In re Compact Disc Minimum Advertised Price Antitrust Litigation

Full title: In re COMPACT DISC MINIMUM ADVERTISED PRICE ANTITRUST LITIGATION.

Court: United States District Court, D. Maine.

Date published: Jun 13, 2003

216 F.R.D. 197 (D. Me. 2003)

Opinion

         In determining whether to approve proposed settlements of multidistrict antitrust litigation over music compact disc (CD) pricing, the District Court, Hornby, J., held that: (1) notice program provided the best practicable notice under the circumstances and complied with the requirements of both Clayton Act and class action rule; (2) private class action on behalf of non-natural persons countrywide and residents of States, Commonwealths and Territories not represented by their respective Attorneys General met requirements of class action rule for certification; (3) settlement agreement among retailers and distributors, private plaintiffs, and 43 States, Commonwealths and Territories concerning retail store sales was fair, reasonable, and adequate, and would thus be approved; (4) ten percent was a reasonable percentage for overall recovery of attorney fees and (5) proposed settlement which created a program of 50% discounts for music club members who purchased CDs from the defendant music clubs during the class period would not be approved since it did not have any significant benefit to the members of the class as a whole.

         Orders in accordance with opinion.

          John Brautigam, Maine Assistant Attorney General, Augusta, ME, Liaison Counsel for the Plaintiff States.

          Linda Gargiulo, Assistant Attorney General, New York City, Lizabeth Leeds, Assistant Attorney General, Tallahassee, FL, Lead Counsel for the Plaintiff States.

          Alfred C. Frawley, III, Gregory P. Hansel, Preti, Flaherty, Beliveau & Haley, LLC, Portland, ME, Liaison Counsel for the Private Plaintiffs.

          Joseph C. Kohn, Michael J. Boni, Steven M. Steingard, Kohn, Swift & Graf, PC, Philadelphia, PA, Lead Counsel for the Private Plaintiffs.

          Michael Jaffe, Wolf Haldenstein Adler Freeman & Herz, LLP, New York City, for Trowbridge Plaintiffs.

          William J. Kayatta, Jr., Clifford H. Ruprecht, Pierce Atwood, Portland, ME, Liaison Counsel for the Distributor Defendants.

          Joseph H. Groff, III, Jensen, Baird, Gardner & Henry, Portland, Liaison Counsel for Retailer Defendants.

         DECISION AND ORDER ON NOTICE, SETTLEMENT PROPOSALS, CLASS CERTIFICATIONS AND ATTORNEY FEES

          HORNBY, District Judge.

         I. SUMMARY

          I conducted a final fairness hearing on May 22, 2003 on the proposed settlement of this multidistrict antitrust litigation over music compact disc (" CD" ) pricing. I now APPROVE the settlement agreements among retailers and distributors, private plaintiffs, and 43 States, Commonwealths and Territories concerning retail store sales. I APPROVE notice, CERTIFY a settlement class, APPROVE attorney fees and disbursements for the private plaintiffs' lawyers and for the States, and AWARD injunctive relief to prevent similar behavior in the future and to end generally the litigation over past practices. ¹ This settlement will put cash in the hands of millions of consumers and music CDs in libraries and schools throughout the country, and will ensure that the challenged distributor/retailer practices will not resume. Specifically, as a result of this settlement, valued at over $143 million, more than 3.5 million consumers who purchased one or more CDs at retail during the class period (January 1, 1995 through December 22, 2000) will each receive a check for close to $13 (the average overcharge per CD is estimated at $.23); and 5.6 million CDs (1,960 carefully selected titles) will be distributed on a pro rata basis by population to the various States, Commonwealths and Territories for public use and enjoyment (primarily through libraries and educational institutions). The cost of notifying potential claimants, processing claims, distributing checks and distributing the CDs will be $6-8 million. The States, Commonwealths and Territories, and the lawyers representing the class, will receive $14.3 million for legal services, to be divided among them by agreement.

1 I DEFER approval of the various CD distribution plans pending receipt of additional information.

         I DISAPPROVE the settlement agreement among private plaintiffs, music clubs and distributors concerning music club sales. That moots the requests for final class certification and attorney fees in that case. I have rejected the music club settlement agreement because I do not find that it provides any measurable value to music club members. That lawsuit therefore will proceed in the ordinary course.

         II. PROCEDURAL BACKGROUND

         In 1996, the Attorneys General of several states began investigating whether distributors and retailers were illegally conspiring to inflate or support the prices of prerecorded music products through the adoption and implementation of Minimum Advertised Price (" MAP" ) policies.² The Federal Trade Commission (" FTC" ) initiated a parallel investigation. In 2000, the FTC obtained consent decrees against the distributor defendants. See In re Sony Entm't, Inc., Docket No. C-3971, 2000 WL 1257796 (F.T.C. Aug. 30, 2000). On May 10, 2000, private plaintiffs filed the first of over 50 class action lawsuits around the country against the defendants. On August 8, 2000, States, Commonwealths and Territories, through their respective Attorneys General, filed a complaint in the Southern District of New York.³ Generally, the various lawsuits allege that retailers and distributors conspired to fix the price of CDs ⁴ in violation of federal and state antitrust laws.

2 Generally, distributors suggest retail prices (MSRP) that retailers may choose to follow or not. Often, retail sales occur at a discount from the MSRP. Distributors also offer retailers various advertising funds and benefits. Allegedly, under the MAP program, the distributors set limits on the discounts from MSRP that retailers could advertise anywhere, including their own stores, and failure to comply with the discount advertising limitations resulted in loss of the advertising funds and benefits. Retailers were permitted, however, to put a lower sticker price on the CD and charge a lower price.

3 The 43 plaintiff States, Commonwealths and Territories include: Florida, New York, Alabama, Alaska, Arizona, Arkansas, California, Connecticut, Delaware, Hawaii, Idaho, Illinois, Indiana, Iowa, Kansas, Maine, Maryland, Massachusetts, Michigan, Mississippi, Montana, Nevada, New Mexico, North Carolina, North Dakota, Northern Mariana Islands, Ohio, Oklahoma, Oregon, Pennsylvania, Puerto Rico, Rhode Island, South Carolina, Tennessee, Texas, Utah, Vermont, Virginia, Virgin Islands, Washington, West Virginia, Wisconsin and Wyoming. Minnesota and Missouri have dismissed their claims against the defendants with prejudice as of the effective date of the proposed settlements. The interests of their residents in this litigation are now represented by counsel for the private plaintiffs. The residents of the remaining states, territories and the District of Columbia are also represented by counsel for the private plaintiffs.

4 The accusations include vinyl recordings and cassettes as well, but CDs are by far the greatest segment and I shall refer only to them for ease of discussion.

          On October 20, 2000, the Multidistrict Litigation Panel (" MDL" ) transferred 41 of the federal cases to this court pursuant to 28 U.S.C. § 1407. See MDL Order (Docket No. 1). I consolidated the actions for pretrial purposes. See Practice and Procedure Order, Nov. 1, 2000 (Docket No. 3). By Orders dated November 15, December 12, and December 15, 2000 (Docket Nos. 1, 42, 43), the MDL Panel transferred an additional 12 tag-along federal actions. The distributor defendants moved to vacate the conditional transfer order in one tag-along case, Trowbridge v. Sony Music Entm't, Inc., No. 00-6065 (S.D.N.Y.).⁵ They argued that the issues in that lawsuit-involving music club sales-were unique, and differed substantially from the other transferred cases. By Order dated April 18, 2001 (Docket No. 115), the MDL Panel denied the motion to vacate, concluding that the Trowbridge action involved sufficiently common questions of fact with the lawsuits previously transferred.⁶

5 That case was previously captioned Roy v. Sony Music Entm't, Inc., until the plaintiffs filed a Second Amended Complaint (Docket No. 148) withdrawing Margaret Roy and Vincent Vanyserloo as named plaintiffs, leaving Jo Ann Trowbridge and Erinn Bucklan as the named representatives for the class.

6 Counsel for the plaintiffs in another tag-along action originally filed in Louisiana state court and removed to federal court, Chacon v. Best Buy, Inc. (docketed as Civ. No. 01-23-P-H in this court), moved for remand back to state court. By Order dated March 12, 2001 (Docket No. 100), I denied the request.

         During the ensuing pretrial proceedings before this court, (the " multidistrict litigation" or the " MDL case" ), the plaintiffs have proceeded in three groups: (1) the plaintiff States, Commonwealths and Territories (the " plaintiff States" ); (2) the private consumer plaintiffs who purchased CDs in retail stores (the " private plaintiffs" ); and (3) the private plaintiffs in the Trowbridge music club action who purchased CDs through music clubs (the " Trowbridge plaintiffs" ). Correspondingly, three separate amended complaints have been filed in this District. State Pls.' Third Amend. Compl. (Docket No. 190); Private Pls.' Second Amend. Compl. (Docket No. 191); Trowbridge Second Amend. Compl. (Docket No. 148).

         The plaintiff States and the private plaintiffs allege in their complaints that the defendants, three retailers and five distributors, illegally conspired to raise the retail price of CDs by using MAP policies in violation of federal antitrust laws and various comparable state laws. They have prosecuted their complaints jointly and I shall refer to them collectively as the " MAP cases." ⁷ Specifically, the plaintiff States and the private plaintiffs contend that traditional retailers, including the retailer defendants, requested, and the distributor defendants in response implemented, stringent MAP policies in an effort to halt growing price competition from discount retailers.

7 Natural persons in the plaintiff States are represented by the Attorneys General of their respective states pursuant to the parens patriae authority under section 4C of the Clayton Act, 15 U.S.C. § 15c, and parallel state laws. Non-natural persons in the plaintiff States and natural persons in the non-plaintiff states are represented by counsel for the private plaintiffs.

         Separately, the Trowbridge plaintiffs, citing the same MAP policies, allege in their amended complaint that certain manufacturers and distributors illegally conspired with music clubs ⁸ to support the prices of CDs sold to music club members (the " Trowbridge action" ).

8 On November 20, 2001, over the objection of the manufacturers and distributors, I granted the Trowbridge plaintiffs' motion to file a Second Amended Complaint. The Second Amended Complaint added the following music clubs as defendants: The Columbia House Company and BMG Music Service. Second Amend. Compl. ¶ ¶ 12, 13 (Docket No. 148).

          I appointed liaison counsel for the plaintiff States, the private plaintiffs, the distributor defendants and the retailer defendants, respectively. Order Appointing Liaison Counsel, Nov. 29, 2000 (Docket No. 14). I also appointed the Attorneys General of New York and Florida as lead counsel for the plaintiff States. Oral Order, Nov. 28, 2000, Hr'g Tr. at 62 (Docket No. 23). The defendants did not request appointment of lead counsel, but several law firms sought appointment as lead counsel for the private plaintiffs. I designated Kohn, Swift & Graf, P.C. as lead counsel for the private plaintiffs.⁹ Order Appointing Lead Counsel, Jan. 26, 2001 (Docket No. 54). I assigned specific responsibilities to lead counsel and liaison counsel respectively, so as to move the litigation along as expeditiously and efficiently as possible. Because the Trowbridge plaintiffs had only a single lawsuit, that portion of the litigation did not require similar appointments, but their lawyers, Wolf Haldenstein Adler Freeman & Herz, LLP, proceeded under similar directions. See Order Describing Counsel's Duties, May 25, 2001 (Docket No. 131). I ordered the lawyers to negotiate a protocol for discovery, directed the monthly filing (at first with the court under seal and later with lead counsel) of detailed time and expense reports for any counsel expecting to seek attorney fees, instructed the lawyers to be efficient and to avoid duplication of efforts, and scheduled recurrent conferences. See Order, Nov. 29, 2000 (Docket No. 15).

9 Kohn, Swift & Graf and Milberg Weiss Bershad Hynes & Lerach LLP (" Milberg Weiss" ) sought appointment as co-lead counsel. Lead Counsel Application (Docket No. 19). A motion was filed to disqualify Milberg Weiss. I concluded that Kohn, Swift & Graf had the experience, skill, resources and expertise best able to move this matter forward and designated that firm as lead counsel. I deferred action on the motion to disqualify Milberg Weiss, but later disqualified the firm on the basis that it had formerly represented four small music retailers in an antitrust class action against CD distributors. Order Mot. Disqualify, Mar. 12, 2001 (Docket No. 99).

         In addition to the multidistrict proceedings in this federal court, several parallel cases have been pending in various state courts, probably the largest number in California. In an effort to minimize duplicative discovery and undue burden on the parties and non-parties, counsel proposed a joint discovery order requiring that all discovery and pretrial scheduling in this MDL case and in the California state-court proceedings captioned Compact Disk Cases be coordinated before this court. After receiving written proposals by the lawyers, I consulted by telephone (with the lawyers' consent) with Commissioner Bruce E. Mitchell, Coordination Trial Judge in the California Superior Court for the Compact Disk Cases. We proceeded to enter a Joint Coordination Order in our respective proceedings, designed to permit other state courts to join in the Coordination Order if they chose to do so. While I am not aware of any other state court formally entering the Joint Coordination Order, I understand that discovery in the other parallel state court actions has been deferred pending the outcome of this MDL case. I have informed the other presiding judges in writing concerning the progress of this MDL case from time to time.

         Discovery has been extensive and disputed, beginning with initial delays in even obtaining FTC documents. See Conference of Counsel, Jan. 31, 2001, Hr'g Tr. at 14-20 (Docket No. 79). Among other things, the plaintiffs and distributor defendants have jointly subpoenaed over 50 non-party retailers to obtain retail sales data. But the documents and data proved difficult to obtain, and were maintained and retained less uniformly than the parties anticipated. As discovery progressed and retained experts began to yield their conclusions, the parties tentatively began to explore settlement in the fall of 2001. Over the course of several months, settlement agreements were negotiated through direct communication between the parties and a mediation and telephone conferences with Magistrate Judge Margaret J. Kravchuk of this District. In July 2002, the Trowbridge plaintiffs filed a stipulation of settlement. In September 2002, the private plaintiffs and the plaintiff States filed notices of agreement and settlement with the distributor and retailer defendants.

         On October 15, 2002, I held a preliminary fairness hearing. As a result of that hearing, I approved notice to the respective classes of the proposed settlements, and conditionally certified the MAP and Trowbridge classes for settlement purposes. On May 22, 2003, I held a final fairness hearing.

         I now address and decide the issues presented in the final fairness hearing, first for the MAP cases, then separately for the Trowbridge action.

         III. MAP CASES

         A. SUFFICIENCY OF NOTICE

          In parens patriae lawsuits, the Clayton Act requires that the State Attorneys General provide notice to class members.¹⁰ The statute states expressly that notice by publication is appropriate unless the court finds that such notice would violate due process. 15 U.S.C. § 15c(b)(1). For those not represented by the State Attorneys General, Rule 23(c)(2) of the Federal Rules of Civil Procedure requires that class members be provided with " the best notice practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort." Rule 23(e) further requires notice of a proposed settlement " in such manner as the court directs." The notice must describe fairly, accurately and neutrally the claims and parties in the litigation, the terms of the proposed settlement, and the options available to individuals entitled to participate, including the right to exclude themselves from the class. See, e.g., Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 811-12, 105 S.Ct. 2965, 86 L.Ed.2d 628 (1985); see also Hoffmann-La Roche, Inc. v. Sperling, 493 U.S. 165, 174, 110 S.Ct. 482, 107 L.Ed.2d 480 (1989); Greenspun v. Bogan, 492 F.2d 375, 382 (1st Cir.1974). Notice must be reasonably calculated to reach potential class members. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 174, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974); Mullane v. Cent. Hanover Bank & Trust Co., 339 U.S. 306, 315, 70 S.Ct. 652, 94 L.Ed. 865 (1950) (" The means employed must be such as one desirous of actually informing the absentee might reasonably adopt to accomplish it." ).

10 Although the Clayton Act parens patriae provision does not speak of a " class," it is convenient to use that term for both the private class action members and the residents represented by the State Attorneys General. I will do so without further explanation in the opinion.

         Here, given the millions of unidentified CD purchasers during the class period, individual notice was not possible. Instead, the notice program consisted of print, broadcast and electronic publication, and employed a short-form and long-form notice. Both forms described the basic elements of the case, the general terms of the proposed settlement, the legal rights of affected consumers, and the process for filing a claim, but the short form was only a summary. The short-form notice also included dedicated Internet and mailing addresses and a toll-free telephone number that consumers could use to obtain additional information about the proposed settlement or to request a copy of the long-form notice. The long-form notice provided greater detail regarding the terms of the proposed settlement.

         The short-form notice appeared in numerous paid media outlets, including thirteen consumer magazines, two nationally-circulated newspaper supplements and nine newspapers. MAP Final Approval Mem., Ex. H., Kinsella Aff. ¶ ¶ 10, 12-14 (Docket No. 241).¹¹ The notice was also broadcast via one hundred fifty-one radio spots. Id. ¶ 17. The plaintiffs' notice expert, Katherine Kinsella, President of Kinsella Communications, Ltd., calculated that through these efforts notice reached at least 85% of the music-consuming public in the country with an average estimated frequency of more than four times. Id. ¶ 21. The long-form notice was published on the Internet and sent by direct mail to anyone who requested it. In addition, other print and electronic media voluntarily covered the proposed settlement and the claims process extensively. Id. ¶ 22. As a result, over 5 million visitors logged on to the settlement website, www.musiccdsettlement.com, over 200,000 calls reached the toll-free number, over 100,000 requests for long-form packages were processed, and over 3.5 million claims were ultimately filed. MAP Final Approval Mem. at 76-77 and Ex. A, Potter Aff. ¶ ¶ 6-14.

11 The list was expanded as a result of the preliminary fairness hearing to achieve greater coverage of Hispanic and African American consumers.

          Two objectors claim in a joint filing that the short-form notice was defective for failing to specify the amount of attorney fees being sought, the terms of the release being requested, and the specific states being represented by the Attorneys General. Fetcher and McGuigan Objection (Docket No. 224). I reject the objections. The very point of a short-form notice is to capture attention without overwhelming the reader, and permit him/her to obtain more detail if desired, by going to the long form. (The information was contained in the long form.) Trying to put too much detail on the short form would defeat its very purpose. The short form used here, consisting of a series of easily understood questions and answers central to the decision whether to file a claim, was appropriate.¹²

12 I recognize that the National Association of Consumer Advocates (" NACA" ) has advocated putting the total maximum fees, how they are calculated, and where they came from in a plain language short-form notice (actually the first page of a long notice). NACA, Standards and Guidelines for Litigating and Settling Consumer Class Actions, 176 F.R.D. 375, 400 (1997). If that can be accomplished without taking the " short" out of the short-form notice, so much the better. I decide only that it was not mandatory here.

         I conclude that the contents of the notices provided class members with sufficient information to make an informed and intelligent decision about whether to file a claim, seek exclusion from the class, or object to the proposed settlement. I also conclude that the notice distribution was excellently designed, reasonably calculated to reach potential class members, and ultimately highly successful in doing so. I therefore find that the notice program provided the best practicable notice under the circumstances and complied with the requirements of both 15 U.S.C. § 15c(b)(1) and Rule 23 of the Federal Rules of Civil Procedure.

         B. CERTIFICATION OF THE CLASS

         Rule 23 is inapplicable to parens patriae actions, because Attorneys General are specifically authorized by statute to sue on behalf of natural persons residing in their respective states. See 15 U.S.C. § 15c(a)(1). The discussion in this section, therefore, is limited to the private class actions for residents of the remaining States, Commonwealths and Territories, and non-natural persons countrywide.

          To be certified for purposes of settlement, the class must meet the requirements of Rules 23(a) and 23(b)(3) of the Federal Rules of Civil Procedure. Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 613-14, 619-22, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997).

         1. Rule 23(a)

         Rule 23(a) sets forth four prerequisites for class certification: numerosity; commonality; typicality; and adequacy of representation.

          (a) Numerosity.

          To satisfy the requirement of numerosity, the proposed class must be " so numerous that joinder of all members is impracticable." Fed.R.Civ.P. 23(a)(1). Here, the class is estimated to consist of millions of CD purchasers, geographically dispersed throughout the United States, its Commonwealths and Territories. There is no dispute that joinder is impracticable under the circumstances. Thus, I find that this class satisfies the numerosity prerequisite.

          (b) Commonality.

          In order to maintain a class action, there must be " questions of law or fact common to the class." Fed.R.Civ.P. 23(a)(2). Complete identity of legal claims and factual circumstances is not required. Rather, this requirement is satisfied " if all class members are in a substantially identical factual situation and the questions of law raised by the plaintiff are applicable to each class member." In re Disposable Contact Lens Antitrust Litig., 170 F.R.D. 524, 529 (M.D.Fla.1996) (citations omitted). Here, the facts and law common to every class member are nearly the same. All class members are purchasers of CDs during the class period, January 1, 1995 to December 22, 2000. While there may be nominal factual variations among specific claims, the plaintiffs' expert asserts without contradiction that the market for music CDs is national in scope, CDs are fundamentally similar, and the defendants' conduct had a common impact on prices nationally. MAP Final Approval Mem., Ex. G, Del Roccili Decl. at 3-37. Moreover, all class members would have to prove that the defendants conspired to fix prices in violation of the antitrust laws; that this violation caused injury to the class members; and that damages are ascertainable. It is well established that " [a]nti-trust, price-fixing conspiracy cases, by their nature, deal with common legal and factual questions about the existence, scope and effect of the alleged conspiracy." In re Sugar Indus. Antitrust Litig., 73 F.R.D. 322, 334 (E.D.Pa.1976). I conclude that the factual and legal questions in this case are sufficient to support a finding of commonality.

          ( c) Typicality. The claims or defenses of the named plaintiffs also must be " typical of the claims or defenses of the class." Fed.R.Civ.P. 23(a)(3). This requirement is satisfied " if the representative plaintiff's claims are based on the same legal theory and arise from the same practice or course of conduct as the other class members." In re Playmobil Antitrust Litig., 35 F.Supp.2d 231, 241 (E.D.N.Y.1998). In this case, the claims of the named representatives are typical because, like all class members, they allege that during the class period they purchased CDs at supracompetitive prices set by the defendants as part of a price-fixing conspiracy in violation of the antitrust laws. I find that the typicality requirement is satisfied.

          (d) Adequacy of Representation.

          Finally, the representative plaintiffs must fairly and adequately protect the interests of the class. Fed.R.Civ.P. 23(a)(4). This inquiry focuses on both the adequacy of the named plaintiffs and the adequacy of class counsel. It is satisfied if: (i) the interests of the representative parties do not conflict with the interests of any class members; and (ii) the plaintiffs' counsel is qualified, experienced, and able to prosecute the action on behalf of the class vigorously. Andrews v. Bechtel Power Corp., 780 F.2d 124, 130 (1st Cir.1985). As noted previously, the named plaintiffs' injuries are the same as those of the class members. There has been no indication of any conflicts among class members. I conclude, therefore, that the named plaintiffs are adequate representatives of the interests of the class. With respect to class counsel, I find that the quality of their pleadings, diligence in investigating and prosecuting this matter, and extensive experience in class actions and antitrust litigation demonstrate that they have protected the interests of the class competently and adequately, and will continue to do so.

         2. Rule 23(b)(3)

         In addition to the prerequisites of Rule 23(a), the class must also satisfy those of Rule 23(b)(3): namely, (1) questions of law or fact common to class members must predominate over any questions affecting only individual members; and (2) the class action must be superior to other available methods for the fair and efficient adjudication of the matter.

          (a) Predominance.

          The Rule 23(b) predominance inquiry is " readily met in certain cases alleging ... violations of the antitrust laws." Amchem, 521 U.S. at 625, 117 S.Ct. 2231. " The ... inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation." Id. at 623, 117 S.Ct. 2231. Predominance is satisfied " unless it is clear that individual issues will overwhelm the common questions and render the class action valueless." In re NASDAQ Market-Makers Antitrust Litig., 169 F.R.D. 493, 517 (S.D.N.Y.1996).

         It is clear that individual issues in this case will not overwhelm the common questions of law or fact, because the central question is whether the defendants conspired to fix prices and, if so, by how much. There is no doubt that the plaintiffs would present common evidence regarding the existence and scope of the alleged conspiracy at any trials of this matter. Similarly, although a wide range of CDs and prices are potentially involved, the price for CDs is not negotiated disc by disc; common proof of pricing and discounting will apply to many claims. The fact that individual class members' damages may vary due to quantity of purchases or types of CDs purchased does not defeat predominance. See, e.g., In re Master Key Antitrust Litig., 528 F.2d 5, 12 n. 11 (2d Cir.1975); In re Indus. Diamonds Antitrust Litig., 167 F.R.D. 374, 383 (S.D.N.Y.1996) (citing cases); In re Alcoholic Beverages Litig., 95 F.R.D. 321, 327 (E.D.N.Y.1982).

          (b) Superiority.

          With respect to the superiority requirement, a court must consider these factors:

(A) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.

Fed.R.Civ.P. 23(b)(3). Neither the plaintiffs nor the defendants have argued that any individual class member has an interest in maintaining a separate lawsuit or that there is any problem in consolidating the litigation here. Moreover, the Judicial Panel on Multidistrict Litigation ordered the coordination and transfer of these cases for pre-trial proceedings on the basis that centralization of the many federal court actions would serve the convenience of the parties and witnesses, promote the just and efficient conduct of the litigation, avoid unnecessary duplication of discovery and judicial rulings, and conserve the resources of the parties, counsel and the judiciary.¹³ There are no unusual difficulties to be encountered. Finally, the 43 parens patriae actions will proceed on behalf of their " class" members regardless, representing far and away the great bulk of the " class." It makes no sense, therefore, to deny class action status to the remainder. I conclude that the class action is the superior method for resolving this controversy.

13 Cooperation with state courts, see supra pp. 6-7, also supports the conclusion that concentrating the litigation in this forum is advantageous.

         For the foregoing reasons, I find that the requirements of Rule 23 are satisfied, and I certify the following settlement class for purposes of the MAP cases:

All natural persons in the states of Colorado, Georgia, Kentucky, Louisiana, Massachusetts, Minnesota, Missouri, Nebraska, New Hampshire, New Jersey and South Dakota, in the District of Columbia, and in the U.S. Territories of Guam and American Samoa and all non-natural persons or entities in the United States and its Territories who purchased music products from one or more retailers during the period January 1, 1995 to December 22, 2000, but excluding the defendants, their subsidiaries, affiliates, officers, directors, and employees, and excluding those persons or entities who timely and validly requested exclusion from participation in this litigation.

         C. FAIRNESS, REASONABLENESS AND ADEQUACY OF THE SETTLEMENT ¹⁴

14 There are actually four settlement agreements, one with each defendant retailer separately and one with all five distributors collectively. (Docket Nos. 183-86) They are uniform in most respects material to this opinion, however, and I shall treat them collectively as one settlement agreement, except where otherwise noted.

         Rule 23(e) of the Federal Rules of Civil Procedure requires the court to determine whether the proposed terms of a settlement are fair, reasonable and adequate. See, e.g., City P'ship Co. v. Atlantic Acquisition Ltd. P'ship, 100 F.3d 1041, 1043 (1st Cir.1996); see also D'Amato v. Deutsche Bank, 236 F.3d 78 (2d Cir.2001). Similarly, Section 4C of the Clayton Act directs that the parties may dismiss or compromise a parens patriae antitrust action only with court approval. 15 U.S.C. § 15c(c). While the Clayton Act does not specify the standard for entering into a proposed settlement agreement, courts generally have looked to the standards used to analyze settlements under Rule 23. See, e.g., In re Toys " R" Us Antitrust Litig., 191 F.R.D. 347, 352 (E.D.N.Y.2000); New York v. Reebok Int'l Ltd., 903 F.Supp. 532, 535 (S.D.N.Y.1995), aff'd, 96 F.3d 44 (2d Cir.1996).

          There is no single test in the First Circuit for determining the fairness, reasonableness and adequacy of a proposed class action settlement. In making this assessment, other circuits generally consider the negotiating process by which the settlement was reached and the substantive fairness of the terms of the settlement compared to the result likely to be reached at trial. See, e.g., Weinberger v. Kendrick, 698 F.2d 61, 73-74 (2d Cir.1982). Specifically, the appellate courts consider some or all of the following factors: (1) comparison of the proposed settlement with the likely result of litigation; (2) reaction of the class to the settlement; (3) stage of the litigation and the amount of discovery completed; (4) quality of counsel; (5) conduct of the negotiations; and (6) prospects of the case, including risk, complexity, expense and duration. See Molski v. Gleich, 318 F.3d 937, 953 (9th Cir.2003); In re General Motors Corp. Pick-Up Truck Fuel Tank Litig., 55 F.3d 768, 785 (3d Cir.), cert. denied, 516 U.S. 824, 116 S.Ct. 88, 133 L.Ed.2d 45 (1995); Gottlieb v. Wiles, 11 F.3d 1004, 1014 (10th Cir.1993); In re Drexel Burnham Lambert Group, Inc., 960 F.2d 285, 292 (2d Cir.1992); City of Detroit v. Grinnell Corp., 495 F.2d 448, 465, 468-70 (2d Cir.1974). Finally, the caselaw tells me that a settlement following sufficient discovery and genuine arm's-length negotiation is presumed fair. See City P'ship Co. v. Atlantic Acquisition Ltd. P'ship., 100 F.3d 1041, 1043 (1st Cir.1996); see also New York v. Reebok, Int'l Ltd., 903 F.Supp. 532, 535 (S.D.N.Y.1995), aff'd, 96 F.3d 44 (2d Cir.1996).

          1. Value of the Settlement.

          Obviously, the first and fundamental question is how the value of the settlement compares to the relief the plaintiffs might recover after a successful trial and appeal, discounted for risk, delay and expense. On potential recovery, I find persuasive the analysis of the plaintiffs' economist, Dr. Frank Lichtenberg, that if the plaintiffs win on liability, the most likely damages they could prove at trial are $245 million.¹⁵ MAP Final Approval Mem., Ex. C, Lichtenberg Rpt., at 12. That simple statement, however, hides a host of difficult issues in the calculation and proof of damages. First, the plaintiffs have been able to obtain only limited data about CD sales for the class period, missing data from many important retailers and distributors. So far, the plaintiffs have data from five large retailers.¹⁶ The economist correlated those sales data with data available from two of the defendant distributors, BMG and Sony. Through various extrapolations, he estimated overcharges of $39,975,241, for CD sales of those two defendant distributors through the three defendant retailers.¹⁷ Id. at 7-8. Needless to say, the missing data produce very difficult sampling and extrapolation issues for proof about transactions involving other retailers and distributors.¹⁸ All the economist could do was assume that behavior was uniform for the other defendant distributors. He thereby generated an overcharge figure of $103,786,651 for CDs of all five defendant distributors sold through all three defendant retailers. Id. at 8. He then assumed that pricing behavior of the five largest discounters (WalMart, Best Buy, Target, Kmart and Circuit City, for only two of which he had data) mirrored that of the three retailer defendants. As a result, he extrapolated an overall estimated overcharge of $245,393,122, a figure that included the five discounters' CD sales.¹⁹ But the plaintiffs will be able to recover antitrust damages only for sales where they can prove that the retailer in question agreed with someone about the price scheme; otherwise they face the Illinois Brick hurdle of being unable to recover as indirect purchasers. ²⁰ If the plaintiffs can prove a " horizontal" agreement among all the retailers in question, it might not be difficult to prove that it encompassed advertising of all CD discounts. (The defendants disagree.) But if the plaintiffs can prove agreement only between or among certain retailers, the damages are correspondingly reduced; if the plaintiffs can prove agreement only " vertically" between retailer and distributor, then the antitrust damages may be limited to the CDs sold by that retailer obtained from that distributor. Moreover, it may be that any proven agreement(s) will not be product-wide, but limited to certain musical genres or certain promotions. All of these are serious potential hurdles in proving damages.²¹ Finally, the plaintiffs must prove to a jury's satisfaction the economist's inference from the statistics: that the MAP program caused the reduction in price discounts and the resulting higher prices.

15 I do not use the FTC's estimate of $480 million in overcharges, announced (with no analysis provided) during a May 2000 press conference lauding the FTC's accomplishment in obtaining a cease and desist order. See Press Release, Federal Trade Commission, Record Companies Settle FTC Charges of Restraining Competition in CD Music Market (May 10, 2000) (attached as Ex. A to Feldman Objection (Docket No. 216)). None of the objectors who seek to have me measure the settlement against the FTC's estimate has shown me that the FTC did a careful market pricing analysis to generate the number-or that it considered what could actually be proven at trial.

16 Barnes & Noble, Best Buy, Circuit City, Musicland and Tower Records (only two of the three retailer defendants).

17 Oversimplified, what the economist did was calculate average discounts before, during and after the MAP enforcement period. He found a significant decrease in the average discount during the enforcement period and attributed that to the alleged conspiracy. He used that average number and, through weighting, came up with an average of $.23 overcharge per CD. He then used that figure, by projection, for other distributors.

18 The Clayton Act does allow some flexibility in proving damages against a defendant who has agreed to fix prices (" by statistical or sampling methods, by the computation of alleged overcharges, or by such other reasonable system of estimating aggregate damages as the court in its discretion may permit." ). 15 U.S.C. § 15d.

19 The assumption of uniformity seems particularly subject to challenge for the discounters' behavior. Moreover, proof of their agreement (presumably coerced) may be more difficult.

20 In Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977), the United States Supreme Court held that generally only direct purchasers from the antitrust violator could recover damages.

21 The economist has not included sales from small retailers because there is no data and the factors that drive their pricing are quite different. I find that to be a reasonable exclusion. Moreover, to include those sales would require use of the " umbrella" theory of price fixing damage, a subject of controversy unto itself. Compare, e.g., Mid-West Paper Co. v. Cont'l Group, Inc., 596 F.2d 573, 583-87 (3d Cir.1979) (rejecting umbrella theory as a basis for establishing injury by purchasers from competitors of the defendant price-fixers); with In re Beef Indus. Antitrust Litig., 600 F.2d 1148, 1166 n. 24 (5th Cir.1979) (concluding one who deals with non-conspiring competitor of price-fixers has standing to sue price-fixers); see also Fed. Trade Comm'n v. Mylan Labs., Inc., 62 F.Supp.2d 25, 38-40 (D.D.C.1999) (citing cases).

         Against that optimistic potential recovery I must assess the value of the proposed settlement. In addition to cash payments of $67.375 million, the settlement agreement calls for the distributors to provide $75.7 million worth of CDs. The parties therefore maintain that the settlement agreement totals $143.075 million in calculable benefit to the plaintiffs. In addition, the settlement agreement provides for injunctive relief against the distributor defendants for a seven-year period. Since the FTC had previously obtained the identical injunctive relief, that aspect deserves no additional value. ²² Finally, the settlement agreement adds new equitable relief in the form of a five-year injunction against the three retailer defendants. Specifically, the injunction prohibits the retailer defendants for a period of five years from " solicit[ing], demand[ing], request[ing], advocat[ing] or encourag[ing] any Distributor or Wholesaler of Music Product to adopt or implement any policy practice or plan which makes receipt of any cooperative advertising or other promotional funds contingent upon the price or price level at which any Music Product is advertised, promoted, offered or sold." MAP Final Approval Mem. at 9. The FTC did not obtain that relief, so that element does represent value added by this settlement. See In re Toys " R" Us Antitrust Litig., 191 F.R.D. 347, 355 (E.D.N.Y.2000).

22 Arguably there is some incremental value inasmuch as these plaintiffs can now seek to enforce the injunctive relief if the FTC fails to do so.

          Is an All-Cash Settlement Required? I reject the argument (made by some objectors) that only an all-cash settlement should be approved. Providing a settlement in this case that is fair to all class members is exceedingly difficult. Many millions of people purchased CDs during the period; inevitably, not all of them will be aware of the settlement program and their right to file a claim; for a variety of reasons, not all who are aware will choose to participate; and there are practical difficulties in calibrating the dollar recovery to the actual loss of any given consumer, approximately $.23 per CD purchase, according to the plaintiffs' economist. (People do not keep records to remember or prove their purchases by title, at which stores, on which dates, for which amounts.) In this context-where a cash settlement will not reach every injured member of the class and administratively cannot be calibrated reliably to the actual loss suffered by each member-there is positive value to the class members in the so-called " cy pres program" that the parties have developed. Under that program, music CDs, the subject matter of the alleged illegal behavior, will be made widely available free of charge for public enjoyment. Here, each individual State, Commonwealth or Territory will receive a pro rata share (according to population) of the 5.6 millions CDs (1,960 negotiated titles), and will distribute the CDs in accordance with its own particular plan for public use (these have all been filed with the court and provide, for the most part, distribution to libraries and schools). ²³ As a result, members of the public (and thus potentially class members who did not file a claim, as well as those who did) will benefit either in using the CDs themselves or in the general public benefit from recurrent music CD availability.²⁴ Finally, as far as I can tell (and no one, including the objectors, has briefed the issue), a federal court has no authority over what a State does with amounts it receives if a parens patriae case proceeds to judgment. The Clayton Act specifies that the parens patriae lawsuit is " on behalf of natural persons residing in such State," 15 U.S.C. § 15c(a)(1), but says only that any treble damage award goes to the State, § 15c(a)(2). There is no specification of what the State may, or must, do with it.²⁵ Accordingly, given the apparent freedom in what to do with a money judgment, when 43 State Attorneys General recommend a cy pres treatment of part of their respective States' recovery in a settlement, I heed, although I am not bound by, their views on its advisability.

23 For example, the State of Iowa's distribution plan for the estimated 57,436 CDs that it will receive calls for the allocation of the CDs to Iowa public and academic libraries, Iowa college and university nonprofit radio stations, nine residential institutions or facilities operated by the Iowa Department of Human Services, and the Iowa Braille and Sight Saving School. See MAP Final Approval Mem., Ex. F18, at 1. The distribution plan of the Commonwealth of the Northern Mariana Islands provides that its entire allocation of CDs will be distributed to the Commonwealth's three public libraries located on each of the three main islands. See MAP Final Approval Mem., Ex. F38, at 1.

24 Because of cost factors, defendants are obviously willing to turn over greater value (as reflected by value to the recipient) in product than in cash. It is administratively impractical, however, to distribute CDs to individual class members. Allocating them according to the individual's choice of title, with shipping and handling expenses, to 3.5 million addresses would be excessive. The combination of significant cash payments to individual class members, coupled with the cy pres distribution of CDs, is a reasonable and creative resolution.

25 Some states have passed legislation that specifically addresses how monetary recoveries in antitrust actions will be treated. See, e.g., Ariz.Rev.Stat. Ann. § 41-191.01-.02 (2003) (establishing revolving fund for the deposit of monies received by the state as a result of the enforcement of federal or state antitrust statutes and mandating that a portion of any antitrust recovery be set aside for antitrust enforcement activities).

          How Should the CDs be Valued? The objectors argue strenuously that the settlement CDs are not worth the value the settling parties have assigned them. The parties negotiated over how to value the CDs; they also negotiated over which titles would be included (ultimately, 1,960). In fact, the plaintiffs went to some lengths to ensure that the settlement CDs cover a wide range of musical genres and that they are desirable product, not simply " discount bin" items. The distributors initially wanted the CDs valued at suggested retail price; the plaintiffs wanted them valued at wholesale price, typically 30% less than retail. The parties ultimately agreed to value them at 20% less than retail. As a result, the $75.7 million product agreement will provide 5.6 million CDs, given the 1,960 titles selected.

          In truth, it is not easy to assign a value to the resulting 5.6 million CDs. From the plaintiffs' point of view, the benefit could be the retail price consumers would have to pay. (The plaintiffs agree with the objectors that the average list price of the 1,960 titles is $17.20.) MAP Final Approval Mem. at 38-39. That is the approach taken by some courts. See, e.g., In re Toys " R" Us Antitrust Litig., 191 F.R.D. 347, 352 (E.D.N.Y.2000). But realistically, I would have to consider not only the suggested retail price, but also the variety of discounts available in stores or online (and shipping and handling charges). I have no information about the availability of those discounts for volume purchases of this sort (if this were to be a cash purchase), nor do I have any information about volume pricing generally. There would also be severe administrative costs in even trying to assign these more precise values, particularly in a context of changing discounts and pricing. From the distributors' point of view, the cost to them could be the minimal level of providing additional copies of an already engineered recording (I do not have that cost information), or the higher level of the lost sale for each CD, probably best measured at the wholesale price the distributor could recover. Again, however, the variety of discounts in the market suggests that number to be malleable, and not susceptible to precise proof.

         Both the plaintiffs' and the distributors' viewpoints are relevant in assessing value, and neither is determinative. Ultimately, I am satisfied that these are not bargain basement CDs; the list of CDs shows that the plaintiffs have obtained good product covering diverse artists and genres. I am also satisfied that valuing the CDs at suggested retail price would overvalue them because of the volume involved and because of discounting behavior in the market. In fact, the plaintiffs' economist has calculated that the average price in the marketplace is 14.8% less than the MSRP.²⁶ MAP Final Approval Mem., Ex. C, Litchtenberg Rpt., at 6. Thus, the parties' 20% reduction for setting a value on the product is not unreasonable, with a resulting total value of $75.7 million. To be conservative, however, I also consider the consequences if I were to value the CDs at wholesale, 30% less than the average MSRP. Then the total value would be $66.2375 million,²⁷ an average price of $11.83 per CD, for a list of titles that runs all the way to historic recordings and boxed sets.²⁸

26 The economist's estimate is based on data from 2000, after enforcement of the MAP policies ceased. There has been no indication, however, that the average discount in the marketplace has changed.

27 This number results from a mathematical calculation: if $75.7 million is 20% less, then $66.2375 million is 30% less.

28 The parties' agreement, valuing them at MSRP less 20%, results in an average price of $13.76 per CD.

         I find it reasonable, therefore, to accept the parties' negotiated total value for the settlement (before the value-added injunctive relief and before subtracting attorney fees) of $143.075 million, about 58.3% of the potential recovery estimated by the plaintiffs' expert,²⁹ and significantly greater than the recovery for which data exists without extrapolation. If I award 10% of that amount in attorney fees and disbursements (see below), the net recovery to the class is $128,767,500, or 52.5% of the potential damages to be recovered at trial, without assigning any value to the injunctive relief.³⁰           The fundamental question, then, is whether a settlement in excess of 50% is reasonable. The objectors maintain that it is not; that the FTC settlement made this case a slam-dunk victory for the plaintiffs; and that the States and the private class plaintiffs have surrendered too soon for too little. The plaintiffs and the defendants, on the other hand, take pains to show all the difficulties in the case, with the defendants asserting that the case is even more difficult than the plaintiffs recognize. As supervising judge, I am not to prejudge the merits of the case (I may have to decide the merits if the settlement fails), and I am not to second-guess the settlement; I am only to determine if the parties' conclusion is reasonable. See, e.g., Evans v. Jeff D., 475 U.S. 717, 727, 106 S.Ct. 1531, 89 L.Ed.2d 747 (1986); Cotton v. Hinton, 559 F.2d 1326, 1330 (5th Cir.1977). With that background, I proceed to consider the other factors that appellate courts have identified.

29 At wholesale prices, the total value of the settlement would be $133.6125 million ($67.375 million in cash plus $66.2375 million in product), about 54.5% of the potential recovery, and still significantly greater than the recovery for which data exists without significant extrapolation.

30 At the hearing, I pressed the plaintiffs' lawyers on whether I should treble the calculation of recoverable damages for purposes of assessing the fairness, reasonableness and adequacy of settlement. Reluctantly, they agreed that I should. Further research after the hearing, however, reveals that courts generally do not consider either recovery of treble damages or recovery of attorney fees in assessing the settlement. See City of Detroit v. Grinnell Corp., 495 F.2d 448, 458-59 (2d Cir.1974); In re Warfarin Sodium Antitrust Litig., 212 F.R.D. 231, 257-58 (D.Del.2002); In re Lorazepam & Clorazepate Antitrust Litig., 205 F.R.D. 369, 377 n. 12 (D.D.C.2002); In re Ampicillin Antitrust Litig., 82 F.R.D. 652, 654 (D.D.C.1979). The basis for this treatment seems to be a Second Circuit decision, Grinnell, where the court gave two reasons: (1) tradition; and (2) disruption of the settlement process because trebling would require defendants to " admit their guilt." 495 F.2d at 459. I am not persuaded by either reason: tradition is hardly compelling, given the other interests at stake here; and the whole premise of a settlement (except settlements only to avoid legal expense) is defendant recognition that liability is a possibility. Trebling is automatic if antitrust damages are assessed at all, unlike other forms of punitive damages that require additional proof or that may be left to court discretion. Respected commentators who disagree with automatic trebling nevertheless recognize that the statute seems to require it. 2 Phillip E. Areeda et al., Antitrust Law ¶ 303d, at 41, ¶ 330b, at 274 (2d ed.2000). Therefore, if a settlement reflects a potential damage recovery, it should logically reflect the other parts of that recovery (trebling and attorney fees) that the statute awards automatically. Certainly it is at least at the back of lawyers' minds as they negotiate. Nevertheless, my analysis in text follows the precedents. On the other hand, if trebling is counted, the potential recovery after trial and appeal would be $736 million and fees. The total value of the settlement as calculated in text then would represent about 19.4% of the potential trebled recovery and after subtraction of attorney fees and disbursements would represent about 17.5%, without considering value to the injunction against retailers. Obviously those are less attractive percentage numbers than presented in text under the traditional analysis. But even so, given my role as supervising judge not to decide the merits of the case but only to determine if the parties' conclusion to settle on these terms was reasonable, and given the presumption of fairness following arm's-length negotiations and given all the difficulties the litigation presents, I do not find the settlement unreasonable.

          2. Class Reaction.

          The response to the settlement has been generally positive. As a result of the notice program, 11 objections were filed in this court on behalf of 22 affected consumers,³¹ 121 consumers have opted-out or attempted to do so, and over 3.5 million people have submitted claims. Given the size of the class, the number of objections and opt-outs is miniscule.

31 See Feldman Objection (Docket No. 216); Padilla Objection (Docket No. 218); Ferguson Objection (Docket No. 219); Kirk Objection (Docket Nos. 220-21, 239); Marsh Objection (Docket No. 222); Fetcher and McGuigan Objection (Docket No. 224); Simon Objection (Docket No. 225); Walker et al. Objection (Docket No. 226) (objection filed by the Washington Legal Foundation on behalf of 11 class members); Spillane Objection (Docket No. 231); Pearson Objection (Docket No. 232); Brewer Motion (Docket No. 245); see also Plath Letter (Dec. 31, 2002); Ford Letter (Mar. 1, 2003); Stephens Letter (Jan. 29, 2003); Fontan Letter; Jackson Letter. See Binder of Requests for Exclusion and Objections, at tab 7 (Docket No. 262).

          3. Stage of Litigation.

          The case is being settled after enough time for serious discovery and evaluation. The plaintiffs had the FTC's conclusion in 2000; the benefit of the discovery the FTC performed; the benefit of discovery that some States had done starting in 1996, both before and during the FTC investigation; and the discovery performed during two years in this multidistrict litigation, all totaling hundreds of thousands of pages. Collectively those measures generated significant information about the prospects and pitfalls. During the multidistrict litigation, both sides had the opportunity to find out what pricing data were available from retailers and what were not, which sampling technologies would work and which would not. They hired experts to analyze and extrapolate from the available data. Motions to dismiss were resolved. Under closely monitored case management practices, the consolidated cases were well underway to the final stages of motions for summary judgment, class certification and trial either here in Maine or in their home districts from which they were transferred,³² followed by appeal.

32 The United States Supreme Court decided in Lexecon, Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26, 40, 118 S.Ct. 956, 140 L.Ed.2d 62 (1998), that a transferee court such as Maine may not retain the cases for trial, but must ship them back to their respective originating districts when pretrial proceedings are complete. Arguably, Lexecon would not govern the 43 parens patriae lawsuits if the MDL Panel transferred them here for trial. See 28 U.S.C. § 1407(h). Arguably also, the two consolidated amended complaints filed in this District voluntarily by the States and the private plaintiffs respectively amount to a consent to transfer. After all, it would be hard to know where to " send back" the consolidated complaints.

          4. Quality of Counsel and of Representation.

          Counsel for the State Attorneys General and the private plaintiffs are all very well-qualified and experienced, matched only by the qualifications and experience of the defendants' counsel. I was involved in the selection and approval of lead and liaison counsel for the plaintiffs and thereby gave particular attention to competence and experience. I observed the lawyers' performance in written arguments, in several in-person appearances and by telephone conference. Uniformly they did their homework and were prepared and effective advocates. Their performance confirmed my confidence in their abilities and dedication to their clients' causes. I also take note of the fact that 43 State Attorneys General, by virtue of their office assigned to defend the best interests of their respective citizens, have endorsed the fairness, reasonableness and adequacy of the proposed settlement.

          5. Conduct of Negotiations.

          The negotiations went on for a period of nine months. From all that I have personally observed in the case, I am confident that the negotiations were at arm's length. The parties sufficiently persuaded me of the seriousness of their undertaking to suspend, and to continue to suspend, the case management deadlines, something I am usually loath to do. A magistrate judge was involved, at the parties' request, in settlement discussions concerning one or more retailers.

          6. Case Prospects, Including Risk, Complexity, Expense and Duration.

          The reasonableness, fairness and adequacy of the settlement in this matter ultimately depend on the strength of the plaintiffs' case and the opposing defenses. Having monitored this multidistrict litigation closely, having decided motion practice, having stayed aware of the discovery progress and knowing that there is motion practice still to come, I am confident that this is both a complex and a difficult case. Significant and expensive discovery remain for the plaintiffs on liability and damages. More experts will have to be hired at great expense. The defendants are promising to resist the class certification motion for the residents of those states and territories not represented by State Attorneys General. There will be difficult summary judgment motions on proof of agreement,³³ and the proper legal standard ( per se vs. rule of reason) if an agreement can be proven. If they survive summary judgment, the plaintiffs must persuade a jury that there was an agreement among the defendants, against the defendants' argument that the practices here were independently adopted. Moreover, the plaintiffs have no direct precedent for their argument that the case can proceed on a theory of per se illegality.³⁴ Although price agreements are per se illegal, see, e.g., Arizona v. Maricopa County Med. Soc'y, 457 U.S. 332, 102 S.Ct. 2466, 73 L.Ed.2d 48 (1982), by their terms the MAP policies at best were agreements on what prices could be advertised, not what prices could be charged. If the plaintiffs cannot proceed on the per se illegality theory, they will have to satisfy the rule of reason analysis. This will require expert testimony defining the market (geographically and by product) and will give the defendants the opportunity to assert business justifications, see, e.g., Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 483, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992), such as maintaining a diversity of retail outlets so as to market a wide range of music genre CDs, in the face of the impact of the giant discount chains and their focus on the most popular titles. I have previously mentioned some of the damage proof difficulties.

33 If the plaintiffs cannot prove that a particular retailer joined in the alleged agreement, Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977), may prevent them from including that retailer's sales in their damage calculation, although some of the pendent state law claims do not pose the Illinois Brick obstacle.

34 The caselaw on how to treat MAP agreements has been divided, see In re Nissan Antitrust Litig., 577 F.2d 910, 915-17 (5th Cir.1978), and the FTC has vacillated. See 1 ABA Section of Antitrust Law, Antitrust Law Developments 144-45 (5th ed.2002).

          So, the plaintiffs face the following hurdles: significant and expensive additional discovery; difficulty both in gathering data and then in proving antitrust damage causation; hiring more experts and opposing the defendants' experts; defending attacks on the private class status; and defending against serious motions for summary judgment, given the difficult issues in proving first, the agreement, and second, damages. The plaintiffs thereafter will have to prepare for and attend a lengthy trial, perhaps in several locations; ³⁵ motions for judgment as a matter of law at the close of those trials; post trial motion practice; and lengthy appeals. To be sure, some of these are obstacles for the defendants as well, but during all this time, the individual consumers, members of the class, will have no recovery.

35 See supra note 32. There are also a number of pending state lawsuits.

         I have no crystal ball on how the lawsuits and trials will come out, but under these uncertainties and given the time horizons, I find the settlement in excess of 50% of the likely possible damages to be reasonable in value. ³⁶

36 I have assumed that the defendants collectively are able to pay any judgment, although there may be an issue for one or another as reflected, for example, in the retailer payment schedules.

          7. Miscellaneous.

          There are some additional issues of fairness that objectors have raised. First, some have argued that it is unfair to award a uniform payment per class member that does not vary with the number of purchases made, and that it is unfair to allow recovery without some proof of purchase. I reject those challenges. Most people do not routinely keep records of such modest purchases for years after they occurred; the parties did try to determine if the UPC symbols on the CDs (assuming the purchased CDs still exist) could be used to verify transactions, but found that UPC symbols could not perform that function (and the administrative costs would be high in any event); the penalty of perjury language is at least some deterrent against fraudulent claims; a single claim per person avoids the worst of possible defrauding activity, albeit reducing the recovery of large volume purchasers; and at the same time, the likely price loss on any single CD purchase is in pennies. I find the proposed resolution-one claim per person, with a simple signed statement under penalty of perjury-to be a fair method of allocation, given the proof costs and the administrative expenses.³⁷

37 At a loss of $.23 per CD purchased, a recovery of $13 covers the purchase of about 55 CDs over the six-year class period.

         A second and somewhat related challenge is that some people who bought used CDs may have filed claims, thereby diluting the recovery; or, alternatively, that perhaps buyers of used CDs were damaged by the defendants' actions and should have been solicited to file claims. I reject this challenge as highly speculative. There is no evidence from anyone that the market for used CDs was affected in any way, no evidence that retailers' sales of used CDs is an important segment of the market, and no evidence that there are likely to be a large number of used CD claims. I have already ruled that the claims process is a reasonable solution, even though it is conceivable that some spurious claims have crept in among the 3.5 million.

         Finally, there is the matter of $8.5 million that has been set aside from the cash settlement to cover the costs of the notice program, settlement administration and distribution of the settlement proceeds, including the distribution of the CDs to and within the States, Commonwealths and Territories. MAP Final Approval Mem. at 66-67. Approximately $5.1 million has been spent or committed at this stage, and it seems likely that not all of the $8.5 million will be needed. Under the settlement agreement, any extra is allocated pro rata to the States, Commonwealths and Territories under the cy pres distribution program. I was concerned initially (as were some of the objectors) that leftover amounts might be claimed as additional attorney fees, but at the final fairness hearing the plaintiffs' lawyers agreed that would not occur. I was also concerned that the unspent funds would be an excessive amount to distribute under the cy pres program, with no guarantee how the recipients would spend it. The lawyers assured me, however, that if the actual distribution checks (now estimated at $12.63 or higher each) are delayed for a few weeks, they will have a better understanding of the expense needs and will be able to increase the individual checks going to class members so as to reduce any potential leftover cy pres cash amounts to a minimum. See Potter Supp. Aff. ¶ 6 (Docket No. 257). Counsel shall report to the court on the status of these numbers by June 30, 2003, and before distributing any funds to the class.

          8. Distribution Plan.

          I find the distribution plan for the cash to be fair, adequate and reasonable. I have already explained why it is not feasible to provide recovery calibrated to the number of CDs each person purchased during the class period or to require proof of purchase. The claim form here is simple and fair, and will result in 3.5 million class members getting a measurable, uniform, cash recovery.

         I am likely to find the distribution plans for the 5.6 million CDs fair, adequate and reasonable, provided that certain additional information and/or assurances are provided. I have already explained why it is administratively impractical to mail CDs to each class member according to personal choice. ³⁸ Instead, the cy-pres distribution provides for wide circulation of music CDs at a reasonable distribution cost. (The cost of distributing the 5.6 million CDs is estimated to be $700,000, an average cost of 12 cents per CD, reasonable under any analysis. See MAP Final Approval Mem., Ex. A, Potter Aff. ¶ 30.) But the Settlement Agreement provides: " Distribution under the Cy-Pres Distribution Plan shall be to not-for-profit corporations and/or charitable organizations and/or governmental or public entities, with express conditions ensuring that the funds or Non-Cash Consideration be used to further music-related purposes or programs reasonably targeted to benefit a substantial number of the persons who purchased Music Products from one or more Retailers." Settlement Agmts. ¶ 4.9.4 (emphasis added). I do not easily find the corresponding guarantee in the individual state distribution plans for either the CDs themselves or any remaining cash to be distributed. By June 30, 2003, the plaintiffs shall either point me to the relevant provisions or explain how they propose to satisfy that requirement.³⁹

38 See supra note 24.

39 It would not be consistent with the purposes of the settlement for a recipient to be able to sell or give away the CDs upon receiving them, or to be able to use any distributed cash for unrelated purposes.

         I therefore approve the cash distribution plans, and DEFER ruling on the CD distribution plans.

         D. ATTORNEY FEES AND EXPENSES

         The plaintiffs have requested a total attorney fee award of $12,575,000, together with disbursements of $1,873,651.44, for a total award of $14,448,651.44. Jt. Mot. Att'y Fees (Docket No. 204). This amount would come out of the settlement fund. Whatever is awarded will be shared among the States who are participating in the lawsuits and the private lawyers who are representing the class members. The States and the lawyers have not told me how they will divide any award, except that they have agreed among themselves. The State Attorneys General have asked me to approve how their offices propose to use the fee award they receive. Objectors have raised a number of challenges: (1) that States should receive no attorney fees because State Attorneys General are already paid by state tax money and further amounts should not be subtracted from amounts due the residents who pay those taxes; (2) that I should insist on knowing the fees each law firm or State will receive and evaluate that amount against the benefit accruing to the class members or state residents respectively represented; (3) that I should assess the attorney fees by using the lodestar approach and find them excessive (the same for disbursements plus, in the latter case, reject them for inadequate documentation); and (4) that I should appoint an independent auditor to review the fees and disbursements for me.

          1. States' Right to Recover Attorney Fees and Expenses.

          The State Attorneys General are in federal court pursuant to the Clayton Act. That statute provides specifically that their respective States can recover their reasonable attorney fees and disbursements if they succeed on their parens patriae claims. 15 U.S.C. § 15c(a)(2). Congress therefore has made the judgment that even though States use public servants paid by state-raised tax monies to pursue these claims, they may recover attorney fees and disbursements against defendants. The question then becomes whether these fees, that are statutorily available upon judgment, are unavailable if a case settles short of judgment. The objectors have cited me to no case that so holds, and in fact States have been awarded fees for settlements. See, e.g., In re Toys " R" Us Antitrust Litig., 191 F.R.D. 347, 357 (E.D.N.Y.2000) (approving $5.4 million in fees plus costs to the States and class plaintiffs' counsel pursuant to final approval of settlement agreement). It would be a perverse reading of the statute to allow States to recover their attorney fees only upon judgment, thereby building in a strong disincentive to the settlement of lawsuits to the public benefit. To be sure, the fees after judgment are taken from the defendants' pockets, whereas fees before judgment come from the settlement amount and therefore arguably from the residents' pockets.⁴⁰ But that is why a judge is asked to review the settlement and the fees to approve their reasonableness in light of the overall settlement. I conclude that the States are entitled to a fee award. I also conclude, however, that it is none of my business how the respective States subsequently use that award. The Clayton Act provides only that the award is made to the State. I understand that the State Attorneys General might like some assistance in saying that the money should be invested back into similar enforcement activities (and I note that the settlement agreement so provides in paragraph 4.3), but that is a matter for each State, not for this federal court.⁴¹ I therefore decline to approve or disapprove what the States might do with the fees they recover.

40 Over the years there has been discussion of the proper way to settle cases involving attorney fees, and whether the case should first be settled ignoring fees, leaving them to the judge to assess later and additionally against defendants; whether the two numbers should be negotiated separately; or whether the defendants can settle on a global amount, then leave it to the plaintiffs, their lawyers and the court to work out the division. The caselaw and the relevant rules certainly permit the manner in which the plaintiffs' counsel have proceeded here, negotiating a lump sum amount. See Evans v. Jeff D., 475 U.S. 717, 733, 106 S.Ct. 1531, 89 L.Ed.2d 747 (1986); White v. New Hampshire Dep't of Employment Sec., 455 U.S. 445, 453 n. 15, 102 S.Ct. 1162, 71 L.Ed.2d 325 (1982).

41 See supra note 25.

          2. Allocation.

          I also conclude that I need not know how the attorney fees will be divided.⁴² It would not be improper for me to know, but it is not necessary for me to know. See In re Warfarin Sodium Antitrust Litig., 212 F.R.D. 231, 262 (D.Del.2002). My concern is with the overall amount being taken from the total recovery and its reasonableness. I also point out that certain objectors' argument that each attorney fee award must relate directly to the benefit accruing to the particular portion of the class represented by those lawyers receiving that fee is directly contrary to what a supervising judge is expected to do in complex, multidistrict litigation. I appointed particular law firms with special tasks as lead and liaison counsel to avoid overlap and duplication and encouraged those law firms to divide up necessary tasks among the participating lawyers based on experience and efficiency. The result is that lawyers or law firms may have properly spent their time on tasks that are not related to the particular benefit of the portion of the class that originally brought them into the lawsuit.

42 At the hearing, private counsel from California and Minnesota announced that they were no longer pressing their objection to their respective treatments.

          3. Lodestar or Percentage of Funds.

          Under First Circuit law, a district court may assess fees against a common fund recovery (which is what this case is),⁴³ " either on a percentage of the fund basis or by fashioning a lodestar." In re Thirteen Appeals Arising out of the San Juan Dupont Plaza Hotel Fire Litig., 56 F.3d 295, 307 (1st Cir.1995); see also United States v. 8.0 Acres of Land, 197 F.3d 24, 33 (1st Cir.1999). The First Circuit has also recognized, however, that the percentage of the fund method is the prevailing approach used in common fund cases and that there are distinct advantages for using it in such cases. In re Thirteen Appeals, 56 F.3d at 307; see also Third Circuit Task Force Report, Selection of Class Counsel, 208 F.R.D. 340, 422 (2002) (suggested procedures for reviewing fees in non-auction cases). That is the method I will use-is the total fee reasonable when examined as a percentage of recovery? The lodestar approach (reasonable hours spent times reasonable hourly rates, subject to a multiplier or discount for special circumstances, plus reasonable disbursements) can be a check or validation of the appropriateness of the percentage of funds fee, but is not required. In re Thirteen Appeals, 56 F.3d at 307; see also Third Circuit Task Force Report, 208 F.R.D. at 422 (expressing skepticism of even this limited use of the lodestar).

43 Under the Clayton Act, this is a statutory fee-shifting case, but only upon judgment. 15 U.S.C. § 15c(a)(2). When the case settles short of judgment, as this one has, the plaintiffs' attorney fees come out of the recovery and then fit the common fund model.

         Initially upon reading the papers and hearing the argument I was uncertain of the premise for the fees sought here, inasmuch as the lawyers justified the numbers on both a percentage of funds and the lodestar method, and appeared to use unusually precise numbers for the former, going down to tenths of a percent. Ultimately, however, the plaintiffs' lawyers clarified at the hearing that they had started with a 10% basis for the applicable percentage of the fund, subsequently modifying it somewhat.⁴⁴

44 I still do not fully understand how the lawyers came up with their precise fee request ($12,575,000), since it matches neither the percentage of funds number nor the ostensible lodestar number. Some of the caselaw seems to permit the approach they took. See, e.g., In re Toys " R" Us Antitrust Litig., 191 F.R.D. 347, 356-57 (E.D.N.Y.2000) (" Whether evaluated under the conventional lodestar method, or as a percentage of recovery, the amounts requested [$3,258,000 in fees, representing 5.7% of the total settlements and 16% of the cash portion of the settlements, plus $458,784.60 in expenses and $2,176,000 in compensation to the States, representing 3.8% of the total value of the settlements and 11.05% of the aggregate cash value of the settlements, plus $86,912.35 in costs] are reasonable." ) (citations omitted). Perhaps it results from the appellate pronouncements that either percentage of funds or lodestar can support fees in cases of this kind, In re Thirteen Appeals Arising out of the San Juan Dupont Plaza Hotel Fire Litig., 56 F.3d 295, 307 (1st Cir.1995), and resulting lawyer uncertainty as to what approach a court will use. For that confusion, we judges are responsible. But it seems to me that lawyers are obliged first to explain to the court how they came up with the numbers in their fee request, i.e., choose one method and follow it through, then if necessary show that the alternate method would support it. Choosing a number they are happy with, then justifying it on both approaches, is neither a logical method of determining fees nor a device that facilitates judicial review.

         I conclude that 10% is a reasonable percentage for overall recovery, given the size of this case, its difficulty and complexity, and the recovery generated, all of which I have discussed previously in detail. Whether that is high, medium or low seems to depend on whom you ask and how you divide up the cases. It is certainly within the range that has been approved in other cases, ⁴⁵ and I find it appropriate. Although some of the cases and commentaries call upon district judges to give a precise analysis of how they choose the percentage, I do not know how to do so in any meaningful way that would defend 10% over 11% or 9%, etc. Instead, I find the proposed fee reasonable under all the circumstances.

45 See discussion of ranges in In re Copley Pharm., Inc., 1 F.Supp.2d 1407, 1412-13 (D.Wyo.1998) (contingency fee standard of 33%; generally 20-30% in common fund cases; 25% benchmark in Ninth Circuit; 13-20% range where recovery is $51-75 million; 6-10% range and lower where recovery is $75-200 million); In re Prudential Ins. Co. of Am. Sales Practices Litig., 148 F.3d 283, 339 (3d Cir.1998) (4.1%-17.92% where recovery exceeds $100 million); Conley v. Sears, Roebuck and Co., 222 B.R. 181, 188 (D.Mass.1998) (quoting the affidavit of expert Charles Silver, a professor at the University of Texas School of Law: " Although the available empirical evidence does not enable me to make authoritative generalizations about fee award practices in class actions where public law enforcement officials and private attorneys are working cooperatively, the anecdotal evidence known to me suggests that fees ranging from 10-15 percent of the recovery are usual in these cases." ); see also NACA, Standards and Guidelines, supra note 12, at 397-98. This court is not within the Seventh Circuit, where I would have to determine what is the market rate for these legal services. See In re Synthroid Mktg. Litig., 325 F.3d 974, 978-79 (7th Cir.2003); In re Synthroid Mktg. Litig., 264 F.3d 712, 718-22 (7th Cir.2001).

          Applying the 10% against the number I approved as the settlement value, namely $143,075,000, yields $14,307,500.⁴⁶ Applied against the cash-only portion of the settlement, the fee represents 21%. Put still another way, the fee per class member (3.5 million claims) is $4.09, for a class member's net cash recovery of about $13, well below the level of most contingent fee agreements and not even considering the value of the cy pres distribution and injunctive relief. By contrast, the lodestar calculation already exceeded this number by January 2003. Although I have not scrutinized details of the plaintiffs' lawyers' lodestar calculation (because I am using the percentage of funds approach), I do not find the total lodestar number surprising given the length of time some State Attorneys General have been engaged and given the complexity of this MDL litigation and the issues involving data evaluation and analysis. I note also that there has been extensive legal activity since January, and fees and expenses therefore have been and continue to be ongoing since the lodestar number was calculated. In sum, I find the 10% request reasonable because the plaintiffs have obtained an excellent outcome after long, diligent work of high quality.

46 In a single sentence, the lawyers seem to seek a pro rata share of any interest the settlement funds are earning in the bank accounts where they sit. Courts have debated the appropriateness of allowing interest on fees when payment is delayed. It is often said that under the lodestar approach lawyers are permitted to calculate their fees based on current billing rates, thereby foregoing interest, instead of using historical rates and charging interest. For a percentage of funds approach, on the other hand, the rationale for interest disappears. All such variables should be subsumed in the determination of what is an appropriate percentage. Of course, if there were to be substantial delay in payment after a court's approval of fees, interest might be appropriate. I do not see any reason for such an award here, and decline to award interest.

         Now what is included in the approved number may create a problem. The plaintiffs are seeking a total of $14,448,651.44, including both attorney fees and disbursements. The First Circuit has said that a trial court may " set the percentage at a level which not only accounts for fees, but also suffices to cover reimbursable expenses in whole or in part," but only if it " set[s] forth specific reasons for selecting the percentage and ... explain[s] its analysis with particularity." In re Fidelity/Micron Sec. Litig., 167 F.3d 735, 737 (1st Cir.1999). I am not sure what the First Circuit had in mind for such an analysis in deciding whether to make the percentage include disbursements. It cited an Eleventh Circuit case that incorporated the lodestar factors, added additional factors, posited still additional unidentified factors unique to each case and concluded by requiring that a district court " identify all factors upon which it relied and explain how each factor affected its selection of the percentage of the fund awarded as fees." Camden I Condominium Ass'n, Inc. v. Dunkle, 946 F.2d 768, 775 (11th Cir.1991). As I have already stated, I have no idea how to do that.⁴⁷ I believe that 10% is a fair number, but I cannot defend its choice against 9.9% or 10.1% or 9.8% or 10.2%, etc. ⁴⁸

47 Even the factors approved in the Third Circuit Task Force Report, taken directly from Gunter v. Ridgewood Energy Corp., 223 F.3d 190, 195 n. 1 (3d Cir.2000), do not assist in this calculation:

48 See Third Circuit Task Force, Court Awarded Attorney Fees, 108 F.R.D. 237, 245 (1985):

          Suffice it to say here that if the plaintiffs press their request for a total fee and disbursement amount that exceeds the overall 10% figure I have approved, they will have to provide more support for their disbursements than they have to date. To justify taking more from the class, I will have to be satisfied with the hotel accommodations they selected, the air fares they obtained, the meals they charged (multidistrict cases obviously involve much travel and attendant expense, over $150,000 here), the photocopying they have charged (about $130,000 for private counsel alone, and the States have incurred over $285,000 for data/photocopying expenses), etc.⁴⁹ The plaintiffs shall notify the court by June 23, 2003, whether they press their case for disbursements in an amount that would lift the total award above the 10% I have approved.⁵⁰

49 Many of the reasons for preferring the percentage of funds approach over the lodestar analysis apply to the disbursement review as well. It seems a step backward to revert to judicial second guessing of expenses.

50 If the total award stays at $14,307,500, and if the claimed disbursements of $1,873,651.44 are reasonable, then the net attorney fees are $12,433,848.56, 18.5% of the cash recovery, or 8.7% of the total recovery.

          4. Independent Auditor.

          Because I use the percentage of funds approach, the services of an independent auditor to check the lodestar numbers (fees and disbursements) are unnecessary.

         E. OBJECTIONS

         I have dealt with all the serious objections in the course of this discussion. ⁵¹ The rest are without merit and deserve no discussion. ⁵²

51 Some have become moot, such as those objections to the ceiling or floor for payments (neither turned out to be applicable) and those challenging notice due to a low response rate (30,000, according to one objection) before all the 3.5 million claims were filed.

52 Class action litigation stirs the interest and concern of many. Primarily, they are concerned that class actions serve to line lawyers' pockets without significantly benefiting allegedly damaged class members, or that the class action device promotes the filing of spurious lawsuits forcing defendants to pay only in order to avoid the high expenses of class action litigation. What is less frequently observed is that the process provokes into action an interesting group of professional objectors, who seem to have a variety of agendas, some not always apparent. I read all the objections, and heard orally all those who wished to be heard, both at the outset of the hearing, and again in rebuttal at the end. Their contributions varied from useful and important (whether I accepted them totally or not) to the other end of the spectrum. The culmination, however, was a post-hearing " Notice of Objection," in which two objectors, represented by lawyers from Massachusetts and Missouri, " hereby object to any other class member's adoption of oral or written arguments made by [the two objectors] that were not specifically addressed in that class member's filed, written objections." Feldman and Marsh Notice of Objection, at 1 (Docket No. 268). That is a first, in my experience. Generally, adoption and endorsement of someone else's arguments are high forms of flattery; I am not aware of any legal principle that gives a monopoly over arguments to the one who first made them; and I specifically instructed objectors at the hearing not to repeat points that others had made, but merely to state that they agreed with or adopted them!

         IV. TROWBRIDGE (MUSIC CLUB) ACTION

         For the lawsuit challenging music club practices, there are no State Attorneys General as plaintiff, and only a single lawsuit, transferred (over objection) from the Southern District of New York, with two class representatives and one law firm representing them. The analysis, therefore, is slightly different.

         A. SUFFICIENCY OF NOTICE

          I will not repeat my earlier discussion concerning the standards of notice. I add that individualized notice by first-class mail ordinarily satisfies the requirement that class members receive the best notice practicable under the circumstances. See, e.g., Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 175, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974); Peters v. Nat'l R.R. Passenger Corp., 966 F.2d 1483, 1486 (D.C.Cir.1992) (citing cases).

          Here, the notice program provided just such individualized notice by mail. In particular, direct notice was mailed to over 8.1 million potential class members identified from the membership records of the music club defendants. Trowbridge Mem. at 35 (Docket No. 253); Jaffe Decl. ¶ ¶ 3, 33 (Docket No. 254). The notice described, among other things, the class, the litigation, the proposed settlement, how an affected consumer could opt-out or object, and the fund for costs, attorney fees and incentive awards. Trowbridge Mem. Ex. 1A. The notice also directed individuals seeking additional information to a website address and provided counsel's address for mailing or delivering any written comments. In addition, a summary notice was published in the national editions of USA Today and Newsweek. The plaintiffs' notice expert estimated that the publication reached millions of readers. Trowbridge Mem. Ex. 3; Jaffe Decl. ¶ ¶ 3, 34.

         I find that the notice given to the plaintiff class was completely adequate and appropriate for this class action and fulfilled all the requirements of due process and Rule 23 of the Federal Rules of Civil Procedure.

         B. CERTIFICATION OF THE CLASS

         I will not repeat my earlier discussion on the requirements of class certification but turn directly to the facts of this case, involving a proposed class of music club members.

         1. Rule 23(a)

          (a) Numerosity.

          Here, the class is estimated to consist of millions of music club members nationwide who purchased prerecorded CDs from August 11, 1996 to July 31, 2002. There is no dispute that joining all members of the class would be difficult, inconvenient and impracticable. I therefore find that this class satisfies the numerosity prerequisite.

          (b) Commonality.

          The Trowbridge action is based on the same overarching conspiracy as the MAP cases: to fix at supra competitive levels the prices that certain consumers paid for CDs. For the same reasons previously discussed, I conclude that the questions of law and fact common to music club class members are sufficient to support a finding of commonality.

          (c) Typicality.

          In this case, the claims of the named representatives are typical of the claims of all class members, because they are based on the same legal theories and circumstances. Like all music club class members, the named plaintiffs contend that the defendants violated the antitrust laws by conspiring to inflate the prices of CDs sold through music clubs. As a result, the plaintiffs allegedly paid higher prices than they would have paid absent such collusion. I find that the typicality requirement is met.

          (d) Adequacy of Representation.

          First, I find that the named representatives, music club members, are adequate representatives of the class of music club members. The injuries alleged by the named plaintiffs are the same as those of the class members. In addition, the named representatives have followed the proceedings, submitted to depositions, and responded to extensive written discovery requests, including turning over their entire collections of CDs. Trowbridge Mem. at 41; Jaffe Decl. ¶ 51. Second, I find that class counsel is able and experienced and has served as a competent and diligent advocate for the interests of class members.

         2. Rule 23(b)(3)

         As for Rule 23(b)(3), for the same reasons enumerated for the MAP cases I find that questions of law or fact common to class members predominate over any questions affecting only individual members and that the class action is superior to other available methods for the fair and efficient adjudication of this action.

         For the foregoing reasons, I find that the requirements of Rule 23 are satisfied.

         C. Fairness, Reasonableness and Adequacy of the Settlement

          The proposed settlement generates $1,025,000 in cash,⁵³ and a program of 50% discounts ⁵⁴ (free shipping and handling) for music club members who purchased CDs from the defendant music clubs during the class period. The parties propose to devote the cash to the plaintiffs' costs of notice (about $72,000) and incentive payments of $5,000 each to the two named class representatives. Class counsel seeks the remainder (about $940,000) for attorney fees and disbursements (the latter about $105,000). Thus, the only value to class members is the discount program.           I have great difficulty finding that the discount program has any identifiable value. Even if I find value, I cannot quantify it,⁵⁵ and it certainly is not significant value. The objectors presented a substantial amount of documentation that music clubs already make many discounts available to their members. These discount programs differ in some ways from the settlement discount,⁵⁶ but not in any way that permits me to quantify the settlement discount program as either a benefit to the class members or a detriment to the defendants.⁵⁷ At the end of the day, I have no solid evidence that there are members who already use all their coupon discounts and for whom the settlement discounts will therefore add significant value, or that there are members who do not like the terms of current discount offers and therefore will find the settlement discounts more attractive to use. This would be the sort of evidence that might show some value to the members.⁵⁸ As for detriment to the defendants, the other side of the settlement valuation equation, I have no evidence that use of the settlement discounts will displace some other more profitable sales (full price or discount) that the music clubs might make. Without that evidence, for all I know this will be just another contribution to their marketing program that will result in more profitable sales than previously, and cost them nothing as a settlement.⁵⁹ In short, although some members may like and use the settlement discounts, I do not find them to have any significant benefit to the members of the class as a whole or any measurable detriment to the defendants who are using them to settle the case.⁶⁰ (And they cannot be justified as merely an added sweetener to an otherwise demonstrable cash value to the plaintiff class.)

53 Actually, it generates $1.5 million, but $475,000 is retained by the defendants to pay for their portion of the notice expenses.

54 A 50 % discount on one to three new purchases, depending on the number of CDs the music club member purchased during the class period.

55 A day after the hearing, the Trowbridge plaintiffs filed a letter requesting permission to engage an economist and prepare and file at a later date an analysis of value. I declined the request. I think it better for everyone to have a clear understanding of the scope of my concerns before determining whether investment of further resources is appropriate or advisable. I thought my skepticism and concerns had been sufficiently apparent at the preliminary fairness hearing. See Hr'g Tr. at 55, 59, 63, 70 (" I will look forward in the final fairness hearing to having ... details to persuade me that indeed, there is value to what has been provided here." ) (Docket No. 200). Obviously, however, I was not successful then in communicating the seriousness of my concerns. The parties can determine whether, in light of this opinion, they believe any salvage is possible.

56 The differences that the parties highlight are the following: the percentage amount of the discount (the objectors disagree that this is significant, stating that the clubs regularly offer coupons up to 70 or 80% off, or fixed dollar coupons that amount to greater than 50%); the settlement discount is good for 6 months, whereas other coupons are good for only days or weeks (an objector who appeared in person disagreed, stating that he has a drawer full of coupons that do not expire and submitted one as evidence); the settlement discount requires no other purchases, whereas many of the discounts require purchase of one CD at full price, as a result of which, one or several other CDs become available at varying discounts (on the other hand, there was some evidence that every time a member buys a CD at whatever price, he/she receives a coupon for a CD at a steeply discounted value); the settlement discount makes the half-price CD available with no shipping and handling charges, whereas most other offers require the member to pay $2.79 for shipping and handling; the settlement discount will be exercisable against almost the entire catalog unlike other restricted coupons (objectors disagreed, stating that club discounts often apply across the entire catalog); and the settlement discount is available to past club members who are members of the class, and as inactive club members they would not receive discount offers. (I cannot assess this factor. To be sure, this may be a discount they would not otherwise receive, but their lapsed membership may also indicate lack of interest in the club programs generally and thus the settlement discount may be only an opportunity for the defendants to try to recover lost members.)

57 In this respect it is unlike the coupon settlement in In re Mexico Money Transfer Litig., 267 F.3d 743 (7th Cir.2001), where there was no evidence that other discounts were available, and class members could sell or give away their discounts to others.

58 I do recognize that this discount does not carry the disadvantages of some coupon programs that have been criticized in the cases and commentary, in that the discount covers a substantial part (50%) of the purchase price of a modestly priced item and arguably does not require a change in consumer behavior. See, e.g., In re Gen. Motors Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 808-10 (3d Cir.1995) (rejecting $1,000 settlement coupons for the purchase of a new truck); Clement v. Am. Honda Fin. Corp., 176 F.R.D. 15 (D.Conn.1997) (rejecting $75 or $150 coupon requiring class members to purchase or lease another vehicle through AHFC within two years). On the other hand, the discount is not transferable, such that members with no use for it could sell it to members who could economically use it, see, e.g., In re Mexico Money Transfer Litig., 267 F.3d at 748, and there is no requirement that a certain number of the discounts be used for the defendants to satisfy their obligation. See, e.g., In re Motorsports Merch. Antitrust Litig., 112 F.Supp.2d 1329, 1335-36 (N.D.Ga.2000); see generally Note, In- Kind Class Action Settlements, 109 Harv. L.Rev. 810 (1996), for an analysis of the cases and factors.

59 Defense counsel spoke about " cannibalizing," and paragraph 12 of the Declaration of Sharon Siegel " estimates that perhaps 25-35% of sales that would be made in the ordinary course of business could be replaced by sales made pursuant to the Settlement Consideration." (Docket No. 252, Ex. 4). Notably there is no indication that any sales under the settlement program would substitute for more profitable sales. Affidavits of music club personnel refer to likely redemption rates of 8% or 15%, but they are speaking of music club coupon redemption experience in general, and the rates bear no relation to whether this program will " cost" the defendants anything or " benefit" the members at all. The lawyers would like me to value the settlement by taking the dollar amount of the 50% discount ($8.99 on a $17.98 CD) and multiplying it by a predicted number of redemptions, thereby ostensibly generating over $10 million in value to the class. I decline to accept that analysis for the reasons stated in text.

60 I am not rejecting discounts automatically or outright. I recognize that the costs of processing and mailing a check to a class that may exceed 8 million members will outweigh the value of a small check and may require alternative relief; but this discount program has no demonstrated value to the class members.

         In fairness, I recognize that this is a very difficult case for the plaintiffs to prove. The music clubs did not use MAP programs. To the extent this tag-along suit is related to the claims made in the MAP cases, any recoverable damages depend on the success of the MAP claims, and then the additional finding that music club prices were somehow affected in an antitrust way by whatever lack of discounting there was in retail store prices.⁶¹ There are also problems proving that the music club defendants were part of any pricing conspiracy and the nature of any such agreement. Unless the plaintiffs can prove that the music club defendants were part of the conspiracy, they face standing problems under Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977). The likelihood of success has always been very slim. I do not fault the plaintiffs, therefore, for failing to present an analysis of their potential damage recovery if the case proceeds to trial. It is hard to analyze what may not be there.

61 The damage theory of the MAP case is that retailers failed to discount the suggested retail price as much as they would have without MAP. Music clubs have always maintained a variety of discount programs. It would be very difficult to prove that these music club discount programs were collusively affected by the MAP programs used in retail stores or, if so, in what amount.

          In sum, I have a proposed settlement that produces discounts that some members of the class may want to use, to which I cannot attach a value, and which I do not find to be significant. In turn, it purports to settle a lawsuit that, I conclude, has dubious value itself. Should I therefore simply approve the settlement for what it is worth, confident that the case otherwise will generate absolutely no benefit for the class and knowing that the class members are no worse off, and that some of them at least may like the discounts? I conclude that the answer is No, ⁶² because a settlement is not fair where all the cash goes to expenses and lawyers,⁶³ and the members receive only discounts of dubious value. I did consider approving the settlement, while denying attorney fees to class counsel because of the minimal value conferred. Under the settlement agreement as it exists, however, the parties have agreed that any amount of cash I do not award as fees reverts to the defendants, no help to the class members.⁶⁴ I therefore find the settlement agreement to fail the fair, reasonable and adequate test.

62 My skepticism was surely evident at the preliminary hearing on fairness, see supra note 55. Because I was then uncertain of how music clubs worked and what the benefits might be, I permitted the proposed settlement to go forward for reactions. Although class counsel argues that the reaction was generally positive, I was impressed with the criticisms. In addition to the persuasiveness of some of the objectors, there was a notable number of opt-outs (470), a number of which expressed dissatisfaction with either the lawsuit or the settlement. See Jaffe Aff. (Requests for Exclusion) ¶ 3 (Docket No. 263).

63 Plus $10,000 to the two named plaintiffs.

64 Perhaps aware of my discontent, the lawyer for one of the music clubs proposed at the very end of the fairness hearing that the defendants would waive the reversion and agree that any amounts of the cash pool not awarded to class counsel should go to a music charity.

         Because I disapprove the proposed settlement agreement, I do not reach the attorney fees question or the incentive payments to class representatives. I have no doubt that class counsel worked long and hard hours. They simply did not confer significant benefit upon the class. Probably the nature of the claims they were litigating did not justify their work.

         I note that pursuant to paragraph 5 of the Order Granting Preliminary Approval of the Proposed Settlement (Docket No. 197), this disapproval automatically vacates, without further order, the provisional certification of the class.

         V. CONCLUSION

         A. MAP CASES

         1.

          By June 27, 2003, counsel shall present to me a corrected Order concerning claims, reflecting the rulings I made at the hearing on various questioned categories of claims.

         2.

          By June 30, 2003, counsel shall report the distribution expenses and the revised amount of the checks to be distributed to each claimant.

         3.

          By June 30, 2003, counsel shall provide a response on how the cy pres distribution plans ensure appropriate use of the CDs and any remaining cash.

         4.

          Thereafter, counsel shall provide a revised form of Judgment reflecting the rulings made at the hearing and in this Order.

         B. TROWBRIDGE ACTION

         By June 30, 2003, the parties shall propose new dates for all the deadlines that previously have been suspended during settlement negotiations, so that this matter can move along swiftly until it is ready for trial, and resulting transfer back to the Southern District of New York or other disposition.

         SO ORDERED.

In common fund cases of this sort-in which the attorneys' fees and the clients' award come from the same source and the fees are based on a percentage amount of the clients' settlement award-district courts should consider several factors in setting a fee award. Among other things, these factors include: (1) the size of the fund created and the number of persons benefited; (2) the presence or absence of substantial objections by members of the class to the settlement terms and/or fees requested by counsel; (3) the skill and efficiency of the attorneys involved; (4) the complexity and duration of the litigation; (5) the risk of nonpayment; (6) the amount of time devoted to the case by plaintiffs' counsel; and (7) the awards in similar cases. See In re Prudential Ins. Co. of Am. Sales Practices Litig., 148 F.3d 283, 336-40 (3d Cir.1998); In re General Motors Corp. Pick-Up Truck Fuel Tank Products Liability Litigation, 55 F.3d 768, 819-22 (3d Cir.1995); Moore's Federal Practice, Manual for Complex Litigation (Third) § 24.121, at 207 (1997).

The fundamental problem with an approach that does no more than assure that the lower courts will consider a plethora of conflicting and at least partially redundant factors is that it provides no analytical framework for their application. It offers no guidance on the relative importance of each factor, whether they are to be applied differently in different contexts, or, indeed, how they are to be applied at all.