outlook was stable and perceived the likelihood of a downgrade as remote. The record also contained evidence that Moody's reaffirmed Vivendi's Baa2 rating in January 2001 ( see DX-1993), arguably negating the continued force of its June 2000 "possible downgrade" warning.
Moreover, there was evidence that members of the public did actually perceive the risk of a downgrade to be remote or highly unlikely. For example, Dr. Nye testified that market analysts were surprised by the actual downgrades ( see Tr. 3750:24-3752:14), citing documents supporting that conclusion. ( See, e.g., PX-1142 (article quoting AP Online saying the May 3, 2002 downgrade "took investors by surprise" and Barclays
analyst saying "there had been no hint of an imminent downgrade"). Dr. Nye also testified that the downgrades were triggered by Vivendi's disclosure of new, non-public information to the rating agencies that had been previously undisclosed. ( See Tr. 3524-28 (Dr. Nye opining that the downgrade was based on new, non-public information provided by Vivendi to rating agencies); Clip Report from Deposition Video of Mr. Alex Hebert, former special advisor to Vivendi's Finance Division, Vols. 1 2 (testifying that the rating agencies received non-public information and that he sent the ratings agencies information in the months before the May 2002 downgrade)).
If the rating agencies downgraded Vivendi's credit rating in May 2002 upon receipt of new, non-public information, a reasonable investor who believed Vivendi's fraud and who had not been provided with the same non-public information would have thought the possibility of a rating downgrade to be remote in the months leading up to May 2002. Though defendants presented some conflicting evidence suggesting that the market was not surprised by the downgrade ( see, e.g., Tr. 3752:16-3754:22) (Nye cross); DX-240), it was for the jury to weigh that evidence against plaintiffs' conflicting evidence. It was perfectly reasonable or the jury to conclude in the end that reasonable investors would have thought the possibility of a downgrade in May, July or August 2002 to be remote had they believed Vivendi's prior misstatements.
Vivendi also argues that the market knew that Vivendi might sell treasury shares or dispose of some or all of Vivendi Environment, and that it might seek to raise cash quickly in 2002, such that none of the other events on the nine days identified by Dr. Nye could be viewed as remote by a reasonable investor who believed the fraud. Vivendi seems to be arguing that if a company warns that it might sell assets at some point in the future, an investor can never thereafter be surprised by a particular asset sale, even if that sale occurs under suspicious circumstances — i.e., suddenly, out-of-the-blue, and at prices below the asset's intrinsic or fair value. Again, that cannot be correct. Even a reasonable investor who considered it likely that Vivendi would sell treasury shares or other assets or seek to raise cash at some point in 2002 or 2003 could consider it highly unlikely that Vivendi would sell its shares and assets quickly and at prices below their true value due to a liquidity crunch. As with the ratings downgrades, there was sufficient evidence in the record to support an inference that investors were, in fact, surprised by these events. ( See, e.g., Tr. 3515:12-20 (Nye testimony) ("[T]he market took [the treasury share sale] as a signal they needed cash.") From such testimony, the jury was entitled to conclude that each event identified by Dr. Nye fell within the zone of risk concealed by Vivendi's fraud in the sense that an investor who believed the fraud would have thought it "highly unlikely" that these events would unfold at the time they did and under the circumstances they did.
The Court is not persuaded by Vivendi's remaining argument that the Second Circuit's recent decision in Omnicom, 597 F.3d 501, compels this Court to hold that plaintiffs' theory of the case rests on an overly broad conception of the "zone of risk" (and, in effect, requires the Court to reconsider its summary judgment ruling).
In Omnicom, the plaintiffs alleged that Omnicom had engaged in fraudulent accounting for the "Seneca transaction" pursuant to which Omnicom transferred certain deteriorating internet companies to a newly formed entity called Seneca. 597 F.3d at 504. News reports from the time of the transaction suggested that Omnicom was using it as an accounting method to move the internet companies' losses off its books, but the
market price did not drop in response to the transaction or to the reports regarding the transaction's purpose. Id. Plaintiffs alleged that the accounting for the Seneca transaction was fraudulent in three specific ways: (i) Omnicom should have written down the internet companies before engaging in the Seneca transaction, (ii) Omnicom did not properly value the internet companies, and (iii) Omnicom should have accounted for Seneca's losses after the transaction because Omnicom controlled Seneca. Id.
Over a year after the Seneca transaction, one of Omnicom's outside directors, who was also the Chair of its Audit Committee, resigned. Several negative news reports followed, as rumors circulated that The Wall Street Journal ("WSJ") would be publishing a negative article about accounting issues at Omnicom. Id. at 505-06. The WSJ article was published on June 12, 2002, and stated, inter alia, that the director had resigned amid questions he had raised for months regarding the purpose of the Seneca transaction, and amid questions about whether the board had received full information about the transaction. Id. at 506. The article also raised concerns about Omnicom's cash flow and accounting practices in general, noting, for example, that Omnicom "uses more aggressive means that its competitors to calculate the critical statistics of how much of its growth it generates from existing operations." Id. at 507. Omnicom's stock price fell sharply in response to that article and plaintiffs sought to recover that loss. See id. at 508.
The Second Circuit affirmed the district court's dismissal of the claim for failure to prove loss causation. The Second Circuit first found that the June 12 article was not a corrective disclosure because it did not purport "to reveal some then-undisclosed fact with regard to the specific misrepresentations alleged in the complaint concerning the Seneca transaction." Id. at 511. In reaching that conclusion, the Court pointed out that "[t]he use of the Seneca transaction as an accounting method to remove losses from Omnicom's books was known to the market a year before [the director's] resignation." Id. "At best," the Second Circuit explained, "[plaintiffs] had shown that the market may have reacted as it did because of concerns that [the director's] resignation and the negative tone of the June 12 article implied accounting or other problems in addition to the known Seneca transaction." Id. at 512.
The Second Circuit then went on to consider whether the director's resignation could be viewed as a materialization of a foreseeable risk concealed by the allegedly fraudulent Seneca transaction. The Court reasoned that because the use of the Seneca transaction to attempt to move the internet company losses off Omnicom's books was known to the market long before the director's resignation, the risk that was concealed by the fraud under plaintiffs' theory of the case would have to be the fraudulent nature of the accounting for the transaction. Id. at 513. But the Court found that the director's resignation "did not add to the public knowledge any new material fact about the Seneca transaction," and certainly not about whether the accounting for the transaction was proper. Id. at 514. Therefore, the Second Circuit found that the generalized investor reaction of concern in the wake of the June 12 article was "far too tenuously connected — indeed, by a metaphoric thread — to the Seneca transaction" to support liability. Id.
Vivendi now argues that Omnicom "precludes all-embracing theories like Plaintiffs' liquidity risk theory." (Def. 50(b) Reply at 16 n. 24.) According to Vivendi, Omnicom makes clear plaintiffs cannot establish loss causation by showing that the events identified by Dr. Nye revealed something new about Vivendi's "true liquidity condition," but must instead show that these events revealed something new about some of the specific ways that plaintiffs alleged Vivendi misrepresented its liquidity condition (e.g., earnings management, misleading disclosures regarding purchase accounting or Maroc Telecom).
Vivendi reads Omnicom far too broadly. As an initial matter, to the extent Vivendi suggests that Omnicom's discussion of corrective disclosures somehow disposes of this case, that argument is unavailing. This is not a corrective disclosure case. Rather, plaintiffs are proceeding under a materialization of the risk theory. The Court addressed that theory in detail at summary judgment, and found it to be viable. Vivendi V, 634 F. Supp. 2d at 365-69. Vivendi now contends that Omnicom "flatly rejected" plaintiffs' conception of loss causation (as endorsed by this Court at summary judgment). However, Omnicom did not purport to re-define the law of loss causation, or to overrule the previous Second Circuit cases ( Castellano, Suez Equity, Lentell, etc.) upon which this Court's previous analysis was based. Omnicom simply applied this circuit's existing loss causation principles to the particular factual context before it, a context that was quite different from the context of this case. The Second Circuit's application of the principles of loss causation to the facts before it in Omnicom does not suggest that this Court's previous application of those same principles to the facts of this case was incorrect.
In Omnicom, the fraud allegation was limited to a claim that Omnicom had used fraudulent accounting in connection with the Seneca transaction. 597 F.3d at 504. Thus, the subject of the fraud in Omnicom was quite narrow; there were no allegations of broad-based accounting improprieties manifested in a myriad of ways or, as here, of a liquidity risk concealed over a long period of time, but only of accounting fraud in connection with a single transaction. But the event that plaintiff claimed to be a "materialization of the risk" concealed by that alleged improper accounting (a director's resignation and resulting negative press) revealed no new information about the subject of the fraud (the Seneca transaction). The public had known for years that Omnicom had used the Seneca transaction as an accounting method to get certain losses off its books, and when he resigned, the director identified no new facts about that transaction or its accounting. See id. at 514. At best, the resignation revealed "other unknown concerns" about the company, completely unrelated to the subject of the alleged fraud. See id.
Here, in contrast, the alleged fraud is much broader and more far-reaching than in Omnicom. Plaintiffs' fraud allegations are not limited to a single transaction. Rather, plaintiffs theory of the case is that Vivendi concealed a serious and growing liquidity risk throughout the Class Period — including by repeatedly giving misleading assurances to the public about its financial condition despite knowing that its liquidity situation was dire. As a logical matter, the zone of risk concealed by Vivendi's fraud would therefore be larger than the zone of risk concealed by the much narrower fraud alleged in Omnicom.
Vivendi mischaracterizes plaintiffs' theory of the case to the extent they suggest that the subject of the fraud alleged by plaintiffs is limited to specific allegations regarding earnings management, purchase accounting, Maroc Telecom, Telco, and the Cegetel current account.
Moreover, the events allegedly constituting the materializations of the risk in this case ( e.g., ratings downgrades) tie much more directly to the subject of the fraud (a concealed liquidity risk) than did the director's resignation in Omnicom. Whereas a director's resignation may send a somewhat ambiguous negative signal to the market, a ratings downgrade sends a clear signal: it reveals that the rating agencies have changed their views regarding a company's ability to pay its debts. Unlike in Omnicom, where a rational jury could not conclude that the director's resignation revealed anything new about the alleged accounting improprieties, it was perfectly reasonable for the jury in this case to conclude that the events on the nine days identified by Dr. Nye, including several ratings downgrades, revealed new information about Vivendi's liquidity condition that had been concealed by Vivendi's fraud.
(b) Connection between the Events and Share Price Declines
Vivendi's second challenge is that plaintiffs failed to connect the nine events that Dr. Nye opined were materializations of Vivendi's undisclosed liquidity risk with any share price declines. Specifically, Vivendi contends that (i) there were no statistically significant share price declines on three of the dates identified by Dr. Nye — January 7, July 10, and July 15, 2002 — and that Dr. Nye's opinion to the contrary was based on a flawed statistical analysis that used an improper control period, an improper industry index, and made incorrect assumptions regarding volatility; and that (ii) Dr. Nye's analysis of the events on the other six dates was flawed because he "neglected to examine the intra-day movement of Vivendi's stock price." (Def. 50(b) Br. at 37-39.) In effect, Vivendi is arguing that no rational fact-finder could believe Dr. Nye.
Vivendi has not shown that Dr. Nye's analysis was so flawed that no reasonable jury could accept it. Dr. Nye provided reasonable explanations at trial for his control period, industry index, and volatility assumptions. ( See, e.g., Tr. 3493-94 (explaining that although it is usually preferable to use a control period that is outside the Class Period, Vivendi was a different company both before and after the Class Period, such that he felt it would be better to use a control period inside the Class Period and noting that the period he chose was "conservative"); Tr. 3868-75 (describing reasons for his choice of industry index); Tr. 3652-55 (defending volatility assumptions).) Vivendi cross-examined Dr. Nye extensively on each of these choices, and presented an expert witness, Dr. Silber, who criticized Dr. Nye's work. ( See Tr. 3616-3655 (cross-examination regarding choice of industry index, control period, and volatility); see generally Tr. 6243-6411; 6436-6483 (Silber testimony).) This was a typical battle of the experts. It was properly within the province of the jury to determine, after weighing all the evidence, whether to credit Dr. Nye's opinions on these issues or not. Knapp, 46 F.3d at 179; United States v. Artuso, 618 F.2d at 195.
Vivendi has also failed to show that a rational jury could not accept Dr. Nye's analysis on the ground that Dr. Nye allegedly failed to examine intra-day price movement on the nine days he identified. Vivendi went to great lengths at trial to argue that Dr. Nye's analysis was not worthy of belief for this reason. ( See, e.g., Tr. 6892-6898 (Saunders closing); Tr. 3790:6-3977:23 (Nye cross-examination).) But Dr. Nye ultimately stood by his opinions and offered viable justifications for doing so. For example, he testified that on the nine dates in question, he looked at the entire day's price movement even if the negative announcement was made later in the day because he had seen documents from those dates that suggested to him that the bad news had leaked prior to the public announcement and therefore could be responsible for price declines before the announcement. ( See, e.g., Tr. 3794:48; 3794:18-3795:11 (explaining his use of the full day's price decline July 2, 2002, including price declines before the announcement of the downgrade at 2:37 p.m., because the information had leaked into the market before 2:37 p.m.).) Vivendi's experts disagreed that the documents in question revealed that the bad news had leaked prior to the announcement, but offered no price movement analysis
of their own. The jury, in any event, rejected the testimony of Vivendi's experts and the Court is persuaded that a "reasonable jury" was entitled to do so.
(c) Whether the Misstatements Caused Inflation
Vivendi's third main challenge with respect to loss causation is that plaintiffs allegedly failed to prove that the misstatements caused inflation because (i) Dr. Nye's calculations of inflation throughout the Class Period (his "inflation band") had "no rational basis" in the evidence, and because (ii) the inflation band did not correspond to the fifty-seven misstatements, such that 43 of the fifty-seven statements the jury found to violate Section 10(b) actually occurred on days where inflation remained constant or decreased.
Vivendi's contention that Dr. Nye's inflation band lacks a rational basis in the evidence is without merit. Vivendi challenges three aspects of Dr. Nye's calculation of his inflation band: (i) his determination of the date at which inflation reached its maximum (December 13, 2001); (ii) his use of a proxy — purchase accounting — to measure increasing inflation over time between the start of the Class Period and December 13, 2001; and (iii) and his deviation from that proxy in the last 45 days before maximum inflation. The Court rejected each of these challenges in ruling upon defendants' pre-trial motion in limine to exclude Dr. Nye's testimony, finding that Dr. Nye's proposed testimony was sufficiently reliable to go before the jury. ( See Dkt. 929 at ¶ 7.) Nothing has changed since then, except that Dr. Nye actually gave the testimony at trial. It was for the jury to determine whether to credit Dr. Nye's properly admitted opinions.
Vivendi also argues that plaintiffs failed to establish loss causation because their inflation band did not correspond to the fifty-seven alleged misstatements on Table A. Vivendi contends that at minimum, there can be no liability for statements made on days on which inflation remained constant or decreased. Vivendi has not identified a single case to support this proposition, and the Court is aware of none. On the other hand, courts have suggested that a misstatement may cause inflation simply by maintaining existing market expectations, even if it does not actually cause the inflation in the stock price to increase on the day the statement is made. See Castillo v. Envoy Corp., 206 F.R.D. 464, 472 (M.D. Tenn. 2002) ("[T]he lack of statistically significant movement of the stock price following each misrepresentation . . . does not address whether the misstatements caused the stock to be artificially maintained at a level that did not reflect its true value.") (internal quotation marks omitted); see also In re Bristol-Myers Squibb Sec. Litig., No. 00-1990 (SRC), 2005 WL 2007004, at *17-18 (D.N.J. Aug. 17, 2005) ("[A] misstatement could serve to maintain the stock price at an artificially inflated level without also causing the price to increase further."); Swack v. Credit Suisse First Boston, 383 F. Supp. 2d 223, 240 (D. Mass. 2004) ("Defendants' conduct could have tempered a drop in price that would otherwise have occurred, or resulted in a greater increase than the stock would otherwise have enjoyed, absent the deceptive analyst reports."). The Court agrees and holds that a statement can cause inflation by causing the stock price to be artificially maintained at a level that does not reflect its true value.
Vivendi contends that these cases stand only for the proposition that a misstatement may be material even if share price does not increase, and do not support the proposition that a misstatement can be actionable even if it does not cause inflation to increase. While it is true that these cases did not squarely address whether a plaintiff could establish loss causation for statements made on days in which inflation remained constant — to the Court's knowledge, no court has — the Court does not believe that the discussion of the courts in these cases should be read as narrowly as Vivendi posits. Inflation and share price, after all, are related concepts.
Vivendi argues that the "maintenance" theory of inflation cannot be correct since for it to be true, the Court would have to accept the "impossibly unlikely" and "implausible" assumption that "on each day Defendants made a misrepresentation that did not increase inflation, if Defendants had not made that alleged misrepresentation, then inflation would have decreased by the exact same amount that the new misrepresentation simultaneously reinflated it." (Def. Reply in Supp. of Mot. for Partial Summ. Judg. on Loss Causation at 40) (emphasis in original). But Vivendi loses sight of the fact that in securities fraud cases, plaintiffs need not prove the amount of loss caused by each misstatement with complete mathematical precision. See Lattanzio v. Deloitte Touche LLP, 476 F.3d 147, 158 (2d Cir. 2007) (requiring plaintiffs to produce sufficient evidence for the fact finder to "ascribe some rough proportion of the whole loss" to defendant's fraud); Lentell, 396 F.3d at 177 ("We do not suggest that plaintiffs were required to allege the precise loss attributable to [defendant's] fraud. . . ."); Vivendi V, 634 F. Supp. 2d at 364 ("In theory, plaintiffs need only prove that they suffered some damage from the fraud.").
This method of proof makes particular sense in cases involving numerous misstatements over an extended time period on the same general topics. Where a jury has found, as here, a defendant omitted information about its true liquidity risk in fifty-seven statements over two years, it is easy for the company to then point to each particular misstatement and argue that plaintiffs have not proved that that particular statement caused any additional inflation in the share price distinct from the inflation caused by the other fifty-six statements. It may be impossible for an expert witness to reliably disaggregate the impact of any particular misstatement from the continued force of previous statements. The "maintenance" theory of inflation simply reflects the reality that inflation in a company's stock price is difficult to quantify with mathematical precision in any case, and that in a case where a company repeatedly makes statements that omit information about its liquidity risk, it is reasonable to conclude that each misstatement played a role in causing the inflation in the stock price (whether by adding to the inflation or helping to maintain it), even if it is not possibly to quantify the exact impact that each statement had on the inflation.
To hold, as Vivendi argues, that inflation must rise each time a misstatement is reiterated, and that plaintiffs in securities fraud cases must quantify the precise amount of the rise in inflation due to each such misstatement would produce a perverse result: it would make it harder for plaintiffs to prove loss causation when a company makes numerous similar misstatements over a long time period than when a company makes a single, isolated fraudulent statement, even though the former situation involves a more pervasive and widespread fraud. Such a rule could permit a company to avoid Section 10(b) liability by repeating its misstatements so many times that it becomes impossible for an expert to prove that any particular misstatement, viewed in isolation, caused a quantifiable increase in inflation. That runs contrary to the Second Circuit's guidance that plaintiffs need not show the "precise loss attributable to [defendant's] fraud" in order to make out a securities fraud claim. Lentell, 396 F.3d at 177; Lattanzio, 476 F.3d at 168. In the absence of any authority holding that the "maintenance" theory of inflation is not viable and that misstatements can only be actionable on days when inflation actually increases, the Court declines to so hold in this case.
For all of these reasons, the Court declines to enter judgment as a matter of law in Vivendi's favor on the issue of loss causation. The evidence presented at trial was sufficient for a reasonable jury to conclude that Vivendi made material misstatements and omissions that concealed facts regarding Vivendi's liquidity condition that, when disclosed, caused the price of Vivendi's shares to drop; and that the loss suffered by the plaintiffs was a foreseeable consequence of the alleged material misstatements or omissions. Lentell, 396 F.3d at 172.
5. Effect of Jury's Finding of Zero Inflation on Certain Dates during the Class Period
Vivendi also raises a number of challenges to the jury's verdict based on the fact that the jury found no inflation in Vivendi's share price on certain days during the Class Period. By way of background, plaintiffs at trial argued that the daily inflation in Vivendi's share prices was the difference between the actual share price and what the price would have been if Vivendi's true liquidity risk had been disclosed to the public. ( See, e.g., Tr. 3466:12-21 (Nye Testimony).) Dr. Nye presented the jury with a chart in which he had calculated what he believed to be the daily inflation in Vivendi's share price for each trading day in the Class Period. (PX 1486.) On that chart, the inflation increased from the beginning of the Class Period until December 13, 2001 and then decreased in "steps" on nine days in 2002 on which events occurred partially revealing Vivendi's fraud to the market. Dr. Nye opined that by August 14, 2002, the inflation in the stock price was zero because "the fraud is purged, there is no longer any inflation in the stock, no more for the market to find out about that the company knows." ( See id.; Tr. 3570:8-23; 3574:16-25.)
At the end of the trial, the Court instructed the jury as follows with respect to damages:
If you find that plaintiffs have proven all of the elements of their Section 10(b) claim against one or more of the defendants, then you must determine the amount of per share damages, if any, to which plaintiffs are entitled. Plaintiffs have the burden of proving damages by a preponderance of the evidence.
Plaintiffs can recover only actual damages, which is the difference between the price plaintiffs paid for each share of Vivendi stock and the price the share would have cost if no false or misleading statement or omission of material fact had occurred — in other words, the inflation in the stock price.
There may be factors other than the alleged false or misleading statements that affected Vivendi's stock price on any given day. Plaintiffs bear the burden of disaggregating (or separating out) any declines that were caused by other non-fraud related events. Defendants are not liable for any loss resulting from those other non-fraud related events.
Any damage calculation you determine must be expressed in terms of the daily amount of inflation per share . . .
Any damages you award must have a reasonable basis in the evidence. Damages need not be proven with mathematical certainty but there must be enough evidence for you to make a reasonable estimate of damages. (Jury Charge No. 28.)
Vivendi made no objection to this charge. (Dkt. 994 (Def. Revised Objections to the Court's Proposed Chart to the Jury).)
The Verdict Form asked the jury to fill out the daily inflation in the price of Vivendi's shares that was caused by any statements the jury found to have violated Section 10(b). (Verdict Form, Question No. 58.) The jury filled out the Verdict Form, and for the most part, found the daily inflation in Vivendi's share price to be approximately half of what Dr. Nye had opined. However, from September 11 to September 28, 2001, the jury listed the inflation in Vivendi's stock price as zero. The jury also found zero inflation on the nine days between November 2001 to August 14, 2002 on which Vivendi's ordinary shares traded, but Vivendi's ADRs' did not (or vice versa).
Those days were days in which the stock markets were closed in the United States, but not in France (or vice versa). There was nothing in the record that would support treating days on which one type of stock traded but the other did not differently than other days. (As will be discussed below, the jury's finding of zero inflation on these nine days when Vivendi's shares traded on one exchange, but not the other, likely reflects an error or confusion on their part.)
Vivendi now argues that because the jury found inflation in Vivendi's share price to be zero from September 11 to September 28, 2001, the Class Period must end on September 11, 2001, because the jury verdict reflects a "finding of fact" that the complete truth had been revealed to the market by September 11, such that everything that Vivendi had hidden before September 11 was revealed on that date. Vivendi's argument distorts the jury's verdict into something it was not. The contention that the jury made a "finding of fact" that the full truth about Vivendi's liquidity was revealed to the market on 9/11 is absurd. There was no evidence at trial that anything that had previously been hidden about Vivendi's liquidity was revealed to the market on September 11, 2001 — something Vivendi itself concedes. (Def. 50(b) Br. at 31 n. 23 ("Plaintiffs offered no evidence that any aspect of the fraud was disclosed [between September 1, 2001 and September 11, 2001].").
The jury's determination that inflation was zero on September 11 and in the weeks immediately following is more properly understood as the jury's attempt to determine damages according to the Court's instructions. The Court instructed the jury that it was to determine damages by calculating the inflation in the share price. ( See Jury Charge No. 28) In addition, the Court instructed the jury that, "[a]ny damages you award must have a reasonable basis in the evidence," and that "there must be enough evidence for you to make a reasonable estimate of damages." ( Id.) The jury may have concluded that Dr. Nye's had not adequately disaggregated the effects of the World Trade Center attacks on 9/11 and the resultant turmoil in the markets when calculating the daily inflation in Vivendi's stock price (even though he purported to have done so), and that it could not reasonably estimate the daily inflation in Vivendi's share price in the immediate aftermath of 9/11 in light of that turmoil. Consequently, the jury may have concluded that although the evidence at trial proved that Vivendi's fraud had caused some damages on 9/11 and immediately thereafter, plaintiffs had not met their burden of introducing evidence sufficient to enable them to make a reasonable estimate of the daily inflation in that period, such that it could not award any damages in that time period.
Implicitly recognizing the implausibility of its argument that the jury made a "finding of fact" that the full truth of Vivendi's liquidity situation was revealed to the market on 9/11, Vivendi argues in the alternative that if the jury did not intend to find that the truth was fully revealed on 9/11, there must at minimum be a new trial because the jury misunderstood the Court's instructions and "did not understand that inflation could drop to zero only when the truth about the alleged fraud was fully reflected in the market price." (Def. 50(b) Br. at 30.) Again, the premise of Vivendi's argument — that the jury could find zero inflation only when the full truth of the fraud was revealed — is faulty. As noted above, the jury may have thought that the Court's instructions precluded them from awarding any damages for dates on which they were unable to make a "reasonable estimate" of the daily inflation — for instance, due to the turmoil caused by 9/11 — even if they were able to make reasonable estimates as to the damages on other dates. This does not suggest that the jury misunderstood the Court's instructions.
Vivendi's next approach is to argue that a verdict in which inflation drops to zero on dates when not even plaintiffs alleged that the revelation of the fraud caused a share price decline would give certain shareholders impermissible windfall damages, thereby violating the Supreme Court's holding in Dura that a shareholder can only recover for share price declines caused by the disclosure of the fraud. (Def. 50(b) Br. at 30-31, n. 23 (citing Dura, 544 U.S. at 342-43.) In support of this argument, Vivendi offers the example of an investor who bought a Vivendi share on August 31, 2001 (when the jury found $2.14 in inflation per ADR) and sold that share on September 17, 2001 (when the jury found no inflation). Vivendi contends that this shareholder would be entitled to $2.14 in damages, even though there was no evidence that any aspect of the fraud was disclosed between August 31 and September 17. But Vivendi's example is easily rebutted since any shareholder who sold prior to the first date of materialization of the fraud (January 7, 2002) would not be entitled to collect damages under well-established loss causation principles, which permit a party to recover only if they can show that they purchased shares at an inflated price and that the share price fell after the truth concealed by the fraud became known. See Dura, 544 U.S. at 347 (plaintiff failed to establish economic loss element of Section 10(b) violation where he alleged only that he purchased his shares at an inflated price, but not that the company's "share price fell significantly
The Court has modified Vivendi's example somewhat for this discussion, since Vivendi's initial example used Vivendi ordinary share purchases, which, as noted above, are no longer in the class. The Court's example uses ADRs instead, and modifies the dates somewhat in order to do so.
Vivendi also argues that September 11, 2001 should be treated as the "first date of the materialization of the fraud" since that is the first date during the Class Period on which the jury found zero inflation in Vivendi's share price, but this argument is unavailing. There is nothing in the record to support a conclusion that there was a materialization of the fraud on September 11, 2001. The jury's verdict is perfectly consistent with the evidence offered by plaintiffs that the first date of the materialization of the fraud is the first date identified by Dr. Nye: January 7, 2002.
after the truth became known"). Thus, investors who sold Vivendi shares in the aftermath of 9/11 will not get a windfall.
A better example, which does raise legitimate concerns about certain shareholders obtaining a windfall, is as follows: Consider a shareholder, shareholder X, who purchased a Vivendi ADR on December 13, 2001 (the date of maximum inflation). On that date, the jury found the inflation in Vivendi's ADRs' to be $10 per share. Imagine that shareholder X sold his share on April 1, 2002 — i.e., after the first alleged materialization of the risk, but before the full truth about Vivendi's liquidity situation had been revealed to the market, according to Dr. Nye. The jury found inflation to be zero on April 1, 2002 (which was a date on which Vivendi's ADRs traded on the NYSE, but Vivendi's ordinary shares did not trade on the Paris Bourse). In this example, shareholder X would be entitled to $10 in damages, reflecting the difference between the inflation on the date of purchase and the inflation on the date of sale. Now consider a shareholder, shareholder Y, who also bought an ADR on December 31, 2002 (at $10 inflation), but sold his share on either March 31, 2002 or April 2, 2002 (dates on which the jury found inflation to be $8.55 per ADR). There were no disclosures relating to liquidity between March 31 and April 2. Shareholder Y would be entitled to collect $1.45 in damages, reflecting the difference between the inflation on the date of purchase and the inflation on the date of sale. Thus, shareholder X will recover significantly more than shareholder Y, even though nothing was disclosed between the date that shareholder X sold his share and the date that shareholder Y sold his share that would justify treating them differently.
The discrepancy between shareholder X's recovery and shareholder Y's recovery in this example stems from the fact that on April 1, 2002, a day on which Vivendi's ADRs traded, but Vivendi's ordinary shares did not (and on eight other similar days) the jury found zero inflation in the price of the share that did trade. No evidence was presented at trial to justify such a finding. The expert witnesses at trial treated days on which only one of class of Vivendi's stock traded like any other day in the Class Period; their inflation values were unaffected by the closure of one of the stock markets. The jury's finding of zero inflation on nine days when only one class of Vivendi's shares traded most likely reflects either confusion on their part regarding the proper treatment of dates on which Vivendi's shares traded on the Paris Bourse but not on the NYSE (or vice versa), or an administrative error in filling out the Verdict Form.
The Court's jury charges and the Verdict Form did not provide any instructions to the jury regarding the treatment of such days.
However, this minor error does not justify a new trial (and obviously does not suggest in any way that Vivendi is entitled to judgment as a matter of law since it says nothing about the sufficiency of the evidence regarding damages). "A motion for a new trial ordinarily should not be granted unless the trial court is convinced that the jury has reached a seriously erroneous result or that the verdict is a miscarriage of justice." Medforms, Inc. v. Healthcare Mgmt. Solutions, Inc., 290 F.3d 98, 106 (2d Cir. 2002). An error in filling out the amount of the daily inflation on 9 of 454 days in the Class Period is not the type of "miscarriage of justice" or "seriously erroneous result" that would justify throwing out the entire jury verdict in this case, after a complex trial that lasted over three months. Rather, it is a minor error that in no way suggests that the jury fundamentally misunderstood the overall task they had been charged with. It would be a far greater
miscarriage of justice to overturn the jury's entire verdict because the jury made a few relatively insignificant errors in completing a long and complex Verdict Form. The consequences of the jury's erroneous finding of zero inflation on a handful of days in the Class Period can easily be addressed during the claims administration procedure. It may turn out that no shareholders who purchased or sold Vivendi shares on the nine dates in question will submit claims, which would essentially moot the issue. If shareholders who purchased or sold Vivendi shares on these dates do submit claims, the parties can submit briefs and the Court can consider how best to address such claims.
To the extent Vivendi's challenge to the jury's finding of zero damages on days when Vivendi's shares traded on one exchange but not the other resembles an inconsistency challenge (e.g., challenging the logical irreconcilability of the jury's findings of inflation on March 31, 2002 and April 2, 2002 relative to its findings on April 1, 2002), Vivendi waived its right to bring this challenge by failing to raise this issue before the jury was discharged. See Kosmynka, 462 F.3d at 83; DiBella, 403 F.3d at 117. This type of error in filling out the verdict form is precisely the type of error that could have been easily remedied by resubmitting the Verdict Form to the jury before the jury was discharged. Instead, Vivendi sat by silently.
The parties may agree that the most appropriate solution is for the Court to simply correct the jury's error on the nine days in question to award the amount that the jury clearly would have intended to award if it had not made the error. See, e.g., United States ex rel. Miss. Road Supply Co. v. H.R. Morgan, Inc., 542 F.2d 262, 269 (5th Cir. 1976) (correcting computational error made by jury who "clearly intended to award [Mississippi Road] all of the requested damages" but entered an incorrect larger number). Alternatively, Vivendi may move for a partial new trial limited solely to the issue of the daily inflation in Vivendi's share price on the handful of days in question, and the Court will consider such a motion.
6. Forward-Looking Statements
Vivendi also contends that certain of the fifty-seven statements that the jury found to have violated Section 10(b) are "forward-looking statements" as to which the jury could not have found liability under the PSLRA because: (1) they were accompanied by meaningful cautionary language, or (2) the jury found that Vivendi acted recklessly, and PSLRA shields forward-looking statements from liability if they are not made with actual knowledge of their falsity. Vivendi also challenged these statements on this basis before the Court submitted Table A and the Verdict Form to the jury, but the Court concluded at that time that each of the statements on Table A could be properly put to the jury.
The statements challenged by Vivendi on this basis are Statements Nos. 1, 2, 5, 6-3, 8, 14, 18-2, 20, 26, 28-1, 28-2, 47, and 56 on Table A. These were all written statements.
In reaching that conclusion at trial, the Court did not make a definitive determination whether any of the fifty-seven statements were forward-looking statements protected by the PSLRA. The Court determined that even if the statements were forward-looking statements protected by the PSLRA's safe harbor, it was for the jury to determine whether the cautionary language accompanying any of the statements on Table A was sufficiently "meaningful," and therefore the Court did not believe that any of the statements on Table A were inactionable as a matter of law under 15 U.S.C. § 78u-5(c)(1)(A)(i). In addition, plaintiffs alleged that defendants actually knew that all fifty-seven statements were misleading, and the evidence could have supported such a finding. It was therefore appropriate to submit any forward-looking statements to the jury since any statements that the jury found were not accompanied by meaningful cautionary language would be actionable if the jury found that Vivendi made the statements with actual knowledge of their falsity.
(a) The Challeged Statements Do Not Fall Within the PSLRA'S "Safe Harbor"
The PSLRA creates a safe harbor which provides special protection for alleged misstatements that are "forward-looking" in nature. See 15 U.S.C. § 77z-2(c)(1)(B)(ii). The PSLRA defines forward-looking statements to include, inter alia, statements containing "a projection of revenues, income (including income loss), earnings (including earnings loss) per share, capital expenditures, dividends, capital structure, or other financial items;" statements of the "plans and objectives of management for future operations . . . "; and statements of "future economic performance." 15 U.S.C. § 78u-5(i)(1)(A)-(C). The safe harbor shields written forward-looking statements from liability if any one of the following criteria is met: (1) the statement was "identified as a forward-looking statement, and is accompanied by meaningful cautionary statements identifying important factors that could cause actual results to differ materially from those in the forward-looking statement"; (2) the statement was immaterial; or (3) the statement, if made by a business entity, was not made or approved by an executive who had "actual knowledge" that the statement was false or misleading. 15 U.S.C. § 78u-5(c)(1); Slayton v. Am. Express Co., 604 F.3d 758, 765 (2d Cir. 2010). However, the safe harbor, and the closely related bespeaks caution doctrine, do not apply to statements of present or historical facts. See P. Stolz Family P'ship L.P. v. Daum, 355 F.3d 92, 96-97 (2d Cir. 2004) (bespeaks caution doctrine does not protect misrepresentations of present or historical facts).
The PSLRA's "safe harbor" provision is based in part on the judicially created "bespeaks caution" doctrine. Slayton, 604 F.3d at 770 n. 5 (citing Conference Report at 43, 1995 U.S.C.C.A.N. at 742). Courts frequently treat the two doctrines as essentially analogous to one another, and examine the case law relating to both doctrines in analyzing the other. See, e.g., Rombach, 355 F.3d at 173.
In this case, the statements challenged by Vivendi relate primarily to Vivendi's expected EBITDA and cash flow figures. For example, Vivendi contends that the following statement from its October 30, 2000 F-4 is clearly a projection that falls within the PSLRA's safe harbor: "It is also our objective to grow pro forma adjusted EBITDA at an approximate 35% compound annual growth rate through 2002." (Statement 1 on Table A.) However, this statement and others like it are not entitled to the protection of the PSLRA's safe harbor (or the related bespeaks caution doctrine) because plaintiffs are not challenging the accuracy of the forward-looking aspects of the statement. As the Second Circuit has recently directed in Iowa Pub. Emps.' Ret. Sys. v. MF Global, "A statement may contain some elements that look forward and others that do not. . . . But in each instance the forward-looking elements and the non-forward-looking are severable." 620 F.3d 137, 144 (2d Cir. 2010).
This statement is probably Vivendi's best example of a statement that, on its face, can be read to be forward-looking. Other statements that Vivendi contends are forward-looking are statements of present or historical fact on their face and cannot be classified as forward-looking under any reasonable definition of that term. For example, Vivendi challenges the statement that, "`Our third quarter results for the media and communications business, with 24% revenue and 90% EBITDA growth, including organic growth of 8% and 36% respectively, are obviously strong despite the tough environment,' said Jean-Marie Messier, Chairman and Chief Executive Officer of Vivendi Universal. `They reflect both our higher potential for growth and greater resiliency to recessionary environments compared to many of our peers.'" (Statement No. 18.) This statement clearly described results already achieved for in the third quarter, and contains a statement of opinion by Messier regarding what those results reflect about the company in relation to its peers. It is a stretch to argue that any part of this statement is forward-looking, much less to argue that the statement as a whole is forward-looking.
Here, the plaintiffs challenge the non-forward looking elements of Vivendi's statements regarding its EBITDA growth, rather than the non-forward-looking elements. As the First Circuit explained in In re Stone Webster, Inc., Sec. Litig., the safe harbor was "intend[ed] to protect issuers and underwriters from projections and predictions of future economic performance, which are later shown to have been inaccurate." 414 F.3d 187, 213 (1st Cir. 2005) (emphasis added). Thus, courts often describe forward-looking statement as statements whose accuracy can only be verified after they are made. See, e.g., Harris v. Ivax Corp., 183 F.3d 799, 805 (11th Cir. 1999) (statement is forward-looking if its truth or falsity is verifiable only after it is made); In re Ashanti Goldfields Sec. Litig., 184 F. Supp. 2d 247, 266 (E.D.N.Y. 2006) ("[A]s a general rule, statements whose truth cannot be ascertained until some time after the time they are made are `forward-looking' statements."). The negative corollary to that proposition is that statements about present or historical facts, whose accuracy can be determined at the time they were made, are not forward-looking statements falling within the PSLRA's safe harbor. See In re NutriSystem, Inc. Sec. Litig., 653 F. Supp. 2d 563, 579 (E.D. Pa. 2009) ("A statement is not forward-looking if its accuracy can be determined at the time it was made."); see also Illinois State Bd. of Inv., 369 F. App'x 260, 264 (2d Cir. 2010) (safe harbor does not apply to statements or present fact). In other words, the safe harbor does not protect statements which are misleading about historical and present facts at the time they are made, and whose misleading nature can be verified at the time they are made, simply because the statements are couched as predictions of future events. See P. Stolz, 355 F.3d at 97 ("It would be perverse indeed if an offeror could knowingly misrepresent historical facts but at the same time disclaim those misrepresentations with cautionary language.").
In this case, plaintiffs have never argued that Vivendi's 35% EBITDA growth projection or any other projections made by Vivendi were misleading because the company failed to achieve its targets. Under plaintiffs' theory of the case, whether Vivendi actually achieved its 35% target or not is irrelevant. According to plaintiffs, it is the announcement of the target itself that is misleading as a matter of present fact in that Vivendi failed to disclose that a huge one-time purchase accounting benefit was built into the projection — a benefit that existing internal, undisclosed documents showed would account for almost 50% of EBITDA growth. That is, plaintiffs alleged that Vivendi failed to disclose as a matter of present fact that the company was not actually envisioning achieving anything close to 35% EBITDA growth as a result of improved operations. Rather, the growth rate — whether it be 35% or any other number — was the result of "accounting magic" (PX 145), that analysts (or shareholders) will "not have it easy to track." (PX-654). In this sense, the misleading nature of the statement could be verified the moment it was made, and did not depend on any future events. In re Oxford Health Plans, Inc. Sec. Litig., 187 F.R.D. 133, 141 (S.D.N.Y. 1999) ("[P]laintiffs point out that they are not relying on the falsity of Oxford's financial projections and estimates, but rather the defendants' failure to disclose historical and existing material facts about Oxford's computer problems and the impact of those problems on the reliability of the financial statements. The safe harbor and bespeaks caution doctrines do not apply to these omissions.") Cf. Harris v. Ivax Corp., 183 F.3d at 805. Under these
In closing, plaintiffs' counsel did not argue that the 35% EBITDA growth rate would not be reached but that Vivendi failed to disclose that "the target would be reached only . . . by using purchase accounting benefits that created an illusion of business growth." (Tr. 7290: 4-13)
circumstances, the Court finds that this statement and the others challenged by Vivendi are not protected by the safe harbor and was properly placed before the jury.
(b) Waiver
Assuming, arguendo, that one or more of the thirteen challenged statements are properly viewed as forward looking, Vivendi contends that it cannot be liable for any such statement because the jury found that it acted recklessly. Vivendi is correct that under 15 U.S.C. § 78u-5(c)(1)(B), a forward-looking statement is not actionable unless the jury finds that it was made with actual knowledge of its falsity (as opposed to recklessness). However, Vivendi's challenge in this regard is properly characterized as an argument that the jury's finding of recklessness is inconsistent with its finding of liability as to any statements on Table A that are forward-looking. Vivendi's argument is based on a comparison between the jury's general verdict of liability and the jury's response to a special interrogatory regarding Vivendi's state of mind. That is a quintessential inconsistency challenge to the extent that it challenges the logical reconcilability of two separate findings by the jury. If Vivendi is trying to suggest that it is really just challenging the sufficiency of the evidence of liability as to the allegedly forward-looking statements — an assertion that cannot really be squared with the arguments it actually makes in its briefs — that challenge would fail. There was plenty evidence in the record from which a reasonable jury could have concluded that Vivendi, through Hannezo and/or Messier, made each of the fifty-seven misstatements, with actual knowledge of their falsity. That the jury found only recklessness does not negate the existence of this evidence, and cannot therefore justify entry of judgment as a matter of law in Vivendi's favor. See Fed.R.Civ.P 50(a) ("If a party has been fully heard on an issue during a jury trial and the court finds that a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue, the court may . . . grant a motion for judgment as a matter of law"); Zellner, 494 F.3d at 370-71; Tolbert, 242 F.3d at 74.
Construing Vivendi's argument as an argument that the jury's finding of liability on the challenged statements is inconsistent with its special interrogatory responses (finding recklessness), the Court finds concludes Vivendi is not entitled to a new trial because Vivendi failed to object to the inconsistency before the jury was discharged, such that the opportunity to resubmit the issue to the jury and cure the inconsistency was lost. See Kosmynka, 462 F.3d at 83; DiBella, 403 F.3d at 117 (plaintiff waived objection that verdicts were inconsistent by failing to object while the jury was still empanelled); Haskell v. Kaman Corp., 743 F.2d 113, 123 (2d Cir. 1984) (party who failed to object that interrogatory answers were inconsistent with general verdict waived his right to object under Rule 49(b) by failing to object before the jury was discharged) (citing Skillin v. Kimball, 643 F.2d 19, 19-20 (1st Cir. 1981)).
Vivendi contends, relying on Denny, 42 F.3d at 111, that it cannot have waived its right to challenge the inconsistency because it objected to the question in the Verdict Form that produced the inconsistency. However, Denny cannot be read so broadly. Rather than adopt a per se rule of waiver, the Second Circuit held instead held that courts must apply waiver principles on a "case-by-case" basis, 42 F.3d at 110, 111. In Denny, the verdict form was legally erroneous because it submitted two claims, implied warranty and strict liability, which were deemed identical. As the Second Circuit observed:
This was not a case in which an inconsistency could be resolved by resubmitting the verdict form to the jury with a request that it reconcile its answers to particular questions. In such a case, neither the court's instructions as to the law nor the verdict form would be altered. In the instant matter, however, Ford had objected to submitting both the strict liability and implied warranty claims to the jury. That objection had been overruled, and if that ruling is assumed to be correct, the jury's verdict was not inconsistent. . . . A party who has timely objected to jury instructions is not obligated to renew its objections after the jury has rendered a verdict consistent with those instructions.Id. at 111 (emphasis added)
The Court's reasoning in Denny suggests that in deciding whether to find a waiver where a party fails to object to an inconsistent verdict, the focus should be on determining whether an objection lodged at the time of verdict could have cured the alleged inconsistency without requiring a new trial. If resubmission to the jury could cure the inconsistency, that weighs very strongly in favor of a finding of waiver.
Here, unlike in Denny, the jury verdict was, at least in Vivendi's view, inconsistent with the jury charges, which had stated that the jury could only find liability for forward-looking statements that were made with "actual knowledge" of their falsity. ( See Jury Charge No. 25.) Thus, a timely objection at the time the jury verdict was read would not simply have been a renewal of Vivendi's earlier objection to the format of the Verdict Form; it would be a new objection that the verdict was inconsistent with the Court's instructions. In addition, unlike in Denny, any possible inconsistency could have been cured by resubmitting the Verdict Form to the jury and reminding them that, consistent with the Court's original instructions, they could not find liability for any forward-looking statement unless they found that it was made with actual knowledge. Having failed to object before the jury was discharged, Vivendi waived its right to seek a new trial. See Trinidad v. Am. Airlines, Inc., No. 93 Civ. 430 (SAS), 1997 WL 79819, at *2 (S.D.N.Y. Feb. 25, 1997) (plaintiff waived the right to object to inconsistency between general verdict and special interrogatories under Rule 49(b) because he failed to object to the inconsistency before the jury was discharged).
Vivendi's suggestion that it was "not given the opportunity between the conclusion of the Court's reading of the verdict and discharge of the jury to raise any inconsistencies," (Def. 50(b) Reply at 2, n. 2), is unpersuasive. Counsel were provided with photocopies of the Verdict Form while the Court was reading the verdict into the record. That took considerable time, giving counsel ample opportunity to review the Verdict Form and identify potential inconsistencies before the jury was polled. The alleged inconsistency should have been plainly apparent to Vivendi since Vivendi believed the challenged statements to be forward-looking within the meaning of the PSLRA.
7. Puffery
Vivendi also moves for judgment as a matter of law on the grounds that certain of the statements that the jury found were misleading were inactionable "puffery." Puffery is an optimistic statement that is so vague, broad, and non-specific that a reasonable investor would not rely on it, thereby rendering it immaterial as a matter of law. See ECA, Local 135 IBEW Joint Pension Trust of Chicago v. JP Morgan Chase Co., 553 F.3d 187, 206 (2d Cir. 2009). However, the mere fact that a statement uses conclusory, indefinite, and unverifiable terms, rather than expressing a reason in dollars and cents, does not compel a conclusion that it is immaterial as a matter of law. Virginia Bankshares, Inc. v. Sandberg, 501 U.S. 1083, 1093-94 (holding that statement that merger would give shareholders "high value for their shares" could be deemed material, and noting that "such conclusory terms in a commercial context are reasonably understood to rest on a factual basis that justifies them as accurate, the absence of which renders them misleading"); Novak v. Kasaks, 216 F.3d 300, 315 (2d Cir. 2000) (statements that inventory situation was "in good shape" or "under control" made while defendants "allegedly knew the contrary was true," were actionable); City of Monroe Employee Retirement System v. Bridgestone Corp., 399 F.3d 651, 674 (6th Cir. 2005) (statement that "the objective data clearly reinforces our belief that these are high-quality, safe tires" was non-puffery on the ground that it was "an assertion of a relationship between data and a conclusion, one that a finder of fact could test against record evidence"); Basquiat ex rel. Estate of Basquiat v. Sakura Int'l, No. 04 Civ. 1369 (GEL), 2005 WL 1639413, at *5 (S.D.N.Y. July 5, 2005) (statements were not puffery where they included specific detail and were alleged to be knowingly false). "Whether the [statement of] opinion or `soft information' is indeed actionable "depends on all relevant circumstances of the particular case." Vivendi I, 381 F.Supp. at 182 (citing Ganino v. Citizens Utilities Co., 228 F.3d 154, 162 (2d Cir. 2000)).
Vivendi contends the following statements are puffery: Statements Nos. 3-1, 3-2, 5, 6-1, 6-2, 6-3, 8, 12, 13-1, 17, 18-1, 18-3, 19-1, 19-3, 28-1, 28-2, 28-3, 29-1, 29-2, 35-1, 35-2, 36-2, 39, 40, 40-3, 41, 42, 47, 50, and 57.
The Court concludes that none of the statements Vivendi contends are puffery are inactionable as a matter of law. For example, Vivendi contends that the following statements from a 6-K Vivendi filed with the SEC on January 13, 2001 were puffery:
Moreover, our group will have a very sound financial footing in communications, probably the best of all its competitors. With share capital amounting to more than 50 billion euros, after the disposal of Seagram's spirits and wine business, Vivendi Universal is expected to have zero net debt on January 1, 2001 on a pro forma basis. Thanks to our free net cash flow and the opportunities to dispose of some holdings, such as our stake in BSkyB, we will have an additional war chest of 10 billion euros for 2001-2002 before the first euro of debt, and without the creation of new shares. That means we will have the resources to pursue the growth of our businesses in an especially health and efficient way.
On these very sound basses, which differentiate us strongly from our competitors, I have no hesitation whatsoever in saying that the current level of our share price, which has been very adversely affected temporarily by arbitrate operations on the US and European indexes, does not reflect the true value of Vivendi's businesses, nor does it reflect the future prospects of Vivendi Universal.
(PX-720.) These are not the types of vague, generalized statements of optimism that reasonable investors would not rely on. The statements regarding Vivendi's "sound financial footing," are supported by specific statements of fact regarding Vivendi's resources and financial condition — for example, that Vivendi had "zero net debt," and "free cash flow." See City of Monroe, 399 F.3d at 674 (optimistic statements supported by specific facts are not puffery); Basquiat, 2005 WL 1639413, at *5 (same). Reasonable investors could rely on these statements, viewed as a whole, and find them to be materially misleading regarding Vivendi's true liquidity condition. The jury, which received specific instructions on puffery and materiality, was entitled to make the ultimate determination as to whether this statement, and others like it, was material. See Ganino, 228 F.3d at 162 ("Materiality is a mixed question of law and fact."); see also In re Spiegel Inc. Sec. Litig., 382 F. Supp. 2d 989, 1028 (N.D. Ill. 2004) ("Materiality . . . requires delicate assessments of the inferences a reasonable shareholder would draw from a given set of facts and the significance of these facts to him and these assessments are peculiarly ones for the trier of fact.") (internal quotations and citations omitted).
III. Vivendi's Motion for a New Trial Pursuant to Rule 59
Vivendi has also moved for a new trial pursuant to Rule 59 on the following grounds: (1) that the verdict represented an impermissible compromise as to Vivendi's liability; (2) that Vivendi was denied a fair trial because it was allegedly not provided with a list of the statements that plaintiffs contended were misstatements until after the close of the evidence; (3) that the Verdict Form deprived Vivendi of a fair trial because it did not require the jury to make particularized findings as to each element of a Section 10(b) claim and therefore confused the jury; and (4) that plaintiffs made improper and highly prejudicial arguments during their summation.
A. Standard of Review
Under Rule 59, courts may grant a new trial "for any reason for which a new trial has heretofore been granted in an action at law in federal court." Fed.R.Civ.P. 59(a). It is well-established that a district court may grant a new trial under Rule 59 only if it concludes that the jury reached a "seriously erroneous result" or that the "verdict is a miscarriage of justice." Manley v. Ambase Corp., 337 F.3d 237, 245 (2d Cir. 2003). Thus, a court should grant a new trial if it finds that the verdict is "against the weight of the evidence," id., or "if the trial was not fair to the moving party," United States v. Greer, 285 F.3d 158, 170 (2d Cir. 2002). A trial may be unfair to the moving party if substantial errors were made in admitting or excluding evidence, or in charging the jury, or if misconduct by counsel during the course of the trial causes unfair prejudice to the moving part. See Pappas v. Middle Earth Condo. Ass'n, 963 F.2d 534, 540 (2d Cir. 1992); Sharkey v. Lasmo (AUL Ltd.), 55 F. Supp. 2d 279, 289 (S.D.N.Y. 1999). The Rule 59 standard is less stringent than the Rule 50 standard for judgment as a matter of law in two respects: "(1) a new trial under Rule 59(a) `may be granted even if there is substantial evidence supporting the jury's verdict,' and (2) `a trial judge is free to weigh the evidence himself, and need not view it in the light most favorable to the verdict winner.'" Manley, 337 F.3d at 244-45 (quoting DLC Mgmt. Corp. v. Town of Hyde Park, 163 F.3d 124, 133-34 (2d Cir. 1998).
A court may consider a motion for a new trial brought after the liability phase of a case has been tried, but before the damage phase has begun and before a final judgment has been entered. See Wantanabe Realty Corp. v. The City of New York, No. 01 Civ. 10137 (LAK), 2003 WL 22862646, at *5 (S.D.N.Y. Mar. 3, 2003) (addressing motion for a new trial brought after liability phase of bifurcated trial but before judgment had been entered); Flynn v. Goldman Sachs Co., 836 F. Supp. 152, 154 n. 2, 164 (S.D.N.Y. 1992) (reviewing motion for new trial after trial on liability, although damages phase had yet to begin and no judgment had been entered).
B. Compromise Verdict
Vivendi argues that the verdict reflects an impermissible compromise as to Vivendi's liability, requiring a new trial. A verdict must be set aside on compromise grounds if a court concludes that jurors who could not agree on liability compromised by agreeing to find liability but to modify the damages award to reflect their unresolved disagreement. Fox v. City Univ. of New York, 187 F.R.D. 83, 93 (S.D.N.Y. 1999). In order for a district court to grant a new trial on compromise grounds, "the record itself viewed in its entirety must clearly demonstrate the compromise character of the verdict. . . ." Maher v. Isthmian Steamship Co., 253 F.2d 414, 415 (2d Cir. 1958). Thus, a court should not grant a new trial in circumstances where, "[w]hile a compromise may have occurred, there is an equally reasonable and perhaps even better explanation which involves no jury misconduct." Ajax Hardware Mfg. Corp. v. Indus. Plants Corp., 569 F.2d 181, 184 (2d Cir. 1977); see Lewis v. City of New York, 689 F. Supp. 2d 417, 426 (E.D.N.Y. 2010); Shepherd v. Metro-N. Commuter R. Co., 791 F. Supp. 1008, 1011-12 (S.D.N.Y. 1992). A court may infer that a verdict is a compromise "where damages are awarded in an amount inconsistent with the theory of liability offered at trial together with other indicia such as a close question of liability." Atkins v. New York City, 143 F.3d 100, 104 (2d Cir. 1998); Diamond. Enters. USA, Inc. v. Steinsvaag, 979 F.2d 14, 17 (2d Cir. 1992) ("An inadequate damages award, standing alone, does not indicate a compromise among jurors. Besides inadequate damages, there must be other indicia of compromise, such as difficulty in jury deliberations or close questions of liability.") (internal citations omitted).
In this case, the jury's damage award could support an inference of compromise. With some exceptions, the jury roughly halved the daily inflation figures proposed by Dr. Nye. See Gries v. Zimmer, Inc., No 90-2430, 1991 WL 137243, at *10 (4th Cir. 1999) (noting that a 50% reduction in damages "raises suspicions" about a compromise but holding that damage award did not justify granting a new trial on compromise grounds). The Second Circuit's decision in Ajax, 569 F.2d 181, is particularly instructive as to how a district court should evaluate whether a jury's damage award indicates a compromise on the underlying issue of liability. In Ajax, both parties agreed that if defendant breached the contract, plaintiff would be entitled to a liquidated sum of about $160,000. 569 F.2d at 184. But the jury found a breach of contract and awarded only $70,000 in damages. Id. The defendant argued that the verdict was an improper compromise and moved for a new trial, and the district court granted that motion. Id. The Second Circuit reversed the district court's finding that the verdict should be set aside on compromise grounds. Id. The court found that based on the instructions that had been given, the jury may well have believed that it was permissible for it to award any sum of damages that did not exceed $160,000. Id. Consequently, the court found that while a compromise was one possible inference from the facts, it was not the only plausible inference, since it was equally reasonable to conclude that the jury had simply tried to follow the dourt's instructions on damages without engaging in any misconduct. Id. Although the Second Circuit overturned the district court's grant of a new trial on compromise grounds, the Second Circuit upheld the district court's grant of a new trial on the ground that there was no reasonable basis in the record to support a $70,000 damage award, in light of the
parties' agreement that if the contract had been breached, damages should have been $160,000. Id.
In this case, as in Ajax, "a compromise on the issue of liability is not the only reasonable explanation" for the verdict. Id. at 184. At trial, the jury was instructed to "determine the amount of per share damages, if any, to which plaintiffs are entitled," and that their damage calculation was to be "expressed in terms of the daily amount of inflation per share." (Jury Charge No. 28.) The Verdict Form further directed the jury that in determining the amount of daily inflation it could consult PX-86 and DX-1878 "for guidance." (Verdict Form Question No. 58.) The jury was also instructed that, "[a]s the sole judges of the facts, you must decide which of the witnesses you will believe, what portion of their testimony to accept, and what weight to give it" and that they could give the expert testimony "whatever weight, if any, you find it deserves in light of all of the other evidence in this case." (Jury Charges Nos. 11, 14.) The jury was never instructed that it was required to either accept or reject the experts' daily inflation calculations wholesale, and no party ever requested such an instruction. The jury could therefore have understood that it was permissible to award a sum that it found to be reasonable, and certainly that it was permissible to award any sum that fell within the range presented by the experts at trial. See Ajax, 569 F.2d at 184 (finding no compromise where the jury could have concluded by following the court's instructions that it could award any amount of damages that fell below the contract amount). Thus, the half damages award does not suffice to prove that a compromise occurred.
PX-1486 and DX-1878 were the tables of daily inflation amounts prepared by Dr. Nye and Dr. Silber, respectively.
Of course, Ajax also raises the question of whether there was a reasonable basis in the evidence for the jury's calculations of daily inflation in this case. Vivendi takes the position on this motion that the expert testimony on damages was "all or nothing," in the sense that the jury had to either accept or reject Dr. Nye or Dr. Silber's analyses wholesale. Tellingly, Vivendi did not take that position before the jury began deliberating. Vivendi never asked the Court to instruct the jury that it had to accept Dr. Nye or Dr. Silber's analysis wholesale if it determined that damages were warranted. This failure certainly raises a question as to whether Vivendi has waived its right to challenge the jury's failure to adopt either expert's damage figures wholesale.
Waiver aside, it is well-established that the computation of damages is a quintessential fact issue for the jury, and that a jury need not accept an expert's damage calculations wholesale. See, e.g., Popovich v. Sony Music Entm't Inc., 508 F.3d 348, 359 (6th Cir. 2007) (upholding jury's award of damages for breach of contract which was substantially below the figure proposed by plaintiffs' expert and above the estimate proposed by defendant's expert); Am. Nat'l Bank Trust Co. of Chicago v. Reg'l Transp. Auth., 125 F.3d 420, 436-39 (7th Cir. 1997) (upholding jury's determination of fair market value for property that exceeded the sum requested by plaintiffs' counsel in closing argument and differed cfrom the calculations presented by both sides' experts); First Nat'l Bank of Kenosha v. United States, 763 F.2d 891, 896 (7th Cir. 1985) (jury finding that fair market value of a piece of real estate was $1.1 million had a reasonable basis in the record where the jury was presented with widely divergent expert opinions regarding the fair market value, and decided not to accept the view of either expert but
instead "arrived at its figure independently (very possibly, we suspect, by splitting the difference)").
In this case, the Court concludes that there was a reasonable basis in evidence for the daily inflation figures found by the jury. These figures fell within the range of daily inflation presented by the experts for each side. See Popovich, 508 F.3d at 359; First Nat'l Bank of Kenosha, 763 F.2d at 896. Vivendi argued that it was implausible that after removing market- and industry-wide reasons for declines in Vivendi's stock price, the entirety of the company-specific decline on the nine days in question was due to fraud-related events. ( See Tr. 6899:1-15.) But Vivendi's experts did not themselves present any analysis of the extent to which this was so. The jury may, therefore, have decided to reduce Dr. Nye's calculations of daily inflation to account for the possibility that company-specific news unrelated to the fraud was responsible for some portion of the decline in Vivendi's stock prices on the nine days. "Of course, we do not know for sure whether the jury actually made such calculations, but a jury has wide discretion in determining damages, so long as it has a reasonable basis." Dresser Indus., Inc. v. Gradall Co., 965 F.2d 1442, 1447 (7th Cir. 1992) (emphasis in original) (upholding jury award of lost profits even though it exceeded expert's estimates).
The cases cited by Vivendi do not compel a contrary conclusion. None of the cases Vivendi identifies concern a jury's damage award where a broad range of damages could be reasonably calculated, here, based on the jury's determination of the relevance of other company-specific news that was in the marketplace on the days in question. For example, Exodus Partners LLC v. Cooke, No. 04 Civ. 10239 (GEL), 2007 WL 120053 (S.D.N.Y. January 17, 2007), was a breach of contract case in which damages could either be either zero (if the jury found no breach of the contract) or $500,000 (if the jury found there was a breach). The jury awarded $330,000 in damages, and Judge Lynch held that that award lacked any reasonable evidentiary foundation. 2007 WL120053, at *14. Unlike Exodus (and Ajax), which involved contracts for a sum certain, daily inflation is difficult to quantify and inherently imprecise. Moreover, there was conflicting expert testimony in this case regarding the amount of damages. Vivendi, in effect, would have this Court adopt a rule that would preclude a jury from making any adjustments to damage figures calculated by an expert unless there is specific testimony in the record regarding the exact amount by which that expert's figures would be adjusted. The Court declines to do so. Vivendi also argues that there are a number of other "indicia of compromise" in the record that, viewed in conjunction with the half damages award, support the inference that the jury compromised on liability. Vivendi argues that "liability was close and vigorously contested," which the Second Circuit has suggested can buttress an inference of compromise. See Atkins, 143 F.3d at 104. Vivendi contends that it was clear that there was "difficulty in jury deliberations," id., because the jury deliberated for fourteen days over the span of three weeks, and many of the documents requested by the jury related to liability. Vivendi also suggests that the jury's exoneration of Hannezo and Messier is further proof of compromise.
The Court does not believe that any of these factors suffice to "clearly demonstrate" that the verdict reflected an impermissible compromise as to liability. See Maker, 253 F.2d at 419. Although liability in this case was hotly contested, there was a wealth of unusually incriminating evidence such as the Book of Warnings and Vivendi's stark admissions in the Arbitration Documents that supports the jury's verdict. The jury never indicated to the Court that it was deadlocked, and the fourteen days of deliberations must be viewed in context of the length and complexity of the trial, the presence of multiple defendants and the large number of statements on Table A. The jury's request for over 70 exhibits throughout the course of its deliberations supports an inference that it did its job carefully. The fact that the jury adjusted the expert's damage calculations to find zero inflation after September 11, 2001 — presumably in response to defendants' arguments that the markets experienced turmoil in the wake of 9/11 — further suggests that the jury considered the evidence in this case conscientiously. Finally, the jury's finding that Vivendi was liable but that Messier and Hannezo were not does not necessarily reflect a compromise, and could easily be explained by the differing admissibility of the evidence against the three defendants. Even if, arguendo, "one inference is as good as another," the record taken as a whole does not demonstrate that the verdict was an improper compromise. See Maker, 253 F.2d at 419. Accordingly, the Court will not order a new trial on this basis. Id.; see also Ajax, 569 F.2d at 184.
C. Alleged Failure to Identify Misstatements until the Close of Evidence
Vivendi also moves for a new trial on the ground that plaintiffs allegedly concealed their true list of alleged misstatements until after the close of the evidence, depriving Vivendi of a fair trial. To put this discussion in context, some background regarding the genesis of Table A to the Verdict Form (the list of alleged misstatements or omissions upon which the jury was asked to rule) is required. Towards the end of the trial, the Court solicited proposed verdict forms from all parties. Plaintiffs' proposed verdict form would have asked the jury a single question with respect to each defendant on plaintiffs' Section 10(b) claims, as follows: "Did [defendant] knowingly or recklessly make materially misleading or omissions that concealed liquidity risks at the company during the Class Period?" (Pl. Proposed Special Verdict Form (Dec. 14, 2009).) Plaintiffs' proposed verdict form did not identify any specific statements by the company. Defendants, on the other hand, submitted a very complex proposed verdict form that would have asked separate questions about whether plaintiffs had proven that Vivendi made a false or misleading statement regarding its liquidity; whether that statement was material; whether the statement was forward-looking and if so, whether it was accompanied by meaningful cautionary language or actual knowledge; whether the statement was made with an intent to defraud (and if so, whether it was made knowingly or recklessly); whether the statement caused inflation; and whether the revelation of the information concealed by the statement caused a decline in Vivendi's share price. (Defendants' Proposed Verdict Form (Dec. 14, 2009).) Defendants' proposed verdict form would have asked the jury to fill in a blank table at the back of the verdict form in which they would be required to identify the specific statement they had found to be materially false or misleading. ( Id.)
Plaintiffs' Verdict Form would then have asked the jury to determine whether these statements or omissions were made during the entire Class Period or during a certain portion of the Class Period.
The Court reviewed both of those proposals and also reviewed the verdict forms used in several recent securities class actions tried before a jury ( e.g., In re Apollo Group, Inc. et al, No. 04 Civ. 2147 (D. Ariz.); Jaffe v. Household Int'l, Inc., No. 02 Civ. 5893 (N.D. Ill.); In re Clarent Corp. Sec. Litig., No. 01 Civ. 3361 (N.D. Cal.); and In re JDS Uniphase Corp., No. 02 Civ. 1486 (N.D. Cal.)). The Court determined that plaintiffs' proposed verdict form was inadequate to the extent that its general question about whether plaintiffs' had made an "omission" was divorced from the requirements of Rule 10b-5, which makes it unlawful "to make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in light of the circumstances under which they were made, not misleading." 17 C.F.R. 240.10b-5(b) (emphasis added). Under the plain language of Rule 10b-5, an "omission" is not a violation unless plaintiffs can point to statements that were made misleading by the omitted facts. United States v. Finnerty, 05 Cr. 393 (DC), 05 Cr. 397 (DC), 2006 WL 2802042, at *6 (S.D.N.Y. Oct. 2, 2006). On the other hand, Vivendi's approach was also inadequate since it shifted the burden to the jury at the end of a long and complex trial to identify the statements that it found to be misleading — the jury, in effect, would be asked to weed through hundreds of lengthy documents itself and pick out the specific sentences it found to be misleading.
The Court, therefore, asked plaintiffs to propose an alternate verdict form that identified specific misstatements, consistent with the approach used in several of the recent securities cases cited above. Plaintiffs responded to the Court's request by proposing an appendix to the verdict form identifying the statements plaintiffs alleged to be actionable. That appendix went through various drafts and generated some back and forth between counsel and the Court regarding whether certain statements therein should be put to the jury. The Court also directed plaintiffs to provide defendants with an explanation of why they believed each statement on their appendix was materially misleading in order to enable defendants to prepare for closing arguments, which plaintiffs did. The finalized appendix was attached to the Verdict Form as "Table A."
This back and forth occurred during the approximate two-week holiday recess between the close of the evidence and the parties' closing arguments.
Vivendi now contends that this procedure was unfair, and that it was ambushed by plaintiffs' true list of misstatements at the end of trial when it was too late to defend against them. Vivendi's contention that it was ambushed is a gross exaggeration. Vivendi was aware long before trial that plaintiffs were alleging that the fifty-seven statements on Table A were misleading and also knew the reasons why plaintiffs believed these statements to be misleading. Most of the statements on Table A were specifically identified in plaintiffs' First Amended Consolidated Class Action Complaint, dated November 24, 2003, or were very similar to statements identified in the complaint. The vast majority of the statements listed on Table A (53 of 57) had also been identified by plaintiffs in response to an interrogatory from defendants asking plaintiffs to identify the allegedly false and misleading statements on which they were suing. By Vivendi's own admission, plaintiffs' interrogatory response was "enormously detailed." (Tr. 6673:19-24.) To be sure, four of the fifty-seven statements on Table A were not included in plaintiffs' interrogatory response. However, these four statements had been identified in plaintiffs' expert reports — which were filed long before trial — in a manner that clearly put defendants on notice that plaintiffs believed these four statements to be misleading. ( See Tr. 6737:12-6738:25 (describing ways in which defendants were put on notice that plaintiffs believed these four statements to be misleading).) Thus, defendants were aware long before trial that plaintiffs believed the fifty-seven statements identified on Table A were misleading. Defendants also knew long before trial why plaintiffs believed each of the statements was misleading. Vivendi had all the information it needed to effectively prepare a defense — and indeed, Vivendi did prepare and mount a detailed defense at trial.
Plaintiffs' interrogatory response (which was amended once) identified approximately one hundred and fifty statements, and included a detailed description of why plaintiffs believed each statement to be misleading. (See Margolies Decl. in Opp. to Vivendi's Rule 50/59 Motion at Ex. 67.)
It is true that plaintiffs did not narrow down their interrogatory response and finalize a more limited list of statements that they intended to put to the jury before trial (and that the Court never ordered plaintiffs to do so). But any unfairness Vivendi experienced from having to prepare to defend against so many alleged misstatements was not undue; it was simply a result of the fact that, according to plaintiffs (and now the jury), Vivendi concealed its true liquidity risk over a long period of time through a large number of misstatements. Plaintiffs could in fact have presented to the jury every single statement identified in their interrogatory response (to the extent such statements were not subject to exclusion as a matter of law) had they felt this would be a sound trial strategy. That plaintiffs chose instead to present only a subset of those statements to the jury at the end of the trial did not deprive Vivendi of a fair trial since Vivendi knew long before trial that plaintiffs believed all of the statements on Table A were misleading. There was no "miscarriage of justice" here, and the Court declines to order a new trial. See Manley, 337 F.3d at 245.
Vivendi contrasts the way the trial was conducted in this case to the way the trial was conducted in Jaffe v. Household Int'l, Inc., 02 Civ. 5893 (N.D. Ill.), pointing out that in Household, the plaintiffs were required to finalize a list of misstatements they intended to put to the jury before trial. Vivendi neglects to mention that the court in Household required plaintiffs to produce such a list only after the defendant wrote a lengthy letter to the Court before trial protesting plaintiffs' failure to finalize a list of misstatements, and asking the Court to compel plaintiffs to provide them with such a list. Vivendi — which has not hesitated throughout this trial to submit letters protesting any number of issues — made no such request after the Court denied defendants' motion for summary judgment on loss causation and set this case down for trial. Instead, Vivendi simply buried a vague, boilerplate-type assertion that "plaintiffs fail properly to identify the allegedly false or misleading statements of which plaintiffs complain" in a long list of defenses in the parties' lengthy joint pre-trial order. ( See Dixon Decl. Exs. 38, 39.)
D. Verdict Form
Vivendi also seeks a new trial on the ground that Verdict Form allegedly deprived it of a fair trial. Without citing any cases in which a court has ordered a new trial based on a verdict form's failure to require the jury to make particularized findings as to each element (and certain sub-elements) of a claim, Vivendi argues that a new trial must be granted here for that reason. "The use of special or general verdicts, as well as the content and form of any interrogatories submitted to the jury, are matters within the sound discretion of the district court." Micrel, Inc. v. TRW, Inc., 486 F.3d 866, 882 (6th Cir. 2007). "[V]erdict questions must be read in conjunction with the judge's charge to the jury." Vichare, 106 F.3d at 466; see also Romano v. Howarth, 998 F.2d 101, 104 (2d Cir. 1993). A new trial is warranted only if the questions mislead or confuse the jury, or if they inaccurately frame the issues to be resolved by the jury. Romano, 998 F.2d at 104.
Vivendi has not shown that they are entitled to a new trial under this standard. The Court decided on the format of the Verdict Form in this case after much consideration of the parties' proposals. Vivendi's proposed verdict form was extremely complex and, in the Court's view, unwieldy. Plaintiffs, on the other hand, proposed a very simple verdict form. The Court developed a Verdict Form that represented an appropriate compromise between the approaches proposed by both sides in an effort to make sure that the verdict was as straight-forward and manageable for the jury as possible. It was still quite complex, as the jury was asked to make decisions with respect to fifty-seven separate statements and three separate defendants, and was also asked to calculate daily inflation figures for 454 separate days, among other things. Vivendi takes issue with the fact that the Court did not require the jury to check separate boxes with respect to each element of a Section 10(b) violation, and with the fact that the Verdict Form did not include separate questions regarding forward-looking statements. But the Court made clear to the jury in its charges that they were required to find each element of a Section 10(b) violation against a particular defendant in order to find against that defendant on plaintiffs' Section 10(b) claim. (Jury Charge No. 22; see Verdict Form at Question 1-57.) The Court was also clear to the jury that different rules applied to any statements on Table A that were forward-looking, and it specifically instructed the jurors regarding the findings required for a forward-looking statement to be actionable. (Jury Charge No. 25.) It is presumed that the jury understood and followed these instructions, see United States v. Downing, 297 F.3d 52, 59 (2d Cir. 2002), and Vivendi has failed to rebut that presumption.
In addition, defense counsel's closing statements emphasized the elements that plaintiffs had to prove each element of a Section 10(b) claim, presenting a graphic of depicting a hurdler who had to get over four hurdles (material misstatement or omission, scienter, reliance, and loss caused by the fraud) in order to reach the finish line. ( See, e.g., Tr. 6757:25-6760:2; 6759:15-16; 6846:24-6847-2; 6848:17-22; 6866:8-21; 6885:23-6886:1; 6870:13-25.) The jury was certainly aware that it had to find each element of a Section 10(b) violation as against a particular defendant in order to find that defendant liable.
Vivendi argues that the fact that the jury found Vivendi liable but exonerated Messier and Hannezo proves that the jury was confused by the Verdict Form. The Court has already rejected this argument. The differing admissibility of the evidence provides a perfectly reasonable explanation for the differing verdicts against Vivendi and the two individual defendants. Overall, when the Verdict Form is read conjunction with the Court's instructions, the Court finds that it did not mislead, confuse or inaccurately frame the issues and a new trial is not warranted. See J.K. Walkden, Ltd. v. Lord Taylor, No. 93 Civ. 6084 (BSJ), 2001 WL 619040, at *5 (S.D.N.Y. June 6, 2001) (holding that verdict form did not warrant a new trial).
Vivendi also objects to the grouping of related statements from the same document — for example, two excerpts from consecutive paragraphs of the same document — into single entries on Table A to the Verdict Form. Instead, Vivendi would have broken down the fifty-seven entries on Table A into eighty-one separate statements. Vivendi has cited no authority suggesting the Court's approach was improper, and the Court is aware of none. The Court's approach was also consistent with the verdict form used in Jaffe v. Household, which Vivendi cites in other contexts as a model that the Court should have followed.
E. Plaintiffs' Summation
Vivendi's final challenge is that plaintiffs allegedly made three "highly improper" and inflammatory arguments during their summation regarding the "Arbitration Documents" which unfairly influenced the jury's verdict. The challenged statements from plaintiffs' closing are as follows: first, a statement that Vivendi "fought mightily" to exclude the Arbitration Documents from the trial ( see Tr. 7206:19-7207:1); second, a statement that the redacted portions of the Arbitration Documents, which were not shown to the jury, were "even more powerful" than the parts the jury was permitted to see (Tr. 4017:1-4); and third, a statement that Vivendi contends improperly implied to the jury that Vivendi had won the proceedings against Messier in the New York Supreme Court (which it did not). (Tr. 7209:2-12.) Vivendi contends that the first two statements can only have been intended to elicit a prejudicial reaction from the jury that Vivendi was improperly trying to keep relevant information from it. Vivendi argues the third statement improperly suggested to the jury that Vivendi won proceedings in the New York Supreme Court, in violation of this Court's ruling that the results of the New York case could not be raised with the jury.
As noted earlier, the "Arbitration Documents" were filings made by Vivendi, through its former counsel, in a separate arbitration and litigation between Vivendi and Messier regarding Messier's termination and severance. Those documents were heavily redacted at the Court's direction, as the Court found that the vast majority of the statements therein were not relevant to the issues in this trial. Only certain sentences, in which the company made admissions having direct relevance to this case, were left un-redacted and shown and read to the jury.
It is well established that "[n]ot every improper or poorly supported remark made in summation irreparably taints the proceedings; only if counsel's conduct created undue prejudice or passion which played upon the sympathy of the jury, should a new trial be granted." Marcic v. Reinauer Transp. Cos., 397 F.3d 120, 127 (2d Cir. 2005); see also Parrish v. Sollecito, 280 F. Supp. 2d 145, 168 (S.D.N.Y. 2003) ("A new trial is only warranted where the attorney's concluding argument deprived the opposite party of a fair trial.") (internal quotation marks omitted). Whether to order a new trial due to remarks made during summation "is within the broad discretion of the trial court." Parrish. 280 F. Supp. 2d at 168.
The Court has considered the "totality of the circumstances" surrounding the challenged statements from plaintiffs' summation, see Marcoux v. Farm Serv. Supplies, Inc., 290 F. Supp. 2d 457, 472 (S.D.N.Y. 2003), and finds that the challenged statements do not warrant a new trial. During Vivendi's closing, which was delivered before plaintiffs' closing, Vivendi's counsel had showed a slide to the jury which stated, "Plaintiffs Do Not Tell You the Whole Story" in large font. Around the time that slide was shown to the jury, counsel argued that plaintiffs did not tell "the whole truth" and had shown the jury only snippets of documents. Counsel then began discussing the highly-redacted Arbitration Documents which created the possible inference that plaintiffs had a hand in withholding information contained therein from the jury.
While the Court did not draw that inference — it simply understood defense counsel's reference to the mostly blank pages as an attempt to identify the Arbitration Documents he was about to discuss — it was not unreasonable for plaintiffs' counsel to interpret the statement as he did, and some jurors may have interpreted it that way. As a consequence, plaintiffs' counsel responded as he did.
While some of the rejoinder may have been ill-advised, overall, plaintiffs' counsel was simply responding to statements that could have been misconstrued by the jury as implying that plaintiffs had concealed large portions of the Arbitration Documents.
Vivendi also objects to the statement that "[w]hatever is convenient to win the case, that's what they [Vivendi] do, and that, ladies and gentlemen, I think is heads I win in the Supreme Court, tails I win in the federal court, with the same story just spinning it a different way." Vivendi contends this statement clearly implied to the jury that Vivendi won the proceedings in state court. To be sure, counsel's language could have been more precise, and one might have interpreted it as defendants did. However, it seems equally likely that counsel was not suggesting that Vivendi won the New York proceedings, but was simply comparing the position that Vivendi took to try to win those proceedings (that Messier caused a liquidity crisis that threatened the company's very existence) to the position it took in this trial (that there was no liquidity crisis). To the extent any jurors may have drawn an inference that Vivendi won the proceedings in state court, any prejudice was cured by the special instruction the Court gave to the jury regarding the Arbitration Documents after plaintiffs' closing. See United States v. Rodriguez, 968 F.2d 130, 142 (2d Cir. 1992) ("It is a `rare case' in which improper comments in a prosecutor's summation are so prejudicial that a new trial is required.") (quoting Floyd v. Meachum, 907 F.2d 347, 348 (2d Cir. 1990)); Am. Nat'l Fire Ins. Co. v. Mirasco, Inc., 451 F. Supp. 2d 576, 583 (S.D.N.Y. 2006) (instruction that jury was not to speculate about Court's determination regarding a particular issue after counsel made suggestive remarks regarding the Court's decision in closing arguments cured any prejudice caused by counsel's remarks). The Court presumes that the jury in this case heeded that instruction. See Zafiro v. United States, 506 U.S. 534, 540-41 (1993); United States v. Downing, 297 F.3d 52, 59 (2d Cir. 2002) ("Absent evidence to the contrary, we must presume that juries understand and abide by a district court's limiting instructions."). This is a far cry from the type of egregious misconduct that cannot be alleviated by curative instructions. Cf. Koufakis v. Carvel, 425 F.2d 892, 901 (2d Cir. 1970).
The specific instruction was as follows: "During the trial and in closing arguments there have been many references to the statements made on behalf of Vivendi in a separate arbitration [and] litigation between Mr. Messier and Vivendi over his termination agreement. I ruled that certain of his statements were relevant to this case and those statements were admitted into evidence and read to you. You may consider these statements. I also excluded or redacted other portions of documents that were not relevant to this case, and you should not speculate as to what was or was not in those redacted portions or speculate as to what the outcome of that separate dispute was. Nor should you hold against any party the Court's decision on any evidentiary matter, either on this or any other issue." (Tr. 7496:23-7497:10.) The Court had previously given the jury similar instructions more than once during the trial. ( See Tr. 4016:17-4017:20; 4042:2-11.) Thus, the jury was well aware that the excluded portions of the documents were not relevant to this case, and that it was not to speculate about who won the New York Supreme Court proceedings.
In a footnote, Vivendi also renews a laundry list of objections that it raised at the time of plaintiffs' summation. The Court considered those objections at the time they were raised, and sees no reason to reconsider its determination that plaintiffs' closing did not exceed the bounds of permissible argument. To the extent that any isolated statements were not fully supported by the record, the Court finds that it is not reasonably probable that the verdict was influenced by any such misstatements in light of the length of the closing arguments, the large number of issues in dispute, and the Court's clear instruction to the jury that "[w]hat the lawyers have said in their opening statements, in their closing arguments, in their objections or in their questions is not evidence, nor is anything I may have said during the trial or may say during these instructions to be considered by you as evidence." (Tr. 7495:1-5.) See Malmsteen v. Berdon LLP, 595 F. Supp. 2d 299, 310 (S.D.N.Y. 2009) ("While certain of these statements may have been on shaky evidentiary footing, or were otherwise overzealous, they were not `so inflammatory or so unsupported by the record as to affect the integrity of the trial.") (internal citation and quotation omitted).
IV. Class Plaintiffs' Motion for Entry of Final Judgment
Plaintiffs have also moved for the entry of final judgment. Vivendi opposes that motion, arguing that entry of a final judgment is premature for a number of reasons, most significantly, that Vivendi is entitled to rebut the presumption of reliance on the market price of Vivendi's stocks with respect to particular class members. The Court agrees that Vivendi is entitled to rebut the presumption of reliance on an individual basis and that entry of judgment is, therefore premature.
In this case, plaintiffs proved the element of "justifiable reliance" on Vivendi's misrepresentations on a class-wide basis through the fraud-on-the-market theory. This theory, which applies to well-developed securities markets, assumes that the market price of stock reflects all available public information, including material misrepresentations, and assumes that investors in an efficient securities market reasonably rely on the integrity of the market price of securities. Basic v. Levinson, 485 U.S. 224, 245-47 (1998); Hevesi v. Citigroups, Inc., 366 F.3d 70, 77 (2d Cir. 2004). Therefore, a party who purchases securities in an efficient market need not prove that they directly relied upon or even knew of the alleged misrepresentations, since reliance is assumed once the materiality on an omission is established. See id.
It is well-established that the presumption of reliance on the market price of a security under the fraud-on-the-market theory is rebuttable. As the Supreme Court stated in Basic, "Any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price, will be sufficient to rebut the presumption of reliance." 485 U.S. at 248. The Supreme Court identified several ways in which a defendant could rebut the presumption:
For example, if petitioners could show that the "market makers" were privy to the truth . . . and thus that the market price would not have been affected by their misrepresentations, the causal connection could be broken: the basis for finding that the fraud had been transmitted through market price would be gone. Similarly, if, despite petitioners' allegedly fraudulent attempt to manipulate market price, news [of the allegedly concealed information] credibly entered the market and dissipated the effects of the misstatements, those who traded Basic shares after the corrective statements would have no direct or indirect connection with the fraud. Petitioners also could rebut the presumption of reliance as to plaintiffs who would have divested themselves of their Basic shares without relying on the integrity of the market. For example, a plaintiff who believed that Basic's statements were false . . . and who consequently believed that Basic stock was artificially underpriced, but sold his shares nevertheless because of other unrelated concerns, e.g., potential antitrust problems, or political pressures to divest from shares of certain businesses, could not be said to have relied on the integrity of a price he knew had been manipulated.Id. at 248-49. This excerpt suggests that some of the means of rebutting the presumption of reliance can be proven on a class-wide basis, and some necessarily require individualized inquiry. For example, a party may rebut the presumption of reliance by showing that "the market" already knew the truth that was allegedly omitted from a company's statements, such that any omissions cannot have been material — this is known as the "truth on the market" defense. See Ganino, 228 F.3d at 167 ("A defendant may rebut the presumption that its misrepresentations have affected the market price of its stock by showing that the truth of the matter was already known."). Such a rebuttal is properly done as part of a class-wide trial since questions about what "the market" as a whole knew are common to all class members. In this case, Vivendi tried to rebut the fraud-on-the-market presumption through a "truth on the market" defense, arguing that whole market knew Vivendi's true liquidity condition ( see, e.g., Tr. at 6866:25-6867:1; 6869:3-4), but the jury was not persuaded.
Similarly, a party might attempt to rebut the fraud-on-the-market presumption by presenting evidence that the market of the security in question was not actually efficient, such that the presumption should not apply. Questions about market efficiency are appropriate resolved at a class wide trial.
On the other hand, certain means of rebutting the presumption of reliance require an individualized inquiry into the buying and selling decisions of particular class members. For example, the Supreme Court stated in Basic that the presumption of reliance would be rebutted if the defendant could show that a particular investor would have purchased a company's stock even if she had known of the fraud, or that a particular investor purchased even though she did actually know of the fraud. 485 U.S. at 248; In re WorldCom Inc. Sec. Litig., 219 F.R.D. 267, 292 (S.D.N.Y. 2003). Alternatively, if a particular investor relied upon information not generally available to the public, it may be argued that that particular investor did not rely upon the integrity of the market. See Swack v. Credit Suisse First Boston, 230 F.R.D. 250, 262 (D. Mass. 2005) (citing Grace v. Perception Tech. Corp., 128 F.R.D. 165, 169 (D. Mass. 1989)). Logically, any attempt to rebut the presumption of reliance on such grounds would call for separate inquiries into the individual circumstances of particular class members. See id.; 7 NEWBERG ON CLASS ACTIONS 22:61 ("[R]ebuttal of individual reliance will not defeat class certification and may be resolved after trial on common issues. With the exception of attempted rebuttals of reliance based on information widely available to class members which may give rise to a common issue for the class, a rebuttal of reliance by a particular class member must necessarily be on an individual basis because there can be no class presumption of non-reliance.") For this reason, courts in securities fraud actions have consistently recognized that issues of individual reliance can and should be addressed after a class-wide trial, through separate jury trials if necessary. See, e.g., Swack, 230 F.R.D. at 263-64 (granting motion to certify class despite defendants' contention that individual issues of non-reliance predominated, and reasoning that action should proceed in a two phases, a first phase dealing with class-wide issues and a second phase dealing with individualized issues of reliance or damages); Jaffe v. Household, No. 02 Civ. 5893, 2005 WL 3801463, at *4 (N.D. Ill. Apr. 18, 2005) (denying discovery into individual claimants' reliance on the ground that issues relating to individual reliance could be adjudicated after class-wide issues have been determined); In re Lucent Techs. Inc. Sec. Litig., No. 00 Civ. 621 (JAP), 2002 WL 32818345, at *2 (D.N.J. May 9, 2002) (delaying discovery concerning plaintiffs' investment history until a later stage in which "individualized rebuttal proceedings may be pursued to determine whether a claimant may recover, once the matter of liability has been adjudicated"); In re ICN/Viratek Sec. Litig., No. 87 Civ. 4296 (KMW), 1996 WL 34448146, at *1 (S.D.N.Y. July 15, 1996) (granting class plaintiffs' motion to bifurcate trial into a trial on class-wide issues,
followed, if necessary, by separate trials on individualized reliance and damages); Biben v. Card, 789 F. Supp. 1001, 1003 (W.D. Mo. 1992) (granting defendant's motion to bifurcate trial into a class-wide liability phase and a proof-of-claim phase at which individualized issues relating to damages and reliance could be addressed); see also Fisher v. The Plessey Co., Ltd., 103 F.R.D. 150, 156 (S.D.N.Y. 1984) ("[D]efendants may seek to rebut a presumption of reliance by demonstrating that individual debenture holders [of publicly traded securities publicly] had access to and knowledge of the omitted information, and therefore placed no reliance on the tender documents. . . . If necessary, the Court can hold separate hearings on the issue of reliance.")
Plaintiffs' suggest that Vivendi already used its opportunity to rebut the presumption of reliance, but the record does not support that contention. While Vivendi attempted to rebut the presumption of reliance on a class-wide basis at trial by showing that the allegedly omitted information was immaterial, Vivendi did not challenge the individual reliance of each class member at trial — indeed, Vivendi could not have done so, as Vivendi does not yet know the identity of most class members. There is no evidence that Vivendi ever waived its right to contest individualized reliance in this case, and it seems clear that plaintiffs themselves contemplated that issues of individual reliance might have to be addressed in separate proceedings after a trial on class-wide issues. Though the Court did not issue a formal order before trial clarifying that issues
For instance, in a brief submitted to this Court on February 23, 2009 regarding issues of consolidation, plaintiffs stated the following: "If plaintiffs are successful in establishing liability and causation and securing damage awards from the jury at the single trial, then any individualized issues relating to the class representatives, absent class members, individual plaintiffs or GAMCO, to the extent any exist, can be resolved in a subsequent proceeding, the scope and nature of which can be determined at a later date." (Dkt. 774.) Shortly before trial, defendants sought to compel the trial testimony of two representatives from Gamco who were not part of the class. In response, plaintiffs filed a motion in limine in which they asked the Court to bar testimony by any Gamco representatives, and "further request[ed] that the Court exclude all evidence concerning any individualized issues, e.g. reliance, to the extent that they exist, even for the named plaintiffs in this class." (See Dkt. 938, 940.) They argued that "[r]esolution of any individualized issues relating to the named plaintiffs in the class action should await the conclusion of the class trial." ( Id.) The Court granted plaintiffs' motion to exclude the testimony of the two Gamco representatives orally, but by oversight, failed to address plaintiffs' separate request to exclude all issues of individualized reliance from the trial. (See Tr. at 5:10-20.) Neither party sought clarification from the Court as to whether plaintiffs' request to exclude all issues of individualized reliance had been ruled upon.
of individual reliance were reserved for after the class trial, the absence of such an order is not dispositive. See Evans v. Connecticut, 168 F.R.D. 118, 120 (D. Conn. 1996) (absence of court order bifurcating trial into liability and damages phase did not bar plaintiff in Title VII action from seeking back pay and reinstatement after a trial on liability issues even though he failed to present evidence regarding such issues during trial). Here, it seems clear that all parties were on notice that individual reliance issues might require resolution in separate proceedings after the class trial.
Plaintiffs have not identified a single case in which a court in a securities fraud class action has precluded a defendant from contesting issues of individualized reliance after a class-wide trial on liability issues, where issues of individual reliance were not litigated as part of the class-wide trial and where the defendant asserted an intent to challenge individual reliance issues after the class-wide trial.
The Court's conclusion that Vivendi is entitled to an opportunity to rebut the presumption of reliance on an individual basis does not answer the question of what procedures should be used during the individual reliance phase. See In re ICN/Viratek, 1996 WL 34448146, at *4 (directing the parties to submit further briefing as to what the Seventh Amendment requires with respect to the procedures during the individual reliance and damages phase of a bifurcated securities fraud action). And it remains to be seen whether Vivendi will actually be able to prove that any individual claimants would have purchased Vivendi shares even if they had known of the fraud. "It has been noted that `it is hard to imagine that there ever is a buyer or seller who does not rely on market integrity. Who would knowingly roll the dice in a crooked crap game?'" Basic, 485 U.S. at 246-47 (quoting Schlanger v. Four-Phase Sys. Inc., 555 F.Supp. 535, 538 (S.D.N.Y. 1982)). But Vivendi apparently believes there are such investors, and is entitled to an opportunity to try to show that.
The Court notes that whether short sellers can benefit from the fraud on the market presumption is unsettled. Compare Schleicher v. Wendt, No. 02 Civ. 1332, 2009 WL 761157, at *9 (S.D. Ind. Mar. 20, 2009) (rejecting argument that short sellers cannot rely on the fraud-on-the-market presumption on the ground that "[t]heir decisions about the value of the stock, however, can still be based on the integrity of the market price") and Jones v. Intelli-Check, Inc., 274 F. Supp. 2d 615, 632-33 (D.N.J. 2003) (holding that short sellers are not entitled to the presumption of reliance provided by the fraud on the market theory).
In light of the Court's conclusion that Vivendi is entitled to rebut the presumption of reliance on an individual basis, entry of final judgment at this juncture would be premature See Jaffe v. Household Int'l Inc. et al, No. 02 Civ. 05893 (N.D. Ill.), Docket Entry. No. 1697 (striking motion for entry of judgment after jury verdict on class-wide claims in securities class action on the ground that it would be premature to enter judgment before the second phase of the case had concluded); Taylor v. Bd. of Educ., 288 F.2d 600, 602 (2d Cir. 1961) (judgment was not final where it resolved issues of liability but left the quantum of relief unresolved). The issues that remain to be resolved here are neither "routine" nor "ministerial." Cf. Morgan v. United States, 968 F.2d 200, 204 (2d Cir. 1992) (treating district court's order which remanded case to magistrate judge to make an arithmetic correction to the judgment as final and appealable on the ground that the "ministerial or arithmetic computation" that remained was "merely routine" and "[would] not spark an appeal," such that "permitting the earlier review will not thwart the policy against piecemeal appeals"). In addition, it may be that the methods for calculating an individual claimant's damages will be hotly contested and may trigger additional appeals. This further counsels against entry of judgment at this stage. In light
Although plaintiffs have identified two securities fraud actions in which courts have entered final judgments after a class-wide trial based on the jury's computation of daily inflation figures, see In re Apollo Group, Inc. Sec. Litig., No. 04 Civ. 2147, 2008 WL 410625 (D. Ariz. Feb. 13, 2008); Backman v. Polaroid Corp., 893 F.2d 1405, 1408 (1st Cir. 1998), there is no indication that the defendants in either of those cases sought to challenge issues of individual reliance. Thus, the issues that remained to be addressed after the class trial in those cases may well have been "routine" and "ministerial" calculations of class members' damages. That is not the case here.
of the Court's conclusion that entry of judgment is premature, the Court will deny, without prejudice to renew plaintiffs' motion for pre-judgment interest and for approval of plaintiffs' proposed claims administration procedure and class notice.
CONCLUSION
For the foregoing reasons, the Court concludes that claims brought by purchasers of Vivendi's ordinary shares must be dismissed under Morrison. Going forward, the class shall consist of all persons from the United States, France, England and the Netherlands who purchased or otherwise acquired American Depositary Shares of Vivendi between October 30, 2000 and August 14, 2002.
Vivendi's motion [969] for judgment as a matter of law, or in the alternative, for a new trial, is denied for the reasons described above, except that the Court grants judgment as a matter of law in Vivendi's favor with respect to Statement No. 55 on the Verdict Form. Class Plaintiffs' motion for entry of judgment [1013] is denied on the ground that it is premature to enter final judgment prior to the damages phase of this litigation. In light of the Court's conclusion that entry of judgment is premature, Class Plaintiffs' motions for an award of pre-judgment interest [1018] and for approval of their proposed notice and claims administration procedure [1024] are denied without prejudice and with leave to re-file at a later date as appropriate.
SO ORDERED.