Opinion
MDL Docket No. 1500 02 Civ. 5575 (SWK).
September 28, 2006
MEMORANDUM OPINION
On September 29, 2006, David Pikus, the special master for attorney's fees, submitted his final report and recommendation ("RR") to this Court and served the RR on all relevant parties. Following the passage of a comment period that yielded only positive feedback, the Court now adopts the RR.
Special Master Pikus recommends an award of $147,500,000.00 in fees and $3,417,237.51 in expenses. He reached this recommendation by employing a two-tiered percentage structure, which averages out to 5.9% of the common fund. This tiered percentage structure rewards class counsel for negotiating a settlement that more than doubled relevant industry predictions, despite the receipt of several smaller, yet compelling, settlement offers at earlier stages of negotiations, and reflects the intent of lead plaintiff's original fee agreement with class counsel. The recommended award, both as a percentage and in overall terms, properly rewards class counsel for their vigorous advocacy of the class's interests over four years of litigation. The Court was particularly impressed by the parties' spirit of cooperation and civility during intensive settlement negotiations. The attorneys' exemplary behavior throughout the litigation decreased the burden on the judiciary and reduced the potential costs to their respective clients.
While Special Master Pikus rightly recommends an award that avoids recourse to a mere benchmark, it is notable that the averaged 5.9% award is within the range of percentage awards that have recently been approved by the courts of this Circuit in megafund settlements. See, e.g., Wal-Mart Stores, Inc. v. Visa U.S.A. Inc., 396 F.3d 96, 121, 123 (2d Cir. 2005) (approving a 6.5% award); In re Worldcom, Inc. Sec. Litig., 388 F. Supp. 2d 319, 353, 360 (S.D.N.Y. 2005) (awarding 5.5%). After considering the factors adopted in Goldberger v. Integrated Res., Inc., which have been thoroughly analyzed in the RR, the Court finds Special Master Pikus's recommendation to be fair and reasonable. 209 F.3d 43, 50 (2d Cir. 2000) (citation omitted).
The administration of the award is to be carried out by reference to the accompanying Order approving the RR and awarding attorney's fees and reimbursement of costs.
ORDER APPROVING REPORT RECOMMENDATION OF SPECIAL MASTER AND AWARDING ATTORNEYS' FEES AND REIMBURSEMENT OF COSTS
The above-referenced matter came before the Court on Lead Securities Counsel's Application for an Award of Attorneys' Fees and Reimbursement of Expenses from the fund created by the settlement with Defendants in this class action (the "Fee Application");WHEREAS, prior to the hearing, a Notice of Proposed $2.65 Billion Settlement of Securities Class Action, Certification of a Settlement Class, Final Approval Hearing, Application for Attorneys' Fees and Expenses, and Proposed Plan of Allocation ("Notice") was directed to all members of the Securities Class at the address of each such person as set forth in the records of Gilardi Co., LLC (the "Settlement Administrator"), by first class mail, postage prepaid, and a Summary Notice of Proposed $2.65 Billion Settlement of Securities Class Action, Certification of a Settlement Class, Final Approval Hearing, Application for Attorneys' Fees and Expenses, and Proposed Allocation ("Summary Notice") was published once in the national editions of Investor's Business Daily, The Wall Street Journal, The New York Times, and USA Today pursuant to the specifications of the Court;
WHEREAS, the Notice and Summary Notice informed the Securities Class of the application of Lead Securities Counsel, made on behalf Securities Class Counsel in the litigation, for an award of attorneys' fees and reimbursement of expenses from the common fund created by the Settlement (the "Settlement Fund");
WHEREAS, the Court appointed David H. Pikus, Esq. as Special Master to review the Fee Application, and Mr. Pikus has submitted to the Court a Report Recommendation of the Special Master recommending an award of $147,500,000 in fees and $3,417,237.51 in expenses;
WHEREAS, having considered the Fee Application, the Report Recommendation of the Special Master, and any and all objections to the Fee Application or Report Recommendation of the Special Master,
NOW, THEREFORE, IT IS HEREBY ORDERED THAT:
1. The Court finds and concludes that the Special Master's recommended award of fees and expenses is fair and reasonable to the members of the Securities Class, and that the expenses for which he recommends reimbursement were reasonable and necessary.
2. Securities Class Counsel are awarded attorneys' fees in the amount of $147,500,000.00, plus interest at the same rate as earned by the Settlement Fund from October 7, 2005, the date it was funded. Lead Securities Counsel is authorized to withdraw said amount from the Settlement Fund and allocate the awarded fees among itself and other Securities Class Counsel based on Lead Securities Counsel's determination of the relative contributions of the firms to the prosecution and settlement of this class action on behalf of the Securities Class.
3. Securities Class Counsel are awarded reimbursement of expenses in the amount of $3,417,237.51, plus interest at the same rate as earned by the Settlement Fund from October 7, 2005, the date it was funded. Lead Securities Counsel is authorized to withdraw said amount from the Settlement Fund for distribution to itself and other Securities Class Counsel based on Lead Securities Counsel's determination of the expenses each paid or may be obligated to pay in connection with the prosecution and settlement of this case.
4. Lead Securities Counsel may hereafter submit one or more supplemental requests for reimbursement of expenses incurred in this matter after the filing with the Court of Lead Securities Counsel's Fee and Expense Application.
5. The finality of the Judgment entered with respect to the Settlement shall not be affected in any manner by this Order, or any appeal from this Order.
6. There is no just reason for delay in the entry of this Order, and immediate entry of this Order by the Clerk of the Court is expressly directed pursuant to Rule 54(b) of the Federal Rules of Civil Procedure.
IT IS SO ORDERED.
REPORT RECOMMENDATION OF THE SPECIAL MASTER
By orders entered February 6 and 28, 2006, I was appointed Special Master of the Court for the purpose of reviewing and making recommendations to the Court upon various applications for fees and expenses filed by attorneys and law firms representing individual claimants or the class in this action. Class counsel requested that the Court enter an award of seven percent of the total recovery, or $175 million, plus reimbursement of costs and disbursements. Although counsel seek an award based upon the percentage method, they report accrued fees, based upon their customary hourly billing rates, in the amount of $46,861,731.25. I respectfully recommend that the applications be granted to the extent of awarding $147,500,000 in fees and $3,417,237.51 in expenses.Background of the Action
Various plaintiffs commenced the underlying actions in 2002, in the wake of the merger between Time Warner and AOL, as well a national recession and a plunging stock market. Plaintiffs alleged that America Online Inc. ("AOL") and its successor merged entity improperly accounted for dozens of advertising transactions, inflating revenue for fifteen quarters between 1998 and 2002. These transactions were, according to the Amended Complaint, designed to create the appearance that they were generating revenue, when in fact they provided only illusory benefits to the company. See Apr. 6, 2006 Opinion Order (approving settlement).
The underlying matter is a class action filed on behalf of a putative class consisting of all persons who, between January 27, 1999 and August 27, 2002, acquired shares of AOL or the combined company. The defendants consisted of AOL Time Warner, its constituent entities, the accounting firm of Ernst Young, LLP, various officers and directors of the company and, initially, certain underwriters.
The Minnesota State Board of Investment was appointed as lead plaintiff. The Minneapolis law firm of Heins Mills Olson, P.L.C. became lead counsel.
The potential breadth of the class was enormous. These securities were traded on national exchanges and very widely held. Plaintiff described the security as "one of the most widely held stocks in the world." (Mem. of Lead Securities Pl. Minn. State Board of Investment in Supp. of Motion for Class Certification at 12.) Class counsel estimated that there could be as many as 4.7 million class members. Approximately 830,000 claimants filed timely proofs. Some 10,082 potential claimants opted out, but this ostensibly large number constituted less than 0.2 percent of the potential class members. Apr. 6 Opinion Order at 24.
The action survived various motions to dismiss, which were complex, time-intensive and granted in limited respects. The parties launched into massive discovery, primarily in the form of document production. Class counsel organized a multi-tiered review structure which, they report, utilized the services of dozens of lawyers and encompassed a staggering total of 15.5 million documents. Pleadings were amended in the wake of motions to dismiss and as evidence was unearthed. Counsel were poised to conduct hundreds of depositions in the fall and spring of 2005-06.
Another Special Master, Paul D. Wachter, was appointed to supervise these gargantuan discovery activities. Through his efforts and the professionalism of counsel on both sides, priorities turned to settlement. The record reveals that these were no ordinary negotiations. Counsel prepared intensively on both sides and presented evidence and argument to each other and to the Special Master. Counsel grappled with the unsettled issue of loss causation, made all the more difficult by the recession and burst of the so-called "technology bubble" which coincided with the wrongs alleged in the amended complaint. Daily records produced by class counsel reflect the time-intensity of the efforts.
On July 29, 2005, the parties fixed the boundaries of a settlement in a "Memorandum of Understanding," which led to a new round of negotiations over a full Stipulation of Settlement. On September 28, 2005, the Court held a preliminary approval hearing to address the settlement documents and, shortly thereafter, certified the class for settlement purposes, while granting preliminary approval to the settlement itself. Ultimately, the settlement called for a payment by defendants directly to the class in an amount of $2.5 billion, to which the Department of Justice agreed to add $150 million from its own settlement. The Court held a final approval hearing on February 22, 2006.
Although a handful of written objections were filed, no objectors appeared at the hearing. Class counsel largely credited their own hard work and creativity for the settlement's magnitude and achievement, and cited a number of serious hurdles and risks in upcoming litigation in endorsing the compromise. Defendants' counsel, in turn, reiterated confidence in their defenses, ascribing their clients' support of the settlement to a desire to obviate distractions and move forward with their business activities. See Feb. 22, 2006 Transcript at 9. In any event, "both parties concluded that the Settlement was the best and most efficient outcome for their clients in light of the costs of litigation and mutability of applicable legal standards." Apr. 6 Opinion Order at 18-19.
On April 6, 2006, the Court entered its Opinion Order approving the settlement. The Opinion Order is testament to the quality, efforts and reputation of class counsel. Lead counsel "has garnered judicial praise for its representation in previous actions, and has continued to show its client commitment and exceptional lawyering in this case." Id. at 18. The Court was continuously "impressed with the quality of representation provided by Lead Plaintiff's Counsel, its prosecution of the lawsuit, and its negotiation of the Settlement." Id. at 13.
In this final phase of the non-ERISA litigation, class counsel seek their award of $175,000,000 in fees plus expenses. Counsel have submitted summaries of their so-called "lodestar," calculated at current hourly rates and of their expenses, broken down by firm. They have also submitted contemporaneously recorded time records, reflecting the services performed by each timekeeper, with corresponding dates and duration.
Counsel submit these records for use as a potential "cross-check" against their fee request, which is based on a percentage of the total recovery. Counsel derive their request largely from a comparison with fees awarded in other oversized class action settlements and support it with a Declaration of John C. Coffee, Jr., a Columbia Law School professor and eminent authority on corporate and securities law. Professor Coffee has been a frequent supporter of class counsel fee applications. Counsel and their expert stress the size of their recovery and the risks attendant to continued litigation.
For example, he filed declarations in support of fee applications or otherwise assisted class counsel in Cendant Corp. Litig. (D.N.J.), Visa Check/Mastermoney Antitrust Litig. (E.D.N.Y.), In re Royal Ahold Securities "ERISA" Litig. (D. Md.), Alcatel Alsthom Securities Litig. (E.D. Tex.) and McNamara v. Bre-X Minerals, Ltd. (N.D. Tex).
They distinguish other cases in which the stock price decline was preceded by the issuer's own restatement of financial results. Here, by contrast, class counsel and their experts had to root out and decipher an abstruse web of different artifices used to inflate the company's accounting for advertising revenues. Counsel also assert that parallel governmental activities did not undergird or materially assist their efforts. As the Court itself recognized, there were unsettled aspects of the relevant law on loss causation, as well as potential problems proving scienter. Apr. 6 Opinion Order at 27. Further, defendants argued that the stock decline was "the result of a number of factors — including the general decline in Internet stock values — unrelated to the allegations of fraud." Id. The confluence of these and other factors, class counsel urge, justifies the enormity of their fee request.
Counsel apprised the class, with their notice of the settlement, of their intention to seek from the common fund $175,000,000 in fees, plus $4,501,332.36 in expenses. Only four objections were filed, none from any major institutional holders.
I circulated a draft of this Report Recommendation to class counsel and to the objectors, and provided them with an opportunity to comment. The draft differed in some respects from this final Report Recommendation, both in terms of a somewhat lower award (a unitary recommendation of five percent) and various provisional findings designed to elicit comment or explanation, including specific analyses of attorney lodestar charges.
This procedure was expressly authorized in a prior version of Rule 53. Although it was removed during a recent rewrite of the Rule, the Advisory Committee made clear that the procedure remains allowable. See Advisory Committee Notes, R. 53, 2003 Amendments.
Prior to the release of the draft, lead plaintiff had not commented at all upon the fee application. Rather than draw an inference either way from lead plaintiff's silence, I also forwarded the draft to the Attorney General of Minnesota in his statutory capacity as attorney for the Minnesota State Board of Investment, inviting his comment. Cf. In re Cendant Corp. Litig., 264 F.3d at 280-81 ("acquiescence [which is the most that a failure to object shows] is not the same thing as `prior approval'") (ellipse in original); see also Eash v. Riggins Trucking Inc., 757 F.2d 557, 559 n. 1 (3d Cir. 1985); U.S. v. State of La., 751 F. Supp. 608 (E.D. La. 1990) (courts have inherent power to appoint amici curiae).
Chief Deputy Attorney General Kristine L. Eiden responded with an informative letter dated August 3, 2006. She noted that lead plaintiff initially had negotiated a tiered percentage fee structure with lead counsel, based upon the size of the recovery and the stage at which it was achieved. The Chief Deputy apprised me that this agreement was amended to provide simply that lead counsel "shall be compensated as determined by the court on behalf of the class based on the law applicable to the court's jurisdiction." She further indicated that the Attorney General had been monitoring the litigation and the fee application and had intended to seek the appointment of a special master if the Court had not appointed one.
Her letter endorsed the draft fee award, which reflected, to an even larger degree, reductions in counsel's fee and expense requests:
The MSBI was supportive of the appointment of a special master to carefully review the application for fees and expenses filed by attorneys and law firms. The Report demonstrates you have made a thorough and careful review of the application and seek to `harmonize the interest of counsel and the class.' We believe your recommended fee award serves the interest of the MSBI and the class well.
Two other previous objectors, Margaret Keffer and Paul Heyburn, responded to the draft. Ms. Keffer stated, in pertinent part:
I think Mr. Pikus did a good job and I accept his recommendation.
Mr. Heyburn, on the other hand, expressed no specific view on the draft. Instead, he suggested yet another analytical approach and requested that I compile and present certain information for 55 different lawyers, including various credentials and their year-by-year time summaries not only on this case, but on their other caseloads as well. I declined his request as unnecessarily time-consuming and not required for a lodestar "cross-check" of a percentage fee. Even if this approach were appropriate, it should have been suggested in his initial objection, not at this late stage. Finally, Mr. Heyburn's standing to participate in the fee process is questionable at best in light of the Court's holding that he had no requisite loss and therefore no standing to object to the settlement. See Apr. 6, 2006 Opinion Order at 38, n. 17.
Neither counsel nor any objector requested a hearing. But in addition to considering the submissions from the Minnesota Attorney General, Ms. Keffer and Mr. Heyburn, I accommodated lead counsel's request to clarify certain aspects of the fee request in the wake of the draft. With commendable candor, counsel acknowledged that certain requested expense reimbursements were more appropriately part of a firm's overhead, and withdrew those components from consideration. Counsel also provided additional substantiation for other expense requests. At the same time, counsel strongly, and, in part convincingly, urged me to increase the amount of my fee recommendation, stressing that the draft percentage was unreasonably lower than that of certain cases which they considered comparable, and that they undertook serious risks in pursuit of this litigation. Counsel noted that they rejected sizable, tempting and justifiable settlement offers before finally obtaining the $2.5 billion proposal from defendants — specifics not readily apparent from counsel's prior submissions.
Counsel also confirmed that they voluntarily abrogated the tiered fee agreement made with lead plaintiff at the outset of the case. That agreement provided for 12 percent of the gross recovery up to $25 million, and two percent increases in each $25 million tier above that amount, up to a maximum of 17 percent. The agreement contained other incentives that are not relevant to the current award. The agreement, though recognized by both attorney and client as excessive, nevertheless reflected ex ante anticipation of a sizable percentage fee and provided for increasing, rather than decreasing percentages, as the recovery climbed.
Discussion
The Two Permissible MethodsThe so-called "common fund doctrine," coupled with Fed.R.Civ.P. 23(e), authorizes the Court to award fees and expenses to class counsel and certain other claimants in a duly certified class action. See In re "Agent Orange" Product Liability Litigation, 818 F.2d 216, 222 (2d Cir. 1987), cert. denied sub nom. Schwartz v. Dean, 484 U.S. 926 (1987). The methodology has evolved in a circular history. In the early days of the common fund doctrine, attorneys fees were generally awarded as a percentage of the recovery. Winkelman v. General Motors Corp., 48 F. Supp. 504 (S.D.N.Y. 1942), aff'd sub nom. Singer v. General Motors Corp., 136 F.2d 905 (2d Cir. 1943). That preference shifted in the 1970s, as the federal courts became concerned that ballooning settlement numbers were generating excessive legal fees under the percentage system. See City of Detroit v. Grinnell Corp., 495 F.2d 448, 468 (2d Cir. 1974).
Led by the Third Circuit, the courts shifted to the rate-based lodestar method. See Lindy Bros. Builders, Inc. v. American Radiator Standard Sanitary Corp., 487 F.2d 161, 167 (3d Cir. 1973). Using this procedure, "the district court scrutinizes the fee petition to ascertain the number of hours reasonably billed to the class and then multiplies that figure by an appropriate hourly rate. Once that initial computation has been made, the district court may, in its discretion, increase the lodestar by applying a multiplier based on `other less objective factors,' such as the risk of the litigation and the performance of the attorneys." Goldberger v. Integrated Resources, Inc., 209 F.3d 43, 47 (2d Cir. 2000) (citations omitted); see also "Agent Orange" Product Liability Litigation, 818 F.2d at 222 ("the court may, in its discretion, increase or decrease [the hourly fees] by examining such factors as the quality of counsel's work, the risk of the litigation and the complexity of the issues").
As the broadening scope and complexity of litigation consumed ever-increasing legal manpower over the ensuing decades, the lodestar fell into disfavor. Critics have contended that the lodestar created a temptation for lawyers to run up the number of hours for which they could be paid, created an unanticipated disincentive to early settlements and compelled district courts "to engage in a gimlet-eyed review of line-item fee audits." Goldberger, 209 F.3d at 48-49; see also Court Awarded Attorney Fees, Report of the 3d Cir. Task Force, 108 F.R.D. 237 (Oct. 8, 1985); Kirschoff v. Flynn, 786 F.2d 320 (7th Cir. 1986).
The common fund cases thus completed their circular journey back to the percentage regime. By the recent turn of the century, the courts of appeals had fully restored the percentage method, some mandating its use, e.g., Swedish Hospital v. Shalala, 1 F.3d 1261, 1265 (D.C. Cir. 1993), with others, including the Second Circuit, retaining the lodestar as an alternative. See Goldberger, 209 F.3d at 50; In re Washington Public Power Supply System Securities Litigation, 19 F.3d 1291 (9th Cir. 1990); Rawlings v. Prudential-Bache Properties, Inc., 9 F.3d 513 (6th Cir. 1993).
Regardless of the method used, the fee determination process represents a departure from the norms of the justice system. The defendants, having settled, have little interest in the division of the spoils. Individual class members, whose numbers are scattered and who often lack a singular interest sufficient to prompt intervention, rarely object. See Thomas E. Willging, Laural L. Hooper Robert J. Niemic, Empirical Study of Class Actions in Four Federal District Courts at 76 (Fed. Jud. Ctr. 1996).
Because the adversarial system breaks down at this point of litigation, just as the interests of the class and its counsel begin to diverge, the Court effectively becomes a fiduciary whose charge is to protect the class against excessive fees. See Goldberger, 209 F.3d at 52; see generally Court Awarded Attorney Fees, 108 F.R.D. 237. Regardless of the method used, "a fee award should be assessed based on scrutiny of the unique circumstances of each case and `a jealous regard to the rights of those who are interested in the fund.'" Goldberger, 209 F.3d at 53, quoting Grinnell, 495 F.2d at 469.
General Principles of the Two Approaches
Although the Second Circuit has provided detailed guidance on the implementation of the two permissible methods, it has left broad discretion to the district courts in the choice between the lodestar and percentage processes. Goldberger, 209 F.3d at 50; Wal-Mart Stores, Inc. v. Visa U.S.A. Inc., 396 F.3d 96 (2d Cir. 2005), cert. denied sub nom. Leonardo's Pizza by the Slice, Inc. v. Wal-Mart Stores, Inc., 544 U.S. 1044 (2005).
The lodestar method has been criticized for generating avoidable hours, discouraging early settlement and burdening district judges with tedious audits of time records. Goldberger, 209 F.3d at 48-49; Alan Hirsch Diane Sheehey, Awarding Attorneys Fees Managing Fee Litigation at 73-74 (Fed. Jud. Ctr. 2d ed. 2005). Even in the commercial marketplace, which, like the lodestar, has long been wedded to the billable hour, small signs of change are emerging. See Douglas McCollam, The Billable Hour: Are Its Days Numbered? 183 N.J.L.J. 285, 286 (Jan. 30, 2006) ("reviled and ubiquitous, the billable hour is the cockroach of the legal world"). The billable hour inherently "encourages firms to overstaff matters, lard their bills with marginally useful services, and draw out cases that might be brought to a swifter conclusion." Id. at 285.
Even so, the conclusion of the cited commentary and others is that the billable hour remains predominant in the world of commercial litigation. See also Karen Dean, The Billable Hour's Staying Power, 185 N.J.L.J. 203 (Jul. 17, 2006) (citing recent survey showing that 94 percent of corporate law departments favored hourly billing from outside counsel). In fact, some authorities suggest that concerns over the lodestar's susceptibility to excessive hours and delays are more theoretical than real. See 7B Charles Wright, Arthur Miller M. Kane, Federal Civil Practice Procedure 2d § 1803 at 508 (West 1986).
Although the Second Circuit has vested the lower courts with the option of using either the percentage or lodestar methods, every significant Southern District opinion facing the issue since Goldberger has embraced the percentage approach, without much case-specific analysis of the choice. Ling v. Cantley Sedacca, L.L.P., 2006 WL 290477 (S.D.N.Y. Feb. 8, 2006); In re Worldcom, Inc. Erisa Litigation, 2005 WL 3116188 (S.D.N.Y. Nov. 22, 2005); Hicks v. Stanley, 2005 WL 2757792 (S.D.N.Y. Oct. 24, 2005); In re WorldCom, Inc. Securities Litigation, 388 F. Supp.2d 319 (S.D.N.Y. 2005); FTR ex rel. Cel-Sci Co v. Adv. Fund II Ltd., 2005 WL 2234039 (S.D.N.Y. Sep. 14, 2005); Spann v. AOL Time Warner Inc., 2005 WL 1330937 (S.D.N.Y. Jun. 7, 2005); In re Elan Securities Litig., 385 F. Supp.2d 363 (S.D.N.Y. 2005); In re Bristol-Myers Squibb Sec. Litigation, 361 F. Supp.2d 229 (S.D.N.Y. 2005); Denney v. Jenkens Gilchrist, 230 F.R.D. 317 (S.D.N.Y. 2005); In re Alloy, Inc. Securities Litigation, 2004 WL 2750089, *2 (S.D.N.Y. Dec. 2, 2004); In re Global Crossing Securities ERISA Litigation, 225 F.R.D. 436 (S.D.N.Y. 2004); In re Interpublic Securities Litigation, 2004 WL 2397190 (S.D.N.Y. Oct. 26, 2004); In re AMF Bowling, 334 F. Supp.2d 462 (S.D.N.Y. 2004); Klein ex rel. SICOR, Inc. v. Salvi, 2004 WL 596109 (S.D.N.Y. Mar. 30, 2004); In re Independent Energy Holdings PLC, 2003 WL 22244676 (S.D.N.Y. Sep. 29, 2003); In re Lloyd's American Trust Fund Litigation, 2002 WL 31663577 (S.D.N.Y. Nov. 26, 2002); Baffa v. Donaldson Lufkin Jenrette Sec. Co., 2002 WL 1315603 (S.D.N.Y. Jun. 17, 2002); In re Dreyfus Aggressive Growth Mutual Fund Litig., 2001 WL 709262 (S.D.N.Y. Jun. 22, 2001); In re American Bank Note Holographics, Inc., 127 F. Supp.2d 418 (S.D.N.Y. Jan. 2, 2001); Varljen v. H.J. Meyers Co., Inc., 2000 WL 1683656 (S.D.N.Y. Nov. 8, 2000). Each of these decisions seems to have been guided mainly by the generic simplicity of a percentage selection, in contrast to the grueling calculation of the lodestar.
The generalized rationale that has led other courts to employ the percentage method obviously has considerable attraction here, as in any so-called "megafund" case. This process spares the Court the agonizing and time-consuming parsing of reams of attorney timesheets and hindsight recalculation of the proper amount of work and charges. The larger the case, the more complex is the lodestar documentation.
At the same time, the Court of Appeals has never made size of the case or judicial workload the sole determinants of the choice. Indeed, in contrast to other circuits which have entirely jettisoned the lodestar in favor of the percentage method, the Goldberger court very clearly retained the lodestar option, without limiting its use to cases of a certain magnitude or genre.
See, e.g., In re General Motors Corp. Liab. Litig., 55 F.3d 768, 821-22 (3d Cir. 1995), cert. denied sub nom. General Motors Corp. v. French, 516 U.S. 824 (1995); Swedish Hosp. Corp. v. Shalala, 1 F.3d 1261, 1271 (D.C. Cir. 1993); Camden I Condominium Ass'n v. Dunkle, 946 F.2d 768, 774 (11th Cir. 1991).
While the sheer magnitude of these blockbuster cases lends itself to the relative simplicity of a percentage award, such outsized awards magnify the subjectivity of the percentage assessment. Class counsel themselves admit that double-digit "benchmark" awards often found in smaller cases become atypical when applied to multibillion dollar recoveries. (Mem. in Supp. of Lead Securities Counsel's Application at 10.) To this extent, the lodestar method, rooted in market-based hourly billing rates, is a more objective measure of compensation. See City of Detroit v. Grinnell Corp., 495 F.2d 448, 470-71 (2d Cir. 1974) (only the lodestar method can "claim objectivity"); see also In re Bolar Pharmaceutical Co. Sec. Litig., 966 F.2d 731, 732 (2d Cir. 1992) (fee computed under lodestar method is "strongly presumed to be reasonable"). That advantage is partially lost, however, when a significant multiplier is interposed.
Furthermore, because Goldberger requires that the courts choose between the two alternatives, counsel do not know until the case is over which method of compensation will obtain. Accordingly, a number of traditional objections to the lodestar process — that it leads to unnecessary hours and disincentives to early settlement — are not necessarily obviated even if the Court ultimately employs a percentage formula. And the lodestar method can address those problems, by elimination of unwarranted hourly charges or, on the flip side, rewarding efficient work with a multiplier.
Application of These Principles
Ultimately, as Goldberger instructs us, the goal is a reasonable fee, which may be established by either method. Under the circumstances of this case, the percentage approach is preferable.
Of course, the generic reasons apply. The percentage method reduces, though it does not eliminate, an expensive and mind-numbing line-by-line review of counsel's bills, thereby freeing judicial resources and allowing class and claimants to be paid, and the case to be closed, more quickly. It also replicates the contingent fee system that has developed in this country for cases of significant risk or of plaintiffs who are unwilling or unable to afford the cost prior to conclusion.
Although this case involved a mix of class members, some of which were sufficiently large and sophisticated that they could have afforded an hourly fee structure, it is doubtful that any single member would have made such an investment on its own. The original retainer agreement between lead plaintiff and counsel, though subsequently modified to leave compensation entirely to the Court's discretion, was based on tiered percentages.
In addition, because class counsel here comprised 26 different law firms, a proper lodestar calculation would require the Special Master and the Court not only to scrutinize a particularly large mass of records but also to perform an allocation function as well, which can be done internally among counsel under the percentage approach. See In re Prudential Ins. Co. of America Sales Practice Litig., 148 F.3d 283, 329 n. 96 (3d Cir. 1998), cert. denied sub nom. Krell v. Prudential Ins. Co. of America, 525 U.S. 1114 (1999).
Counsel seek recovery under the percentage method. Lead plaintiff finds it acceptable as well. Although the four objectors request a cut in the fees sought, none objects to use of a percentage.
This settlement also consisted entirely of monetary compensation. It did not include coupons, injunctive relief or other non-monetary elements of ambiguous valuation, which could undermine the reliability of a percentage analysis. Compare, e.g., Shaw v. Toshiba America Information Systems, 91 F. Supp.2d 942 (E.D. Tex. 2000).
Further, the lodestar method works most efficiently where only modest adjustment is required for the time spent and the hourly billing rates. Here, where at least some potential would exist for adjustment of the appropriate time spent and the hourly rates, more than typical modification would likely be necessitated. And, as explicated below, this case involved higher-than-average risk on its merits, a scenario more suitable for a percentage determination.
Tipping the scale here is my opinion that, even under a lodestar analysis, a positive multiplier would be justified to compensate counsel for the risks they undertook and the manner in which they achieved this impressive recovery. As a multiplier injects an element of subjectivity into the fee process, there becomes less reason to reject a percentage award for lack of objectivity. As this Circuit's law requires, I have cross-checked my tentative percentage calculation against the lodestar. See Wal-Mart, 396 F.3d at 123; Goldberger, 209 F.3d at 50.
The Percentage of the Recovery
This Circuit has rejected a "benchmark" recovery percentage for class action counsel. Unlike some other circuits, which have embraced generically applicable percentages, often as high as 25 percent, see, e.g., Six Mexican Workers v. Arizona Citrus Growers, 904 F.2d 1301, 1311 (9th Cir. 1990), the Second Circuit requires a searching analysis based upon the circumstances of each case. Goldberger, 209 F.3d at 53.
The traditional factors for computing common fund attorneys fees thus remain relevant. In Goldberger, the Court defined those elements as (1) the time and labor expended by counsel; (2) the magnitude and complexities of the litigation; (3) the risk of the litigation; (4) the quality of representation; (5) the requested fee in relation to the settlement; and (6) public policy considerations. Wal-Mart, 396 F.3d at 121 ( citing Goldberger, 209 F.3d at 50). Goldberger emphasizes two other important points, namely that "moderation" is the touchstone and that the district courts should strive to replicate market compensation. 209 F.3d at 52 ("market rates, where available, are the ideal proxy for [class counsel's] compensation"). At the same time, the Second Circuit recognizes that, in this type of case, where the fee is set retrospectively without the challenge of a traditional adversary, we "cannot know precisely what fees common fund plaintiffs in an efficient market for legal services would agree to, given an understanding of the particular case and the ability to engage in collective arm's-length negotiation with counsel." Id. Accordingly, the Court of Appeals requires that a "searching assessment" be performed by the district court in each case based upon the circumstances of that case. Goldberger, 209 F.3d at 52.
This framework is similar to that employed in New York, the host state for this litigation. see Matter of Freeman, 34 N.Y.2d 1, 9, 355 N.Y.S.2d 336 (1974); Steiger v. Dweck, 305 A.D.2d 475, 476, 762 N.Y.S.2d 84 (2d Dept. 2003) ("[i]n determining the reasonableness of an application for attorneys' fees, the court should consider the time spent, the difficulties involved in the case, the nature of the services rendered, the amount in controversy, the professional standing of counsel, the fee customarily charged by other similarly situated attorneys for similar services, the contingency or certainty of compensation, and the results obtained"). See also American Bar Assn. Model R. of Professional Conduct 1.5(a).
The dilemma in applying a marketplace rule to this case, and others like it, is that class actions typically do not involve the same retention process and client bargaining opportunities prevalent in other species of commercial cases. Most often, class counsel fees are set retrospectively by the courts.
Not surprisingly, there is no ready database to which the Court can turn to ascertain the prevailing market arrangement for the type of service performed by class counsel on a percentage basis. One vestigal utility of the lodestar approach is its tailored array of hourly billing rates which presumably reflect the market value of each attorney's service. Because, by contrast, all percentage awards are judicially determined, there is generally no arm's length bargaining at the inception of a class action that would yield a market-based formula.
Nevertheless, we are not bereft of market precedent. Data are available from studies of cases in which "auctions" were conducted for the selection of class counsel. See Laural L. Looper Marie Leary, Auctioning the Role of Class Counsel in Class Action Cases: A Descriptive Study (Fed. Jud. Ctr. 2001), reprinted at 209 F.R.D. 519 (2002). The study concluded that, in such cases, the fee percentages tended to be lower than those awarded retrospectively by the courts:
Attorneys' fees were generally less than the reported percentages in other class actions in the respective circuits. The majority of fee awards was less than 9% (may or may not include expenses) of the total recovery and ranged from a low of approximately 5% in In re Auction Houses to a high of 22.5% in In re Oracle.
Class counsel herein, by contrast, cite typical percentages that range between 5.5 and 25 percent, a range higher than that found in the auction study. (Mem. in Supp. of Lead Securities Counsel's Application at 10.)
Although the auction approach has fallen into disfavor, e.g., In re Cendant Corp. Litig., 264 F.3d 201 (3d Cir. 2001), cert. denied sub nom. Mark v. Calif. Pub. Employees' Retirement System, 535 U.S. 929 (2002), there is no reason to doubt the data generated in those cases. While it is true that fee auctions drew theoretical criticism for downplaying the quality element in counsel selection, counsel ultimately selected in various cases, including Cendant, were renowned practitioners in the field.
In WorldCom, a case bearing parallels to the present matter, at least in terms of its size and nature and the experience and ability of the attorneys, counsel for the class entered into a negotiated retainer agreement with the lead plaintiff there, the New York State Common Retirement Fund. Class counsel's time records in the present case show that they closely followed WorldCom as a model for their services. The WorldCom retainer agreement contained a tiered percentage structure, depending upon the size of the recovery and the stage at which it was achieved:
Total Recovery Commencement of From completion 15 days before discovery of fact the last date through the discovery to set for trial completion of final 16 days through end of fact discovery before the last case (including date set for all appeals) trial
The complete retainer agreement was filed with the WorldCom court as part of counsel's fee application and is available at http://www.worldcomlitigation.com/courtdox/retainer.pdf.
Tier 1: 12% 13% 14% $0-$100 million of recovery of recovery of recovery
Tier II: Plus Plus Plus $100-$250 11% 12% 13% million of any amount in of any amount in of any amount in this range this this range range
Tier III: Plus Plus Plus $250-$500 9% 11% 12.5% million of any amount in of any amount in of any amount in this range this this range range
Tier IV: Plus Plus Plus $500-$1 billion 5.5% 6.5% 7.5% of any amount in of any amount in of any amount in this range this range this range
Tier V: Plus Plus Plus Over $1 billion 4.0% of any 4.75% of any 5.5% amount amount of any amount in in excess of excess of excess of $1,000,000,000 $1,000,000,000 $1,000,000,000
Judge Cote ultimately approved counsel's request for fees totaling $336.1 million, or 5.5 percent of the total recovery. On the grid, a figure of four percent applied to the settlement portion above $1 billion.
A negotiated retainer arrangement also existed in Cendant Corp. Litig., 264 F.3d 201. There, two leading class action firms agreed to a fee scale, for settlement during discovery, of 17.5 percent for a recovery up to $100 million; 10 percent between $100 and $300 million; 7.5 percent between $300 and $500 million; and 5 percent above $500 million.
The Third Circuit remanded the case because, inter alia, counsel had not obtained lead plaintiff's prior review of their fee application, as required by the retainer letter. But this does not affect the vitality of the letter as a heuristic example of market-based bargaining.
Although two samples are no talismans, and the Second Circuit has frowned upon benchmarks in any event, the WorldCom and Cendant retainer letters furnish informative market-based guidance. And they do so not only with actual bargained-for percentages, but also with sliding scales based on both the magnitude and juncture of the recovery.
To be sure, there have been other negotiated fee arrangements with different, and potentially more lucrative, fee structures. See John C. Coffee, Jr., Litigation Governance: A Gentle Critique of the Third Circuit Task Force Rept., 74 Temple L. Rev. 805, 809 n. 16 (2001) (citing two cases in which fee percentages increased with the size of the award). Neither of those cases appears comparable to the instant matter. In fact, in one of those cases, which ultimately yielded a $75 million class recovery, counsel's fee contract provided for the greater of tiered percentages or three times their lodestar.
The WorldCom and Cendant retainer letters are consistent with a number of judicial decisions which reduce the fee percentages as the amounts of recovery increase. See, e.g., Prudential, 148 F.3d at 340. The agreements also incorporated negotiated percentages noticeably lower than those which class counsel herein cite as prevailing in megafund cases. These agreements, and the Federal Judicial Center study, provide evidence that the percentages awarded retrospectively by various courts tend to exceed those that would be generated through give-and-take bargaining in the commercial marketplace.
The original retainer agreement in the present case deviated from typical practice in at least two respects: the percentages were higher and they increased, rather than decreased, with the size of the award. Some experts have promoted this concept to provide incentives for counsel to pursue cases where diminishing returns might otherwise cause premature settlement. See John C. Coffee, Jr., 74 Temple L. Rev. at 809.
That the case-specific, "moderation" — based approach enunciated in Goldberger has served its purpose is borne out by statistics in the decision's wake. The Second Circuit has maintained its historical position as the number two host of securities class action settlements. The fee award structure has continued to attract high quality counsel whose efforts have generated ever-increasing settlement numbers. The bidding lawyers in WorldCom, Cendant and various fee auctions, though agreeing to single-digit percentages, included preeminent class action practitioners. This vindicates the wry observation of one district judge who, after slashing a fee request by nearly two-thirds, noted that "[i]f it amounts to punishment, I am confident there will be many attempts to self-inflict similar punishment in future cases." In re Visa Check/Mastermoney Antitrust Litig., 297 F. Supp.2d 503, 525 (E.D.N.Y. 2003), aff'd sub nom. Wal-Mart Stores, Inc. v. Visa U.S.A. Inc., 396 F.3d 96 (2d Cir. 2005), cert. denied sub nom. Leonardo's Pizza by the Slice, Inc. v. Wal-Mart Stores, Inc., 544 U.S. 1044 (2005).
In 2004, for example, of 117 class action settlements in the federal system nationwide, 22 were venued in the Second Circuit. Only the Ninth Circuit, with 29, hosted more. Laura E. Simmons Ellen Ryan, Post-Reform Act Securities Settlements (Cornerstone Research 2005) at 16. The median settlement increased from $6.1 million in the prior cumulative period to $8.2 million in 2004. Id.
Chosen co-counsel in both WorldCom and Cendant were Bernstein Litowitz Berger Grossmann LLP and Barrack, Rodos Bacine.
This proposition is corroborated not only in circuit-wide statistics, but also in this very action. No fewer than 18 separate law firms sought the role of lead or liaison counsel. The competitors for the lead counsel role included many prominent names in class action service, such as Milberg Weiss Bershad Hynes Lerach, Bernstein Litowitz Berger Grossman, Berger Montague and Cohen Milstein Hausfeld Toll. Twenty-six separate law firms ultimately enlisted in the effort and now seek compensation from the common fund. The competition became so intense that chosen counsel reviewed the prospect of disqualifying a leading competitor, Milberg Weiss, and much later filed a motion against one of Milberg's successor firms to restrain alleged interference. In foresight or in hindsight, record-setting compensation was not needed to lure top lawyers to the representation of this class.
The Goldberger Factors
(1) The Time And Labor Expended By Counsel.
Counsel report spending in the aggregate 135,185.55 hours on this litigation. This equated to fees of $46,861,731.25. More than half of the hours, 70,036.50, were dedicated by lead counsel, with a total time value of $23,989,373.75.
This case was necessarily time-intensive. Lead counsel alone claims to have utilized 55 lawyers. That firm was assisted by 25 other firms, which are part of this application for compensation. At least 15.5 million documents were produced and reviewed. Motions were filed for dismissal and for summary judgment. Some of the issues were familiar ones, such as the standards for alleging securities fraud and accounting irregularities. Others were less common, including some of the devices allegedly used by the issuer to inflate its reported performance. Loss causation was a difficult and somewhat unsettled issue.
Although it is does not appear that any depositions were actually conducted, preparation was underway for hundreds of examinations, some covering complicated subject matter and sophisticated witnesses. Counsel interacted frequently with their opposition, their experts and Special Master Wachter.
Accordingly, it is unsurprising that this case would reasonably consume tens of thousands of hours. Large cases inherently tend to produce at least some inefficiencies, and that was the case here. The record abounds with conferences among large numbers of lawyers, attendance by multiple lawyers at various proceedings and vast amounts of time spent on particular tasks. I hasten to add that I make no finding that any of the firms deliberately engaged in overstaffing or lodestar overfeeding. It is easier in hindsight to suggest greater efficiencies than it is to manage complex litigation during the tumult of approaching deadlines. But efficiency is an important goal of fee assessment, and a consideration here.
The issues of time and labor are more particularly addressed in the lodestar cross-check (see pp. 45-57, infra). Suffice it to say, the lodestar cross-check suggests that the labor time value reported by class counsel should be reduced by approximately 14.7 percent for unnecessary or excessive time, imprecise billing and work on the fee applications.
At the same time, the percentage of the recovery also will reflect the pressures under which counsel labored, and the long days and travel they frequently had to endure. In sum, this was a demanding case which necessitated great time and labor, though not as much as claimed in counsel's lodestar.
(2) The Magnitude and Complexities of the Litigation.
The Court has already recognized the magnitude and complexities of this litigation. "The breadth of resources dedicated to the prosecution of this lawsuit reflects the complexity of the issues involved . . ." Apr. 6 Opinion Order at 21. In addition to legal developments occurring during the pendency of this action, e.g., Dura Pharmaecuticals, Inc. v. Broudo, 125 S. Ct. 1627 (2005); Lentell v. Merrill Lynch Co., 396 F.3d 161 (2d Cir. 2005), cert. denied, 126 S. Ct. 421 (2005), the arcanum of the challenged accounting transactions would test the mettle of any veteran securities lawyer. The task of reviewing, evaluating and utilizing the millions of documents produced in this case was daunting. The experts' submissions in connection with summary judgment motions relied on linear regression analyses and other esoterica that would challenge the interdisciplinary skills of the most mathematically proficient of lawyers.
Counsel point to the magnitude of the recovery as another justification for their fee. They correctly observe that this is one of the largest securities settlements in history. Concomitantly, their fee request too is of epic dimension. Counsel also reported to me their rejection of significant prior settlement offers, tenaciously holding out for the $2.5 billion ultimately achieved.
But the size of a recovery does not necessarily correlate with the quality of the representation. After all, "a large settlement can as much reflect the number of potential class members or the scope of the defendant's past acts as it can indicate the prestige, skill, and vigor of the class's counsel." City of Detroit v. Grinnell Corp., 560 F.2d 1093, 1099 (2d Cir. 1977).
The amount of this settlement is in no insignificant measure attributable to the size of the class. This is a particular circumstance that the courts need to filter to avoid excessive fee awards. Id.
In this case, the damages were magnified by the sheer amount of securities outstanding and the resulting number of trades. A total of 830,000 claims were filed. Other than some additional cost of identifying and notifying class members, which was largely the function of claims administrators whose charges will be fixed within the disbursements, it is difficult to discern much additional complexity of the work or risk stemming from the class size. By the same token, the size of the class and the attendant risk to the defendants gave the defendants inherent incentive to settle the action.
In sum, the issues in the case were significant and complex. The ultimate size of the recovery, however, was also influenced by the wide dispersion and popularity of these securities.
(3) The Risk of the Litigation.
Risk has been described by some authorities as the preeminent factor in determining either a percentage or a lodestar multiplier. See Agent Orange, 818 F.2d at 236. That is because the Court's major focus in fashioning a fee award is encouraging the bar to undertake future risks for the public good in tomorrow's cases.
Counsel depict this case as high risk because of the many challenges faced, especially proving loss causation, scienter and damages. (Decl. of Samuel D. Heins, ¶ 14, Dec. 1, 2005.) Counsel note that prosecuting the case "on an entirely contingent basis was, by any measure, an extremely risky proposition. (Heins Decl., ¶ 3.)
Taking a case on a contingent basis, however, inheres in virtually any class action.
Counsel's analysis of "risk" is accurate as far as it goes. The legal principles of loss causation are still evolving and were unsettled in certain respects at the commencement and during the progress of this litigation. See, e.g., Dura Pharmaceuticals, 125 S. Ct. 1627. Counsel also faced the challenge of distilling the impact of defendants' accounting irregularities from the overlapping national recession and specific downturns in the Internet and media markets. In this respect, the case against AOL Time Warner carried more risk than Goldberger and greater risk than that generally inherent in class actions.
But "risk" is not as monocentric as counsel assume. In addition to the possibility of losing the action entirely, attorneys often face the risk of nonpayment, especially in large class actions that could drive defendants into bankruptcy or otherwise make a judgment uncollectible. In contrast to the fate suffered by WorldCom, Enron or Integrated Resources and Drexel Burnham Lambert in Goldberger, the $2.65 billion settlement paid by defendants herein proved more of a glancing blow than a knockout punch. The company's merger era market capitalization of $193 billion was more than 70 times the amount paid in settlement. The prime corporate defendant remained solvent throughout. Co-defendant Ernst Young was and remains one of the largest accounting firms in the world. There is no suggestion that any of the individual defendants, a virtual Who's Who of civic and financial leaders, were impecunious. Thus, while there was an elevated risk of loss in this case, the prospects for collection of a judgment or settlement were promising from the start.
Risks also run both ways. Even in a case with viable defenses, a bulging ad damnum clause poses significant risks to a defendant, ranging from a catastrophic judgment to embarrassments causing competitive disadvantage or decline in stock price. Although defendants' survival was not necessarily threatened here, a verdict higher than the settlement amount was a viable prospect, carrying with it a powerful incentive toward settlement. Cases of this magnitude usually end in settlement because neither side wants to assume the ultimate risk of a jury verdict or summary judgment. See Janet Cooper Alexander, Do the Merits Matter? A Study of Settlements in Securities Class Actions, 43 Stanford L. Rev. 497, 578 (1991). Given the stakes here, and the appointment of a skilled Special Master who shepherded the parties through the settlement process, the class faced not only a dogged defense team, but, conversely, defendants with reasons to settle.
Ultimately, the risk to be measured in a regime of moderation is the risk that the class will not prevail, as opposed to the risk that counsel will be relegated to an hourly fee or a percentage below perceived benchmarks. This Report Recommendation evaluates incentive as a separate issue. Appropriately measured, class counsel here faced higher-than-average risk on the merits, typical risk of facing trial and lower-than-average risk of non-collection.
(4) The Quality of Representation.
Here, too, the Court has already provided the answer, even if the results do not already speak for themselves. The Court's opinions herein have repeatedly commended the skill and performance of class counsel. The Court has cited "exceptional lawyering in this case." Apr. 6, 2006 Opinion Order at 18. Satisfaction of the class is reflected in the paucity of objections to the final settlement.
Counsel defeated a volley of motions to dismiss, thwarting quality challenges from many of the nation's heralded securities defense law firms. They adroitly navigated some 15.5 million documents. They marshaled the intricate facts of this case. Their efforts ultimately helped persuade a media giant and a "Big Four" accounting firm to part with $2.5 billion in the negative glare of a settlement.
The quality of the service is unassailable. That being said, quality performance does not always justify an outlying fee. Counsel were selected herein on the basis of their quality, formidability and track record of superior results. Their hourly billing rates, which stand well above the median, reflect an inherent and anticipated quality factor. See Alba Conte and Herbert B. Newberg, Newberg on Class Actions § 14:6 (Thomson-West, 4th ed. 2002) ("market rate for legal fees depends in part on the risk of nonpayment a firm agrees to bear, in part on the quality of its performance, in part on the amount of work necessary to resolve the litigation, and in part on the stakes of the case").
Counsel also benefited to a degree from work performed by others. While counsel transformed the facts and arguments into successful advocacy, the irregular transactions were first exposed by reports in the media, and the ensuing accounting analysis was largely the work of expert consultants.
Parallel government investigations were also undertaken, though the parties — even at the settlement hearing — disputed their relative significance. AOL was cited by the SEC for limited violations in 2000 and entered into a deferred prosecution agreement with the United States Attorney for the Eastern District of Virginia on December 14, 2004. Several peripheral figures were indicted, various AOL executives implicated in the class action lost their jobs and the company restated its accounting during the pendency of the action.
(5) The Requested Fee in Relation to the Settlement.
As noted, these blockbuster cases generally entail lower fee percentages than those seen in smaller cases, to avoid conferring a windfall upon the attorneys merely because of the size of the class or the recovery. See Prudential, 148 F.3d at 340. With that in mind, this recommendation must obey Goldberger's command for "jealous regard to the rights of those who are interested in the fund." Goldberger, 209 F.3d at 53.
A single-digit percentage award will fall within the range allowed in similarly sized settlements and is commensurate with competitively-based fee arrangements. See Looper Leary, Auctioning the Role of Class Counsel in Class Action Cases, 209 F.R.D. at 531, 595-96. As a result, the overwhelming majority of the funds recovered will reach the hands of the class members.
(6) Public Policy Considerations.
The competing poles of public policy consideration are the encouragement of counsel to accept worthy engagements and the discouragement of excessive lawyer compensation. These two objectives can be reconciled.
This case had ostensible merit. Unlike other cases where the class award consisted significantly of injunctive relief, stock, price rollbacks or hard-to-value coupons, the class will receive its entire multibillion dollar award in cash. Thus, it is reasonable and proper to authorize a nine-figure fee for counsel in this case without offending policies against excessive compensation.
Awards of this nature should prove more than sufficient to galvanize the best of the class action bar into action in future cases. By the same token, these awards should have a deterrent effect on errant corporate leaders, signaling that counsel will be well paid for their efforts to redress securities infractions.
This Report recommends a percentage with which few should quarrel. Even one of the few objectors to the fee application previously acknowledged that counsel should receive compensation that could be as high as $125 million, an amount not markedly less than the proposed award.
Fee Recommendation
A fee of $147,500,000 will strike a fair balance between the interests of the class and its counsel. This fee is derived through a simplified two-tier structure based loosely on the concept initially agreed by the parties. A percentage of five percent is applied to the first $1 billion of recovery and a percentage of 6.5 percent is applied to the $1.5 billion balance. This yields an overall percentage of 5.9 percent.This large dollar award joins the ranks of the largest class action fee awards. It represents a 3.15 multiplier upon the already large $46,861,731.25 lodestar. And, in contrast to the requested fee of $175 million, this reduction of nearly 16 percent leaves an additional $27.5 million for distribution to the class. The proposed percentage is applied to the gross recovery amount, not the net after disbursements. See Newberg, § 14:6 ("most courts seem to apply the percentage of recovery methodology to the gross settlement fund"). The percentage is not applied to the $150 million contribution from the government settlement, as class counsel did not generate that recovery.
Several conflating methodologies support this recommendation. First, it closely approximates the percentages agreed after arms-length bargaining in the WorldCom and Cendant retainer agreements, and falls within the range cited in the 2001 Federal Judicial Center auction study. See Looper Leary, Auctioning the Role of Class Counsel in Class Action Cases. Application of the WorldCom formula here would result in a fee award of $138.5 million. This amount is more consistent with my recommendation than with counsel's request of $175 million. The Cendant formula would yield $152.5 million, nearly identical to this recommendation.
That study did not include the numerous ten-figure settlements approved in the ensuing five years, and included cases from other Circuits that have "eased into a practice of `systematically' awarding fees in the 25% range, `regardless of type of case, benefits to the class, numbers of hours billed, size of fund, size of plaintiff class, or any other relevant factor.'" Cf. Goldberger, supra at 51 ( quoting Court Awarded Attorney Fees, 108 F.R.D. at 274 n. 32). Given the now generally accepted principle that fee percentages should decline as the megafund recovery increases, the recommended award in this case is consistent with the range found in the study.
Second, it results in hourly billing rates ranging between $677.25 and $2,047.50. One leading publication reported with some fanfare that the $1,000 hourly rate barrier was recently broken by a prominent lawyer, a former United States Attorney General. Hourly Billing Rates Continue to Rise, National L.J. (Dec. 12, 2005). The recommended award herein effectively places 161 class attorneys (including several associates) in that stratospheric company. Indeed, it yields an average recovery — among partners and associates — of $1,091 per hour. If, by contrast, class counsel's fee request were approved in full, they would reap an award averaging $1,294.55 per hour. The latter amount, which has no parallel in the hourly marketplace, is more difficult to reconcile with Goldberger's "preference for moderation." 209 F.3d at 57.
This figure is derived by multiplying the lawyer's requested hourly rate by the effective multiplier of 3.15 for the unadjusted lodestar. For example, Samuel D. Heins' effective rate would be $2,047.50.
Inclusion of paralegals would slightly reduce this number.
As a third and related matter, the recommendation comports with a lodestar cross-check. See pp. 38-56, infra. The award, based on counsel's unadjusted lodestar, yields a 3.15 multiple. Based upon an adjusted lodestar, the multiple becomes 3.69 (though higher if adjustments were made to counsel's billing rates). The reduction of nearly 16 percent closely approximates the adjustments made to counsel's lodestar of approximately 15 percent.
Fourth, it conforms to legal and ethical requirements that a fee be "reasonable," using prevailing standards in the legal community as a key measure. As the lodestar analysis reveals (pp. 51-57, infra), counsel's standard hourly billing rates already fall at the high end of the spectrum. This is not a case so unusual in risk as to comfortably tolerate an across-the-board $1,300 hourly fee.
A less conventional, but nonetheless informative check confirms this proposition. One authoritative survey reports average annual revenue-per-lawyer nationally at $382,169. Altman, Weil, Inc., Survey of Law Firm Economics (2005) at 32. Another source, focusing on revenue-per-lawyer statistics of the nation's 100 leading law firms, reports a median result of $719,000. See The AmLaw 100, American Lawyer (May 1, 2006). Applying the fee award here proportionately to lead counsel (the only firm which provided utilization rate information in its submission) would generate an award of $67,850,000 for the approximate three year period between filing of the action and settlement, or $22,616, 666.67 per year. Allocated among the approximately 50 lawyers that the firm reports utilizing in this period, and adjusted for the 53 percent of total firm time devoted to this matter (Decl. of Samuel D. Heins, ¶ 103, Dec. 1, 2005), the revenue per lawyer would total $853,459.12 — a figure twice the national median and beyond the midpoint of the country's top firms. This places lead counsel in the company of the country's most financially successful law firms, and well ahead of its regional peers, reinforcing the generosity of the proposed fee.
Lead counsel report utilizing as many as 55 lawyers, but only 50 devoted meaningful time.
If anything, this unrefined test likely underestimates the financial benefit of this award to the law firm. For example, the average revenue-per-lawyer, as sampled in lead counsel's region, is $279,448. Altman Weil at 32. In addition, not all 50 lawyers worked on the case in any given year. It should be emphasized again, in any event, that this inexact computation was not a primary factor in the fee recommendation, but, rather, one of several checks on its reasonableness.
Counsel understandably focus on the particular Goldberger factors that particularly favor an award of the level they seek, such as the magnitude and complexities of the litigation; the risk of the litigation; the quality of representation; and the requested fee in relation to the settlement. Counsel should also be commended for proposing a single-digit percentage fee, consistent with recent downward trends. Contrast Visa Check/Mastermoney Antitrust Litig., 297 F. Supp.2d 503 (finding counsel's 18 percent fee request "absurd" and "wholly out of character for a group of counsel whose commitment to the [class] until now, been admirable and unflagging").
At the same time, counsel overlook or downplay several key considerations imposed by the Court of Appeals: the discouragement of benchmarks; a focus not on risks inherent in most class actions, but, rather, on risks apposite to the case at hand; rejection of automatic correlation of size with risk and effort; and, most critically, an overshadowing "preference for moderation." Goldberger, 209 F.3d at 57.
I have carefully considered the cases cited by counsel, a number of which have generated higher percentage recoveries for class lawyers. It would be facile, however, to determine a fee by averaging the awards in similarly sized cases. The universe of such cases is still too small to serve as a reliable resource. See Newberg, § 14:6 ("there are fewer such cases to study"). The overall data do not sufficiently weight the nascent trend toward markedly smaller percentage settlements. See Wayne Schneider, Courts Don't Have to Award Excessive Fees to Incentivize Class Counsel in Federal Securities Class Actions, 20 NAPPA Report 8, 8-9 (May 2006).
More importantly, this circuit has been "disturbed by the essential notion of a benchmark." Goldberger, 209 F.3d at 51. Goldberger's search for an "efficient market" does not necessarily lead to an averaging of awards made by various courts in comparably sized cases. Virtually all of the megafund fee awards cited by counsel were fixed retrospectively in quasi-monopolistic fashion, not by the robust competition that normally defines an efficient market. Each case also carries its own risks and rewards. In a number of comparably sized cases, the primary malefactors were bankrupt. In some respects, on the other hand, cases such as Enron and WorldCom, built on allegations of accounting gimmickry, helped pave the way for the result here. Thus, the compendium of ex post fee awards submitted by counsel is utilized in this Report Recommendation, like the lodestar, as a cross-check, rather than the foundation, of the proposed award.
Counsel rely heavily upon the size of their recovery, in seeking a very large fee and in citing these comparably sized cases as precedent. Size of the recovery is not necessarily, however, a reliable measure of the risk undertaken, the quality of the legal work or the effort involved. The average settlement has increased markedly over the years and records are continuously broken. Burgeoning awards may be more attributable to external factors than to the work of counsel in a particular case. See, e.g., PricewaterhouseCoopers Securities Litig. Study (2005) at 3 (suggesting that the clout of lead plaintiffs under the PLSRA, steep declines in the stock prices of multibillion dollar corporations and growing third-party liability have converged to generate larger settlements).
And "it is not ten times as difficult to prepare, and try or settle a 10 million dollar case as it is to try a 1 million dollar case." Goldberger at 52, quoting Union Carbide Consumer Prod. Bus. Securities Litig., 724 F. Supp. 160, 167-68 (S.D.N.Y. 1989). "The huge fees in a huge case might be less a function of the amount or quality of the attorneys' work, or even of the risk undertaken, and more simply a function of the fact that the lawyers managed to find and bring a case with huge damages." Task Force on Contingent Fees, Tort Trial Insurance Practice Section of the American Bar Association, Report on Contingent Fees in Class Action Litigation, 25 Rev. Litig. 459, 470 (2006). This was, to a degree, true in this case. The size of the class and the market capitalization of the primary defendant conflated to magnify the amounts at stake and, concomitantly, the size of the settlement. While counsel can proudly and appropriately take their share of the credit, the magnitude is also a function of the underlying circumstances.
There is an emerging consensus in favor of tapered percentages for awards in megafund cases, prompted by record-breaking recoveries carrying the potential for windfall fees. See, e.g., Wal-Mart, 396 F.3d at 123 ("the sheer size of the instant fund makes a smaller percentage appropriate"); Prudential Insurance Co. of America Sales Practices Litig., 148 F.3d at 340; In re Smithkline Beckman Securities Litig., 751 F. Supp. 525, 534 (E.D. Pa. 1990); but see In re Linerboard Antitrust Litigation, No. MDL 1261, 2004 WL 1221350 (E.D. Pa. Jun. 2, 2004). Class counsel here candidly acknowledge this trend, and proffer their seven percent request as consistent with the practice. See also Schneider, 20 NAPPA Report at 9. Counsel have laudably exercised some self-moderation in limiting their request to a single-digit fee. But see In re Quintus Securities Litig., 201 F.R.D. 475, 490 (N.D. Cal. 2001), rev'd on other grounds, 306 F.3d 726 (2002) ("[b]enevolence of counsel is no substitute for hard bargaining").
At the same time, there is continued debate over how, if at all, to stratify percentages within a given case. See John C. Coffee, Jr., 74 Temple L. Rev. at 809 (suggesting that percentages should increase at certain milestones to discourage class counsel from settling at points where counsel's marginal return would begin to recede). Professor Coffee has urged that this case in particular qualifies for his approach:
[T]he declining percentage method `can create an incentive to settle quickly and cheaply when the returns to effort are highest' and fails to reward counsel for `investing additional time and maximizing plaintifs' recovery.' In re Auction Houses Antitrust Litig., 197 F.R.D. 71, 80 (S.D.N.Y. 2000). This is not just an abstract point, because the facts of this case show plaintiffs' attorneys to have held out for a higher recovery in the face of high risk — at precisely the point where a `declining percentage' formula would most prejudice them.
(Decl. of John C. Coffee, Jr., ¶ 42, Nov. 30, 2005.) Counsel's rejection of multiple offers in excess of $1 billion reflects merit in this approach in this case. See also Linerboard Antitrust Litigation, 2004 WL 1221350.
An award as high as seven percent, however, would be excessive here. Although there was some heightened risk on the case's merits, the risk of non-collection was low. In contrast to the seven years devoted by counsel in Wal-Mart, class counsel spent three meaningful years herein. And although there clearly was not the level of spadework present in Goldberger in terms of prior convictions and governmental investigations, there were parallel SEC and Justice Department proceedings which had some impact on the course of events. See Feb. 22, 2006 Transcript at 9 (defendants contending that the government investigations were "quite far reaching and went on for quite a long time").
Counsel cite expert predictions early in the case of a settlement in the range of $1 billion. Their ultimate achievement of a $2.5 billion settlement, they urge, justifies recognition for exceeded expectations and tenacity.
This contention is, however, a double-edged sword. True enough, if the $1 billion estimates are reliable, we have further evidence of a noteworthy result. At the same time, these estimates weigh against a significant bonus, at least for the first tranche of the recovery. Multipliers and high percentage awards are intended, in the main, to account for risk and to reward exceptional accomplishments. At least to the extent of the first $1 billion, counsel ask a sizable premium for what they themselves describe as an expected result.
These $1 billion estimates are given some credence, as they have been embraced by class counsel and Professor Coffee, and, based on the record and the frequency of settlements in securities class actions, appear reasonable. But because they are neither scientific nor tested by the adversarial process, they are accorded less than full weight. See Goldberger, 209 F.3d at 56 (devaluing counsel's contention, based on expert's declaration, that they recovered 90 percent of the damages claimed).
This factor, however, supports a higher percentage for the recovery portion above $1 billion, rather than the declining percentages seen in the WorldCom and Cendant retainer agreements. In those agreements, higher percentages were applied to the first dollars of recovery.
Lead plaintiff's comments are also significant. The Minnesota State Board of Investment is a large, sophisticated entity generally represented by the Attorney General of Minnesota. It suffered the largest loss and is presumably the claimant most affected by the fees payable.
Recent securities law reforms contemplate that the lead plaintiff will exercise control over class counsel and their fees, as happened in WorldCom. See Elliott J. Weiss John S. Beckerman, Let the Money Do the Monitoring: How Institutional Investors Can Reduce Agency Costs in Securities Class Actions, 104 Yale L.J. 2053, 2106-07 (1995); see also Sherleigh Assocs. LLC v. Windmere-Durable Holdings, Inc., 184 F.R.D. 688, 694-95 (S.D. Fla. 1999) (recognizing that selection of appropriate lead counsel helps to assure the reasonableness of fees and expenses); In re Party City Sec. Litig., 189 F.R.D. 91, 116 (D.N.J. 1999) (recognizing that counsel fees should be "the result of hard bargaining"); In re Network Assocs., Inc. Sec. Litig., 76 F. Supp.2d 1017, 1033 (N.D. Cal. 1999) (holding that "lead plaintiff owes a fiduciary duty to obtain the highest quality legal representation at the lowest price").
Lead plaintiff's embrace of a reduction of the fees requested, as postulated in the draft, reinforces the conclusion that counsel's request should be trimmed. While plaintiff's approval of a reduced fee derives inherently from self-interest in maximizing its present recovery, an organization of its sophistication, with widespread stockholdings, would have a concurrent interest in promoting incentives to entice counsel to accept future worthy cases.
The original retainer agreement, though invalidated by both signatories, would have yielded a payment here of at least $425 million — a plainly excessive amount unlikely to pass judicial muster. While the amounts specified in that agreement are therefore not instrumental in my recommendation, the agreement conceptually recognized the challenges of the case and, in conjunction with counsel's tenacious rejection of attractive earlier settlement offers, justifies upward reconsideration of the draft recommendation.
This recommended adjustment to counsel's fee request should not detract from recognition of the fine work that counsel have accomplished and their fidelity to the class in the face of certain risks. Counsel in this case deserve handsome compensation and it is the intention of this Report Recommendation to provide it, within the bounds of moderation and a competitive marketplace.
The Lodestar Cross-Check
The lodestar analysis has two fundamental components: examining the attorney time records for reasonableness and necessity, and establishing appropriate hourly billing rates. Goldberger, 209 F.3d at 47.
The Time and Services Billed
Dissecting even a portion of 135,000 hours of itemized billings is at best an imperfect exercise. A review of counsel's hourly-based recordings, however, suggests that some adjustment would be warranted. This adjustment stems primarily from excessive or unnecessary time spent on certain tasks, full billing of out-of-town travel time, imprecise time recording, charges for administrative services or the lack of correlation in certain billing records. There is no evidence of fraudulent billing practices or purposeful inflation of the lodestar. Similarly, some excesses are inevitable in a case of this nature and magnitude and, though subject to adjustment, should not reflect negatively on the loyalty or integrity of counsel.
In evaluating the lodestar, I have reviewed the summaries provided by all counsel of their time and expenses; undertaken a cursory reading of counsel's actual time records; sampled five months for a full lodestar analysis; reviewed the case file at the courthouse; and focused on certain tasks which consumed significant hours. I then reduced the lodestar by extrapolating these samplings where appropriate, and accounting for other questionable billing, taking care not to double count adjustments. For example, where the hourly rates were adjusted downward, reductions for excessive time were computed at the lower hourly rates.
Sampling and extrapolation is an acceptable method of assessing lodestar. See In re Continental Illinois Securities Litig., 962 F.2d 566, 572-73 (7th Cir. 1992); see also New York Ass'n for Retarded Children, Inc. v. Carey, 711 F.2d 1136, 1146 (2d Cir. 1983) (because "it is unrealistic to expect a trial judge to evaluate and rule on every entry in an application . . . courts have endorsed percentage cuts as a practical means of trimming fat from a fee application"). This practice should apply with even more efficiency where, as here, the lodestar is used as a cross-check rather than the primary calculus.
In this case, I examined five months of lead counsel's services, chosen at random from those which involved significant activity: February, April and October of 2003, July of 2004 and April of 2005. I also gave particular scrutiny to the time records from August 2005. Based on these reviews, I concluded that thousands of hours were either excessively expended or insufficiently documented, amounting to $2,487,282 in lodestar.
These extrapolations are justified through a perusal, in less detail, of the work performed outside the months in the sampling, and a survey of the time records produced by the other law firms. Travel time continued to be billed at its full scope and rate. Numerous activities were conducted by multiple lawyers. Counsel's billing records for these other months reflect, for instance:
• Reviews by multiple attorneys of the same books and articles regarding the corporate merger.
• Extensive time on various amendments to the Complaint, including at least 1,339 hours spent on the first amendment alone, for which I applied a reduction of $130,000.
• Considerable time spent on subleasing office space to accommodate the expansion of the lead firm for this matter.
• Attendance at seminars with generic utility.
• Administration of litigation funds.
• Work related to counsel's fee applications.
Some of the work was duplicative or excessive, whereas other services, such as those underlying the fee applications, are not compensable at all under controlling law. See City of Detroit v. Grinnell Corp., 560 F.2d 1093; Washington Public Power Supply System, 19 F.3d 1291.
Certain disallowances consisted of conferences between or among lawyers where one participant's time was recorded but not another's, or where excessive numbers of personnel were assigned to a particular task. If the former problem stems from inadequate recordkeeping, the remedy of disallowing the recorded time may seem harsh. But the law on this subject is clear: a lawyer applying for a fee bears the burden of recording all corresponding entries. See, e.g., Carrero v. New York City Housing Authority, 750 F. Supp. 660 (S.D.N.Y. 1990).
Travel time is another consideration. A lawyer's reasonable travel time for client business purposes should be recoverable, as the practice in this jurisdiction is to charge at a unitary hourly rate and a lawyer not engaged in travel for a client's business could presumably otherwise be representing another client at his or her hourly billing rate. Once again, however, as with any attorney's time, the key is "reasonableness."
Applying this standard to the present case, long distance travel should not be viewed identically with local travel to Foley Square or opposing counsel's office in Manhattan. This is not to suggest that long distance travel should not be reimbursed at all. A long distance trip, however, affords opportunities to perform work for the present or other clients, especially with advances in technology such as laptop computers and cellular phones. Travel time not productively spent should be adjusted.
For long-distance travel in which the time records contain no evidence of other work performed en route, I applied a reduction of 50 percent for billable hours, consistent with the methodology adopted by the district court and affirmed by the Court of Appeals in Agent Orange, 818 F.2d at 230.
Certain other time entries bear telltale signs of rounding and estimation, rather than precise recording. A noticeable illustration is a pattern of some timekeepers of recording much of their daily time in even-hour increments, rather than a random mixture of daily hours ending in .00, .25, .50 and .75. To account for such imprecise reporting, I have reduced the total fees attributable to each of those timekeepers by 15 percent. See New York Ass'n for Retarded Children, 711 F.2d at 1146, 1147 (upholding percentage reductions and placing the burden on fee applicants to keep and produce "accurate records" of work done and time spent). It bears emphasis that this reduction is recommended due to imprecision, not evidence of fabrication.
A final issue involves the continued dedication of thousands of hours, mostly associate time, reviewing and analyzing documents after the Memorandum of Understanding for settlement of the case was executed on July 29, 2005. It is understandable that a firm would not want to furlough associates based on a Memorandum of Understanding that might never reach a formal settlement stage. On the other hand, few private clients paying hourly fees would absorb the cost of such unabated work over a two-month period. I reduced lead counsel's lodestar by a total of 4,998.25 associate hours to account for this. To the extent that the continued work may be attributable to the ongoing ERISA litigation, compensation for these hours may be sought in a subsequent fee application.
The reductions computed for lead counsel totaled $2,487,282. As noted, most of these problems resurfaced in a perusal of the records of the 25 assisting law firms. These other lawyers recorded time for certain non-compensable or non-essential work, in numerous cases engaged in rounded timekeeping and continued work largely unabated for some time after the Memorandum of Understanding was signed. The magnitudes were similar. On the other hand, these other firms generally did not bill for excessive time on the Amended Complaint or the fee application. Thus, the detailed reductions computed for lead counsel are extrapolated to cover other counsel's lodestar as well, excepting the work on the Amended Complaint and the fee application.
The total consisted of $749,787 for excessive time; $130,000 for excessive hours on the Amended Complaint; $464,806 for rounded billing; $123,045 for preparing the fee application; and $1,019,643 for unnecessary services after the settlement memorandum.
Once again, there is no evidence of purposeful manipulation of the lodestar by any counsel. These are, however, charges that a gimlet-eyed fee auditor would assuredly trim in the demanding marketplace. The overall result is a reduction of 14.7 percent of the hours charged for all counsel.
Billing Rates
A proper lodestar analysis also requires the establishment of appropriate hourly billing rates. In this regard, I have reviewed authoritative sources, see Altman Weil, Survey of Law Firm Economics; Hourly Billing Rates Continue to Rise, National L.J. (Dec. 12, 2005); considered the submissions of class counsel, including rate summaries they provided for opposing counsel; and drawn upon my own experience as a practitioner. See Goldberger, 209 F.3d at 56.
Three issues arise over lead counsel's billing rates: (1) the use of current hourly rates, even for work performed in prior years; (2) the use of New York rates by lead counsel, a Minneapolis firm; and (3) the comparison of the rates charged by counsel with prevailing rates in the community. The answer to the first question is that it is acceptable to use counsel's current rates to compensate them for the lengthy delay in payment. Missouri v. Jenkins, 491 U.S. 274 (1989).
The second and third issues are not as easily resolved. The lodestar calculation requires the assessment of appropriate hourly billing rates. Although some decisions, e.g., WorldCom, have simply cross-checked the awarded percentages against counsel's own reported lodestar, without adjustment, a lodestar review normally requires an assessment of the billing rates. Compare, e.g., Ling v. Cantley Sedacca, 2006 WL 290477; see also McDonald v. Pension Plan of the NYSA-ILA Pension Trust Fund, 450 F.3d 91 (2d Cir. 2006); Goldberger, 209 F.3d at 56. Without at least some scrutiny of the lodestar hours and billing rates, the cross-check becomes less meaningful.
Considerable authority supports the charge by out-of-town counsel at rates prevailing in the host venue of the class action. See Court Awarded Attorney Fees, 108 F.R.D. at 261. The Second Circuit endorsed this practice in Polk v. New York State Dept. of Correctional Services, 722 F.2d 23, 25 (2d Cir. 1983), a statutory fee-shifting case under 42 U.S.C. § 1988. See also Arbor Hill Concerned Citizens Neighborhood Ass'n v. County of Albany, 369 F.3d 91 (2d Cir. 2004). At the same time, it is apparent that the Court of Appeals first adopted the general rule for the purpose of avoiding "unwarranted windfalls," in an era of fewer so-called "national" law firms and less readily available fee data. The rule designed to avert a windfall should not be employed to create one.
In Agent Orange, 818 F.2d 226, the district court faced a then less common situation involving multiple parties of national scope and dozens of non-local counsel. Deviating from the general rule enunciated in Polk, the district judge computed a "national rate" applicable to all counsel.
The Court of Appeals affirmed this procedure as within the lower court's discretion:
An examination of the alternatives to the use of national rates in large multiparty class actions of this sort readily establishes the necessity for affording district courts this discretion. Use of our forum rule would distort dramatically the purposes of the lodestar calculation itself — to ensure fair and just compensation to counsel and to prevent the award of windfall fees. This distortion would occur because, in cases in which the vast majority of attorneys involved are non-local, the forum rule necessarily will either overcompensate or undercompensate a substantial number of non-local attorneys. Undercompensation could deny counsel their right to fair and just fees; overcompensation would not be consistent with the need to prevent windfalls. Adherence to the forum rule in cases in which the inherent limitations of the rule are magnified, i.e., where few local counsel and vast numbers of non-local counsel are involved, therefore, makes little sense.818 F.2d at 232-33.
In this case, lead counsel and a number of supporting firms were based outside New York, in cities where hourly billing rates are markedly lower. See, e.g., Altman Weil at 90-91 (reporting $185 and $100 median hourly billing rates for partners and associates, respectively, in the sample for lead counsel's state of Minnesota, in comparison with $290 and $200, respectively, in New York). Even the rates charged by the supporting New York firms tended to sit on the high end of the fee spectrum. Id. at 91 (reporting median rates of $290 and $200, respectively, for partners and associates in New York). Were the rates recalculated to reflect these medians and home state standards, the lodestar would be further reduced and, concomitantly, the multiplier would increase.
Lead counsel, responding to the draft of this Report, pointed out that other district courts have utilized their regular hourly rates in cross-checking percentage fee awards.
In this particular case, however, the multiplier would not rise to such a level as to commend a more drastic cut in the fees sought. Because the lodestar is utilized here as a cross-check, rather than the primary calculus, it is not necessary to resolve the thorny questions of whether to adjust the billing rates and which regional rates should apply. Suffice it to say, these data reinforce the conclusion that the recommended cut in fees is not unfair.
One concern that surfaced in my initial review, and was reflected in the draft, involved large increases in lead counsel's hourly billing rates during the course of the litigation. It was satisfactorily explained that certain lawyers had previously forgone billing rate increases for several years and that this was not an attempt to exploit the circumstances of this case.
The raw lodestar is subject to an upward or downward multiplier to account for particular circumstances of the case. A multiplier is not, however, presumed. To the contrary, the unadjusted lodestar is "strongly presumed to be reasonable . . ." Bolar Pharmaceutical, 966 F.2d at 732. A multiplier bestows excessive compensation if the lodestar fee already reflects the fair value of the lawyers' services. See Agent Orange, 818 F.2d at 237.
Stated somewhat differently, an hourly rate inherently reflects risk, quality and other factors associated with an attorney's competitiveness, skills and overhead. Adding a multiplier to a base rate that already accounts for these factors would result in windfall compensation to lawyers at the expense of their class clients. See In re Bolar Pharmaceutical Co. Securities Litigation, 800 F. Supp. 1091, 1096 (E.D.N.Y. 1992).
Counsel's request of $175,000,000 reflects a multiplier of 3.73 (against the unadjusted lodestar), which they contend is reasonable. They assert that a typical multiplier sits between 3 and 4.5. (Mem. in Supp. of Lead Securities Counsel's Application at 14.)
Courts and commentators have differed greatly in their assessments of the dominant range of multipliers. One district judge computed an average multiplier of 1.44, after conducting an analysis of 49 federal securities law actions litigated within the Second Circuit. See In re McDonnell Douglas Equipment Leasing Securities Litig., 842 F. Supp. 733 (S.D.N.Y. 1994); compare William J. Lynk, "The Courts and the Plaintiff's Bar: Awarding the Attorney's Fee in Class-Action Litigation," 23 J. Legal Studies 185, 196 (1994) (reporting an average multiplier of 1.69 for class actions studied between 1973 and 1990); Wayne Schneider, "Courts Don't Have to Award Excessive Fees," 20 NAPPA Report at 10 (range of multipliers between 1.37 and 3.1 in seven representative class action settlements in 2004 and 2005); In re Cendant Corp. PRIDES Litig., 243 F.3d 722, 742 (3d Cir. 2001), cert. denied sub nom. Kirby McInerney Squire, LLP v. Joanne A. Aboff Family Trust, 534 U.S. 889 (2001) (in surveying cases with common funds over $100 million, court found multiplier of 1.35 to 2.99 common); In re NASDAQ Market-Makers Antitrust Litig., 187 F.R.D. 465, 489 (S.D.N.Y. 1998) (multipliers between 3 and 4.5 are common); Newberg, § 14:6 ("[m]ultiples ranging from one to four frequently are awarded in common fund cases when the lodestar method is applied"); see also Wal-Mart, 396 F.2d at 123 (3.5 multiplier); WorldCom, 388 F. Supp.2d at 354 (4.0 multiplier); In re Avon Products, Inc. Securities Litig., Fed. Sec. L. Rep. (CCH) ¶ 97061, 1992 WL 349768 (S.D.N.Y. 1992) (decision used percentage method, but stated that under the lodestar approach a risk multiplier of 2 to 3 would have been appropriate).
There is little definitive guidance in the evolving science of lodestar cross-checking on whether to use the adjusted or unadjusted lodestar for this purpose. Leading recent opinions have differed in their approaches. Contrast Visa, 396 F.3d at 123; WorldCom, 388 F. Supp.2d at 354, with Ling v. Cantley Sedacca, 2006 WL 290477; see also McDonald, 450 F.3d 91.
The proposed 5.9 percent award, or $147,500,000, results in a multiplier of 3.69 against the adjusted lodestar of $39,973,056.76. This amount is tantalizingly close to the 3.73 multiplier which counsel urge is appropriate here. It would represent a multiplier of approximately 3.15 against the unadjusted lodestar of $46,861,731.25.
Either multiplier lies within the trending range in a case of this nature. These approximate the 3.5 multiplier in Visa and the 4.0 multiplier in WorldCom, and sit high within the emerging spectrum. The multiplier also does not account for the additional work performed by counsel since submitting their application ( see Heins Decl. ¶ 99), which would reduce the number modestly. The result here would appropriately recognize risk, investment and quality.
This multiplier recognizes, in particular, the risk and quality stemming from the issues of loss causation, scienter and damage calculation, tempered by the collectibility of the amounts sought and the inherent likelihood of settlement in a case of this magnitude. See Alexander, 43 Stanford L. Rev. at 578 (suggesting that large securities class actions almost invariably settle). Loss causation was a serious issue upon which summary judgment motions were pending. Lead counsel assumed some unusual financial risks in the face of these uncertainties, nearly doubling the size of the firm to handle the workload of this case and advancing millions of dollars in disbursements, including additional space and equipment to accommodate the document review.
It bears emphasis that the lodestar computation here is a cross-check, calculated with less precision than would be required if lodestar were the primary methodology. Thus, although a multiplier of 3.69 is on the higher side, the lodestar cross-check is sufficiently within bounds to sustain the fairness of the award.
Objections
In arriving at the recommended percentage, I have considered the four objections filed by putative class members to counsel's application. Indeed, the very fact that only four objections were received from a class potentially comprising hundreds of thousands — with no demurral from any institutional investors — influences my recommendation. See Wal-Mart, 396 F.2d at 118-19.
Objections were filed by Margaret M. Keffer, Pat L. Canada, Paul Heyburn and Cynthia R. Levin Moulton, and rejoined by lead counsel. Notwithstanding class counsel's contention that two of the objectors lack standing, I will address each objection to ensure a full record.
Ms. Keffer claims to have lost $2,700 on her purchase of 100 shares. Her pro se objection stands as a reminder that, beyond the lawyers, institutions and sophisticated shareholders who increasingly spearhead these cases, are many minor shareholders who, as an aggregate group, owned a large percentage of stock and will also absorb the impact of the fee award, albeit collectively. This is among the various circumstances that informed my proposed reduction of nearly 16 percent in counsel's fee request.
Ms. Keffer's objection otherwise lacks specifics for this Report Recommendation to address. See In re Prudential Securities Inc. Ltd. Partnerships Litig., MDL No. 1005, M-2-67 (MP), 1995 WL 798907 (S.D.N.Y. Nov. 20, 1995). Her grievance actually appears to be directed more substantively at the settlement itself. Lamenting that the settlement amounts to only three cents on her invested dollar, she argues that "1/3 would be fair or at least acceptable." She had the right, perhaps not entirely practical, to opt out of the class. Realistically speaking, there is nothing now that the Court could do in this fee application to satisfy her demand. Even if the Court awarded nothing more than $45 million in lodestar and expenses, Ms. Keffer's recovery would come little closer to the one-third she seeks.
Pat Canada urges that counsel's fee be limited to $100 million. Class counsel point out that, in commenting earlier upon the settlement, Mr. Canada suggested a fee of $125 million — an amount close to this recommendation. Mr. Canada urges the Court to "follow the lead of the Court" in WorldCom, surmises that "the true lodestar is less than $20 million," and notes that the cases was neither tried nor even prepared for trial and that parallel government investigations made the "job of class counsel much easier."
While the recommended compensation is close to Mr. Canada's desire, some of his considerations are more on target than others. As noted, the WorldCom decision and the retainer letter in that case have both been instructive. On the other hand, $20 million is an unreasonably low assessment of the lodestar. And although the case never reached the ultimate stages of trial preparation, the collection, coding and evaluation of the 15.5 million documents was part of the trial groundwork. Finally, though counsel's case was helped here by parallel regulatory and criminal investigations, this was more a case of tandem riding than of piggybacking.
Mr. Heyburn demurs on the bases that "[t]his settlement is not fair, adequate or reasonable," that the class representatives underperformed by failing to "take into account the application of the laws of the 50 states," that the fees sought represent "an exorbitant sum," and that "[t]here is no comparison with a lodestar/multiplier calculation or a summary of hours spent working on the case." The first two objections are essentially challenges to the settlement itself, which, of course, have already been addressed and rejected by the Court. The third objection is conclusory. The final objection has been addressed through my evaluation.
Ms. Moulton similarly urged that the Court defer any fee evaluation until counsel submitted detailed time and expense records. Though disputing the reasonableness of counsel's demand, she offered no proposal of her own, merely alleging that, at least at the time of her objection, the Court lacked sufficient evidence.
I have received and reviewed class counsel's time records and expense summaries. I have used the time records as a cross-check against the lodestar and recommended a $27.5 million reduction in the fees sought. Counsel's expense application has also been reduced by approximately a half million dollars, as set forth below.
Disbursements
Counsel sought reimbursement of their out-of-pocket disbursements in ten categories: document and database management; delivery, courier and postage; hotel, meals and transportation; telephone/facsimile; computer research/document retrieval; experts; professional and outside services; special master services; process server fees; and additional case expenses. The request totaled $4,501,332.43, or approximately 11 percent of the adjusted lodestar.
After reviewing summaries of the expenses, broken down by firm, I requested and received supplemental information and documentation from counsel. In other instances, I am recommending the approval of certain categories, based on my scrutiny of the record, counsel's timesheets and other comparably sized cases. The charges in each of those categories appear reasonable and commensurate with the types of services performed. Counsel also voluntarily accepted several of my draft recommendations and reduced their disbursement requests accordingly.
The approved-in-full categories consist of delivery, courier and postage ($20,799.44); telephone/facsimile ($32,749.61); professional services (largely outside investigators) ($94,575.75); and process server fees ($14,314.67).
The category of "document and database management services" contains seven sub-categories. I recommend allowance in full of cable services ($1,319.40); computer consulting ($38,653.30); and telephone services at depository ($4,607.57). Counsel have voluntarily withdrawn their requests for depository equipment and furniture ($142,191.44), depository lease, storage and security ($148,459.21) and document destruction ($654.00). I have, however, taken these risks and expenses into account in determining an appropriate fee percentage and reviewing the attendant lodestar multiplier.
For "document production services," which represents photocopying, counsel have now agreed to seek the lesser of actual cost or 20 cents per page, the rate set forth in Rule 39 of the Second Circuit. The total allowed in this category is $605,715.20.
At my request, counsel provided me with a further breakdown of the category of hotels, meals and transportation and reconfigured their data, representing that they have complied with the following guidelines recommended in the draft report:
a. Out-of-town travel expenses are limited to a maximum of four lead counsel lawyers and two other class lawyers per hearing.
b. Only out-of-town meals will be reimbursed, subject to the above limitations.
Accordingly, I recommend approval of $437,819.25 in transportation, $373,355.45 for hotels and $94,364.43 for meals.
The $166,909.03 sought for computer research and document retrieval has been consensually reduced to $66,543.96. Some of the original charges were for access to electronic legal research platforms such as Lexis or Westlaw, or to the federal courts' PACER system. Lexis and Westlaw offer subscription plans, in which certain access is provided for a base fee and surcharges are applied for access to data not included in the basic subscription. Counsel voluntarily agreed to amend their request to eliminate the base subscription fees. PACER charges are allowed in full.
Counsel seek reimbursement for experts in four sub-categories: class benefits ($24,950.00); computer forensics ($7,128.26); financial accounting and auditing ($1,201,331.86); and economists ($396,102.73). Counsel have provided me with further substantiation which justifies the allowance of those expenses.
Of the "additional case expenses," books and publications ($673.80), court reporting ($134.38), court costs ($2,042.25) and witness and mileage fees ($56.20) are all justified. Counsel have voluntarily withdrawn their requests for training facilities ($11,538.55) and "miscellaneous" ($314.76). No recommendation or report is necessary for categories of expenses for which authorization was already given to counsel in prior court orders.
As a result of their voluntary acceptance of the recommendations in the draft report, counsel have commendably reduced their expense request by a total of at least $486,108.28. A copy of counsel's modified submission is annexed as Appendix A.
Conclusion
This recommendation of $147,500,000 harmonizes the interests of counsel and the class. It passes the tests of moderation and reasonableness while at the same time setting a mark that almost certainly will tempt well qualified counsel to accept comparable risk in the future. Just as importantly, it reflects a number that a sophisticated, hard-bargaining client, facing similar future claims in its business operations, should consider fair compensation for counsel who assume this type of risk and achieve this level of result.
In sum, the prospect of a collectible fee of $147.5 million for approximately three years of substantive work is more than sufficient to attract high quality securities litigation counsel to the cause of a deserving class. Such an award, which places counsel among the nation's revenue-generating elite and results in at least 161 effective attorney billing rates exceeding $1,000 per hour, fairly rewards those attorneys for a job well done and creates strong incentives to vie for future cases.