In re Initial Public Offering Securities Litig

United States District Court, S.D. New YorkOct 13, 2004
227 F.R.D. 65 (S.D.N.Y. 2004)

Nos. 21 MC 92(SAS), 01 Civ. 3857(SAS), 01 Civ. 8408(SAS), 01 Civ. 7048 (SAS), 01 Civ. 9417(SAS), 01 Civ. 6001(SAS), 01 Civ. 0242(SAS).

October 13, 2004.

Melvyn I. Weiss, Robert A. Wallner, David A.P. Brower, Ariana J. Tadler, Milberg Weiss Bershad Schulman LLP, New York City, Stanley Bernstein, Rebecca M. Katz, Bernstein Liebhard Lifshitz, LLP, New York City, Liaison Counsel for Plaintiffs.

Gandolfo V. DiBlasi, Penny Shane, Sullivan Cromwell, New York City, Liaison Counsel for Defendants (Underwriters).

Jack C. Auspitz, Morrison Foerster LLP, New York City, Liaison Counsel for Defendants (Issuers).


SCHEINDLIN, District Judge.West Page 71


Between January 11 and December 6, 2001, thousands of investors filed class action lawsuits, alleging that 55 underwriters, 310 issuers and hundreds of individuals associated with those issuers had engaged in a sophisticated scheme to defraud the investing public. In brief, the scheme consisted of a requirement, imposed by the underwriters, that IPO allocants purchase shares in the aftermarket, often at escalating prices, and pay undisclosed compensation. In addition, the underwriters prepared analyst reports that contained inaccurate information and recommendations because the analysts operated under a conflict of interest. As a result of the scheme, plaintiffs allege that they collectively lost billions of dollars. These actions were consolidated before this Court for pre-trial supervision. At the suggestion of the Court, the parties selected six cases to be used as test cases for determining whether these suits can proceed as class actions. Upon plaintiffs' motion, I now address whether these test cases can be certified as class actions.

Defendants have submitted thousands of pages of briefs, affidavits, exhibits and reports in opposition to the motion. Although they raise every conceivable argument, their major contention is that individual issues predominate over common issues with respect to almost every aspect of proof. In particular, defendants note that each plaintiff differs with respect to her knowledge of the alleged scheme when she invested ( e.g., whether she was an allocant or an aftermarket purchaser or both and whether and to what extent she was exposed to press reports and other public disclosures); the nature of her investment ( e.g., whether she was a long term investor, a short seller, a day trader, or a momentum trader); the timing of her investment ( e.g., the purchase price of the stock and the effect of any artificial inflation at the time of purchase); the amount of her damages ( e.g., the subsequent dissipation of any artificial inflation by the time of sale); and the traceability of her shares to a particular offering and registration statement. Because of these differences, defendants argue, common issues cannot predominate, and class certification must be denied. Defendants also contend that it would be impossible to ascertain which investors should be in the class and which must be excluded.

In their zeal to defeat the motion for class certification, defendants have launched such a broad attack that accepting their arguments would sound the death knell of securities class actions. Yet class-wide adjudication under Rule 23 of the Federal Rules of Civil Procedure is particularly well-suited to securities fraud cases. In opposing certification, defendants do not truly seek separate adjudications of each individual claim. In reality, they seek no adjudication because the prospect of 310 million individual lawsuits (based on a hypothetical average class membership of one million investors), represents an impossible burden for all parties — the individual plaintiffs, the defendants and the courts. Thus, if certification is denied, defendants will have essentially defeated the claims without ever having been compelled to defend the suits on the merits. Of course, if plaintiffs fail to satisfy the stringent requirements of Rule 23, then a class cannot be certified, even if that results in plaintiffs' inability to press their claims.

See Fed.R.Civ.P. 23(b)(3) Advisory Committee Note (acknowledging that class action is an appealing tool for adjudicating cases of "fraud perpetrated on numerous persons by the use of similar misrepresentations").

See In re Worldcom, Inc. Sec. Litig., 219 F.R.D. 267, 304 (S.D.N.Y. 2003) ("Few individuals could even contemplate proceeding with this litigation in any context other than through their participation in a class action, given the expense and burden that such litigation would entail.").

See In re Methyl Tertiary Butyl Ether ("MTBE") Prods. Liab. Litig., 209 F.R.D. 323, 353 (S.D.N.Y. 2002) (denying class certification).

"The class action device was designed to promote judicial efficiency and to provide aggrieved persons a remedy when individual litigation is economically unrealistic, as well as to protect the interests of absentee class members." This underlying purpose of Rule 23 provides much-needed guidance in focusing on the real issues. While highly competent counsel, with unlimited resources, have the capability to advance an almost unlimited array of complex arguments against certification, the Court must not lose sight of the ultimate question: whether class adjudication of the issues raised in these complaints is clearly superior to any other form of dispute resolution. Although defendants' arguments have raised a number of thorny problems, forcing this Court to take a hard look at the pleadings and the many submissions made in support of and in opposition to this motion, the balance tips strongly in favor of certification. Trying these cases will be an arduous task, but that is no reason to close the courthouse door to the alleged victims of a sophisticated and widespread fraudulent scheme. Accordingly, for the reasons set forth below, class certification, to the extent noted, is granted in each of the six focus cases.



A. The Alleged Scheme

Plaintiffs seek recovery for securities fraud pursuant to the Securities Act of 1933 (the "Securities Act") and the Securities Exchange Act of 1934 (the "Exchange Act"). Plaintiffs allege that defendants engaged in a comprehensive scheme to defraud investors by artificially inflating the prices of the issuers' stocks. The alleged scheme is described at length in my February 19, 2003 Opinion denying defendants' motion to dismiss. Familiarity with that Opinion is assumed.

See In re Initial Public Offering Sec. Litig. ("In re IPO"), 241 F. Supp. 2d 281 (S.D.N.Y. 2003).

B. The Focus Cases

These proceedings sweep together for pre-trial management 310 consolidated class actions, each with a distinct group of defendants (many of whom are overlapping) but alleging the same scheme to defraud investors. Because the question of whether a class can be certified under the rigorous standard set forth by Rule 23 is common to all of these consolidated actions, judicial efficiency counsels in favor of a test case approach. Accordingly, the parties have presented for the Court's consideration six cases, involving the following issuers: Corvis Corp. ("Corvis"); Engage Technologies, Inc. ("Engage"); Firepond, Inc. ("Firepond"); iXL Enterprises, Inc. ("iXL"); Sycamore Networks, Inc. ("Sycamore"); and VA Software Corp, formerly known as VA Linux Systems, Inc. ("VA Linux") (collectively, the "focus cases").

Plaintiffs selected Corvis, Engage, Firepond, Sycamore and VA Linux. See 10/14/03 Letter to the Court from Melvyn I. Weiss, liaison counsel for plaintiffs. Defendants selected iXL. See 11/26/03 Letter to Weiss from Mark Holland, counsel for defendant Merrill Lynch.

The parties have agreed that "[t]he rulings on the class certification motions in the selected cases will govern those cases only." However, most of the issues this Opinion addresses would undoubtedly be raised in a motion for class certification with respect to the remaining 304 consolidated actions. This Opinion is intended to provide strong guidance, if not dispositive effect, to all parties when considering class certification in the remaining actions.

7/11/03 Case Management Term Sheet, Part II.D.

See Transcript of 6/17/04 Hearing at 3:1-6 ("THE COURT: [O]nce there is a decision on these motions, . . . we can talk about what common issues . . . would only reappear in every other class certification motion, and what unique issues there may be in the remaining action.").

C. The Parties' Submissions

On September 2, 2003, plaintiffs moved for class certification and submitted Plaintiffs' Memorandum of Law in Support of Their Omnibus Motion for Class Certification ("Plaintiffs' Omnibus Mem."). Defendants responded with six opposition briefs: the Underwriter Defendants' Memorandum in Opposition to Plaintiffs' Motion for Class Certification in Corvis ("Corvis Mem."); the Memorandum in Opposition to Omnibus Motion for Class Certification, and to Certification of the Proposed Class in Engage Technologies, Inc. ("Engage Mem."); the Underwriter Defendants' Memorandum of Law in Opposition to Plaintiffs' Motion for Class Certification in Firepond ("Firepond Mem."); the iXL Underwriter Defendants' Memorandum in Opposition to Plaintiffs' Motion for Class Certification ("iXL Mem."); the Underwriter Defendants' Opposition to Plaintiffs' Motion for Class Certification in Sycamore ("Sycamore Mem."); and Credit Suisse First Boston ("CSFB") LLC's Memorandum in Opposition to Plaintiffs' Motion for Class Certification in VA Linux ("VA Linux Mem."). Plaintiffs replied on April 19, 2004, with Plaintiffs' Corrected Reply Memorandum of Law in Support of Their Omnibus Motion for Class Certification ("Plaintiffs' Reply"). Defendants responded on May 10, 2004, with the Underwriter Defendants' Sur-Reply Memorandum in Opposition to Plaintiffs' Motion for Class Certification ("Defendants' Sur-Reply"), and plaintiffs submitted Plaintiffs' Response to Underwriter Defendants' Sur-Reply Memorandum in Opposition to Class Certification ("Plaintiffs' Response") on May 19, 2004.

The iXL Mem. is dated February 23, 2004. The Corvis, Engage, Firepond and VA Linux memoranda are all dated February 24, 2003. The corrected Sycamore Mem. is dated April 7, 2004.

After oral argument on June 17, 2004, I directed plaintiffs to submit a letter brief refining their proposed class definition, which plaintiffs submitted on July 6, 2004 ("Class Def. Letter"). Defendants responded to the proposed definition on July 20, 2004, in a letter brief of their own ("Class Def. Opp."). Finally, on September 7, 2004, I ordered plaintiffs to submit a proposed trial plan, which plaintiffs submitted on September 15, 2004 ("Trial Plan"). Defendants opposed the Trial Plan in a letter brief dated September 22, 2004, and plaintiffs replied on September 28, 2004.

The parties have also submitted many expert reports regarding the hotly contested issues of loss causation and damages. Plaintiffs submitted an expert report by Professor Daniel Fischel on January 20, 2004 ("1/20/04 Fischel Report"). Defendants countered with reports by the following experts: Dr. Christopher B. Barry in support of iXL Mem. ("Barry Report"); Dr. Paul A. Gompers in support of Sycamore Mem. ("Gompers Report"); Dr. Allan W. Kleidon in support of Sycamore Mem. ("Kleidon Report"); Dr Maureen O'Hara in support of Corvis Mem. and VA Linux Mem. ("2/23/04 O'Hara Report"); Dr. Erik R. Sirri in support of iXL Mem. ("Sirri Report"); and Dr. René M. Stultz in support of Firepond Mem. ("Stultz Report"). Plaintiffs submitted a rebuttal report by Professor Fischel dated April 15, 2004 ("4/15/04 Fischel Report"). Defendants countered with a report by Dr. Bradford Cornell, dated May 10, 2004 ("Cornell Report"). In my June 21, 2004 Order, I directed plaintiffs to "submit a supplemental report from Fischel in which he analyzes the causal link between the alleged tie-in agreements and their effect on stock price in light of all tie-in purchases in the six focus cases known to plaintiffs' counsel (both in the form of aftermarket trades and pre-opening bids-and-asks). Fischel is advised to pay particular attention to the duration of any inflationary effect caused by this activity." Plaintiffs submitted a final Fischel report on July 12, 2004 ("7/12/04 Fischel Report"), and defendants countered with a report from Dr. O'Hara dated July 23, 2004 (the "7/23/04 O'Hara Report").

The Barry Report, 2/23/04 O'Hara Report and Sirri Report are all dated February 23, 2004. The Gompers Report, Kleidon Report and Stultz Report are all dated February 24, 2004.

In re IPO, No. 21 MC 92, 2004 WL 1635575, at *1 (S.D.N.Y. June 23, 2004) (emphasis in original).

D. The Proposed Class Periods

For each of these consolidated actions, "[t]he Class consists of all persons and entities that purchased or otherwise acquired the securities of [Specific Issuer] during the Class Period and were damaged thereby," subject to various exclusions. Plaintiffs propose class periods for each case that span the period between the initial public offering ("IPO") and December 6, 2000. For the purposes of this Opinion, plaintiffs' proposed class periods are adopted for plaintiffs' Exchange Act claims; however, plaintiffs' proposed class periods must be shortened with respect to plaintiffs' claims pursuant to section 11 of the Securities Act in each of the six focus cases.

Class Def. Letter at 1. The exclusions from this class definition are discussed in detail in Part IV.A.4., infra.

See 4/19/02 Consolidated Amended Class Action Complaint For Violations of the Federal Securities Laws for Corvis ¶ 58; 4/19/02 Consolidated Amended Class Action Complaint For Violations of the Federal Securities Laws for Engage ¶ 55; 4/19/02 Consolidated Amended Class Action Complaint For Violations of the Federal Securities Laws for Firepond ¶ 53; 4/19/02 Amended Class Action Complaint For Violations of the Federal Securities Laws for iXL ¶ 64; 4/19/02 Consolidated Amended Class Action Complaint For Violations of the Federal Securities Laws for Sycamore ¶ 68; 4/19/02 Consolidated Amended Class Action Complaint For Violations of the Federal Securities Laws for VA Linux ¶ 52.

See infra Part IV.B.4.

E. Focus Case-Specific Facts

a. The Corvis IPO

1. Corvis

Corvis held its IPO on July 28, 2000, with CSFB serving as the lead underwriter, offering 31,625,000 shares at $36.00 per share. Defendants note that "[p]rior to the IPO, Corvis had issued a significant number of unregistered shares . . . which would have been freely tradeable at the time of the IPO, provided that the shareholders satisfied SEC Rule 144." Corvis reported a number of outstanding unregistered shares of stock that had been issued to other companies and to Corvis affiliates before the Corvis IPO. On the first day of trading, the stock opened at $74.00, peaked at $98.00 and closed at $84.72, a 135% increase over the offering price. By the end of the first day of trading, 28,137,100 shares had changed hands in 35,755 transactions.

See Corvis Corp. IPO Facts, Ex. A to 2/24/04 Declaration of Fraser L. Hunter, Jr. in Support of Corvis Opposition ("Hunter Corvis Decl.").

Corvis Mem. at 20 n. 19.

See 7/27/00 Form S-1/A filed by Corvis ("Corvis Form S-1/A"), Ex. N to Hunter Corvis Decl. at II-2-5.

See id.

See Corvis Corp. (CORV) Market-Wide Trading Data for Day One, Ex. D to Hunter Corvis Decl.

Plaintiffs allege that 195 of the institutional allocants in the Corvis IPO, to whom 12,193,450 shares were allocated, entered into tie-in agreements with the allocating underwriter. Plaintiffs further allege that purchase orders from these allocants accounted for 1,469,600 of the 2,569,600 purchase orders placed during the pre-open bid session, during which the opening share price rose to more than twice the $36.00 offering price. During the ten business days from July 28 through August 10, 2000, Corvis allocants purchased a total of 11,582,004 shares in the aftermarket. The same investors sold a total of 1,543,240 shares during that time.

See Summary of Alleged Tie-In Agreements ("Fischel Tie-In Summary"), Ex. A to 7/12/04 Fischel Report.

See Purchase Orders for the Focus Case Stocks in the Pre-Open Bid Session ("Fischel Purchase Order Summary"), Ex. B to 7/12/04 Fischel Report.

See Aftermarket Trading in Corvis Corp. by CSFB Allocated Accounts, Ex. D-1 to 7/12/04 Fischel Report.

Following the IPO, Corvis's stock climbed to its highest price, $108.06 per share, on August 4, 2000, the same day Broadwing disclosed a $44,000,000 investment position in Corvis and announced that it would buy $200,000,000 in equipment. By the end of September 2000, the stock had fallen to just over $60.00 per share. On October 2, 2000, Corvis filed a prospectus for 2,446,074 newly registered shares acquired through the exercise of employee stock options. Corvis explicitly incorporated the disclosures made in its IPO prospectus into its October 2, 2000 prospectus. During November 2000, the stock slid from $64.00 to $28.81 per share. Corvis experienced a slight rebound in December 2000, reaching $40.38 per share on December 6. According to Professor Fischel, Corvis underperformed when compared to various market benchmarks by 27 to 64 percentage points from July 28 to December 6, 2000, and by 35 to 67 points thereafter. On May 10, 2001, when the first Corvis case, PRFT Partners v. Corvis Corp., No. 01 Civ. 3994, was filed, shares of Corvis closed at $7.380 per share. b. Corvis Class Representatives

See Corvis Corp. Chronology of Events from July 28, 2000 to December 6, 2000 ("Corvis Chronology"), Ex. F to Hunter Corvis Decl., at 1.

See id. at 3.

See 10/2/00 Form S-8 for Corvis ("October 2, 2000 Prospectus"), Ex. N to Hunter Corvis Decl.

See id. at II-1.

See Corvis Chronology at 5-6.

See id. at 6.

See 1/20/04 Fischel Report ¶¶ 26, 29. "Underperformed" and "overperformed," as the terms are used by Fischel, mean that stock prices either declined or rose relative to the price movements of various benchmark indices. See 1/20/04 Fischel Report ¶¶ 23-26. For example, if the benchmark suggests that the market lost 50% of its value during the period, and a stock lost 80% of its value, then the stock is said to have "underperformed" the benchmark by 30 percentage points. Conversely, if another stock lost only 20% of its value over that period, it "overperformed" by 30 percentage points. As discussed in Part IV.B.2., infra, plaintiffs assert that price underperformance after December 6, 2000 shows that price dissipation continued throughout and after the class period.

See NASDAQ: Charts, quote.dll?page=chartingmode=basicsintraday=offtimeframe =4ycharttype=ohlcsplits=offearnings=offmovingaverage =Nonelowerstudy=volumecom parison=offindex=drilldown =offsymbol=CORVselected=CORV (Aug. 20, 2004).

The Private Securities Litigation Reform Act of 1995 ("PSLRA"), Pub.L. No. 104-67, 109 Stat. 737 (1995), amended the Securities Act and Exchange Act to require that "[e]ach plaintiff seeking to serve as a representative party" in a class action under those sections submit a certification containing information about the proposed representative's claim and capacity to serve. See 15 U.S.C. § 77z-1(a)(2)(A); 15 U.S.C. § 78u-4(a)(2)(A). There are discrepancies between the transactions reported by the proposed class representatives in their PSLRA Lead Plaintiff Certifications and the transactions listed by defendants in their submission entitled Proposed Class Representative Transactions in Corvis During 92 Trading Days of the Proposed Class Period ("Corvis Transactions Data"), Ex. K to Hunter Corvis Decl. For example, Michael Huff lists only five purchases of Corvis stock while the Corvis Transactions Data reveals fifteen purchases. Compare 6/26/01 Huff PSLRA Lead Plaintiff Certification, Ex. K to Hunter Corvis Decl., with Corvis Transactions Data. For purposes of this Opinion, I rely on the Corvis Transactions Data.

(1) Satswana Basu

Satswana Basu, who also seeks to act as a class representative in six other IPO cases, purchased and sold 95,198 Corvis shares, and sold and covered short 20,000 Corvis shares between November 17 and December 6, 2000, resulting in a $736,869.20 loss during that time. Basu also purchased Corvis shares after December 6, 2000.

See Corvis Transactions Data.

(2) Michael Huff

Between August 24 and September 26, 2000, Michael Huff bought and sold 12,000 shares of Corvis stock for a $22,755.00 profit. On September 28 and 29, 2000, however, Huff purchased a total of 6,000 shares for $472,832.50, which he had not sold as of December 6, 2000. Had he sold the shares that day, when the stock closed at $40.38 per share, Huff would have suffered a total pre-December 6, 2000 loss of $207,797.50.

See id. Unrealized losses may serve as the basis for a securities fraud claim. See Federman v. Empire Fire and Marine Ins. Co, 597 F.2d 798, 801-02 n. 1 (2d Cir. 1979) (acknowledging that where "[t]he amended complaint alleged that . . . [o]n the stock retained, Federman sustained an unrealized loss of approximately $900.00, . . . [t]here is no question . . . that the plaintiffs complied with the purchaser-seller standing requirement").

(3) Sean Rooney

Sean Rooney purchased 1,000 shares of Corvis stock on August 7, 2000 at $107.50 per share and another 500 shares on August 11, 2000 at $90.00 per share. Rooney sold 500 shares on December 5, 2000, at $39.00 per share, leaving him with 1,000 unsold shares on December 6, 2000. Had he sold his shares that day, when the stock closed at $40.38 per share, he would have suffered a total pre-December 6, 2000 loss of $92,620.00. Rooney also received an allocation of 500 shares in the IPO.

See Corvis Transactions Data.

See 11/25/03 Deposition of Sean P. Rooney, Ex. M to Hunter Corvis Decl., at 77:25-78:10.

2. Engage

a. The Engage IPO

Engage held its IPO on July 20, 1999, with Goldman Sachs acting as lead underwriter, offering 6,938,000 shares at $15.00 per share. The IPO prospectus for Engage notes that 1,225,324 shares of common stock were already outstanding well before the IPO, on April 30, 1999. On the first day of trading, the stock opened at $28.00, peaked at $47.00 and closed at $41.00, a 173% increase over its offering price. By the end of the first day of trading, 14,887,200 shares had changed hands.

See 7/19/99 Engage Prospectus, Ex. B to 2/24/04 Declaration of Gandolfo V. DiBlasi ("DiBlasi Decl."), at 5 n. 1.

See id. The prospectus also notes that, under Rule 144 promulgated pursuant to the Securities Act, 17 C.F.R. § 230.144, beneficial owners of pre-IPO Engage stock became eligible to trade their shares on the open market 90 days after the IPO — i.e., on October 18, 1999. See id.

See CBS MarketWatch: Historical Quote, =7%2F20%2F99symb=ENGAsiteid=mktw (Sept. 3, 2004).

Plaintiffs allege that forty-nine of the institutional allocants in the Engage IPO, to whom 786,900 shares were allocated, entered into tie-in agreements with the allocating underwriter. Plaintiffs further allege that purchase orders from these allocants made up 693,000 of the 1,251,000 total purchase orders placed during the pre-open bidding session, during which the opening price for Engage was set at $28.00, $13.00 above the offering price. During the ten business days from July 20, 1999 through August 2, 1999, Engage allocants purchased a total of 3,313,660 shares in the aftermarket. The same investors sold a total of 135,850 shares during that time.

See Fischel Tie-In Summary.

See Fischel Purchase Order Summary.

See id.

See Aftermarket Trading in Engage Technologies, Inc. by Goldman Sachs Allocated Accounts July 20, 1999 — August 2, 1999, Ex. D-2 to 7/12/04 Fischel Report.

Following the IPO, the price for Engage stock fell, but the stock traded consistently in the mid-twenties through the end of 1999. On January 16, 2000, approximately 6,700,000 non-IPO shares associated with "employee stock options, corporate acquisitions, and other transactions became freely tradable in large numbers in the secondary market." From January to February 2000, Engage prices climbed to all-time highs, peaking at over $180.00 per share. The price declined rapidly in March and April of 2000, and the stock split two for one on April 4, 2000. By August 2000, the stock was trading around the offering price of $15.00 per share when adjusted to reflect the split. The price dropped below the offering price for the first time in October 2000, and continued to decline through December 6, 2000, when it was trading at a split-adjusted price of approximately $3.19 per share. Fischel asserts that Engage underperformed when compared to various market benchmarks by 35 to 72 percentage points from July 20, 1999 to December 6, 2000, and by 34 to 68 points thereafter. On September 7, 2001, when the first Engage case, Chin v. Engage Tech., Inc., No. 01 Civ. 8404, was filed, Engage stock closed at $0.190 per share.

See Engage Mem. at 5.

Id. at 26. See also DiBlasi Decl. ¶ 11.

See Engage Mem. at 5. See also Engage Price Chart, Ex. C to DiBlasi Decl.

See Summary of Trading by Kasbarian ("Kasbarian Trading Summary"), Ex. C to 2/24/04 Declaration of David M.J. Rein ("Rein Decl."), at 2.

See Engage Price Chart.

See 1/20/04 Fischel Report ¶¶ 26, 29.

See NASDAQ: Charts, chartingmode=basicsintraday=offtimeframe=4ycharttype =ohlcsplits=offearnings=offmovingaverage=Nonelowerstudy =volumecomparison=offin dex=drilldown=offsymbol =ENGAselected=ENGA (August 20, 2004).

b. Engage Class Representatives

(1) Stathis Pappas

Stathis Pappas was an allocant in Engage's IPO, receiving 100 shares of Engage stock on July 20, 1999, at $15.00 per share. Pappas made no aftermarket purchases in Engage. It does not appear from the facts before this Court that Pappas ever sold any of those shares, which were worth approximately $3.19 per share on December 6, 2000. Had he sold his shares that day, he would have suffered a $1,181.26 loss on the transaction. According to defendants, Pappas is a member of more than fifteen potential classes in these consolidated actions.

See Pappas PSLRA Lead Plaintiff Certification, Ex. F to Rein Decl.

See 12/9/03 Deposition of Stathis Pappas ("Pappas Dep."), Ex. B to Rein Decl., at 214:4-7 ("[I]f I sold the 30 cent stock or whatever it is today, I haven't looked at it because it depresses me.").

See Engage Mem. at 31.

(2) Krikor Kasbarian

Between March 31 and August 29, 2000, Krikor Kasbarian purchased 20,000 shares of Engage stock for $1,008,687.50 and sold 30,000 post-split shares for $411,450.00, resulting in a total pre-December 6, 2000 loss of $597,237.50. On his October 8, 2001 PSLRA certification, Kasbarian failed to disclose several transactions. In October, 2000, Kasbarian destroyed records documenting his Engage trades. He held no Engage stock on December 6, 2000.

See Kasbarian Trading Summary at 2.

See Engage Mem. at 35 ("His lead plaintiff certifications omitted many of his trades in Engage as well as other securities on which he is suing in this litigation, and his sworn explanations for those omissions are inconsistent."); see also Comparison of PSLRA Certifications of Krikor Kasbarian to His Trading Records, Ex. E to Rein Decl. For example, Kasbarian failed to disclose that on August 25, 2000 he purchased 10,000 shares for $113,687.50, which he sold on August 29, 2000 for $117,562.50, resulting in a profit of $3,875.00 on the transaction.

See 12/10/03 Deposition of Krikor Kasbarian, Ex. A to Rein Decl., at 110:21-111:9.

See Kasbarian Trading Summary at 2.

Kasbarian engaged in six transactions involving Engage after December 6, 2000. In July 2001, Kasbarian bought Engage twice and sold twice, each sale within a few days of the respective purchase. Kasbarian made a profit of $.06 per share on the first set of trades and broke even on the second set. Kasbarian failed to disclose these trades on his PSLRA certification. Following submission of the certification, Kasbarian made a final pair of trades in Engage stock, purchasing 4,000 shares on October 25, 2001, and selling those shares for a $.03 profit per share on October 30, 2001. According to defendants, Kasbarian is also a member of more than fifteen potential classes in these consolidated actions and seeks to serve as a class representative in the litigation.

See id.

See 12/10/03 PSLRA Lead Plaintiff Certification for Kasbarian, Ex. D to Rein Decl., at 1.

See Kasbarian Trading Summary at 2.

See Engage Mem. at 31.

3. Firepond

a. The Firepond IPO

Firepond held its IPO on February 4, 2000, with Robertson Stephens acting as lead underwriter, offering approximately 5,000,000 shares at $22.00 per share. On that date, 27,751,713 unregistered shares were already outstanding. On the first day of trading, Firepond's stock opened at $52.00, peaked at $102.31 and closed at $100.25, an increase of 356% over its offering price. By the end of the first day of trading, 9,284,900 shares had changed hands.

See 2/4/00 Firepond Prospectus, Ex. A to 2/24/04 Declaration of Brendan J. Dowd ("Dowd Decl."), at 1.

See id. at 58. With respect to the number of unregistered shares eligible for trading after the IPO, the Firepond Prospectus bizarrely states that "4,486,242 shares will be available for resale in the public market in reliance on Rule 144(k) immediately following this offering, of which 4,552,074 shares are subject to lock-up agreements." Id. This remarkable calculation leaves considerably fewer than zero unregistered shares eligible for trading on the date of the IPO. However, the Prospectus also notes that, under Rule 144, thousands of shares would become tradeable 90 days after the IPO — i.e., on May 4, 2000. See id.

See CBS MarketWatch: Historical Quote, =FIREsiteid=mktw (Sept. 3, 2004).

Plaintiffs allege that 152 of the institutional allocants in the Firepond IPO, to whom 3,153,100 shares were allocated, entered into tie-in agreements with the allocating underwriter. Plaintiffs further allege that purchase orders from these allocants made up 3,965,100 of the 4,125,100 million total purchase orders placed during the pre-open bidding session, during which the opening price for Firepond rose to $30.00 above the offering price. During the ten business days from February 4 through February 17, 2000, Firepond allocants purchased a total of 13,970,988 shares in the aftermarket. The same investors sold a total of 12,611,116 shares during that time.

See Fischel Tie-In Summary.

See Fischel Purchase Order Summary.

See Aftermarket Trading in Firepond, Inc. by Robertson Stephens Allocated Accounts, Ex. D-3 to 7/12/04 Fischel Report.

Following the IPO, Firepond's stock fell slightly, closing at $71.00 on February 17, 2000. The price then rebounded, reaching a high of $97.44 on February 29, only to rapidly decline to $15.88 by April 17. Defendants claim that the number of outstanding shares increased by nearly 2,000,000 between the IPO and April 30, 2000. The stock once again rebounded, peaking at $40.69 on June 29, 2000, but soon resumed its decline, trading in the upper teens again by August 1. On August 2, 2000, Firepond's 180-day lock-up expired, and more than 26,000,000 shares became tradeable. The price of the stock began to fall dramatically in the fall of 2000, declining from over $17.00 a share on September 14 to a closing price of $6.69 on December 6, 2000. Fischel asserts that Firepond underperformed when compared to various market benchmarks by 27 to 63 percentage points from February 4, 2000 to December 6, 2000, and 33 to 65 points thereafter. On July 31, 2001, when the first Firepond case, Barrett v. FirePond, Inc., No. 01 Civ. 7048, was filed, Firepond closed at $0.66.

See Firepond, Inc. Closing Prices 2000, Ex. B to Dowd Decl.

See Firepond Mem. at 38.

See Firepond, Inc. Closing Prices 2000.

See 2/4/00 Firepond Prospectus at 53.

See Firepond, Inc. Closing Prices 2000.

See 1/20/04 Fischel Report ¶¶ 26, 29.

See Bigcharts — Historical Quotes, =1symbol=fireclose_date=7%2F31%2F01x=0y=0 (August 20, 2004).

b. Firepond Class Representatives

(1) Zitto Investments

Zitto Investments ("Zitto") purchased 345 shares of Firepond stock between February and April of 2000. It sold 100 shares in February for a profit but sold 245 shares in August 2000 for less than $20.00 per share, resulting in a net loss of $7,258.75.

See Class Representative Trading Summary, App. B to Firepond Mem, at 1.

(2) James and Diane Collins

James and Diane Collins ("the Collinses") made four purchases of Firepond stock between March 14 and May 1, 2000, totaling 500 shares for $24,300.00. The Collinses had not sold any shares of Firepond stock prior to December 6, 2000, when the stock closed at $6.69 per share. Had the Collinses sold their shares that day, they would have suffered a $20,955.00 loss on the transaction.

See id.

(3) Joseph Zhen

Between February 11 and March 17, 2000, Joseph Zhen purchased 2,600 Firepond shares for $192,956.25 and sold 1,600 shares for $126,725.00, resulting in a gain of $7,982.69. Zhen still held 1,000 shares when the price dropped dramatically in late March. He sold his remaining shares on April 17, 2000 for $16,093.80, bringing his total pre-December 6, 2000 loss to $50,137.45. Zhen omitted thirteen of his seventeen Firepond trades, some of which resulted in profits, from his September 20, 2001 PSLRA certification. During discovery, Zhen failed to produce trading records for transactions in securities other than Firepond.

See id.

See Firepond Mem. at 36.

See 12/5/03 Deposition of Joseph Zhen, Ex. D to Dowd Decl., at 96:15-97:10.

4. iXL

a. The iXL IPO

iXL held its IPO on June 2, 1999, with Merrill Lynch serving as lead underwriter, offering close to 7,000,000 shares at $12.00 per share. The iXL IPO Prospectus notes that "no restricted securities will be eligible for immediate sale on the date of this prospectus," and that "121,828 restricted securities issuable pursuant to stock options will be eligible for sale 90 days after the date of this prospectus [on August 31, 1999.]" During the first day of trading, the stock opened at $15.13, peaked at $24.50 and closed at $17.88, an increase of 49% above the offering price. On the first day of trading, 14,008,117 shares changed hands.

See 6/2/99 iXL Prospectus, Ex. E to 2/24/04 Declaration of Robert G. Houck ("Houck Decl."), at E2.

See 6/2/99 iXL IPO Prospectus at 97.

See iXL Enterprises Daily Stock Prices and Volume, Ex. I to Houck Decl., at 12.

Plaintiffs allege that thirty-seven of the institutional allocants in the iXL IPO, to whom 1,222,750 shares were allocated, entered into tie-in agreements with the allocating underwriter. During the ten business days from June 3 through June 16, 1999, iXL allocants purchased a total of 3,080,089 shares in the aftermarket. The same investors sold a total of 2,956,325 shares during that time.

See Fischel Tie-In Summary. Fischel states that he was not provided with any pre-opening bid information for iXL. See 7/12/04 Fischel Report ¶ 6.

See Aftermarket Trading in iXL Enterprises, Inc. by Merrill Lynch Allocated Accounts, Ex. D-4 to 7/12/04 Fischel Report.

In the weeks following the iXL IPO, the share price rose slightly, closing at $19.13 on June 28, 1999. That day, 4,000,000 shares were registered with the SEC for use in future acquisitions. A month later, on July 27, 1999, iXL issued a positive earnings announcement, and Merrill Lynch upgraded the stock from near-term "accumulate/buy" to near-term "buy/buy." The following day, shares closed at $29.13. In August 1999, the price dropped to $22.00. Shares rebounded to $37.00 in mid-November and, on January 20, 2000, the price reached $58.75, the stock's high-water mark.

See iXL Enterprises Daily Stock Prices and Volume at 12.

See 6/28/99 Form S-4 for iXL, Ex. E to Houck Decl., at E24-E26.

See 7/27/99 iXL Press Release: "iXL Enterprises Reports Record Revenue Sequential Quarterly Revenue [sic] for iXL, Inc. Subsidiary Increases 41 Percent," Ex. D to Houck Decl., at D8-D11.

7/28/99 Merrill Lynch Report for iXL, Ex. G to Houck Decl., at G14.

See iXL Enterprises Daily Stock Prices and Volume at 12-14.

In February 2000, more than 50,000,000 shares became tradeable due to the expiration of a lock-up that had taken effect shortly after the IPO. Between mid-February and late June, 2,400,000 shares that had been subject to lock-up agreements were sold. On February 18, 2000, Merrill downgraded iXL to "accumulate" and warned of the risk of sales of unlocked shares. Merrill reclassified its long-term rating to "buy" on March 20, 2000. In September 2000, Merrill downgraded iXL first to "accumulate" and then to "neutral." The price of iXL shares closed at $1.25 on December 6, 2000. Fischel asserts that iXL underperformed when compared to various market benchmarks by 63 to 99 percentage points from June 2, 1999 to December 6, 2000, and by 36 to 62 points thereafter. On October 25, 2001, when the first iXL case, Turner v. iXL Enterprises, Inc., No. 01 Civ. 9417, was filed, iXL closed at $0.32 per share.

See iXL Mem. at 10-11.

See id. at 11.

2/18/00 Merrill Lynch Report for iXL, Ex. G to Houck Decl., at G68.

3/20/00 Merrill Lynch Report for iXL, Ex. G to Houck Decl., at G72.

9/1/00 Merrill Lynch Report for iXL, Ex. G to Houck Decl., at G84.

9/5/00 Merrill Lynch Report for iXL, Ex. G to Houck Decl., at G86.

See iXL Mem. at 11.

See 1/20/04 Fischel Report ¶¶ 26, 29.

See =0time=100freq=1compidx=aaaaa%3A0comp=NO_SYMBOL_CHOSENma=0maval=9uf=01f=11f2=01f3=0type=64style=320size=3sid= 149463o_symb=DE%3A922795startdate=12%2F6%2F00enddate=10%2F2 5%2F2001show=symb=DE%3A922795draw.x=43draw.y=14default=tru ebackurl=%2Fintchart%2Fframes%2Fframes%2Easpprms=qcd (Oct. 11, 2004).

b. iXL Class Representatives

(1) John Miles

Between August 19, 1999 and November 2, 2000, John Miles purchased 16,400 iXL shares for $138,233.41. Miles sold 2,400 shares by August 16, 2000 for $53,452.04, but still held 14,000 shares as of December 6, 2000. Had he sold his remaining shares that day, when the stock closed at $1.25 per share, he would have suffered a total pre-December 6, 2000 loss of $67,281.37.

See John Miles iXL Trades, Ex. J to Houck Decl., at J18.

(2) John Rowe

Between June 4 and August 31, 1999, John Rowe purchased 1,335 iXL shares for $25,685.23. He sold 335 shares between September 17 and November 22, 1999 for $12,029.82, at a profit, leaving him with 1,000 unsold shares as of December 6, 2000. Had he sold his remaining shares that day, when the stock closed at $1.25 per share, Rowe would have suffered a total pre-December 6, 2000 loss of $12,405.41.

See J. Chris Rowe iXL Trades, Ex. J to Houck Decl., at J29. Rowe actually sold the remaining shares on December 29, 2000 for $938.71. See id.

5. Sycamore

a. The Sycamore IPO

Sycamore held its IPO on October 21, 1999, with Morgan Stanley acting as the co-lead underwriter, offering 7,475,000 shares at $38.00 per share. On the first day of trading, Sycamore shares opened at $270.88, the day's high price, and closed at $184.75, an increase of 386% above the offering price. Almost 10,000,000 shares changed hands on the first day of trading.

See 10/21/99 Sycamore Prospectus, Ex. 6 to 2/24/04 Declaration of Brant W. Bishop ("Bishop Decl.").

See CBS MarketWatch: Historical Quote, mb=SCMRsiteid=mktw (Sept. 3, 2004).

Plaintiffs allege that eighty-seven institutional allocants in the Sycamore IPO, to whom 1,077,625 shares were allocated, entered into tie-in agreements with the allocating underwriter. Plaintiffs further allege that purchase orders from these allocants made up 561,900 of the 2,868,035 total purchase orders placed during the pre-open bidding session, during which the opening price for Sycamore rose to $232.88 above the offering price. During the first ten days following Sycamore's IPO, Sycamore allocants purchased a total of 2,297,115 shares. The same investors sold a total of 745,832 shares during that time.

See Fischel Tie-In Summary.

See Fischel Purchase Order Summary.

See Aftermarket Trading in Sycamore by Morgan Stanley Allocated Accounts, Ex. D-5 to 7/12/04 Fischel Report.

Defendants assert that on the first day of public trading, 5,734,183 previously issued non-IPO Sycamore shares were not subject to lock-up. At least 368,587 of these shares became tradeable 90 days after the Sycamore IPO — i.e., on January 19, 2000. Following the IPO, the price of Sycamore stock climbed steadily, closing at $203.00 on October 26, 1999. On January 19, 2000, millions of additional shares that had been issued prior to the Sycamore IPO were released from lock-up, and by the end of that week, the stock price reached $280.00. The stock split three for one on February 14, 2000. On March 2, 2000, the stock soared to a price of $569.81 per share, and the next day 8,985,186 more shares, issued one year earlier, were released from lock-up. A secondary offering of 10,200,000 Sycamore shares occurred on March 14, 2000. On April 18, 2000, 26,965,355 additional shares were released from lock-up. After rising and falling several times, the stock price eventually declined, trading around $300.00 per share in October of 2000, and closing at a split-adjusted price of $169.31 per share on December 6, 2000. Fischel asserts that Sycamore underperformed when compared to various market benchmarks by 32 to 68 percentage points from October 21, 1999 to December 6, 2000, and by 32 to 64 points thereafter. On July 2, 2001, when the first Sycamore case, Pond Equities v. Sycamore Networks, Inc., No. 01 Civ. 6001, was filed, Sycamore closed at $8.63 per share.

See 10/21/99 Sycamore Prospectus at 53 (explaining that 65,471,542 of the 71,205,725 shares issued and sold by Sycamore were subject to lock-up, leaving 5,734,183 free shares).

See id. at 52. In fact, millions of shares otherwise subject to 180-day lock-up agreements became tradeable on January 19, 2000, because Sycamore shares had traded at more than twice the IPO price, satisfying a release condition of many of the Sycamore lock-up agreements. See id.; Sycamore Mem. at 37; 3/14/00 Secondary Prospectus filed by Sycamore, Ex. 7 to Bishop Decl., at 54 (noting that nearly 30 million restricted shares were eligible for trading on January 19, 2000).

See Sycamore Significant Days Summary, Ex. 13 to Gompers Report, at 2.

See 10/21/99 Sycamore Prospectus at 52 (noting that the expiration of the lock-up would occur 180 days following the 10/21/99 Sycamore IPO).

See Sycamore Significant Days Summary at 3.

See id. at 45. The stock prices in this subsection are adjusted for the 3-1 stock split.

See id. at 8.

See 3/14/00 Sycamore Prospectus, Ex. 7 to Bishop Decl., at 54 (noting that 8,985,186 shares were freed from lock-up on March 3, 2000).

See id.

See id. at 67.

See Sycamore Significant Days Summary at 39-42.

See 1/20/04 Fischel Report ¶¶ 26, 29.

See 7%2F2%2F01symb=SCMRsiteid=mktw (Oct. 11, 2004).

b. Sycamore Class Representatives

(1) Barry Lemberg

Barry Lemberg purchased 100 Sycamore shares for $175.00 per share on March 3 and an additional 100 shares on April 13, 2000 for $71.13 per share. Lemberg had not sold those shares as of December 6, 2000, when the stock closed at $56.44 per share. Had Lemberg sold the shares that day, he would have suffered a $13,325.00 loss on the transaction. While Lemberg alleges that he is entitled to damages relating to 1,200 shares, he purchased only 200 shares before December 6, 2000.

See 7/20/01 PSLRA Lead Plaintiff Certification of Barry Lemberg.

This price reflects the stock's closing price without adjusting for the 3-1 stock split on February 14, 2000.

See Deposition of Barry Lemberg ("Lemberg Dep."), Ex. 8 to Bishop Decl., at 72:6-73:6. Lemberg purchased 1,000 shares on February 15, 2001. See id. at 73:7-10.

Lemberg's testimony and questionnaire responses conflict as to whether he received an IPO allocation, and although Lemberg alleged that he had never personally bought and sold the same stock on the same day, the record reveals that he had conducted a same day transaction on at least one occasion. He testified that he does not remember having any involvement in preparing the Complaint. Moreover, defendants claim that he "does not understand financial markets well," and is "unable to describe the specific behavior of Morgan Stanley that was improper."

Compare 9/12/03 Questionnaire of Barry I. Lemberg ("Lemberg Questionnaire"), Ex. 1 to Ex. 8 to Bishop Decl., at 3 (answering "No" when asked whether he received any allocation of stock relating to the current litigation), with Lemberg Dep. at 42:5-10 (answering "Yes" to the same question).

Compare Lemberg Questionnaire at 3 (answering "No" when asked whether, from 1998 to 2000, he had ever engaged in "buying and selling same stock [sic] on the same trading day"), and Lemberg Dep. at 27:14-16 (same), with Active Assets Account Report for Barry Lemberg, Ex. 2 to Lemberg Dep., at 8 (reporting that Lemberg both bought and sold CacheFlow stock on 11/24/99), and Lemberg Dep. at 32:16-33:11 (acknowledging the discrepancy between Lemberg's previous affirmation and the CacheFlow transaction).

See Lemberg Dep. at 94:17-96:11.

Sycamore Mem. at 49.

(2) Vasanthakumar Gangaiah

Vasanthakumar Gangaiah purchased and sold 11,600 shares of Sycamore stock between March 1 and December 6, 2000, resulting in a $69,276.21 loss. Defendants assert that Gangaiah provided "conflicting and false" testimony about his investment accounts, and "falsely claimed never to have received an IPO allocation." Defendants also question Gangaiah's understanding of the litigation, noting that he does not know what kind of damages are being sought in the case or which people are members of the class.

See 3/01/00 to 12/31/00 E*Trade Account Statements for Vasanthakumar Gangaiah, Ex. 6 to 11/30/03 Deposition of Vasanthakumar Gangaiah ("Gangaiah Dep."), Ex. 9 to Bishop Decl.

Sycamore Mem. at 46. When initially asked when he opened his first investment account in the United States, Gangaiah responded that his first account, an E*Trade account, was opened in 1999. See Gangaiah Dep. at 21:9-14. However, Gangaiah later testified that he had opened a DLJ Direct account in 1998, prior to the E*Trade account, which he claimed had been closed after a couple of months. See id. at 27:1-9. Defendants observe that Gangaiah's documents show that his DLJ Direct account had not been closed in 1998, but was active as late as June 2000. See Sycamore Mem. at 46. Defendants allege that while Gangaiah claimed that he held only the E*Trade and DLJ investment accounts, he actually held another account with BancBoston Robertson Stephens ("BancBoston"). See id.

Sycamore Mem. at 47. Gangaiah testified that he had never received an IPO allocation. See Gangaiah Dep. at 40:10-12. Defendants assert, however, that Gangaiah's BancBoston account was "opened for the express purpose of receiving 750 shares in the IPO (one of the IPOs at issue in this litigation)." Sycamore Mem. at 46-47.

See Sycamore Mem. at 47.

(3) Frederick Henderson

Frederick Henderson received an IPO allocation of 50 shares of Sycamore stock at $38.00 per share, which he flipped on October 22, 1999 for $200.00 per share. Between May 2 and October 13, 2000, Henderson purchased an additional 14,000 shares for $1,550,081.25. He sold 4,000 shares on November 16, 2000 for $266,675.00, for a loss, leaving him with 10,000 unsold shares as of December 6, 2000. Had he sold them that day, when the stock closed at $56.44, Henderson would have suffered a total pre-December 6, 2000 loss of $710,906.25.

See id. at 19, 38.

See 8/14/01 PSLRA Lead Plaintiff Certification of Frederick B. Henderson. This result factors in the $8,100 profit Henderson made when he flipped his 50 allocated shares.

Henderson does not recall seeing the Amended Complaint before it was filed. He acknowledges that he "has no idea how or why December 6, 2000 was selected as the end of the class period, nor . . . whether it should be the end of the class period."

See 12/2/03 Deposition of Frederick Henderson, Ex. 3 to Bishop Decl., at 40:4-11.

Id. at 176:18-177:7.

6. VA Linux

a. The VA Linux IPO

VA Linux held its IPO on December 9, 1999, with CSFB serving as its lead underwriter, offering 5,060,000 shares at $30.00 per share. At that time, VA Linux had already issued 35,301,586 unregistered shares. The VA Linux Prospectus notes that VA Linux's "directors and officers as well as other stockholders and optionholders" had agreed to subject themselves to a 180-day lock-up on their unregistered shares, but does not explicitly state whether each and every one of the 35,301,586 unregistered shares was subject to lock-up. VA Linux registered 24,863,635 additional shares simultaneously with its IPO as part of its various stock option benefit plans. VA Linux's filings for these stock option benefit plans explicitly incorporate the contents of the VA Linux IPO Prospectus and set their prices at $30, the same as the IPO price. On the first day of trading, VA Linux stock opened at $299.00 and peaked at $320.00, with a trading volume of 7,685,600 shares. Shares closed that day at $239.25, an increase of 698% over the offering price.

See 12/9/99 VA Linux Prospectus, Ex. H to 2/24/04 Declaration of Fraser L. Hunter, Jr. in Support of VA Linux Opposition ("Hunter VA Linux Decl.").

See id. at 65.


See id.

See CBS MarketWatch: Historical Price, b=LNUXsiteid=mktw (Sept. 3, 2004).

Plaintiffs allege that 147 of the institutional allocants in the VA Linux IPO, to whom 2,174,850 shares were allocated, entered into tie-in agreements with the allocating underwriter. Plaintiffs further allege that purchase orders from these allocants made up 294,230 of the 426,230 total purchase orders placed during the pre-open bid session, when the opening price was set at $299.00, $269.00 above the offering price. During the first ten days following VA Linux's IPO, VA Linux allocants purchased a total of 1,572,138 shares in the aftermarket. The same investors sold a total of 165,473 shares during that time.

See Fischel Tie-In Summary.

See Fischel Purchase Order Summary.

See Aftermarket Trading in VA Linux Systems, Inc. by CSFB Allocated Accounts, Ex. D-6 to 7/12/04 Fischel Report.

On February 3, 2000, VA Linux announced its acquisition of, Inc. in a $913,300,000 stock deal. In March and April 2000, VA Linux announced acquisitions, using stock and cash, of Trusolutions, Inc., NetAttach and Precision Insight, Inc. On June 7, 2000, a total of 22,940,202 shares became tradeable at the end of a lock-up period. On November 6, 2000, the company announced that it would miss its earnings estimates. That day, four firms monitoring VA Linux issued analyst reports, and the stock plummeted from $30.00 to $17.38. The stock closed at $7.94 on December 6, 2000. Fischel asserts that VA Linux underperformed when compared to various market benchmarks by 30 to 66 percentage points from December 9, 1999 to December 6, 2000, and by 26 to 58 points thereafter. On January 11, 2001, when the first VA Linux case, Makaron v. VA Linux Sys., Inc., No. 01 Civ. 0242, was filed, VA Linux closed at $9.031 per share.

See VA Linux Chronology of Events from December 9, 1999 to December 6, 2000 ("VA Linux Chronology"), Ex. I to Hunter VA Linux Decl., at 6. The, Inc. deal closed on June 7, 2000. See VA Software Corp. 10-Q filed on 6/12/00, ex. H to Hunter VA Linux Decl., at 9.

See id. at 9, 11.

See 12/9/99 VA Linux Prospectus at 65.

See VA Linux Chronology at 28.

See id. at 28-29.

See CBS MarketWatch: Historical Price, b=LNUXsiteid=mktw (Sept. 13, 2004).

See 1/20/04 Fischel Report ¶¶ 26, 29.

See NASDAQ: Charts, basicsintraday=offtimeframe=5ycharttype=ohlcsplits =offearnings=offmovingaverage=Nonelowerstudy =volumecomparison=offin dex=drilldown=offsymbol =LNUXselected=LNUX (August 20, 2004).

b. VA Linux Class Representatives

(1) Harold Zagoda

On December 9, 1999, Harold Zagoda purchased 10,000 shares of VA Linux stock at $300.00 per share and received an allocation of 300 shares at $30.00 per share. He sold 800 shares on December 13, 1999 for $200.06 per share. On June 3, 2000, he purchased 1,000 shares at $61.50 per share; on July 31, he purchased 1,500 shares at $32.00 per share; and on October 27, he purchased 1,000 shares at $27.50 per share. He held 13,000 shares on December 6, 2000, when the stock closed at $7.94 per share. Had he sold his remaining shares that day, Zagoda would have suffered a total pre-December 6, 2000 loss of $2,882,732.00.

See Proposed Class Representative Transactions in VA Linux During 252 Days of the Proposed Class Period ("VA Linux Transactions Data"), Ex. L to Hunter VA Linux Decl.

Based on the record before the Court, Zagoda continued to hold these shares until January 11, 2001, the date when Makaron v. VA Linux Sys., Inc., No. 01 Civ. 0242, was filed.

(2) Spiros and Mary Gianos

Spiros and Mary Gianos purchased 3,000 shares of VA Linux stock on December 10, 1999 for $764,180.00 and an additional 1,000 shares on December 13, 1999 for $200,000.00. Between December 13, 1999 and April 6, 2000, the Gianoses sold their 4,000 shares for $367,183.00, resulting in a loss of $596,997.00.

See id.

(3) Anita Budich

Anita Budich purchased 13 shares of VA Linux stock on December 15, 1999, at $230.63 per share, which she still owned on December 6, 2000. Had she sold them that day, when the stock closed at $7.94 per share, Budich would have suffered a $2,894.97 loss on the transaction.

See id. Budich reported on her PSLRA Certification that she purchased 23 shares of VA Linux stock for $227.00 per share. See 1/16/01 PSLRA Lead Plaintiff Certification of Anita J. Budich. However, for purposes of this Opinion I rely on the values provided in the VA Linux Transactions Data.

F. Industry-Wide Events Affecting All Focus Cases During the Class Period

In opposition to these motions, defendants note that throughout the class period, various events affected the markets for each of the six focus cases, and indeed for all of these 310 consolidated actions. First, the market for Internet and technology stock underwent an unprecedented boom in the late 1990s, ignited by the emergence and visibility of the Internet coupled with a streak of economic optimism and experimentation. Stock prices soared. Eventually, though, the "huge market bubble in Internet stocks burst, propelling prices of those stocks down 90% in a few months," prompting a period of "market chaos." Second, various reports were published describing the use of tie-in agreements by some allocating underwriters in IPOs. In mid-June of 1999, amidst a period of staggering decline in many Internet stock prices, MSNBC published an investigative report stating that underwriters had employed "a widely practiced marketing scheme known as a `tie-in'" to artificially inflate the price of Internet IPO stock. According to the MSNBC article, this practice extended to "the industry's leading, most prestigious firms." The article described a scheme whereby "the opportunity to get in on IPO offerings at cheap, pre-market prices" was "conditioned on an unwritten, oral agreement" that the customer would "give back to the underwriters what amounts to a blank check to buy many more shares . . . the minute the deal goes public." The MSNBC article attained some notoriety within the investment industry. Then, on July 14, 2000, the Wall Street Journal published a front page article reporting that Gary Tanaka, a partner in the growth-oriented mutual fund group Amerindo, had made an "agreement to buy shares [of an internet company] in the aftermarket" to increase his fund's allocation in the IPO and that "market experts" believed such agreements "raise regulatory questions."

See Gompers Report ¶¶ 12, 13.

See Selected Internet Related Indices 10/22/99-12/29/00, Ex. 6 to Gompers Report.

iXL Mem. at 29.

In re Merrill Lynch Co., Inc. Research Reports Sec. Litig., 289 F. Supp. 2d 416, 419 (S.D.N.Y. 2003).

Christopher Byron, IPO Quid Pro Quo Squeezes Investors, MSNBC, June 16, 1999, Ex. A to Houck Decl., at A2.

Id. at A3.

Id. at A2.

Byron reiterated the claims made in his MSNBC article on television during a CNBC interview held on June 18, 1999. See Bill Griffeth, Interview with Christopher Byron on Possible Price Manipulation of Internet Stocks, CNBC, June 18, 1999, Ex. A to Houck Decl., at A12. That same day, the well-visited financial website Raging Bull featured Byron's article as the second listing on its "News Links of the Week" page. See Lycos Finance's Raging Bull: News Links of the Week for June 18, 1999, Ex. A to Houck Decl., at A8. A discussion of the article also appeared as a post in a Yahoo! chat room. See MSFT_Sux, Yahoo Groups Internet Stock Talk: Potential Dangers of Playing Internet IP, June 18, 1999, Ex. A to Houck Decl., at A10.

Greg Ip et al., The Color Green: The Internet Bubble Broke Records, Rules and Bank Accounts, Wall St. J., July 14, 2000, Ex. A to Houck Decl., at A1.

On August 25, 2000, the SEC's Division of Market Regulation issued Staff Legal Bulletin No. 10, warning securities distributors that "solicit[ing] their customers to make additional purchases of the offered security after trading in the security begins" violates Regulation M, promulgated pursuant to the Exchange Act. The Bulletin called tie-in agreements "a particularly egregious form of solicited transaction" that "undermine the integrity of the market as an independent pricing mechanism for the offered security." The Bulletin explained that "[u]nderwriters have an incentive to artificially influence aftermarket activity because they have underwritten the risk of the offering, and a poor aftermarket performance could result in reputational and subsequent financial loss."

SEC Div. of Market Reg.: Staff Legal Bulletin No. 10 (Aug. 25, 2000), Ex. A to Houck Decl., at A25.

The release of the SEC Bulletin prompted Barron's to publish an article on September 11, 2000, stating that despite its "tame" tone, "[t]he bulletin targets one of the main grease guns now lubricating the IPO machine." The Wall Street Journal followed suit on December 6, 2000, publishing an article detailing the "new IPO playbook on Wall Street," in which investors who agree to buy in the aftermarket are the ones who receive the largest allocations. The article also acknowledged that while the SEC, in its August Bulletin, was "blowing the whistle" to quell the practice of using tie-in agreements, the understandings between investors and brokers were usually informal and unwritten, making them difficult to police. However, despite their "limited and implicit" nature, claimed some investors, "tie-ins . . . have a significant market impact . . . provid[ing] the rocket fuel that sometimes boosts IPO prices into orbit on the first trading day." Prices for all six of the focus stocks remained depressed, and never returned to their offering levels. Finally, defendants note that various press reports described apparent conflicts of interest among analysts and underwriters. SEC Chairman Arthur Levitt acknowledged the problem on April 13, 1999, issuing an "early warning signal" that the high number of "rosy stock analyses appear to be shaped by the lucrative investment banking ties that analysts' firms have with the companies they're supposed to watch with a critical eye." Following Levitt's warning, allegations of analyst conflicts continued to appear in the press. For example, the British Sunday Times reported on April 23, 2000, that "[t]he `Chinese Walls' that once separated researchers and bankers have all but disappeared in today's banking world and researchers have often become blatant pitchmen for bank deals," and an August 1, 2000 article in The Philadelphia Inquirer quoted a finance professor to the effect that the analyst conflict problem "seems to have gotten worse in recent years," with analysts issuing approximately fifty "buy" recommendations for every "sell" recommendation, a far cry from the six to one ratio that existed in the early 1990s.

Jack Willoughley, Pump It Up: The Dirty Little Secret Behind the IPO Boom, Barron's, Sept. 11, 2000, Ex. A to Houck Decl., at A28.

Susan Pulliam and Randall Smith, Trade-Offs: Seeking IPO Shares, Investors Offer to Buy More in After-Market, Wall St. J., Dec. 6, 2000, at A1, Ex. K to 1/20/04 Fischel Report.


See generally (Oct.11, 2004).

See Press Reports on Analyst Conflicts, Ex. B to Houck Decl., at B1 (documenting fifteen articles discussing analyst conflicts published from October 27, 1985 through August 1, 2000).

A Briefing for Investors: Analysts' Rosy Reports Draw SEC Chief's Fire, L.A. Times, Apr. 14, 1999, at C4, Ex. B to Houck Decl., at B46.

Garth Alexander, The Tipsters Who Never Say Sell, Sunday Times (of Britain), Apr. 23, 2000, at 8, Ex. B to Houck Decl., at B59.

Miriam Hill, Even Research Comes With Spin, The Philadelphia Inquirer, Aug. 1, 2000, at C1, Ex. B to Houck Decl., at B66.


A. The Requirements of Rule 23

Federal Rule of Civil Procedure 23 governs class certification. To be certified, a putative class must meet all four requirements of Rule 23(a) as well as the requirements of one of the three subsections of Rule 23(b). In this case, as in most cases seeking money damages, plaintiffs bear the burden of demonstrating that the class meets the requirements of Rule 23(a) — referred to as numerosity, commonality, typicality, and adequacy — and that the action is "maintainable" under Rule 23(b)(3). Under Rule 23(b)(3) — the only applicable subsection of Rule 23(b) — "common" issues of law or fact must "predominate over any questions affecting only individual members," and a class action must be demonstrably "superior" to other methods of adjudication.

See Caridad v. Metro-North Commuter R.R., 191 F.3d 283, 291 (2d Cir. 1999).

See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 614 (1997).

1. Rule 23(a)

a. Numerosity

Rule 23 requires that the class be "so numerous that joinder of all members is impracticable." "Impracticability does not mean impossibility of joinder, but refers to the difficulty or inconvenience of joinder." Although precise calculation of the number of class members is not required, and it is permissible for the court to rely on reasonable inferences drawn from available facts, numbers in excess of forty generally satisfy the numerosity requirement. The numerosity of plaintiffs' proposed classes, each of which includes thousands of investors, is undisputed.

In re Independent Energy Holdings PLC Sec. Litig., 210 F.R.D. 476, 479 (S.D.N.Y. 2002) (citing In re Avon Sec. Litig., No. 91 Civ. 2287, 1998 WL 834366, at *5 (S.D.N.Y. Nov. 30, 1998)).

See Trief v. Dun Bradstreet Corp., 144 F.R.D. 193, 198 (S.D.N.Y. 1992).

b. Commonality

Commonality requires a showing that common issues of fact or law affect all class members. A single common question may be sufficient to satisfy the commonality requirement. "The critical inquiry is whether the common questions are at the core of the cause of action alleged."

See Fed.R.Civ.P. 23(a)(2); see also Trief, 144 F.R.D. at 198.

See German v. Federal Home Loan Mortgage Corp., 885 F. Supp. 537, 553 (S.D.N.Y. 1995).

D'Alauro v. GC Services Ltd. P'ship, 168 F.R.D. 451, 456 (E.D.N.Y. 1996) (quotation omitted). See also In re "Agent Orange" Prod. Liab. Litig., 818 F.2d 145, 166-67 (2d Cir. 1987).

The commonality requirement has been applied permissively in securities fraud litigation. In general, where putative class members have been injured by similar material misrepresentations and omissions, the commonality requirement is satisfied.

See In re Blech Sec. Litig., 187 F.R.D. 97, 104 (S.D.N.Y. 1999).

See, e.g., Blackie v. Barrack, 524 F.2d 891, 902 (9th Cir. 1975) ("Confronted with a class of purchasers allegedly defrauded over a period of time by similar misrepresentations, courts have taken the common sense approach that the class is united by a common interest in determining whether a defendant's course of conduct is in its broad outlines actionable. . . ."); In re Baldwin-United Corp. Litig., 122 F.R.D. 424, 426 (S.D.N.Y. 1986) ("The nub of plaintiffs' claims is that material information was withheld from the entire putative class in each action, either by written or oral communication. Essentially, this is a course of conduct case, which as pled satisfies the commonality requirement of Rule 23. . . .").

c. Typicality

The typicality requirement "is not demanding." A named plaintiff's claims are "typical" pursuant to Rule 23(a)(3) where each class member's claims arise from the same course of events and each class member makes similar legal arguments to prove the defendants' liability. "The rule is satisfied . . . if the claims of the named plaintiffs arise from the same practice or course of conduct that gives rise to the claims of the proposed class members."

Forbush v. J.C. Penney Co., 994 F.2d 1101, 1106 (5th Cir. 1993) (citing Shipes v. Trinity Indus., 987 F.2d 311, 316 (5th Cir. 1993)).

See Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 155 (2d Cir. 2001).

Marisol A. v. Giuliani, 929 F. Supp. 662, 691 (S.D.N.Y. 1996), aff'd, 126 F.3d 372 (2d Cir. 1997).

In addition, a putative class representative's claims are not typical if that representative is subject to unique defenses. The test is whether the defenses will become the focus of the litigation, overshadowing the primary claims and prejudicing other class members. Accordingly, the commonality and typicality requirements "`tend to merge' because `[b]oth serve as guideposts for determining whether . . . the named plaintiff's claim and the class claims are so inter-related that the interests of the class members will be fairly and adequately protected in their absence.'"

See Baffa v. Donaldson, Lufkin Jenrette Sec. Corp., 222 F.3d 52, 59 (2d Cir. 2000).

See Landry v. Price Waterhouse Chartered Accountants, 123 F.R.D. 474, 476 (S.D.N.Y. 1989).

Caridad, 191 F.3d at 291 (alterations in original) (quoting General Tel. Co. of the Southwest v. Falcon, 457 U.S. 147, 157 n. 13 (1982)).

d. Adequacy

Plaintiffs must also show that "the representative parties will fairly and adequately protect the interests of the class." To do so, plaintiffs must demonstrate that the proposed class representatives have no "interests [that] are antagonistic to the interest of other members of the class." Courts have also considered "whether the putative representative is familiar with the action, whether he has abdicated control of the litigation to class counsel, and whether he is of sufficient moral character to represent a class."

Fed.R.Civ.P. 23(a)(4). See also Banyai v. Mazur, 205 F.R.D. 160, 164 (S.D.N.Y. 2002).

Baffa, 222 F.3d at 60. An antagonistic interest arises when there is a "fundamental conflict or inconsistency between the claims of the proposed class members" that is "so palpable as to outweigh the substantial interest of every class member in proceeding with the litigation." In re NASDAQ Market-Makers Antitrust Litig., 169 F.R.D. 493, 514-15 (S.D.N.Y. 1996). See also Robinson, 267 F.3d at 170 (holding that Rule 23(a)(4) requires "absence of conflict" between named representatives and class, as well as "vigorous prosecution").

Noble v. 93 University Place Corp., No. 02 Civ. 1803, 2004 WL 944543, at *4 (S.D.N.Y. May 3, 2004) (citations omitted).

Class representatives cannot satisfy Rule 23(a)(4)'s adequacy requirement if they "have so little knowledge of and involvement in the class action that they would be unable or unwilling to protect the interests of the class against the possibly competing interest of the attorneys." However, it is well established that "in complex litigations such as securities actions, a plaintiff need not have expert knowledge of all aspects of the case to qualify as a class representative, and a great deal of reliance upon the expertise of counsel is to be expected."

Baffa, 222 F.3d at 61 (quotations and citation omitted).

In re AM Int'l Inc., Sec. Litig. 108 F.R.D. 190, 196-97 (S.D.N.Y. 1985).

The requirements of adequacy and typicality tend to bleed into one another. But "[r]egardless of whether the issue is framed in terms of the typicality of the representative's claims . . . or the adequacy of [their] representation . . . there is a danger that absent class members will suffer if their representative is preoccupied with defenses unique to [her]."

Gary Plastic Packaging Corp. v. Merrill, Lynch, Pierce, Fenner Smith, Inc., 903 F.2d 176, 180 (2d Cir. 1990).

e. Ascertainability

Although "`Rule 23(a) does not expressly require that a class be definite in order to be certified[,] a requirement that there be an identifiable class has been implied by the courts.'" "This implied requirement is often referred to as `ascertainability.'"

In re MTBE, 209 F.R.D. at 336 (quoting Zapka v. Coca-Cola Co., No. 99 Civ. 8238, 2000 WL 1644539, at *2 (N.D. Ill. Oct. 27, 2000)).

Id. (citing Van West v. Midland Nat'l Life Ins. Co., 199 F.R.D. 448, 451 (D.R.I. 2001); In re Copper Antitrust Litig., 196 F.R.D. 348, 358 (D. Wis. 2000)).

"An identifiable class exists if its members can be ascertained by reference to objective criteria." "Class members need not be ascertained prior to certification, but `the exact membership of the class must be ascertainable at some point in the case.'" It must thus be "administratively feasible for a court to determine whether a particular individual is a member" of the class. "The Court must be able to make this determination without having to answer numerous fact-intensive questions."

Id. at 337 (quoting Zapka, 2000 WL 1644539, at *2). See also Clay v. American Tobacco Co., 188 F.R.D. 483, 490 (S.D. Ill. 1999); Gomez v. Illinois State Bd. of Educ., 117 F.R.D. 394, 397 (N.D. Ill. 1987).

Id. (quoting Rios v. Marshall, 100 F.R.D. 395, 403 (S.D.N.Y. 1983)).

Rios, 100 F.R.D. at 403 (citing 7 CHARLES ALAN WRIGHT, ARTHUR R. MILLER MARY KAY KANE, FED. PRAC. PROC. § 1760 at 581 (1972)).

Daniels v. City of New York, 198 F.R.D. 409, 414 (S.D.N.Y. 2001) (quotation omitted). A plaintiff's failure to plead a class that can be ascertained without subjective individual inquiries may cause individual issues ( i.e., whether each individual class member belongs in the class) to predominate. Consequently, the question of ascertainability takes on great importance in 23(b)(3) class actions, which may only be certified if common questions — not individual inquiries — predominate. See infra Part IV.B.1.

2. Rule 23(b)

If plaintiffs can demonstrate that the proposed class satisfies the elements of Rule 23(a), they must then establish that the action is "maintainable" as defined by Rule 23(b). Rule 23(b) provides that "an action may be maintained as a class action if the prerequisites of subdivision (a) are satisfied, and in addition" one of three alternative definitions of maintainability is met. Plaintiffs argue that these putative class actions are maintainable under subsection (b)(3), which requires "that questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy." Rule 23(b)(3) thus has two elements: "predominance" and "superiority."

a. Predominance

"In order to meet the predominance requirement of Rule 23(b)(3), a plaintiff must establish that the issues in the class action that are subject to generalized proof, and thus applicable to the class as a whole . . . predominate over those issues that are subject only to individualized proof." "The 23(b)(3) predominance requirement is `more stringent' and `far more demanding than' the commonality requirement of Rule 23(a)." Courts frequently have found that the requirement was not met where, notwithstanding the presence of common legal and factual issues that satisfy the commonality requirement, individualized inquiries predominate. Nonetheless, the Supreme Court has noted that "[p]redominance is a test readily met in certain cases alleging consumer or securities fraud. . . ."

In re VISA Check/Master Money Antitrust Litig., 280 F.3d 124, 136 (2d Cir. 2001) (quotation marks and citation omitted).

Maneely v. City of Newburgh, 208 F.R.D. 69, 76 (S.D.N.Y. 2002) (quoting Amchem Prods., 521 U.S. at 623-24).

See, e.g., Augustin v. Jablonsky, No. 99-CV-3126, 2001 WL 770839, at *13 (E.D.N.Y. Mar. 8, 2001) (finding individualized issues of proximate causation predominate despite plaintiffs' showing of commonality under Rule 23(a)(2)); Martin v. Shell Oil Co., 198 F.R.D. 580, 592-93 (D. Conn. 2000) (finding individualized proof of breach, causation, and trespass predominates where commonality was not contested); In re MTBE, 209 F.R.D. at 350 (finding individual issues predominate although defendants conceded commonality).

Amchem Prods., 521 U.S. at 625.

b. Superiority

The superiority prong of Rule 23(b)(3) requires a court to consider whether a class action is superior to other methods of adjudication. The court should consider, inter alia, "the interest of the members of the class in individually controlling the prosecution or defense of separate actions" and "the difficulties likely to be encountered in the management of a class action."

See Fed.R.Civ.P. 23(b)(3); see also Eisen v. Carlisle Jacquelin, 417 U.S. 156, 164 (1974).

3. Rule 23(g)

Rule 23(g) requires a court to assess the adequacy of proposed class counsel. To that end, the court must consider the following: (1) the work counsel has done in identifying or investigating potential claims in the action, (2) counsel's experience in handling class actions, other complex litigation, and claims of the type asserted in the action, (3) counsel's knowledge of the applicable law, and (4) the resources counsel will commit to representing the class. The court may also consider "any other matter pertinent to counsel's ability to fairly and adequately represent the interests of the class."

See Noble, 2004 WL 944543, at *5 (citing Rule 23(g)(1)(C)(i)). See also In re Drexel Burnham Lambert Group, Inc., 960 F.2d 285, 291 (2d Cir. 1992) ("[C]lass counsel must be qualified, experienced and generally able to conduct the litigation.") (quotation marks and citation omitted).

Rule 23(g)(1)(C)(ii).

Defendants do not contest the qualifications of class counsel, who easily meet the requirements of Rule 23(g).

B. The Standard of Proof

All of these requirements are aimed at answering two questions: Can the claims be managed as class actions, and should they be managed as class actions? In this regard, the term "claims" encompasses not only plaintiffs' claims, but also any affirmative defenses that defendants may assert.

See Waste Mgmt. Holdings, Inc. v. Mowbray, 208 F.3d 288, 295 (1st Cir. 2000) ("[A]ffirmative defenses should [also] be considered in making class certification decisions."); Castano v. American Tobacco Co., 84 F.3d 734, 744 (5th Cir. 1996) (explaining that "a court must understand the claims, defenses, relevant facts, and applicable substantive law in order to make a meaningful determination of the certification issues").

Courts must therefore exercise their judgment to further Rule 23's goals of promoting judicial economy and providing aggrieved persons a remedy when it is not economically feasible to obtain relief through multiple individual actions. The Second Circuit requires a "liberal" construction of Rule 23. Thus, "to deny a class action simply because all of the allegations of the class do not fit together like pieces in a jigsaw puzzle would destroy much of the utility of Rule 23." Accordingly, in securities cases, "when a court is in doubt as to whether or not to certify a class action, the court should err in favor of allowing the class to go forward."

See Califano v. Yamasaki, 442 U.S. 682, 701 (1979). See generally 5 JAMES WM. MOORE ET AL., MOORE'S FEDERAL PRACTICE § 23.03 (3d ed. 2004) ("Rather than attempting to interpret Rule 23 through a `liberal' or `strict' construction, the best policy is to take a flexible approach, and to interpret Rule 23 so as to promote the purposes underlying the rule. The class action device was designed to promote judicial efficiency and to provide aggrieved persons a remedy when individual litigation is economically unrealistic, as well as to protect the interests of absentee class members. The underlying purposes of Rule 23 may supply meaningful guidance for courts called on to resolve disputes concerning the proper interpretation of the Rule.").

See Korn v. Franchard Corp., 456 F.2d 1206 (2d Cir. 1972); In re Lloyd's Am. Trust Fund Litig., No. 96 Civ. 1262, 1998 WL 50211, at *5 (S.D.N.Y. Feb. 6, 1998) ("The Second Circuit has directed district courts to apply Rule 23 according to a liberal rather than a restrictive interpretation.").

Green v. Wolf Corp., 406 F.2d 291, 300 (2d Cir. 1968).

In re Indep. Energy, 210 F.R.D. at 479 (citation omitted). To that end, the Second Circuit has recently reaffirmed that while class certification decisions are reviewed for abuse of discretion, it is "`noticeably less deferential to the district court when that court has denied class status than when it has certified a class.'" Moore v. PaineWebber, Inc., 306 F.3d 1247, 1252 (2d Cir. 2002) (quoting Caridad, 191 F.3d at 291).

Notwithstanding the general liberality in this circuit towards class certification motions, the Supreme Court unequivocally requires district courts to undertake a "rigorous analysis" that the requirements of Rule 23 have been satisfied. The burden rests on plaintiffs to make this showing.

Falcon, 457 U.S. at 161.

See Caridad, 191 F.3d at 291.

The question remains, however, as to what constitutes a rigorous analysis. Must plaintiffs prove their case? Must a district court make factual and legal findings that the proposed class satisfies the Rule? Given that class certification decisions only became appealable in 1998, and that the Supreme Court did not even articulate the "rigorous analysis" standard until 1982, the guidance from higher courts is scant.

See Fed.R.Civ.P. 23(f) and Advisory Committee Note.

In Eisen v. Carlisle Jacquelin, a securities and antitrust suit that originated in this district, the Supreme Court made its first significant pronouncement on class certification. In that case, the district court, after conducting a hearing on the merits of plaintiffs' claims, imposed 90% of the cost of the class notice on defendants. Finding fault in the lower court's approach, the Supreme Court explained that "nothing in either the language or history of Rule 23 . . . gives a court any authority to conduct a preliminary inquiry into the merits of a suit in order to determine whether it may be maintained as a class action. Indeed, such a procedure contravenes the Rule. . . ."

417 U.S. at 177. See also Miller v. Mackey Int'l, Inc., 452 F.2d 424, 427 (5th Cir. 1971) ("In determining the propriety of a class action, the question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits, but rather whether the requirements of Rule 23 are met.") quoted in Eisen, 417 U.S. at 178.

Many lower courts have understood this passage in Eisen to mean that on a Rule 23 motion, as on a Rule 12(b)(6) motion, a court must assume the allegations contained in the complaint to be true and draw all inferences in plaintiffs' favor. In several cases decided shortly after Eisen, for instance, the Second Circuit held that on a Rule 23 motion, "the facts will be taken as alleged in the complaint or as they appear without dispute in the record before us."

Shayne v. Madison Square Garden Corp., 491 F.2d 397, 398 (2d Cir. 1974). Accord Shelter Realty Corp. v. Allied Maint. Corp., 574 F.2d 656, 661 n. 15 (2d Cir. 1978) ("[W]e have previously held that it is proper to accept the complaint allegations as true in a class certification motion."). See also Green, 406 F.2d at 294 n. 1 (stating, in a case decided prior to Eisen: "We have assumed for the purpose of this appeal, as we are required to do, that the `facts' stated in the complaint and accompanying papers submitted to Judge Ryan are true.").

But such a view — if it was ever correct — is no longer the prevailing view. As Judge Frank Easterbrook recently stated, "[t]he proposition that a district judge must accept all of the complaint's allegations when deciding whether to certify a class cannot be found in Rule 23 and has nothing to recommend it."

Szabo v. Bridgeport Machs., Inc., 249 F.3d 672, 675 (7th Cir. 2001). See id. at 677 (explaining that Eisen does not require a court to accept the allegations of a complaint as true, but only prohibits a court from saying something like "I'm not going to certify a class unless I think that the plaintiffs will prevail."); Gariety v. Grant Thornton, LLP, 368 F.3d 356, 366 (4th Cir. 2004) (" Eisen simply restricts a court from expanding the Rule 23 certification analysis to include consideration of whether the proposed class is likely to prevail ultimately on the merits."); Johnston v. HBO Film Mgmt., Inc., 265 F.3d 178, 186-89 (2d Cir. 2001) (finding it "not only . . . appropriate, but also necessary" to look beyond the pleadings at class certification); Waste Mgmt. Holdings, 208 F.3d at 298 ("[A] district court must formulate some prediction as to how specific issues will play out in order to determine whether common or individual issues predominate."); Castano, 84 F.3d at 744 (holding that Eisen does not suggest "that a court is limited to the pleadings when deciding on certification. . . . A district court certainly may look past the pleadings to determine whether the requirements of Rule 23 have been met.").

Just four years after Eisen, the Supreme Court explained in Coopers Lybrand v. Livesay that although district courts should avoid weighing the merits of a plaintiff's claims at class certification, "class determination generally involves considerations that are `enmeshed in the factual and legal issues comprising the plaintiff's cause of action.'" Four years later, the Court imposed its "rigorous analysis" test. Repeating the just-quoted language from Livesay, Justice Stevens wrote in General Telephone Company of the Southwest v. Falcon that "sometimes it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question. . . . [A]ctual, not presumed, conformance with Rule 23(a) remains . . . indispensable." In light of this language, it would be error to presume plaintiffs' allegations to be true.

437 U.S. 463, 469 (1978) (quoting Mercantile Nat'l Bank v. Langdeau, 371 U.S. 555, 558 (1963)). In a footnote, the Court quoted with approval a leading civil procedure treatise:

Evaluation of many of the questions entering into determination of class action questions is intimately involved with the merits of the claims. The typicality of the representative's claims or defenses, the adequacy of the representative, and the presence of common questions of law or fact are obvious examples. The more complex determinations required in Rule 23(b)(3) class actions entail even greater entanglement with the merits. . . .

15 C. WRIGHT, A. MILLER E. COOPER, FEDERAL PRACTICE AND PROCEDURE § 3911, at p. 485, n. 45 (1976) quoted in Livesay, 437 U.S. at 469 n. 12.

Falcon, 457 U.S. at 161.

Id. at 160.

The tricky question that remains, however, is: If a court may not take the allegations of the complaint as true, what showing must plaintiffs make in support of their class certification motion? On this question, the Supreme Court has been silent.

At least two Courts of Appeal have implied that plaintiffs' showing on a class certification motion must satisfy the requirements of Rule 23 by a preponderance of the evidence or a similar standard. In Szabo v. Bridgeport Machines, Inc., the Seventh Circuit held that where it is necessary to make legal or factual inquiries on a Rule 23 motion, the court should "receive evidence (if only by affidavit) and resolve the disputes before deciding whether to certify the class," even if such a resolution requires a "preliminary inquiry into the merits." Szabo likened a district court's finding under Rule 23 to the sorts of "inquiries routinely [undertaken] under Rule 12(b)(1) and 12(b)(2) before deciding whether [the courts] possess jurisdiction over the subject matter of the case and the persons of the defendants, the location of the proper venue, application of forum non conveniens, and other preliminary issues." If such situations are truly analogous to class certification, then a district court would need to find that the proposed class satisfies each of the elements of Rule 23 by a preponderance of the evidence.


See, e.g., Luckett v. Bure, 290 F.3d 493, 496-97 (2d Cir. 2002) (noting that a district court must find the existence of subject matter jurisdiction by a preponderance of the evidence); Metropolitan Life Ins. Co. v. Robertson-Ceco Corp., 84 F.3d 560, 567 (2d Cir. 1996) (same, for personal jurisdiction); GTFM Inc. v. International Basic Source, Inc., No. 01 Civ. 6203, 2002 WL 42884, at *2 (S.D.N.Y. Jan. 11, 2002) (same, for change of venue); Ultra Sucro Co. v. Illinois Water Treatment Co., 146 F. Supp. 393, 398 (S.D.N.Y. 1956) (I. Kaufman, J.) (same, for forum non conveniens).

Even more recently, the Fourth Circuit held — in a securities fraud case — that a district court must make "findings" in resolving a Rule 23 motion, even if such findings overlap with the merits. The court explained:

Gariety, 368 F.3d at 366.

The . . . concern that Rule 23 findings might prejudice later process on the merits need not lead to the conclusion that such findings cannot be made. The jury or factfinder can be given free hand to find all of the facts required to render a verdict on the merits, and if its finding on any fact differs from a finding made in connection with class action certification, the ultimate factfinder's finding on the merits will govern the judgment. A model for this process can be observed in the context of the preliminary injunction practice. Courts make factual findings in determining whether a preliminary injunction should issue, but those findings do not bind the jury adjudging the merits, and the jury's findings on the merits govern the judgment to be entered in the case.


The court's analogy to a preliminary injunction hearing suggests that on a class certification motion in the Fourth Circuit, a plaintiff must establish the elements of Rule 23 by evidence sufficient to establish a "likelihood of success on the merits" — a burden similar to the Seventh Circuit's apparent requirement that plaintiffs prove that they satisfy Rule 23 by a preponderance of the evidence.

In the Fourth Circuit, a preliminary injunction may be granted only where the court finds that "the moving party clearly establishes entitlement to the relief sought." Hughes Network Sys. Corp. v. Interdigital Communications Corp., 17 F.3d 691, 693 (4th Cir. 1994). While such a determination entails some factfinding, the moving party need not establish her claims by a preponderance of the evidence; rather, "a probable right, and a probable danger that such right will be defeated, without the special interposition of the court, is all that need be shown." United States v. Fang, 937 F. Supp. 1186, 1197 (D. Md. 1996) (construing Hughes) (citations omitted).

Both the Supreme Court and the Second Circuit, however, have suggested that requiring a plaintiff to establish the elements of Rule 23 — especially when those elements are "enmeshed" in the merits — by a preponderance of the evidence would work an injustice. In Eisen, the Court noted that

a preliminary determination of the merits may result in substantial prejudice to the defendant, since of necessity it is not accompanied by the traditional rules and procedures applicable to civil trials. The court's tentative findings, made in the absence of established safeguards, may color the subsequent proceedings and place an unfair burden on the defendant.

More recently, in Caridad v. Metro-North Commuter Railroad, the Second Circuit reminded district courts that they "must not consider or resolve the merits of the claims of the purported class." Rather, a plaintiff is only required to make "some showing." Differing from the Seventh and Fourth Circuits, the Second Circuit clearly held that "a weighing of the evidence is not appropriate at this stage in the litigation." If a district court is forbidden to weigh the evidence on class certification, a fortiori, plaintiffs need not establish the elements of Rule 23 by a preponderance of the evidence.

191 F.3d at 293 (citing Eisen, 417 U.S. at 177).

Id. at 292.

Id. at 293.

Even more recently, in In re VISA Check/Master Money Antitrust Litigation, the Second Circuit reiterated the "some showing" standard. Juxtaposing the requirements of Falcon and Caridad, the court held that "[a]lthough a trial court must conduct a `rigorous analysis' to ensure that the prerequisites of Rule 23 have been satisfied before certifying a class, `a motion for class certification is not an occasion for examination of the merits of the case.'" In the context of expert reports, for example, VISA Check teaches that a district court "may not weigh conflicting expert evidence or engage in `statistical dueling' of experts." Instead, the sole job of a district court in assessing expert evidence on a class certification motion is to "ensure that the basis of the [plaintiff's] expert opinion is not so flawed that it would be inadmissible as a matter of law."

280 F.3d at 134-35 (quoting Caridad, 191 F.3d at 291).

Id. at 135 (quoting Caridad, 191 F.3d at 292).

Id. (citing cases).

In sum, under the binding caselaw in this Circuit, a district court may not simply accept the allegations of plaintiffs' complaint as true. Rather, it must determine, after a "rigorous analysis," whether the proposed class comports with all of the elements of Rule 23. In order to pass muster, plaintiffs — who have the burden of proof at class certification — must make " some showing." That showing may take the form of, for example, expert opinions, evidence (by document, affidavit, live testimony, or otherwise), or the uncontested allegations of the complaint.


1. Commonality

A. Rule 23(a)

The common issues of liability presented in these six class actions are overwhelming. Any plaintiff seeking damages — whether proceeding individually or as a class member — will have to establish the following facts, all of which defendants vigorously dispute:

• The participation of each defendant in the alleged scheme.
• The existence and terms of tie-in agreements, and the process by which defendants induced allocants to enter into tie-in agreements.
• Defendants' failure to disclose the existence, extent and purpose of the tie-in agreements, and the materiality of that omission.
• The existence and magnitude of excess compensation, and how such payments were induced. If excess compensation was paid in unusual forms, such as wash sales, that those actions amounted to payment of excess compensation.
• Defendants' failure to disclose excess compensation, and the materiality of defendants' omission.
• Where defendants conducted a secondary public offering ("SPO") ( e.g., Corvis, Sycamore and iXL), that the SPO offering price was derived from prices that were artificially inflated through market manipulation, that defendants failed to disclose the price inflation, and the materiality of that omission.
• That plaintiffs are entitled to a presumption of reliance.
• That plaintiffs bought their shares in an efficient market.
• That analysts reporting on the specific securities had conflicts of interest, that the analysts failed to disclose such conflicts, and that such omissions were material.
• That analyst coverage was used in the marketing of defendants' IPOs.
• That price-targets set in analyst reports were the product of manipulated prices.
• That tie-in agreements and analyst reports materially affected stock prices.
• That defendants acted with scienter in manipulating stock prices.
• That defendants' manipulation actually caused inflation of stock prices.
• That the artificial inflation of stock prices caused by the unlawful scheme dissipated over time.
• The true value and actual price of the stock at the time plaintiffs purchased and sold stock.
• That press reports and regulatory announcements were neither sufficiently clear nor specific to place plaintiffs on inquiry notice of the alleged scheme with respect to each issuer. Proof of these facts may require extensive discovery and expert testimony, and, if the three and one-half years that have elapsed since the filing of the first suit in these consolidated actions, Makaron v. VA Linux, are any indication, the disposition of all of the 310 consolidated class actions will take years.

The following is a list of common issues for all plaintiffs asserting claims under section 10(b) and Rule 10b-5. Those plaintiffs who also fall into plaintiffs' section 11 classes will need to prove additional common issues with respect to those claims, including whether class members who bought in the aftermarket can trace their shares to a defective registration statement and whether class members who bought after a 12-month earnings statement had been issued can prove that the earnings statement failed to cure the registration statement's misrepresentations and omissions.

See Trial Plan. This list includes some of the common questions anticipated by plaintiffs and the Court. Defendants, of course, may present defenses that raise issues common to all class members and that can be decided based on common proof.

By contrast, only a few authentically individualized issues remain. Most prominent is the need to calculate damages individually, but even that may be accomplished by applying a common formula to each individual claim. Indeed, the quantum of damages is the only element plaintiffs must prove on an individual basis. All other individual questions ( e.g., actual knowledge and inquiry notice) will arise because of issues defendants choose to raise. In fact, many of defendants' anticipated defenses will require proof that is relevant to large groups within the class; for example, if defendants assert that a plaintiff's claim is barred because the June 16, 1999, MSNBC article placed that plaintiff on notice that the IPO market was tainted by fraud, the determination of that issue is relevant to all plaintiffs who purchased after the MSNBC article. Similarly, if defendants argue that a plaintiff's claim is barred because her trading history shows the payment of undisclosed compensation through "wash" sales, the questions of whether the wash sale constitutes undisclosed compensation and whether a particular pattern of trading activity constitutes a wash sale bear on all similarly situated class members.

See infra Part IV.B.3.

As a result, plaintiffs have satisfied the Rule 23(a) commonality requirement. Defendants' contention that individual issues will predominate at trial is addressed in the discussion of the Rule 23(b)(3) predominance requirement.

See infra Part IV.B.

2. Typicality

"When it is alleged that the same unlawful conduct was directed at or affected both the named plaintiff and the class sought to be represented, the typicality requirement is usually met irrespective of minor variations in the fact patterns underlying individual claims." "The factual background of each named plaintiff's claim need not be identical to that of all the class members as long as `the disputed issue of law or fact occup[ies] essentially the same degree of centrality to the named plaintiff's claim as to that of other members of the proposed class.'" For example, where plaintiffs allege a market manipulation scheme, typicality may be satisfied despite fluctuations in the amount of inflation over time, even though such fluctuations create differences between class members and class representatives in terms of how much, if any, of their loss was caused by an alleged scheme.

Robidoux v. Celani, 987 F.2d 931, 936-37 (2d Cir. 1993).

In re Worldcom, 219 F.R.D. at 280 (alteration in original) (quoting Caridad, 191 F.3d at 293).

See In re Worldcom, 219 F.R.D. at 280; cf. In re Sumitomo Copper Litig., 262 F.3d 134, 140-41 (2d Cir. 2001);

Plaintiffs' proposed class representatives allege that they were harmed in the same "unitary scheme" as the rest of the class. All class members, including the proposed class representatives, bought shares allegedly inflated by defendants' wrongdoing, and each was damaged thereby. The disputed issues are central to the claims of all proposed class representatives and the class members they seek to represent.

Plaintiffs' Reply at 15.

Pappas, one of plaintiffs' proposed Engage class representatives, was an allocant and did not purchase shares in the aftermarket. He is clearly qualified to serve as plaintiffs' Engage section 11 class representative. See infra Part IV.B.4. Pappas will therefore likely be preoccupied by showing that the omission of the alleged scheme ( i.e., tie-in agreements, undisclosed compensation and analyst conflicts) from the registration statement was material, and not that the alleged scheme actually drove up prices in the aftermarket. Similarly, to the extent that Pappas asserts market manipulation claims, defendants may challenge his ability to prove loss causation (because he bought shares before any artificial inflation created by tie-in agreements affected the market price). Accordingly, Pappas is not a suitable 10b-5 class representative.

Some class representatives, though, are not qualified to represent plaintiffs' section 11 classes because they would be preoccupied by the unique defense that they cannot trace their shares to a defective registration statement. See infra Part IV.B.4.b.

However, even where a class representative's motivation to prove the underlying fraud is typical of all class members, she may nonetheless be excluded as atypical if she is "subject to unique defenses which threaten to become the focus of litigation." Defendants contend that some of plaintiffs' proposed class representatives are subject to unique defenses, and thus atypical, because they: (1) were allocants, engaged in tie-in agreements, or were knowledgeable institutional investors, and thus can be charged with knowledge of the alleged scheme; (2) purchased stock after publication of certain articles or after the SEC Bulletin, and thus were on inquiry notice of the alleged scheme; (3) were short sellers, momentum traders or day traders, and therefore cannot avail themselves of a presumption of reliance on stock prices; or (4) purchased stock after the close of the class period or after filing suit, and therefore cannot be said to have relied on the integrity of the market, because they were willing to buy after learning of the alleged scheme.

Gary Plastic, 903 F.2d at 180. While Gary Plastic also found that a holder of certificates of deposit ("CDs") who discovered that interest on those CDs was being fraudulently underpaid but nevertheless allowed the CDs to roll over several times was subject to such a unique defense, id., such a finding is not mandated here for proposed class representatives who purchased shares after the close of the class period. In an efficient stock market — as opposed to the market for CDs — information that share prices have been fraudulently manipulated is immediately reflected in share prices, and the resulting artificial inflation disappears. See infra Part IV.B.1.

See Engage Mem. at 24; Firepond Mem. at 28-31; VA Linux Mem. at 40 n. 39.

Defendants argue that such notice would both bar some class representatives' claims under the section 10(b) statute of limitations, see iXL Mem. at 18-19, and prevent others from enjoying a presumption of reliance, see Corvis Mem. at 18-19.

See Firepond Mem. at 31-32.

See Engage Mem. at 33-34.

Defendants' arguments are unavailing. The question of participation resulting in actual knowledge is adequately addressed by the revised class definition. The question of whether any particular publication placed a class representative on inquiry notice of the alleged scheme early enough that her claim would be barred under the section 10(b) statute of limitations is itself a question common to all class members. Defendants may also choose to challenge the rebuttable presumption of reliance with respect to any individual class representative on the grounds that some publication ( e.g., the MSNBC article or the SEC Bulletin) placed her on inquiry notice of the alleged scheme. This will not raise a unique defense. To the contrary, a determination that such publications were not sufficient to place the class representative on inquiry notice would inure to the benefit of all class members who, like the class representative, bought after the publication was issued. Conversely, if defendants succeed in rebutting the presumption of reliance as to any class representative, then each similarly situated class member would be forced to prove reliance individually, thereby causing individual questions to predominate for those investors and mandating amendment of the class definition or decertification.

See infra Part IV.A.4.

See infra Part IV.B.1.b.(3).

Similarly, defendants' attacks on the proposed class representatives' reliance on the integrity of the market because of "unique" investment strategies do not defeat typicality. The classes as pled include many investors with similar investment strategies, so any "unique defenses" based on those strategies are in fact common questions. Finally, defendants' argument that class representatives who purchased after the close of the class period should be excluded because the "fact that [the] proposed plaintiff purchased shares both after learning of the fraud and after filing suit rebuts the fraud-on-the-market presumption" makes no sense. Just because a stock is manipulated at one point in its trading history does not mean that the stock is forever tainted; a plaintiff may legitimately believe that, although his past losses were caused by market manipulation, the effect of that manipulation has dissipated and the stock price once again reflects all available information about its true value.

See, e.g., In re Worldcom, 219 F.R.D. at 281-82 (rejecting similar arguments, noting that "[e]ach of these methods of making investment decisions is representative of methods used by many other investors. Each of the methods reflects an evaluation of the publicly available information about Worldcom. . . .").

See infra Part IV.B.1.b.(2).

Engage Mem. at 34 (construing Rolex Employees Ret. Trust v. Mentor Graphics Corp., 136 F.R.D. 658, 664 (D. Or. 1991)).

Plaintiffs' proposed class representatives' claims arise from the same course of events, and require the same legal arguments, as those of the class at large. Consequently, because defendants have not established that any proposed class representative in the six focus cases will assert atypical claims or be subject to unique defenses that "overshadow the primary claims and prejudic[e] other class members," plaintiffs have satisfied the Rule 23(a) typicality requirement with respect to their Exchange Act claims. However, certain proposed class representatives are atypical with respect to plaintiffs' section 11 classes because they are subject to the unique defense that they cannot trace their shares to an allegedly defective registration statement, as discussed in Part IV.B.4.b. below.

In re MTBE, 209 F.R.D. at 338 n. 22.

3. Adequacy

a. Antagonistic Interests

Defendants attack as inadequate any proposed class representative who is either a proposed class representative or a class member in another of these consolidated actions. Defendants submit that this dual role creates a conflict of interest because plaintiffs with interests in multiple cases may seek to increase any settlement designation in favor of one action at the expense of another.

See Engage Mem. at 31-32. These representatives include Pappas and Kasbarian, who seek to represent the Engage class, see id; Zitto, a class representative in Firepond, see Firepond Mem. at 35; and Basu, a class representative in Corvis, see Corvis Mem. at 17.

Defendants cite two cases in support of their theory, but both are inapposite. First, in duPont v. Wyly, the court found duPont to be an inadequate class representative because he was also the plaintiff in a personal action he brought against University Computing Company ("UCC"), a defendant in Wyly. Given that recovery in either case could have rendered UCC judgment-proof, the court found that duPont could not represent the class as he would have an interest in ensuring his own recovery in his personal law suit. Second, in Boro Hall v. Metropolitan Tobacco Co., Jamaica Tobacco, a proposed class representative, had not only brought a personal antitrust action against Metro Tobacco, but was also a competitor of other class members. Furthermore, Metro Tobacco counterclaimed against Jamaica Tobacco, giving Jamaica Tobacco an incentive to settle that other class members would not share.

61 F.R.D. 615, 624 (D. Del. 1973).

See id.

74 F.R.D. 142, 144 (E.D.N.Y. 1977).

See id.

It is overwhelmingly likely that the interests of the proposed class representatives, even in a settlement posture, will be in maximizing the possible recovery of all classes in which the class representative is a member. For a class representative to profit by reducing the recovery of the class she represents for the sake of another class, her monetary interest in the benefitted class would have to be many times greater than her interest in the class she represents, because the money sacrificed by one class is likely to be distributed among three hundred different classes, each with thousands of class members. There is no evidence that any class representative has such a disproportionately small interest in the class that he, she, or it seeks to represent. Furthermore, this Court will review the fairness of any settlement to ensure that it is reasonable and adequate, and to prevent inequitable distribution. For these reasons, membership in more than one of these consolidated classes does not result in an interest so antagonistic as to prevent the adequate representation of absent class members.

See Joel A. v. Giuliani, 218 F.3d 132, 138 (2d Cir. 2000).

b. Familiarity with the Action

Defendants argue that the proposed Sycamore class representatives cannot fulfill their roles as fiduciaries to class members because they are unfamiliar both with their case and their duties as class representatives. For example, defendants claim that Henderson, a Sycamore class representative, "does not understand the scheme alleged" in the Complaint. However, at his deposition Henderson described the alleged laddering, biased analyst reporting, and the inflated commissions allegedly received by underwriter defendants. Henderson also described his responsibilities as class representative to include retaining the best available counsel, remaining involved in the litigation, and ensuring that class members are kept informed about the litigation and that their interests are protected. Given his familiarity with the case and his responsibilities, Henderson satisfies Rule 23(a)(4)'s adequacy requirement. Gangaiah's testimony demonstrates a similar level of familiarity with the case and an understanding of his responsibilities. Lemberg, on the other hand, is clearly not a sophisticated investor and was unable to describe the operation of the alleged market manipulation. "Regardless, `it is unreasonable to expect an ordinary investor . . . to have the requisite sophistication and legal background to assist counsel in assessing liabilities under the securities laws.'" Even Lemberg, with his basic understanding of the case, satisfies Rule 23(a)(4)'s adequacy requirement. No evidence suggests that any proposed class representative is so lacking in her understanding of and involvement in her case that she is inadequate.

See Sycamore Mem. at 44-50 (discussing the inadequacies of class representatives Gangaiah, Henderson, and Lemberg). Sycamore's claim that class representatives have "conflicting views of the theory of [the] case," id., is essentially identical to their claim that the representatives do not understand the nature of the litigation or their responsibilities.

Id. at 47.

See Henderson Dep. at 107-11, 121-34.

See id. at 165-66.

See Gangaiah Dep. at 59-66, 98-99.

See Lemberg Dep. at 89.

In re College Bound Consol. Litig., No. 93 Civ. 2348, 1994 WL 236163, at *4 (S.D.N.Y. May 31, 1994) (quoting Klein v. A.G. Becker Paribas Inc., 109 F.R.D. 646, 651-52 (S.D.N.Y. 1986)).

See Lemberg Dep. at 89, 145.

c. Abdication of Control to Class Counsel

Defendants argue that the proposed class representatives have abdicated to class counsel their roles as fiduciaries for the class and so are inadequate to serve as representatives. Essentially, defendants raise the concern that by relinquishing responsibility for prosecuting the case to class counsel, the proposed representatives will be unable to protect the class members should a conflict of interest arise between class counsel and class members.

See Firepond Mem. at 32-36.

Defendants' concern is unwarranted. To the extent that it relates to the representatives' purported lack of participation or control, this argument is merely an extension of the familiarity objection, which I have already rejected. Even if, as defendants claim, counsel and the class representatives have conflicting views of the case, "a great deal of reliance upon the expertise of counsel is to be expected." For this reason, "[t]he ultimate responsibility to ensure that the interests of class members are not subordinated to the interests of either the class representatives or class counsel rests with the district court" — not the proposed class representatives.

In re AM Int'l., 108 F.R.D. at 196-97.

Maywalt v. Parker Parsley Petroleum Co., 67 F.3d 1072, 1078 (2d Cir. 1995). The foregoing discussion also disposes of defendants' concern that the "class representatives and their own counsel have conflicting views of the theory of this case." Sycamore Mem. at 44.

d. Moral Character

"Although credibility may warrant denying certification, `it is generally inappropriate to deny certification based on questions going to the credibility of named plaintiffs.'" Defendants assert that certain class representatives are inadequate because of their failure to disclose all transactions in the relevant security, inconsistencies in their sworn statements and testimony, and, in the case of Kasbarian, destruction of trading records after learning of the alleged fraud (but prior to filing suit).

Saddle Rock Partners v. Hiatt, No. 96 Civ. 9474, 2000 WL 1182793, at *5 (S.D.N.Y. Aug. 21, 2000) (quoting In re Frontier Ins. Group, Inc. Sec. Litig., 172 F.R.D. 31, 41 (E.D.N.Y. 1997)).

See, e.g., Engage Mem. at 35 (discussing Kasbarian); Firepond Mem. at 36-37 (discussing Zhen); Sycamore Mem. at 46-50 (discussing Gangaiah and Lemberg).

Defendants' argument has no merit. There is no evidence that any of the conduct here was the result of bad faith or an attempt to deceive defendants or the court. For example, Kasbarian's destruction of trading records occurred before he was aware of the possibility of a lawsuit; he had no reason to believe he would need those records. Such conduct does not render Kasbarian inadequate to prosecute the interests of the class. Furthermore, none of the inconsistencies or omissions complained of by defendants, such as failing to disclose certain specific transactions, affect the merits of the class representatives' manipulation claims. Given the complexity of these actions, minor testimonial inconsistencies and omissions are likely to occur. Only if "the problems alleged call the validity of the plaintiffs' entire case into question" do such credibility issues merit denial of class certification. The temporary omission of certain transactions from class representatives' disclosures does not call into question the overall validity of their claim that they lost money because of defendants' manipulation of securities markets. Denial of certification on credibility grounds is not warranted. 4. Ascertainability

See Kasbarian Dep. at 113.

Harrison v. Great Springwaters of Am., Inc., No. 96 Civ. 5110, 1997 WL 469996, at *6 (E.D.N.Y. June 18, 1997).

A number of courts have reached the same conclusion. See, e.g., Saddle Rock Partners, 2000 WL 1182793, at *5; In re Frontier Ins. Group, 172 F.R.D. at 41; Kalodner v. Michaels Stores, Inc., 172 F.R.D. 200, 210 (N.D. Tex. 1997); In re Consumers Power Co. Sec. Litig., 105 F.R.D. 583, 605-06 (E.D. Mich. 1985); see also Plaintiffs' Reply at 111-18.

Ascertainability, not ascertainment, is a prerequisite to class certification. Accordingly, at this stage of the proceedings, plaintiffs need not present an airtight method of identifying every class member who may be entitled to a recovery. Rather, the goal at this stage is to define a class that excludes, with broad strokes, segments of the proposed class that are not so entitled. Precise identification of every class member may be accomplished at a later stage.

See, e.g., Dunnigan v. Metropolitan Life Ins. Co., 214 F.R.D. 125, 135 (S.D.N.Y. 2003) ("Class members need not be ascertained prior to certification, but must be ascertainable at some point in the case.").

See id.; Dorchester Investors v. Peak Trends Trust, No. 99 Civ. 4696, 2002 WL 272404, at *6 (S.D.N.Y. Feb. 26, 2002) (acknowledging that class members who knew of the alleged scheme cannot recover); id. at 6 n. 6 (certifying class even in light of "Defendants['] argu[ment] that because neither those who sell short nor those who knew about the short selling scheme can be readily identified, the class should not be certified," stating that "[t]he Court rejects this argument because, as discussed above, certifying the class is the most efficient method available for [a securities] class action. The parties must use the available discovery mechanisms to determine who falls in or out of the class. . . .").

Defendants impliedly argue that because the Federal Rules of Civil Procedure were revised in 2003 to eliminate the availability of "conditional certification," perfect ascertainment of class members should be a prerequisite of class certification. This is not so. The 2003 revisions to Rule 23 do not require identification of every class member prior to certification. Rather, to certify a class, a court must simply be "satisfied that the requirements of Rule 23 have been met." The court may alter or amend the certification order, or even decertify the entire class, at any point before final judgment if the need arises. To require the identification of all class members at the class certification stage would impermissibly require a determination, on the merits, of the validity of each proposed class member's claim.

See Class Def. Opp. at 2 n. 1 (challenging plaintiffs' citation of Dorchester, 2002 WL 272404, on the ground that Dorchester "merely held — when conditional certification was still permissible — that questions about the class period and the knowledge of a small number of investors should not bar an initial certification ruling that would be modified as appropriate in light of discovery").

Fed.R.Civ.P. 23(c)(1)(C) Advisory Committee Note.

See id. ("[F]or example, proceedings to define the remedy may demonstrate the need to amend the class definition or subdivide the class. . . . A determination of liability after certification . . . may show a need to amend the class definition. Decertification may be warranted after further proceedings.").

See In re MTBE, 209 F.R.D. at 337 n. 20 (quoting Forbush, 994 F.2d at 1104-06).

"[I]t is axiomatic that one cannot commit a fraud . . . against oneself." This truism takes on special importance when the participation of certain investors ( i.e., those who engaged in laddering and paid undisclosed compensation) is integral to the alleged scheme. It should be noted that this inquiry — which seeks to ascertain which investors could not have been defrauded because of their actual knowledge of the alleged scheme — is not the same as the question of which investors knew enough that they could not have relied on the market price of securities as an accurate measure of their intrinsic value. That question is one of predominance, not ascertainability. Plaintiffs concede that investors who knowingly participated in the alleged scheme have no right to recover. Plaintiffs' revised class definition seeks to exclude these investors.

In re RCS Engineered Products Co., Inc., 102 F.3d 223, 226 (6th Cir. 1996). See also Gurary v. Winehouse 190 F.3d 37, 45 (2d Cir. 1999) (a plaintiff "must establish that he or she engaged in a securities trade in ignorance of the fact that the price was affected by the alleged manipulation.").

See Hevesi v. Citigroup, Inc., 366 F.3d 70, 77 (2d Cir. 2004).

See infra Part IV.B.1.b.(3).

See Plaintiffs' Reply at 5 ("As a result, the Class definition eliminates anyone who knowingly participated in the alleged misconduct.").

Of course, plaintiffs' suggestion is not binding; it is the Court, not plaintiffs, that ultimately decides what class to certify. See, e.g., Gibson v. Local 40, Supercargoes Checkers of the ILWU, 543 F.2d 1259, 1264 (9th Cir. 1976) (narrowing proposed class definition); Taylor v. Safeway Stores, Inc., 524 F.2d 263, 269 (10th Cir. 1975) (same) overruled on other grounds by Ruckelshaus v. Sierra Club, 463 U.S. 680 (1983) ; Martin v. American Med. Sys., Inc., No. IP 94-2067-C-H/G, 1995 WL 680630, at *6 (S.D. Ind. Oct. 25, 1995) ("The court need not merely accept or reject plaintiffs' proposed class definition without considering modifications of that definition.").

The first and most important inquiry in determining which groups of investors to exclude on the basis of actual knowledge is the question of what the scheme entails. In this case, plaintiffs have alleged that defendants engaged in the following scheme to manipulate the market:

15. The Underwriter Defendants set about to ensure that there would be large gains in aftermarket trading on shares following initial public offerings by improperly creating artificial aftermarket demand They accomplished this by conditioning share allocations in initial public offerings upon the requirement that customers agree to purchase, in the aftermarket, additional shares of stocks in which they received allocations, and, in some instances, to make those additional purchases at pre-arranged, escalating prices ("Tie-in Agreements").

. . .

16. By extracting agreements to purchase shares in the aftermarket, the Underwriter Defendants created artificial demand for aftermarket shares, thereby causing the price of the security to artificially escalate as soon as the shares were publicly issued.
17. Not content with record underwriting fees obtained in connection with new offerings, the Underwriter Defendants sought, as part of their manipulative scheme, to further enrich themselves by improperly sharing in the profits earned by their customers in connection with the purchase and sale of IPO securities. The Underwriter Defendants kept track of their customers' actual or imputed profits from the allocation of shares in the IPOs and then demanded that the customers share a material portion of the profits obtained from the sale of those allocated IPO shares through one or more of the following types of transactions: (a) paying inflated brokerage commissions; (b) entering into transactions in otherwise unrelated securities for the primary purpose of generating commissions; and/or (c) purchasing equity offerings underwritten by the Underwriter Defendants, including, but not limited to, secondary (or add-on) offerings that would not be purchased but for the Underwriter Defendants' unlawful scheme. (Transactions "(a)" through "(c)" above will be, at varying times, collectively referred to hereinafter as "Undisclosed Compensation").

MA ¶¶ 14, 16-17.

Clearly, the laddering scheme plaintiffs allege includes three necessary components: the tie-in agreements, the undisclosed compensation, and the escalation in share prices caused by artificial demand Accordingly, an investor can only be said to have full knowledge of the alleged laddering scheme if she is aware of all three components.

Defendants complain that "[i]dentifying claimants with knowledge would be a massive undertaking in light of plaintiffs' assertion that thousands of [investors] participated in the alleged manipulation in hundreds of offerings." Defendants base their assertion of widespread participation on plaintiffs' own allegations, which read in pertinent part:

30. Institutional and retail investors, who have received allocations in initial public offerings from various firms, have noted that it was common knowledge that the clients who were forced to pay Undisclosed Compensation to the underwriters, in the form of commissions or otherwise, and who agreed to purchase in the aftermarket received allocations in IPOs.
31. This industry-wide understanding was sometimes expressed by the Underwriter Defendants and other times implied, but nevertheless invariably communicated between those with the power to make allocations of shares in initial public offerings (the underwriters) and customers seeking the allocations.

Defendants' Sur-Reply at 5. Defendants assert that exclusion of class members who had actual knowledge of the alleged scheme would require "claimant-by-claimant inquiry at trial." Id. at 6. This argument involves elements both of ascertainability and predominance. Essentially, defendants argue that if plaintiffs cannot at this stage provide a means for excluding all class members with actual knowledge, then trial of these actions will be dominated by individual inquiries into whether each class member had knowledge. However, if plaintiffs can plead an ascertainable class, then individual determinations of actual knowledge — ultimately inquiries into who belongs in the class — will not predominate at trial. Whether the analysis is denominated as "ascertainability" or "predominance," the outcome is the same.

MA ¶¶ 30-31.

Defendants' concerns are unfounded. First, a close look at these paragraphs is absolutely necessary in view of defendants' argument. Paragraph 30 reveals that the allegation is only that "investors [who are allocants] have noted that. . . ." Thus, the pleading is not that "everyone knew of the scheme" but rather that some allocants "noted" that certain information was common knowledge. This is not a judicial admission by plaintiffs that "everyone knew of the scheme." In addition, one must look closely at what these investors say was common knowledge. They say that it was common knowledge that investors who paid undisclosed compensation and agreed to purchase in the aftermarket received allocations. This is not surprising. But they do not say that it was common knowledge that the price of stock was artificially inflated through illegal tie-in arrangements that required a large percentage of allocants to pay undisclosed compensation and to agree to make a certain number of purchases in the aftermarket at escalating prices in order to obtain an allocation. That is the guts of the scheme now alleged and nothing in paragraph 30 pleads that such a scheme was commonly known by the investing public.

Indeed, even the broadest reading of paragraphs 30 and 31 does not support a conclusion that analyst conflicts — one part of the coherent fraudulent scheme alleged — were common knowledge.

The same is true of paragraph 31. The paragraph begins with the words "this industry-wide understanding." This raises the question — to what does the word "this" refer? The natural reading is that it refers back to the immediate prior paragraph so that "this industry-wide understanding" is that investors who paid undisclosed compensation and agreed to purchase in the aftermarket received allocations. Paragraph 31 merely pleads that the underwriters made it known that those who paid undisclosed compensation and agreed to purchase stock in the aftermarket received allocations — not that such investors were aware of an illegal scheme to inflate stock prices.

Second, even if plaintiffs' allegations are construed as broadly as possible, they do not suggest that many investors knew of the entire scheme alleged. Nowhere do plaintiffs allege that allocants were aware that such agreements were part of an industry-wide scheme to inflate share prices through the creation of artificial demand Many allocants may have been defendants' unwitting tools, each performing certain acts ( i.e., paying undisclosed compensation and agreeing to purchase in the aftermarket) that only when aggregated constituted a cohesive scheme to defraud investors. Even to the extent that allocants might have suspected illegality, that wrongdoing could be ascribed a clear and direct goal — the enrichment of the Underwriter defendants through payment of excessive compensation and increased business ensured by tie-in agreements — not the indirect scheme to defraud investors by artificially driving up securities prices alleged here. As plaintiffs' counsel has noted, it is unlikely indeed that investors who had full knowledge of the alleged scheme would retain their shares for any length of time after the securities' immediate price gains if they knew that the heavy demand had artificially inflated the price, and that the artificial inflation would inevitably dissipate over time. Finally, plaintiffs' counsel has explained that, contrary to defendants' assertions that the scheme was "common knowledge" and "invariably communicated" to allocation-seekers, only a limited population of allocants actually paid undisclosed compensation or consummated tie-in agreements:

See Transcript of 6/17/04 Hearing, 49:3-8 ("[MR. BROWER:] Simply, persons who participated in the scheme, those who had knowledge are unlikely to have lost money net trading in the stock in which they received an allocation, because even if they did aftermarket transactions as part of tie-in agreements, they still wouldn't trade themselves to a loss.").

MR. WEISS: We're not saying that all allocants were subjected to this kind of requirement for laddering and kickbacks. There's a certain small universe, but we say that the universe was sufficient to be able to doctor this market and to create huge additional compensation that was undisclosed for these underwriters; a scheme, information that was never disclosed by . . . the defendants throughout the class period. The population of those allocants who participated is a relatively small population. . . .

Id. at 54:10-19.

This position is more consistent with information gleaned through the discovery process than is the notion that every customer who ever expressed an interest in an allocation somehow became privy to the alleged scheme.

See Plaintiffs' Reply at 47 ("[T]he uncontroverted testimony of all 17 [named] Plaintiffs [several of whom received allocations] is that each was entirely ignorant of the manipulation when each purchased his or her shares.").

After reviewing plaintiffs' new proposed class definition, and considering the traits most likely to separate investors who knew of the alleged scheme from those who did not know, the following represents an ascertainable class:

The Class consists of all persons and entities that purchased or otherwise acquired the securities of [Specific Issuer] during the Class Period and were damaged thereby. Excluded from the Class are:
(1) Defendants herein, each of their respective parents, subsidiaries, and successors, and each of their respective directors, officers and legal counsel during the Class Period, and each such person's legal representatives, heirs, and assigns, members of each such person's immediate family, and any entity in which such person had a controlling interest during the Class Period;
(2) all persons and entities that, with respect to [Specific Issuer's] initial public offering: (a) received an allocation, (b) placed orders to purchase shares of that issuer's securities in the aftermarket within four weeks of the effective date of the offering, (c) paid any undisclosed compensation to the allocating underwriter(s), and (d) made a net profit (exclusive of commissions and other transaction costs), realized or unrealized, in connection with all of such person's or entity's combined transactions in [Specific Issuer's] securities during the Class Period; and
(3) all persons and entities who satisfy all of the requirements of subparagraph (2) with respect to any of the 309 initial public offerings that are the subject of these coordinated actions, if that offering occurred prior to [Specific Issuer's] offering.

The structure of the class definition is based on that of the proposed definition submitted by plaintiffs in response to my June 21 Order. See Class Def. Letter at 1. However, the content of this definition differs substantially from plaintiffs' proposal.

This definition applies equally to plaintiffs' section 11 classes, because knowledge also precludes recovery under that statute. See 15 U.S.C. § 77(k)(a) (granting right of recovery to "any person acquiring such security (unless it is proved that at the time of such acquisition he knew of such untruth or omission)").

As I have previously noted, the ascertainability inquiry does not demand ascertainment at the class certification stage. Certain investors not automatically stricken by the class definition may later prove to have actual knowledge of the alleged scheme. Any securities fraud class action runs the risk of including individual investors who may be ineligible for recovery for any number of reasons, including actual knowledge of the alleged fraud. If the possibility that certain class members might eventually be excluded were sufficient to preclude class certification, there could never be a securities fraud class action. At trial, defendants may choose to bear the burden and the cost of proving that any particular investor had access to nonpublic information that gave that investor actual knowledge of the alleged scheme.

It is well-established that defendants in a class action may contest the claims of individual class members even when they are included within the class definition. See, e.g., Basic Inc. v. Levinson, 485 U.S. 224, 249 (1988) (acknowledging possibility of rebutting presumption of reliance as to individual class members); In re Ski Train Fire in Kaprun, Austria on November 11, 2000, 220 F.R.D. 195, 208 (S.D.N.Y. 2003) (certifying wrongful death class action, noting "[t]he determination of which heir (or heirs) of the decedents is entitled to a judgment can be decided during the damages phase of the trial.").

One of the interesting side effects of the class action form is that, in some cases, it effectively transfers the burden of proving individual facts from plaintiffs to defendants. As the Supreme Court has noted, "the Advisory Committee had dominantly in mind [in enacting Rule 23(b)(3)] vindication of the rights of groups of people who individually would be without effective strength to bring their opponents into court at all." Amchem Prods., 521 U.S. at 617 (quotation marks and citation omitted); accord Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 813 (1985) ("The plaintiff's claim may be so small . . . that he would not file suit individually. . . ."). Just as a plaintiff who suffered a small loss may not have the resources or motivation to bring an individual suit, a defendant in a 23(b)(3) class action is forced to make the difficult decision whether it is worth expending resources to prove that a small-stake class member should be excluded from recovery. In either situation, the inevitable cost of litigating an issue forces a party to weigh the expected benefit (subtracting transaction costs) against the cost of losing if she does not litigate the issue. Rule 23 simply transfers that cost-benefit analysis from the alleged victim to the alleged wrongdoer.

The class definition broadly excludes those investors who exhibit the hallmarks of full participation in the alleged scheme. Defendants are alleged to have defrauded investors by manipulating the conduct of allocants, both in terms of the compensation they paid and their aftermarket activity, thereby creating a market where the aggregate demand caused by tie-in agreements artificially inflated the price of the stock. The class excludes those who engaged in the acts alleged to have driven up securities prices, and who exhibited their knowledge of the overall scheme by selling their shares for a profit before the effects of the scheme dissipated. The class definition further excludes those who had knowledge of the scheme in one case from participating in the classes for any subsequent IPOs in these consolidated actions, because an investor who has knowledge of the alleged fraud in one offering cannot erase that knowledge thereafter.

See MA ¶¶ 14-17.

Defendants assert that applying plaintiffs' proposed exclusions, which are similar (although not identical) to those just enumerated by the Court, will present serious manageability problems because the information to be gathered with respect to each allocant "would be scattered in multiple formats among many firms" and "would have to be repeated for each of the few thousand allocants in a single case, and for each of the thousands in 309 cases." However, the requirement is ascertainability — not ascertainability with case. Plaintiffs in a class action meet their burden by pleading a class whose membership is ascertainable, even if actual ascertainment might prove "slow and burdensome. Here, plaintiffs note that the class definition factors "are all mathematically certain and objectively determinable," and that the documentary evidence required to apply the proposed definition "is legally required to be retained by broker-dealers" and includes "customer monthly statements, trade confirmations, and order tickets."

Class Def. Opp. at 4. Defendants also launch other attacks on the class exclusions, noting particularly that "plaintiffs propose no method to count `repeat conduct' by multi-faceted `entities,'" and that some non-party brokers may no longer retain records of allocants' trading behavior. Id. However, plaintiffs are not required to anticipate and address all possible problems that may arise as discovery continues; all that is required at this stage is a method of ascertaining class members that can be manageably applied. If defendants' predictions come true, and the effect is disabling, then they may properly move to amend this certification order or to decertify the class. See Fed.R.Civ.P. 23(c)(1)(C).

Dunnigan, 214 F.R.D. at 136.

Class Def. Letter at 2.

Although defendants mount several attacks on whether the proposed class definition will successfully exclude those with actual knowledge of the alleged scheme, only three require comment. First, defendants complain that the class definition "is based only on investor conduct in the `309' IPOs and thus fails to account for knowledge acquired or shown through participation in the 87 follow-on offerings" conducted by 82 issuers in these consolidated actions. Defendants make a good point. An investor who participated in an IPO or traded in its aftermarket in ignorance of the alleged scheme, but later exhibited knowledge of the alleged scheme in connection with a follow-on offering, should be charged with knowledge of the scheme only after her knowing participation. Consequently, the class exclusions apply to participants in follow-on offerings, but only exclude those participants with respect to trades executed after they satisfy the class exclusion criteria.

Class Def. Opp. at 3 n. 3 (quoting Plaintiffs' Reply at 2).

Second, defendants claim that the class definition does not adequately exclude investors who had knowledge of the fraudulent scheme through participation in "the `more than 900' IPOs that plaintiffs allege were manipulated as part of this purported industry wide scheme," but 600 of which are not part of these consolidated actions. I note, however, that defendants have vehemently opposed suggestions that they produce any discovery whatsoever with respect to any IPOs other than those consolidated here. Defendants may not now have it both ways; if plaintiffs do not obtain full discovery in the IPOs that are not in suit, then defendants are barred from making this argument. On the other hand, if defendants now believe that it would be beneficial to alter the boundaries of this case to give plaintiffs access to discovery in the approximately 600 remaining IPOs, this issue can be revisited. Otherwise, defendants' argument is without merit.


See, e.g., Transcript of 3/5/03 Hearing at 46:3-6 ("MR. DiBLASI: Your Honor, the notion that [plaintiffs' counsel] intends to introduce discovery issues relating to the [6]00 other IPOs makes absolutely no sense to any of the underwriters, and we will oppose that every way we can."); 10/29/03 Letter from DiBlasi to Weiss at 2 ("The burdens associated with Plaintiffs' proposed change in approach [ i.e., requiring defendants to provide additional discovery outside the 309 cases] would be enormous and unfair."); Transcript of 12/11/03 Hearing at 25:16-18 ("MR. ICHEL: The parties are spending incredible amounts of time on discovery in just six class focus cases. So by bringing in additional IPOs beyond the 310 it makes it unmanageable."). Plaintiffs have also relied on the limitation to the 309 consolidated actions. See, e.g., 8/19/03 Letter to the Court from Stanley Bernstein, counsel for plaintiffs; 8/23/03 Letter from Bernstein; 10/14/03 Letter from Bernstein. It makes no sense to allow defendants to offer evidence that potential class members participated in the alleged scheme in the 600 IPOs not at issue here while simultaneously precluding plaintiffs from taking discovery on those IPOs to determine the extent of each defendant's participation in the alleged scheme.

Third, defendants note that "the proposal would not exclude those participants who supposedly paid undisclosed compensation through `churned' or `wash' transactions or high volume trades or those who allegedly obtained allocations by purchasing shares in `undesired add-on offerings.'" Defendants are simply wrong. Allocants who paid undisclosed compensation — in whatever form — are excluded if they also purchased in the aftermarket and profited from their investments. The only question is whether such transactions amount to undisclosed compensation. That is a common question of law, not an ascertainability problem.

Class Def. Opp. at 4 n. 3 (quoting MA ¶¶ 42-43).

Accordingly, plaintiffs' class is ascertainable.

B. Rule 23(b): Predominance

Defendants challenge plaintiffs' proposed class on the grounds that individualized questions will predominate at trial. Defendants' predominance arguments fall into four major categories: transaction causation, loss causation, damages, and section 11 liability. As discussed earlier, plaintiffs' cases offer a wealth of common issues. With the exception of defendants' arguments regarding section 11 tracing (which are limited to the duration of the section 11 classes), none of defendants' arguments defeat plaintiffs' showing of predominance.

See supra Part IV.A.1.

1. Transaction Causation

"Like reliance, transaction causation refers to the causal link between the defendant's misconduct and the plaintiff's decision to buy or sell securities. It is established simply by showing that, but for the claimed misrepresentations or omissions, the plaintiff would not have entered into the detrimental securities transactions." Plaintiffs may avail themselves of a rebuttable presumption of reliance under the following theories.

Emergent Capital Inv. Mgmt., LLC v. Stonepath Group, Inc., 343 F.3d 189, 197 (2d Cir. 2003) (citations omitted).

a. The Affiliated Ute Presumption

"In securities fraud claims, reliance is presumed when the claim rests on the omission of a material fact." This presumption of reliance is not conclusive. Rather, "once the plaintiff establishes the materiality of the omission . . . the burden shifts to the defendant to establish . . . that the plaintiff did not rely on the omission in making the investment decision." To satisfy this burden, a defendant must prove "that `even if the material facts had been disclosed, plaintiff's decision as to the transaction would not have been different from what it was.'"

In re Worldcom, 219 F.R.D. at 291 (citing Affiliated Ute Citizens v. United States, 406 U.S. 128, 153-54 (1972). Accord Castellano v. Young Rubicam, Inc., 257 F.3d 171, 186 (2d Cir. 2001); Press v. Chemical Inv. Servs. Corp., 166 F.3d 529, 539 (2d Cir. 1999)).

See DuPont v. Brady, 828 F.2d 75, 78 (2d Cir. 1987).

Id. at 76.

Id. at 78 (quoting Rochez Bros. v. Rhoades, 491 F.2d 402, 410 (3d Cir. 1974)).

Defendants attempt to distinguish Affiliated Ute on the following grounds: " Affiliated Ute was not a class action, did not involve alleged market manipulation, was not deemed applicable to the manipulation and misrepresentation claims asserted in Basic, and would [still] require" that plaintiffs demonstrate the materiality of the omissions and their ignorance of the omitted facts. While a court need not address every argument it rejects, a few observations are in order. The Second Circuit has applied Affiliated Ute in the class action context. Moreover, while Basic adopted the "fraud on the market" presumption, it contains no language disfavoring Affiliated Ute where both market manipulation and material omissions are alleged. Rather, Basic approved the Affiliated Ute presumption and presented the "fraud on the market" presumption alongside it in the panoply of securities fraud-related presumptions. Finally, the materiality of the alleged omissions here ( i.e., the total nondisclosure of the alleged scheme) has not been disputed. Plaintiffs are entitled to an Affiliated Ute presumption of reliance to the extent their 10b-5 claims derive from material omissions.

Class Def. Opp. at 5 n. 5.

See Handwerger v. Ginsberg, 519 F.2d 1339, 1341-42 (2d Cir. 1975) (holding, in class action context, that Affiliated Ute eliminated "[t]he requirement of proving individual reliance . . . at least as to claims of fraudulent omissions brought under s 10(b) and Rule 10b-5.").

See Basic, 485 U.S. at 243 ("There is, however, more than one way to demonstrate the causal connection. Indeed, we previously have dispensed with a requirement of positive proof of reliance, where a duty to disclose material information had been breached, concluding that the necessary nexus between the plaintiffs' injury and the defendant's wrongful conduct had been established. See Affiliated Ute Citizens v. United States. . . ."); id. at 245 ("Requiring a plaintiff to show a speculative state of facts, i.e., how he would have acted if omitted material information had been disclosed, see [ Affiliated Ute] . . ., would place an unnecessarily unrealistic evidentiary burden on the Rule 10b-5 plaintiff who has traded on an impersonal market.").

b. The Fraud on the Market Presumption

Plaintiffs may also avail themselves of a presumption of reliance, under the "fraud on the market" theory, for claims arising from alleged misrepresentations and market manipulation.

The 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 no less significant than in a case of direct reliance on misrepresentations.

Id. at 241-42 (quotation marks and citation omitted, alterations in original).

"The fraud-on-the-market doctrine, as described by the Supreme Court in Basic v. Levinson, creates a rebuttable presumption that (1) misrepresentations by an issuer affect the price of securities traded in the open market, and (2) investors rely on the market price of securities as an accurate measure of their intrinsic value." A defendant, of course, may rebut the fraud on the market presumption by showing that it made no material misrepresentations because the alleged misrepresentations were already known to the market — a so-called "truth on the market" defense.

Hevesi, 366 F.3d at 77. In Hevesi, the Second Circuit noted that "[a]lthough the fraud-on-the-market doctrine clearly applies to statements made by issuers, as in Basic, we have never addressed whether it also applies to reports by analysts." Id. (emphasis in original). In this case, plaintiffs' allegations that conflicts of interest led analysts to issue improperly glowing reports on the manipulated securities do not reflect the whole of plaintiffs' theory of liability; rather, such fraudulent reports are alleged in connection with a larger scheme to artificially inflate prices. Accordingly, plaintiffs' claims are not dependent upon a finding that they are entitled to a presumption of reliance on analyst reports; if plaintiffs prove that the scheme as a whole artificially inflated prices, then they may employ the fraud-on-the-market presumption to prove that they relied on those prices "as a measure of their intrinsic value." Id.

See Ganino v. Citizens Utils. Co., 228 F.3d 154, 167 (2d Cir. 2000) ("Under [the truth on the market theory], a misrepresentation is immaterial if the information is already known to the market because the misrepresentation cannot then defraud the market. A defendant may rebut the presumption that its misrepresentations have affected the market price of its stock by showing that the truth of the matter was already known. However, the corrective information must be conveyed to the public `with a degree of intensity and credibility sufficient to counter-balance effectively any misleading information created by' the alleged misstatements.") (citing Provenz v. Miller, 102 F.3d 1478, 1492 (9th Cir. 1996); Associated Randall Bank v. Griffin, Kubik, Stephens Thompson, Inc., 3 F.3d 208, 213-14 (7th Cir. 1993)).

(1) Market Efficiency

The fraud on the market presumption only applies if the market for the security is open and developed enough that it quickly incorporates material information into the price of the security — in other words, the market must be an "efficient" one. Defendants object that plaintiffs have not met their evidentiary burden of showing that the markets for the stocks in the focus cases were efficient.

See, e.g., Freeman v. Laventhol Horwath, 915 F.2d 193, 197 (6th Cir. 1990) ("The fraud on the market theory rests on the assumption that the price of an actively traded security in an open, well-developed, and efficient market reflects all the available information about the value of a company.") (citation omitted). Definitions of the relevant economic terms are provided in Cammer v. Bloom, 711 F. Supp. 1264, 1276 n. 17 (D.N.J. 1989): "`An open market is one in which anyone, or at least a large number of persons, can buy or sell. . . . A developed market is one which has a relatively high level of activity and frequency, and for which trading information (e.g., price and volume) is widely available. . . . An efficient market is one which rapidly reflects new information in price. These terms are cumulative in the sense that a developed market will almost always be an open one. And an efficient market will almost invariably be a developed one.'" (quoting BROMBERG LOWENFELS, 4 SECURITIES FRAUD AND COMMODITIES FRAUD, § 8.6 (Aug. 1988)).

See 6/8/04 Letter from DiBlasi to the Court at 2 ("Plaintiffs have offered the Court no evidence that the relevant markets were efficient at the time of the offerings or later in the `internet bubble' environment.").

The Second Circuit has not adopted a test or method for determining whether the market for a security is efficient. Nonetheless, the record in this case contains several strong indications that the market in which the focus stocks traded was efficient. Three facts stand out as particularly probative: first, all the focus stocks were traded on the NASDAQ National Market; second, the focus stocks were traded actively at high volumes throughout the class period; and third, the focus stocks were the subjects of numerous analyst reports and extensive media coverage. Under any conceivable test for market efficiency, these three facts are sufficient to meet plaintiffs' Rule 23 burden to make "some showing" that the stocks in question traded on an efficient market.

The Cammer court identified five factors that would be useful in proving an efficient market: (1) a large weekly trading volume; (2) the existence of a significant number of analyst reports; (3) the existence of market makers and arbitrageurs in the security; (4) the eligibility of the company to file an S-3 registration statement; and (5) a history of immediate movement of the stock price caused by unexpected corporate events or financial releases. See 711 F. Supp. at 1286-87. Several courts have used this approach. See Binder v. Gillespie, 184 F.3d 1059, 1065 (9th Cir. 1999) (adopting the Cammer approach); Hayes v. Gross, 982 F.2d 104, 107 (3d Cir. 1992) (same); Freeman, 915 F.2d at 199 (same); Krogman v. Sterritt, 202 F.R.D. 467, 474 (N.D. Tex. 2001) (employing three factors from O'Neil v. Appel, 165 F.R.D. 479 (W.D. Mich. 1996), in addition to the Cammer approach); O'Neil, 165 F.R.D. at 503 (suggesting additional factors from the economic literature to supplement the Cammer approach). But see In re PolyMedica Corp. Sec. Litig., No. CIV.A. 00-12426-REK, 2004 WL 1977530, at *14 (D. Mass. Sept. 7, 2004) (rejecting Cammer and similar cases for unfairly reading economic definitions into Basic's efficient market requirement, noting that "the relevant question is whether the market . . . is one in which market professionals generally consider most publicly announced material statements about [the issuer], thereby affecting [its] stock market price") (quotation marks, alterations and citations omitted). Here, whether the Cammer test or the broader definition of efficiency adopted by In re PolyMedica is applied, the outcome is the same.

Federal courts have repeatedly held that a listing on NASDAQ or a similar national market is a good indicator of efficiency. See, e.g., Stevelman v. Alias Research, Inc., No. 5:91-CV-682, 2000 WL 888385, at *4 (D. Conn. June 22, 2000) ("For stocks . . . that trade on a listed exchange such as NASDAQ, [the] reliance element of a 10b-5 cause of action is presumed."); Levine v. Skymall, Inc., No. CIV. 99-166-PHX-ROS, 2002 WL 31056919, at *5 (D. Ariz. May 24, 2002) ("Although not dispositive, the fact that SkyMall stock is traded on the NASDAQ stock market's National Market System also contributes to finding that the market is efficient."); RMED Intern., Inc. v. Sloan's Supermarkets, Inc., 185 F. Supp. 2d 389, 404-05 (S.D.N.Y. 2002) ("Indeed, research has failed to reveal any case where a stock traded on the AMEX was found not to have been traded in an open and efficient market. . . . Rather, to the contrary, numerous courts have held that stocks trading on the AMEX are almost always entitled to the presumption.") (citations omitted); O'Neil, 165 F.R.D. at 504 ("The market system upon which a particular stock trades provides some insight as to the likelihood that the market for that stock is efficient. . . .").

Ultimately, whether the relevant markets were efficient is a question of fact to be resolved at trial. The present finding — that plaintiffs have made "some showing" that the focus markets were efficient — is solely for the purposes of adjudicating the pending motion for class certification, and is not binding on the finder of fact. Based on the evidence presented at trial, the finder of fact may conclude that the relevant markets were efficient, in which case all class members will benefit from a presumption of reliance. On the other hand, the finder of fact may conclude that one or more of the relevant markets was inefficient, in which case those plaintiffs who traded in such markets would be required to make individual showings of reliance.

See, e.g., Basic, 485 U.S. at 249 n. 29 ("Proof [rebutting a presumption of reliance] is a matter for trial, throughout which the District Court retains the authority to amend the certification order as may be appropriate. See Fed.R.Civ.P. 23(c)(1) and (c)(4). See 7B C. WRIGHT, A. MILLER, M. KANE, FEDERAL PRACTICE AND PROCEDURE 128-132 (1986). Thus, we see no need to engage in the kind of factual analysis the dissent suggests that manifests the `oddities' of applying a rebuttable presumption of reliance in this case."); In re Ashanti Goldfields Sec. Litig., No. CV 00-0717, 2004 WL 626810, at *16 (E.D.N.Y. Mar. 30, 2004) ("[P]roof of market inefficiency . . . or rebuttal of the presumption of reliance is best left to the trial phase of litigation.") (citing Basic, 485 U.S. at 248 n. 29); RMED Int'l, Inc. v. Sloan's Supermarkets, Inc., No. 94 Civ. 5587, 2002 WL 31780188, at *4 (S.D.N.Y. Dec. 11, 2002) ("Whether or not a market for a stock is open and efficient is a question of fact.") (citing In re Laser Arms Corp. Sec. Litig., 794 F. Supp. 475, 490 (S.D.N.Y. 1989) (holding that "[w]hether in fact Laser Arms traded in an efficient market is a question of fact. Therefore, resolution of that issue must await presentation of further proof at trial."), aff'd, 969 F.2d 15 (2d Cir. 1992)). While these cases all pre-date the 2003 amendments to Rule 23 forbidding conditional certification, the new Rule still permits a court to decertify a class or amend the certification as necessary.

For example, the finder of fact might accept defendants' suggestion that the relevant markets were inefficient because they were part of the "internet bubble." See 6/8/04 DiBlasi Letter at 2.

(2) Investment Strategies

"[I]t has been noted that `it is hard to imagine that there ever is a buyer or seller who does not rely on market integrity. Who would knowingly roll the dice in a crooked crap game?'" Defendants believe there are such investors. Indeed, they claim that so many reckless gamblers engaged in a `crooked crap game,' and that exposing their folly would be such an arduous task, that any adjudication of their claims would require innumerable individual inquiries.

Basic, 485 U.S. at 246-47 (quoting Schlanger v. Four-Phase Sys. Inc., 555 F. Supp. 535, 538 (S.D.N.Y. 1982)).

Defendants assert that "thousands of day and momentum traders were not concerned about the integrity of a stock's market price," and argue that "[f]or both types of traders the integrity of the market price was irrelevant to the investor's decision to purchase." According to defendants, "subjective inquiries" into whether these traders actually relied on market integrity would cause individual issues to predominate. But day and momentum traders have the same incentives to prove defendants' liability as all other class members, and their presence in a securities class does not create intra-class conflicts.

iXL Mem. at 20-21 (quotations omitted). Defendants define "day traders" as follows: "Day traders . . . focus solely on price volatility. They `hope that their stocks will continue climbing or falling in value for the seconds to minutes they own the stock.'" iXL Mem. at 12-13 (quoting SEC, Day Trading: Your Dollars at Risk, (Aug. 23, 2004)). Defendants define "momentum traders" by comparison: "`Momentum traders' likewise buy `stocks simply because they're going up in price.' . . . Though they may hold stock longer than a day trader, they similarly wish to take advantage of price movement — even movement due to a bubble or manipulation." Id. at 13 (quoting Perkins Perkins, The Internet Bubble 25 (1999)).


See, e.g., In re Oxford Health Plans, Inc., 191 F.R.D. 369, 377 (S.D.N.Y. 2000) (rejecting purported intra-class conflicts between in-and-out investors and those that held their shares throughout the period, noting that "`common questions [of fact and law] . . . bind class members with more force than the varying questions related to price inflation drive them apart.'" (quoting In re Gaming Lottery Sec. Litig., 58 F. Supp. 2d 62, 69-71 (S.D.N.Y. 1999)) (alterations in original); Saddle Rock Partners, 2000 WL 1182793, at *5 (addressing defendants' concerns that a proposed day trading class representative would not adequately assert the element of reliance, stating that "the fact that he may have traded Maybelline shares on the basis of short term price drops which he believed to reflect market inefficiencies indicates that he may have been relying on the integrity of the market to establish the more stable, longer term price.").

Similarly, defendants challenge plaintiffs' proposed classes on the grounds that they may contain short sellers, and that, "[b]ecause short sellers do not rely on the market price, they do not enjoy a presumption of reliance." Defendants cite the Third Circuit's decision in Zlotnick v. TIE Communications in support of this contention. But Zlotnick does not control. Not only is Zlotnick a Third Circuit case (and therefore not binding on this Court), it pre-dates the Supreme Court's seminal opinion in Basic. Indeed, in cases like this one, courts in the Third Circuit and elsewhere have almost unanimously rejected the Zlotnick exception. One such court noted that:

Moreover, under defendants['] view of the case, any plaintiff seeking to represent a class of investors of a large, publicly traded corporation would be unable to satisfy reliance, and, hence, typicality, as a matter of law. . . . It can be stated without fear of gainsay that the shareholders of every large, publicly traded corporation includes [sic] institutional investors, short-sellers, arbitragers etc. The fact that these traders have divergent motivations in purchasing shares should not defeat the fraud-on-the-market presumption absent convincing proof that price played no part whatsoever in their decision making. If defendants believe that this stretches the concept of reliance beyond the intent of the statute, their course of attack is to overrule Basic, not render its holding meaningless.

"Short selling is accomplished by selling stock which the investor does not yet own; normally this is done by borrowing shares from a broker at an agreed upon fee or rate of interest. At this point the investor's commitment to the buyer of the stock is complete; the buyer has his shares and the short seller his purchase price. The short seller is obligated, however, to buy an equivalent number of shares in order to return the borrowed shares. In theory, the short seller makes this covering purchase using the funds he received from selling the borrowed stock. Herein lies the short seller's potential for profit: if the price of the stock declines after the short sale, he does not need all the funds to make his covering purchase; the short seller then pockets the difference. On the other hand, there is no limit to the short seller's potential loss: if the price of the stock rises, so too does the short seller's loss, and since there is no cap to a stock's price, there is no limitation on the short seller's risk. There is no time limit on this obligation to cover. `Selling short,' therefore, actually involves two separate transactions: the short sale itself and the subsequent covering purchase." Zlotnick v. TIE Communications, 836 F.2d 818, 820 (3d Cir. 1988).

iXL Mem. at 21.

See id. at 21 (citing Zlotnick, 836 F.2d at 823).

See Argent Classic Convertible Arbitrage Fund L.P. v. Rite Aid Corp., 315 F. Supp. 2d 666, 676 n. 13 (E.D. Pa. 2004); Moskowitz v. Lopp, 128 F.R.D. 624, 630-31 (E.D. Pa. 1989) (citing Zlotnick for its discussion of the elements of the fraud-on-the-market presumption, but nevertheless applying the presumption to a class including short-sellers); In re W. Union Sec. Litig., 120 F.R.D. 629, 637 (D.N.J. 1988) ("While Zlotnick can arguably be seen as a cutting-back on the potential scope of [ Peil v. Speiser, 806 F.2d 1154 (3d Cir. 1986), in which the Third Circuit accepted the fraud on the market theory], we find its validity somewhat questionable in light of Basic, supra. Not only is Basic a later opinion of a superior court, it also makes several positive references to Peil, supra, the scope of which Zlotnick arguably constricts."). Defendants proffer only one recent case applying the Zlotnick exception, and that case used the exception solely to reject a proposed class representative whom the court had already found inadequate because he was subject to the unique defense that he had sold all of his shares for a profit. See Wiekel v. Tower Semiconductor Ltd, 183 F.R.D. 377, 392 (D.N.J. 1998).

Moskowitz, 128 F.R.D. at 631 (emphasis in original).

This analysis is far more persuasive than defendants' application of the Zlotnick exception, and it comports with the policy and practice of certifying securities class actions in this Circuit. Accordingly, the presence of short sellers does not undermine plaintiffs' showing of predominance.

See, e.g., In re Ames Dep't. Stores Inc. Stock Litig., 991 F.2d 953, 967 (2d Cir. 1993) (noting the general applicability of the fraud-on-the-market presumption to all investors, stating that "[b]ecause the fraud on the market may taint each purchase of the affected stock, each purchaser who is thereby defrauded (and, since the presumption is rebuttable, not all purchasers necessarily are defrauded by the information) is defrauded by reason of the publicly disseminated statement."); In re Worldcom, 219 F.R.D. at 296 ("The existence of short selling, even voluminous short selling . . . does not suggest that the presumption of reliance should not apply to those [short sellers] who purchased the [securities] and lost money.").

(3) Knowledge of Fraudulent Scheme

A presumption of reliance may be rebutted by a showing that the plaintiff had knowledge of the omitted fact or fraudulent scheme. "[I]f the plaintiff has been furnished with the means of knowledge and he is not prevented from using them he cannot say that he has been deceived by the misrepresentations of the other party."

Frigitemp v. Financial Dynamics, 524 F.2d 275, 282 (2d Cir. 1975) (applying common law principles of fraud in the context of Rule 10b-5) (citing Shappirio v. Goldberg, 192 U.S. 232, 241-42 (1904)).

Defendants note that "pervasive press reports mirrored the allegations in these cases," pointing to several occasions exposing class members, through the national media or official releases, to information which, defendants claim, "would have made any reader aware of the allegations here and put them on notice to inquire further." Defendants maintain that determining "which purchasers knew what, and when . . . will require . . . subjective inquiry into each claimant's state of mind."

Id. at 7. Defendants also assert that determining which class members had actual knowledge of the scheme through their participation in various IPOs will require subjective individual inquiries. This argument was addressed at Part IV.A.4., supra.

Id. While there is no debate regarding the content and distribution of these publications, which are described in Part II.F., supra, determining whether they placed investors on inquiry notice is a task better left to the finder of fact. Indeed, I note that the MSNBC article, while it includes the language cited by defendants, also features a flat denial from defendant Goldman Sachs that its firm engaged in any unlawful tie-in schemes. See Ex. A to Houck Decl., at A5-A6 ("An official at Goldman, which also underwrote the eToys IPO, would say only that the firm `does not make the allocation of IPO securities conditional on an undertaking to buy securities in the aftermarket.'"). Finally, the words "Corvis," "Engage," "Firepond," "iXL," "Sycamore" and "VA Linux" do not appear in this article; nor do defendants cite any other articles specifically connecting any of the six focus cases to the alleged scheme. See Defendants' Sur-Reply at 7; Exs. A, B to Houck Decl. (comprising various articles purportedly alluding to the alleged scheme).

Defendants' Sur-Reply at 8.

However, the question of whether publicly available information "would have made any reader aware of the allegations here" presents an important class-wide common issue. If any of defendants' proffered publications is determined to have been so relevant, clear and widely disseminated that knowledge of the alleged scheme must be imputed to the universe of investors in the stock market, then reliance cannot be proven individually or collectively. Furthermore, differences among class members in terms of access to publicly available information ( e.g., whether certain investors actually saw all publicized materials, or whether they had access to sophisticated investment advice in interpreting the releases) are insufficient to defeat certification or rebut plaintiffs' presumed reliance.

Id. at 6.

The issue of when plaintiffs were placed on inquiry notice by publicity is also a common question in the context of defendants' statute of limitations arguments. See iXL Mem. at 18-19. Although the two arguments have different ramifications ( i.e., a finding that a plaintiff knew of the fraud forces the plaintiff to prove reliance individually, while a finding that the plaintiff was on inquiry notice so long before suit that her claim is time-barred precludes recovery completely), the common questions they present are exactly the same. This situation is easily distinguished from that considered by the Second Circuit in Moore, 306 F.3d 1247, 1253, which (1) concerned fraudulent misrepresentations in connection with individual insurance contracts (and thus did not invoke any securities fraud-related presumptions of reliance), and (2) concerned oral misrepresentations agents made directly to customers, not the market-wide public dissemination of written information. Similarly, Zimmerman v. Bell, 800 F.2d 386, 390 (4th Cir. 1986) (holding that, where specific information regarding the alleged fraud was abundant, "individual class members must demonstrate that the omitted information was not otherwise available to them") is inapposite. Not only is Zimmerman a Fourth Circuit case, it pre-dated the fraud on the market presumption of reliance created in Basic, and concerned a situation where the alleged fraud was explicitly revealed by numerous publications.

Defendants cite a number of cases in support of their contention that, because some of the proposed class members may be charged with knowledge of the alleged scheme, the class should not be certified. See Defendants' Sur-Reply at 7 n. 3. Defendants' citations are inapposite. See, e.g., Seibert v. Sperry Rand, 586 F.2d 949, 952 (2d Cir. 1978) (finding that alleged scheme was a "matter of public knowledge" where "[a]ffidavits submitted by both parties show that [defendants'] difficulties were reported countrywide in the press and on radio and television, were discussed in Congress, and were analyzed in published administrative and judicial opinions[, and] that a nationwide consumer boycott was being conducted against [defendant], accompanied by massive media advertising"); Frigitemp, 524 F.2d at 282 (rejecting appellant plaintiffs' claims based on the fact that "shareholders gave up shares in the belief that the financial structure of the corporate entity would be strengthened," where "[t]he defendants did nothing to induce that belief, the truth of which was peculiarly within the knowledge of the appellants."). In fact, defendants' citation to Siebert, which involved vast dissemination of the facts underlying each plaintiffs' claim, demonstrates the value of adjudicating common facts ( e.g., whether all plaintiffs should be charged with knowledge based on publicly available information) in a single proceeding. Siebert held, at the summary judgment stage, that publicity regarding an alleged fraud was so widely disseminated that plaintiffs could not avail themselves of a presumption of reliance based on material omissions; here, the Court or a finder of fact may similarly determine that such publicity placed class members on constructive notice of the scheme, and thus bar the class, or members of the class who purchased after such reports were disseminated, from invoking a presumption of reliance.

See, e.g., In re Data Access Sys. Sec. Litig., 103 F.R.D. 130, 139 (D.N.J. 1984) ("There will always be some individuals who read the financial statements directly, others who read secondary analyses . . . and many others who relied on advice of stockbrokers or friends. If defendants' argument were to prevail that factual differences of this nature were sufficient to defeat class action certification, there could never be a class action of securities purchasers.").

2. Loss Causation

In addition to transaction causation, plaintiffs must prove loss causation; that is, they must show a "causal link between the alleged misconduct and the economic harm ultimately suffered by the plaintiff." Plaintiffs may submit an expert report suggesting a methodology for determining such a link. "A district court must ensure that the basis of [such an] expert opinion is not so flawed that it would be inadmissible as a matter of law." At the class certification stage, the question "is whether plaintiffs' expert evidence is sufficient to demonstrate common questions of fact warranting certification of the proposed class, not whether the evidence will ultimately be persuasive;" a district court should therefore refrain from "weigh[ing] conflicting expert evidence or engag[ing] in `statistical dueling' of experts." Under Rule 23(b)(3), plaintiffs must present a methodology for determining loss causation that may be commonly applied to all members of the class.

Emergent Capital, 343 F.3d at 197.

See VISA Check, 280 F.3d at 134-35 (examining submission of expert report to show loss causation in the antitrust class action context). In an antitrust case, "a plaintiff must make some showing of actual injury attributable to something the antitrust laws were designed to prevent" — a requirement akin to the loss causation requirement in securities fraud cases. J. Truett Payne Co. v. Chrysler Motors Corp., 451 U.S. 557, 562 (1981).

VISA Check, 280 F.3d at 135.

Id. (citing Caridad, 191 F.3d at 292-93).

See id; see also In re Sumitomo Copper, 182 F.R.D. at 91 (granting class certification upon finding that "plaintiffs' econometric methodologies have a reasonable probability of establishing" plaintiffs' claims by common proof). But see Newton v. Merrill, Lynch, Pierce, Fenner Smith, Inc., 259 F.3d 154, 189 (3d Cir. 2001) (where determining the existence of loss would require individual analysis of each investor's trades, "[t]he individual questions . . . are overpowering.").

Unlike damages, which require a showing of the quantum of loss, loss causation requires only that there be a causal connection between the alleged wrongdoing and plaintiffs' loss. In a market manipulation case, plaintiffs can satisfy their burden by presenting a means to determine that the scheme caused an increase in price that dissipated throughout the class period. To satisfy Rule 23 in the context of loss causation, plaintiffs need not precisely quantify the proportion of each plaintiff's loss attributable to dissipation; they need only provide a mechanism showing that the alleged scheme actually caused some loss to all class members. Plaintiffs must, however, provide a mechanism for proving that inflation dissipation occurred throughout the class period. Otherwise, investors who purchased after all artificial inflation created by the alleged scheme had dissipated would be differently situated ( i.e., they would be forced to prove loss causation individually using alternatives to the class method of proof), and individual questions would dominate the loss causation inquiry.

See Emergent Capital, 343 F.3d at 197.

See In re IPO, 297 F. Supp. 2d at 675 ("It is that dissipation — and not the inflation itself — that caused plaintiffs' loss."). In this case, the alleged "misstatements and omissions did nothing more than conceal the Underwriters' alleged market manipulation;" hence, plaintiffs need not proffer separate loss causation methodologies for their misstatement and omission allegations. Id. Defendants' complaint that plaintiffs' methodology only links tie-in agreements, but not analyst conflicts or undisclosed compensation, to loss causation is also rejected. If the alleged scheme, taken as a whole, caused plaintiffs' loss, then there is no need to parse the scheme into its component parts and determine whether each alleged component caused inflation.

See, e.g., Dukes v. Wal-Mart, Inc., 222 F.R.D. 189, 192 (N.D. Cal. 2004) (finding that even where plaintiffs' expert "conceded that he could not calculate whether 0.5 percent or 95 percent of the employment decisions at Wal-Mart might be determined by stereotyped thinking," report was nonetheless admissible to prove that stereotyped thinking caused plaintiffs' injuries).

Plaintiffs submit the expert opinion of Professor Fischel to provide a method of proving that the alleged scheme inflated stock prices as early as the beginning of trading, and that the inflation dissipated throughout the class period. Fischel's methodology for proving loss causation depends on two separate analyses: first, an analysis of the initial inflation caused by alleged tie-in agreements; and second, an analysis of the dissipation of that inflation over time.

See 1/20/04 Fischel Report; 4/15/04 Fischel Report; 7/12/04 Fischel Report.

Fischel empirically demonstrates the effect of tie-in agreements on demand and price through an analysis of the pre-open bid sessions for five of the six focus cases. During the pre-open bid session, in which a new issue takes dealer quotes before actual trading begins, the lead underwriter opens the bidding and investors may enter bids to purchase shares. The level of demand in the preopen bid session affects bid prices. Fischel describes and analyzes price changes in the "inside bid" — the highest bid at any given time — with respect to the bidding activity of the lead underwriter and investors alleged to have executed tie-in agreements. Institutional investors with alleged tie-in agreements constituted much of the demand for shares in each pre-open bid session, and purchase orders executed after these sessions accounted for a substantial portion of all shares issued in the IPO. Demand by investors with tie-in agreements remained strong throughout the pre-open bid session. Fischel notes that, "consistent with the substantial purchase orders at the end of the pre-open bid session, the opening price for each of the focus case stocks was substantially higher than the offer price."

See 7/12/04 Fischel Report. Fischel did not have access to data regarding the iXL pre-open bid session. See id. ¶ 6.

See Sirri Report ¶¶ 14-16.

See 7/12/04 Fischel Report ¶¶ 7-12.

See id. ¶ 8 (using FirePond as an example, "[i]nstitutional allocants with alleged tie-in agreements accounted for 95 percent of the total demand for FirePond stock [as measured by the proportion of purchase orders from those with alleged tie-in agreements] at the beginning of the pre-open bid session . . . amounting to 52 percent of the 5,666,666 shares issued in the FirePond IPO").

See id. ¶ 10 (by the end of the session, purchase orders from allocants with alleged tie-in agreements constituted 96% of the total demand at that time, which amounts to 70% of the total number of shares issued). By comparison, only 55.6% of the Firepond IPO shares were actually allocated to investors with alleged tie-in agreements. See id. ¶ 5.

Id. ¶ 11. See also supra Part II.D. (noting, for example, that Sycamore stock was offered at $38 per share, but opened at $270.88).

Fischel also observes that the lead underwriter in each pre-open bid session set the initial bid substantially higher than the offering price, and that "[t]his is consistent with . . . knowledge of the volume of pending purchase orders." Fischel asserts that:

7/12/04 Fischel Report ¶ 9.

The literature has documented that even before the opening of trading, significant price discovery takes place and that a large proportion of the change in price from the offer price to the opening price is captured in the first quote entered by the lead underwriter. This evidence supports the conclusion that the alleged tie-in agreements affected prices before trading began.

Id. (footnote omitted).

Fischel also notes that "activity in the lead underwriter's bid during the pre-open bid session was [frequently] followed by an increase in the inside [best] bid."

Id. ¶ 12. Although Fischel does not fully explain the import of this observation, it seems to reflect the same type of "price discovery" — that is, knowledge of expected demand — he infers with respect to the initial underwriter bids.

Trading activity of allocants with alleged tie-ins was not limited to purchase orders executed at the beginning of trading. Rather, Fischel notes that allocants "purchased substantial quantities of shares in each focus [stock's] aftermarket." Fischel also undertakes a regression analysis to show that the size of each allocation correlates to the quantity of stock that allocant purchased in the aftermarket. To explain how such purchases might inflate prices, Fischel notes that "Keim and Madhavan . . . find that a buyer-initiated trade of only 0.16 percent of a company's outstanding stock is associated with a permanent price increase of 4.7 percent in the stock price."

Id. ¶ 13. Fischel does not clarify whether this reference is to allocants with tie-in agreements, or to the entire population of allocants.

See id. ¶ 15. Of course, this analysis tends more to show the existence of tie-in agreements than the existence of artificial inflation. Nonetheless, it does highlight the possible breadth of the alleged scheme.

Id. ¶ 23 (citing Keim, Donald B. Ananth Madhavan, The upstairs market for large-block transactions: analysis and measurement of price effects, 9 Rev. Fin. Studies (Spring 1996), 1-36, at 19) (emphasis in original). Defendants' expert, Dr. O'Hara, challenges Fischel's assertion of a permanent price increase, noting that, in the market microstructure field, "[p]ermanent effects refer to the impact of the trade on beliefs," not to indelible or long-enduring price effects. 7/23/04 O'Hara Report at 4 n. 2. This is exactly what Fischel intends to show — that the alleged tie-ins changed investors' beliefs in the true value of the securities and that, over time, the artificial price inflation dissipated. The quantity of alleged tie-in purchases distinguishes this case from West v. Prudential Sec., Inc., 282 F.3d 935 (7th Cir. 2002), in which the Seventh Circuit rejected plaintiffs' allegation that the purchasing behavior of eleven investors privy to secret information raised market prices. Here, the number of alleged tie-in purchases as a proportion of overall share demand is considerable, and Fischel offers a method to detect inflation attributable to those tie-in purchases.

Having thus established a mechanism for proving that the alleged scheme caused artificial inflation, Fischel turns to the problem of how to determine the duration of that inflation and its rate of dissipation. Fischel adopts the "Comparable Index Approach," in which the overall performance of an issue is compared to a benchmark index averaging the price movements of comparable stocks. Defendants' expert asserts, and Fischel concedes, that the Comparable Index Approach is usually invoked to determine damages, not loss causation. Specifically, "[t]he premise of the Comparable Index Approach is that all changes in the price of a company's stock not accounted for by movements in the comparable company stock prices and not accounted for by movements in the general market are attributed to the alleged fraud." Thus, as generally applied, the Comparable Index Approach is used to calculate damages where loss causation has already been proven or is assumed; that is, it " assumes loss causation rather than detects it."

See 1/20/04 Fischel Report ¶¶ 16-19; 4/15/04 Fischel Report ¶¶ 6-7. Fischel proposes that the rate of dissipation can be shown using either the Comparable Index Approach or the "Event Study Approach," both of which compare the observed fluctuations in a security's price to the expected returns if that stock had not been manipulated ( i.e., was neither undergoing artificial inflation, and thus overperforming its expected value, nor dissipating that inflation, and thus underperforming). Both theories adopt the same formula for determining the degree of variation from expected returns. Through this mechanism, they quantify performance deviations over time, creating a "value line" that can be compared to the actual "price line" of a stock to show the existence of loss and degree of damages to investors at any given time.

See Cornell Report ¶¶ 4-8; 4/15/04 Fischel Report ¶ 6 (the Comparable Index Approach "could be used to compute the value line" of a stock) (emphasis added).

Cornell Report ¶ 4.

Id. (emphasis in original).

However, Fischel has already provided a method to show that the alleged scheme artificially inflated stock prices. Plaintiffs' loss causation calculation does not depend on the Comparable Index Approach. It is, however, a component of the analysis. Once artificial inflation has been established by the mechanisms discussed earlier ( i.e., by a lead underwriter making a high initial bid in the pre-open bid session and raising its own bid; by creation of artificial demand through tie-in agreements, which causes prices to rise; and by permanent changes in beliefs caused by buyer-initiated trading), all that remains is detecting the dissipation of that inflation. Here, each of the focus stocks ultimately plummeted in value to levels far below their offering prices and not far above zero, the lowest possible value. Some loss causation may be inferred simply from the disappearance of the original inflation. After all, when an artificially inflated stock tumbles to a fraction of its offering price, it is logical to assume that the artificial inflation has dissipated. The Comparable Index Approach need not carry the load of proving the existence of inflation or dissipation.

See In re IPO, 297 F. Supp. 2d at 674 ("In market manipulation cases, therefore, it may be permissible to infer that the artificial inflation will inevitably dissipate."). For example, of the focus cases, Engage, Firepond and iXL all traded below $10 per share on December 6, 2000. The stocks continued to underperform after the close of the class period ( e.g., on the date the first suit was filed in each case, Corvis closed at $7.38, Engage at $0.19, Firepond at $0.66, Sycamore at $8.63 and VA Linux at $9.031). See supra Part II.E.

Fischel proposes only to use the Comparable Index Approach to determine the duration of dissipation. Under this framework, Fischel finds that each of the focus stocks significantly overperformed on the first day of trading and underperformed in the long term when compared to various benchmark indices. Fischel attributes the securities' initial overperformance to artificial inflation in the immediate aftermarket and their later underperformance, at least in part, to a gradual dissipation of that inflation. The rate of dissipation, and its existence, can be inferred from the fact that, in the long run, the focus stocks consistently declined further in price than comparable market benchmarks, which presumably reflected the same market-wide variables. Fischel notes that the markets for the six focus cases significantly underperformed market benchmarks even after December 6, 2000, implying that the stock price continued to shed inflation throughout and after the close of the class period. As a result, Fischel has established a method by which a finder of fact could conclude both that stock prices were artificially inflated and that the inflation dissipated throughout the class period, continuing even after December 6, 2000.

See 7/12/04 Fischel Report ¶¶ 20-22.

See id. ¶ 21; 1/20/04 Fischel Report ¶¶ 23-29; 4/15/04 Fischel Report ¶ 22.

See 7/12/04 Fischel Report ¶ 20 ("The fact that underperformance continued for each of the six focus case stocks after December 6, 2000 strongly suggests that the inflation that existed at the time of each stock's IPO had not fully dissipated by the end of the class periods.").

Dr. O'Hara notes that, in fact, Corvis occasionally overperformed market benchmarks well after its IPO but during the class period, when a simplistic application of Fischel's underperformance measure of dissipation would imply constant underperformance. See 7/23/04 O'Hara Report ¶ 17. However, there could be any number of reasons for such an unexpected price increase, including materially misleading analyst reports. While quantifying the actual amount of inflation at every point during the class period is a necessarily fact-intensive inquiry, it is not one that plaintiffs must undertake to prove loss causation. Plaintiffs must merely show some loss, and the significant protracted underperformance of the focus stocks throughout and after the class period satisfies plaintiffs' burden at this stage.

Fischel's theory is not fatally flawed. Although defendants present a cadre of experts clamoring to apply alternative methods of determining loss causation, now is not the time to "weigh conflicting evidence or engage in `statistical dueling' of experts." Defendants are free to attack Fischel's theory at trial or present alternative theories if they choose.

It seems unusual that defendants in a securities fraud case would go to such trouble to provide methods for measuring and detecting the harm caused by their alleged wrongdoing. Clearly, though, defendants' efforts to adduce their own theories of loss causation seek to persuade the Court that any valid theory of loss causation — like those defendants proffer — would require intensive trade-by-trade analysis and be characterized by instantaneous dissipation of artificial inflation. Defendants' alternative theories of loss causation, then, are offered to serve twin goals: first, to challenge manageability and predominance by presenting a laborious and time-consuming method for detecting inflation; and second, to provide a method that assumes an almost instantaneous rate of dissipation that, if adopted, would exclude most class members from recovery. The class certification decision is not the appropriate place to choose the winning theory of loss causation. The only issue now is whether plaintiffs' theory must be rejected as a matter of law.

VISA Check, 280 F.3d at 135. But see Newton v. Merrill Lynch, Pierce, Fenner Smith, Inc., 259 F.3d 154, 178-79 (3d Cir. 2001) (affirming finding that individual loss causation inquiries predominated where securities broker-dealers were accused of failing to procure the best price possible for their clients). Newton, however, dealt with very different conduct from that alleged here. In Newton, the district court found "that defendants' practice did not detrimentally affect the value of plaintiffs' securities across the entire market [and that there was] no resemblance to cases where economic injury naturally flowed from defendant's alleged conduct." Id. at 178. Here, plaintiffs allege a coordinated scheme that artificially inflated prices throughout the entire market.

Plaintiffs have satisfied their burden at this stage to articulate a theory of loss causation that is not fatally flawed. Moreover, because plaintiffs' theory posits protracted dissipation throughout the proposed class period, it presents common questions of liability — namely, whether tie-in agreements artificially inflated stock prices and the duration of any such inflation ( i.e., whether the inflation dissipated abruptly or over the course of the entire class period). Defendants' alternative theories of loss causation, which generally require intensive trade-by-trade analysis of transitory price effects, would, if adopted by the jury, answer that question in the negative. Defendants provide various criticisms of Fischel's "methodology," but these attacks go to the weight of Fischel's conclusions and must be reserved for trial. Defendants also point out that Fischel's analysis may not be able to quantify the amount of inflation or dissipation at any given time. However, as I have already noted, loss causation only requires that plaintiffs establish some inflation and dissipation, not the precise size of the inflation or amount of the loss. That inquiry relates to damages, not loss causation, and is therefore addressed in the next section.

See Kleidon Report ¶¶ 44-71; 2/24/04 O'Hara Report ¶¶ 23-47; Sirri Report ¶¶ 34-37; Stultz Report ¶¶ 8-43.

See Emergent Capital, 343 F.3d at 197 ("Of course, if the loss was caused by an intervening event, like a general fall in the price of Internet stocks, the chain of causation will not have been established. But such is a matter of proof at trial. . . ."). For example, defendants argue that: Fischel's benchmark comparisons may suffer from selection bias, see 7/23/04 O'Hara Report ¶ 14; that Fischel's theory does not explain the lack of correlation between amount of wrongdoing (as a percentage of shares issued) and magnitude of initial overperformance, see 7/23/04 O'Hara Report ¶ 28; and that Fischel ignores possible confounding factors and important events both in the course of focus case trading and in the larger context of the Internet bubble, see generally Barry Report; Gompers Report. See also Sycamore Mem. at 13 ("Sycamore's price performance over the proposed class period is readily explained by market forces that impacted stocks in general, and the optical networking sector in particular. . . ."). However, to prove loss causation, plaintiffs need not show that the alleged scheme was the sole cause of loss. Plaintiffs may satisfy their burden by showing that, because of the alleged scheme, they lost more than they would have lost had the stock price been affected only "by market forces that impacted stocks in general." Id.

See, e.g., 7/23/04 O'Hara Report at 6 n. 5, ¶¶ 17, 18, 31, 34-36.

3. Damages

If plaintiffs are successful in proving liability, they will have to provide a methodology for calculating damages. In any publicly traded securities market, some investors own many shares and some own only a few; some maintain their portfolios for years, and some trade shares daily. Thus, the extent of the harm suffered by each class member as a result of the alleged misconduct is, by definition, an individualized inquiry.

See Blackie, 524 F.2d at 905 ("The amount of damages [in a 10b-5 class action] is invariably an individual question. . . ."); see also In re Rent-Way Sec. Litig., 218 F.R.D. 101, 119 (W.D. Pa. 2003) ("The problems presented by `in and out sellers' are bound to inhere in any securities action alleging a fraud on an open securities market. This is all the more true in cases such as this one where the alleged fraudulent scheme was of longer duration and/or involved a multiplicity of alleged misrepresentations.").

However, where common questions otherwise predominate, the need for individualized damages inquiries is not enough to scuttle the class action. Rather, the Second Circuit has recognized several methods by which a court may address the problem of individual damages while securing the benefits of the class action device for common issues of liability:

See VISA Check, 280 F.3d at 139 ("Common issues may predominate when liability can be determined on a class-wide basis, even when there are some individualized damage issues."); In re Worldcom, 219 F.R.D. at 302 ("When liability can be determined on a class-wide basis, individualized damage issues are not ordinarily a bar to class certification."); see also Fed.R.Civ.P. 23(b)(3) Advisory Committee Note ("[A] fraud perpetrated on numerous persons by the use of similar misrepresentations may be an appealing situation for a class action, and it may remain so despite the need, if liability is found, for separate determination of the damages suffered by individuals within the class.").

There are a number of management tools available to a district court to address any individualized damages issues that might arise in a class action, including: (1) bifurcating liability and damage trials with the same or different juries; (2) appointing a magistrate judge or special master to preside over individual damages proceedings; (3) decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; (4) creating subclasses; or (5) altering or amending the class.

VISA Check, 280 F.3d at 141 (footnotes omitted).

"Particularly where damages can be computed according to some formula, statistical analysis, or other easy or essentially mechanical methods, the fact that damages must be calculated on an individual basis is no impediment to class certification." Although there are extreme cases in which calculation of damages may present such an intolerable burden that it renders class certification inappropriate, "such cases rarely, if ever, come along."

Klay v. Humana, Inc., No. 02-16333, 2004 WL 1938845, at * 13, ___ F.3d ___ (11th Cir. Sept. 1, 2004) (footnotes omitted); see also id. ("In assessing whether to certify a class, the Court's inquiry is limited to whether or not the proposed methods for computing damages are so insubstantial as to amount to no method at all[.] Plaintiffs need only come forward with plausible statistical or economic methodologies to demonstrate impact on a class-wide basis.") (citation and alterations omitted).

See, e.g., Windham v. Am. Brands, Inc., 565 F.2d 59, 70 (4th Cir. 1977) ("The district court estimated — conservatively, we think — that in the absence of a practical damage formula, determination of damages in this case would consume ten years of its time. The propriety of placing such a burden on already strained judicial resources seems unjustified."). Nonetheless, the Windham court noted that "in cases where the fact of injury and damage breaks down in what may be characterized as `virtually a mechanical task,' `capable of mathematical or formula calculation,' the existence of individualized claims for damages seems to offer no barrier to class certification on grounds of manageability." Id. at 68 (citing, inter alia, Blackie, 524 F.2d at 905). Windham was a complicated antitrust case where "plaintiffs could plead no common impact or injury from an alleged conspiracy to control prices in tobacco auctions. Such an allegation was precluded by the great variety of geographical markets, daily price fluctuations and individualized systems for grading product quality." Rios, 100 F.R.D. at 408 n. 13. Unlike the plaintiffs in Windham, though, plaintiffs here seek to prove damages for each class member through a common formula that their expert says can be developed after the completion of discovery. See 4/15/04 Fischel Report ¶¶ 4-7.

Klay, 2004 WL 1938845, at * 14.

"Before and after the enactment of the PSLRA, absent class members in securities fraud cases have been awarded a common fund of damages computed by the trier of the fact, based usually on expert testimony. . . ." For example, a jury may be asked to compute the "true value" of a stock over time, including fluctuations due to various price-affecting events, and consequently determine by what degree the stock was inflated at any given time during the class period. Thus, important common questions regarding damages, as well as loss causation, may be resolved by asking the jury to trace a "graph delineating the actual value of the stock throughout the class period. When compared with a comparable graph of the price the stock sold at, the determination of damage will be a mechanical task for each class member."

In re Oxford Health Plans, 244 F. Supp. 2d at 251.

This type of damages calculation is common in securities cases. See, e.g., Sirota v. Solitron Devices, Inc., 673 F.2d 566, 576-77 (2d Cir. 1982); In re Seagate Tech II. Sec. Litig., 843 F.Supp. 1341, 1348-49 (N.D. Cal. 1994); Effective Use of Damages Experts in Securities Class Actions, 1332 Practicing Law Institute, Corporate Law and Practice Course Handbook Series 805, 811 (Sept.-Oct. 2002) (discussing use of expert testimony to determine the "ribbon" of artificial inflation between the true value and market price of shares over time).

Blackie, 524 F.2d at 909 n. 25; see also In re Rent-Way, 218 F.R.D. at 119 (granting class certification and noting that plaintiffs "will be able to present a workable framework for determining aggregate damages and price inflation at . . . trial through the use of expert witnesses who will extrapolate these figures based on trading data during the class period.").

Plaintiffs suggest just such an approach. Plaintiffs have proposed using both the "Event Study Approach" and the "Comparable Index Approach" to determine the effect that any given event during the class period had on stock prices. While assessing the effect of each salient event over hundreds of days in any given class period may be a laborious and time-consuming task, it nonetheless provides a common basis for calculating the damages of all class members. By contrast, an alternative approach that would force each class member to prove in individual proceedings how various events impacted the stock price when she purchased and sold stock would be staggeringly inefficient, would provide countless opportunities for juries to render inconsistent verdicts, and, if the cost were placed on individual class members seeking to prove damages, would likely present a formidable (if not complete) barrier to recovery.

See Plaintiffs' Reply at 83-87 ("The difference between the two lines [representing true value and actual price] shows the amount that a purchaser paid above the actual value of the stock, and therefore the damages that the purchaser incurred by paying more for the stock than it was worth.") (citing Green v. Occidental Petroleum Corp., 541 F.2d 1335, 1344-45 (9th Cir. 1976) (Sneed, J. concurring)).

For an explanation of these theories, see supra n. 29.

See, e.g., Carnegie v. Household Int'l, Inc., 376 F.3d 656, 661 (7th Cir. 2004) (Posner, J.) ("The realistic alternative to a class action is not 17 million individual suits, but zero individual suits, as only a lunatic or a fanatic sues for $30.").

Accordingly, by suggesting a method by which a jury could determine the true value of securities over time, plaintiffs present the common question of magnitude of damages. At this stage of the proceedings, plaintiffs have met their burden to establish that common questions predominate.

However, I note that although plaintiffs have presented a method by which damages could be commonly proved in the same trial as the remainder of plaintiffs' claims, the Court is not bound to limit the proceedings to a single trial. If, as the case develops, it becomes apparent that another method of determining or apportioning damages would be superior to a unitary proceeding, then other avenues of adjudication may be pursued. See VISA Check, 280 F.3d at 141.

4. Section 11 Claims

a. Tracing

"Aftermarket purchasers who can trace their shares to an allegedly misleading registration statement have standing to sue under § 11 of the 1933 Act." A plaintiff successfully traces her shares if she demonstrates that her stock was actually "issued pursuant to a defective [registration] statement;" it is "insufficient that [her] stock `might' have been issued pursuant to a defective [registration] statement." This requirement has been strictly applied, even where its application draws arbitrary distinctions between plaintiffs based on the remote genesis of their shares.

Demaria v. Andersen, 318 F.3d 170, 178 (2d Cir. 2003) (citing Barnes v. Osofsky, 373 F.2d 269, 272 (2d Cir. 1967)).

Lorber v. Beebe, 407 F. Supp. 279, 287 (S.D.N.Y. 1975). Accord Krim v., 210 F.R.D. 581, 586 (W.D. Tex. 2002); Harden v. Raffensperger, Hughes Co., 933 F. Supp. 763, 766 (S.D. Ind. 1996); In re Quarterdeck Office Sys. Sec. Litig., No. CV 92-3970, 1993 WL 623310, at *2 (C.D. Cal. Sept. 30, 1993); Kirkwood v. Taylor, 590 F. Supp. 1375, 1379 (D. Minn. 1984).

See Barnes, 373 F.2d at 272-73 (strictly applying tracing requirement despite acknowledging "that this construction gives § 11 a rather accidental impact as between one open-market purchaser of a stock already being traded and another"); see also In re Crazy Eddie Sec. Litig., 792 F. Supp. 197, 202 (E.D.N.Y. 1992) ("If Congress wishes to ease the burden on securities holders such as plaintiffs, it can do so.").

Tracing may be established either through proof of a direct chain of title from the original offering to the ultimate owner ( e.g., if the owner was an allocant in the IPO, or took actual physical possession of share certificates directly from an allocant), or through proof that the owner bought her shares in a market containing only shares issued pursuant to the allegedly defective registration statement. The modern practice of electronic delivery and clearing of securities trades, in which all deposited shares of the same issue are held together in fungible bulk, makes it virtually impossible to trace shares to a registration statement once additional unregistered shares have entered the market. Even where the open market is predominantly or overwhelmingly composed of registered shares, plaintiffs are not entitled to a presumption of traceability.

See Lorber, 407 F. Supp. at 287; Abbey v. Computer Memories, Inc., 634 F. Supp. 870, 873 (N.D. Cal. 1986). Similarly, the presence of identical shares that were traded before an offering and remain in the market after the offering forecloses the possibility of a section 11 class. See Klein v. Computer Devices, Inc., 591 F. Supp. 270, 273 n. 7 (S.D.N.Y. 1984) ("The open-market purchaser . . . must be able to trace his particular securities to the registration statement when it covered additional securities of an outstanding class. If the purchaser bought identical securities already being traded on the open market, he must look elsewhere for relief.") (citations omitted).

See Lorber, 407 F. Supp. at 287; Abbey, 634 F. Supp. at 873-75; see also 2/20/04 Declaration of Jeffrey Waddle, Senior Counsel and Vice President of the Depository Trust Clearing Corporation, in support of iXL Mem. ("Waddle Decl.") at 1-2.

See Barnes, 373 F.2d at 273 ("an action under § 11 may be maintained `only by one who comes within a narrow class of persons, i.e. those who purchase securities that are the direct subject of the prospectus and registration statement'") (quoting Fischman v. Raytheon Mfg. Co., 188 F.2d 783, 786 (2d Cir. 1951)); see also Barnes, 373 F.2d at 272 ("[I]t seems unlikely that the section developed to insure proper disclosure in the registration statement was meant to provide a remedy for other than the particular shares registered. . . . Beyond this, the over-all limitation of § 11(g) that `In no case shall the amount recoverable under this section exceed the price at which the security was offered to the public,' and the provision of § 11(e) whereby . . . an underwriter's liability shall not exceed `the total price at which the securities underwritten by him and distributed to the public were offered to the public,' point in the direction of limiting § 11 to purchasers of the registered shares, since otherwise their recovery would be greatly diluted when the new issue was small in relation to the trading in previously outstanding shares."). See generally Krim, 210 F.R.D. at 586 (even where market consisted of 91% IPO stock, court held named plaintiffs seeking class certification did not have standing because "[p]laintiffs must demonstrate all stock for which they claim damages was actually issued pursuant to a defective statement, not just that it might have been, probably was, or most likely was, issued pursuant to a defective statement.") (emphasis in original); In re Quarterdeck, 1993 WL 623310, at *2-3 (same result where 97% of market was IPO stock); Abbey, 634 F. Supp. at 874-75 (same result where 82% of market was IPO stock).

Defendants assert that the actual tracing of each plaintiff's stock is "a necessarily individualized inquiry." Furthermore, defendants proclaim that, insofar as each class member must individually prove that her shares were issued pursuant to the relevant registration statement, the necessity of trying individual issues should disqualify the class under the Rule 23(b)(3) predominance requirement.

Sycamore Mem. at 35. See also Corvis Mem. at 20; Engage Mem. at 26; Firepond Mem. at 38-39; iXL Mem. at 35-37; VA Linux Mem. at 42.

See, e.g., Corvis Mem. at 20.

Defendants are correct. If the classes for each of the focus cases are to extend from the date of the IPO to the last day of plaintiffs' proposed class period, December 6, 2000, then each class will include plaintiffs who purchased their shares after untraceable shares entered the market. While some individual class members who purchased after the end of the class period might be able to trace their shares successfully, the resulting inquiry would fragment the class action into myriad mini-trials on the subject of tracing. Plaintiffs' proposed section 11 classes are suitable only for those periods in which class members' ability to trace their shares is susceptible to common proof. Such generalized proof is possible if plaintiffs' section 11 class periods are limited to exclude all purchases made after untraceable securities entered the market. As a result, the section 11 class periods for each of the focus cases must end at the time when unregistered shares became tradeable.

See, e.g., Harden, 933 F. Supp. at 766-67 (only those with section 11 standing "may properly be considered members of the class"); In re Quarterdeck, 1993 WL 623310, at *2-4 (denying class certification, finding that named plaintiffs lacked standing because they could not trace their shares to the allegedly defective registration statement, and finding that named plaintiffs' lack of standing constituted a "unique defense" violating the typicality requirement of Rule 23(a)(3)).

Because of the impossibility of tracing shares once they have mingled with unregistered shares, reserving the tracing issue until a future claims process would be of limited utility. See Waddle Decl.; In re Crazy Eddie, 792 F. Supp. at 201-02. Class members who purchased when only registered shares existed in the market would automatically satisfy the tracing requirement, and class members who purchased shares once untraceable shares entered the market would, because of the anonymity of fungible bulk storage, almost certainly be unable to satisfy their requirement. Thus, common sense requires limitation of section 11 classes to those periods in which plaintiffs will be able to satisfy their burden to show traceability and to exclude potential plaintiffs whose claims would almost invariably be futile. Plaintiffs' counsel has noted bitterly the possible unfairness of this standard:

MR. WEISS: Because most people today keep their stock at the brokerage firm, the street name, they throw it all into this common fungible account. So [the underwriters'] conduct is designed to make it virtually impossible, once they introduce new shares into the market through [Rule] 144, to be able to distinguish one share from the other. . . . [Y]ou are giving them an incentive to avoid section 11 liability. THE COURT: What do I do about the case law, which according to the defense, whether it's one percent or less than one percent, once that problem occurs, the cases, they . . . uniformly say it's over. MR. WEISS: . . . [T]his is different from the other cases because the conduct of the underwriter[s] . . . they are creating an environment that makes it impossible for somebody to take advantage of section 11. . . . I am trying to couple the conduct of the underwriters, who are actually in charge of handling the shares physically. . . . THE COURT: Has any court adopted this theory? MR. WEISS: I don't think so.

Transcript of 6/17/04 Hearing at 108:15-109:23. The advent of fungible bulk storage has made plaintiffs' tracing requirement a stringent one indeed; however, it is not the domain of this Court to abrogate such a requirement. That is a job for Congress.

For each focus case, the filed registration statement summarizes the number and status of outstanding shares, and tells investors when outstanding shares will qualify to enter the market, including information as to when lock-ups will expire and at what point previously issued shares become eligible for trading under Rule 144, promulgated pursuant to the Securities Act. Rule 144 provides, in pertinent part, that affiliated holders of restricted securities who have satisfied the statutory holding period must wait until the issuer "has been subject to the reporting requirements of [either] section 13 . . . or section 15(d) of the [Exchange Act] . . . for a period of at least 90 days" and "has filed all the reports required to be filed thereunder during the 12 months preceding such sale. . . ." In either case, the issuer becomes subject to the filing requirements of the Exchange Act when its filed registration statement becomes effective.

See 10/2/00 Form S-8 filed by Corvis ("Corvis Form S-8"), Ex. N to Hunter Corvis Decl., at 2; 7/19/99 Engage Prospectus at 5; 2/4/00 Firepond Prospectus at 58; 6/2/99 iXL IPO Prospectus at 97; 10/21/99 Sycamore Prospectus at 52; 12/9/99 VA Linux Prospectus at 65.

See 15 U.S.C. §§ 78m, 78o(d).

In the Corvis and VA Linux cases, additional stock offerings were consummated before the 90-day Rule 144 holding period expired. Where there are multiple public offerings of a security, a plaintiff is entitled to a presumption that she has satisfied the tracing requirement of section 11 only if every such offering was defective. However, in both cases, the additional offerings explicitly incorporated the contents of the IPO prospectuses. Thus, to the extent that the IPO registration statements are defective, so are the additional registration statements.

See Corvis Mem. at 20 n. 19 (shares issued pursuant to employees' exercise of stock options); VA Linux Mem. at 43 n. 41 (same).

See, e.g., Bernstein v. Crazy Eddie, Inc., 702 F. Supp. 962, 972 (E.D.N.Y. 1988); In re Eagle Computer Sec. Litig., No. C-84-20382(A), 1986 WL 12574, at *9 (N.D. Cal. Mar. 31, 1986).

See Corvis Form S-8 at 2; S-8 Forms filed by VA Linux on 12/9/99, Ex. H to Hunter VA Linux Decl.

Under Rule 144(k), non-affiliates who hold unregistered shares may sell their shares without restriction after they have held the shares for a period of two years. Defendants imply that some unregistered shares might have been tradeable at the time of the IPOs in Corvis, Firepond and Sycamore. No such inference is supported by the facts. In Corvis, all outstanding shares issued before 1999 (and therefore tradeable under Rule 144(k) before 2001) were issued to affiliates, and defendants have produced no evidence that such shares were ever transferred to non-affiliates. In Firepond, all Rule 144(k) shares were subject to 180-day lock-up agreements. In Sycamore, the company did not exist two years prior to the IPO, making it impossible for any non-affiliate to have held unregistered shares for the two years required by Rule 144(k).

See Corvis Form S-1/A at II-2-5. It is possible that some shares held by Corvis affiliates were sold to non-affiliates more than two years before the Corvis IPO. It unjustifiedly aggravates an already onerous burden to force plaintiffs to prove a negative ( i.e., that no such shares were transferred prior to the expiration of the Rule 144, 90-day holding period). If, however, defendants can prove that such transfers occurred and unregistered shares were tradeable earlier, then the Corvis section 11 class will be shortened accordingly.

See 2/4/00 Firepond Prospectus at 58.

Sycamore's web page shows that the company was founded in February 1998. See Sycamore Networks: Corporate Information: News Events, (August 8, 2004).

Shares issued in the context of stock-based acquisitions (like those in the VA Linux case) cannot circumvent the required holding periods of Rule 144. Before trading unregistered stock, a recipient must hold the stock for "[a] minimum of one year. . . ." Thus, a recipient of stock in VA Linux's first acquisition — of Trusolutions, Inc., on March 28, 2000 — would not have been able to sell that stock until March 28, 2001, well after the end of plaintiffs' proposed class period.

Accordingly, plaintiffs' section 11 class periods are appropriately limited to the periods between each IPO and the time when unregistered shares entered the market. In Corvis, Engage, Firepond, iXL and Sycamore, unregistered shares became tradeable 90 days after the IPO pursuant to Rule 144. In VA Linux, all outstanding shares appear to have been subject to 180-day lock-up agreements, so the VA Linux section 11 class period extends for 180 days after the IPO. Thus, plaintiffs' section 11 class periods are limited to the following: Corvis, July 28, 2000 to October 26, 2000; Engage, July 20, 1999 to October 18, 1999; Firepond, February 4, 2000 to May 4, 2000; iXL, June 2, 1999 to August 31, 1999; Sycamore, October 21, 1999 to January 19, 2000; and VA Linux, December 9, 1999 to June 12, 2000.

See 12/9/99 VA Linux Prospectus at 65.

b. Adequacy and Typicality of Section 11 Class Representatives

A class representative's lack of standing under section 11 qualifies as a "unique defense" sufficient to defeat the typicality of a proposed class representative. Moreover, because section 11 grants a right of recovery only to plaintiffs who sold their securities below the offering price, and limits that recovery to the difference between the sale and offering prices (or the difference between the offering price and the value of shares still held at time of suit), plaintiffs who sold all their traceable stock at prices above the offering price have no right to recover under section 11. Defendants posit that any proposed class representative who sold her shares at a price in excess of the offering price should be excluded because, absent any possibility of section 11 recovery, her claims are not typical of section 11 class members who have a right to recover damages. Defendants are correct. Besides the fact that such a class representative would be subject to unique defenses with respect to her section 11 claims, the foreclosure of any hope for recovery calls into question her motivation to fairly and adequately protect the interests of the class.

See In re Quarterdeck, 1993 WL 623310, at *4.

See In re IPO, 241 F. Supp. 2d at 351.

See Sycamore Mem. at 40.

Consequently, representatives of plaintiffs' proposed section 11 classes must (1) have purchased shares during the appropriate class period, and (2) have either sold the shares at a price below the offering price or held the shares until the time of suit.

Accordingly, the following class representatives are appropriate representatives for their section 11 classes: for Corvis, Huff and Rooney; for Engage, Pappas; for Firepond, the Collinses, Zhen and Zitto; and for VA Linux, Budich and Zagoda. Because plaintiffs have no suitable class representatives for their iXL and Sycamore section 11 classes, their motion to certify those classes must be denied.

C. Rule 23(b)(3): Superiority

Plaintiffs must show that a "class action is superior to other available methods for the fair and efficient adjudication of the controversy." Rule 23 suggests a number of nonexclusive factors the trial judge can weigh to determine superiority, including "the interest of members of the class in individually controlling the prosecution." In a case with thousands or millions of claimants, though, a class member's interest in aggregating the claims substantially outweighs her interest in individual control of the litigation. "The more claimants there are, the more likely a class action is to yield substantial economies in litigation." "[I]n enacting Rule 23(b)(3), `the Advisory Committee had dominantly in mind vindication of the rights of groups of people who individually would be without effective strength to bring their opponents into court at all.'" In a securities class action where millions of shareholders are damaged by fraudulent conduct, none but the very largest individual investors have the capital to prosecute their claims individually. This is especially true in a case such as this one, where expert reports, voluminous briefing and vast discovery are par for the course. However, when investors' claims are aggregated, even an investor who bought a single share has the chance to recover for defendants' alleged wrongdoing. This benefit of the class action form is not easily overcome.

Carnegie, 376 F.3d at 661. See also id. ("It would hardly be an improvement to have in lieu of [a] single class action 17 million suits each seeking damages of $15 to $30."). While individual plaintiffs here seek substantially more money ( e.g., Spiros and Mary Gianos, proposed class representatives for VA Linux, lost $597,085.00 in connection with their purchases of VA Linux stock), the cost of litigating a securities fraud action against multiple well-funded defendants is staggering. See, e.g., In re Cendant Corp. Litig., 243 F. Supp. 2d 166, 172-74 (D.N.J. 2003) (finding that class counsel's 35,000 hours of attorney time and $55,000,000 requested fee were "not clearly excessive" in a securities fraud class action that the Third Circuit found was "a simple case in terms of liability") (quoting In re Cendant Corp. Litig., 264 F.3d 201, 285 (3d Cir. 2001)).

Klay, 2004 WL 1938845, at *23 (quoting Amchem Prods., 521 U.S. at 617) (internal quotation omitted).

Cf. In re Worldcom, 219 F.R.D. at 304 ("Few individuals could even contemplate proceeding with this litigation in any context other than through their participation in a class action, given the expense and burden that such litigation would entail.").

See, e.g., Castano, 84 F.3d at 748 (stating that the "most compelling rationale for finding superiority in a class action [is] the existence of a negative value suit"); In re Inter-Op Hip Prosthesis Liab. Litig., 204 F.R.D. 330, 348 (N.D. Ohio 2001) ("Negative value claims are claims in which the costs of enforcement in an individual action would exceed the expected individual recovery.").

Any consideration of superiority must be framed in this context. Thus, the mere possibility of complexity or unmanageability does not defeat a class action. Because a securities fraud class action offers the opportunity for redress of wrongs where victims would otherwise be unable to press their claims, "a class action has to be unwieldy indeed before it can be pronounced an inferior alternative — no matter how massive the fraud or other wrongdoing that will go unpunished if class treatment is denied — to no litigation at all."

See In re MTBE, 209 F.R.D. at 349 ("[a] court may not decline to certify a class for the sole reason that it may become unmanageable.")

Carnegie, 376 F.3d at 661.

Moreover, the superiority of the class action form to alternative means of adjudication cannot — and should not — be considered in a vacuum. "In many respects, the predominance analysis . . . has a tremendous impact on the superiority analysis . . . for the simple reason that, the more common issues predominate over individual issues, the more desirable a class action lawsuit will be as a vehicle for adjudicating the plaintiffs' claims." Any consideration of superiority must therefore be subjective; it must weigh the benefits and costs of allowing the class action to proceed versus the benefits and costs of individual adjudication.

Klay, 2004 WL 1938845, at *22.

See id. at *27 ("[W]e are not assessing whether this class action will create significant management problems, but instead determining whether it will create relatively more management problems than any of the alternatives (including, most notably, 600,000 separate lawsuits by the class members).") (emphasis added).

In this case, class adjudication is clearly superior to any other form of adjudication. Although preparation and trial of 310 class actions, each of which includes the multitude of common questions presented here, is daunting, preparation and trial of 310 million individual suits with virtually identical allegations would be impossible for all participants — plaintiffs, defendants and the courts. Rule 23 is intended to facilitate, not prevent, litigation of a multitude of claims with substantially identical allegations.

See id. at *23; Fed.R.Civ.P. 23(b)(3) Advisory Committee Note (acknowledging that class action is an appealing tool for adjudicating cases of "fraud perpetrated on numerous persons by the use of similar misrepresentations").

Defendants make little effort to propose alternative means of adjudication that might be superior to the class action form. The two alternative forms defendants suggest — individual prosecution of claims and NASD arbitration — are both impractical for the reasons just described. Because of the costs arbitration or litigation impose on small-stakes securities fraud plaintiffs, neither could result in any recovery for the vast majority of investors included in the class definition, even if defendants' liability is ultimately proved. As neither the defendants nor the Court can suggest a means of adjudicating plaintiffs' claims that would be superior or even comparable to the efficiency and fairness of a class action, plaintiffs have satisfied the superiority requirement of Rule 23(b)(3).

See iXL Mem. at 40.


Accordingly, plaintiffs' motion for class certification is granted in part and denied in part for each of the six focus cases. Plaintiffs' Exchange Act classes are certified to the extent they include investors who acquired shares between the date of the IPO and December 6, 2000 and who satisfy the Court's revised class definition. Plaintiffs' section 11 classes for Corvis, Engage, Firepond and VA Linux are certified as to all investors that satisfy the revised class definition and acquired shares before unregistered shares entered the market, and sold those shares for a loss at prices below the offering price. All of plaintiffs' proposed class representatives except Pappas are suitable to prosecute the Exchange Act claims. Huff, Rooney, Pappas, the Collinses, Zhen, Zitto, Budich and Zagoda are appropriate class representatives for their respective section 11 classes. Because plaintiffs have proposed no suitable class representatives for their iXL or Sycamore section 11 classes, those classes cannot be certified.