In re American Intern. Group, Inc. Securities Litigation

Full title: In re AMERICAN INTERNATIONAL GROUP, INC. SECURITIES LITIGATION. This…

Court: United States District Court, S.D. New York.

Date published: Feb 22, 2010

265 F.R.D. 157 (S.D.N.Y. 2010)

Opinion

265 F.R.D. 157 (S.D.N.Y. 2010) In re AMERICAN INTERNATIONAL GROUP, INC. SECURITIES LITIGATION. This Document Relates To: All Actions. No. 04 Civ. 8141(DAB). United States District Court, S.D. New York. February 22, 2010

[Copyrighted Material Omitted] [Copyrighted Material Omitted]

OPINION

DEBORAH A. BATTS, District Judge.

Lead Plaintiffs Ohio Public Employees Retirement System (" OPERS" ), State Teachers Retirement System of Ohio (" STRS Ohio" ), and Ohio Police & Fire Pension Fund (" OP & F" ) (collectively, " Lead Plaintiffs" ), bring a proposed class action against Defendants American International Group (" AIG" ), its former Chairman and CEO Maurice " Hank" Greenberg, its outside auditors PricewaterhouseCoopers LLP (" PwC" ), and numerous other corporate and individual defendants associated with AIG (hereinafter " Defendants" ) for violations of Sections 11 and 15 of the Securities Act of 1933, 15 U.S.C. §§ 77k and 77 o, Sections 10(b), 20(a), and 20A of the Securities Exchange Act of 1934, 15 U.S.C. §§ 78j(b), 78t(a) and 78t-1, and SEC Rule 10b-5. ¹

1 Although The Florida State Board of Administration, Plaintiff in related case 05 Civ. 7886(DAB), has opted-out of this suit, that matter remains consolidated with this case for purposes of discovery. ( See June 8, 2006 Stipulation of Discovery.)

Lead Plaintiffs allege that during the proposed Class Period of October 28, 1999 to April 1, 2005, Defendants made numerous and repeated material omissions and misstatements, including in AIG's quarterly and annual financial statements, leading to the artificial inflation of AIG's stated financial assets. Lead Plaintiffs further allege that these misstatements and the inflated prices misled investors who purchased AIG securities and ultimately caused them substantial economic harm when the price of AIG securities declined due to the public disclosure of those omissions and misstatements.

Lead Plaintiffs now move for class certification under Federal Rule of Civil Procedure 23(a) and (b)(3). For the following reasons, Lead Plaintiffs' Motion to certify the class of AIG bondholders and Defendants Wachovia Securities, Merrill Lynch, Gen Re, Ferguson, Houldsworth, and Napier is DENIED. However, after making certain modifications to the definition of the class as described herein, Lead Plaintiffs' Motion to certify the class of equity stockholders is GRANTED.

I. BACKGROUND

A. Parties

Lead Plaintiffs are three public pension funds that are instrumentalities of the State of Ohio, serve hundreds of thousands of current and former state employees, and maintain combined assets of well over $100,000,000,000.00. (Third Amended Complaint (hereinafter " TAC" ), ¶ ¶ 38-41.)

Defendant AIG is a Delaware corporation, with its principal place of business in New York, New York, engaging in the business of domestic and international insurance and insurance-related activities. (TAC, ¶ 44.) AIG's securities are registered under § 12(b) of the Exchange Act and are listed on the New York Stock Exchange and several other exchanges around the world. (TAC, ¶ 44.)

Defendant Maurice R. " Hank" Greenberg was, until March 14 and June 8, 2005, respectively, AIG's Chief Executive Officer and Chairman of the Board. (TAC, ¶ 45.) Defendant Greenberg is President, Chief Executive Officer, and Chairman of Defendant C.V. Starr & Co., Inc., a private Delaware corporation with its principal place of business in New York, New York, engaging in the business of providing commercial casualty insurance. (TAC, ¶ ¶ 46, 105.) Defendant Greenberg is also a director of Defendant Starr International Company, Inc. (" SICO" ), a Bermuda-based private company incorporated in Panama, primarily responsible for providing AIG executives with incentive-based compensation. (TAC, ¶ ¶ 47, 93-94.)

Defendant Howard I. Smith was a director of AIG's Board, and until March 21, 2005, AIG's Vice Chairman, Chief Financial Officer, and Chief Administrative Officer. (TAC, ¶ ¶ 53.) Defendant Martin J. Sullivan is a Director of AIG's Board, and until March 14, 2005, when he was elected Chief Operating Officer as a replacement for Greenberg, AIG's Vice Chairman and Co-Chief Operating Officer. (TAC, ¶ 59.) Defendant Thomas R. Tizzio was a Director of AIG's Board and Senior Vice Chairman of AIG's General Insurance business segment until May 2003. (TAC, ¶ 64.) Defendant Michael J. Castelli was Vice President and Comptroller of AIG until April 15, 2005, when he was placed " on leave." (TAC, ¶ ¶ 67, 69.) Defendant Christian M. Milton was AIG's Vice President of Reinsurance until March 21, 2005, when he was terminated. (TAC, ¶ 71.) Defendant Michael L. Murphy was AIG's Legal Counsel and a senior executive in its Bermuda office until March 27, 2005, when he was terminated. (TAC, ¶ ¶ 72, 74.)

Defendant John A. Graf was Executive Vice President of AIG from August 2001 to September 2004. (TAC, ¶ 77.) Defendant Frank J. Hoenemeyer was a Director of AIG, a member of the Board's Audit Committee, and chair of the Audit Committee during the year of 2003. (TAC, ¶ 81.) Defendant Eli Broad was an AIG Director from 1999 to 2002 and a member of the AIG Finance Committee. (TAC, ¶ 89.)

Defendant Union Excess Company, Ltd. is a Barbados-domiciled company primarily in the business of reinsuring risks arising from AIG subsidiaries. (TAC, ¶ 112.) Defendant Richmond Insurance Company, Ltd. is a Bermuda-based reinsurance company affiliated with Defendant AIG. (TAC, ¶ 114.)

Defendant General Reinsurance Corp. (" Gen Re" ) is a wholly-owned subsidiary of Berkshire Hathaway, in the business of providing global reinsurance, as well as risk assessment, transfer, and management operations. (TAC, ¶ 124.) Defendant Ronald E. Ferguson was Gen Re's Chairman and Chief Executive Officer until October 2001 and June 2002, respectively. (TAC, ¶ 124.) Defendant John B. Houldsworth was Chief Executive Officer of Cologne Re Dublin, a subsidiary of Gen Re, and Defendant Richard Napier was Senior Vice President at Gen Re's Stamford, Connecticut office until June 8, 2005 (collectively the " Gen Re Defendants" ). (TAC, ¶ ¶ 128, 133.)

Defendant PricewaterhouseCoopers LLP (" PwC" ) is a limited liability partnership in New York, New York that served as AIG's outside auditor and principal accounting firm prior to and during the Class Period. (TAC, ¶ 118.) Defendant Wachovia Securities, Inc. is a Virginia-based corporation and underwriter of AIG debt securities. (TAC, ¶ ¶ 138-39.) Defendant Merrill Lynch & Co is a New York-based corporation and underwriter of AIG debt securities. (TAC, ¶ ¶ 141-42.)

B. The Contingent Commissions Payments and Bid-Rigging Scheme

Lead Plaintiffs allege that since the 1990s, AIG and other insurance companies paid non-party Marsh & McLennan Companies, Inc. (" Marsh" ) billions of dollars in illegal contingent commissions in return for Marsh steering business towards them, all the while styling these payments as related to various " services." (TAC, ¶ ¶ 232, 236.) On January 1, 2003, AIG and Marsh began implementing one such contingent commission agreement, resulting in billions of dollars of revenue for AIG during the Class Period. (TAC, ¶ ¶ 237, 245.) As a result, AIG's SEC filings during the Class Period contained false and materially misleading statements regarding its premium revenues as well as the level of competition in the insurance industry. (TAC, ¶ ¶ 245-46.)

In addition, in conjunction with the contingent commission arrangements, AIG took part in an illegal scheme to rig bids for quotes that Marsh presented to its clients, thus ensuring that AIG would be selected by Marsh to win these bids and provide coverage to the clients. (TAC, ¶ ¶ 247-49.) This bid-rigging also made AIG's public statements during the Class Period false and materially misleading in regard to premium revenues and competition in the insurance industry. (TAC, ¶ 250.)

C. Accounting Frauds

Lead Plaintiffs further allege that Defendants engaged in approximately eleven other " core" transactions and up to seventeen additional " non-core" transactions that are the subject of the Third Amended Complaint. (February 6, 2009 Tr., 10:21-11:13.)

Among the most substantial of these additional transactions, Lead Plaintiffs allege that in late 2000 and 2001,² AIG and Gen Re completed a reinsurance transaction between their subsidiaries in which AIG booked $500,000,000.00 in premium revenue and $500,000,000.00 in claims reserves over the fourth quarter of 2000 and the first quarter of 2001. (TAC, ¶ ¶ 331-32, 386-87, 397.) However, while AIG nominally agreed to assume $100,000,000.00 of Gen Re's insurance risk as part of the transaction, in reality that risk was nothing more than a fiction. (TAC, ¶ ¶ 331-32, 386-87, 397.)

2 Lead Plaintiffs also allege that in 1998 and 1999, AIG sold non-traditional insurance products to non-party Brightpoint, which had the effect of illegally " smoothing" Brightpoint's earnings and violating generally accepted accounting principles (hereinafter " GAAP" ). (TAC, ¶ ¶ 620-21.)

Additionally, on March 30, 2005, Defendants Richmond Insurance and Union Excess, who had engaged in prior reinsurance transactions with AIG, were disclosed by AIG to be under its control, thus requiring AIG to account for liabilities that it had previously transferred to Richmond Insurance and Union Excess. (TAC, ¶ ¶ 359, 465-66, 470-71.) AIG announced that this would lead to a reduction of approximately $1,100,000,000.00 in AIG shareholder equity. (TAC, ¶ 359.)

Further, Lead Plaintiffs allege that during the Class Period AIG made " top side" adjustments to its reserves in the tens of millions of dollars, including in late 2000 and early 2001. (TAC, ¶ ¶ 505-06.) AIG is also alleged to have engaged in multiple schemes to hide AIG underwriting losses and to mischaracterize its revenues from so-called " life settlement" policies. (TAC, ¶ 518.)

AIG also failed to disclose the ongoing SEC investigation into these transactions until September 2003, when the SEC filed a civil action against it. (TAC, ¶ ¶ 651-57.) Between February 2000 and January 2002, AIG further engaged in the marketing and sale of other such improper insurance products, including to non party PNC Bank. (TAC, ¶ ¶ 658-77.) AIG then misled the public about the scope of the resulting investigations by the SEC, the U.S. Department of Justice, and the New York Attorney General. (TAC, ¶ ¶ 716-21.)

AIG completed its acquisition of HSB Group on November 22, 2000, while its stock price was alleged to have been inflated by the then existing frauds. (TAC, ¶ ¶ 179-80.) Lead Plaintiffs also allege that Defendant Greenberg attempted to manipulate the price of AIG stock on several occasions, including during AIG's August 29, 2001 acquisition of American General, and on multiple occasions in February 2005, prior to Greenberg's departure as AIG's Chief Executive Officer. (TAC, ¶ ¶ 729-30, 749-50, 759.) As a result of AIG's acquisition of American General and HSB Group during the time when AIG's stock was inflated, HSB Group and American General shareholders who engaged in the resulting stock for stock transfer on the basis of the relative prices of the companies' stock are alleged to have received materially fewer shares of AIG stock than they would have, had the truth about AIG's misstatements and omissions been known. (TAC, ¶ 179.)

D. Disclosures

Lead Plaintiffs allege that on six days during the Class Period, declines in the price of AIG securities occurred directly as a result of AIG's public disclosure of the foregoing illegal conduct and misleading statements and omissions. (TAC, ¶ 762; Lead Pits' Mem. of Law, 19.) First, on October 14, 2004, the New York Attorney General announced that AIG had been implicated in a scheme to rig bids and pay contingent commissions, contributing to a decline of 10.43% in AIG stock from $66.99 to $60.00. (TAC, ¶ ¶ 222-26, 762.) Then, on October 15, 2004, Defendant Greenberg, in a call with analysts, admitted that AIG had received a subpoena from the New York Attorney General's office in September 2004 and had launched an internal investigation, which contributed to a decline of 3.58% in AIG stock from $60.00 to $57.85. (TAC, ¶ ¶ 302-03, 762.)

Lead Plaintiffs also allege that on March 17, 2005, the media's widespread reports that the SEC and New York Attorney General had expanded their probes into AIG to include additional accounting issues, helped cause a stock decline of 3.34% from $62.90 to $60.80. (TAC, ¶ ¶ 346-48, 578, 762.) Subsequently, on March 30, 2005, AIG issued a press release relating to the accounting for its Gen Re reinsurance transaction and announcing a reduction in shareholder equity of $1,770,000,000.00, contributing to a 1.79% decline in AIG stock from $58.20 to $57.16. (TAC, ¶ ¶ 357-62, 465-67, 579-80, 762.) On March 31, 2005, the media reported that PwC had received a subpoena related to AIG and that the Government's investigation into AIG might widen beyond the scope of AIG's March 30, 2005 press release, contributing to a stock price decline of 3.06% from $57.16 to $55.41. (TAC, ¶ ¶ 364, 581, 762.)

Finally, on April 1, 2005, several media reports were released regarding regulators' expansion of the scope of their investigation and the New York Attorney General's threat to indict AIG for alleged document removal and destruction in its Bermuda offices, contributing to a 8.05% decrease in stock price from $55.41 to $50.95. (TAC, ¶ ¶ 321, 366-70, 497-98, 576, 762.)

In addition to these six dates on which Lead Plaintiffs seek to show a decline in the price of AIG's securities attributable to the disclosure of AIG's alleged misstatements and omissions, Lead Plaintiffs also highlight AIG's May 31, 2005 10-K Form for the year 2004, in which AIG restated its earnings by more than $3,900,000,000.00 over the previous four years and disclosed approximately two dozen improper transactions in violation of GAAP. (TAC, ¶ ¶ 25, 587-610; Lead Pits' Mem. of Law, 2.)

E. Procedural History

The original Complaint in this matter was filed on October 15, 2004. On February 7, 2005, the Honorable Laura Taylor Swain granted Lead Plaintiffs' Motion for Consolidation and for appointment of Lead Plaintiffs and Lead Counsel. On April 19, 2005, Lead Plaintiffs filed a Consolidated Amended Class Action Complaint and on September 27, 2005, Lead Plaintiffs filed a Consolidated Second Amended Complaint.

Upon reassignment from Judge Swain, in Orders dated April 27, 2006, May 31, 2006, and March 2, 2007, the Honorable John J. Sprizzo granted Defendant Evan Greenberg's Motion to Sever, reserved judgment on Evan Greenberg's Motion to Dismiss, and denied the Motions to Dismiss of 22 other Defendants. On June 20, 2006, Lead Plaintiffs voluntarily dismissed Defendant Evan Greenberg and on May 6, 2008, Lead Plaintiffs entered into a Stipulation and Order of Dismissal with prejudice of Defendant Union Excess Reinsurance Company, Ltd. On December 15, 2006, Lead Plaintiffs filed the current 486-page Consolidated Third Amended Complaint. On October 3, 2008, Lead Plaintiffs moved for approval of their proposed settlement with Defendant PwC.

On January 12, 2009, the case was assigned to this Court. On February 25, 2009, Lead Plaintiffs moved for approval of their proposed settlement with Defendant Gen Re.

On October 2, 2009, the Court signed a Stipulation and Order of Voluntary Dismissal, without prejudice, of Defendants Maurice R. Greenberg, Howard I. Smith, Christian M. Milton, Michael J. Castelli, C.V. Starr & Co. Inc., and Starr International Company, Inc. On November 10, 2009, the Court granted Lead Plaintiffs' Motions of Voluntary Dismissal of Defendants Michael Murphy and Frank Hoenemeyer, and on January 21, 2010, the Court granted Lead Plaintiffs' Motion of Voluntary Dismissal of Defendant Richmond Insurance Company.

II. DISCUSSION

A. Legal Standard

" [A] district judge may certify a class only after making determinations that each of the Rule 23 requirements has been met." In re Initial Public Offerings Securities Litigation, 471 F.3d 24, 41 (2d Cir.2006). The " standard of proof applicable to evidence proffered to meet the requirements of Rule 23[is] a preponderance of the evidence." In re Flag Telecom Holdings, Ltd. Securities Litigation, 574 F.3d 29, 35 (2d Cir.2009) (quoting Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 202 (2d Cir.2008)). The Rule 23 requirements here include numerosity, commonality, typicality, and adequacy of representation under Rule 23(a)(1) through (4), and predominance and superiority under Rule 23(b)(3). (Lead Pits' Mem. of Law, 4.)

In making these determinations, the Court " resolves factual disputes relevant to each Rule 23 requirement" and must find that " whatever underlying facts are relevant to a particular Rule 23 requirement have been established" in order to satisfy the applicable legal standard for that requirement. In re Initial Public Offerings, 471 F.3d at 41. While merits issues may overlap with the requirements of Rule 23, " a district judge should not assess any aspect of the merit s unrelated to a Rule 23 requirement." Id. Furthermore, trial courts are provided broad discretion to determine whether or not to grant class certification because " the district court is often in the best position to assess the propriety of the class and has the ability ... to alter or modify the class, create subclasses, and decertify the class whenever warranted." In re Sumitomo Copper Litigation v. Credit Lyonnais Rouse, Ltd., 262 F.3d 134, 139 (2d Cir.2001).

B. Lead Plaintiffs' Motion for Class Certification

1. Standing to Bring the AIG Bond Claims

Before addressing the Rule 23 requirements for class certification, the Court first resolves the issue of standing for the AIG debt securities. Lead Plaintiffs seek to represent the class with respect to the § 11 claims in Counts One, Two, and Three, the § 15 claim in Count Four, and the § 10(b) and Rule 10b-5 claim in Count Five. (TAC, ¶ ¶ 1082-1128.) Of the five AIG bonds at issue, Count One concerns the 2.85% bonds, Count Two the Zero Coupon bonds, Count Three the 2.875% and 4.25% bonds, Count Four the Zero Coupon, 2.85%, 2.875%, and 4.25% bonds, and Count Five the Zero Coupon and 0.5% bonds. (TAC, ¶ ¶ 1082-1128.) AIG contends that Lead Plaintiffs do not have standing to bring claims related to Counts One through Four on behalf of the proposed class. (February 10, 2009 Tr., 77:8-78:17; February 27, 2009 Tr., 23:8-24).³

3 With regard to the § 10(b) and Rule 10b-5 claim in Count Five of the Third Amended Complaint, AIG asserts that Lead Plaintiffs cannot establish that AIG's Zero Coupon and 0.5% bonds traded in an efficient market, that the fraud-on-the-market presumption does not apply, and thus that common issues do not predominate over individual issues of reliance for the debt securities. (AIG Mem. of Law, 23-25.) Accordingly, the Court considers AIG's challenge to the § 10(b) and Rule 10b-5 bond claim on the basis of the predominance requirement in Section II.B.3.a.(2).(a).(ii), infra.

There is conflicting case law in the Second Circuit on whether a court may certify a class of purchasers of a security or fund that was not also purchased by the Lead Plaintiffs. However, the most recent cases, which have addressed this question as one of standing, rather than under the Rule 23 requirements, have found that where purported class representatives " bought only ... ordinary shares, as opposed to debt ... they lack standing to bring claims on behalf of those class members who purchased ... debt notwithstanding the allegations that the same general course of conduct allegedly engaged in by defendants caused injury to all putative class members." In re Parmalat Securities Litigation, 2008 WL 3895539, at *3 (S.D.N.Y. Aug.21, 2008); see also Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 2005 WL 2148919, *4 (S.D.N.Y. Sep.6, 2005) (" ... courts have held that a class action plaintiff does not have standing to bring claims on behalf of purchasers of different securities where those claims are based on different factual allegations and legal theories" ); In re Salomon Analyst Level 3 Litigation, 350 F.Supp.2d 477, 497 (S.D.N.Y.2004) (dismissing bondholder claims for lack of standing where " no named plaintiff is alleged to have purchased any [bonds]" and rejecting the holding in " a handful of cases" certifying classes where none of the named representatives purchased one or more of the securities at issue because those cases did not " address the issue of standing" and " under the specific circumstances of the case before them [,]" a plaintiff could be deemed adequate under Rule 23(a)(3)).

Lead Plaintiffs nonetheless contend that it " is unquestionably the case" that " Lead Plaintiff[s] allege[ ] a single pattern of fraud based on identical legal theories," and thus, they " need not have purchased every security purchased by any Class member in order to adequately represent the Class's interests." (Lead Pits' Mem. of Law, 14) (emphasis in original.) In addition to the fact that Lead Plaintiffs address their argument to adequate representation under Rule 23(a)(3), rather than to standing, it is not the case that Lead Plaintiffs allege a single pattern of fraud based solely on one legal theory. Private actions under §§ 10(b) and 11 are based on distinct theories of liability and require a Plaintiff to plead and prove different sets of facts. See Herman & MacLean v. Huddleston, 459 U.S. 375, 382, 103 S.Ct. 683, 74 L.Ed.2d 548 (1983) (noting that a § 11 claim is " limited in scope, ... places a relatively minimal burden on a plaintiff [,] ... must be brought by a purchaser of a registered security, must be based on misstatements or omissions in a registration statement, and can only be brought against certain parties" whereas a § 10(b) and Rule 10b-5 claim is, " [i]n contrast ... a ‘ catchall’ antifraud provision ... [that] requires a plaintiff to carry a heavier burden to establish a cause of action [,] ... can be brought by a purchaser or seller of ‘ any security’ against ‘ any person’ who has used ‘ any manipulative or deceptive device or contrivance’ in connection with the purchase or sale of a security [,][and] ... [m]ost significantly, ... [requires that] the defendant acted with scienter" ); (" ... Section 11 and Section 10(b) address different types of wrongdoing ..." ).

Accordingly, because Lead Plaintiffs' § 10(b) and Rule 10b-5 claims related to AIG stock are based upon different legal theories and would require proof of different facts than their § 11 claims related to AIG bonds, Lead Plaintiffs do not have standing to bring the § 11 bond claims of the proposed class by virtue of their § 10(b) and Rule 10b-5 claims.

Despite the fact that they did not purchase the Zero Coupon, 0.5%, 2.85%, or 2.875% bonds. Lead Plaintiffs also contend that they may represent the class with respect to all of these bonds because they purchased AIG's 4.25% Notes during the Class Period and thus have standing to bring the § 11 claims. (Lead Pits' Mem. of Law, 13.)

However, under § 11 of the Securities Exchange Act, 15 U.S.C. § 77k(a), if a security registration statement contains an untrue statement or omission of material fact, a plaintiff acquiring the security " may sue" the issuers of that registration statement " unless it is proved that at the time of such acquisition [it] knew of such untruth or omission." See also Mayer v. Oil Field Systems Corp., 803 F.2d 749, 755 (2nd Cir.1986) (" Nor may [Plaintiff] recover under § 11 ... of the 1933 Act unless the defendant misrepresented or omitted a material fact and the plaintiff had no knowledge of the untruth or omission." ); In re Livent, Inc. Noteholders Securities Litigation, 151 F.Supp.2d 371, 441 (S.D.N.Y.2001) (" [plaintiffs] may not recover under § 11 ... if they knew of the alleged untruth or omission at the time of purchase." ). Furthermore, knowledge of a misstatement or omission is " sufficient to defeat a § 11 ... claim; defendants need not demonstrate plaintiffs' actual knowledge of the truth." In re Livent, 151 F.Supp.2d at 441.

The Registration Statement and Prospectus pursuant to which Lead Plaintiffs purchased the 4.25% bonds, and which incorporated select financial data for the years 2000 through 2003 as well as AIG's 2003 10-K Form, were filed by AIG with the SEC on March 22, 2004 and became effective on April 20, 2004. (TAC, ¶ ¶ 1014-17.) Lead Plaintiffs purchased the 4.25% bonds on March 15, 2005, after they had moved to be appointed Lead Plaintiffs and after the initial Complaint was filed on October 15, 2004. (TAC, Ex. 1, at 17.) That initial Complaint alleged that AIG had filed false and misleading financial information in its SEC statements during the Class Period, including the 2000 and 2003 periods. (Compl. ¶ ¶ 28-31, 36-39; ¶ 45 (" These financial statements ... were false and misleading" ); ¶ 47 (" Due to accounting improprieties, [AIG] presented its financial results and statements in a manner which violated GAAP, including ... fundamental accounting principles." ))

Consequently, because Lead Plaintiffs were necessarily aware of the allegations in the original Complaint, including the false and misleading financial information in the Class Period financial statements, they cannot now plausibly contend that they were unaware that the Registration Statement and Prospectus contained false and misleading financial information at the time that they purchased AIG's 4.25% Notes on and after March 15, 2005. (TAC, Ex. 1, at 17.)

Attempting to circumvent this defense to their § 11 claims, Lead Plaintiffs contended on February 27, 2009 that In re Adelphia Communications Corp. Securities and Derivative Litigation, 2007 WL 2615928, *7 (S.D.N.Y. Sep.10, 2007), demonstrates that § 11 only bars recovery for " those untruths or omissions that the purchaser was aware of" and thus that Lead Plaintiffs should not be precluded from bringing suit on the bonds, given that the March 17, March 30, March 31, and April 1, 2005 disclosures occurred after Lead Plaintiffs purchased the 4.25% bonds on March 15, 2005. (February 27, 2009 Tr., 53:10-54:15.) The Court, however, finds In re Livent, Inc. Noteholders Securities Litigation, 151 F.Supp.2d 371 (S.D.N.Y.2001), and not In re Adelphia, to be apposite in the circumstances present here.

In In re Livent, plaintiffs brought § 11 claims based upon misstatements made in connection with a November 1997 Registration Statement and a December 1997 Prospectus. The Amended Complaint made clear that plaintiffs were aware of the misstatements in those documents before they purchased the debt securities because the Amended Complaint stated that they had relied on defendant's " publicly issued statements," including an " August 10, 1998 Press Release and its disclosure of ‘ serious irregularities in the Company's financial records' and the need to restate its financial results." 151 F.Supp.2d at 441. The Court in In re Livent then found that, even though plaintiffs had affirmatively alleged that they " did not know ... of the misstatements and omissions of material fact contained in the Registration Statement and Prospectus," this allegation was negated by their admitted reliance on the 1998 Press Release, which therefore warranted dismissal of their § 11 claims. Id.

In contrast to In re Livent, in In re Adelphia, the Defendants had argued " that because Plaintiffs knew some statements were false, they [could not] recover for damages caused by other statements held out as true when the securities were purchased." 2007 WL 2615928, at *7. The Court rejected this argument, noting that the language of the statute would " not support such an interpretation" and distinguished In re Livent by finding that the § 11 defense would not apply where " a purchaser purport[s] to rely on a different statement not yet disavowed." Id. (emphasis added.) Thus, the Court in In re Adelphia acknowledged that while purchasers may not rely on a Registration statement that they know to contain misstatements and then later sue under § 11, as the In re Livent Court had found, they may bring suit where they purchase bonds while alleging that they have relied on public statements that have not yet been disavowed, even if they are aware of other, distinct misstatements and omissions that might affect the price of the bonds.

Consequently, unlike In re Adelphia, the Lead Plaintiffs here knew that some or many of the statements contained in the Registration Statement were false when they relied on that Statement in purchasing the 4.25% bonds on March 15, 2005. Further, Lead Plaintiffs do not allege that in purchasing the 4.25% bonds they relied on other statements, not then known to be untrue. Lead Plaintiffs knew of the untruth of the 2000 and 2003 financial statements, which were incorporated into the 4.25% Registration Statement, at the time they purchased the 4.25% bonds. Thus, as in In re Livent, Lead Plaintiffs " rel[ied] on financial statements already known to be false" and nowhere " purport[ed] to rely on a different statement not yet disavowed." In re Adelphia, 2007 WL 2615928, at *7 (stating and distinguishing the facts in In re Livent ).

Further, a claim for violation of § 15 of the Securities Act " imposes derivative liability on persons who ‘ control’ those liable under §§ 11 and 12" and " [t]hus, if there is no underlying § 11 or § 12 liability, there can be no § 15 liability." Milman v. Box Hill Systems Corp., 72 F.Supp.2d 220, 227 n. 8 (S.D.N.Y.1999) (citing SEC v. First Jersey Secs., 101 F.3d 1450, 1472 (2d Cir.1996); 15 U.S.C. § 77 o ). Therefore, because they lack standing for their § 11 bond claims, Lead Plaintiffs also lack standing for the derivative § 15 bond claim.

While Lead Plaintiffs do not have standing to bring the §§ 11 and 15 claims, the Court finds that Lead Plaintiffs have standing to bring their § 10(b) and Rule 10b-5 bond claim under Count Five of the Third Amended Complaint, given that the claim is brought under the same legal theory as the remaining § 10(b) and Rule 10b claims, and relies on many of the same material misstatements from AIG's public financial statements. ( See TAC, ¶ ¶ 1125-28.)

Accordingly, because Lead Plaintiffs lack standing to bring suit under §§ 11 and 15, the Court will not certify the class for the claims set forth in Counts One through Four.

2. Rule 23(a)

a. Rule 23(a)(1): Numerosity

" The numerosity requirement in Rule 23(a)(1) does not mandate that joinder of all parties be impossible-only that the difficulty or inconvenience of joining all members of the class make use of the class action appropriate." Central States Southeast and Southwest Areas Health and Welfare Fund v. Merck-Medco Managed Care, L.L.C., 504 F.3d 229, 244-45 (2d Cir.2007). Here, Defendants have not challenged the numerosity requirement, and Lead Plaintiffs assert, and Defendants do not contest, that the proposed class contains over two million shareholders. (February 10, 2009 Tr., 12:13-18.)

Accordingly, the Court finds that the numerosity requirement of Rule 23(a)(1) has been met.

b. Rule 23(a)(2): Commonality

" The commonality requirement is met if plaintiffs' grievances share a common question of law or of fact." Central States Southeast and Southwest Areas Health and Welfare Fund v. Merck-Medco Managed Care, LLC, 504 F.3d 229, 245 (2d Cir.2007) (internal quotation omitted).

Here, Defendants have not challenged the commonality requirement, as there are many common questions of law and fact, including, but certainly not limited to, whether Defendants made particular material misstatements and omissions in their publicly filed documents and whether Defendants acted with the requisite level of scienter for § 10(b) and Rule 10b-5 ⁴ liability. Similarly, there are common questions of law for Lead Plaintiffs' § 20(a) ⁵ and § 20A ⁶ claims as well, including whether the § 20(a) Defendants controlled AIG and whether the § 20A Defendants, Broad and Graf, possessed material, nonpublic information at the time they allegedly traded in AIG securities.

4 In order to prove a § 10(b) and Rule 10b-5 claim, a plaintiff must show: " (1) a material misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation." Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, 552 U.S. 148, 157, 128 S.Ct. 761, 169 L.Ed.2d 627 (2008).

5 In order to establish a prima facie case of liability under § 20(a), " a plaintiff must show a primary violation by the controlled person and control of the primary violator by the targeted defendant, and show that the controlling person was in some meaningful sense a culpable participant in the fraud perpetrated by the controlled person." S.E.C. v. First Jersey Securities, Inc., 101 F.3d 1450, 1472 (2d Cir.1996).

6 To establish a § 20A claim, a plaintiff must show: " (1) a predicate violation of the Exchange Act or its rules and regulations; (2) that the defendant traded the security at issue contemporaneously with the plaintiff; and (3) that the defendant was in possession of material, nonpublic information at the time of the trade." In re Monster Worldwide, Inc. Securities Litigation, 549 F.Supp.2d 578, 584 (S.D.N.Y.2008).

Accordingly, the Court finds that the commonality requirement of Rule 23(a)(2) has been met.

c. Rule 23(a)(3): Typicality

" To establish typicality under Rule 23(a)(3), the party seeking certification must show that each class member's claim arises from the same course of events and each class member makes similar legal arguments to prove the defendant's liability." In re Flag Telecom Holdings, Ltd. Securities Litigation, 574 F.3d 29, 35 (2d Cir.2009) (quoting Robidoux v. Celani, 987 F.2d 931, 936 (2d Cir.1993)).

(1). Unique Defense to Reliance Based on Post-Disclosure Purchases

Defendant AIG contends that Lead Plaintiffs cannot satisfy the typicality requirement of Rule 23(a)(3) because they purchased additional AIG shares following the October 2004 disclosures and again after the close of the Class Period. (AIG Mem. of Law, 11-12.)

While " there is no per se rule to this effect," courts have sometimes found that " a person that increases his holdings in a security after revelation of an alleged fraud involving that security is subject to a unique defense that precludes him from serving as a class representative." Rocco v. Nam Tai Electronics, Inc., 245 F.R.D. 131, 136 (S.D.N.Y.2007) (rejecting as atypical a proposed class representative who had purchased defendant's stock after the public disclosure of some, but not all, of defendant's fraudulent statements and omissions). However, courts in this jurisdiction have differed over whether a class representative's purchase of shares after the disclosure of corrective information will create a unique defense to which it is subject. See, e.g., In re Monster Worldwide, Inc. Sec. Litig., 251 F.R.D. 132, 135 (S.D.N.Y.2008) (" the fact that a putative class representative purchased additional shares in reliance on the integrity of the market after the disclosure of corrective information has no bearing on whether or not [the representative] relied on the integrity of the market ... before the information at issue was corrected or changed" ) (quoting In re Salomon Analyst Metromedia, 236 F.R.D. 208, 216 (S.D.N.Y.2006), rev'd on other grounds, 544 F.3d 474 (2d Cir.2008)).

In Rocco, the Court rejected as atypical a proposed class representative who had purchased the defendant's stock not only prior to, but also on four occasions after the disclosure of one of the defendant's two frauds. Rocco, 245 F.R.D. at 133-34 & n. 2. Critically, the Court in Rocco found " particularly harmful to his efforts for certification as class representative" that he had " bought [defendant's] stock despite information he had about [defendant's] allegedly manipulated finances" under the second fraud, which had not yet been publicly corrected. Id.

Here, Lead Plaintiffs purchased thousands of AIG shares on multiple occasions prior to any disclosures by AIG, including during each year from 1999 to 2004, as well as after the October 2004 disclosures and again after the March 2005 disclosures. (Leffell Decl., Ex. 5, Lead Plts' Amend. Resp. to Interrogatories, 6-10, 12-18, 21-29.) Unlike the facts in Rocco, at the times Lead Plaintiffs made their post-disclosure purchases, (Leffell Decl., Ex. 5, Lead Plts' Amend. Resp. to Interrogatories, 11, 17-18, 27-29), there is no evidence that they were aware of any ongoing, publicly undisclosed fraud or that they purchased the stock with anything other than public information. Thus, although Lead Plaintiffs purchased AIG stock after the October 14 and 15, 2004 disclosures and again after the March and April, 2005 disclosures, they could do so in reliance on the market because the price of AIG stock had declined to incorporate the latest information in those disclosures and, unlike Rocco, there is no evidence that Lead Plaintiffs were aware of further, undisclosed frauds. Accordingly, the Court rejects Defendant AIG's argument that Lead Plaintiffs are atypical due to unique defenses.

(2). Class Conflicts

Defendant AIG further argues that Lead Plaintiffs should not be certified as the class representatives or, alternatively, that the class should be divided into subclasses because the Lead Plaintiffs' interests in maximizing their own recovery may be adverse to or in conflict with two categories of class members. (AIG Mem. of Law, 13-18.)

In In re Seagate Technology II Securities Litigation, 843 F.Supp. 1341 (N.D.Cal.1994), a district court found that there were potential class conflicts between a group of class members who sold a security on a particular day on which the price of the security fell and those who purchased the security on the same day. In particular, the Court noted that those who purchased on that day would have an incentive to present evidence showing that the amount of inflation still present in the price from undisclosed frauds was as sizeable as possible, while those selling on that day would seek to demonstrate that most or all of the inflation had already leaked out of the price, through previous disclosures. Id. at 1359. Where there were multiple partial disclosures of the misstatements or omissions, the Seagate Court suggested that the severity and number of such conflicts were likely to increase, given that on each day that a partial disclosure occurred, the overall amount of price inflation would decrease, thus separating the interests of those who sold on that day from those who held over or purchased on that day. Id. at 1364-65.

Here, AIG contends that the Lead Plaintiffs' interests are divergent from some other members of the class because they purchased stock in AIG after the October 14 and 15, 2004 disclosure of the bid-rigging and contingent commission fraud, while other class members purchased their AIG stock prior to these disclosures and sold their stock prior to the March and April 2005 disclosures of the various accounting misstatements and omissions. Accordingly, those who primarily held over the October 2004 disclosure dates would have an interest in maximizing the price inflation from the bid-rigging and contingent commission fraud, while Lead Plaintiffs and others who held over the March and April 2005 dates would have an interest in maximizing the price inflation from the accounting frauds. AIG argues, therefore, that Lead Plaintiffs may be able to manipulate their discovery and presentation of evidence to support a larger price effect arising from the accounting frauds. Furthermore, according to AIG, Lead Plaintiffs may be able to demonstrate a smaller price effect arising from the bid-rigging and contingent commission fraud, or even disprove these frauds altogether, thus harming the interests of the class members that only held stock over the October 2004 disclosure dates. (AIG Mem. of Law, 13-16).

Defendant AIG also contends that the Lead Plaintiffs' interests conflict with two groups of class members who traded their shares of HSB Group or American General for AIG shares during AIG's acquisition of each company. (AIG Mem. of Law, 16-18; TAC ¶ ¶ 1029-48). During the mergers, a stock swap took place in which AIG shares were traded to HSB and American General stockholders and HSB and American General shares to AIG shareholders based upon the relative price of the shares. Consequently, to the extent that AIG shares were overpriced due to the alleged frauds, those who had a greater percentage of HSB and American General shares than AIG shares at the time of the mergers received fewer AIG shares in return for their shares, while those with a greater percentage of AIG shares at the time of the mergers received a greater number of HSB and American General shares than they otherwise would have received had the misstatements and omissions not occurred. Given that Lead Plaintiffs held a greater percentage of AIG shares than HSB shares and a lesser percentage of AIG shares than American General shares at the times of the respective mergers, AIG argues that Lead Plaintiffs will have an interest in minimizing the amount of price inflation at the time of the HSB merger and maximizing the amount of price inflation at the time of the American General merger, thus prejudicing those in the reverse positions. (AIG Mem. of Law, 16-18).

While Seagate has been widely cited, courts in this jurisdiction have nearly universally declined to follow its holding or reasoning. See, e.g., In re Alstom SA Securities Litigation, 253 F.R.D. 266, 277 (S.D.N.Y.2008) (" Courts have ... repeatedly recognized that putative intra-class conflicts relating to the time at which particular class members purchased their securities, and which could potentially motivate different class members to argue that the securities were relatively more or less inflated at different time periods, relate to damages and do not warrant denial of class certification" ); In re Flag Telecom Holdings, Ltd. Securities Litigation, 245 F.R.D. 147, (S.D.N.Y.2007) (declining to find an inter-class conflict where " the decline in value in [the defendant's] stock may have been caused by both the alleged fraud relating to the reciprocal transactions and the alleged misstatements relating to pre-sales found in the Registration Statement" because " [u]nder the doctrine of proximate causation, the harm may have been caused by either or both of these alleged acts of deception." ) (emphasis in original), rev'd on other grounds, 574 F.3d 29 (2d Cir.2009); Cromer Finance Ltd. v. Berger, 205 F.R.D. 113, 127 (S.D.N.Y.2001) (" [T]he fact that conflicts among class members may arise at the settlement or damages stage of the litigation does not require the denial of class certification or the disqualification of these representatives" and " [i]f they are needed later, the court has substantial discretion to create sub-classes." ); In re Oxford Health Plans, Inc., 191 F.R.D. 369, 377-78 (S.D.N.Y.2000) (finding that " the holding in Seagate has been widely discredited" and rejecting " an assumed Seller/Purchase conflict" ); In re Gaming Lottery Securities Litigation, 58 F.Supp.2d 62, 69-70 (S.D.N.Y.1999) (finding that " [t]he position of the Seagate court is most decidedly the minority position" ).

AIG seeks to distinguish these cases on two grounds. First, AIG contends that none of these cases involved a class asserting multiple fraudulent schemes with distinct time periods, and that instead, each involved a " single common scheme, which all class members had an interest in proving." (AIG Mem. of Law, 16 n. 16.) In addition, AIG asserts that these cases relied upon the view that the amount of price inflation was merely an issue of damages, whereas after the Supreme Court's decision in Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336, 125 S.Ct. 1627, 161 L.Ed.2d 577 (2005), which explicated the importance of loss causation, the conflict " bears on liability, not just damages." (AIG Mem. of Law, 16 n. 16.)

The decision in In re Flag Telecom, however, rejected claims of an intra-class conflict in a post- Dura case that involved distinct frauds, both chronologically, and substantively. 245 F.R.D. 147 (S.D.N.Y.2007). In In re Flag Telecom, the defendant corporation was alleged to have released both a false Prospectus and Registration Statement exaggerating the extent of product pre-sales to mislead investors about the demand for the company's product, and to have engaged in ongoing false and misleading accounting statements about the state of the company's financial conditions. Id. at 159. As here, the defendants in In re Flag Telecom argued that the plaintiff's demonstration that the disclosure of one of these frauds had led to a price decrease, thus establishing loss causation, might negate the loss causation element of the other fraud. As the Court in In re Flag Telecom pointed out, however, " the decline in value of Flag stock may have been caused by both [of] the alleged fraud[s] ...," which would therefore mean that " the sets of claims are not antagonistic to each other because proof of one does not negate an essential element of the other." Id. at 159-60 (emphasis in original). Accordingly, the Court acknowledged that a hypothetical conflict between the elements of two distinct frauds was insufficient at the class certification stage to find lead plaintiff atypical.

The Court finds no facts here that persuade it to reach a conclusion different from that of the Court in In re Flag Telecom. In In re Flag Telecom, the Court had acknowledged that while the prospectus and registration fraud claims were brought under §§ 11, 12(a), and 15 of the Securities Act of 1933, the misleading accounting statement claims were brought under §§ 10(b) and 20(a) of the Securities Exchange Act of 1934. Id. at 151, 159. Additionally, whereas in In re Flag Telecom, the first fraud implicated the extent of demand for the defendant's products rather than its accounting practices and net financial health, here both the bid-rigging/contingent commissions and the financial accounting frauds ultimately implicated the accuracy and truthfulness of AIG's financial statements. ( See TAC, ¶ ¶ 245-46 (alleging that as a result of the contingent commissions AIG's SEC filings during the Class Period contained false and materially misleading statements regarding its premium revenues); ¶ 250 (alleging that as a result of the bid-rigging, AIG's public statements during the Class Period were false and materially misleading with respect to its premium revenues); ¶ ¶ 587-88 (summarizing the May 31, 2005 restatement correcting previous misstatements and omissions in AIG's financial statements)). Further, while the prospectus fraud in In re Flag Telecom completely preceded the accounting fraud, AIG's did-bid-rigging/contingent commissions fraud and the accounting frauds overlapped from at least the third quarter of 1999 to the fourth quarter of 2004. ( See TAC, ¶ ¶ 815, 826, 831, 836, 840, 851, 857, 862, 868, 878, 883, 888, 894, 906, 913, 919, 926, 936, 943, 950, 957, 960.)

Nevertheless, despite the fact that the frauds in In re Flag Telecom were based upon different causes of action, involved the dissemination of distinct categories of material misinformation, and were not contemporaneous, the Court found no class conflict. A fortiori here, where both of the frauds are brought under the same causes of action, where there is substantial overlap in the timing of the frauds, and where both frauds resulted from misinformation disseminated in financial statements, the Court does not find a class conflict. See also, e.g., In re Alstom SA Securities Litigation, 253 F.R.D. 266 (S.D.N.Y.2008) (finding no class conflict where plaintiffs alleged that defendants had " artificially inflated the price of Alstom's securities by making materially false and misleading statements relating to" both " (1) the demand and revenue earned from Alstom's sale of cruise ships ... and (2) the profitability of [subsidiary] ATI" even where the two frauds did not occur simultaneously and involved different segments of Alstom's business.)

As to the purported merger conflict, AIG does not cite to a case from within this jurisdiction where a court has divided a class on the basis of a purported conflict resulting from a stock-for-stock merger. (AIG Mem. of Law, 18.) One Court within this Circuit has found that a merger conflict like the ones alleged here " generally relates to the measure of damages and is otherwise insufficient to defeat consolidation" because such a conflict " is speculative and hypothetical." In re Olsten Corp. Securities Litigation, 3 F.Supp.2d 286, 291-92, 296 (E.D.N.Y.1998) (rejecting conflict argument against adequacy of prospective class representative who " was a holder of Quantum shares prior to the Olsten-Quantum merger [,]" and thus " appear[ed] to have actually benefitted from the fraud against the shareholders of Olsten Corporation" due to Quantum's inflated price at the time of the merger). The Court agrees and finds the purported merger conflicts here to be no less hypothetical and speculative than the alleged conflict arising from the timing of the so-called accounting and bid-rigging/contingent commission frauds. Accordingly, the Court also rejects AIG's claim of a class conflict on the basis of the HBS and the American General mergers.⁷

7 To the extent AIG makes the same conflict arguments with regard to the Rule 23 requirement of adequate representation, that argument is also rejected.

d. Rule 23(a)(4): Adequacy of Representation

" Adequacy entails inquiry as to whether: (1) plaintiff's interests are antagonistic to the interest of other members of the class and (2) plaintiff's attorneys are qualified, experienced and able to conduct the litigation." In re Flag Telecom Holdings, Ltd. Securities Litigation, 574 F.3d 29, 35 (2d Cir.2009) (quoting Baffa v. Donaldson, Lufkin & Jenrette Sec. Corp., 222 F.3d 52, 60 (2d Cir.2000)). " The focus is on uncovering conflicts of interest between named parties and the class they seek to represent. In order to defeat a motion for certification, however, the conflict must be fundamental." Id. (internal quotations omitted).

Defendant AIG contends that the Court should deny certification because Lead Plaintiffs' credibility is subject to attack due to their alleged failure to comply with discovery obligations. (AIG Mem. of Law, 18-19.) However, in light of the absence of any evidence that relevant documents were actually destroyed or withheld by Lead Plaintiffs, the Court is satisfied that Lead Plaintiffs have adequate measures in place to preserve relevant documents. ( See February 10, 2009 Tr., 91:1-10) (AIG's counsel, conceding that there is no evidence that anything was destroyed).

Additionally, the Court has observed the conduct of Lead Plaintiffs' counsel in this action, and has also reviewed the attached firm resumes, indicating extensive class action experience, (Ratzman Decl. Exs. 11, 12), and finds that Labaton Sucharow and Hahn Loeser will fairly and adequately represent the interests of the class.

3. Rule 23(b)(3) ⁸

8 In addition to the four requirements of Rule 23(a), a party seeking class certification must satisfy one of the three categories of Rule 23(b). Here, certification is sought pursuant to Rule 23(b)(3), which requires that:

a. Predominance

To satisfy Rule 23(b)(3), the Court must find that " questions of law or fact common to class members predominate over any questions affecting only individual members." The predominance requirement " tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation," and is " far more demanding" than the commonality requirement under Rule 23(a). Moore v. PaineWebber, Inc., 306 F.3d 1247, 1252 (2d Cir.2002) (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)). The predominance requirement is met " if the plaintiff can establish that the issues in the class action that are subject to generalized proof, and thus are applicable to the class as a whole, ... predominate over those issues that are subject only to individualized proof." Cordes & Co. Financial Services, Inc. v. A.G. Edwards & Sons, Inc., 502 F.3d 91, 108-09 (2d Cir.2007).

(1) Common Course of Conduct

AIG argues, separate and apart from its allegations of class conflicts under the typicality and adequate representation requirements, that Lead Plaintiffs' claims do not satisfy the predominance requirement because " the factual issues with respect to class members whose claims are limited to non-disclosure of contingent commission arrangements have little, if anything, in common with those raised by class members whose claims are limited to the alleged accounting fraud." (AIG's Mem. of Law, 20.) Instead, AIG asserts, common questions " barely exist" and factual questions of (1) the misstatements and omissions, (2) scienter, and (3) loss causation are different for the two distinct groups. (AIG's Mem. of Law, 20.) AIG also asserts that predominance cannot be met because there is no " common scheme" or " common course of conduct" between the bid-rigging and contingent commission fraud and the remaining accounting frauds, including the Gen Re reinsurance transaction. (February 10, 2009 Tr., 66:4-25.)

However, the Court again finds the decision in In re Flag Telecom, 245 F.R.D. 147 (S.D.N.Y.2007), rev'd on other grounds, 574 F.3d 29 (2d Cir.2009), persuasive. As explained, supra Section II.B.2.C.(2), the court in In re Flag Telecom found no class conflict despite the fact that the two frauds there were divided among different causes of action, involved the dissemination of distinct categories of material misinformation, and were not contemporaneous with each other. Similarly, with regard to the issue of whether common questions predominated over individual ones, the Court in In re Flag Telecom specifically noted that all class members, regardless of whether they had claims from only one of the two frauds or both, could rely on the fraud-on-the-market presumption, and that the " proper determination of individual damages [could] be determined at trial through the use of expert witnesses." Id. at 171-72. The same holds true here. See also, e.g., In re Alstom SA Securities Litigation, 253 F.R.D. 266 (S.D.N.Y.2008) (finding " questions of law and fact common to the Proposed Class predominate over any questions affecting only individual members" where plaintiffs alleged that defendants had " artificially inflated the price of Alstom's securities by making materially false and misleading statements relating to" both " (1) the demand and revenue earned from Alstom's sale of cruise ships ... and (2) the profitability of [subsidiary] ATI [,]" even where the two frauds did not occur simultaneously and involved entirely different segments of Alstom's business.)

Here, Lead Plaintiffs allege a common course of conduct by Defendants that was directed at overstating AIG's financial results, through the creation of sham transactions, mislabeling, redirecting, and directly misstating and omitting financial information on AIG's quarterly and annual financial statements. By misstating the contingent commissions as legitimate business expenses, and misstating the resulting payments from steered clients as legitimate premium revenue, AIG materially overstated the financial condition of its insurance business and misled investors. As counsel for Lead Plaintiffs argued:

as to the contingent commission/bid-rigging fraud, that was also a financial fraud. That led to an artificial appearance of greater premium income, which then flowed through the various financing statements ... That, in turn, made the company look rosier than it was. That, in turn, inflated the stock for an independent series of reasons than what happened with the accounting manipulations. [But], [t]hey both inflated the stock.

(February 27, 2009 Tr., 42:22-43:6; see also TAC, ¶ ¶ 245-46 (alleging that as a result of the contingent commissions AIG's SEC filings during the Class Period contained false and materially misleading statements regarding its premium revenues); ¶ 250 (alleging that as a result of the bid-rigging, AIG's public statements during the Class Period were false and materially misleading with respect to its premium revenues)).

Accordingly, the Court finds, by a preponderance of the evidence, that there was a common course of conduct permeating the various frauds alleged in the Third Amended Complaint.

(2). Fraud on the Market

(a). The Elements of Fraud on the Market

The Supreme Court's decision in Basic v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988), held that upon a showing of " public material misrepresentations," id. at 247, 108 S.Ct. 978, plaintiffs bringing securities class action claims may rely upon a " presumption, created by the fraud-on-the-market theory and subject to rebuttal by [defendant], that persons who had traded [defendant's] shares had done so in reliance on the integrity of the price set by the market." Id. at 245, 108 S.Ct. 978. Although the Basic Court found that the fraud-on-the-market presumption was consistent with the purposes of the securities laws, as well as " common sense and probability", id. at 245-47, 108 S.Ct. 978, the practical impetus behind the rule arose largely from the fact that " [r]equiring proof of individualized reliance from each member of [a] proposed plaintiff class effectively would have prevented respondents from proceeding with a class action, since individual issues then would have overwhelmed the common ones," id. at 242, 108 S.Ct. 978.

Accordingly, under the predominance requirement of Rule 23(b)(3), in order to show that common issues of reliance predominate, the Lead Plaintiffs will need to meet the requirements of the fraud-on-the-market presumption.

Under the Supreme Court's decision in Basic, [t]he fraud on the market theory is based on the hypothesis that, in an open and developed securities market, the price of a company's stock is determined by the available material information regarding the company and its business .... Misleading statements will therefore defraud purchasers of stock even if the purchasers do not directly rely on the misstatements ... The causal connection between the defendants' fraud and the plaintiffs' purchase of stock in such a case is not less significant than in a case of direct reliance on misrepresentations.

485 U.S at 241-42, 108 S.Ct. 978 (quoting Peil v. Speiser, 806 F.2d 1154, 1160-61 (3d Cir.1986)). According to this hypothesis then, " [a]n investor who buys or sells stock at the price set by the market does so in reliance on the integrity of that price" and " an investor's reliance on any public material misrepresentations, therefore, may be presumed for purposes of a § 10(b) and Rule 10b-5 action." Id. at 247, 108 S.Ct. 978.

The Court in In re Salomon, 544 F.3d 474 (2d Cir.2008) stated:

The only question raised by this appeal is whether the district court properly determined that the Rule 23(b)(3) predominance requirement was met. The Rule 23(b)(3) predominance requirement tests whether a proposed class is sufficiently cohesive to warrant adjudication by representation. To meet the requirement a plaintiff must show that those issues in the proposed action that are subject to generalized proof outweigh those issues that are subject to individualized proof. In this case, the question of whether the predominance requirement is met largely turns on whether and how the Basic fraud-on-the-market presumption applies ...

The Basic Court thereby set forth a test of general applicability that where a defendant has (1) publicly made (2) a material misrepresentation (3) about stock traded on an impersonal, well-developed (i.e. efficient) market, investors' reliance on those misrepresentations may be presumed. This is all that is needed to warrant the presumption.

Id. at 480-81 (internal quotations omitted).⁹

9 In footnote 4 of its opinion, the Salomon Court also noted the Basic requirement " that the plaintiff traded the shares between the time the misrepresentations were made and the time the truth was revealed." 544 F.3d at 481 n. 4. Consistent with the fourth requirement of Basic, set out in footnote 4 of Salomon, Lead Plaintiffs conceded at the August 14, 2009 hearing that, given the absence of evidence or claims that information about the alleged frauds had leaked out prior to disclosure dates, they would accede to AIG's argument that in-and-out traders be excluded from the class. (August 14, 2009 Tr., 237:1-238:7.) Further, even had Lead Plaintiffs not made this concession, the Second Circuit's recent decision in In re Flag Telecom found that to the extent in-and-out traders cannot " establish loss causation ... they must therefore be excluded from the certified class." 574 F.3d at 41.

(i). Public Statement by Gen Re Defendants ¹⁰

10 The Gen re Defendants are General Reinsurance Corp., a wholly-owned subsidiary of Berkshire Hathaway, (TAC, ¶ 124); Ronald E. Ferguson, Gen Re's Chairman and Chief Executive Officer until October 2001 and June 2002, respectively, (TAC, ¶ 124); John B. Houldsworth, Chief Executive Officer of Cologne Re Dublin, a subsidiary of Gen Re, (TAC, ¶ 128); and Richard Napier, Senior Vice President at Gen Re's Stamford, Connecticut office until June 8, 2005. (TAC, ¶ 133.)

Here, the Gen Re Defendants have argued, and the Court agrees, that Lead Plaintiffs " cannot establish class-wide reliance [against the Gen Re Defendants] for the ... reason that they cannot show that the market was relying on any statement or action of the Gen re Defendants." (Gen Re Mem. of Law, 21-23.) As the Court in In Re Salomon made clear, although " there is a private right of action under Section 10(b) against entities other than issuers," such an action must " satisf[y] each of the elements or preconditions for liability" such that the fraud-on-the-market presumption does not apply to a defendant whose " deceptive acts were not communicated to the public, as required by Basic." 544 F.3d at 481 (quoting Stoneridge Investment Partners, LLC v. Scientific-Atlanta, Inc., 552 U.S. 148, 159, 128 S.Ct. 761, 169 L.Ed.2d 627 (2008)).

Here, the Third Amended Complaint nowhere alleges that the Gen Re Defendants made a public misstatement ¹¹ with regard to AIG, ( see generally, TAC, ¶ ¶ 124-137, 314-464, 1139-48, 1170-77) (pleading allegations with regard to the Gen Re Defendants), nor do Lead Plaintiffs provide evidence of any such misstatement in their submissions in support of the Motion for Class Certification.

11 An omission may give rise to the fraud-on-the-market presumption only where a defendant has a duty to disclose. Stoneridge, 552 U.S at 148, 128 S.Ct. 761 (citing Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 153-54, 92 S.Ct. 1456, 31 L.Ed.2d 741 (1972)). Lead Plaintiffs do not allege or show any such duty on behalf of the Gen Re Defendants.

Because Lead Plaintiffs have not established or even pled that the Gen Re Defendants made any public misstatement or omission with regard to AIG, the fraud-on-the-market presumption does not apply to claims against these Defendants, and individual issues of reliance predominate over common issues for the claims against the Gen Re Defendants regarding AIG stock. Accordingly, the Court does not certify the class of claims against the Gen Re Defendants. See In re Salomon, 544 F.3d at 485 (" In re IPO now requires a district court to make a ‘ definitive assessment’ that the Rule 23(b)(3) predominance requirement has been met." ); In re Flag Telecom, 574 F.3d at 39 (" In re IPO makes clear that courts may resolve contested factual issues where necessary to decide on class certification, and when a claim cannot succeed as a matter of law, the Court should not certify a class on that issue." ) (quoting McLaughlin v. Am. Tobacco Co., 522 F.3d 215, 228 (2d Cir.2008)).

(ii). Efficient Bond Market

Although the Court finds in Section II.B.1, supra, that Lead Plaintiffs do not have standing to represent the class on the §§ 11 and 15 bond claims, because Lead Plaintiffs have standing for the § 10(b) and Rule 10b-5 bond claim, the Court must examine whether the members of the class can rely on the fraud-on-the-market presumption in bringing claims against AIG related to price decreases in its Zero Coupon and 0.5% debt securities under Count Five of the Third Amended Complaint.

One of the requirements of the fraud-on-the-market presumption is " an open and developed securities market" in which the " price of a company's stock is determined by the available material information regarding the company and its business." Basic, 485 U.S. at 241, 108 S.Ct. 978. While courts in this jurisdiction have routinely applied the five-factor test set forth in Cammer v. Bloom, 711 F.Supp. 1264 (D.N.J.1989), in " considering the efficiency of equity markets ... [the Second Circuit] has not adopted a test for the market efficiency of stocks or bonds [.]" Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 204 n. 11 (2d Cir.2008).¹² Nevertheless, in cases involving bonds rather than common stock, the Second Circuit has found that a district court may nevertheless " properly use ... the Cammer factors as an analytical tool." Id. at 210. Accordingly, the Court will examine the efficiency of the AIG bond market using the Cammer factors as a framework, but will also consider additional evidence to the extent it is relevant to the efficiency of the market for the AIG bonds.

12 Their prescience in not doing so may be beginning to be recognized. See Paul Krugman's School for Scoundrels, in which he reviews Justin Fox's book, The Myth of the Rational Market; A History of Risk, Reward, and Delusion on Wall Street, August 6, 2009 column in The New York Times (explaining that Fox's book " tells the story of the professors who enabled [the errors and] abuses under the banner of the financial theory known as the efficient-market hypothesis" and that the resulting " complex financial strategies ... played a crucial role in the catastrophe that has now overtaken the world economy" ).

The Cammer factors tending to establish an efficient market include: " (1) a large weekly trading volume; (2) a significant number of securities analysts following and reporting on a company's stock; (3) the presence of market makers and arbitrageurs who are able to react swiftly to company news and drive the stock price; (4) the eligibility of the company to file an S-3 Registration Statement ¹³ for its public offerings; and (5) empirical facts showing a cause and effect relationship between unexpected corporate events or financial releases and an immediate response in the stock price." In re SCOR Holding (Switzerland) AG Litigation, 537 F.Supp.2d 556, 574 (S.D.N.Y.2008) (citing Cammer, 711 F.Supp. at 1286-87).

13 An S-3 Registration Statement is a SEC registration form that may be used in a public offering by a company that has " been filing reports under the Exchange Act for at least thirty-six months and either has outstanding $150 million of voting stock held by nonaffiliates or $100 million of such stock outstanding coupled with an annual trading volume of three million shares per year." Cammer, 711 F.Supp. at 1271 n. 5.

Lead Plaintiffs rely on the analysis of their expert, Dr. John D. Finnerty, Professor of Finance at the Graduate School of Business Administration of Fordham University and managing principal of Finnerty Economic Consulting. (August 13, 2009 Tr., 7:13-18.) Dr. Finnerty analyzed AIG's 0.5% and Zero Coupon bonds, (Finnerty Decl., ¶ ¶ 56-57; August 13, 2009 Tr., 70:14-19), and opined that for those two bonds " the markets for Debt Securities were efficient during the Class Period." ¹⁴ (Finnerty Decl., ¶ 80; August 13, 2009 Tr., 70:20-71:23, 73:7-16.)

14 It is clear from the substance of the expert declarations, as well as their testimony, that their evidence and arguments are largely directed at the Cammer factors. (August 13, 2009 Tr., 69:19-70:12) (listing the Cammer factors).

For the first Cammer factor, Dr. Finnerty contended that both the 0.5% and Zero Coupon bonds were actively traded during the period when the Trade Reporting and Compliance Engine (" TRACE" ),¹⁵ was in place. " Trades of the 0.5% Notes were reported [o]n TRACE beginning March 3, 2003" while " [t]rades of the Zero Coupon Debentures were reported beginning July 1, 2002." (Finnerty Decl., ¶ 60.) During that period, the 0.5% bonds averaged 239 trades per year and .95 trades per trading day, at an average par value ¹⁶ of at least $606,647.00. (Finnerty Decl., ¶ ¶ 62-63, Ex. G.) The Zero Coupon bonds averaged 1,506 trades per year and 5.98 trades per trading day, at an average par value of at least $3,023,783.00. (Finnerty Decl., ¶ ¶ 62-63, Ex. G.) According to Dr. Finnerty, this equates to an average daily dollar trading value of $576,546.00 for the 0.5% bonds and $18,072,978.00 for the Zero Coupon bonds. (Finnerty Decl., ¶ ¶ 63, 65.)

15 According to Dr. Finnerty, TRACE " is a system of reporting that the NASD and FINRA introduced. It replaced the fixed income pricing system ... [and] was designed to enhance the degree of transparency in the bond market by requiring relatively prompt reporting of trades involving corporate and government debt securities." (August 13, 2009 Tr., 72:1-6.)

16 " When a bond or other debt instrument trades at face value, it is said to trade ‘ at par.’ " 6A Fletcher Cyclopedia of the Law of Corporations, § 2674.

Dr. Finnerty then compared the average trades per trading day of the AIG bonds to the " average number of trades per trading day of bonds whose transactions were reported [o]n TRACE and were issued prior to the start of the Class Period and were scheduled to mature after the Class Period." (Finnerty Decl., ¶ 63.) Among such bonds the median number of trades per trading day was. 47, from which Finnerty concluded that the AIG bonds had a " relatively high volume of trading." (Finnerty Decl., ¶ 63, Ex. H.) Dr. Finnerty also found that the average daily dollar trading value for the 0.5% bonds was greater than 70% of those bonds whose transactions were reported in TRACE and which were outstanding during the whole Class Period, and that the average daily dollar value of trading for Zero Coupon bonds was greater than 90% of such bonds. (Finnerty Decl., ¶ 65, Ex. L, M.)

AIG's expert, Dr. Charles C. Cox, a financial economist and former Commissioner and chief economist of the SEC, responded that Dr. Finnerty's own statistics on the trading of the bonds demonstrate that even when they were reported on TRACE, the 0.5% bonds were not traded on approximately 283 days and the Zero Coupon bonds were not traded on 62 days during the Class Period. Dr. Cox opined that this fact is inconsistent with market efficiency. (August 14, 2009 Tr., 173:10-23; Cox Decl., ¶ 62) (citing Finnerty Decl., Ex. G.)

Other than his comparison to the trading volumes of other, TRACE-reported bonds, Dr. Finnerty did not state his basis for opining that the AIG bonds have a large trading volume. Although Dr. Finnerty has shown that the 0.5% and Zero Coupon bonds trade at a higher average volume and in larger average dollar amounts than the majority of TRACE-reported bonds, this cannot support his claim that the AIG bonds trade in an efficient market without some further showing that these other bonds trade in an efficient market themselves. This Dr. Finnerty has not done. Nor has Dr. Finnerty compared the volume and value of trading in the AIG bonds to other securities for which studies or courts have found there exists an efficient market. Accordingly, the Court does not find that Lead Plaintiffs have shown evidence of an efficient market based upon the volume of trading.

Dr. Finnerty addressed the second Cammer factor by arguing that because AIG bonds were tracked by rating agencies such as Moody's, Standard & Poor's, and Fitch, as well as by a wide variety of industry research analysts, there would have been " wide and timely dissemination of information relevant to the valuation of AIG securities." (Finnerty Decl., ¶ ¶ 58-59, Ex. C; August 13, 2009 Tr., 71:5-9.) Dr. Cox argued, however, that rating agencies rate bonds based upon whether or not they are paid by the issuer, regardless of market efficiency, and that " ratings changes from rating agencies lag, not lead news ... [s]o that rating agencies wouldn't be a mechanism for rapidly incorporating information into the price of a security ..." (Cox Decl., ¶ ¶ 63-64; August 14, 2009 Tr., 174:1-11.)

Given Dr. Cox's explanation of the process by which rating agencies rate securities, the Court agrees that the mere fact that a rating agency rates a bond is not indicative of it trading in an efficient market. Further, as Dr. Cox noted, none of the industry analysts who examined AIG's debt securities discussed either the 0.5% or the Zero Coupon bonds specifically, but rather analyzed AIG's bond ratings and financial performance generally. (Cox Decl., ¶ 64.) Accordingly, the second Cammer factor provides little support to the claimed efficiency of the market for AIG's 0.5% and Zero Coupon bonds.

For the third Cammer factor, Dr. Finnerty explained that the AIG bonds were underwritten by Morgan Stanley, a market maker with average daily trading volumes for both bonds in the millions of dollars, and that they were also heavily traded by other major brokers such as Deutsche Bank, Citigroup, Bear Stearns, Lehman Brothers and Goldman Sachs. (Finnerty Decl., ¶ ¶ 67-68; August 13, 2009 Tr., 72:12-73:6.) Dr. Finnerty added that the stark similarity between the TRACE prices and the prices reported by these major brokers is indicative of an efficient market. (Finnerty Decl., ¶ 68.) First, it is not credible that Morgan Stanley had an " average value of daily trading volume during the Class Period [of] $1,841,182.00 for the 0.5% Notes" at the same time that the overall " average daily dollar value of trading during the Class Period for the 0.5% Notes was $576,546.00." ( Compare Finnerty Decl., ¶ 67, with Finnerty Decl. ¶ 65.) Similarly it is not credible that Morgan Stanley had an " average value of daily trading volume during the Class Period [of] ... $38,722,246.00 for the Zero-Coupon Debentures" at the same time that the overall " average daily dollar value of trading during the Class Period ... for the Zero-Coupon Debentures was $18,072,978.00." ( Compare Finnerty Decl., ¶ 67, with Finnerty Decl. ¶ 65.) The failure to explain these discrepancies is troubling to the Court. Further, while the Court finds that Dr. Finnerty has provided some evidence of market makers for the 0.5% and Zero Coupon bonds, he has not compared the level of trading of these brokers with market makers for other securities trading in markets that have been found to be efficient.

For the fourth Cammer factor, Dr. Finnerty testified, and Defendants do not appear to contest, that AIG was permitted to file an S-3 form. (August 13, 2009 Tr., 71:4-5.)

For the fifth and final Cammer factor, Dr. Finnerty opined that the close relationship between the returns on AIG common stock and the return on AIG bonds, as well as the relationship between the release of company-specific news and the reaction of bond prices further demonstrate the efficiency of the AIG bond market. (Finnerty Decl., ¶ ¶ 70-71; August 13, 2009 Tr., 78:7-79:10.) Dr. Finnerty examined whether there was a price change in the 0.5% and Zero Coupon bonds on four dates during the Class Period on which AIG-related news was released: July 29, 2001, January 29, 2002, February 21, 2002, and October 14, 2004. On each of these four dates Dr. Finnerty found either a price increase i n the Zero Coupon bonds when there was favorable news for AIG or a price decrease when there was unfavorable news for AIG. Additionally, on February 21, 2002, Dr. Finnerty found a price decrease in the 0.5% bonds upon unfavorable news for AIG and on July 29, 2001, he found a price increase upon favorable news for AIG. (Finnerty Decl., ¶ ¶ 72-75.)

Nevertheless, the Court finds Dr. Finnerty's analysis of the fifth Cammer factor to be deficient. First, in a separate portion of his analysis, Dr. Finnerty claimed to have found a statistically significant decrease in AIG stock related to the disclosure of AIG misstatements ¹⁷ on the dates of October 14, 2004, October 15, 2004, March 17, 2005, March 30, 2005, March 31, 2005, and April 1, 2005. ( See Section II.B.3.a.(2).(c), infra. ) However, on none of those dates did Dr. Finnerty report the results of event studies on the 0.5% and Zero Coupon bonds, which apparently failed to show statistically significant price changes on those dates, attributable to the disclosures.¹⁸ (August 13, 2009 Tr., 84:23-86:1) (Dr. Finnerty, agreeing that for the 0.5% and Zero Coupon bonds, " generally ... there were no statistically significant price movements on any of the disclosures dates on which [he] focused with respect to AIG stock" ).

17 See Section II.B.3.a.(2).(c).(i), infra, for a discussion of event studies and statistical significance.

18 At his May 15, 2008 deposition, Dr. Finnerty testified:

In addition, while Dr. Finnerty did examine and report the price changes of the bonds on the four dates of July 29, 2001, January 29, 2002, February 21, 2002, and October 14, 2004, he did not report the event studies for any of those dates, in that the studies did not show a statistically significant price change attributable to the AIG news. (Finnerty Dep. Tr., 167:4-23; 171:24-172:13.)

Further, as Dr. Cox noted, Dr. Finnerty's overall finding of a relationship between AIG stock and bonds is inconsistent with the price change on October 14, 2004, the lone disclosure date that Dr. Finnerty examined for both AIG stock and bonds. On that day, given the relationship Dr. Finnerty calculated to exist between the AIG stock and the debt securities, as well as the price decrease in AIG stock, Dr. Cox calculated that there should have been price decreases of 2.31% and 1.42% for the 0.5% and Zero Coupon bonds, respectively. However, Dr. Finnerty's results showed price decreases of 0 and 0.3%, respectively. (Finnerty Decl., ¶ 75; Cox Decl., ¶ 66; August 14, 2009 Tr., 175:12-176:1.) Thus, on one of the key dates during which significant AIG-related disclosures occurred, there was little or no movement in the price of the debt securities, despite large decreases in the price of AIG stock, both in absolute terms, and in the amounts attributable to the disclosures according to Dr. Finnerty's event study. Even more problematic for Dr. Finnerty's argument is that on two of the four dates that he measured the change in bond prices, the 0.5% bonds did not trade at all; a finding that Dr. Cox opined, and the Court agrees, is not indicative of market efficiency. (Cox Decl., ¶ 60; August 14, 2009 Tr., 172:8-173:1) (citing Finnerty Decl. ¶ ¶ 72-75).

With regard to evidence provided by the Parties that does not directly relate to the Cammer factors, Dr. Finnerty further argued that the eligibility and reporting of AIG bonds on TRACE, which became operational on July 1, 2002, " has increased transparency in the corporate bond market" and is an important indicator of market efficiency. (Finnerty Decl., ¶ ¶ 60-66; August 13, 2009 Tr., 71:24-72:7.) In response to Dr. Finnerty, Dr. Cox contended that prior to the implementation of TRACE, the trading of most debt securities took place " on the over-the-counter market between dealers and their customers," essentially privately. (Cox Decl., ¶ 48; August 14, 2009 Tr., 168:1-170:1.) Furthermore, the 0.5% bonds were trading on TRACE for only 2 of the 5 years during the Class Period, while the Zero Coupon bonds were trading on TRACE for 32 of 40 months during the Class Period. (Cox Decl., ¶ 61; August 14, 2009 Tr., 169:21-170:1.) Additionally, Dr. Cox opined that although TRACE is a necessary condition for market efficiency of bonds, it is not sufficient, given both his own analysis of other securities traded under TRACE and studies by financial economists who have found that a market for bonds is not necessarily efficient because it is reported through TRACE. (Cox Decl., ¶ 57; August 14, 2009 Tr., 170:2-18.)

The Court agrees with Dr. Finnerty that during the period in which AIG's bonds were reported on TRACE, there was increased transparency in those markets. However, that finding begs the question of whether those markets were efficient during the period when they were not reported on TRACE. Further, the Court agrees with Dr. Cox that transparency is insufficient by itself to demonstrate the efficiency of a market, and that it is merely one factor that may be considered in concert with the Cammer factors and other evidence of a market's efficiency.

Dr. Finnerty also found that brokers who dealt in the 0.5% and Zero Coupon bonds had relatively low average " bid-ask spreads," ¹⁹ below the median bid-ask spread of convertible securities whose transactions were reported on TRACE. (Finnerty Decl., ¶ 69; August 13, 2009 Tr., 71:5-9.) Because Dr. Finnerty calculated that the 0.5% and Zero Coupon bonds had lower bid-ask spreads than the median bid-ask spreads of similar bonds that were reported on TRACE and were outstanding during the full Class Period, he found that the markets for the bonds were relatively liquid, which he argued is itself an indicator of an efficient market. (Finnerty Decl., ¶ 69.)

19 The " bid-ask spread" is the difference between the prices at which broker-dealers are willing to buy (the " bid" price) and sell (the " ask" price) a security in transactions with customers, and " reflects the liquidity in the market for that security." (Finnerty Decl., ¶ 69.)

The Court questions Dr. Finnerty's apparent assumption that if the market for AIG bonds is more liquid than that of similar bonds, this by itself constitutes evidence of efficiency. This would only constitute evidence of an efficient market, however, if those similar bonds, or other securities with similar liquidity, had been found to trade in an efficient market due at least in part to their liquidity. Dr. Finnerty and Lead Plaintiffs have provided no such evidence.

In examining the five Cammer factors, as well as additional evidence presented by the Parties, the Court finds that, while Lead Plaintiffs have provided evidence of market makers, transparency during the TRACE period, as well as AIG's eligibility to file S-3 Registration Statements, these factors are undermined by the lack of transparency that existed prior to TRACE, the lack of evidence of trading volumes commensurate with securities that have been found to trade in efficient markets, the lack of evidence of research analysts following the specific bonds at issue, the absence of trading on many days during the Class Period, and the complete absence of price decreases in those bonds of statistical significance due to AIG-related events. Further, the Court finds it problematic that on two of the days when significant AIG-related news was announced, the 0.5% bonds were not traded at all. Indeed, the Court questioned Dr. Finnerty about this during the August 13, 2009 hearing:

THE COURT: Let me just ask you something, Dr. Finnerty. If someone were alarmed at the [bad] news, would they not want to sell?

THE WITNESS: They might.

THE COURT: They might?

THE WITNESS: Yes.

THE COURT: Why wouldn't they?

THE WITNESS: [If][t]hey didn't want to take a loss. They might be comfortable that if the prices adjust, they might simply want to continue to hold the bonds. On the other hand, if they were alarmed, yes, they would sell, could sell.

MR. LEFFELL (Counsel for AIG): On the day of this announcement, is your testimony that of all the millions upon millions of dollars of these bonds that were outstanding, nobody wanted to take a loss, and that's why nobody sold. Is that your testimony?

THE WITNESS: There could be some other reason, but nobody sold. They had to have a reason. There are sophisticated investors, and they decided they would rather hold than sell.

MR. LEFFELL: And your testimony is that the one reason that it couldn't be is that the news had no effect on the price of the bonds ... ?

THE WITNESS: I don't know why. I didn't go poll them, but they were comfortable holding the bonds, and that's why they didn't sell them.

(August 13, 2009 Tr., 93:25-94:23.) The Court finds Dr. Finnerty's explanation for the lack of sales to be inconsistent with the theory of an efficient market, and thus does not find that he has provided a credible explanation for the reason that no 0.5% bonds were sold on two of the AIG-related news days.

Finally, in light of the fact that Dr. Finnerty did not report any event studies measuring the statistical effect of AIG-related disclosures on the price of the 0.5% and Zero Coupon bonds, and because the unreported event studies he did perform did not show statistically significant changes in the bond prices attributable to such disclosures, the Court finds that the fifth Cammer factor weighs heavily against Lead Plaintiffs' contention that the bonds traded in an efficient market. As the Second Circuit recently explained:

Evidence that unexpected corporate events or financial releases cause an immediate response in the price of a security has been considered the most important Cammer factor ... and the essence of an efficient market .... Without the demonstration of such a causal relationship, it is difficult to presume that the market will integrate the release of material information about a security into its price. An event study that correlates the disclosures of unanticipated, material information about a security with corresponding fluctuations in price has been considered prima facie evidence of the existence of such a causal relationship.

Teamsters Local 445, 546 F.3d at 207-08. Indeed, Lead Plaintiffs have not cited to, nor is the Court aware of, any court in this jurisdiction that has found debt securities to trade in an efficient market without first having found that the price of those bonds is responsive to company-specific news at a statistically significant level.

Accordingly, Lead Plaintiffs have not shown, by a preponderance of the evidence, that AIG's 0.5% and Zero Coupon bonds traded in an efficient market. As a result, Lead Plaintiffs cannot gain the benefit of the fraud-on-the-market presumption and common issues of reliance do not predominate over individual issues for their § 10(b) and Rule 10b-5 bond claim in Count Five. Accordingly, class certification is denied for the bond claims in Count Five. See In re Flag Telecom, 574 F.3d at 39 (" In re IPO makes clear that courts may resolve contested factual issues where necessary to decide on class certification, and when a claim cannot succeed as a matter of law, the Court should not certify a class on that issue." ) (quoting McLaughlin v. Am. Tobacco Co., 522 F.3d 215, 228 (2d Cir.2008)).

(b). Loss Causation at the Class Certification Stage

AIG also contends that Lead Plaintiffs have not established loss causation and that such a showing is necessary to establish the fraud-on-the-market presumption and reliance at the class certification stage. (AIG Mem. of Law, 21-22; February 10, 2009 Tr., 75:8-76:3.) In support, AIG cites Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261 (5th Cir.2007), for the proposition that " ‘ loss causation must be established at the class certification stage by a preponderance of all admissible evidence’ before a plaintiff can rely on a fraud-on-the-market presumption." (AIG Mem. of Law, 21) (quoting Oscar, 487 F.3d at 269).

However, the Second Circuit's decision in In re Salomon Analyst Metromedia Litigation, 544 F.3d 474 (2d Cir.2008), rejected the holding in Oscar. The In re Salomon Court, while acknowledging that loss causation was an element of securities fraud that would ultimately have to be proven under § 10(b) and Rule 10b-5, held that at the class certification stage, under the Supreme Court's decision in Basic v. Levinson, 485 U.S. 224, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988), " plaintiffs do not bear the burden of showing an impact on price" or loss causation in order to establish reliance. In re Salomon, 544 F.3d at 483.

(c). Rebuttal of Presumption of Reliance

To warrant the presumption of reliance, the Salomon Court determined that the Basic Court set up a test of general applicability: where a defendant has (1) publicly made (2) a material misrepresentation (3) about securities traded in an efficient market, reliance on those representations may be presumed.

However, because the fraud-on-the-market reliance theory is dependent on these assumptions, " 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." Basic, 485 U.S at 248, 108 S.Ct. 978.²⁰ " In re IPO now requires a district court to make a ‘ definitive assessment’ that the Rule 23(b)(3) predominance requirement has been met. This assessment cannot be made without determining whether defendants can successfully rebut the fraud-on-the-market presumption. The Basic Court explained that a successful rebuttal defeats certification by defeating the Rule 23(b)(3) predominance requirement. Hence, the court must permit defendants to present their rebuttal arguments ‘ before certifying a class ...’ " In re Salomon, 544 F.3d at 485 (internal citations omitted).

20 For instance, defendants may rebut the presumption if they show " that the misrepresentation in fact did not lead to a distortion of price." Id.

In opposing class certification, the Greenberg/Starr Defendants contended that the Defendants' experts had rebutted the fraud-on-the-market presumption. Specifically, they contended that by successfully challenging Lead Plaintiffs' purported evidence of a statistically significant price change on four of the six days on which public disclosure of the frauds occurred, they had carried their burden under In re Salomon and shown that the fraud-on-the-market presumption should not apply. (February 10, 2009 Tr., 93:13-106:10.) Lead Plaintiffs have attempted to defeat this argument by suggesting that In re Salomon and the fraud-on-the-market presumption depend solely on price changes in the stock at the time of the alleged misstatements and omissions, rather than at the time those misrepresentations are publicly disclosed to the market. (February 10, 2009 Tr., 44:3-46:13.) Lead Plaintiffs' argument overlooks the fact that if the market on which AIG stock traded was indeed efficient-an assumption and requirement of the fraud-on-the-market presumption-then the absence of a price decline on the day a fraud was disclosed will strongly indicate that there was also no price increase on the day the fraud occurred. Thus, to the extent a Defendant can show that there was no price decrease in AIG stock on the date a misrepresentation was disclosed, the Court views this showing as strong evidence that there was no price change on the date of the misrepresentation, thus rebutting the fraud-on-the-market presumption.

(i). Lead Plaintiffs' Event Studies

Parties to a securities class action who seek to demonstrate that a change in the price of a stock is or is not attributable to misrepresentations or omissions by the defendant frequently resort to the use of statistical " event studies," performed by experts in the field of financial economics. See, e.g., In re Flag Telecom, 574 F.3d at 34, 40; Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 209 (2d Cir.2008); Fogarazzo v. Lehman Bros., Inc., 263 F.R.D. 90, 106-109 & nn. 155, 159 (S.D.N.Y.2009).

Lead Plaintiffs also engaged Dr. John D. Finnerty to perform an event study on the decrease in AIG's stock prices attributable to disclosures of the alleged fraud. " [A]n event study is a standard technique that financial economists use to determine whether a security's reaction to a news announcement (or some other event) is statistically significant." (Finnerty Decl., ¶ 28.) " The basic notion is to disentangle the effects of two types of information on stock prices-information that is specific to the firm under question ... and information that is likely to affect stock prices marketwide (e.g., change in interests rates)." Mark L. Mitchell & Jeffrey M. Netter, The Role of Financial Economics in Securities Fraud Cases: Applications at the Securities and Exchange Commission, 49 Bus. Law. 545, 556-57 (1994). An event study has several components: " defining the event and announcement day(s), measuring the stock's return during the announcement period, estimating the expected return of the stock during this announcement period in the absence of the announcement, and computing the abnormal return (actual return minus expected return) and measuring its statistical and economic significance." Sanjai Bhagat & Roberta Romano, Event Studies and the Law: Part I: Technique and Corporate Litigation, 4 Am. L. & Econ. Rev. 141, 143-44 (2002).

Lead Plaintiffs' expert, Dr. Finnerty, used the Fama-French Three-Factor Model, and a modified version of that model in his event studies in order to measure and test the statistical significance of the alleged abnormal return on AIG's stock on various days on which AIG disclosed information that had previously been misrepresented or omitted. (Finnerty Decl., ¶ 13.) The Fama-French Three-Factor Model identifies and corrects for three factors that explain excess stock returns: " the excess return on the market over treasury bills; the difference between the returns on small-cap stocks and large-cap stocks; and the difference between returns on high book-to-market (basic value) stocks and low book-to-market (gross) stocks." (Finnerty Decl., ¶ 13; August 13, 2009 Tr., 17:19-18:7.)

Dr. Finnerty's event studies, however, used a " modified" Fama-French Three-Factor Model that adds a " fourth factor [to] measure the return on insurance stocks" in order to control for " industry effects." (Finnerty Decl., ¶ 31; August 13, 2009 Tr., 19:4-12.) Finnerty testified that he used this model because:

[financial economists] know from prior empirical research that the return on stocks are affected by factors that cut across an industry. So if I am measuring the returns on a share of stock, I would normally apply a modified model to capture the industry effects. Another way to say it is when I'm testing for abnormal returns, what I would like to be able to do is to make sure that I don't bias the results by including industry-related factors.

(August 13, 2009 Tr., 19:13-20.)

However, although most of the disclosure dates at issue contained only AIG-related disclosures, the disclosures on October 14 and 15, 2004 were part of the announcement of investigations into widespread insurance industry misbehavior, affecting 18 of the 21 companies on Standard & Poor's insurance industry index. For this reason, Dr. Finnerty decided to rely on a model that removes the industry-wide control variable for the October 14 and 15, 2004 event studies, thus returning to the " unmodified" Fama-French Model. (Finnerty Decl., ¶ ¶ 32-33; August 13, 2009 Tr., 19:21-20:22.)

Although Dr. Finnerty initially performed abbreviated event studies for eleven dates during the Class Period, he determined that only six of those dates showed a statistically significant change in price and was not asked by Lead Plaintiffs' counsel to complete a full event study for the remaining five dates. (August 13, 2009 Tr., 99:20-100:6.) The six disclosure dates for which Dr. Finnerty opined that he had found a statistically significant price decline include October 14 and 15, 2004, March 17, March 30, March 31, and April 1, 2005. (Finnerty Decl., ¶ ¶ 32-37.)

The statistical significance of a measured effect, in this case a price decrease attributable to disclosures, is determined by " comparing a p-value to a pre-established value, the significance level." David H. Kaye and David A. Freedman, Reference Guide on Statistics, in Federal Judicial Center, Reference Manual on Scientific Evidence, 123 (2d ed.2000). The " p-value" is the " probability of getting data as extreme as, or more extreme than, the actual data, given that the null hypothesis is true." Id. at 122. The " null hypothesis," in turn, is the hypothesis that there is no price decrease attributable to the disclosures, but that the model's output, showing such a decrease, is merely the result of random chance in the sampling data. Id.

As an example, if one were attempting to measure whether a decrease in the price of AIG stock on a given day is attributable to a disclosure of a particular piece of information, the null hypothesis would be that the change in price attributable to that disclosure is equal to or greater than 0. For example, the price decrease of AIG stock on March 30, 2005 was 1.79%. (Finnerty Decl., ¶ 35.) Using regression analysis, Dr. Finnerty determined that on March 30, 2005, the price decrease of AIG stock attributable to AIG related disclosures was 1.61%. (Finnerty Decl., ¶ 35.)

Given that the results of Dr. Finnerty's regression model show a decrease of 1.61% in the stock price attributable to the AIG related disclosure, the p-value is said to be the probability of finding a decrease of 1.61% or more, even when the null hypothesis is true, i.e., even when the price change attributable to the disclosure was actually equal to or greater than zero. ²¹ To complete the example, given that Dr. Finnerty's model resulted in a p-value of 8.7 for March 30, 2005, this means that if the null hypothesis were correct, and thus that there was no price decrease attributable to the disclosure on March 30, 2005, there is a 8.7% chance that the model would nevertheless show, by chance, a 1.61% or greater decrease in the price attributable to the disclosure. (Finnerty Decl., ¶ 35.)

21 " In general, the p-value depends on the model and its parameters, the size of the sample, and the sample statistics." Kaye and Freedman., at 123.

If the pre-established significance level in our example were 5%,²² then for a p-value of 8.7, the " findings are not significant, and the null hypothesis is not rejected." Kaye and Freedman, at 125. If the pre-established significance level were 10%, then the p-value would be significant, and the null hypothesis would be rejected. However, one must keep in mind that because " p is calculated by assuming that the null hypothesis is correct ... the p-value cannot give the chance that this hypothesis is true" and thus " there is no meaningful way to assign a numerical probability to the null hypothesis." Id. at 122. Similarly, although when " results are significant at the [5%] level, it is tempting to conclude that the null hypothesis has only a 5% chance of being correct ... [t]his temptation should be resisted" given that " [s]ignificance comes no closer to expressing the probability that the null hypothesis is true than does the underlying p-value." Id. at 124-25.

22 " In practice, statistical analysts often use certain preset significance levels-typically .05 or .01. The .05 level is the most common in social science, and an analyst who speaks of ‘ significant’ results without specifying the threshold probably is using this figure. An unexplained reference to ‘ highly significant’ results probably means that p is less than .01." Kaye and Freedman, at 124.

As a result, the fact that a financial model finds a statistically significant price decrease at the 5% level on a particular day, attributable to the disclosure of previously omitted or misstated information, cannot be interpreted as meaning that there is a 95% chance that the measured price decrease attributable to the disclosures is real. Nevertheless, a finding of statistical significance will be quite important to the Court's determination, because a statistically significant price decrease attributable to the disclosures may be interpreted as compelling evidence that the null hypothesis should be rejected. Thus, a finding of a statistically significant effect will strongly weigh against a finding that the Defendant has rebutted the fraud-on-the-market presumption of reliance.

a). The October 14 and 15, 2004 Event Studies

On October 14, 2004, at about 12:15 P.M., then Attorney General Eliot Spitzer announced the filing of civil litigation against Marsh & McLennan Companies for bid-rigging and revealed that AIG and others had paid Marsh hundreds of millions of dollars in contingent commissions in exchange for steering clients. (TAC, ¶ ¶ 222-26; Finnerty Decl., ¶ 32; August 13, 2009 Tr., 33:19-39:18.) Dr. Finnerty's unmodified French-Fama Model event study showed a 3.79% price decrease in its stock price that day, attributable to the disclosure of AIG's involvement in the bid-rigging and contingent commissions, which was statistically significant at the 1% level. (Finnerty Decl., ¶ 32.)

On October 15, 2004, in a 10:00 A.M. conference call with market analysts, Defendant Greenberg revealed that AIG had received a subpoena from the New York Attorney General in September and had launched an internal investigation into the alleged bid-rigging scheme. (TAC, ¶ ¶ 302-03; Finnerty Decl., ¶ 33; August 13, 2009 Tr., 39:19-41:17.) Dr. Finnerty's unmodified French-Fama Model event study showed a 2.71% price decrease in its stock price that day, attributable to the further disclosure of AIG's involvement in the bid-rigging and contingent commissions, which was statistically significant at the 1% level. (Finnerty Decl., ¶ 33.)

In an attempt to rebut the fraud-on-the-market presumption for the October 14 and 15, 2004 disclosure dates, AIG contends that because Dr. Finnerty did not initially disclose the results of additional event studies he performed for those two days using the modified French-Fama Model, ( compare Finnerty Decl., ¶ ¶ 32-33, with Finnerty Reb. Decl., ¶ ¶ 7, 11), and the fact that the modified model led to results that were not statistically-significant at the 5% level, ²³ the Court should not credit Dr. Finnerty's findings for those dates using the unmodified model. (AIG Mem. of Law, 22.)

23 Dr. Finnerty's application of the modified French-Fama Model to October 14, 2004 showed a price decrease of 1.68% attributable to the disclosure of AIG's involvement in the bid-rigging and contingent commissions, statistically significant at the 10% level. Dr. Finnerty's application of the modified French-Fama Model to October 15, 2004 showed a price decrease of 1.11% attributable to the further disclosure of AIG's involvement in the bid-rigging and contingent commissions, which was not statistically significant at the 10% level. (Finnerty Reb. Decl., Ex. A, at 2.)

Dr. Finnerty only reported the unmodified results for the October 14 and 15, 2004 dates in his initial declaration. (Finnerty Decl., ¶ ¶ 32-33; August 13, 2009 Tr., 65:24-66:17; August 13, 2009 Tr., 102:20-103:5.) From this incomplete reporting procedure, and the fact that the unreported October 14 and 15, 2004 modified Fama-French Model did not show a price decrease attributable to the disclosures at the 5% level of statistical significance, Dr. Cox opined that Dr. Finnerty's " procedures [do not] come up to the level of a scientific methodology." (August 14, 2009 Tr., 160:1-10; see also Cox Decl., ¶ ¶ 45-46.)

While the Court finds Dr. Cox's criticism to be valid, and believes the better analysis would have had Dr. Finnerty submit the October 14 and 15, 2004 modified Fama-French Model results in his initial declaration, it does not find Dr. Finnerty's failure to do so to be dispositive.

The Court finds the logic of Dr. Finnerty's decision of which models to rely on for each date to be sound and persuasive, given that had Dr. Finnerty corrected for, or " netted out" the industry-wide effects of October 14 and 15, 2004, he would have been correcting for, or netting out the very effects he was trying to measure. In contrast, on the remaining four days, by correcting for industry-wide effects, Dr. Finnerty properly netted out misleading changes in price caused by industry events, trends or projections. Indeed, AIG's expert, Dr. Cox, did not contend that Dr. Finnerty used the wrong models on October 14 and 15, 2004. (August 14, 2009 Tr., 208:14-21.) The fact that the modified model led to results that were not statistically significant at the 5% level is not surprising, given that the model was actually designed to correct for or ‘ net out’ industry-wide influences on price, such as the Attorney General's disclosures regarding the bid-rigging scheme.

The Court also notes Dr. Finnerty's testimony that he had decided on the correct model beforehand because he had " worked in other cases that involved the same issue, and [he] had made that determination that for ... those particular dates running the modified model would involve some bias." (August 13, 2009 Tr., 103:19-25.) The Court finds Dr. Finnerty's testimony about the timing of his determination to utilize the unmodified model for the October 2004 dates to be credible, and notes that Defendants themselves have not challenged his testimony's veracity or credibility on this specific point.

Accordingly, the Court accepts Dr. Finnerty's event study for October 14 and 15, 2004, which shows a price decrease attributable to AIG's disclosures at the 1% statistical significance level, and finds that Defendants have not rebutted the fraud-on-the-market presumption for prospective class members who held their stock over those dates.

b). The March 17, 2005 Event Study

On March 17, 2005, the Wall Street Journal reported that the New York Attorney General and the SEC had expanded their probes of AIG to include additional accounting issues, including its accounting of nearly $1,200,000,000.00 in reinsurance transactions with Richmond Insurance and Union Excess. (TAC, ¶ ¶ 347-48; Finnerty Decl., ¶ 34; August 13, 2009 Tr., 43:3-45:6.) Further, the Wall Street Journal also reported that AIG had revealed the possibility of a finding by its outside accountant, PwC, that there was a material weakness in its accounting controls. (TAC, ¶ 578; August 13, 2009 Tr., 43:16-20.) Dr. Finnerty's modified French-Fama Model event study showed a 1.99% price decrease in AIG's stock price that day, attributable to the disclosure of AIG's accounting issues, which was statistically significant at the 5% level. (Finnerty Decl., ¶ 34.) ²⁴

24 Dr. Finnerty also applied the unmodified French-Fama Model as a check, which showed a price decrease of 2.26% attributable to the disclosure of AIG's accounting issues, again statistically significant at the 5% level. (Finnerty Decl., ¶ 34.)

Dr. Cox did not contest that Lead Plaintiffs can show a price decrease for the disclosures of March 17, 2005. (February 10, 2009 Tr., 75:21-76:3; August 14, 2009 Tr., 180:10-181:1.) As a result, the Court finds that Defendants have not rebutted the fraud-on-the-market presumption for prospective class members who held over March 17, 2005.

c). The March 30 and 31, 2005 Event Studies

On March 30, 2005, AIG issued a press release announcing that at least one insurance transaction with Gen Re had not been properly documented, that it would delay filing its 2004 form 10-K while reviewing its accounting, and that the aggregate of accounting adjustments could reach $1,700,000,000.00. (TAC, ¶ ¶ 357-62, 465-67, 759-80; Finnerty Decl., ¶ 35; August 13, 2009 Tr., 45:7-49:6.) Dr. Finnerty's modified French-Fama Model event study showed a 1.61% price decrease in AIG's stock price that day, attributable to the disclosure of AIG's accounting problems, which was statistically significant at the 10% level. (Finnerty Decl., ¶ 35.) ²⁵

25 Dr. Finnerty also applied the unmodified French-Fama Model as a check, which showed a price decrease of 1.84% attributable to the disclosure of AIG's accounting problems, again statistically significant at the 10% level. (Finnerty Decl., ¶ 35.)

On March 31, 2005, the Wall Street Journal reported that PwC had received a subpoena in connection with AIG's Gen Re transaction and that the Government's investigation into AIG's accounting practices could widen beyond those areas noted in the March 30, 2005 press release. (TAC, ¶ 365; Finnerty Decl., ¶ 36; August 13, 2009 Tr., 49:7-51:12.) Dr. Finnerty's modified French-Fama Model event study showed a 1.54% price decrease in AIG's stock price that day, attributable to the further disclosure of AIG's accounting problems, which was statistically significant at the 10.2% level. (Finnerty Decl., ¶ 36.) ²⁶

26 Dr. Finnerty also applied the unmodified French-Fama Model as a check, which showed a price decrease of 1.87% attributable to the further disclosure of AIG's accounting problems, statistically significant at the 10% level. (Finnerty Decl., ¶ 36.)

Defendant AIG contends that it has rebutted the presumption of fraud-on-the-market for March 30 and March 31, 2005, given that Dr. Finnerty's event study does not show a statistically significant price decrease on those days, measured at the 5% ²⁷ level. (AIG Mem. of Law, 21-22.) In support of this argument, AIG's expert, Dr. Cox, contended that 5% is the minimum level of statistical significance that conventional statistical methodology would accept in making a finding that the price decrease is attributable to AIG's disclosures. (August 13, 2009 Tr., 130:8-132:2; Cox Decl., ¶ ¶ 42-44, 46.) Dr. Cox also noted that Dr. Finnerty " previously wrote that at least 5% was required for statistical significance, but [Dr. Finnerty] claims that his thinking has changed." (Cox Decl., ¶ 44) (citing Finnerty Decl., ¶ ¶ 77-82.)

27 See footnote 22, supra.

Thus, although Dr. Finnerty's event study showed a price decrease attributable to AIG's disclosures on March 30 and 31, 2005, measured at and near, respectively, the 10% level of statistical significance, (August 13, 2009 Tr., 46:19-25; August 13, 2009 Tr., 49:24-50:5; Finnerty Decl., ¶ ¶ 35-36), Dr. Cox argued that the Court should not accept that portion of the study as competent evidence, given its purported conflict with standard methodology.

On August 13, 2009, Dr. Finnerty testified that, for statistical significance in the financial economics field, " the standard is ten percent or better." (August 13, 2009 Tr., 31:14-21.) Dr. Finnerty based his conclusion largely on a literature review of the top three finance journals in the United States, covering empirical results from the three and three-quarters' years prior to his November 2008 rebuttal declaration. (August 13, 2009 Tr., 27:20-28:6; Finnerty Reb. Decl., ¶ ¶ 14-15.) In conducting this review, Dr. Finnerty found that:

68.6 percent ... of all of those articles that reported empirical results and specified the minimum standard applied ... the standard that the authors applied was for a cut off of ten percent or better. About one in four used a cut off of five percent or better and only a very small number used a cut off of one percent or better. So the conclusion that I drew is that the standard that is applied in the field of financial economics today in reporting the results of empirical testing is to use a minimum critical threshold of ten percent.

(August 13, 2009 Tr., 30:5-19; see also Finnerty Reb. Decl., ¶ ¶ 15-16 (stating that he " review[ed] 842 articles from the three journals, of which 494 articles reported statistical significance in connection with their empirical tests" and noting that the three journals reported statistical significance at the 10% level or better in 67.7%, 72%, and 62.5% of the articles that reported statistical significance in connection with their empirical tests" )).

During cross-examination, however, Dr. Finnerty conceded that he could not testify, one way or the other, whether, or to what extent, the articles that reported at the 10% significance level also drew conclusions in the text of the article about a hypothesis based on that 10% significance level. (August 13, 2009 Tr., 105:9-106:25.) Indeed, Dr. Cox testified that in his review of the 58 articles from Dr. Finnerty's own literature review that explicitly referred to event studies, he found that " there were at most four that had some conclusion that arguably could be attributed to a ten percent level, but even among those, the primary conclusions in the article were at the five percent or more level." ²⁸ (August 13, 2009 Tr., 133:3-24; August 13, 2009 Tr., 138:22-139:2.) Upon cross examination, Dr. Cox readily conceded that " quite a few" of these 58 articles reported statistical significance at the 10% level, (August 14, 2009 Tr., 191:14-19) yet continued to assert that no more than four of the articles drew conclusions at the 10% level. (August 14, 2009 Tr., 199:1-7). Further, upon reviewing particular articles, Dr. Cox explicitly rejected opposing counsel's suggestion that most of those articles that reported results at the 10% level were drawing conclusions based upon those results. (August 14, 2009 Tr., 203:12-20, 204:5-23, 205:10-24, 205:25-206:8; but see August 14, 2009 Tr., 200:13-201:3, 202:8-18.)

28 Although Lead Plaintiffs objected to the admission of Dr. Cox's study during the midst of his testimony, the Court required AIG to turn over documents related to the study and explained to Lead Plaintiffs' counsel that " there would be no prejudice to you as a result of the fact you [may] come back early and often ... and if you need to, you can have the expert recalled." (August 14, 2009 Tr., 156:2-9.) Lead Plaintiffs cross-examined Dr. Cox on the study on August 14, 2009 and, in the more than six months since cross-examination, Lead Plaintiffs have not requested that he be recalled.

The Court finds the distinction between reporting results and drawing conclusions about hypotheses based on those results to be crucial to the question before it. The Court is not charged with merely reporting levels of statistical significance in this case, but must draw conclusions about the likelihood of causal relationships and make rulings based upon those conclusions.²⁹ The Court finds that Dr. Finnerty's testimony and studies at most demonstrate that it has recently become acceptable in the financial economics field generally to report results at the 10% level. This does not demonstrate, however, that it is consistent with standard methodology in financial economics, or in conducting event studies specifically, to draw conclusions at the 10% level. Dr. Cox' testimony, as well as his review of the articles used by Dr. Finnerty, further support this finding that there is a distinction between reporting and drawing conclusions based on a 10% level of statistical significance. Accordingly, the Court finds that Defendant AIG, through its expert Dr. Cox, has rebutted the fraud-on-the-market presumption for the dates of March 30 and 31, 2005.

29 As already explained, supra, Dr. Finnerty's finding of a price decrease attributable to the disclosures at the 10% level of statistical significance is not the same as a finding that there is a 90% chance that there was a price decrease attributable to the disclosures. See David H. Kaye and David A. Freedman, Reference Guide on Statistics, in Federal Judicial Center, Reference Manual on Scientific Evidence, 122 (2d ed.2000) (explaining that the p-value, or level of statistical significance " cannot give the chance that th[e] [null] hypothesis is true" but " merely gives the chance of getting evidence against the null hypothesis as strong or stronger than the evidence at hand- assuming the null hypothesis to be correct." ) (emphasis added).

d). The April 1, 2005 Event Study

On April 1, 2005, The Wall Street Journal reported that Defendant Greenberg would resign as Chairman of AIG on March 29, 2005 due to a threat by the Attorney General to indict AIG, that regulators were examining possible accounting improprieties by AIG, and that AIG employees may have destroyed or removed documents. (TAC, ¶ ¶ 366-70, 497-98, 576; Finnerty Decl., ¶ 374; August 13, 2009 Tr., 51:13-57:4.) Dr. Finnerty's modified French-Fama Model event study showed a 5.71% price decrease in AIG's stock price that day, attributable to the further disclosure of AIG's accounting issues, which was statistically significant at the 1% level. (Finnerty Decl., ¶ 37.) ³⁰

30 Dr. Finnerty also applied the unmodified French-Fama Model as a check, which showed a 6.16% price decrease attributable to the further disclosure of AIG's accounting issues, again statistically significant at the 1% level. (Finnerty Decl., ¶ 37.)

However, on April 4, 2005 the Wall Street Journal issued a correction to its April 1, 2005 story, stating that it had incorrectly reported that AIG documents had been removed from AIG's Bermuda Office. (Finnerty Decl., ¶ 37.) As a result, Dr. Finnerty calculated a modified measure of the price change on April 1, 2005, this time subtracting 100% of the April 4, 2005 morning increase in AIG stock price from the April 1, 2005 decrease in order to correct for the effect of the incorrect April 1, 2005 disclosure by the Wall Street Journal. (Finnerty Decl., ¶ 38-39.) The resulting 3.80% price decrease attributable to the disclosures was still statistically significant at the 1% level, which led Dr. Finnerty to opine that the accurate disclosures on April 1, 2005 were still responsible for a price decrease. (Finnerty Decl., ¶ 39; August 13, 2009 Tr., 52:11-54:13.)

Dr. Cox did not contest that Lead Plaintiffs can show a price decrease for the disclosures of April 1, 2005.³¹ (February 10, 2009 Tr., 75:21-76:3; August 14, 2009 Tr., 180:10-181:1.) The Court finds that Defendants have not rebutted the fraud-on-the-market presumption for prospective class members who held over April 1, 2005.

31 Prior to their voluntary dismissal without prejudice, the Greenberg/Starr Defendants and their expert, Dr. David Tabak, argued that Dr. Finnerty could not show a statistically significant price decrease for the disclosures of April 1, 2005. (August 13, 2009 Tr., 98:21-101:17; Tabak Decl., 37-39.)

Accordingly, the Court finds that Defendants have rebutted the fraud-on-the-market presumption for March 30, 2005 and March 31, 2005, but have failed to rebut the presumption for October 14, 2004, October 15, 2004, March 17, 2005, and April 1, 2005. Only those prospective class members who held AIG stock over one or more of the days for which the Court finds Defendants have not rebutted the fraud-on-the-market presumption may be included in the certified class.

Therefore, with the exceptions of the March 30 and 31, 2005 price decreases, the claims against the Gen Re Defendants, and the § 10(b) and Rule 10b-5 bond claim, all of which are excluded from the issues certified, the Court finds that common questions of fact and law predominate over individual questions, such that the predominance requirement of Rule 23(b)(3) has been satisfied for the remaining claims.

b. Superiority

The Court must also find that " a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Rule 23(b)(3). The relevant matters to this finding include: " (A) the class members' interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action." Rule 23(b)(3)(A)-(D).

Here, Defendants do not challenge the superiority of the class action form to other methods of resolving the controversy, and the four factors described in Rule 23(b)(3) overwhelmingly support a finding of superiority. The proposed class is substantial in size, consisting of more than two million class members, and thus the desirability of conducting the litigation in the class action form substantially outweighs the interest of members of the class in individually controlling the prosecution of separate actions. Finally, although there are likely to be difficulties encountered in managing a class action of this magnitude, such difficulties are necessary evils in avoiding the undoubtedly much larger difficulties that would arise from resolving the controversy through thousands of smaller suits.

Accordingly, the Court finds that the class action form is clearly superior to other forms of resolving this controversy, and that the superiority requirement of Rule 23(b)(3) has been satisfied.

C. Rule 23(C)(1)(B)

For the foregoing reasons Lead Plaintiffs have satisfied each of the Rule 23(a) and (b)(3) requirements and the Court GRANTS the Motion for Class Certification. Rule 23(C)(1)(B) states that " [a]n order that certifies a class action must define the class and the class claims, issues, or defenses, and must appoint class counsel under Rule 23(g)."

1. Class Definition Modified

a. Exclusion of Claims

Due to lack of standing, the First, Second, Third, and Fourth Counts of the Third Amended Complaint are dismissed. In addition, the Count Five § 10(b) and Rule 10b-5 bond claims are excluded from among the class claims. Similarly, all claims against Defendants Wachovia Securities, Merrill Lynch, Gen Re, Ferguson, Houldsworth, and Napier are excluded from among the class claims.

b. New Class Definition

In their Motion for Class Certification, Lead Plaintiffs provided a proposed class definition as:

All persons who purchased or otherwise acquired AIG securities including equity, fixed income and all other securities during the class period, [ October 28, 1999 to April 1, 2005] ... and who were damaged thereby.

(Lead Plts' Mem. of Law, 6) (quoting TAC, ¶ ¶ 149-50.) For the foregoing reasons, the proposed class definition is modified to exclude all bondholders, as well as all in-and-out-traders who did not hold AIG stock over at least one of the days of October 14, 2004, October 15, 2004, March 17, 2005, or April 1, 2005. Accordingly, the new class shall consist of:

All shareholders who purchased or otherwise acquired AIG equity shares during the Class Period of October 28, 1999 to April 1, 2005, and who possessed that stock over one or more of the dates of October 14, 2004, October 15, 2004, March 17, 2005 or April 1, 2005, as well as all persons and entities who held the common stock of HSB Group at the time HSB Group was acquired by AIG in a stock for stock transaction (the " HSB Group Subclass" ), and all persons and entities who held the common stock of American General at the time American General was acquired by AIG in a stock for stock transaction (the " American General Subclass" ), and were damaged thereby.

Further, as defined at paragraph 152 of the Third Amended Complaint, " members of the immediate families of the Individual Defendants, any parent, subsidiary, affiliate, officer, or director of the Defendant AIG, any entity in which any excluded person has a controlling interest, and the legal representatives, heirs, successors and assigns of any excluded person" are also excluded from the class.

2. Class Counsel

For the reasons explained in Section II.B.2.d, supra, the Court finds that the law firms of Labaton Sucharow and Hahn Loeser will fairly and adequately represent the interests of the class and hereby appoints them as Co-lead Counsel for the class as certified.

III. CONCLUSION

Lead Plaintiffs' Motion for Class Certification, as defined by the Court, supra, is GRANTED.

SO ORDERED.

the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.

(Lead Pits' Mem. of Law, 4) (quoting Rule 23(b)(3)).

Q: In analyzing whether a market for a security responds quickly to new information, is it your usual practice to perform tests of statistical significance?

A: Yes, as part of the analysis.

Q: Well, do you think your analysis would be weaker if you'd not performed tests of statistical significance?

A: I think it would be less complete if one did not at least try to perform statistical analysis.

Q: Well, why then did you deviate from your normal practice and in this case decide not to examine the statistical significance of the price movement in the-two debt securities you looked at?

A: We did do that analysis.

Q: You did a statistical significant-test of statistical significance in that analysis?

A: Yes, we did.

Q: All right. We'll get to that.

A: It's not in the report, but we did it.

Q: You did it and didn't disclose it?

A: We did it. We did not include it in the report.

Q: Why did you not include it in the report?

A: We discussed it with counsel and we decided not to put it in the report.

Q: Did it support anybody's particular position in the case? What-stated otherwise, what were the results?

A: The results generally indicated that the statistical reactions were not statistically significant at conventional levels.

Q: And when you say " conventional levels?"

A: 10 percent or better.

(Finnerty Dep. Tr., 73:12-75:9.)