OPINION AND ORDER

          SCHEINDLIN, District Judge.

         I. BACKGROUND

         The allegations in these consolidated cases are comprehensively described in my Opinion dated February 19, 2003. Familiarity with that Opinion is assumed. In short, plaintiffs allege that defendants fraudulently inflated the share prices of 310 technology stocks during and after their initial public offerings (" IPOs" ) through an elaborate scheme characterized by tie-in agreements, undisclosed compensation and analyst conflicts. According to plaintiffs, several investment banks (the " Underwriters" ) required substantial investors seeking allocations in the IPOs to participate in the scheme. The companies going public (the " Issuers" ) and their directors and officers (the " Individual Defendants" ) allegedly profited from the scheme-despite low offering prices as compared to the stocks' immediate prices in the aftermarket-by taking advantage of the artificially inflated stock to raise capital, enter into stock-based transactions, or sell their individual holdings at high prices. Plaintiffs allege that the value of their holdings plummeted when this artificial inflation dissipated.

         After extensive settlement negotiations facilitated by an experienced mediator, plaintiffs have agreed to a settlement solely with the Issuers and Individual Defendants in 298 of these coordinated cases. Plaintiffs now move for an Order (a) preliminarily approving the terms of the proposed partial settlement, (b) certifying the Settlement classes for the purposes of the proposed settlement only, (c) approving the form and program of class notice described in the Settlement Stipulation, and (d) scheduling a hearing before the Court to determine whether the proposed Settlement Stipulation should be finally approved. For the reasons set forth below, plaintiffs' motion is granted, conditioned on certain modifications to the proposed bar order. Separate hearings will be scheduled to determine the form, substance and program of notification and to determine whether the proposed Settlement Stipulation should be finally approved.

         II. LEGAL STANDARD

         A. Class Action Settlements

          Unlike settlements in ordinary suits, the settlement of a class action must by approved by the court. The court owes a duty to class members to ensure that the proposed settlement is " fair, reasonable and adequate." In making this determination, the court's " primary concern is with the substantive terms of the settlement; " accordingly, the court must " compare the terms of the compromise with the likely rewards of litigation." The trial judge must " apprise herself of all facts necessary for an intelligent and objective opinion of the probabilities of ultimate success should the claim be litigated." The court should not go so far as to effectively conduct a trial on the merits, but should make " findings of fact and conclusions of law whenever the propriety of the settlement is seriously in dispute." The court must also scrutinize the negotiating process leading up to the settlement. " A presumption of fairness, adequacy, and reasonableness may attach to a class settlement reached in arm's-length negotiations between experienced, capable counsel after meaningful discovery."

          In determining whether a settlement is " fair, reasonable and adequate," courts in this Circuit look to the following factors: (1) the complexity, expense and likely duration of the litigation; (2) the reaction of the class to the settlement; (3) the stage of the proceedings and the amount of discovery completed; (4) the risks of establishing liability; (5) the risks of establishing damages; (6) the risks of maintaining the class through the trial; (7) the ability of the defendants to withstand a greater judgment; (8) the range of reasonableness of the settlement fund in light of the best possible recovery; and (9) the range of reasonableness of the settlement fund to a possible recovery in light of all the attendant risks of litigation. Ultimately, the approval of the proposed settlement of a class action is a matter of discretion for the trial court. In exercising that discretion, though, " it is axiomatic that the law encourages settlement of disputes."

         B. Certification of Settlement Classes

          The use of a settlement class allows the parties to concede, for purposes of settlement negotiations, the propriety of bringing the suit as a class action and allows the court to postpone formal certification of the class until after settlement negotiations have ended. The United States Supreme Court has expressly approved the use of the settlement class device, while also warning that the device raises special concerns. A settlement-only class must meet all the requirements of Rule 23, with one important exception: because the case will never go to trial, the court need not consider the manageability of the proceedings should the case or cases proceed to trial. In the settlement context, the " specifications of [Rule 23]-those designed to protect absentees by blocking unwarranted or overbroad class definitions-demand undiluted, even heightened, attention." However, because manageability of the class action at trial is not considered when approving a settlement class, a court may approve a settlement class broader than a litigation class that has already been certified.

         As courts and commentators have noted, when settlement occurs early in the case the parties have less information on the strengths and weaknesses of the claims, and thus the court and class members may be hampered in their ability to determine the fairness of the settlement: Extended litigation between or among adversaries might bolster confidence that the settlement negotiations were at arm's length. If, by contrast, the case is filed as a settlement class action or certified for settlement with little or no discovery, it may be more difficult to assess the strengths and weaknesses of the parties' claims or defenses, to determine the appropriate definition of the class, and to consider how class members will actually benefit from the proposed settlement.

         The use of this device may also raise questions about collusion and the ability of plaintiffs' counsel to represent the interests of the entire class. Thus, because of these concerns, when a settlement class is certified after the terms of settlement have been reached, courts must require a " clearer showing of a settlement's fairness, reasonableness and adequacy and the propriety of the negotiations leading to it."

         C. Preliminary Approval of Class Action Settlements

          Review of a proposed class action settlement generally involves a two-step process: preliminary approval and a " fairness hearing." First, the court reviews the proposed terms of settlement and makes a preliminary determination on the fairness, reasonableness and adequacy of the settlement terms. " Where the proposed settlement appears to be the product of serious, informed, non-collusive negotiations, has no obvious deficiencies, does not improperly grant preferential treatment to class representatives or segments of the class and falls within the range of possible approval, preliminary approval is granted."

         If the court preliminarily approves the settlement, it must direct the preparation of notice of the certification of the settlement class, the proposed settlement and the date of the final fairness hearing. Class members (and non-settling defendants whose rights may be affected by the proposed settlement) then have an opportunity to present their views of the proposed settlement, and the parties may present arguments and evidence for and against the terms, before the court makes a final determination as to whether the proposed settlement is " fair, reasonable and adequate."

         III. THE PROPOSED SETTLEMENT

         The proposed settlement stipulation, its accompanying documents and schedules, the proposed class notice and the proposed settlement order and judgment are complex and lengthy. For clarity, I will briefly describe their primary terms:

         A. Certification of Settlement Classes

         The settling parties stipulate pursuant to Rules 23(a) and (b)(3) to the certification of 298 classes for settlement purposes. They urge the approval of settlement classes broader in scope than the classes certified in my October 13, 2004 opinion. Each proposed settlement class includes: All Persons who purchased or otherwise acquired shares of the Issuer during the class period described in Schedule 2 to Exhibit 1 to Exhibit E [to the Settlement Stipulation] and were damaged thereby. Excluded from the respective Classes are all of the defendants named in the respective Actions and their related parties as set forth on Schedule 3 to Exhibit 1 to Exhibit E. Also excluded from the Classes will be any Person who requests to be excluded from the Classes in accordance with the requirements set forth in the Notice.

         B. The One Billion Dollar Guarantee

         The Issuers' insurers have agreed to provide an undertaking that guarantees that all plaintiff class members in the 298 settling actions (hereinafter, " plaintiffs" ) will recover at least one billion dollars. Specifically, the insurers agree to pay plaintiffs the amount of one billion dollars less the total of all of plaintiffs' recoveries from the Underwriters in (i) these consolidated cases, (ii) the IPO antitrust litigation, and (iii) the Issuers' " excess compensation" claims, which, as discussed below, are assigned by the Issuers to plaintiffs as part of the settlement. Because the proposed settlement provides for a guarantee, not the actual payment of money, none of the proceeds will be distributed to any proposed class member until after the conclusion of all of the above-mentioned proceedings with respect to the Underwriters.

         If the plaintiffs secure no additional recovery from the Underwriters, the 298 settling classes will be compensated under a plan of designation submitted by class counsel. The funds designated for each class vary considerably based on " preliminary estimates of damages by Plaintiffs' consultants," and range from $110,462 in In re CNET Networks (Ziff-Davis), No. 01 Civ. 7669 to $17,147,382 in In re Corvis Corp., No. 01 Civ. 3857. Similarly, recoveries vary widely among the classes on a per-share basis, ranging from $0.002 per share in In re CNET Networks to $0.507 per share in In re McAfee.com Corp., No. 01 Civ. 7034. The average per-share recovery across all 298 class actions is $0.087 per share. The proposed plan of designation does not reflect any deduction for attorneys' fees, which presumably will reduce recoveries in each of the 298 cases.

         Plaintiffs have also proposed a system for distributing funds among the 298 settling classes in the event that some classes ultimately recover damages from the Underwriters but others do not. This contingency system is necessary because, under the Settlement Stipulation, every dollar recovered from the Underwriters is applied to reduce the overall payment owed by the Issuers' insurers under the billion dollar guarantee, regardless of which class receives the payment. Hypothetically, if plaintiffs in In re Accelerated Networks, Inc. Sec. Litig., No. 01 Civ. 5644, ultimately recover one billion dollars from the Underwriters, and plaintiffs in the other 297 settling classes recover nothing from the Underwriters, then the billion dollars of guaranteed money is entirely offset by the recovery in that single case, and plaintiffs' proposed plan of designation is left unfunded.           Accordingly, plaintiffs propose that some of the recovery for any " lucky" classes, such as the In re Accelerated class in my hypothetical, should be distributed to the " unlucky" classes to compensate for the loss of the guaranteed settlement fund. Thus, plaintiffs " propose setting aside from such recoveries up to $1 million per case" -up to a maximum of one-third the total recovery for each " lucky" class-to compensate the classes that would otherwise be excluded from recovery. As applied to my hypothetical, the In re Accelerated class would thus keep $703 million of its billion dollar recovery, and each of the remaining 297 classes would receive one million dollars from the proceeds. If additional classes also recover damages, then they will split the burden of paying the " unlucky" classes equally (although their contribution will be capped at one-third of the recovery of each) with the In re Accelerated class. Recoveries under this secondary distribution plan would be equivalent across all " unlucky" settling classes, with a maximum payment to each nonrecovering class capped at one million dollars, regardless of the amount that class would have received under the Designation Letter. This creates an odd ambiguity regarding the actual amount guaranteed to each class. For example, the class in In re Corvis Corp., No. 01 Civ. 3857, stands to receive over $17 million under the Designation Letter if none of the 298 settling classes derives any recovery from the Underwriters. However, if the In re Accelerated class recovers a billion dollars and the In re Corvis class is shut out of direct recovery, then the In re Corvis class will receive a maximum of one million dollars. The opposite phenomenon obtains for classes set to receive small recoveries under the Designation Letter. For example, the In re CNET class stands to recover $110,462 if no plaintiff class recovers any additional damages, but one million dollars if the In re Accelerated class alone recovers one billion dollars.

         C. The Assignment of Claims

         In addition, the non-bankrupt Issuers have agreed to assign their interest in all claims against the Underwriters for " excess compensation" to a litigation trust, to be represented by plaintiffs' counsel. The Issuers retain their claims against the Underwriters for underpricing, contribution or indemnification, or antitrust violations, but agree that they will not assert such claims except in certain defined circumstances. The Issuers also agree that they will release the Underwriters from such claims if plaintiffs' counsel secure a reciprocal release for the Issuers from the Underwriters in the context of a settlement with the proposed settlement classes.

         D. Payment of Notice Costs

         The Issuers' insurers have agreed to pay up to $15 million in costs related to giving notice of the proposed settlement to the Settlement classes.

         E. The Bar Against the Underwriters' Claims

         The proposed settlement is conditioned on the entry of a court order barring the Underwriters and other non-settling defendants from commencing or continuing a claim, cross-claim or third party claim " arising out of, relating to, or in connection with the securities involved in" these coordinated cases against any Issuer for contribution, indemnification or reimbursement. The proposed bar order would expressly state that " [t]o the extent not already provided by statute, [the Underwriters] shall have the right to reduce their liability in a manner to be determined by the trial court."

         F. The Issuers' Agreement to Cooperate

          The Issuers have also agreed to provide " such reasonable cooperation as they and Plaintiffs' Executive Committee agree is appropriate and necessary to provide, in an efficient and reasonable manner, further information concerning the claims in the Complaints...." If the Issuers and plaintiffs disagree on this point, they may refer the matter to a mediator subject to de novo review by this Court. In turn, plaintiffs agree not to initiate any formal discovery of the Issuers or Individual Defendants.

         G. The Release of Claims Against the Individual Defendants

         Under the terms of the proposed settlement, plaintiffs would release the Individual Defendants from all claims without receiving any monetary compensation from them.

         IV. DISCUSSION

         A. The Settlement Discussions

         The proposed settlement is clearly the product of " serious, informed, non-collusive negotiations." The settling parties are represented by experienced and talented counsel that share expertise in this field and an extensive knowledge of the details of this case. In addition, their negotiations were facilitated by retired United States District Judge Nicholas H. Politan. The proposed settlement is clearly the result of arm's length bargaining.

         B. Certification of the Settlement Classes

         Plaintiffs request certification of classes broader than those I certified on October 13, 2004. The Underwriters oppose the settling parties' request, arguing that:

[i]n Ortiz, the Supreme Court underscored its disapproval of attempts to certify large settlement classes by substitution of fairness for rigorous application of Rule 23: " Rule 23 requires protections ... against inequity and potential inequity at the precertification stage, quite independently of the required determination at postcertification fairness review.... A fairness hearing [cannot] swallow the preceding protective requirements of Rule 23."

         However, nothing in Ortiz v. Fibreboard Corp., 527 U.S. 815, 119 S.Ct. 2295, 144 L.Ed.2d 715 (1999) expresses " disapproval of attempts to certify large settlement classes" in general; rather, it holds that such large settlement classes must meet the requirements of Rule 23, with the exception of the " manageability" factor of Rule 23(b)(3)(D). In reality, " [b]road class action settlements are common, since defendants and their cohorts would otherwise face nearly limitless liability from related lawsuits in jurisdictions throughout the country. Practically speaking, ‘ [c]lass action settlements simply will not occur if the parties cannot set definitive limits on defendants' liability.’ "

          The proposed settlement classes include two groups excluded from the litigation classes: (1) plaintiffs asserting claims under section 11 of the Securities Act of 1933 who purchased their shares after untraceable shares entered the market; and (2) stock purchasers who are excluded from the class because their trading behavior-characterized by receiving IPO allocations, making tie-in purchases, paying undisclosed compensation, and profiting from their trades-suggests that they had actual knowledge of the alleged scheme. The Underwriters correctly note that some putative class members were excluded from plaintiffs' section 11 classes on the ground of predominance, an explicit requirement of Rule 23(b)(3), rather than solely because of manageability concerns. However, such class members fail the predominance requirement precisely because " each class member must individually prove that her shares were issued pursuant to the relevant registration statements." Accordingly, when the problem of manageability at trial-which, in the case of section 11 tracing, would have required individual proceedings to determine traceability-is removed, the predominance defect is no longer fatal. Indeed, the legal question of tracing is identical for all section 11 plaintiffs, whenever they bought their shares; it is only the difficulty of answering that question for plaintiffs excluded from the section 11 litigation classes that defeats a finding of predominance in the context of a class certified for litigation.

          Similarly, plaintiffs excluded from the class because their trading behavior suggests participation in the scheme were excluded under the implied " ascertainability" requirement of Rule 23. However, the basis of defendants' arguments against the ascertainability of plaintiffs' class was based on the notion that " identifying claimants with knowledge would be a massive undertaking." That massive undertaking would never occur under the terms of the settlement. Essentially, the Issuers have solved the problem of ascertaining which class members may have known of the alleged scheme by agreeing to concede the issue. The necessity of overwhelming mini-trials to determine knowledge thus evaded, the remaining problem of ascertainability- i.e., which investors belong in each settlement class-is easily resolved, and can be accomplished by simple evaluation of class members' trading records. Indeed, the exclusion of such plaintiffs from the litigation classes was reached after " considering the traits most likely to separate investors who knew of the alleged scheme from those who did not know" -not, importantly, after determining conclusively that those excluded from the class were not entitled to any recovery. Those excluded plaintiffs may still bring individual actions against the Underwriters, in which they will need to prove their own ignorance and reliance. Again, the legal question for excluded plaintiffs and members of the litigation classes is identical: did they know of the alleged fraud? It is only the process of determining the answer to that question, and thus ascertaining which class members might be eligible for recovery, that renders some members unsuitable for inclusion in the litigation class.

         Defendants have focused on the differences between the proposed settlement classes and the already-certified litigation classes; however, other than the two differences I have just described, the settlement and litigation class definitions are quite similar. In my Opinion of October 13, 2004, I granted plaintiffs' Omnibus motion for class certification in six focus cases with certain modifications to the class definitions. In doing so, I addressed the question of whether plaintiffs' proposed classes in the six cases comport with the requirements of Rule 23. As I wrote in that Opinion, " 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." Defendants have proffered no arguments that any of the 298 settling classes differs so substantially from the six certified classes in the focus cases that the reasoning of my October 13, 2004 Opinion should not apply.

         Moreover, the inclusion of class members who were excluded from the certified litigation classes is fair to all class members. Although these groups of potential class members were excluded from the certified litigation class on the grounds of both predominance and ascertainability, their inclusion in the proposed settlement classes raises two fairness concerns: (1) whether extinguishing the rights of settlement class members who would not have been included in the litigation classes is fair to those members; and (2) whether distributing the benefits of the settlement to class members who are not part of the litigation classes is fair to those members who are part of both the litigation and settlement classes.

         The first problem is easily solved. Every class member who purchased relevant securities after the close of the certified litigation classes for plaintiffs' section 11 claims and lost money-and thus could be eligible for recovery under section 11-is also a member of plaintiffs' certified classes under section 10 of the Securities Exchange Act of 1934 and Rule 10b-5 promulgated thereunder. Thus, no rights of unrepresented class members are extinguished by the inclusion of all section 11 plaintiffs, regardless of the traceability of their shares. With respect to those potential class members excluded from the class definition because of their trading behavior, the calculus is even easier: because such investors have exhibited a substantial likelihood of knowing participation in the alleged scheme, they likely have no rights to extinguish by their inclusion in the settlement classes.

          The second question-whether distribution of settlement benefits to class members excluded from the litigation classes is fair to members of both classes-is more complex. With regard to plaintiffs' section 11 claims, no additional plaintiffs will be rewarded by the expanded class definition, because all section 11 plaintiffs are already part of the section 10 classes. However, because some plaintiffs would have been able to participate in class actions seeking recovery for their section 11 claims and others would have to proceed individually (if at all) to prove that their shares were traceable to a defective registration statement, class members are differently situated with respect to the damages they could expect to recover if the class action proceeded in litigation. This concern, though, does not mandate the exclusion of section 11 class members who are not entitled to a presumption of traceability from the settlement classes.

         Accordingly, because the proposed settlement classes satisfy the requirements of Rule 23 and the settlement does not prejudice any class members, I find that the proposed settlement classes pose no barrier to class certification.

         C. The Proposed Settlement Falls Within the Range of Possible Approval

          It is tempting to think of this settlement in strictly monetary terms. In most circumstances, one billion dollars is a huge amount of money. However, a close look at the details of these consolidated actions renders the one billion dollar guarantee of minimal magnitude and troubling in its potential execution.

         It is misleading even to use the term " billion dollar guarantee," because that term conflates what are really 298 individual settlements. Each of those settlements-not just the " unitary" settlement which the Issuers have agreed to fund up to one billion dollars-must fall within the range of possible approval. It cannot be emphasized enough that there is no single action called " In re Initial Public Offering Securities Litigation; " rather, these proceedings are simply a means of coordinating the pretrial management of 310 separate, albeit similar, securities class actions. Indeed, a close examination of plaintiffs' proposed designation of funds among classes shows that individual class settlements vary widely, as noted earlier, from $110,462 to $17,147,382. But the true value of the guarantee may, in some cases, be considerably lower than that reflected in plaintiffs' designation letter. For example, the vast class of In re Corvis, representing the purchasers of over 340 million shares, might, under plaintiffs' contingency plan for redistributing additional recoveries among nonrecovering classes, collect only one million dollars-a per-share recovery of less than $0.003. To a small holder of Corvis shares, the so-called " billion dollar guarantee" could prove to be somewhere between invisible and illusory.

         While " there is no reason, at least in theory, why a satisfactory settlement could not amount to a hundredth or even a thousandth part of a single percent of the potential recovery," a court must nonetheless evaluate any proposed class action settlement to see whether it falls within the " range of possible approval." Such a determination, though, need not focus solely on the monetary benefits of a proposed settlement. It should additionally weigh the value of any nonmonetary or intangible benefits associated with the agreement.           Despite the apparent magnitude of the billion dollar guarantee, this settlement is not solely-or even primarily-about monetary recovery. Indeed, if more than one billion dollars is recovered from the Underwriters in any of the 298 classes (or in more than one), then the monetary value of the settlement is nil. The real value of the Settlement Stipulation to the members of each of the 298 classes comes in the form of three distinct benefits: the comfort that at the end of the day there will be a minimal recovery to class members and their counsel; the assignment of the Issuers' claims against the Underwriters to the plaintiff classes; and the cooperation of the Issuers in plaintiffs' actions against the Underwriters.

         The value of each of these benefits should not be understated. Although the potential monetary payout from the billion dollar guarantee is uncertain (and may, in fact, amount to nothing at all if recovery from the Underwriters exceeds one billion dollars), the guarantee of some recovery confers a distinct benefit to the classes. In addition to affording a minimal recovery for class members, the billion dollar guarantee provides a reliable source of future revenue for class counsel. In proceedings as complex as these, the potential risk for plaintiffs' counsel is enormous. Outlay of expenses and attorney time for the four years during which these cases have been pending must already be astronomical, and will continue to grow as class counsel pursues the Underwriters. The presence of a guaranteed monetary payout to the class-and, through calculation of fees, to its attorneys-ensures that counsel will receive some compensation for their efforts. In litigation as potentially draining as these coordinated cases, providing a reasonable fallback source of recovery ensures that class counsel will retain the capacity and zeal to pursue the classes' claims against the remaining defendants, which is, in itself, a benefit to the class.

          Similarly, the assignment of claims and the cooperation of the Issuers in plaintiffs' cases against the Underwriters are themselves valuable. If plaintiffs are correct that the Issuers' IPOs were substantially underpriced and that the Underwriters received substantial excess compensation in bringing the Issuers public, then the assignment of the Issuers' claims for excess compensation could be valuable indeed. Furthermore, from a pragmatic standpoint, the value of 298 willing allies in litigation, as opposed to the specter of hundreds of uncooperative opponents, is significant. The Issuers know far better than the plaintiff classes precisely what occurred in the period leading up to and including their IPOs, and their willingness to open their files and aid plaintiffs in amassing evidence against the Underwriters may ease the plaintiffs' discovery burden enormously.

         The Underwriters have raised one thorny question with respect to the distribution of money among the classes in the event that recovery from the Underwriters offsets any or all of the compensation guaranteed by the settling defendants. Specifically, under plaintiffs' plan, if some classes receive no direct recovery from the Underwriters, then the classes that do recover may have to pay up to one-third of their own recovery-or up to one million dollars per non-recovering class-to members of other classes. This poses two problems: (1) whether a settlement that mandates that class members might someday forfeit some of their potential recovery can be fair to any class; and (2) how the redistribution provision could be applied when these consolidated pretrial proceedings are over and the 298 cases are reassigned to other judges in this District, who will, in turn, have to approve any individual plans of allocation. There are two answers to these questions.

          First, the chance that some classes will be shut out of recovery while others recover large amounts from the Underwriters is quite slim. Although these coordinated proceedings are intended to resolve common issues before hundreds of separate class actions go to trial, the fact remains that very few securities class actions are ever tried. It is far more likely that all of these coordinated cases will be resolved in a unitary manner, similar to the unitary settlement now offered by the Issuers and Individual Defendants. Thus, it is unlikely that plaintiffs' redistribution plan will ever be applied. For the same reason, it is unlikely that these coordinated cases will ever be reassigned to other judges.

          Second, a clear and precise notice will alleviate concerns that class members might not understand that they may be required to surrender a portion of their later recovery to members of other classes. Such a notice will also aid any other judges who must determine exactly what this partial settlement entails, should it be finally approved. The current proposed class notice does not describe how the one billion dollar guarantee will be administered, and makes no mention of how the funds will be designated among classes, allocated among class members, increased or decreased in light of recoveries from the underwriters and potentially reassigned from one class to others. Nonetheless, the unusual terms of the one billion dollar guarantee do not mean that it cannot be approved, nor that a clear notice describing its terms cannot be drafted. Indeed, the one billion dollar guarantee is, at this stage, simply an insurance agreement between the settling classes and the settling defendants. Not only does it provide a backup source of funds for class members, it engages in risk-pooling, in which class members individually acknowledge that they may have to pay up to one-third of their later recoveries to other classes, in return for the chance to receive such redistribution if their own class is shut out. This type of risk-pooling occurs every time a private individual buys an insurance policy, and I see no reason why it should not be allowed in this unique context. However, the very novelty of these coordinated proceedings mandates that the notice to class members be painstakingly clear and complete, and no notice will be approved unless it meets this demanding standard. In sum, the combination of the one billion dollar guarantee, the assigned claims, and the cooperation of the settling defendants falls within the range of possible approval.

         D. The Underwriters' Objections

         1. Collusion

          The Underwriters contend that the Court should deny preliminary approval of the proposed settlement because the parties' agreement encourages " collusion" between plaintiffs and the Issuers that would " distort the fact-finding process" and " preclude a fair determination of the Underwriter Defendants' percentage of responsibility for any damages awarded in any IPO trial." The Underwriters argue that the sliding-scale nature of the billion dollar guarantee would provide an improper incentive for the Issuers to implicate the Underwriters while simultaneously reducing the Issuers' exposure.

         This argument lacks merit. The " improper" financial motive decried by the Underwriters is always present in a multi-defendant case, and the Issuers would have an incentive to implicate the Underwriters regardless of whether the proposed settlement is approved. It is a common practice for defendants to point the finger at each other and to minimize their own misconduct. Indeed, the Underwriters will now have the advantage of blaming the " empty chair" -decrying the Issuers' dishonesty and greed. Nothing in this settlement increases the settling parties' motive to distort the truth regarding the Underwriters' role in the alleged scheme beyond the motives the settling parties would otherwise have to reduce their proportion of liability if they proceeded to trial. To the contrary, the proposed settlement arguably decreases the Issuers' motive to distort the truth. By providing a comprehensive release to the settling parties, the proposed settlement allows them to speak freely about the existence of, and their involvement in, the alleged scheme, something they might not do if they feared sizeable damage awards from hundreds of juries.

         The Underwriters suggest the " cooperation clause" evidences the parties' improper collusion. The settling parties, however, have cited a number of cases in which courts have approved similar cooperation clauses. Furthermore, it makes sense that the settling parties would incorporate such a clause into their settlement. The plaintiffs stand to benefit from the ability to obtain documents and information without the cost and delay inherent in formal discovery. The Issuers, on their part, have committed only to provide informally to plaintiffs what both plaintiffs and the Underwriters could obtain through discovery. For these reasons, I find that the terms of the proposed settlement do not inherently create an impermissible motive for the parties to " collude," as the Underwriters suggest.

         2. The Proposed Bar Order

          The bar order proposed by the settling parties reads in pertinent part as follows:

Accordingly, the Court hereby bars all such claims for or in the nature of contribution: (a) against the Protected Persons; and (b) by the Protected Persons against any person or entity other than any person or entity whose liability to the Settlement Class has been extinguished pursuant to the Issuers' Settlement Stipulation and this Order and Final Judgment. Further, the Court hereby bars and enjoins each defendant Underwriter or other non-settling defendant or the successors or assigns of either (" Barred Parties" ) from commencing or continuing against the Protected Persons, in any forum, (i) a Claim Over, or (ii) any claim, cross claim, or third party claim arising out of, relating to, or in connection with the securities involved in, or the settlement of, such Action, including the negotiation of, execution of, or implementation of the settlement. To the extent not already provided by statute, Barred Parties shall have the right to reduce their liability in a manner to be determined by the trial court.

         The Underwriters object to the wording of this bar order on the grounds that it is broader than the express language of section 78u-4(f)(7)(A), which authorizes only a bar of " all future claims for contribution arising out of the action." They also correctly point out that the proposed settlement expressly authorizes plaintiffs' counsel (as trustees of the litigation trust) to pursue the Issuers' assigned claims for excess compensation or underpricing relating to the securities at issue in the settlement against the Underwriters. Thus, were I to issue a bar order worded in this way, plaintiffs could pursue the Issuers' assigned claims (such as those for excess compensation) against the Underwriters, but the Underwriters could not pursue any claims against the Issuers.

         The settling parties attempt to defend this one-sided arrangement by relying on the final sentence of their proposed bar order-" [t]o the extent not already provided by statute, Barred Parties shall have the right to reduce their liability in a manner to be determined by the trial court" -to leave room for the Court to compensate for any injustice caused by the broadly worded bar order. This proposed catch-all sentence does not cure the fundamental unfairness created by a non-mutual bar order. The PSLRA's settlement discharge provision both mandates a mutual bar order and limits the scope of that bar order to contribution claims. This Court has no authority to deviate from the express wording of the statute. Any bar order that the Court may issue (should the proposed settlement be finally approved) will be limited to the express wording of section 78u-4(f)(7)(A). It will provide a mutual bar for claims of contribution; it will not bar the parties from pursuing other claims.

         However, I do not accept the Underwriters' suggestion that the need for this modification to the proposed settlement somehow precludes preliminary approval of the settlement. While a court cannot re-write the terms of the proposed settlement stipulation, this Court has in the past conditioned its approval of class action settlements on certain modifications to the terms of those settlements. I find that it is appropriate to do so here. My preliminary approval of the proposed settlement is therefore conditioned on the settling parties' adoption of a bar order consistent with this Opinion.

         3. Settlement Discharge

          The Underwriters contend that the proposed settlement and settlement order violate the settlement discharge provisions of the PSLRA because they amount to an impermissible " one-way contribution scheme." This argument characterizes the one billion dollar guarantee as a claim for contribution from the Underwriter defendants that must be barred under the PSLRA after the partial settlement is finally approved. There is no basis for such a conclusion.

          It is true that the proposed settlement allows the settling defendants to reduce their payments to plaintiffs in the amount of any recovery or settlement that plaintiffs receive from the Underwriters. Yet this potential reduction is not a " claim for contribution" by the settling defendants against the Underwriters. Under the terms of the proposed settlement, the settling defendants do not need to assert any contribution claim against the Underwriters to obtain a reduction of their obligations to plaintiffs; indeed, they need make no payment to plaintiffs until plaintiffs' efforts to recover against the non-settling parties come to an end. Once that occurs, the amount of the payment (if any) is calculated automatically.

         Implicitly conceding that the one billion dollar guarantee cannot be characterized as a direct " claim for contribution," the Underwriters nevertheless allege that it amounts to an impermissible " indirect claim" for contribution against them because it allows the settling defendants to benefit from plaintiffs' recovery from the Underwriters. There is no basis for this argument in the text of the statute. Nothing in the PSLRA prohibits the settling defendants from guaranteeing a certain financial recovery to the settling plaintiffs. As the many cases cited by the Issuer Defendants make clear, sliding-scale settlements in securities actions where the settling defendants' obligation to the plaintiffs depended on the plaintiffs' later recovery from non-settling parties-and in which therefore the precise dollar value of the settlement was not known at the time of the settlement-were known and approved by state and federal courts prior to the enactment of the PSLRA. In drafting the PSLRA, Congress easily could have, but did not, bar settlements in which the settling parties guaranteed a payment the size of which depended on plaintiffs' recovery from non-settling defendants. The absence of any express rule in the PSLRA against sliding-scale settlement guarantees such as the one proposed here militates strongly against the Underwriters' position.

         The PSLRA's legislative history also makes clear that the one billion dollar guarantee does not run afoul of the PSLRA's settlement discharge provision. In enacting the PSLRA, Congress hoped to deter frivolous securities litigation by modifying the joint and several liability standard for private actions under the federal securities laws:

The Committee heard considerable testimony about the impact of joint and several liability on private actions under the Federal securities laws. Under joint and several liability, each defendant is liable for all of the damages awarded to the plaintiff. Thus, a defendant found responsible for 1% of the harm could be required to pay 100% of the damages. Former SEC Commissioner J. Carter Beese, Jr., observed that " [t]his principle has a legitimate public policy purpose, but, in practice, it encourages plaintiffs to name as many deep-pocket defendants as possible, even though some of these defendants may bear very little responsibility for any injuries suffered by the plaintiff...." Where peripheral defendants are sued, the pressure to settle is overwhelming-regardless of the defendant's culpability.... The Committee modifies joint and several liability to eliminate unfairness and to reconcile the conflicting interests of investors in a manner best designed to protect the interests of all investors-those who are plaintiffs in a particular case, those who are investors in the defendant company, and those who invest in other companies.           This guidance clarifies that section 78u-4(f)(7)(A) could not have been intended to discourage partial settlements but rather was designed to buttress the modified joint-and-several liability standard. By mandating that settling covered parties enjoy a bar order, Congress sought to facilitate-not impede-early settlements. The settlement discharge provision ensures that settling parties will not be exposed (after settlement) to liability for more than their share of responsibility (as determined by the parties through settlement) because of future claims for contribution by non-settling defendants. This in turn makes an early partial settlement easier to achieve because the settling party need not broker a global settlement (that is, a settlement that resolves the potential claims for contribution by non-settling co-defendants) before settling with plaintiffs. Keeping this in mind, and given the statute's legislative history, it makes no sense to interpret section 78u-4(f)(7)(A)(ii)'s language as anything more than a prohibition against unfairly one-sided litigation in which parties evade their fair share of responsibility by settling early and then pursuing contribution claims against parties that, but for the bar order, would in turn have pursued their own contribution claims against the settling parties.

         For these reasons, I find that the one billion dollar guarantee provided by the proposed settlement does not amount to a " future claim" for " contribution" by the settling defendants against the Underwriter defendants that would be barred by section 78u-4(f)(7)(A)(ii).

         4. The Judgment Reduction Provision

          Another basis for the Underwriters' opposition to the proposed settlement is their contention that its structure deprives the non-settling defendants of their statutory judgment reduction rights. The PSLRA states that:

If a covered person enters into a settlement with the plaintiff prior to final verdict or judgment, the verdict or judgment shall be reduced by the greater of

(i) an amount that corresponds to the percentage of responsibility of that covered person; or

(ii) the amount paid by that covered person."

         Based on a tortured interpretation of this statute and a misleadingly selective quotation of its legislative history, the Underwriters suggest that the PSLRA precludes the parties from entering into a settlement such as this one if it might conceivably prevent a court from considering the amount paid by settling parties when reducing a verdict or judgment entered against the non-settling defendants. However, the Underwriters' argument is predicated on the baseless notion that, under the PSLRA, a partial settlement must predict the future- i.e., it must tell any non-settling defendants precisely the value of a credit against their future liability, well before anybody knows the extent of that liability. Such a position makes little sense. In a case presenting similar questions, the Second Circuit explained:

We also conclude that the non-settling defendants are not entitled to any greater degree of certainty about the amount of their judgment credit than they already have.... [T]he district court's orders informed the parties well before trial of the method that will be utilized to calculate the set-off. The non-settling defendants know that they will get a credit amounting to the greater of (1) the settlement attributed to common damages, or (2) the settling defendants' proportionate share of the total damages. They therefore know that the settling defendants' wrongdoing is relevant, and can develop their trial strategy accordingly. We find no error in the district court's decision to leave the determination of the actual amount of the judgment credit for calculation at trial because the non-settling defendants will get at least the full settlement amount as a credit, unless the settlement damages are not common, an issue which is obviously contingent on the outcome of the trial.

         Here, the monetary value of the partial settlement will itself not be determined until the proceedings in all 298 cases are complete. At that point, as in Gerber, defendants will be entitled to a judgment credit in each case that is no less than the amount designated to that case under the settlement.

         Obviously, because of the sliding-scale nature of the proposed settlement, there is a substantial likelihood that there will be no monetary value to this settlement. Given the sheer number of class actions governed by this proposed settlement and the magnitude of damages alleged in each, the actual monetary value of this proposed settlement will almost certainly be either one billion dollars or nothing. To reach any other value, one of two things would need to happen: either (1) only a few of the settling cases would receive any recovery while most would be shut out (unlikely given the similarity of the allegations in each case); or (2) each of the cases covered by the proposed settlement would yield a remarkably low recovery compared to both the damages alleged and the average recovery in securities class actions. Thus, the Underwriters' judgment reduction rights will be easy to calculate when all 298 cases have concluded: it will either be the amount designated to each class under the settlement or the percentage of fault allocated by the trier of fact.

         Moreover, nothing in the PSLRA or its legislative history suggests that it proscribes a partial settlement that defers the calculation of the settling parties' obligation to the plaintiff until after a verdict or judgment against the non-settling parties is entered. Rather, as the Senate Report quoted above makes clear, in enacting the PSLRA Congress was primarily interested in ensuring that a non-settling party would not be exposed to liability for more than its percentage of responsibility for plaintiffs' damages. Because the proposed settlement in no way prevents the Underwriters from reducing any verdict or judgment entered against them to " an amount that corresponds to [their] percentage of responsibility" under section 78u-4(f)(7)(B)(i), the Underwriters have not convincingly argued that the proposed settlement undermines their judgment reduction rights under the PSLRA.

         5. The Side Agreements

          Finally, the Underwriters argue that the proposed settlement should not be approved because the Issuers have not disclosed all side agreements identified pursuant to Rule 23(e)(2). The Underwriters also argue they should be permitted an opportunity to take documentary and deposition discovery regarding these side agreements " so that the full extent of the agreements between the Plaintiffs, Issuers and the Issuers' insurers are known."

          The Advisory Committee's 2003 Notes indicate that Rule 23(e)(2) is intended to ensure that the parties disclose undertakings related to a class action settlement because those " side agreements" may have influenced the terms of the settlement by trading away possible advantages for the class in return for advantages for others. The same Note states that " [f]urther inquiry into the agreements identified by the parties should not become the occasion for discovery by the parties or objectors."

         After the Underwriters' Opposition was filed, the Issuers submitted unsigned copies of an " Agreement among Insurers" and a " Insurers-Insureds Agreement" along with exhibits, schedules and other accompanying documents for the Court's in camera review. These two agreements were previously identified pursuant to Rule 23(e)(2) on June 16, 2004, and have not been disclosed to plaintiffs or to the Underwriters. I have now reviewed the " Agreement among Insurers" and " Insurers-Insureds Agreement" and their attachments in their entirety. The former, as one might expect, describes the respective financial obligations of the Issuers' insurers in connection with the proposed partial settlement. The latter resolves a variety of issues between the Issuers and their insurers relating to the Issuers' insurance policies.

         Counsel for the Issuers and Individual Defendants has represented that these two agreements were not negotiated with the plaintiff classes, and have not been seen by their counsel. The agreements themselves do not affect the rights of, or consideration to, the proposed Settlement classes. For these reasons, I am not concerned that any of the disclosed agreements " influenced the terms of the settlement by trading away possible advantages for the class in return for advantages for others." The Underwriters' request for discovery relating to the disclosed agreements is therefore denied.

         E. The Class Notice

         Plaintiffs' moving papers include a comprehensive proposal for the form and program of class notification. Before the proposed settlement is finally approved, though, the class notice must be modified to conform with this Opinion. Additionally, the moving parties have not supplied the Court with any information regarding the relative costs and benefits of existing publication media. Rather than consider the adequacy of plaintiffs' Proposed Notice, a hearing will be held to discuss in detail the form, substance and program of notice. Prior to that hearing, plaintiffs' counsel shall circulate a revised Proposed Notice. In addition, the settling parties should make written submissions prior to the hearing on such issues as the relative cost of giving notice by means of television, radio and the Internet, as well as paper-based media, and how plaintiffs' proposed notice compares with court-approved notice programs in other very large class actions.

         V. CONCLUSION

         For the reasons set forth herein, plaintiffs' motion is granted except as specifically noted above. Accordingly, it is hereby ORDERED that:

1. The proposed Settlement classes are certified consistent with the settling parties' proposed class definitions.

2. Pursuant to Rule 23 of the Federal Rules of Civil Procedure, and for the purposes of the proposed settlement only, all current Lead Plaintiffs and all proposed Class Representatives in the settling actions are hereby certified as Class Representatives.

3. I preliminarily approve the proposed settlement contingent on modifications of the proposed bar order consistent with this Opinion.

4. A conference is hereby scheduled for March 18, 2005, at 10:00 A.M., in Courtroom 15C, for purposes of (a) making final determinations as to the form, substance and program of notice, and (b) scheduling a Rule 23 fairness hearing. The settling parties are hereby directed to submit on or before February 28, 2005, a revised Settlement Stipulation and a proposed " Preliminary Order In Connection With Settlement Proceedings" that is consistent with this Opinion and my class certification Opinion of October 13, 2004. The revised Settlement Stipulation should include a revised proposed class notice that is consistent with this Opinion. Any objections the Underwriters may have to the proposed Settlement Stipulation and " Preliminary Order In Connection With Settlement Proceedings" that they have not already put before the Court in connection with this or any other motion shall be submitted on or before March 10, 2005.

         SO ORDERED

that the 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. The matters pertinent to the findings include: ... (D) the difficulties likely to be encountered in the management of a class action.

The use and availability of [sliding-scale settlements] may enable plaintiffs to obtain recovery-enhancing information that would otherwise have remained hidden at the time of settlement or judgment.... Defendants generally have an incentive to withhold damaging evidence for as long as possible in the hope that they will be able to take advantage of this informational asymmetry to negotiate a favorable settlement or that they will succeed in concealing the information until after a trial judgment is entered. Adverse information is ordinarily disclosed only in response to narrowly tailored discovery requests or when a defendant or his attorney fears that not revealing the information will lead to the imposition of sanctions. A recent study of the discovery process found that attempts to conceal information are often successful. The study reported that in thirty-nine percent of settled cases, defendants' lawyers " believed they still knew something significant about the case that opposing counsel had not discovered." It also found that even in cases that went to trial, defendants' lawyers thought they " knew something of consequence," that had not been revealed in thirty-two percent of their cases.