In Re Educational Testing Service Praxis Principles of Learning & Teaching: Grades 7-12 Litigation , 447 F. Supp. 2d 612 ( 2006 )

447 F.Supp.2d 612 (2006)

In re EDUCATIONAL TESTING SERVICE PRAXIS PRINCIPLES OF LEARNING AND TEACHING: GRADES 7-12 LITIGATION.

MDL No. 1643.

United States District Court, E.D. Louisiana.

August 31, 2006.

A. Stephen Hut, Jr., Wilmer Cutler Pickering Hale and Dorr LLP, Washington, DC, Keith M. Pyburn, Jr., Fisher & Phillips, LLP, New Orleans, LA, for Educational Testing Service Praxis Principles of Learning and Teaching, Grades 7-12 Litigation.

Jeffery N. Luthi, Washington, DC, pro se.

FINAL APPROVAL OF SETTLEMENT ORDER

VANCE, District Judge.

Before the Court are settlement class counsel's motion for final approval of the class action settlement and for an award of attorneys' fees, costs and incentive payments. The Court has considered all of the evidence submitted at the fairness hearing held on July 12, 2006, as well as the objections and legal memoranda submitted by the parties. For the reasons stated more fully below, the Court finds the settlement of this class action to be fair, reasonable and adequate, and the Court awards attorneys' fees, costs and incentive payments as provided in this order.

I. BACKGROUND

A. Factual Background

Defendant Educational Testing Service is a not-for-profit corporation that designs administers, and scores a wide range of standardized educational tests. ETS is the world's largest private educational testing organization, and it administers over 12 million examinations annually.

Among the educational tests that ETS designs, administers and scores are the Praxis Series examinations. The Praxis tests are a series of tests used by many states in the teacher licensing process. Passage of one or more of the Praxis tests is required for licensing in 39 states and U.S. jurisdictions. The Praxis tests are administered six times per year, at 650 test centers in all 50 states.

Plaintiffs sued ETS over the administration of only one of the Praxis series of tests—the Praxis Principles of Learning and Teaching: Grades 7-12 (the "PLT: 7-12") test. During 2003-2004, 19 states contracted with ETS to use the PLT: 7-12 as part of their teacher licensing process. The PLT: 7-12 is also relevant to colleges, universities and professional organizations as a measure of teaching credentials. Individuals who take the PLT: 7-12 can request that ETS report their scores directly to various states, colleges and universities.

ETS incorrectly scored the PLT: 7-12 over the course of nine test administrations between January 2003 and April 2004. As a result of this error, approximately 27,000 people who took the PLT: 7-12 during that time period received a score that was lower than it would have been had the exams been graded properly. The scoring error caused about 4,100 test takers to receive a "false failure," i.e., they were notified that they had received a failing score in at least one state in which their score was reported, when in fact they should have received a passing score. The scoring error caused approximately 23,000 other test takers to receive passing scores that were lower than they would have been had the tests been scored correctly.

After a client state questioned ETS about scoring results for the PLT: 7-12, ETS began an investigation that ultimately led it to discover the scoring error. On or about July 10, 2004, ETS began to notify affected test takers by telephone and letter that they had incorrectly been told that they had failed the PLT: 7-12 test when, after rescoring, they had actually passed the test. Although ETS re-scored all of the PLT: 7-12 tests taken during the relevant period, it provided the adjusted scores to only those people who had received false failures. ETS did not provide the rescored results to test takers whose initially-reported score was sufficient to pass the exam in all of the states to which it was reported.

As a result of the scoring error, the plaintiffs in this litigation allege that many test takers who received a false failing score were unable to obtain, or to obtain in a timely manner, their teaching credentials, and they therefore could not secure or retain employment as certified teachers. Plaintiffs also allege that the scoring error delayed some test takers' completion of bachelor's and/or master's degrees and that it caused some test takers to abandon teaching and pursue alternate majors and careers. In addition, many of those test takers allegedly retook the PLT: 7-12 and in the process incurred additional registration fees and test preparation expenses. Further, plaintiffs allege that some test takers who initially received a passing score in each of the jurisdictions to which their score was reported were harmed by having artificially low scores reported to states and educational institutions.

B. Procedural Background

On December 16, 2004, the Judicial Panel on Multidistrict Litigation transferred nine actions relating to the PLT: 7-12 scoring error from federal district courts in Pennsylvania, Louisiana and Ohio to this Court under 28 U.S.C. § 1407 for consolidated or coordinated pretrial proceedings. These cases joined four existing cases already in the Eastern District of Louisiana. On January 4, 2005, the Court consolidated all actions for pretrial purposes. A number of additional actions were later consolidated with those cases, and a total of 29 actions are currently consolidated before this Court. On January 24, 2005, the Court appointed Dawn Barrios, of Barrios, Kingsdorf, & Casteix, L.L.P., as plaintiffs' lead counsel, and Richard Arsenault of Neblett, Beard, & Arsenault, as plaintiffs' liaison counsel. The Court directed plaintiffs to file a master complaint by March 10, 2005. The Court further directed the parties to make voluntary Rule 26(a) disclosures by March 10, 2005, and to serve and respond to interrogatories and document requests by April 11, 2005, and May 11, 2005, respectively.

On March 9, 2005, the parties submitted a Joint Report detailing their preparations for discovery. The following day, plaintiffs submitted an Administrative Master Complaint bringing claims under four theories: (1) breach of contract, (2) negligence, (3) negligent misrepresentation, and (4) a violation of § 2 of the Sherman Act. ETS moved to dismiss plaintiffs' state law claims for negligence and negligent misrepresentation, as well as plaintiffs' requests for emotional distress and punitive damages on both their contract and tort claims. ETS also moved to dismiss plaintiffs' claims under the Sherman Act. Plaintiffs opposed the motion to dismiss, arguing that the claims were proper under the applicable state contract and tort laws, as well as under the Sherman Act.

On June 3, 2005, the Court issued Pretrial Order # 4, directing the parties to brief the choice of law as to each state law claim at issue in ETS's pending motion to dismiss. The Court further limited plaintiffs' depositions to 16 and ETS's depositions to 25, including five unnamed class members. On June 24, 2005, the Court issued Pretrial Order # 5 detailing the schedule for discovery over the succeeding months.

Also on June 24, 2005, plaintiff Michelle Kochensky filed a Second Amended and Restated Class Action Complaint, containing segregated claims on behalf of Pennsylvania plaintiffs for Pennsylvania Unfair Trade Practices Act violations, breach of fiduciary duty and civil rights violations, as well as the same claims asserted in the Administrative Master Complaint, except for the Sherman Act claim. On July 12, 2005, ETS filed a motion to dismiss the segregated claims.

On August 29, 2005, Hurricane Katrina struck New Orleans. On December 1, 2005, the Court granted ETS's motion for partial dismissal as to the Sherman Act claim. While ETS's other motions to dismiss remained pending, the parties notified to the Court that they had reached a tentative settlement. On February 10, 2006, plaintiffs moved for preliminary approval of the settlement and for certification of a class for the purpose of settlement. The Court held a hearing on the motions on February 22, 2006. On March 13, 2006, the Court certified a settlement class, appointed class counsel, and preliminarily approved the proposed settlement. The Court at that time appointed Patrick Juneau as Special Master, appointed Bourgeois Bennett, LLC, CPAs to serve as Court Appointed Disbursing Agent (CADA), and approved the proposed Notice and Claim Forms. The Court approved the Special Master's hourly rates for services to be rendered on behalf of the class. The Court also approved deadlines of June 3, 2006 for exclusions (or opt-outs), and July 3, 2006 for objections and notices of intention to appear at the fairness hearing. Finally, the Court scheduled the fairness hearing on the settlement for July 12, 2006.

On May 23, 2006, Kochensky filed a motion to compel discovery by class counsel, in connection with her evaluation of the settlement. Class counsel opposed the motion on both substantive and procedural grounds. On June 12, 2006, the Court, for the most part, denied the motion to compel discovery.

On June 27, 2006, plaintiffs filed a motion for final approval of the settlement. Class counsel also filed a motion for attorneys' fees and costs, and for approval of an incentive award for class representatives. The Court received objections to the settlement from plaintiffs Edmund Heustis, Michelle Kochensky and Brian Welsh. On July 12, 2006, the Court held a fairness hearing to determine the appropriateness of the proposed settlement and to provide for attorneys' fees, costs and incentive payments to class representatives.

C. The Settlement Class, Class Representatives and Class Counsel

On March 13, 2006, the Court certified a class for purposes of settlement. The class consists of the following persons:

All persons who took the Praxis Principles of Learning and Teaching: Grades 7-12 examination between January 1, 2003 and April 30, 2004. Specifically excluded from the Class are all persons who have executed full and final releases of their Causes of Action with ETS.

The Court also appointed four plaintiffs to serve as class representatives: Kathleen Jones, Paul Perrea, Raffael Billet, and Janet Riehle. Each of these plaintiffs took the PLT: 7-12 during the relevant time period and received a false failure because of the scoring error. Moreover, each of the proposed class representatives was a named plaintiff in an action against ETS concerning the scoring error. The Court also appointed Dawn Barrios of Barrios, Kingsdorf, & Casteix, L.L.P.; Richard Arsenault of Neblett, Beard & Arsenault; Philip Bohrer of The Bohrer Law Firm, L.L.C.; Phyllis Brown of The Law Offices of Phyllis Brown; Sherrie Savett of Berger & Montague; Steven Bell, of Counsel to the Simon Law Firm; Walter Leger of Leger & Mestayer; and Joseph Bruno of Bruno & Bruno as class counsel.

D. The Settlement Agreement

Under the terms of the proposed Settlement Agreement, ETS would pay $11.1 million into a settlement fund for the benefit of participating class members who suffered damages as a result of the scoring error. This would come in two installments: one for $250,000.00, which was to be made shortly after the Preliminary Approval Order, and, the other for $10,850,000.00, which would be made within seven days of the Court's entry of a final order and judgment. The Agreement also provides that ETS would furnish a free score report, valued at $35, to any class member who did not receive a false failure.

The Agreement provides various duties for the Special Master. These duties include to (1) establish, conduct and manage the claims administration process; (2) create claim forms; (3) create the formulation of fair, equitable and reasonable procedures for the claims process, the allocation of the settlement to participating class members, and distribution of the settlement fund; (4) evaluate all claims and distribute the settlement fund in accord with the criteria established; (5) recommend to the Court the necessary reserves from the settlement fund to effectuate adequate and reasonable management of the allocation process, including fee and cost reserves; (6) conduct hearings of participating class members' allocation objections and allocation protocol; (7) submit to the Court a report on the above, along with recommendations for the Court's consideration in proceeding with the allocation and distribution process following the effective date; and (8) assist the Court and settlement class counsel, as necessary and in accordance with the Agreement. The Agreement further gives the Special Master authority to implement reasonable procedures to prevent payment of fraudulent claims and otherwise to assure an acceptable level of reliability and quality control in claims processing. The Agreement provides for the CADA to assist the Special Master in maintaining records, preparing and filing tax returns, managing the financial aspects of claims, fees, costs and expenses, the computerized generation of data regarding evaluation of claims, and disbursement and administration of the settlement fund.

The Agreement requires the Special Master to provide a location for class members to send proofs of claim, or claim forms. The Agreement authorizes three separate claim forms: (1) an individualized damage claim form for all class members to be used in claiming any itemized damages caused by the inaccurate test scores; (2) an expedited payment claim form for class members who received a false failure and choose to request a payment of $500 as an alternative to requesting individualized relief; (3) a score report claim form for class members who did not receive a false failure and who wish to request a free score report, valued at $35. The Court's Preliminary Approval Order set the deadline for submission of these claim forms of July 3, 2006.

After evaluation of the submitted claim forms, the Special Master is to determine eligibility for, and the amount of, individual allocations. The Agreement forbids the Special Master to consider the amount of the settlement fund in making these individual determinations. If the total sums awarded to participating class members exceed the available funds, the awards are to be reduced and paid on a pro-rata basis. The Special Master is to prepare a report and recommendation regarding his methodology and allocation of settlement funds, detailing the individualized allocation process. No disbursement may be made without approval of this report by the Court. At the completion of the allocation process, any residual sums either may be reallocated and distributed to participating class members by the Special Master or be subject to a cy pres award. No amount of the settlement fund will revert to ETS.

The Settlement Agreement also indicates class counsel's intention to apply for attorneys' fees not to exceed 40% of the settlement fund, and incentive awards for class representatives not to exceed $2,000.00 each. Because these provisions are subject to the discretion of the Court, and not the agreement of the parties, the Court discusses them separately from the Settlement Agreement, infra.

E. Notice

At the fairness hearing, the Court received testimony from the Notice Administrator, Todd Hilsee, who described the forms and procedure used to notify class members of the proposed settlement and their rights with respect to it. The Notice Administrator testified that the notice was mailed and actually delivered to 96% of the approximately 27,000 class members, which, according to Hilsee, was one of the highest success rates with direct mail he had ever been able to achieve. The Notice Administrator also described the other steps taken to give notice, including designing the notice forms in clear, plain language, according to the sample forms developed by the Federal Judicial Center; developing a neutral web site containing information about the proposed settlement; issuing a national press release about the proposed settlement; and providing numerous educational associations, state departments of education, and teachers' associations with information about the settlement. The Court is satisfied that notice to the class fully complied with the requirements of Rule 23.

F. Settlement Administration

After the Court appointed the Special Master and the CADA, the Special Master prepared a report that recommended that 2.5% of the settlement fund be reserved for litigation costs; 7% of the settlement fund be reserved for settlement administrative costs and taxes; and 40% of the settlement fund be reserved for attorneys' fees. The Special Master and the CADA then began to disburse funds from the $250,000 initial payment, in order to administer the settlement account, send notice to the class members, and receive claim forms from participating class members. As of July 10, 2006, the total amount of those disbursements was $132,334.89. This included the fees and costs of the CADA, the Notice Administrator and the Special Master, as well as copying and mailing costs and bank fees.

On March 27, 2006, the claims period opened. As of the July 22, 2006 revised cut-off date^[1] for the receipt of claim forms, the Special Master received 4,310 total claim forms, broken down by category, as follows:

Total Claims:                             4,310
Total False Failures:                     2,026
Expedited Claim Forms:                 1,365
Individualized Claim Forms:              661
Total Non—False Failures:           2,284
Rescores Only:                         1,710
Individualized Claim Forms:              105
Individualized Claim Forms
with Rescore Requests:                 469

In addition to claims forms, the settlement administrators received eight opt-outs. Shortly thereafter, four of the opt-outs rescinded their exclusions, leaving four opt-outs in a class of about 27,000.

G. Objections

The Court received three purported objections. Each of the objections was filed after the deadline stated in the order preliminarily approving the settlement. The order required objections to be filed on or before ten days before the fairness hearing.^[2] The detailed notice disseminated to Class Members, however, indicated that objections must be postmarked no later than July 3, 2006. The Court found at the hearing that the objections were timely postmarked in accordance with the class notice and that they could be considered on the merits.

On the morning of the fairness hearing, objector Heustis sought to withdraw his objection to the settlement. In order to guard against side agreements, the Court must give its approval before an objector may withdraw an objection to a class action settlement. See Fed.R.Civ.P. 23(e)(4)(B). After reviewing his objection and the document seeking withdrawal, the Court found that this "objection" was actually only a request for clarification. The Court therefore approved the withdrawal of the objection at the fairness hearing.

The remaining objectors were represented by the same attorney, who was given the opportunity to be heard in support of the objections at the fairness hearing and to cross-examine the witnesses called by the settlement proponents.

II. CLASS ACTION SETTLEMENT

A. Legal Standard—Fair, Reasonable and Adequate

Federal Rule of Civil Procedure 23 governs the settlement of class actions. See Henderson v. Eaton, 2002 WL 31415728, *2 (E.D.La.2002) (citing Pearson v. Ecological Science Corp., 522 F.2d 171, 176-77 (5th Cir.1975)). A class action may not be dismissed or compromised without the district court's approval. See Fed.R.Civ.P. 23(e). Before a court approves a settlement, the court must find that the proposed settlement is "fair, reasonable, and adequate." Fed.R.Civ.P. 23(e)(1)(c); see also Cope v. Duggins, 203 F.Supp.2d 650, 652-53 (E.D.La.2002) (citing Cotton v. Hinton, 559 F.2d 1326, 1330 (5th Cir. 1977)).

The Court must "ensure that the settlement is in the interest of the class, does not unfairly impinge on the rights and interests of dissenters, and does not merely mantle oppression." Reed v. General Motors Corp., 703 F.2d 170, 172 (5th Cir.1983) (quoting Pettway v. American Cast Iron Pipe Co., 576 F.2d 1157, 1214 (5th Cir.1978)). Because the parties' interests are aligned in favor of a settlement, the Court must take independent steps to ensure fairness in the absence of adversarial proceedings. Reynolds v. Beneficial Nat'l Bank, 288 F.3d 277, 279-80 (7th Cir. 2002) (noting that the class action context "requires district judges to exercise the highest degree of vigilance in scrutinizing proposed settlements."); see also Manual for Complex Litigation (Fourth) § 21.61 (2004). The Court's duty of vigilance does not, however, authorize it to try the case in the settlement hearings. Cotton, 559 F.2d at 1330.

The Fifth Circuit has identified six factors that the Court should consider in assessing whether a settlement is fair, adequate, and reasonable: (1) the existence of fraud or collusion behind the settlement; (2) the complexity, expense, and likely duration of the litigation; (3) the stage of the proceedings and the amount of discovery completed; (4) the factual and legal obstacles to prevailing on the merits; (5) the range of possible recovery; and (6) the opinions of the class counsel, class representatives, and absent class members. See Reed, 703 F.2d at 172 (citing Parker v. Anderson, 667 F.2d 1204, 1209 (5th Cir. 1982)). In considering these factors, there is a strong presumption in favor of finding the settlement fair. See Cotton, 559 F.2d at 1331 ("Particularly in class action suits, there is an overriding public interest in favor of settlement.") (citing United States v. Allegheny-Ludlum Industries, Inc., 517 F.2d 826 (5th Cir.1975)).

B. Discussion

The Court considers each of the applicable factors below.

Fraud or Collusion

There is no evidence that the settlement involves fraud or collusion. The parties indicated that the amount of the settlement was the largest to date in a testing case against ETS. In addition, the parties presented evidence of their arm's-length negotiations, including the testimony of the mediator who facilitated the settlement. Further, the objectors do not allege fraud or collusion. In the absence of any indication that the settlement is fraudulent or collusive, this factor favors approval of the settlement.

Complexity, expense, and likely duration of the litigation

This case involved a scoring error that occurred over a 16-month period, which inflicted damage on a discrete number of individuals. The resulting contract and tort claims were not in and of themselves particularly complex, nor was the transaction between the parties document-intensive. Further, the defendant all but admitted nonperformance of the contract in question. In this sense, the case was not one of sprawling complexity.

Nevertheless, application of this factor favors approval of the settlement. In this context, the Court considers whether settling now avoids the risks and burdens of potentially protracted litigation. Ayers v. Thompson, 358 F.3d 356, 369 (5th Cir. 2004). Here, the nature of the claims suggested that the amount of individual damages would be relatively small, making the cost of litigating individual claims potentially prohibitive. Yet, the issues involved in class certification threatened to delay the case at great expense and/or to result in a denial of certification. This follows because, in deciding whether to certify a class for trial, the Court would have had to consider "whether the case, if tried, would present intractable management problems." See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). Because plaintiffs sought to certify a nationwide class that potentially implicated the laws of many states, it is not at all clear that this case would have survived such an inquiry. Further, resolution of the threshold question of which state laws to apply would have delayed resolution of the certification issue.

In addition, the defendant's strategy on the class certification front was to admit nonperformance of the contract and to argue that the only issues to be tried were individual causation and individual damages. Hence, the argument goes, class certification should fail because individual issues would predominate. This factor also posed the risk that protracted certification proceedings would result not just in delays but also in a denial of certification, which could make the expense of litigating individual claims prohibitive.

Finally, the defendant undoubtedly would have sought an immediate appeal of an order certifying a class for trial. Under Fed.R.Civ.P. 23(f), the Court of Appeals has discretion to grant or deny such an appeal. If the Court of Appeals granted an appeal, the litigation could be delayed for one or two years during the pendency of the appeal. Further, the Court of Appeals may well have decertified the class. If the class certification order stood, the parties would still be a year or more away from a trial on the merits. Under the circumstances, an early settlement that obviated these delays and the risks of having to litigate the claims makes sense. Accordingly, consideration of the second factor favors approval of the settlement.

Stage of the proceedings and amount of discovery completed

This factor asks whether the parties have obtained sufficient information "to evaluate the merits of the competing positions." Ayers, 358 F.3d at 369. Thus, the question is not whether the parties have completed a particular amount of discovery, but whether the parties have obtained sufficient information about the strengths and weaknesses of their respective cases to make a reasoned judgment about the desirability of settling the case on the terms proposed or continuing to litigate it. In re Train Derailment Near Amite, La., 2006 WL 1561470 at *22 (E.D.La.2006) (citing In re Combustion, Inc., 968 F.Supp. 1116, 1127 (W.D.La. 1997)). The Fifth Circuit has explicitly rejected the proposition that the parties must always undertake extensive formal discovery and build a voluminous record before settlement. Cotton, 559 F.2d at 1332; see also In re Corrugated Container Antitrust Litigation, 643 F.2d 195, 211 (5th Cir.1981). Rather, when the settlement proponents have taken affirmative steps to gather data on the claims at issue, and the terms of the settlement or settlement negotiations are not patently unfair, the Court may rely on counsel's judgment that the information gathered was enough to support a settlement. In re Corrugated Container, 643 F.2d at 211.

The parties stated at the fairness hearing that, although they had not taken any depositions, they had participated in sufficient written discovery, both formal and informal, as well as motion practice, to assess the strengths and weaknesses of their own and their opponents' cases. Plaintiffs' counsel reviewed 60,000 documents. In addition, the parties retained experts and consultants, including statisticians, psychometricians, and CPAs, to assist them in evaluating their claims and developing damage models. It was no secret to either the plaintiffs or the defendant that plaintiffs had a strong case on contractual nonperformance. Further, motion practice showed that plaintiffs faced significant legal issues concerning whether non-economic damages were available and whether the class was certifiable. ETS also informally provided information on the test taker population at issue that shed light on the potential number of claimants who could not mount claims for lost income for one reason or another. See Range of Possible Recovery, infra.

The Court is satisfied that the parties were sufficiently informed to assess the strengths and weaknesses of their positions and to make a reasoned evaluation of whether and on what terms to settle. Accordingly, the Court finds that application of this factor strongly favors settlement.

Factual and legal obstacles to prevailing on the merits

The Court has already discussed the risks involved in class certification and the importance of class certification to the plaintiffs' ability to pursue their claims on the merits efficiently. As to the plaintiffs' substantive claims, plaintiffs had a strong chance of proving a breach of contract. Nevertheless, there was a significant risk that the plaintiffs would not be able to recover non-contract damages or for noneconomic losses. First, although the plaintiffs sought punitive damages, the law on punitive damages for the type of conduct at issue was unfavorable to the class. Indeed, most of the applicable state laws require much more egregious conduct than negligence, and usually demand proof of malice or recklessness, to recover punitive damages. See, e.g., Hutchison v. Luddy, 582 Pa. 114, 870 A.2d 766, 770 (2005) (holding that a defendant's actions must have been intentional, malicious, or reckless to merit punitive damages); Oh. Rev.Code § 2315.21(c)(1) (holding that a defendant must have acted with malice or aggravated or egregious fraud to merit punitive damages); Hodges v. S.C. Toof & Co., 833 S.W.2d 896, 901 (Tenn.1992) (holding that a defendant must have acted intentionally, fraudulently, maliciously, or recklessly to merit punitive damages). As to other forms of non-contract damages or noneconomic losses, ETS raised serious issues concerning whether these damages would be foreclosed under the laws of the states at issue because of the economic loss rule, the absence of a duty independent of the contract, and/or the absence of physical injury. In Arnold v. ETS, a testing case resolved under the laws of New Jersey and Georgia, plaintiffs' claims for non-economic damages did not survive a motion to dismiss. Arnold, No. Mer-L-2856-04, Mot. Tr. (N.J.Super.Ct. March 18, 2005). The court reasoned that the defendant owed no duty to the plaintiff independent of its contractual duty and that the case involved a simple breach of contract. Id. The court further noted that New Jersey law does not permit the recovery of punitive damages or mental anguish for basic breach of contract claims, and it dismissed those claims as well. Id. Whether or not, as the settlement proponents suggest, other courts would have found New Jersey law controlling because it is the site of ETS's headquarters, other courts may well have found the reasoning in Arnold persuasive.

Even if the plaintiffs surmounted the legal impediments to the recovery of noncontract or non-economic damages, there were also serious factual issues to be overcome to prevail on those claims. As for the 23,000 people who did not receive a false failure, but simply received a lower than actual score, the ability to demonstrate credibly that the plaintiff suffered emotional distress caused by the scoring error would be no small problem. As to economic losses, plaintiffs who did not receive a false failure would face an uphill battle to demonstrate a loss of employment when they passed the test. The probability of such a recovery would be low at best.

As class counsel admitted at the fairness hearing, plaintiffs who received a false failure faced causation and mitigation issues that would have been obstacles for many. This follows because this group included test takers who were college undergraduates who lacked the credentials to teach, test takers who retook the test and passed in time not to lose an employment opportunity, and/or who found or could have found other jobs at equal or higher pay. Considering the legal and factual issues facing plaintiffs on class certification, non-contract liability, causation, and damages, the probability of success factor favors approval of the settlement.

Range of possible recovery

In considering the range of possible recovery, the Court need not consider recoveries that are beyond the range of the most minimal probability. Thus, engaging in an exercise that posits on the high end a recovery in which all class members would recover significant lost income, mental anguish and punitive damages is too unrealistic to be useful. The 23,000^[3] non-false failures would have severe difficulties proving damages. These individuals would be unlikely to show that their score resulted in the loss of employment opportunities because they passed the exam. Similarly, putting aside the legal issues confronting claims for non-economic damages, serious proof problems are likely to confront plaintiffs who claim they suffered emotional distress because their passing scores should have been higher. The most obvious loss to the non-false failure plaintiffs is the cost to find out their correct scores, which has a value of $35 per score report.

As to the 4,070 false failures, all of these test takers were notified that they had actually passed the PLT by July of 2004. For this reason, the outer limit of potential income loss for these individuals could not realistically extend beyond the 2004-05 school year. Of these false failures, about one-half took the test in 2003 and one-half took the test in 2004. It is clear that most of the 2003 false failures could not show two years of lost income. Almost 54% (1,312) of these false failures occurred at tests administered in the last five months of 2003, which was too late for the hiring cycle for the 2003-04 school year. Almost 43% (489) of the false failures who originally took the test in the first six months of 2003 retook the test and passed before ETS notified test takers of the scoring error in July 2004. They, too, would be extremely unlikely to be able to show two years of lost income. These numbers suggest that about 73% (1,801) of the 2003 false failures could not show two years of lost income.

As to the one-half of the false failures that occurred in 2004, the outer limit of their lost income claims is most likely one year, since the error was reported in July 2004. The evidence at the hearing suggested, however, that a substantial portion of these individuals could not show lost income at all. About a quarter of the 2004 test takers had already retaken the test and received a passing score before ETS notified them of the scoring error in July 2004. The likelihood that they could prove a year's lost wages is not high. The parties represented that the other 2004 test takers would also have difficulty proving any lost income to the extent that the 2004-05 hiring cycle was still open after ETS disclosed the scoring error in July 2004.

Further, the false failures include class members who could not show lost income for other reasons. Almost 1,700 false failures were either undergraduates who were not in a position to take a teaching job, or individuals who were already employed as teachers. The false failures also included an undetermined number of individuals who found nonteaching jobs at higher pay. All of the foregoing factors suggested that the vast majority of the 27,000 class members' claims would not include a lost income component or noneconomic damages.

The foregoing analysis suggests that positing best case scenarios in the hundreds of millions of dollars as the upper range of recovery in this case is a pointless exercise. A more realistic way of evaluating the settlement involves using data that can be extracted from this case. The Court required ETS to furnish data on the individual settlements it entered into before the class action. See Barbara J. Rothstein & Thomas E. Willging, Managing Class Action Litigation: a Pocket Guide for Judges 10 (Federal Judicial Center 2005); see also MCL (4th) § 21.62 (a court should compare the settlement to settlements in individual cases). Accordingly, ETS produced data on the settlements it reached with individual claimants before the present litigation,^[4] which had a mean value of $9,149.12 and a median value of $4,469.98. At the fairness hearing, the parties explained that these numbers are likely to be higher than the average claim of a class member. (See Fairness Hr'g. Tr. at 29-31, 47-49). In particular, the early claimants were likely to be "motivated by a real perception that they has suffered, a real grievance to go out and retain lawyers." Id. at 47. ETS also stated its customary willingness to offer greater settlement amounts to these earlier claimants. Id. Finally, ETS noted its strong desire to settle claims that were large but below the $75,000 threshold for federal diversity jurisdiction, in order to avoid the costs associated with litigating in state courts. Id. at 48.

The Court will use the median value of the individual settlements as a benchmark. This follows because, as a matter of statistics, it is clear that this data set is skewed towards the low end. Only six of the forty-three values are greater than $20,000; only thirteen are greater than $10,000. And as ETS explained at the fairness hearing, the top values were unusually meritorious claims. One would expect a false failure with over $20,000 of demonstrable injuries to bring a claim quickly. Under the circumstances, the median value of $4,469.98 is more representative of the value of a typical claim than the mean value.

The second source of data is the data on claims submitted to date. Under the settlement, class members can ask for any type of damages they can substantiate, whether economic, noneconomic, or both. An analysis of the amount of damages actually sought in this case is instructive as to what the class members might have recovered at trial. The Special Master received 2,026 claims forms from the false failures, of which 1,365 were expedited claims forms seeking a $500 lump-sum, and 661 were individualized claims forms seeking some other amount. This means that roughly 67% of the responding false failures assert that their claims are worth no more than $500. One way to estimate trial recovery is for the Court to extrapolate from the foregoing data that 67% of all of the false failures had claims worth $500, and that the remaining 33% of false failures had individualized claims roughly comparable to the median value of the individual claims already settled. The median value of these settled claims is a suitable proxy for trial recovery because ETS indicated that it paid whatever amounts these claimants could substantiate. Assuming that 67% of all of the false failures, or 2,727 individuals, would recover $500, this yields a payout of $1,363,500. If the remaining 33% of false failures, or 1,343 individuals, have individualized damages of $5,000,^[5] this yields a recovery of $6,715,000. The sum of these two values, $8,078,500, represents an estimate of the total trial value of the claims of the false failures.

The Court will also generate an estimate of the value of the claims of the non-false failures. Because of the low response rate (9.9%) among non-false failures, it is more problematic to extrapolate from the submitted claim forms, since so many nonfalse failures are not claiming any damages at all. Nevertheless, the Court will compute a high-end estimate of these claims by performing the same type of extrapolation that it did with the false failures. Roughly 25% (574) of the responding non-false failures (2,284) sought individual damages. There are approximately 23,000 non-false failures. Extrapolating from the submitted forms yields 5,780 non-false failures who might have proved individualized damages. Assuming, generously, the same $5,000 in damages for each of these individuals leads to a recovery of $28,900,000. Assuming that each non-false failure would receive the value of a score report ($35) in damages adds an additional $805,000. Adding these numbers gives a total upper-end estimate of the trial value of the claims of the nonfalse failures of $29,705,000.

An estimate at the high end of the range of recovery is thus $37,783,500, or $8,078,500 + $29,705,000. But this value is not realistic for a number of reasons. First, it assumes a scenario in which every false failure recovers, when ETS clearly demonstrated that an unknown but significant number of false failures could not have shown any significant injury. Second, this scenario assumes a typical recovery by the false failures based on settlements of unusually strong claims. Third, this scenario assumes that five thousand non-false failures have substantial claims. Given the nature of these claims, the parties believed and the Court agrees that relatively few of the non-false failures would be capable of documenting any but de minimis injuries. Indeed, the overwhelming majority of these individuals (90.1%) have not even filed claims to recover from the settlement fund, even though they received the same notice as the false failures. Fourth, this estimate assumes recovery by non-false failures in the same amount as that of the false failures; if anything, one would expect any recovery by the non-false failures to be lower than that of the false failures. These problems with the estimated value of individualized claims by non-false failures are significant because this is the largest single component of the Court's estimate of potential trial recovery. Thus, while the Court considers the $37 million as at the upper end of the range of recovery, it is clear that the likelihood of such a recovery at trial is low.

If the Court assigned the 5,780 non-false failures theoretically seeking individual damages a 10% chance of recovering $5,000 apiece, the overall recovery would be $11,773,500. This is probably a more realistic number, since only 2.5% (574) of these 23,000 class members have sought to recover specific damages from the settlement fund. This recovery figure is fairly close to the amount of the settlement. Whether comparing the proposed settlement to the $37 million high-end estimate or the $11.7 million lower-end estimate, the Court is mindful that "compromise is the essence of a settlement." Cotton, 559 F.2d at 1330. A settlement entails "a yielding of absolutes and an abandoning of highest hopes." Id. (quoting Milstein v. Werner, 57 F.R.D. 515, 524-25 (S.D.N.Y.1972)) (internal punctuation omitted); see also In re: Warfarin Sodium Antitrust Litigation, 391 F.3d 516, 538-39 (3rd Cir.2004) (finding a settlement of 33% of available damages "well within a reasonable settlement range"). In sum, the Court's assessment of the range of recovery between a high end of $37 million and a lower end of $11.7 million favors approval of the proposed settlement of $11.1 million, especially considering that the cash settlement amount does not include the $35 value of the rescoring option available to all nonfalse failures.

Opinions of class counsel, class representatives, and absent class members

The opinions of all of the affected parties are favorable towards the settlement. Although the Court is careful not to overvalue the opinions of the attorneys involved, the attorneys have given a reasonable basis for their opinions. The Court therefore finds their opinions worthy of consideration.

The opinions of the absent class members appear to favor the settlement. Overall, approximately 16% of the absent class members have submitted claims forms to participate in the settlement. Among the false failures, the participation rate is unusually high. As of the stipulated cut-off date of July 22, 2006, the Special Master received 1,365 expedited claim forms and 661 individualized claim forms from the false failures. This yields an overall participation rate of 49.7% among the false failures. And 33.5% of the false failures submitted expedited claim forms, indicating their belief that the lump-sum payment was preferable to seeking any individualized recovery. This is a strong indication of class support for the settlement.

In addition, the opinions of class members are clear from the extremely low number of opt-outs. Only 4 class members chose to opt out before the deadline, or 0.014% of all class members and 0.10% of all false failures. Perhaps more tellingly, this is 0.09% of those who responded in any way to the notice. Thus, of the class members whose opinions are actually known, only 0.09% perceived that they were better off opting out. If the Court adds the number of objectors to the number of opt-outs, the percentage of class members expressing any dissatisfaction with the settlement agreement increases imperceptibly to 0.14%. Although the Court is careful not to infer too much from a lack of objectors, that the class is made up of a relatively sophisticated group whose claims arise out of a professional examination suggests that the lack of objections is an indicium of support and not of apathy. See In re: Warfarin Sodium Antitrust Litigation, 391 F.3d at 536 (finding a lack of objectors "particularly telling" because they were sophisticated businesses with potentially large claims). Thus, consideration of this factor strongly favors approval of the proposed settlement.

B. Objections

The objections filed by Kochensky and Welsh fall into three categories: (1) objections to the settlement, (2) objections to the motion for attorneys' fees, and (3) objections to the motion for incentive fees. The Court considers the latter two objections, infra, at §§ III and IV, respectively.

The objections to the settlement are (1) that the settlement is inadequate; (2) that false failures should receive $500 expedited payments irrespective of whether they returned a timely and valid expedited payment form; (3) that the provisions for compensating the Special Master and the Court Appointed Disbursing Agent (CADA) are uncapped and lack adequate safeguards; (4) that false failures who receive payments must bear the costs of the litigation while non-false failures who receive only a $35 score report bear none; and (5) that Pennsylvania false failures claimants are required to surrender their claims for treble damages and attorneys' fees for no extra consideration.

The first objection is that the settlement is inadequate. The Court has already discussed the adequacy of the settlement, supra, and concluded that the settlement amount is not inadequate. At the hearing, objectors' counsel demonstrated that his objection to the adequacy of the settlement was based on wildly optimistic suggestions of upper-end recoveries that, given the significant legal hurdles ahead, were at the bottom end of the probability scale.^[6] Accordingly, the objection to the adequacy of the settlement is overruled.

The second objection is that a claims process requiring false failures to complete and return an expedited claim form before receiving the $500 expedited payment is unnecessary. Objectors' counsel asserts that all false failures should receive an expedited payment as part of the settlement. At the fairness hearing, class counsel, the Special Master, notice expert Todd Hilsee, and the Court Appointed Disbursing Agent detailed the reasons for requiring claims forms instead of simply mailing out checks. (Fairness Hr'g Tr. at 152-56). Indeed, the disbursing agent indicated that a claims process provided for greater accountability and protected against fraud. Id. at 166-67. Requiring the claimant to submit a claim form is standard procedure in cases of this kind. Id. at 150; see also MCL (4th) § 21.61. The form is simple, easy to understand, and reduces administrative costs. In addition, objector's counsel implied that it might be possible to mail expedited payments with the settlement notice. Of course, before the final approval of the settlement, the funds for drawing these payments were not yet available. As Todd Hilsee pointed out in his testimony, because plaintiffs had the choice of either individualized damages or an expedited payment, to send the expedited payments with the notice has the potential of encouraging plaintiffs to forego individualized recovery for far less than value, merely by cashing the check. The obvious undesirability of this suggestion gives the unmistakable appearance that the objection was captious. The objection to the claims process for expedited payments is overruled.

The third objection is that the provisions for compensating the Special Master and the Court Appointed Disbursing Agent are uncapped and lack adequate safeguards for review and Due Process. The implication is that compensation for the Special Master and CADA have the potential to deplete unreasonably the funds available to the class. This objection is without merit. The Court has approved the Special Master's hourly rate, which, at $250 per hour, is reasonable. Further, the Court must review and approve any fees and expenses paid to the Special Master or CADA out of the settlement fund. (See R. Doc. 134 at 7-8). To date, the Court has approved disbursements of $23,802.05 in fees and expenses to the Special Master and $35,385.15 in fees and expenses to the CADA. These payments amount to 0.54% of the total settlement fund. There is every expectation that the work of the Special Master and the CADA will not drag on but will be completed before the year's end. In addition, the Special Master has in effect donated office space to handle claims processing, saving the class members' the expense of renting office space. The Court is satisfied that the provisions for compensating the Special Master and the CADA do not threaten the fairness, adequacy or reasonableness of the settlement. This objection is overruled.

The fourth objection is that the false failures bear the costs of litigation while non-false failures who receive only a $35 score report bear none. This argument flows from the fact that the attorneys' fees sought are a percentage of the settlement fund exclusive of the redeemable value of the score reports. The Court rejects this argument. If the value of the score report were added to the amount of the settlement, the result is to increase the attorneys' fee award. Further, the false failures each received a free score report already, before the litigation, with no deduction for attorneys' fees. The free score report provision in the settlement merely provides the same benefit for the other test takers. This objection is overruled.

The fifth objection is that the Pennsylvania false failure class members are required to surrender claims for treble damages and attorneys' fees without consideration. This objection lacks merit. The Pennsylvania class members were not required to surrender any claims. Any class member had the right to opt out of the settlement to pursue any unique claims. Further, the Court's review of the briefing on the motions to dismiss the Pennsylvania claims indicates that these claims faced a risk of summary dismissal at least as great as, if not greater than, the non-contract and non-economic damages claims of the other plaintiffs. There is no basis for a finding that the settlement is unfair to Pennsylvania claimants. Accordingly, this objection is overruled.

For all of the foregoing reasons, the Court finds the settlement to be fair, reasonable and adequate under Rule 23(e)(1)(C) of the Federal Rules of Civil Procedure.

III. ATTORNEYS' FEES AND COSTS

A. Attorneys' fees

Class counsel requests attorneys' fees of 40% of the $11.1 million settlement fund, which does not include the value of the $35 rescoring option available to class members who were not false failures. The court must independently analyze the reasonableness of the attorneys' fees proposed in the settlement agreement. See Strong v. BellSouth Telecomm., Inc., 137 F.3d 844, 849-50 (5th Cir.1998); see also Fed.R.Civ.P. 23(e). In a common fund settlement, in which the plaintiffs' attorneys are paid out of the settlement proceeds, the interests of the attorneys conflict with those of the class. Put simply, the more money the attorneys get, the less the class gets.

The Fifth Circuit has established twelve factors to consider in calculating reasonable fees and costs. Johnson v. Georgia Highway Express, Inc., 488 F.2d 714 (5th Cir.1974). Based on Von Clark v. Butler, 916 F.2d 255, 258 n. 3 (5th Cir.1990), the twelve Johnson factors are: (1) the time and labor required; (2) the novelty and difficulty of the question; (3) the skill requisite to perform the legal service; (4) the preclusion of other employment by the attorney due to the acceptance of the case; (5) the customary fee; (6) whether the fee is fixed or contingent; (7) time limitations imposed by the client or the circumstances; (8) the amount involved and the results obtained; (9) the experience, reputation, and ability of the attorneys; (10) the "undesirability" of the case; (11) the nature and length of the professional relationship with the client; and (12) awards in similar cases. Using the Johnson factors, the court should (1) ascertain the nature and extent of the services supplied by the attorney; (2) determine the value of the service according to the customary fee and the quality of the work; and (3) adjust the compensation on the basis of the other Johnson factors that may be of significance in that particular case. See Von Clark, 916 F.2d at 258 (citations omitted). The first two tasks compute the "lodestar," which is essentially the reasonable number of hours expended on litigation multiplied by a reasonable hourly rate. See Strong, 137 F.3d at 851; see also MCL (4th) §§ 14.121-14.122. The Court then uses the Johnson factors to adjust the lodestar upward or downward using a multiplier. Strong, 137 F.3d at 851; see also Longden v. Sunderman, 979 F.2d 1095, 1099-1100 (5th Cir.1992).

The lodestar method has been under increasing criticism because of the practical difficulties in applying it. See MCL (4th) § 14.121. The method has been called "difficult to apply, time-consuming to administer, inconsistent in result, and capable of manipulation." Id. Furthermore, "the lodestar method creates inherent incentive to prolong the litigation until sufficient hours have been expended." Id.; see also Theodore Eisenberg and Geoffrey P. Miller, Attorney Fees in Class Action Settlements: An Empirical Study, 1 Journal of Empirical Legal Studies 27, 31 (2004) (cataloging the shortcomings of the lodestar method) (hereinafter "Eisenberg and Miller"). For these reasons and others, the vast majority of Courts of Appeals have approved of, or mandated, the use of the percentage method. See, e.g., In re Thirteen Appeals Arising out of San Juan Dupont Plaza Hotel Fire Litigation, 56 F.3d 295, 307 (1st Cir.1995); Gwozdzinnsky v. Sandler Associates, 159 F.3d 1346, 1998 WL 538064 (2d Cir.1998); In re General Motors Corp. Pick-Up Truck Fuel Tank Products Liability Litigation, 55 F.3d 768, 821-22 (3d Cir.1995); Rawlings v. Prudential-Bache Properties, Inc., 9 F.3d 513, 515-17 (6th Cir.1993); Florin v. Nationsbank of Georgia, N.A., 34 F.3d 560, 564-65 (7th Cir.1994); In re Wash. Pub. Power Supply Sys. Sec. Litig., 19 F.3d 1291, 1295 (9th Cir.1994); Gottlieb v. Barry, 43 F.3d 474, 487 (10th Cir.1994); Camden I Condominium Ass'n, Inc. v. Runkle, 946 F.2d 768, 774 (11th Cir.1991); Swedish Hosp. Corp. v. Shalala, 1 F.3d 1261, 1271 (D.C.Cir.1993).

This Court shares the concerns expressed by the majority of courts that the lodestar method is not only burdensome but also acutely susceptible to manipulation. It is fairly easy for the parties to back into the percentage they desire by setting rates and imposing hourly billing requirements high enough to be certain to generate the fee desired. The Court is then faced with the task of second guessing the hours expended on scores of tasks by numerous lawyers over years of litigation. Although the percentage fee method is not a device of lapidary precision, at least it does not involve a mode of analysis that promises more than it can deliver. The Court recognizes that there is no one-size-fits-all percentage that is reasonable in all cases. See MCL (4th) § 14.121 (2004). Rather, the Court agrees with the conclusion of a task force of the Third Circuit Court of Appeals that "a percentage fee, tailored to the realities of the particular case" is the best method of determining reasonable attorneys' fees. Third Circuit 2001 Task Force Report on Selection of Class Counsel, 74 Temp. L.Rev. 689, 707 (2001).

The Fifth Circuit has not expressly adopted the percentage method as an alternative to the lodestar, but it has not overruled a case in which a district court resorted to the percentage fee method. Further, in Longden v. Sunderman, 979 F.2d at 1100, 1100 n. 11, the Court of Appeals upheld an attorneys' fee award that was calculated as a percentage of the total recovery, while at the same time noting that the trial court had also considered all of the attorneys' time and expense records, as well as the Johnson factors. Under Fifth Circuit law, the Court has the flexibility to calculate fees based on the percentage method, as long as it combines its determination with some analysis under the lodestar method. See MCL (4th) § 14.121. Based on all of these considerations, the Court will compute a percentage fee award that takes into account the Johnson factors. The Court will use the lodestar method as a cross-check.

BENCHMARK PERCENTAGE

Many courts begin their fee analysis by determining an initial "benchmark" percentage, which they then adjust for the particular circumstances of the case. See, e.g., Camden I Condominium Ass'n, 946 F.2d at 774-75 (citing Paul, Johnson, Alston & Hunt v. Graulty, 886 F.2d 268, 272 (9th Cir.1989)); In re Catfish Antitrust Litigation, 939 F.Supp. at 501; see also MCL (4th) § 14.122. In determining the benchmark, the Court will consider fee awards in similar cases, which the Court notes is the last of the Johnson factors. The Manual for Complex Litigation states that a fee of 25% of a common fund "represents a typical benchmark." MCL (4th) § 14.121. Further, data on fee awards in class action settlements is available in a recent study that compares two comprehensive class action data sets from 1993-2002. See Eisenberg and Miller at 28. The data sets are (1) data based on published decisions from "all state and federal class actions with reported fee decisions between 1993 and 2002, inclusive, in which the fee and class recovery could be determined with reasonable confidence"; and (2) information reported on more than 600 common fund cases from 1993 and 2002 in Class Action Reports (CAR). Id. The study finds a "strong correlation between the fee amount and the client recovery." Id. at 52. The study further indicates that a scaling effect exists, for "[a]s client recovery increases, the fee percent decreases." Id. at 54. The authors of the study suggest that the results can assist courts in determining fee awards:

[B]ecause our study finds an overwhelming correlation between class recovery and attorney fees, the court can conduct a simple initial inquiry that looks only at these two variables in any case where the size of the class recovery can be estimated. The court need only compare the request in a given case with average awards in cases of similar magnitude. If the request is relatively close to average awards in cases with similar characteristics, the court may feel a degree of confidence in approving the award. If the request is significantly higher than amounts awarded in past cases, the court should inquire further.

Id. at 72. Eisenberg and Miller divided the cases into ten ranges of recovery (deciles) and then gave the mean and median fee percent, as well as the standard deviation, for each fee percent. Id. at 73. The Court finds the study's data on the average percentage fee awarded in the recovery range comparable to this case useful in arriving at a benchmark percentage fee.^[7]Id.

The $11.1 million recovery in this case falls within the 40-50% decile of client recovery, which includes recoveries between $9.7 million and $15 million. Id. at 73. Based on the Class Action Reports (CAR) data set, the mean fee percent for nonsecurities cases in this decile was 27.3%, with a standard deviation of 5.2%. Id. The data set generated from published decisions shows a mean fee percent for cases in this decile of 22.7%, with a standard deviation of 8.4%. Id.

Based on this data, the Court will use an initial benchmark of 25%, which is roughly the average of the mean fee percentages of the two data sets for settlements of this size. The 25% benchmark is also the "typical benchmark" cited by the Manual for Complex Litigation (Fourth) § 14.121. The Court will next determine whether the benchmark should be adjusted based on the particular circumstances of this case. In doing so, the Court will consider the other Johnson factors.

THE JOHNSON FACTORS

The Time and Labor Required

The Court does not find that the amount of time and labor required in this case warrant an adjustment of the benchmark percentage. This case was settled well before trial, well before intensive discovery began, and after limited motion practice. The magnitude of the work required in connection with written discovery, the motion to dismiss, class certification, negotiating the settlement, preliminary approval of the settlement, and the fairness hearing is adequately reflected in the benchmark percentage fee. Accordingly, the Court finds that consideration of this factor does not warrant an increase in the or decrease in the benchmark fee award.

The Novelty and Difficulty of the Question and the Skill Requisite to Perform the Legal Service

This case did not involve factually complex issues. Indeed, ETS admitted that it incorrectly scored the plaintiffs' PLT exams. Although there were unresolved legal issues concerning the availability of tort recovery or recovery of nonpecuniary damages, they were not of such difficulty or demanding of such skill as to warrant an adjustment of the benchmark percentage. The Court finds that consideration of this factor does not warrant an increase or decrease in the percentage fee award.

The Preclusion of Other Employment by the Attorney Due to the Acceptance of the Case

The Court does not doubt that, at times, litigating this matter was time-consuming. It is equally clear, however, that the matter was not so all-consuming as to justify an increase in the fee award. This is especially true given that a good portion of the work took place after the parties confected the initial agreement on settlement. Given the number of attorneys involved, the relatively straightforward nature of these claims, and the absence of high intensity litigation, the Court finds that consideration of this factor does not warrant a departure from the benchmark fee calculation.

The Customary Fee

Class counsel argues that the customary fee in personal injury litigation is 33 1/3% to 50%, and they point to class action cases in Louisiana federal courts in which the court awarded fees of 40% of the common fund. Plaintiffs point to no facts to suggest that these cases are either representative or comparable to this one, and several factors suggest that they are not. The nature of the claims is fundamentally different in that contested personal injury claims arising from chemical spills can be more time intense and complex than straightforward contract claims. In this case, the operative facts on liability were not at issue between the parties. There was no need for complex medical or scientific proof of liability, causation or damages. Further, the size of the fund in this case supports the percentage fee award given. Under the circumstances, the Court will not rely on anecdotes from three personal injury cases to contradict published reports, based on over 1,000 cases, indicating that the customary fee for class actions of this size is between 22% and 27%.

Whether the Fee is Fixed or Contingent

Consideration of this factor is designed to "demonstrat[e] the attorney's fee expectations when he accepted the case." Johnson, 488 F.2d at 718. It is clear that the attorneys who undertook the original cases against ETS did so on a contingency basis, and that class counsel did so as well. But because the Court's fee award is comparable to a reasonable contingency fee award, the Court finds that consideration of this factor does not justify an increase to the fee award.

Time Limitations Imposed by the Client or the Circumstances

Under this factor, the Court is to give a premium for "priority work that delays the lawyer's other legal work." Johnson, 488 F.2d at 718. The Court finds that there are no facts in this case to suggest that this factor justifies an adjustment in the benchmark fee.

The Amount Involved and the Results Obtained

Counsel obtained a very favorable settlement for the plaintiff class. See § II-A, supra. As noted, the amount of the settlement is the largest ever in a testing case against ETS. This is not a case in which the class receives only illusory benefits in the form of coupons or discounts. Rather, counsel has achieved a substantial settlement in an efficient manner that minimizes the drain on the parties' and the Court's resources. Counsel also devised a plan for distribution of the fund and payment of claims that is practical, streamlined and fair. Accordingly, the Court finds that consideration of this factor warrants an increase in the fee award.

The Experience, Reputation and Ability of the Attorneys

The Court is satisfied that the experience and reputation of the attorneys involved are of the highest order. In addition, the Court notes that the ability of class counsel, as evidenced by their performance in this case, is equally first-rate. However, the Court finds that the quality of counsel is already accounted for in the enhancement the Court grants for the quality of the result and that this factor warrants no further enhancement.

The "Undesirability" of the Case

Class counsel asserts that this case was extremely undesirable, mostly based on the risk of non-recovery. The Court finds that although this is not the type of unpopular case that might stigmatize the lawyer who takes it, see Johnson, 488 F.2d at 719, there are some aspects of this case that made it undesirable. The risk of nonrecovery of non-contract damages was significant. Further, the relatively small size of the individual claims made undertaking expensive litigation against a well-financed corporate defendant on a contingent fee a not-very-attractive proposition. Class certification would change this dynamic, but, here, there was a serious risk that a class would not be certified. Accordingly, the Court finds that the risks inherent in this case warrant an increase in the fee award.

The Nature and Length of the Professional Relationship with the Client

There is no evidence of any special or lengthy professional relationship between class counsel and the class members. The relationship did not antedate the litigation, nor will it likely continue beyond the closure of this case. The Court finds that consideration of this factor does not warrant an increase in the fee award.

In conclusion, two of the Johnson factors merit an increase in the fee award: the results obtained factor and the undesirability of the case factor. The Court finds that these factors, taken together, warrant an increase in the fee award to 29% of the fund. This percentage is still within one standard deviation of the mean fee awards for cases of this size (the range of reasonableness suggested by Eisenberg and Miller), and it reflects the reality that only two of the Johnson factors supported an increase in the fee award. See Eisenberg and Miller at 74-76. A much stronger showing on the Johnson factors would have been necessary to justify a 40% fee.

THE LODESTAR CROSS-CHECK

Despite its limited usefulness, the Court will calculate a rough lodestar for purposes of checking the percentage fee award. The attorneys involved reported approximately 7,500 hours, at hourly rates of up to $500 per hour and usually over $400 per hour. These rates are significantly above-market in this jurisdiction, particularly for consumer contract cases involving individual damages that are on average less than $50,000. Class counsel has submitted no evidence to substantiate such above-market rates. Class counsel have not shown that they typically bill by the hour, and instead have indicated they usually charge contingent fees. Moreover, Patrick Juneau, the Court-appointed Special Master, who is performing services in this same case and who is of the same caliber and experience as the attorneys before the Court, is charging only $250 per hour for his own time. Given the Court's knowledge of the local legal market, the Court finds that a reasonable blended rate for the work involved here is $250 per hour. This rate reflects the prevailing hourly rate for partner-level attorneys in this jurisdiction ($350 per hour), averaged with the prevailing rate for associate-level attorneys in this jurisdiction ($150 per hour). The Court uses a blended rate to reflect that not all of the work involved should have been performed by the highest level attorneys. Rather, given the stage of this proceeding, most of the legal work before settlement involved tasks such as research, drafting, and document review, which reasonably should have been performed by associate-level attorneys. In addition, class counsel states that paralegals have spent over 900 hours working on behalf of the class. The Court will use a rate of $75 per hour for paralegal compensation. Using these hours and rates, the Court calculates a lodestar baseline of $1,942,500.

The Court must adjust the lodestar by the Johnson factors to determine an appropriate multiplier. The Court has already found that only two of the Johnson factors warrant an increase in the baseline fee. The Court finds that a multiplier of 1.6 is appropriate. This multiplier adequately accounts for the results obtained and the risks involved. Further, the Court notes that this multiplier produces a fee that is higher than the average fee for recoveries of a similar size.

The Court awards a total fee of $3,219,000, or 29% of the settlement fund. This fee award is further confirmed by the lodestar calculation. The fee award in this case is to be distributed among plaintiffs' counsel by agreement of the attorneys. Longden, 979 F.2d at 1101 (citation omitted). If the attorneys cannot agree to a fair distribution of the fees, the Court will appoint a Special Master, to be compensated out of the fee award, to assist the Court in determining an appropriate fee distribution.

B. Costs

Based on the recommendation of the Special Master, the Court finds that there is an adequate basis for an award of 2.5% of the settlement fund for litigation expenses. Counsel will submit costs to the Special Master, who will determine whether they were for the benefit of the class. Any residual funds will revert to the class.

IV. Incentive Compensation

Federal courts often approve incentive awards in class action lawsuits to compensate named plaintiffs for the services they provide and burdens they shoulder during litigation. See In re Catfish Antitrust Litigation, 939 F.Supp. at 503-04. Here, the class representatives were involved in initiating the litigation, and were willing to be subjected to discovery and to submit to depositions on behalf of the class. The Court finds that there is an adequate basis to award $2,000 incentive payments to each of the Class Representatives. See generally National Association of Consumer Advocates, Standards and Guidelines for Litigating and Settling Consumer Class Actions, 176 F.R.D. 375, 387 (West 1998).

Objectors Kochensky and Welsh argue that they, too, are entitled to incentive compensation for their contributions to the class. The Court notes that, of the seven objections offered by objectors' counsel, the Court found only one to be meritorious. This objection was to the amount of the attorneys' fee award. Although the Court did not require this objection to recognize that the proposed fee was too high, the objection is a favorable factor in the Court's evaluation of the objectors' entitlement to incentive compensation. Balanced against this factor is that objectors' other objections added nothing to the litigation and, if anything, only prolonged it. The Court devoted a good portion of the fairness hearing to entertaining these objections, many of which it found to be of a cavilling sort. Under the circumstances, the Court finds that it is not appropriate to award incentive payments to the objectors.

V. CONCLUSION

For the foregoing reasons, IT IS ORDERED that the Motion for Final Approval of the Class Action Settlement be GRANTED and the SETTLEMENT be APPROVED. IT IS FURTHER ORDERED that Class Counsel be awarded fees of 29% of the common fund and costs of 2.5% of the common fund. IT IS FURTHER ORDERED that the class representatives be awarded incentive payments of $2,000 each.

NOTES

[1] At the fairness hearing, the parties discussed pushing back the few late deadline original cut-off date of July 3, 2006 to accommodate a filings. The parties subsequently agreed to set the at July 22, 2006, or 10 days after the fairness hearing

[2] The objection filed by Edmund Heustis was received by the clerk of court on July 5, 2006. The objection filed by Michelle Kochensky was received on July 6, 2006. The objection filed by Brian Welsh was received on July 10, 2006.

[3] The Court does not have an exact number of false failures or non-false failures. Based on information provided by ETS, the best estimate of the number of false failures is 4,070, and the best estimate of the number of nonfalse failures is 23,000. (R. Doc. 155-3).

[4] ETS actually furnished two summaries of settlements with individual claimants. At the fairness hearing, ETS indicated that one set of settlements included those in which the claimant executed a release of claims against ETS; the other set of settlements included those in which the claimant did not. The settlement amounts in the first set were, on average, higher than those in the second set. In computing the range of recovery, the Court chooses to consider only the first set of settlements.

[5] In order to include only those settled claims that were for more than $500, the Court has removed the lowest two values from the list of individual settlements, because they were below $500. The median value of the settlement amounts rises to $4,491.03, which the Court rounds to the next thousand, or $5,000.

[6] During the fairness hearing, objectors' counsel suggested that the range of recovery might be $200 million. As the Court has discussed at length, the likelihood of a recovery scenario in which each of the false failures recovered approximately $50,000 is extremely slim.

[7] Eisenberg and Miller suggest that a fee request within one standard deviation of the mean is presumptively reasonable, and that one falling between one and two standard deviations from the mean may require further justification. Id. at 74.