Summary
finding that "[t]he reaction of the class to the settlement has been almost entirely positive" where "[n]one of the institutional investors have objected to the size of the settlement"
Summary of this case from Hill v. State St. Corp.Opinion
Case No. 02-md-1335-PB, ALL CASES, Opinion No. 2007 DNH 156.
December 19, 2007.
MEMORANDUM AND ORDER
Lead Plaintiffs brought this class action against Tyco International, Ltd. ("Tyco"), its former auditor, PricewaterhouseCoopers, LLP ("PwC"), and five individual defendants: former Tyco Chief Executive Officer L. Dennis Kozlowski, former Chief Financial Officer Mark H. Swartz, former General Counsel Mark A. Belnick, and former directors Frank L. Walsh and Michael A. Ashcroft (collectively the "Individual Tyco Defendants"). They successfully negotiated a proposed settlement with Tyco and PwC, and now petition for final approval of the proposed settlement. The law firms representing Lead Plaintiffs ("Co-Lead Counsel") have also applied for an award of attorneys' fees and reimbursement of expenses incurred in connection with the prosecution and settlement of this action. For the reasons discussed herein, I approve both motions.
Lead Plaintiffs are the Plumbers and Pipefitters National Pension Fund ("Plumbers"), United Association General Officers Pension Plan ("UAGO"), United Association Local Union Officers Employees Pension Fund ("UALOE"), Teachers' Retirement System of Louisiana ("TRSL"), and the Louisiana State Employees' Retirement System ("LASERS").
I. BACKGROUND
I describe below only those facts and events necessary to place this settlement in context. The full extent of the alleged problems at Tyco during the relevant time period are described in detail in my orders certifying the class and disposing of defendants' motions to dismiss. See generally In re Tyco Int'l, Ltd. Multidistrict Litig., 236 F.R.D. 62 (D.N.H. 2006); In re Tyco Int'l, Ltd. Multidistrict Litig., No. MDL-02-1335-B, 2004 WL 2348315 (D.N.H. Oct. 14, 2004). In brief, plaintiffs allege that during the class period of December 13, 1999, through June 7, 2002, defendants misrepresented the value of multiple companies that Tyco acquired and misreported Tyco's own financial condition in ways that artificially inflated the value of Tyco stock. These fraudulent accounting practices, plaintiffs allege, enabled the Individual Tyco Defendants to reap enormous profits by looting the company through a combination of unreported bonuses, forgiven loans, excessive fees, and insider trading. The looting, in turn, allegedly fostered a coverup by means of continued accounting fraud, materially false and misleading statements, and the omission of material information in various registration statements to cover up the misconduct, all of which further violated the federal securities laws. Meanwhile, PwC allegedly failed to conduct its audits of Tyco's financial statements in accordance with Generally Accepted Auditing Standards ("GAAS") and falsely certified that Tyco's financial statements were fairly presented in accordance with Generally Accepted Accounting Principles ("GAAP").
A. Consolidated Complaint
B. Motion to Dismiss
See generally In re Tyco Int'l, Ltd. Sec. Litig. 185 F. Supp. 2d 102 Tyco I
In July 2005, after the Supreme Court clarified the requirements for establishing loss causation in securities fraud actions, see Dura Pharms., Inc. v. Broudo, 544 U.S. 336 (2005), defendants moved to revisit the motions to dismiss on the question of loss causation. Under Dura, defendants argued, plaintiffs' theory of loss causation (that revelations of looting by the corporate principals caused investors to conclude that they could no longer credit the company's denials of accounting misconduct) was no longer sufficient. I denied this motion.
PwC also filed a motion for summary judgment on loss causation. After briefing and oral argument, I denied this motion. PwC then filed a motion for certification of an interlocutory appeal of this ruling, which had not yet been resolved when PwC agreed to settle.
C. Class Certification
Plaintiffs moved to certify the class on January 14, 2005. Among Tyco's objections to certification was a novel "equity conflict" argument regarding the class members who presently hold a greater share of Tyco's stock than they did during the class period. Tyco argued that these "equity holders" stood to lose more as shareholders than they had to gain as class members because any payment by Tyco to the class would correspondingly reduce the value of their present holdings. Thus, argued Tyco, the interest of equity holders in protecting their present holdings conflicted with the interest of Lead Plaintiffs in recovering damages, and should therefore defeat class certification. I denied this motion because the equity holders (even though they had an interest in preventing others from recovering) nevertheless had a strong interest in recovering on their own claims against Tyco and, more fundamentally, because this potential harm to a subgroup of the class should not bar the remaining class members from being able to proceed as a class. Tyco then filed an unsuccessful appeal of my class certification order in the Court of Appeals for the First Circuit. See In re Tyco Int'l, Ltd. Sec. Litig., No. 06-8022 (1st Cir. Sept. 22, 2006).
D. Discovery and Other Motion Practice
It would be difficult to overstate the volume of discovery in this case. Co-Lead Counsel propounded over 700 requests for admission, documents requests, and interrogatories; participated in over 220 depositions in New York, Florida, Massachusetts, and New Hampshire; and reviewed some 82.5 million pages of documents produced by defendants. This volume of discovery was necessitated by the breadth of plaintiffs' allegations, which spanned more than one hundred different allegedly fraudulent corporate acquisitions by Tyco. Moreover, because of the complexity of the alleged fraud, Co-Lead Counsel needed to retain expert consultants and forensic accountants to assist them in interpreting the information they obtained through discovery.
This discovery process was paired with aggressive, skillfully argued, and unusually challenging motion practice. Defense counsel matched the tenacity of Co-Lead Counsel, missing no opportunity to raise nonfrivolous objections or file nonfrivolous appeals of adverse decisions. This motion practice frequently explored complex, cutting-edge issues in which the state of the law changed even as this case was being litigated. In particular, the parties heavily litigated loss causation when the law in this area was in a state of flux. Plaintiffs' theory of the case requires the fact finder to draw a causal connection between the revelations of apparent corporate looting by Tyco's principals (amounts that were only a small portion of Tyco's corporate profits) and the precipitous decline in Tyco stocks. This theory is novel and exposed plaintiffs to a nontrivial risk of dismissal at the 12(b)(6) stage.
E. Settlement Negotiations
On January 13, 2004, Judge Donald E. Ziegler, retired U.S. District Judge for the Western District of Pennsylvania, assisted with an initial mediation attempt between Lead Plaintiffs and Tyco. That mediation failed because of the tremendous gap between Lead Plaintiffs' request for multibillion-dollar damages and Tyco's desire to settle for mere nuisance damages.
Starting in the fall of 2005, Lead Plaintiffs and Tyco agreed to a new mediator and a new mediation process. The new mediator was Judge Stanley Sporkin, retired U.S. District Judge for the District of Columbia and former Director of the Enforcement Division of the Securities and Exchange Commission ("SEC"). After some initial failures to bridge the gap, Judge Sporkin adopted a new mediation strategy, appointing Eugene R. Sullivan, former chief judge of the U.S. Court of Appeals (Armed Forces), as "advocate" for Tyco and Marvin E. Jacob, certified mediator in the U.S. Bankruptcy and District Courts for the Southern District of New York and a former Associate Regional Administrator for the New York Regional Office of the SEC, as "advocate" for plaintiffs.
Judge Sporkin has numerous other qualifications that are not necessary to recite here. He has received numerous awards and honors during his long and distinguished career in government service. His expertise in this area is widely acknowledged.
The parties engaged in multiple mediation sessions under this new mediation setup, some lasting multiple days. In March 2007, after several relatively small moves by each side over time, Judge Sporkin made another serious push to resolve the case, working with each side in an effort to bring the decision-makers to the bargaining table. Although the March 2007 mediation did not immediately result in a settlement agreement, both sides continued discussions until May 15, 2007, when they reached an agreement in principle to settle the claims for $2.975 billion in cash.
Lead Plaintiffs then used their agreement with Tyco in new negotiations with PwC mediated by Judge Nicholas Politan, retired U.S. District Judge for the District of New Jersey. In June of 2007, Lead Plaintiffs and PwC reached an agreement in principle to settle for the sum of $225 million.
F. Damages Calculations
During the settlement negotiations, Dr. Mark E. Zmijewski prepared a report for plaintiffs estimating the damages suffered by the class. I describe his report in some detail because it is relevant to my analysis of the settlement.
Dr. Zmijewski is the Leon Carroll Marshall Professor of Accounting and Deputy Dean at the University of Chicago Graduate School of Business. He formerly served as Executive Director of the Center for Research in Securities Prices at the University of Chicago Graduate School of Business.
Dr. Zmijewski used an event study methodology to quantify "abnormal" stock and bond movements that could support plaintiffs' argument for loss causation and measure how and when Tyco's equities and bonds were trading at artificially inflated prices. He first disaggregated marketwide inflationary effects from firm-specific effects, and then determined whether the firm-specific effects on each trading day were statistically significant.
Based on plaintiffs' theory of the case, which posited that both revelations of accounting fraud and revelations of corporate looting could qualify as corrective disclosures, Dr. Zmijewski identified eight corrective disclosure dates: January 3, 2002; January 23, 2002; January 29, 2002; April 25, 2002; May 28, 2002; June 3, 2002; June 6, 2002; and June 7, 2002. He then weighted each corrective disclosure based on an estimate of the degree to which other confounding factors (e.g., lowered earnings forecasts, reported losses, or other stock-deflating news announced on the same day as doubts about Tyco's accounting procedures) influenced the market on the day of the disclosure. Dr. Zmijewski then identified inflation-creating dates, i.e., those dates on which Tyco's stock rose and defendants' alleged fraud resulted in the disclosure of false information. Using the corrective disclosure and inflation-creating dates, Dr. Zmijewski calculated inflation per share on a daily basis for the entire class period. He did this by assuming that the stock price was "clean" or uninflated at the end of the class period and then working backward day by day to the beginning of the class period.
A corrective disclosure date is one on which information that defendants allegedly withheld fraudulently from the market became known to the market and the stock price experienced a corresponding, statistically significant downward movement.
Dr. Zmijewski used the same methodology, with adjustments to reflect the differing availability of data, for estimating bond inflation. There were two broad types of Tyco debt instruments traded during the class period: coupon debt (consisting of thirty-two instruments) and zero-coupon debt (consisting of two instruments). Because the price series for most Tyco bonds were missing some information, Dr. Zmijewski selected one bond from each of these two groups that had the most complete price series and used this to conduct the event study. To properly account for the difference in price level for some of the bonds, he measured the inflation in constant percentage terms. Dr. Zmijewski then used the same corrective disclosure/inflation-creating date method to calculate inflation on a daily basis, with the added adjustment of capping the percentage inflation of the bonds so that bond inflation did not exceed stock inflation. He treated one bond — the Tyco Liquid Yield Option Note — separately from the others, because this instrument tracked Tyco's stock price rather than debt prices.
Dr. Zmijewski then used the stock and bond inflation estimates to calculate aggregate damages. He calculated stock inflation damages according to two different models: an institutional model and an individual model. The institutional model relied on quarterly holdings information and estimated damages from each institution. Based on the assumption that gains from a particular share within an institution should not be netted against losses on other shares or opening balances at the beginning of the class period, Dr. Zmijewski estimated total institutional equity damages of $7.1 billion. The individual model relied on the "actual trader model," using empirical trading data from brokerage accounts of individual traders to predict individual trader behavior. By combining the total non-institutional trading volume on each day with his trader behavior model, Dr. Zmijewski was able to simulate the trading pattern that occurred. This simulation yielded an estimate of $3.7 billion in equity damages for individual stock traders. For bonds, he multiplied the face value of the debt issued times the inflation at the time of issuance, yielding aggregate bond damages of $905 million. Combining the institutional and individual equity damages with the bond damages yielded total measurable aggregate damages of $11.73 billion.
Dr. Zmijewski also calculated an alternative, far more conservative damages model based only on Tyco's earning restatements, which he speculated would be similar to defendants' preferred damages model. The earning restatements model merely looked at the difference between Tyco's actual stock price and Tyco's stock price had its financials been reported accurately, without estimating damages caused by interaction between accounting fraud, management looting, and loss of investor confidence in the company. Accordingly, the earning restatements model yielded the far smaller estimate of $2.7 billion in equity damages, or $8.1 billion less than the $10.8 billion in equity damages produced by the event study model.
Had the case proceeded to trial, defendants represent that they would have offered expert witnesses to show that the class suffered no damages or only minimal damages. Defendants' brief in support of final approval of this settlement did not, however, provide specific alternative calculations.
G. Settlement Terms
As agreed upon by the parties, the proposed settlement provides for the payment by Tyco of $2.975 billion in cash, plus interest, and the payment by PwC of $225 million in cash, plus interest. Tyco's payment will be the largest cash payment ever made by a corporate defendant in the history of securities litigation. PwC's payment will be the second-largest auditor settlement in securities class action history. In all, the proposed settlement is the third largest securities class action recovery in history, behind only Enron and WorldCom.
The proposed settlement also provides for the assignment to Tyco of the class's claims against the Individual Tyco Defendants Kozlowski, Swartz, and Walsh (none of whom are part of the settlement). In return for this assignment, Tyco agreed that 50% of any net recovery against those individuals, both on its own claims and on the class's assigned claims, will be transferred to the settlement fund.
H. Plan of Allocation
The settlement fund has already been paid into escrow accounts at a number of major banks. Under the proposed plan of allocation, the settlement funds (less administrative and notice costs, taxes and related expenses, and attorneys' fees and expenses, but plus the interest earned by the fund) will be distributed to class members who submit timely, valid Proof of Claim forms. The Garden City Group ("GCG"), a claims administrator retained by Co-Lead Counsel, will calculate each claimant's share according to the information on the Proof of Claim forms, apportioning each recovery according to Dr. Zmijewski's damages calculations.
Co-Lead Counsel retained two additional independent experts, R. Alan Miller and Dr. Kenneth D. Gartrell, to opine on the reasonableness and fairness of the plan of allocation. Mr. Miller and Dr. Gartrell both opined that the Plan is reasonable and fair from the perspective of investors who acquired Tyco stock during the class period.
Mr. Miller is President of the Philadelphia Investment Banking Company ("PIBC"), which he co-founded in 1983. Among other things, PIBC frequently performs evaluations of businesses and securities for investors. Mr. Miller regularly testifies as an expert witness on securities-related issues.
Dr. Gartrell is a financial economist and economic consultant. He formerly served as a Managing Director at LECG, LLC, an international economic consulting firm, was the founding partner of Ken Gartrell Company, and was a Principal in the economic consulting firms of The Brattle Group and Putnam, Hayes Bartlett.
I. Notice to the Class
After contacting nominee purchasers (i.e. banks, brokers, and other purchasers of record who bought and sold Tyco securities on behalf of others) to identify the beneficial owners on whose behalf the nominees acted, GCG (the claims administrator retained by Co-Lead Counsel) mailed some 2.4 million claim packets containing the Notice of Proposed Settlement to class members and their nominees. For claim packets that were returned undelivered, GCG followed up by periodically searching for the recipients' new addresses and re-mailing the packets until the packets were successfully delivered.
In addition to the mailed claim packets, GCG published summary notices in USA Today, The New York Times, The Wall Street Journal, The Financial Times, South Florida Sun-Sentinel, Palm Beach Post, Chicago Times, The Union Leader (Manchester, NH), and over the PR Newswire. GCG also set up a website and toll-free telephone hotline to assist class members in submitting their claims and provide answers to their questions regarding the settlement.
J. Reaction of the Class to the Notice
Proof of Claim forms are due by December 28, 2007, so the Proof of Claim numbers are not final. Keeping that in mind, out of the 2.4 million copies of the Notice mailed to potential class members, GCG had received 74,655 Proof of Claim forms as of October 12, 2007, and 288 requests for exclusion from the class as of the September 28, 2007 deadline for exclusion.
The court has received twenty-eight objections to the proposed settlement. Of these, four are from institutional investors and the remainder are from individual investors, a handful of whom are not actually members of the class. These objections are described and discussed individually in the analysis section below.
K. Fairness Hearing
The Fairness Hearing took place on November 2, 2007. Co-Lead Counsel appeared, as did counsel for additional plaintiffs. Counsel for Tyco, PwC, Belnick, and Ashcroft were also present. Counsel for two objectors, the Commonwealth of Pennsylvania State Employees' Retirement System and Public School Employees' Retirement System ("SERS/PSERS") and U.S. Trust, requested and were given the opportunity to be heard.
II. ANALYSIS
A. Adequacy of Notice
The Federal Rules of Civil Procedure require, upon certification of the class, that "the best notice that is practicable under the circumstances" be given, "including individual notice to all members who can be identified through reasonable effort." Fed.R.Civ.P. 23(c)(2)(B). The Federal Rules of Civil Procedure further require, upon the successful negotiation of a proposed settlement, that notice be given "in a reasonable manner to all class members who would be bound by the proposal." Fed.R.Civ.P. 23(e)(1). The Due Process Clause requires that notice be "reasonably calculated to reach potential class members." In re Compact Disc Minimum Advertised Price Antitrust Litig., 216 F.R.D. 197, 203 (D. Me. 2003); see also Eisen v. Carlisle Jacquelin, 417 U.S. 156, 174 (1974); Mullane v. Cent. Hanover Bank Trust Co., 339 U.S. 306, 315 (1950). Additionally, the Private Securities Litigation Reform Act ("PSLRA") requires that the notice of settlement provide statements of plaintiff recovery, potential outcome of the case, attorneys' fees or costs, identification of plaintiffs' representatives, and the reasons for settlement. 15 U.S.C. § 78u-4(a)(7).
In this case, Co-Lead Counsel and GCG jointly developed an effective notice program. As detailed above, the notice program involved pre-mailing contacts with nominee purchasers, repeated efforts at mailing the claim packets to class members, publication notice in eight national and regional newspapers, and a website and toll-free telephone hotline.
This notice program compares favorably with the notice programs in other securities cases. See In re Cabletron Sys., Inc. Sec. Litig., 239 F.R.D. 30, 35-36 (D.N.H. 2006) (approving a notice program that distributed notice packets to individual investors and nominees, published a summary notice in one national newspaper, and provided a toll-free telephone hotline); see also In re WorldCom, Inc. Sec. Litig., 388 F. Supp. 2d 319, 332-33 (S.D.N.Y. 2005) (approving a notice program that distributed notice packets to potential class members and published summary notices in two national newspapers and over two wire services). Accordingly, I find that Co-Lead Counsel's notice program met or exceeded all relevant requirements. To the extent that any objectors have claimed that the notice program was not adequate, I overrule their objections. I further find that the contents of the notice packets were acceptable and met all relevant requirements.
B. Adequacy of the Settlement
Under Rule 23(e) of the Federal Rules of Civil Procedure, I may approve this class action settlement only if I conduct a fairness hearing and find that the terms of the settlement are "fair, reasonable, and adequate." Fed.R.Civ.P. 23(e)(2). In the First Circuit, this requires a wide-ranging review of the overall reasonableness of the settlement that relies on neither a fixed checklist of factors nor any specific litmus test. See In re Lupron Mktg. Sales Practices Litig., 228 F.R.D. 75, 93 (D. Mass. 2005) ("[T]he First Circuit has not established a formal protocol for assessing the fairness of a settlement."); Compact Disc, 216 F.R.D. at 206 ("There is no single test in the First Circuit for determining the fairness, reasonableness and adequacy of a proposed class action settlement."); Bussie v. Allmerica Fin. Corp., 50 F. Supp. 2d 59, 72 (D. Mass. 1999) ("This fairness determination is not based on a single inflexible litmus test but, instead, reflects its studied review of a wide variety of factors bearing on the central question of whether the settlement is reasonable in light of the uncertainty of litigation.").
Although the district court must carefully scrutinize the settlement, there is a presumption in favor of the settlement if the parties negotiated it at arms-length, after conducting meaningful discovery. City P'ship Co. v. Atl. Acquisition Ltd. P'ship, 100 F.3d 1041, 1043 (1st Cir. 1996); Nilsen v. York Cty., 382 F. Supp. 2d 206, 212 (D. Me. 2005); Compact Disc, 216 F.R.D. at 207. Moreover, public policy generally favors settlement — particularly in class actions as massive as the case at bar. See WorldCom, 388 F. Supp. 2d at 337.
Some courts have relied on the Second Circuit's Grinnell factors to determine the fairness of a settlement:
(1) the complexity, expense and likely duration of the litigation; (2) the reaction of the class to the settlement; (3) the stage of the proceedings and the amount of discovery completed; (4) the risks of establishing liability; (5) the risks of establishing damages; (6) the risks of maintaining the class action through the trial; (7) the ability of the defendants to withstand a greater judgment; (8) the range of reasonableness of the settlement fund in light of the best possible recovery; (9) the range of reasonableness of the settlement fund to a possible recovery in light of all the attendant risks of litigation.City of Detroit v. Grinnell Corp., 495 F.2d 448, 463 (2d Cir. 1974) (internal citations omitted); see also In re Cendant Corp. Litig., 264 F.3d 201, 232 (3d Cir. 2001) (Cendant I) (applying the Grinnell factors, as is required in the Third Circuit);WorldCom, 388 F. Supp. 2d at 337 (applying the Grinnell factors, as is required in the Second Circuit); Lupron, 228 F.R.D. at 93 (applying the Grinnell factors as a matter of preference).
In Compact Disc, the District of Maine recently used a modified version of the Grinnell factors:
(1) comparison of the proposed settlement with the likely result of litigation; (2) reaction of the class to the settlement; (3) stage of the litigation and the amount of discovery completed; (4) quality of counsel; (5) conduct of the negotiations; and (6) prospects of the case, including risk, complexity, expense and duration.Compact Disc, 216 F.R.D. at 206.
Although I have the discretion to use the Grinnell list of factors verbatim, I find that a more concise list of the considerations at play, modeled on those used in Compact Disc, best fits the facts of this case. Accordingly, I discuss the following considerations in turn: (1) risk, complexity, expense and duration of the case; (2) comparison of the proposed settlement with the likely result of continued litigation; (3) reaction of the class to the settlement; (4) stage of the litigation and the amount of discovery completed; and (5) quality of counsel and conduct during litigation and settlement negotiations.
1. Risk, Complexity, Expense, and Duration of the Case
It is difficult to overstate the complexity of this case. As the depth and breadth of discovery suggests, the facts were not easy to ascertain. Plaintiffs had to scrutinize documents from more than one hundred different corporate acquisitions by Tyco to identify whether those acquisitions involved fraudulent accounting, how the transactions were audited by PwC, and what measures defendants may have taken to cover up the fraudulent accounting.
This was not just a complex case, however. It was also a risky case for both sides, in large part because of an uncertain legal environment. Plaintiffs' theory of the case put them at the cutting edge of a rapidly changing area of the law. In particular, the still-developing law of loss causation in securities cases created significant risk and uncertainty for plaintiffs. According to plaintiffs' theory, the misstatements by Tyco and PwC proximately caused investors' losses, as shown by drops in Tyco's stock prices following the eight corrective disclosures. None of those disclosures, however, involved a specific admission of fraud by Tyco or PwC. Instead, the disclosures mostly related to the integrity of Tyco's management. Although I denied defendants' motion to dismiss and PwC's motion for summary judgment, I did so only after careful consideration and extensive briefing. Moreover, the law remains in flux and it is by no means certain that plaintiffs would have prevailed if they had taken the case to trial and attempted to defend any favorable verdict on appeal.
This case involved a greater risk of non-recovery than other multibillion-dollar securities class action settlements. WorldCom involved a number of complex issues, but the fraud was so extensive that there were many targets to pick off, providing funds for continued litigation. Cendant was, as the Third Circuit observed, a "simple case in terms of liability with respect to Cendant, and the case was settled at a very early stage, after little formal discovery," which made the risks of non-recovery negligible. Cendant I, 264 F.3d at 286. In AOL Time Warner, the risks were "above-average" and loss causation was a similarly unsettled issue, but the chain of causation involved fewer logical leaps and was therefore easier to prove than in this case. See In re AOL Time Warner, Inc. Sec. "ERISA" Litig., No. 02 Civ. 5575 (SWK), 2006 U.S. Dist. LEXIS 78101, at *47-*48 (S.D.N.Y. Sept. 28, 2006). Reflecting these risks and complexities, the parties on both sides necessarily incurred considerable expense in litigating the case.
Counsel for both Tyco and PwC made it clear by their words and deeds during the course of the litigation that they intended to vigorously contest plaintiffs' claims. Defendants selected several of the country's most skilled advocates as their representatives and those advocates responded by uncovering and effectively advancing every non-frivolous legal argument that could conceivably be presented on their clients' behalf. In short, the case took nearly five years to resolve because it was factually complex, turned on several novel and difficult legal issues, and was aggressively and effectively litigated by defendants who were determined to spare no expense in protecting their interests.
If the case survived summary judgment and went to trial, plaintiffs would face additional risk, uncertainty, and delay. Proving loss causation would be complex and difficult. Moreover, even if the jury agreed to impose liability, the trial would likely involve a confusing "battle of the experts" over damages. If, faced with conflicting expert testimony, the jury chose to embrace the most conservative estimate of damages, then the ultimate award might turn out to be less than the proposed settlement. Defendants would appeal any adverse verdicts and those appeals would further delay the resolution of the case. In a case as complex as this one, any appeal would present the plaintiffs with a substantial risk of reversal.
In addition to being risky and complex, this was a lengthy and expensive undertaking for Co-Lead Counsel. Over the past five years, Co-Lead Counsel have put in more than 488,000 hours of attorney time at a market value of over $172 million. They have also incurred more than $29 million in yet-to-be-reimbursed expenses. Continued litigation would drive costs even higher. Thus, risk, complexity, expense, and duration all weigh in favor of approving the proposed settlement.
2. Comparison of the Proposed Settlement with the Likely Result of Continued Litigation
Assuming that all eligible shareholders file claims and that Dr. Zmijewski's event study model accurately characterizes the damages to the class, the proposed settlement represents approximately 27% of the alleged damages to the class. In light of the substantial risk, uncertainty, and delay associated with proceeding to summary judgment and trial, the $3.2 billion settlement amount is an outstanding recovery for the class.
3. Reaction of the Class to the Settlement
The reaction of the class to the settlement has been almost entirely positive. None of the institutional investors have objected to the size of the settlement; indeed, even the institutional investors who objected to other aspects of the settlement at the Fairness Hearing lauded this as an excellent recovery for the class.
Only a small number of individual investors have argued that the settlement should be larger. As discussed in more detail below, however, these objections are based on calculations that overstate the provable damages to the class and are without merit.
4. Stage of the Litigation and the Amount of Discovery Completed
This settlement came after extensive discovery and motion practice. As stated above, plaintiffs reviewed 82.5 million pages of documents and conducted over 220 depositions. At this stage, they have most of the crucial facts in their possession, making them well-positioned to understand the merits of their case. Had the parties not agreed on this settlement, the next steps would be summary judgment and (assuming that plaintiffs survived summary judgment) trial. 5. Quality of Counsel and Conduct During Litigation and Settlement Negotiations
This settlement was the product of arms-length negotiations by highly skilled and diligent counsel on both sides, and Lead Plaintiffs ably discharged their responsibilities to monitor Co-Lead Counsel and ensure that Co-Lead Counsel acted in the best interests of the class. As Judge Sporkin observed in his statement summarizing the settlement negotiations, "the advocacy on both sides of the case was outstanding," and both sides were prepared to try the case if settlement talks failed. Moreover, Judge Sporkin was impressed by "the deep involvement of the class representatives in overseeing the prosecution of the case, and with their commitment to that obligation." I concur with Judge Sporkin's assessments on both points.
In summary, I find that the settlement is fair, reasonable, and adequate.
C. Reasonableness of the Plan of Allocation
Like the settlement itself, the plan of allocation must be fair, reasonable, and adequate. WorldCom, 388 F. Supp. 2d at 344. Co-Lead Counsel have established that the plan of allocation, which compensates class members according to the nature and timing of their Tyco securities transactions, has a reasonable basis. It is calculated according to Dr. Zmijewski's model, and was reviewed by experts R. Alan Miller and Dr. Kenneth D. Gartrell, who concluded both that the underlying methodology was sound and that the distribution plan was fair and reasonable. The fact that these independent experts evaluated the plan of allocation and agreed that it was fair weighs strongly in its favor.
The plan of allocation also deals appropriately with the issue of what to do with excess funds. The plan calls for the continued re-distribution of unclaimed funds to class members according to their pro rata shares, until the costs of such re-distributions make it economically unfeasible to continue doing so. If and when that point is reached, then the balance of the fund will be subject to a cy pres remedy designated by Co-Lead Counsel with the consent of Tyco and PwC. This approach is consistent with the latest draft of the American Law Institute's Principles of the Law of Aggregate Litigation. See American Law Institute,Principles of the Law of Aggregate Litigation, Discussion Draft No. 2, § 3.07 (Apr. 6, 2007) (recommending that a cy pres approach be used only if individual distributions of the surplus to class members are not economically viable or individual class members cannot easily be identified). Although the settlement agreement does not specifically require that I approve the chosen beneficiary of any cy pres remedy, Co-Lead Counsel have assured me that they will submit the planned remedy for my review should a cy pres remedy become necessary, and I will require them to obtain my approval before proceeding with any such plan.
Based on the foregoing, I find that the plan of allocation is fair and reasonable.
D. Objections to the Settlement and Plan of Allocation
None of the institutional investors have objected to either the size of the settlement or the allocation plan. Some individuals, however, have raised objections. I discuss these below.
Some objectors agreed to withdraw their objections after Co-Lead Counsel offered to compensate them for their services to the class from any attorneys' fees award. If I approve these objectors' stipulations of withdrawal, it is because the final resolution is fair to the class, without regard to any compensation the objectors may receive. Nevertheless, Co-Lead Counsel remain free to pay such compensation from the fee award if they so choose, and I need not approve such compensation for it to take place.
1. Objections to the Terms of the Proposed Settlement
Chris Andrews (Doc. No. 1108) objected to the settlement because he disagreed with certain strategic decisions made by Co-Lead Counsel, who had retained Andrews as a consultant to investigate certain factual matters. Andrews has since resolved this dispute with Co-Lead Counsel and has stipulated to the withdrawal of his objection. Had Andrews pursued this objection, I would have overruled it, so I approve his stipulated withdrawal.
James Hill (Doc. No. 1130) objected to the settlement because it could be interpreted to release claims against the Boston Stock Exchange arising from alleged stock market manipulation of Tyco securities traded on that exchange. In response to Hill's concerns, Co-Lead Counsel clarified the terms of the release and Hill withdrew his objection. I approve his stipulated withdrawal.
Peter and Rita Carfagna and Barry Friedman (Doc. No. 1121) objected to the 50/50 split between the class and Tyco for successful prosecution by Tyco of any officer assigned claims. After Co-Lead Counsel clarified the strategic reasons for this decision (namely, the split serves the interests of the class because Tyco has more information regarding the fraud perpetrated by the officer defendants than plaintiffs, and it would be difficult for plaintiffs to collect judgments on the officers' assets), the Carfagnas and Friedman withdrew their objection. I approve their stipulated withdrawal.
James Hayes (Docs. No. 1134 and 1169) argues that the recovery is inadequate because, he alleges, Tyco's stock was artificially inflated by $51 per share at its peak, not the $8.29 per share calculated by Dr. Zmijewski. Hayes' calculation is overly simplistic; although it does adjust for the overall decline in the market, it fails to account for other confounding factors such as the Tyco split plan, Tyco's credit problems, and poor performance by Tyco subsidiaries. Hayes' estimate therefore does not conform to the stringent requirements that plaintiffs would have had to meet to establish loss causation. See Dura, 544 U.S. at 346. Because Hayes' calculation overestimates the provable damages and his objection generally fails to cast doubt on Dr. Zmijewski's event study methodology, I overrule his objection.
Richard and Maryann Wronko (Doc. No. 1132) argue that the recovery is inadequate because it does not fully capture the decrease in value of their Tyco stock and does not include the "likely punitive damages that would have been awarded." The Wronkos' calculation of likely compensatory damages, however, grossly overestimates the potential recovery because it does not account for any possible confounding factors. See Dura, 544 U.S. at 346. Their argument as to punitive damages is problematic because punitive damages are not available for violations of the Securities Exchange Act of 1934. See 15 U.S.C. § 78bb(a) ("no person permitted to maintain a suit for damages under the provisions of this chapter shall recover, through satisfaction of judgment in one or more actions, a total amount in excess of his actual damages on account of the act complained of"); Manchester Mfg. Acquisitions, Inc. v. Sears, Roebuck Co., 802 F. Supp. 595, 605 (D.N.H. 1992). The Wronkos also make two unrelated objections. These objections also have no merit. I therefore overrule the Wronkos' objections. 2. Objections to the Class Period
Specifically, the Wronkos object to the fact that they received the Notice of Proposed Settlement only one week before the deadline for objections, and also that the settlement's termination options give PwC more favorable treatment than Tyco if PwC chooses to withdraw from the settlement.
Joel Douglas (Doc. No. 1105), Richard Dayton (Doc. No. 1115), David Kaiser (Doc. No. 1109), Susan Schaffer (Doc. No. 1138), Marvin and Sonia Greenbaum (Doc. No. 1136), and M.A. Hartley (Doc. No. 1141) objected to the class period, arguing that it should start earlier, end later, and/or include holder claimants. After Co-Lead Counsel explained the basis for the existing class period, all but Kaiser and Hartley withdrew their objections. Kaiser offers no principled reason for modifying the class period, and his proposal to include holder claimants in the class has no legal merit. See Dura, 544 U.S. at 346 (describing the requirements for establishing loss causation); Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 737-38 (1975) (denying standing to holder claimants in private federal securities actions). I therefore overrule his objection. Hartley's objection was untimely and, for the same reasons as Kaiser's, has no legal merit. I therefore overrule Hartley's objection as well. Had Douglas, Dayton, Schaffer, and the Greenbaums pursued their objections, I would have overruled them, so I approve their stipulated withdrawals. 3. Objections to the Plan of Allocation
Robert and Patricia Weinberg (Doc. No. 1110) objected to the cy pres provisions of the settlement because these provisions do not call for input by members of the class. The Weinbergs withdrew their objection after Co-Lead Counsel agreed to consult with Mr. Weinberg regarding what charitable organizations should receive cy pres funds. Weinberg, who is a leader and prominent member of the District of Columbia bar, is well-positioned to offer useful advice to Co-Lead Counsel on this issue. Moreover, because any proposed plan for the distribution of cy pres funds must be submitted to me for approval, there is an additional layer of review protecting the interests of the class if such a distribution becomes necessary. I therefore approve the Weinbergs' stipulated withdrawal.
John Nemfakos (Doc. No. 1137) objects to the different treatment of call options and put options. He also objects to the provision limiting losses from option trading to 1% of the total settlement amount. I overrule both objections. Call options are discounted relative to put options because the purchase of a call option includes a time premium — a wasting asset that will evaporate even if the stock price remains steady. The limitation of losses from option trading is because these derivative securities suffer from much greater volatility than stocks and bonds, making it more difficult to establish loss causation.
A call option is a right to purchase a security at a price and date certain in the future. A put option gives the buyer the right to sell the security to the counter-party at a price and date certain in the future.
4. Miscellaneous Objections
H. Paul Block and Bernice Block (Doc. No. 1156), who assert that they held previously-acquired Tyco stock during the class period but made no transactions during the class period, have filed a late objection to the settlement. Although their objection is untimely, I consider it on the merits. Based on their initial premise that the Securities Litigation Uniform Standards Act ("SLUSA") forces federal law claims to be pre-empted by state law claims, they argue that: (1) holder claimants — although not included as class members — should have standing to object to the proposed settlement; (2) SLUSA should extinguish the federal claims and allow the state claims to move forward; and (3) the proposed settlement should be declared invalid. Their arguments rest on a faulty foundation, however, because their concept of preemption is backward. SLUSA preempts a broad range of state law class-action claims in favor of federal law claims. As the Supreme Court recently held in Merrill Lynch, Pierce, Fenner Smith, Inc. v. Dabit, 547 U.S. 71, 126 S.Ct. 1503, 164 L.Ed.2d 179 (2006), SLUSA does so even when extinguishing those state claims would deny any remedy at all to certain classes of plaintiffs. See id. at 82-89, 126 S.Ct. 1503. For holder claimants like the Blocks, who have no remedy under federal law, this is a harsh result, but the law is clear on this point. See Blue Chip Stamps, 421 U.S. at 737-38 (denying standing to holder claimants in private federal securities actions). Because all three of the Blocks' arguments rely on their incorrect understanding of preemption, none of them have merit. I therefore overrule their objection.
Frank DeCiancio (Doc. No. 1176) filed a late objection to the settlement, asserting that Co-Lead Counsel's calculations of Tyco stock inflation appear illogical. His concerns are adequately addressed by Dr. Zmijewski's detailed description of how he made his damage calculations. Accordingly, I overrule DeCiancio's objection.
E. Attorneys' Fees and Expenses
An attorney who recovers a common fund for the benefit of others is entitled to "a reasonable attorney's fee from the fund as a whole." Boeing Co. v. Van Gemert, 444 U.S. 472, 478 (1980). By assessing attorneys' fees and litigation expenses against a common fund, the court spreads these costs proportionately among those benefitted by the suit. Id. Moreover, providing adequate compensation encourages capable plaintiffs' attorneys to aggressively litigate complex, risky cases like this one rather than settling lower and earlier than would be in the best interests of the class members they represent. See In re GMC Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 801-02 (3d Cir. 1995) (discussing the theoretical foundations for fee awards in class action cases). In this case, Co-Lead Counsel, after negotiations with Lead Plaintiffs, have requested attorneys' fees amounting to 14.5% of the settlement fund. They also seek reimbursement for $28,938,412.74 in expenses. I address the fees and the expenses separately, and conclude that both merit approval.
1. Fees
In line with PSLRA cases in other circuits and past common fund cases in this circuit, I use the percentage of fund ("POF") method with a lodestar cross-check to evaluate the fee request. In re Rite Aid Corp. Sec. Litig., 396 F.3d 294, 305-06 (3d Cir. 2005); United States v. 8.0 Acres of Land, 197 F.3d 24, 33 (1st Cir. 1999); In re Thirteen Appeals Arising out of the San Juan Dupont Plaza Hotel Fire Litig., 56 F.3d 295, 307-08 (1st Cir. 1995). The POF method is appropriate in common fund cases because it "rewards counsel for success and penalizes it [counsel] for failure." GMC Pick-Up Truck, 55 F.3d at 821. Using a lodestar cross-check ensures that the fees are also reasonable in light of the actual amount of work performed. Vizcaino v. Microsoft Corp., 290 F.3d 1043, 1050 (9th Cir. 2002).
Under the POF method, the fee award is calculated as a reasonable percentage of the settlement amount. In re Thirteen Appeals Arising out of the San Juan Dupont Plaza Hotel Fire Litig., 56 F.3d 295, 305 (1st Cir. 1995).
The lodestar ordinarily is calculated by multiplying the number of hours reasonably incurred by the reasonable hourly rate for the services rendered. Gisbrecht v. Barnhart, 535 U.S. 789, 802 (2002).
(a) POF Method
A district court in the First Circuit has "extremely broad" latitude to determine an appropriate fee award under the POF method. Thirteen Appeals, 56 F.3d at 309. Unlike the Second and Third Circuits, the First Circuit does not require courts to examine a fixed laundry list of factors. See id. at 307-09. Cf. Rite Aid, 396 F.3d at 301; Goldberger v. Integrated Res., Inc., 209 F.3d 43, 47 (2d Cir. 2000). I therefore draw loosely on the factors employed by the Second and Third Circuits that are most relevant to my analysis. In particular, I consider: (1) fee awards in similar cases, (2) the complexity, duration, and risk involved in the litigation, (3) the manner in which the fee request was negotiated between Co-Lead Counsel and Lead Plaintiffs, (4) the reaction of the class, and (5) public policy considerations. Based on the totality of these factors, I conclude that the requested 14.5% award is reasonable and appropriate.
(i) Comparison to Other Cases
Co-Lead Counsel argue that I should compare their fee request with the fees that were awarded in connection with the sixteen post-PSLRA securities fraud cases with settlements at or above $400 million. These cases, listed according to the size of the recovery for the class, are: WorldCom ($6.13 billion), Cendant ($2.19 billion), AOL Time Warner, ($2.65 billion), Nortel I ($1.14 billion), Royal Ahold ($1.09 billion), Nortel II ($1.04 billion), McKesson ($960 million), Lucent ($517 million), Bankamerica ($485 million), Dynegy ($474 million), Raytheon ($460 million), Waste Management II ($457 million), Adelphia ($455 million), Global Crossing ($448 million), Freddie Mac ($410 million), and Qwest ($400 million).
See generally In re Nortel Networks Corp. Sec. Litig., No. 01-CV-1855 (RMB) (S.D.N.Y. Jan. 29, 2007) (order and final judgment) (Nortel I); In re Nortel Networks Corp. Sec. Litig., No. 05-MD-1659 (LAP) (S.D.N.Y. Dec. 26, 2006) (order and final judgment) (Nortel II); In re Adelphia Communs. Corp. Secs. Derivative Litig., No. 03 MDL 1529 (LMM), 2006 U.S. Dist. LEXIS 84621 (S.D.N.Y. Nov. 16, 2006); In re Royal Ahold N.V. Sec. ERISA Litig., 461 F. Supp. 2d 383 (D. Md. 2006); Ohio Pub. Employees Ret. Sys. v. Freddie Mac, No. 03-CV-4261 (JES) (S.D.N.Y. Oct. 26, 2006) (order and judgment); AOL Time Warner, 2006 U.S. Dist. LEXIS 78101; In re Qwest Communs. Int'l, Inc. Sec. Litig., No. 01-cv-01451-REB-CBS, 2006 U.S. Dist. LEXIS 71267 (D. Colo. Sept. 28, 2006); In re McKesson HBOC, Inc., Sec. Litig., No. 99-CV-20743 RMW (PVT) (N.D. Ca. Feb. 24, 2006) (order awarding attorney's fees and reimbursement of expenses);WorldCom, 388 F. Supp. 2d 319; In re Dynegy, Inc. Sec. Litig., No. H-02-1571 (S.D. Tex. Jul. 08, 2005) (order awarding attorneys' fees and reimbursement of expenses); In re Raytheon, No. 99-12142-PBS (D. Mass. Dec. 6, 2004) (order and final judgment); In re Global Crossing Sec. ERISA Litig., 225 F.R.D. 436 (S.D.N.Y. 2004); In re Lucent Techs., Inc. Sec. Litig., 327 F. Supp. 2d 426 (D.N.J. 2004); In re Cendant Corp. Sec. Litig., 243 F. Supp. 2d 166 (D.N.J. 2003) (Cendant II); In re Bankamerica Corp. Sec. Litig., 228 F. Supp. 2d 1061 (E.D. Mo. 2002); In re Waste Mgmt., Inc. Sec. Litig., No. H-99-2183 (S.D. Tex. May 10, 2002) (findings of fact and conclusions of law) (Waste Mgmt. II).
Two objectors — SERS/PSERS and U.S. Trust — appeared at the Fairness Hearing and argued that I should limit my comparative analysis to a subset of cases in which the settlements exceeded $1 billion. These "super mega-fund" cases are: WorldCom, Cendant, AOL Time Warner, Nortel I, Royal Ahold, and Nortel II. The objectors contend that the comparison set should be limited to such cases because super mega-fund cases are a distinct subclass in which the size of the recovery is explained more by the size of the class than the work expended by counsel. As a result, the objectors argue, super mega-fund cases — Tyco included — require comparatively lower POF awards to fairly compensate counsel than will be required in cases with smaller settlements.
A comparison of Table 1 with Table 2 explains why this argument is potentially significant. Table 1, which lists the POF awards for the top securities fraud settlements in descending order, reveals that the POF award requested in this case is in line with the POF awards for the other cases in the class (seventh out of seventeen). In contrast, Table 2, which lists the POF awards only in super mega-fund cases, suggests that this case is an outlier when it is compared only with the other cases in the subclass (first out of seven). Thus, it is important to determine whether the objectors are correct in contending that I should limit my comparative analysis to super mega-fund cases. Table 1 Case Name POF Tyco (proposed) 14.50%
Adelphia 21.40% Freddie Mac 20.00% Bankamerica 17.83% Lucent 17.00% Global Crossing 16.18% Qwest 15.00% Royal Ahold 12.00% Raytheon 9.00% Dynegy 8.73% Nortel II 8.00% Waste Mgmt. II 7.93% McKesson 7.79% AOL Time Warner 5.57% WorldCom 5.48% Nortel I 3.00% Cendant 1.73% Table 2 Case Name POF Tyco (proposed) 14.50% Royal Ahold 12.00% Nortel II 8.00% AOL Time Warner 5.57% WorldCom 5.48% Nortel I 3.00% Cendant 1.73% The objectors' contention that super mega-fund cases warrant lower POF awards than smaller cases because they require proportionally less work may well be true as a general matter.See generally In re Prudential Ins. Co. Am. Sales Practice Litig. Agent Actions, 148 F.3d 283, 339 (3d Cir. 1998) ("the basis for this inverse relationship [between the settlement amount and the appropriate POF] is the belief that in many instances the increase [in recovery] is merely a factor of the size of the class and has no direct relationship to the efforts of counsel" (internal quotations omitted)). However, the generalization on which the objectors' argument depends does not hold in this case. The best measure of the effort required to produce a particular result in a given case is the lodestar. In this case, the lodestar expressed as a percentage of the settlement amount is 5.38%. As Appendix 1 demonstrates, this "lodestar percentage" is substantially higher than the lodestar percentages for all but one of the super mega-fund cases and it is much more closely aligned with the lodestar percentages in the larger set of cases that Co-Lead Counsel have proposed for comparison purposes. In other words, whether or not super mega-fund cases generally require proportionally less effort than smaller cases, the generalization is not true in this case. Accordingly, the record does not support the objectors' contention that the proposed POF award should be compared only with the POF awards in super mega-fund cases.In summary, I agree with Co-Lead Counsel that the appropriate set of securities fraud settlements to use for comparison purposes is a set of the sixteen largest settlements rather than a subset of only super mega-fund cases. When I compare the proposed POF award with the POF awards in this set of cases, the proposed POF award does not stand out as unusual.
I do not attach undue significance to this factor. Settlement size is at best a crude indicator of comparability. Each case, regardless of its size, presents its own set of challenges. The work required to resolve the case, the risk of an adverse result, and the quality of the outcome will all vary from case to case. Whether the proposed POF award is reasonable ultimately will depend on an assessment of these largely subjective factors. See Rite Aid, 396 F.3d at 303 (cautioning against "overly formulaic approaches in assessing and determining the amounts and reasonableness of attorneys' fees").
(ii) Complexity, Duration, and Risk
I have already described the factual and legal complexity of this case in explaining why I approved the settlement. The same considerations apply here. This was an enormously complex case, and counsel assumed substantial risk in pursuing it. The number of mergers and acquisitions that were scrutinized and the novelty and difficulty of the legal issues that were presented leave this case with few comparable precedents. Moreover, it is unsurprising that the case took five years to resolve given its difficulty and the fact that plaintiffs were opposed by determined and well-funded adversaries who were represented by highly skilled counsel.
It is noteworthy that three different plaintiffs' firms — entities that generally would not work together unless they found it absolutely necessary — had to work cooperatively, spreading the risk among themselves so that they would have the collective resources required to carry this case through to completion. Co-Lead Counsel, it must be remembered, took this case on a wholly contingent basis. Had they lost on summary judgment or fallen short of establishing liability at trial, they would have lost the tens of millions of dollars in expenses and all of the attorney time that they collectively invested in the case. The $172 million lodestar that Co-Lead Counsel gambled on this case was more than twice as big as in WorldCom ($83 million lodestar), more than four times as big as in AOL Time Warner ($40 million lodestar), and more than twenty-one times as big as in Cendant ($8 million lodestar). It would be inappropriate for me to ignore these differences in evaluating the risk that Co-Lead Counsel assumed in taking on this case.
It is also important to bear in mind that the sheer amount of discovery in this case was staggering. The document production alone — 82.5 million pages — dwarfs every other major securities class action. In comparison, AOL Time Warner involved a mere 15.5 million pages of document production. See AOL Time Warner, 2006 U.S. Dist. LEXIS 78101, at *9. Indeed, Co-Lead Counsel had to review, catalog, and analyze so many documents that, partway through the discovery process, they were forced to hire a technology firm to develop more advanced computerized metrics for sorting through the production.
Accordingly, the unusual complexity, duration, and risk involved in this case are all factors that weigh in favor of the proposed fee award.
(iii) Negotiation of Fee Request
The 14.5% fee request was not unilaterally selected by Co-Lead Counsel. Instead, at the direction of Lead Plaintiffs, Co-Lead Counsel retained two retired judges — Judge Abner J. Mikva (retired Chief Judge of the Court of Appeals for the D.C. Circuit, former White House Counsel, and former Member of Congress) and Judge Alfred M. Wolin (retired U.S. District Judge for the District of New Jersey) — to evaluate Co-Lead Counsel's proposed fee request. Judges Mikva and Wolin carefully considered Lead Plaintiffs' allegations, the duration and procedural history of the case, the amount of completed discovery, the hours spent and expenses incurred by Co-Lead Counsel, and the legal issues at play in the case. Based on their evaluation of this information, Judges Mikva and Wolin concluded that a 14.5% fee would be reasonable. Lead Plaintiffs agreed with this recommendation, Co-Lead Counsel assented to it, and the recommendation became the fee proposal now before me.
This deliberative process is, so far as I know, unique. It is, however, an innovation that is consistent with the spirit of the PSLRA. Under the PSLRA's scheme, the court relies on properly-selected lead plaintiffs to act as agents for the class.See Cendant I, 264 F.3d at 282 (in cases governed by the PSLRA, "lead plaintiff is in the best position, under the PSLRA's scheme, to determine (at least initially) what its lead counsel's fee should be"). The fact that the 14.5% recommendation stems from such a process weighs strongly in favor of its reasonableness.
(iv) Reaction of the Class
Only a tiny percentage of the class has objected to the proposed fee request. Of the 2.4 million Notice recipients, only eleven raised objections, and only four of those objections were filed by institutions. The eleven objections were almost all based on more generalized concerns about the magnitude of the fees and do not require further analysis. To the extent that they raise more specific issues, the objections lack merit, as is discussed in more detail below. Thus, the reaction of the class weighs in favor of approval.
(v) Public Policy Considerations
As I have noted, POF awards generally decrease as the amount of the recovery increases. See Rite Aid, 396 F.3d at 302;Prudential, 148 F.3d at 339. This is because the magnitude of the recovery in many instances is due principally to the size of the class and "has no direct relationship to the efforts of counsel."Prudential, 148 F.3d at 339 (internal quotations omitted); see also Theodore Eisenberg Geoffrey P. Miller, Attorney Fees in Class Action Settlements: An Empirical Study, 1 J. Empirical Legal Stud. 27, 64 (2004) (study of class action fees from 1993 to 2002, finding that as a general rule, the fee percent decreases as client recovery increases, and attributing this pattern to the economies of scale that result from aggregating smaller claims into a single larger action). Other factors may also weigh in favor of a reduced POF award in a particular case. For example, if a settlement is induced by groundwork laid by state regulators and other government entities, rather than the efforts of plaintiffs' counsel, it would be appropriate to reduce the percentage award as the recovery increases. See Prudential, 148 F.3d at 338, 342. Similarly, where a case settles at an early stage, before plaintiffs' counsel have expended hundreds of thousands of hours of work on discovery and motion practice, it is appropriate to reduce the percentage award accordingly. See Cendant II, 243 F. Supp. 2d at 172 (where the case was "a simple case in terms of liability" and settled "at a very early stage, after little formal discovery," fee awarded was 1.73% POF with a 6.875 lodestar multiplier).
In this case, countervailing public policy considerations weigh against any reduction of the POF award. This was an extraordinarily complex and hard-fought case. Co-Lead Counsel put massive resources and effort into the case for five long years, accumulating nearly $29 million in yet-to-be-reimbursed expenses and expending more than 488,000 billable hours (constituting a lodestar of over $172 million) on a wholly contingent basis. But for Co-Lead Counsel's enormous expenditure of time, money, and effort, they would not have been able to negotiate an end result so favorable for the class. Because Co-Lead Counsel's continued, dogged effort over the past five years is a major reason for the magnitude of the recovery, and because this case could not have reached a similarly satisfactory resolution earlier, public policy favors granting counsel an award reflecting that effort.
Without a fee that reflects the risk and effort involved in this litigation, future plaintiffs' attorneys might hesitate to be similarly aggressive and persistent when faced with a similarly complicated, risky case and similarly intransigent defendants. See WorldCom, 388 F. Supp. 2d at 359 ("In order to attract well-qualified plaintiffs' counsel who are able to take a case to trial, and who defendants understand are able and willing to do so, it is necessary to provide appropriate financial incentives."). Of course, not every case needs to proceed as far as this one did to reach a good result for the class. See, e.g.,Cendant II, 243 F. Supp. 2d at 173 ("Lead Counsel's efforts to settle this matter at a relatively early stage has proved to be prescient. . . . waiting to reach settlement might have resulted in many years of delay in the class members' recovery."). But for cases like this one, in which a satisfactory settlement only became possible after years of hard-fought motion practice and searching discovery, it would be against public policy for me to set an unreasonably low POF award that would encourage future plaintiffs' attorneys to settle too early and too low. Additionally, approving this fee award is unlikely to open the floodgates to ever-higher levels of attorney compensation. Few cases will involve the combination of incredible legal and factual complexity, high risk, massive lodestar, and multi-billion-dollar recovery that characterized this case. Accordingly, I find that it would be inappropriate to artificially reduce the percentage award based on the size of the recovery alone.
(b) Lodestar Cross-Check
Several circuit courts have encouraged district judges to use the lodestar method as a cross-check on proposed POF awards. See, e.g., Rite Aid, 396 F.3d at 305; Vizcaino, 290 F.3d at 1043;Goldberger, 209 F.3d at 43. When the lodestar is used in this way, the focus is not on the "necessity and reasonableness of every hour" of the lodestar, but on the broader question of whether the fee award appropriately reflects the degree of time and effort expended by the attorneys. See Thirteen Appeals, 56 F.3d at 307. Such a results-oriented focus "lessens the possibility of collateral disputes [regarding time records] that might transform the fee proceeding into a second major litigation." Id.
In the present case, the lodestar cross-check confirms that the proposed POF award is reasonable. First, I am satisfied that the time charges and hourly rates that were reported in Co-Lead Counsel's fee application are reasonable. I reach this conclusion based on: (1) Co-Lead Counsel's detailed submissions; (2) my familiarity with the work that the case required; (3) the fact that the institutional objectors pointedly declined to challenge either counsels' hourly rates or the time expended (except for certain very specific objections that I resolve below); and (4) Co-Lead Counsel's representation at the Fairness Hearing, which was not contradicted by defense counsel, that Tyco's counsel's lodestar was equal to or higher than Co-Lead Counsel's lodestar.
Second, taking the lodestar amount as an accurate indication of the work that was reasonably required to produce the settlement, the resulting lodestar multiplier of 2.697 appropriately compensates counsel for the risk that they assumed in litigating the case. As Table 3 indicates, the lodestar multiplier in this case is among the lowest of the lodestar multipliers for the top securities fraud settlements (twelfth of seventeen). Because, as I have explained, the risk of an adverse result in this case was higher here than in many of the other large securities fraud cases, the relatively low lodestar multiplier in this case is a good indication that the proposed award is not excessive. Table 3 Case Name Lodestar Multiplier Tyco (proposed) 2.697
In the context in which I use the term in this case, lodestar multiplier is calculated by dividing the fee award by the lodestar amount. In re AT T Corp. Secs. Litig., 455 F.3d 160, 164 (3d Cir. 2006).
2. Objections to Fees
(a) Institutional Objectors
SERS/PSERS and U.S. Trust, joined by the Public Employee Retirement System of Idaho and the West Virginia Investment Management Board, were the only institutional objectors to the fee award. Although the institutional objectors are uniformly pleased with the proposed settlement, they object to the size of the fee award. In addition to their objections to the comparison set of settlements supplied by Co-Lead Counsel, the objectors raise generalized concerns about the danger that a 14.5% award in this case will create a new trend toward ever-higher compensation for plaintiffs' attorneys. This is a legitimate concern — particularly for institutional investors who are likely to be class members in other securities class actions in the future. Nevertheless, as I have described above, the unusual complexity, great legal uncertainty, high risks, massive but justifiable lodestar, high amount of work done in relation to the recovery for the class, and historic magnitude of the recovery in this case put it in a category of its own. Accordingly, the risks that this fee award would drive attorney compensation ever higher in future cases are minimal.
SERS/PSERS also argues that the time spent on the state cases and on seeking Lead Plaintiff and Lead Counsel status should not be counted toward the lodestar because these activities did not benefit the class. This objection lacks merit. First, both activities were of at least some benefit to the fund. See Bankamerica, 228 F. Supp. 2d at 1068 (stating that, to recover fees from a common fund, "attorneys must demonstrate that their services were of some benefit to the fund or enhanced the adversarial process." (quoting Petrovic v. Amoco Oil Co., 200 F.3d 1140, 1156 (8th Cir. 1999))). In particular, the state actions provided additional settlement leverage that Co-Lead Counsel used to negotiate a higher overall recovery from Tyco. The time spent recruiting Lead Plaintiffs also assisted the class because, as Judge Sporkin observed, the Lead Plaintiffs ultimately selected were knowledgeable institutional investors who worked diligently to ensure that Co-Lead Counsel proceeded efficiently and with maximum benefit to the class as a whole. In any event, the time spent on these two sets of activities collectively comprise less than 2% of the lodestar, making them inconsequential for purposes of the lodestar cross-check.
SERS/PSERS argues that the work done by contract attorneys should be treated as an expense to be reimbursed, rather than being included in the lodestar. This objection lacks merit. The lodestar calculation is intended not to reflect the costs incurred by the firm, but to approximate how much the firm would bill a paying client. An attorney, regardless of whether she is an associate with steady employment or a contract attorney whose job ends upon completion of a particular document review project, is still an attorney. It is therefore appropriate to bill a contract attorney's time at market rates and count these time charges toward the lodestar. See Sandoval v. Apfel, 86 F. Supp. 2d 601, 609-11 (N.D. Tex. 2000) (holding that under a fee-shifting statute, the fees of contract attorneys and paralegals are compensable at market rates as part of the attorneys' fees, not just as overhead expenses of the firm); see also Missouri v. Jenkins, 491 U.S. 274, 286-88 (1989) (using the same reasoning to conclude that market-rate billing of paralegal hours should count toward an attorney fee award).
Alternatively, SERS/PSERS argues that the lodestar contributions of contract attorneys were excessive, in that much of the work done by contract attorneys should have been performed by lower-billing paralegals. First, as the Supreme Court has noted, there are many types of work that lie "in a gray area of tasks that might appropriately be performed either by an attorney or a paralegal." Jenkins, 491 U.S. at 288 n. 10. Depending on the particular circumstances, it may or may not be cost-efficient to preclude attorneys from doing such work, particularly if it is intermingled with work that only an attorney can perform. Second and more importantly, even if I assumed that all of the contract attorney work done on behalf of Milberg Weiss LLP and Schiffrin Barroway Topaz Kessler LLP (the two heaviest users of contract attorneys in this case) should have been billed at legal assistant rates, this would reduce the overall lodestar by only about 12.5%, shifting the lodestar multiplier from 2.697 to 3.083. This minor change would not make any practical difference in the lodestar cross-check. See Rite Aid, 396 F.3d at 306 ("The lodestar cross-check calculation need entail neither mathematical precision nor bean-counting."); Thirteen Appeals, 56 F.3d at 307 (the lodestar cross-check does not demand that the judge "determine the necessity and reasonableness of every hour expended"). Accordingly, the objection lacks merit.
Based on my review of the firms' lodestar calculations, legal assistants are billed at about two-thirds of the average rate for contract attorneys.
(b) Individual Objectors
Chris Andrews (Doc. No. 1108), Carfagnas/Friedman (Doc. No. 1121), Lynne Sell (Doc. No. 1126), Vondell Tyler and Ernest J. Browne (Doc. No. 1133), Rinis Travel Service, Inc. (Doc. No 1136), and the Weinbergs (Doc. No. 1110) objected to the fee award based on generalized concerns that the fees were too high. Some also sought assurances that Co-Lead Counsel would continue to be involved as necessary after the settlement is concluded. All of these objections were withdrawn after Co-Lead Counsel provided further assurances to the objectors and informed the objectors of the reduction in the fee request from the 17.5% fee described in the Notice of Proposed Settlement to the 14.5% fee that is now being sought. In light of these assurances, I find that there is no need to address these withdrawn objections further.
Charles L. Glass is the only individual objector to the fee award who has not withdrawn his objection. His objection consists entirely of generalized concerns about the magnitude of the award that are addressed by my discussion above. I therefore overrule his objection.
(c) Allegations by Phillip Crawford
Phillip Crawford, Jr. has made numerous allegations in many separate letters addressed to this Court (Docs. No. 1161, 1163, 1165, 1166, 1167, 1168, and 1175). Although not a class member, Crawford was a contract attorney employed by Peak Counsel, which conducted document review for Milberg Weiss LLP, one of the three firms that acted as Co-Lead Counsel in this case. Most of his allegations are either clearly irrelevant or clearly incorrect, and need not be addressed. Crawford's two remaining allegations are: first, that the fee award is inappropriate because Milberg Weiss allegedly failed to police overbilling by its contract attorneys, and second, that some of the work done by contract attorneys could have been performed by non-lawyers and therefore should have been billed at a lower rate.
Neither allegation has merit. Regarding Crawford's first allegation, Milberg Weiss has submitted affidavits from the relevant personnel at their firm. Together, these affidavits establish that Milberg Weiss conducted a thorough investigation of Crawford's allegations of overbilling by contract attorneys, removed from their fee application all time billed by both Crawford and the one contract attorney whose records were apparently inconsistent, and instituted appropriate procedures to police the billing of contract attorneys. These affidavits satisfy me that Milberg Weiss took all reasonable measures to ensure that their fee application accurately reflected the work put into the case. As for Crawford's second allegation, it fails for the same reason as SERS/PSERS's similar objection, supra. Even if true, his second allegation would not decrease the lodestar enough to call into question the appropriateness of either the lodestar multiplier or the lodestar percentage cross-check.
In summary, none of the objections made to the proposed fee award cause me to question my determination that the proposed award is reasonable.
3. Expenses
Co-Lead Counsel have requested reimbursement of $28,938,412.74 in expenses. In the exhibits to their fee and expense request, Co-Lead Counsel have provided detailed breakdowns of their expenses. They have also provided summary tables, breaking the expenses down by category. They are not seeking reimbursement for computer research charges, overtime, secretarial services, rental space related to document review, supplies, press releases, or certain other miscellaneous expenses. I find that Co-Lead Counsel have provided sufficient documentation of their expenses and that, in light of the legitimate needs arising from the size and complexity of this case, the expense request is reasonable. See In Re San Juan Dupont Plaza Hotel Fire Litig., 111 F.3d 220, 233-38 (1st Cir. 1997). Accordingly, I approve reimbursement of the requested amount.
4. Objections to Expenses
Three class members objected to the expense request. Andrews (Doc. No. 1108) objected mainly based on his belief that the case should have been settled earlier and higher, before Co-Lead Counsel incurred significant expenses. Carfagnas/Friedman (Doc. No. 1121) objected to the reimbursement of certain categories of expenses. Rinis Travel (Doc. No. 1135) objected to the lack of detail in the initial expense request. All of these objections have since been withdrawn. Moreover, to the extent these objections may have had merit, they were mooted by Co-Lead Counsel's decision not to pursue reimbursement for certain expenses and their provision of a detailed, well-organized breakdown of expenses in their fee and expense request.
III. CONCLUSION
For the foregoing reasons, I:
(1) Overrule the objections to the proposed settlement, plan of allocation, and award of attorney's fees and expenses;
(2) Approve the Settlement Agreement (consisting of the terms and conditions of the Stipulation of Settlement dated July 6, 2007, including Amendment No. 1 as filed with the Court on July 12, 2007 and Amendment No. 2 as filed with the Court on October 24, 2007) and the plan of allocation; and
(3) Approve attorneys' fees of 14.5% of the settlement, plus reimbursement of $28,938,412.74 in expenses.
A more detailed final judgment will issue along with this Memorandum and Order.
SO ORDERED.
APPENDIX A Lodestar Percentages for Tyco and the Previous Top Securities Class Action Settlements
ExhibitAPPENDIX B Lodestar Percentage vs. POF for Tyco and the Previous Top Securities Class Action Settlements
The regression line in the above plot was calculated according to the previous top sixteen securities settlements, not includingTyco. This regression line has a coefficient of determination (R2) of 0.8768, indicating that the data is well-fitted to the line. The Pearson's correlation coefficient ® for lodestar percentage and POF (again, for the previous top sixteen securities settlements, not including Tyco) is 0.936, indicating that the two variables are strongly correlated with one another.