Sylvester v. Cigna Corp.

ORDER ON PROPOSED CLASS ACTION SETTLEMENT SINGAL, Chief District Judge. Before the Court is Plaintiffs’ Motion for Approval of Settlement and for Class Certification (Docket # 55) as supplemented and amended by Plaintiffs’ Motion for Approval of Amended Settlement Agreement and Notice Plan (Docket # 66) and Plaintiffs’ Second Supplemental Memorandum in Support of Approval of Amended Settlement Agreement and Notice Plan (Docket # 81). In connection with this motion, Defendants have filed a Memorandum in Support of Plaintiffs’ Motion for Approval of Amended Settlement Agreement and Notice Plan (Docket # 65). In addition, Counsel for Plaintiffs have filed their Application for Award of Attorneys’ Fees (Docket # 56), along with the Class Counsel’s Supplemental Memorandum in Support of Application for Award of Attorneys’ Fees (Docket # 67) and Class Counsel’s Second Supplemental Memorandum in Support of Application for Award of Attorneys’ Fees (Docket # 83). For the reasons explained below, the Court DENIES the motions seeking approval of the proposed settlement (Docket # s 55, 66 & 81) and, therefore, finds the applications for attorneys’ fees (Docket #s 56, 67 & 83) are MOOT. I. BACKGROUND A. The Claims Plaintiffs filed their initial complaint (Docket # 1) against Defendants on July 15, 2003. Plaintiff then filed an amended complaint (Docket #24) on January 7, 2004. This Amended Complaint asserts claims against more than fifty different corporate entities that allegedly underwrote or administered health insurance plans to which Plaintiffs had subscribed (hereinafter collectively referred to as “Defendants” or “CIGNA”). The claims asserted include statutory claims under both the Employee Retirement Income Security Act (“ERISA”), 29 U.S.C. § 1132, and the Racketeer Influenced and Corrupt Organizations Act (“RICO”), 18 U.S.C. § 1962, as well as common law claims, such as breach of contract, fraud, fraudulent concealment, negligent misrepresenta *37 tion, unjust enrichment, accounting and punitive damages. Although the Amended Complaint names eight individual Plaintiffs, it asserts that these individual Plaintiffs wish to act as representatives for an entire class of people eligible to bring such claims against CIGNA. They describe that class as: “All subscribers to Healthsource Plans at any time between January 1, 1986 and December 31, 1997,who made payments to medical product or service providers under a percentage coinsurance plan, where an agreement existed between any of [the Defendants] and the provider to accept a discounted fee and where [the Defendants] did not use that discount in calculating the subscriber’s coinsurance payment as required by the Healthsource Plans (the “Class”).” 1 (Am.ComplJ 114.) As described in the Amended Complaint, Defendants allegedly misled the Class into making coinsurance payments that exceeded the amount actually due under the terms of their CIGNA plans. CIGNA allegedly reaped the benefit of this overpayment and thereby paid out less on the benefit claims filed by Plaintiffs. By way of illustration, Plaintiffs’ Amended Complaint offered the following description of what it described as Defendants’ “scheme”: [A]ssume an inpatient procedure was performed during the Class Period on a [CIGNA] Plan subscriber with a list price of $1,000. The subscriber pays her percentage coinsurance, for example 20% or $200, unaware that [CIGNA] negotiated a secret provider discount whereby the actual allowed cost of the procedure is $600. Since the subscriber has paid $200 out of [the] actual allowed cost of $600, the [CIGNA] Plan now pays only $400 to the provider. As a result of this scheme, the subscriber has paid 33.33% of the actual cost, instead of 20%, and the [CIGNA] Plan has paid only 66.67% of the cost, instead of the 80% called for under the terms of the [CIGNA] Plan. Since the subscriber’s correct coinsurance obligation is, in this example, 20% of the actual cost of $600, or $120, [CIGNA] has overcharged the subscriber $80 for the covered service. ... [A]fter clearly breaching its contractual obligations to the subscriber, [CIGNA] then knowingly sent via U.S. mail a false Explanation of Benefits stating that the allowed cost was $1000, while it concurrently sent a Remittance Advice to the medical provider showing the actual discounted amount for the covered service was $600. (Am. Compl. ¶¶ 79 & 80.) The Amended Complaint further claimed that the illustration just described had been “replicated ... in hundreds of thousands or more ... claims between 1986 and 1998.” (Am. ComplV 81.) B. Discovery Leads to Settlement Negotiations Between October 2003 and March 2004, the parties engaged in discovery that included production of approximately 112,-000 pages of documents and CIGNA’s claims data as well as depositions of three class representatives and two of Defen *38 dants’ corporate representatives. The Court had previously allowed the parties to proceed simultaneously with discovery on the merits and discovery related to class certification finding that the issues were “significantly intertwined.” (Order on Defs.’ Obj. to and Request for Amendment (Docket # 17) at 2.) However, the Court had set a deadline of March 5, 2004 for Plaintiffs to file their motion for class certification. Apparently, during discovery, Plaintiffs’ Counsel determined that the harm caused by Defendants’ alleged “scheme” was not as damaging or widespread as they had initially imagined. 2 Rather, based on analysis of data provided by CIGNA, the expert actuaries hired by Plaintiffs determined that the total damages were approximately $2.9 million. (See Dec. 6, 2004 Hearing Tr. (Docket # 64) at 11.) Defendants also hired an actuary to perform an analysis of the claims data they had produced. Defendants’ expert estimated that Plaintiffs’ damages were approximately $2.3 million. (See id.) In light of this information, counsel for both sides apparently began discussing the potential of settling the case in March 2004 (prior to filing a motion for class certification). These discussions culminated in a two-day mediation session held on June 23, 2004 and July 7, 2004. The parties emerged from the mediation with a handwritten agreement in principle to settle the claims of the class “in exchange for the creation of a settlement fund, funded by the settling Defendants, in the amount of 3.4 million dollars ($3,400,-000).” (Pis.’ Mot. for Class Cert, and Final Settlement Approval (Docket # 55) Ex. 3 at 3.) Although it was not evidenced in this handwritten agreement, there was apparently negotiation and discussion regarding whether the entire settlement fund would be distributed to the Class or whether distributions from the fund would be limited to the “claims made” with the balance then reverting to Defendants. (Groff Aff. (Docket # 82) ¶¶ 13 & 15.) In any event, Plaintiffs’ counsel believed the $3.4 million fund represented both the principal amount of damages plus some amount of interest owed on those damages and, therefore, they had essentially “got everything [they] wanted except perhaps more interest.” (April 28, 2005 Hearing Tr. (Docket # 91) at 10 (statement by Mr. Hansel).) C. The Original Settlement Agreement Following the successful mediation, the parties drafted a detailed proposed settlement agreement, dated July 23, 2004. (See July 23, 2004 Settlement Agreement (Ex. 2 to Docket # 43) (“Settlement Agreement”).) The Settlement Agreement contemplated that, in addition to payment to class members, the settlement fund would need to cover three other payments: (1) “the approved costs and fees of the Settlement Administrator,” (2) “the approved attorney fees and costs of Class Counsel” and (3) “the approved award to the Class Representatives.” (Settlement Agreement ¶ 1.17.) The Settlement Agreement defined the amount remaining after these three deductions as the “Residual Corpus.” (Id.) Under the terms of the Settlement Agreement, the actual amount distributed to the class would “[i]n no event ... exceed the amount of the Residual Corpus.” (Settlement Agreement ¶ 1.20.) However, this was not the only limitation set on the *39 amount payable to the class. Rather, the Settlement Agreement also capped that maximum payment to each class member at forty dollars ($40). 3 Apparently, the parties arrived at this maximum payment by estimating that the Residual Corpus would be approximately two million dollars ($2,000,000) and then dividing that amount by 50,000 class members — a CIGNA-sup-plied estimate. (See, e.g., July 27, 2004 Hearing Transcript (Docket # 50) at 5 (representation of Mr. Hansel).) In any event, under the terms of this Agreement, a Class Member would be required to return a claim form that included the Class Member’s name, address and social security number in order to receive any share of the settlement. (See Class Member Claim Form (Ex. 3 to Docket #43).) In addition, the Claim Form required each Class Member to sign the following statement under the penalty of perjury: I HAVE NO REASON TO DOUBT THAT I WAS A COVERED SUBSCRIBER OR MEMBER OF A HEALTH PLAN ADMINISTERED AND/OR UNDERWRITTEN BY A HEALTHSOURCE COMPANY AT SOME TIME DURING THE PERIOD BETWEEN JANUARY 1, 1986 AND DECEMBER 31, 1997, AND THAT I ACTUALLY MADE A PERCENTAGE CO-INSURANCE PAYMENT TO A HEALTH CARE PROVIDER UNDER THE TERMS OF A HEALTH PLAN UNDERWRITTEN AND/OR ADMINISTERED BY A HEALTHSOURCE COMPANY. (Id.) In addition to this claim form, the mailed notices also included an opt-out form that could be returned in lieu of the claim form. With respect to attorneys’ fees, the Settlement Agreement explicitly notes that “the Settlement in the total amount of three million four hundred thousand dollars ($3,400,000) was negotiated at arms length and determined first and independently, before the Parties conducted any negotiations relating to, or reached any agreement regarding attorneys’ fees and costs.” (Settlement Agreement ¶ 10.8.) Taking this representation at face value, 4 the parties apparently conducted a later round of negotiations that resulted in the Settlement Agreement including the following provision: Fee, Costs, and Expenses. Subject to approval or modification by the Court, Class Counsel’s reasonable costs of suit and attorneys’ fees shall, within fourteen (14) business days after Final Approval, be paid by the Settlement Administrator from the Corpus of the Settlement Trust. Defendant will not oppose a request by Class Counsel for reimbursement of reasonable out of pocket litigation costs up to $250,000, and for an attorneys’ fee of thirty percent (30%) of the net settlement fund, after deduction for the total settlement fund of reimbursable litigation costs and the expenses of settlement administration. *40 Plaintiffs will request an attorneys’ fee of thirty percent of the total settlement fund of three million four hundred thousand dollars ($3,400,000), and reimbursement of out of pocket litigation costs up to $250,000. Court approval of an attorneys’ fee and/or costs in an amount less than stated herein shall not affect any other provisions of the Settlement Agreement, which shall remain fully effective and enforceable. (Settlement Agreement ¶ 10.9.) This type of provision in which a defendant agrees not to oppose an attorney’s fee request that falls below a negotiated limit is commonly referred to as a “clear sailing” clause. See, e.g., Weinberger v. Great N. Nekoosa Corp., 925 F.2d 518, 520 n. 1 (1st Cir.1991) (“In general, a clear sailing agreement is one where the party paying the fee agrees not to contest the amount to be awarded by the fee-setting court so long as the award falls beneath a negotiated ceiling.”). Finally, the Settlement Agreement contained what is commonly referred to as a reverter clause. Pursuant to this provision, any portion of the settlement fund that was not spent or claimed as of “the 61st calendar day following the date of mailing of the last Class Member check” would be returned to CIGNA. (Settlement Agreement ¶ 10.10.) Since the Settlement Agreement contemplated a “claims made” settlement in which payments were only made to those Class Members that returned valid claim forms, there existed a potential that substantial unclaimed funds would revert back to CIGNA. The Settlement Agreement was presented to the Court for preliminary approval at a hearing on July 27, 2004. Much of the discussion at the July 27, 2004 hearing focused on ensuring that notice to all Class Members was adequate. The parties proposed to have the Settlement Administrator mail notices to a list of Class Members compiled by CIGNA. The Settlement Administrator would then follow up on any returned mail by attempting to get updated addresses through the Postal Service or TransUnion, a major credit bureau. The Notice Plan also called for reminder postcards to be mailed to all Class Members prior to the deadline for filing a claim form. In addition to this plan for mailed notices, the Court suggested that there be a publication notice. Ultimately, the parties adopted this suggestion and published a notice of settlement in USA TODAY on October 7, 2004. With respect to the substantive terms of the Settlement Agreement, the Court’s understanding was that each Class Member who filed a claim form would be eligible to receive a payment out of the settlement fund of up to $40. At the July 27, 2004 hearing, counsel for both sides represented to the Court that the class was made up of approximately 50,000 members who were to receive notification of the settlement via mail, along with a claim form and an opt-out form {See July 27, 2004 Hearing Tr. (Docket # 50) at 4-5 (representation of Mr. Hansel), 13 (representation of Mr. Armando).) Based on these representations and the terms of the Agreement, the Court expected that the payout to the class members would be approximately $2 million ($40 per claim x 50,000 claimants). 5 *41 Based upon this anticipated result, the Court issued an order preliminarily approving the proposed settlement as well as the form and matter of notice after the July 27, 2004 hearing. (See Order Preliminarily Approving Settlement (Docket # 47).) D. The First Fairness Hearing The parties next appeared befobe the Court on December 6, 2004 for a fairness hearing. In preparation for that hearing, Plaintiffs filed their Motion for Approval of Settlement and for Class Certification (Docket # 55) on November 22, 2004. In that fling, Plaintiffs’ counsel informed the Court for the first time that there were actually only 39,942 Class Members according to the records maintained by CIG-NA. 6 This twenty percent drop in the number of Class Members, combined with the $40 maximum payout, meant that that maximum payout to the Class under the terms of the Settlement Agreement would be approximately $1.5 million (or, more precisely, $1,597,680.00). It also guaranteed that over four hundred thousand dollars ($400,000) would revert back to CIG-NA. Thus, the $3.4 million settlement fund was “illusory” (in the words of Plaintiffs’ Counsel) and the actual settlement fund did not exceed $3 million. (Pis.’ Mot. for Class Certification and Final Settlement Approval at 13.) At the December 6, 2004 hearing, the Court expressed sincere concerns about these belated revelations regarding the actual size of the class and openly doubted whether discovery prior to the settlement discussions had been sufficient. (See, e.g., Dec. 6, 2004 Hearing Tr. (Docket # 64) at 15.) In both their written submissions and their oral presentation at the hearing, Plaintiffs’ Counsel urged the Court to essentially re-write the Settlement Agreement and increase the maximum amount payable to each Class Member so that the maximum amount reflected the new information regarding the size of the class. 7 For their part, Defendants opposed any increase to the $40 maximum payout to each class member. The payout problems caused by the twenty percent drop in actual Class Members was compounded by a very low response rate on returned claim forms. In fact, mailed notices had been successfully sent to approximately 81.6 percent of the 39,942 Class Membérs. In response to the mailings, 215 Class Member had returned *42 opt-out forms. 8 As of November 17, 2004, only 3,518 Class Member had returned claim forms. By the Court’s calculations, this meant that approximately 28,859 Class Members had received mailed notices and done nothing. 9 In short, if the Court had approved the Settlement Agreement on December 6, 2004, the total payout to the Class would have totaled only $140,720 ($40 x 3,518 Returned Claim Forms). Under the terms of the Settlement Agreement, it was also contemplated that there would also be a payout totaling $40,000 to the eight Class Representatives as well as a disbursement to the Settlement Administrator totaling $75,000. 10 In addition, Class Counsel had filed an application seeking $1,020,000 in attorneys’ fees and reimbursement for $189,062.10 in expenses and costs. 11 (See Class Counsel’s Application for Award of Attorneys’ Fees (Docket # 56) at 1 & 23.) Even assuming that the Court had given wholesale approval to these other requested sums, because of the reverter clause, the entity receiving the largest disbursement from the settlement fund would have been CIGNA. In fact, more than $1.9 million would have reverted to CIGNA— making the payout to Defendants significantly larger that the combined payout to all other interested parties (the Class, the Class Representatives, Class Counsel and the Settlement Administrator). Thus, at the time of the First Fairness Hearing the de facto settlement fund had shrunk from the originally contemplated $3.4 million to approximately $1.4 million, with less than $200,000 actually going to the persons harmed by CIGNA’s alleged conduct. The Court’s dissatisfaction with this outcome was readily apparent at the first fairness hearing. Ultimately, the Court continued the hearing while requesting additional information from the parties. E. The Amendment to the Settlement Agreement On January 14, 2005, the parties filed an Amendment to the Settlement Agreement (Ex. 1 to Docket # 66) (together with the Settlement Agreement, the “Amended Settlement”). Pursuant to this Amendment, the parties agreed to raise the maximum amount paid out to each Class Member who filed a Valid Claim Form from $40 to $51.97. The parties arrived at this larger amount by taking the estimated Residual Corpus of $2,075,937.90 and dividing it among that actual number of Class Members (39,942). (See Defs.’ Mem. in Supp. of Pis. Mot. for Approval of Am. Settlement Agreement (Docket # 65) at 1 n.l.) In short, this Amendment addressed and resolved the issues regarding the parties’ *43 misplaced reliance on CIGNA’s initial estimate of 50,000 class members. In light of this Amendment, it was now theoretically possible that the Class could recover the entire Residual Corpus if each Class Member returned a claim form. Because of the $11.97 increase in maximum payout to each Class Member, the parties proposed to have a new round of notices mailed to any Class Member who had not already returned a valid claim form. Under this revised notice plan, mailed notices would also be re-sent to the 215 Class Members who had elected to opt-out thereby giving them an opportunity to opt back in to the proposed settlement. In addition, the parties agreed to publish an amended notice in USA TODAY. Plaintiffs’ Counsel agreed to pay any additional costs incurred with these revised notices. F. The Second Fairness Hearing The Parties appeared before the Court seeking approval of this Amended Settlement Agreement on January 24, 2005. At the opening of this hearing, Class Counsel initially indicated that they were before the Court seeking final approval of the Amended Settlement. At the Court’s suggestion, the parties agreed to proceed with a request for preliminary approval of the Amended Settlement Agreement and delay a request for final approval until the next fairness hearing, which was set for April 28, 2005. After this concession, the remainder of the January hearing was spent discussing possible revisions to the substance of the notice being provided to the Class. In particular, the Court suggested that Class Members should be advised, in plain language, about the ramifications of the re-verter clause. In response to the Court’s suggestion, the parties ultimately agreed to add the following sentence to the amended notices: “Unclaimed settlement funds will be retained by CIGNA.” (See Jan. 24, 2005 Hearing Tr. (Docket # 75) at 29.) Counsel for both sides, and Defense Counsel in particular, were reluctant to include references to the reverter clause in the newly revised notice, thinking such information was unnecessary. However, as the Court explained at the hearing, the Court believed that this information may well be relevant to some Class Members and thereby encourage the filing of additional claim forms. Driving the Court’s concerns was what the Court considered to be an unexpectedly low response rate. At the time of the January hearing, the Settlement Administrator had reportedly received a total of 4,335 valid claim forms. This represented approximately 10.8 percent of the class. Even with the larger $51.97 payout on each Class Member, this return rate on claims forms would have yielded a total payout to the Class of only $225,289.95 (plus any additional amounts paid to the Class Representatives). G. The Third Fairness Hearing On April 28, 2005, the parties appeared before the Court a third time seeking final approval of the Amended Settlement Agreement that they presented to the Court in January of this year. At this latest hearing, the Court heard presentations from both sides in favor of approving the settlement. In addition, Plaintiffs presented the testimony of Professor Edward Sherman, a law professor and past dean of Tulane Law School, to support their request for final approval. Per the terms of that Amended Agreement, Plaintiffs’ Counsel incurred a total of $21,363.39 in unreimbursable costs as a result of the second round of mail and *44 publication notices. Following this second round of notices, the Settlement Administrator reported that a total of 7,875 valid claim forms had been returned. {See Tilghman Aff. (Docket # 84) ¶ 12.) With each of these Class Members receiving a check from the settlement fund for $51.97, the total payout to the Class would now total $409,263.75 (plus any additional sums awarded to the Class Representatives). Class Counsel represents that the 19.7 percent return rate achieved at this point is “an excellent response rate for a large consumer class action.” {See Pis.’ Second Supp. Mem. (Docket # 81) at 4.) To support their assessment, they cite the observations of Mr. Tilghman, the Settlement Administrator. 12 In two separate declarations filed with the Court, Mr. Tilghman has explained that, in his experience, “claim return rates are 10% or less in the vast majority of settlements that require filing a notice of claim.” (Tilghman Aff. (Docket # 84) ¶ 13 & Tilghman Aff. (Ex. L to Docket # 65) ¶ 11.) By this standard, the 19.7 percent return rate achieved in this case may well be above average. 11. DISCUSSION The Court now faces the following question: Is the current proposed settlement fair, reasonable and adequate? In answering this question, the Court fulfills its “judicial duty to protect the members of a class in class action litigation from lawyers for the class who may, in derogation of their professional and fiduciary obligations, place their pecuniary self-interest ahead of that of the class.” Reynolds v. Beneficial Nat’l Bank, 288 F.3d 277, 279 (7th Cir.2002). It is the burden of the proponents of the settlement to prove that the proposed settlement is fair, reasonable and adequate. See 7B Wright, Miller & Kane, Federal Practice & Procedure: Civil 3d § 1797.1 (2005). In the context of the multi-district litigation over compact disc pricing, another court in this district faced this same question. In that case, Judge Hornby explained the applicable standard and laid out some specific factors to be considered: There is no single test in the First Circuit for determining the fairness, reasonableness and adequacy of a proposed class action settlement. In making this assessment, other circuits generally consider the negotiating process by which the settlement was reached and the substantive fairness of the terms of the settlement compared to the result likely to be reached at trial. Specifically, the appellate courts consider some or all of the following factors: (1) comparison of the proposed settlement with the likely result of litigation; (2) reaction of the class to the settlement; (3) stage of the litigation and the amount of discovery completed; (4) quality of counsel; (5) conduct of the negotiations; and (6) prospects of the case, including risk, complexity, expense and duration. In re Compact Disc Minimum Advertised Price Antitrust Litigation, 216 F.R.D. 197, 206-07 (D.Me.2003) (internal citations omitted). Generally, courts have recognized a “presumption of fairness” for settlements that are deemed to be the result of “arm’s-length negotiations” following *45 “meaningful” or “sufficient” discovery. Wal-Mart Stores, Inc. v. Visa U.S.A, Inc., 396 F.3d 96, 117 (2d Cir.2005) (quoting the Manual for Complex Litigation, Third § 30.42 (1995)); see also In re Compact Disc MAP Litigation, 216 F.R.D. at 207 (“[T]he case law tells me that a settlement following sufficient discovery and genuine arm’s-length negotiation is presumed fair.”). A. The Settlement Does Not Fall Within the Presumption of Fairness The provisions of this settlement agreement, along with facts of this case, do not allow the Court to apply a presumption of fairness to its review of the proposed settlement. Rather, the Amended Settlement Agreement and the process by which that agreement was reached give rise to an inference that the settlement was not the product of arm’s-length negotiations and sufficient discovery. Even accepting the representations of the mediator regarding the two days of mediation, the Court remains troubled that key provisions of the Amended Settlement Agreement, which were finalized by the parties after the mediation session, combine to form an agreement that appears collusive on its face and in practice. Specifically, the Court remains troubled by the combination of the reverter clause and the clear sailing provision. In concert, the Court believes that these two provisions give rise to inferences that there was a lack of arm’s length negotiations and a lack of zealous advocacy for the Class by Plaintiffs’ counsel. In fact, Professor Sherman, an expert presented by Plaintiffs in support of the settlement, admitted that reverter clauses are generally “suspect” and need to be viewed cautiously since they “undercut!] the deterrent effect of class actions” 13 and can “suggest a lack of full diligence in representing the class.” (April 28, 2005 Hearing Tr. at 30 & 38 (testimony of Prof. Sherman).) Of course, the presence of clear sailing clauses correlates generally with reverter clauses. See William D. Henderson, Clear Sailing Agreements: A Special Form of Collusion in Class Action Settlements, 77 TUL. L. REV. 813, 835 (2003) (“[I]t is important to recognize that it would be relatively rare for a plaintiffs attorney to agree to a reverter-fund settlement without also having the security of a clear sailing agreement to reduce the uncertainty in his fee award.”). However, by including a clear sailing provision, plaintiffs’ counsel’s financial incentives are “de-couplefd] ... from those of the class” thereby “increasing the risk that the actual distribution will be misallocated between attorney’s fees and plaintiffs’ recovery.” International Precious Metals Corp. v. Waters, 530 U.S. 1223, 120 S.Ct. 2237, 147 L.Ed.2d 265 (2000) (Mem.) (Justice O’Connor’s Statement respecting denial of writ *46 of certiorari). 14 Thus, it stands to reason that the presence of both a reverter clause and a clear sailing clause should be viewed with even greater suspicion and not be presumed fair to the class. Because of the problems inherent in a class settlement agreement that includes both a reverter clause and a clear sailing clause, the Court believes that the presence of these two provisions in any settlement agreement should present a presumption of unfairness that must be overcome by the proponents of the settlement. 15 The current proposed distribution bears out the problems generally associated with reverter clauses and clear sailing provisions. Based on the most recent estimates provided to the Court, the total payout to the class will be $409,263.75. In addition, if approved by the Court, the Class Representatives may receive an additional $40,000. The two payments combined would yield a total proposed payout to the class of $449,159.81. This amount represents approximately 13 percent of the total $3.4 million settlement fund. In contrast, because of the reverter clause and clear sailing provision, both Class Counsel and Defendants stand to receive substantial portions of the original $3.4 million settlement fund. Specifically, Plaintiffs’ Counsel has filed a request for $1,020,000 in attorneys’ fees and an additional $211,238.25 in expenses. These requests account for approximately 36 percent of the $3.4 million settlement fund. For their part, Defendants have filed only a limited objection to this request that complies with the clear sailing provision of the Settlement Agreement and thereby does not object to Plaintiffs counsel receiving approximately 30 percent of the settlement fund. *47 Assuming the Court were to award Plaintiffs counsel the fees and expenses requested, $1,644,601.94 would still revert to Defendants under the terms of the Amended Settlement Agreement. 16 Thus, approximately 48 percent of the original settlement fund would go back to CIGNA. (Of course, if the Court were to limit the award of attorneys’ fees and expenses, an even larger sum would revert to Defendants pursuant to the terms of the Amended Settlement Agreement.) Given the parties’ estimations that actual damages to the 39,932 Class Members totaled between $2.3 million and $2.9 million (before interest), the Court simply cannot conclude that this proposed division of the $3.4 million settlement fund was the product of an arm’s-length negotiation. In short, the Court believes that the reverter clause and clear sailing clause raise a presumption of unfairness and the proposed distribution in light of these provisions does not overcome that presumption. In addition, the Court notes that there is a basis for questioning whether the parties have satisfied the “sufficient discovery” prerequisite for a presumption of fairness. The most glaring example of the insufficiency of the discovery prior to settlement involves the size of the Class. As became apparent after the first round of mailed notices were sent, neither side was even aware of the actual number of Class Members at the time they negotiated the original Settlement Agreement. Rather, Plaintiffs’ Counsel proceeded to finalize a settlement agreement without even knowing how many clients they were representing. Despite the apparent lack of discovery on this point, the parties proceeded to rely on an inaccurate estimate of the size of the Class in order to set a maximum recovery of $40 for each Class Member. Regardless of the later adjustments made to reflect the actual class size of 39,942, the change necessitated by this post-negotiation revelation simply highlights a crucial gap in the discovery undertaken prior to the settlement negotiations. In short, despite the number of hours and dollars expended on discovery prior to March 2004, the Court concludes that the discovery completed prior to the negotiation and drafting of the Settlement Agreement does not weigh in favor of applying a presumption of fairness. Rather, because the combination of a reverter clause and a clear sailing clause suggests a lack of arm’s-length negotiations that is not overcome by the facts and proposed distribution of the fund, the Court essentially proceeds on a presumption of unfairness. B. The Settlement is Unfair, Unreasonable and Inadequate Of course, the proponents of the proposed settlement may overcome a presumption of unfairness by showing that the proposed settlement is, in fact, fair, reasonable and adequate considering the non-exhaustive list of factors laid out above. Below, the Court discusses each of the relevant factors it has examined and explains its basis for ultimately concluding that the proponents of this settlement have not met their burden of showing that the settlement is fair, reasonable and adequate. 1. Stage of Litigation and Amount of Discovery Completed This litigation was in a relatively early stage when the settlement was originally negotiated. Plaintiffs had not yet filed their class certification motion although *48 the deadline was fast approaching when the parties began their settlement negotiations. The Court has already explained one apparent gap in the discovery completed prior to the Settlement Agreement being finalized. 17 In short, the stage of the litigation and amount of discovery completed suggest that Plaintiffs’ Counsel has not yet completed a full investigation of the allegations laid out in the Amended Complaint. Nonetheless, the serious discovery conducted prior to the settlement negotiations clearly led Plaintiffs’ Counsel to conclude that the legal and factual obstacles they could encounter during the litigation were larger than initially anticipated and that the damages from Defendants’ alleged conduct were smaller than initially estimated. While settlement negotiations may have been a rational route to consider under these circumstances, the Court does not believe that either the stage of this litigation or the amount of discovery completed necessarily suggest that the proposed settlement is fair, reasonable and adequate. 2. Quality of Counsel and Conduct of Negotiations Because of the early stage of the litigation, this Court has not had many opportunities to review and consider the quality of counsel in the context of this case. Thus, the Court’s review is, for the most part, limited to the context of the motions and hearings surrounding the proposed settlement. Undoubtedly, Plaintiffs’ Counsel is made up of a rather large team of experienced lawyers. There are eight attorneys hailing from three different firms listed as attorneys of record for Plaintiffs. At least some members of this team have previous experience serving as class counsel and have been involved in one or more previous cases pressing similar claims against CIGNA on behalf of smaller classes of subscribers. These cases, one in South Carolina and one in New Hampshire, both resulted in settlements. In short, the Court has no reason to doubt that Plaintiffs’ Counsel is qualified and has the necessary experience to represent the Class in this matter. That said, at some point in negotiating the proposed settlement, it appears Plaintiffs’ Counsel lost sight of their fiduciary duties to the Class while still managing to protect their own pecuniary interests. Plaintiffs’ Counsel negotiated a clear sailing provision in which Defendants essentially agreed that Plaintiffs’ Counsel are entitled to fees and expenses totaling 30 percent of the entire $3.4 million rever-sionary fund. 18 With this agreement in hand, Plaintiffs’ Counsel seems content to rationalize the low claims rate and correspondingly low payout to the Class by *49 citing the age-old adage, “You can lead a horse to water, but you can’t make him' drink.” (Dec. 6, 2004 Hearing Transcript at 62 (statement by Mr. Hansel).) While there may be some truth to this saying, it does not capture the fiduciary relationship between a lawyer and a group of clients. As fiduciaries to the Class, it is the responsibility of Plaintiffs’ counsel to ensure that the settlement provides real value (or, to extend the metaphor of the just quoted aphorism, ‘actual drinks’) to their 39,769 clients. In this case, it appears that Plaintiffs’ Counsel secured a $3.4 million pool and also secured (to the extent possible) approximately a third of that pool for themselves. Nonetheless, they did not obtain the same level of security for their clients. This arrangement has satiated Plaintiffs’ counsel, but, in this Court’s assessment, it has ultimately left the Class parched. The Court has reviewed the affidavit submitted by the mediator, and also considered counsels’ own representations that the negotiations were hard-fought and contentious. These representations include Defense counsel’s candid statement at the last hearing that “CIGNA would have been perfectly willing to settle this case without a reversion but the amount of the settlement would have reflected still Cigna’s assessment of the value of this claim and it would not have been 3.4 million dollars.” (April 28, 2005 Hearing Tr. at 80 (statement by Mr. Armando).) Nonetheless, it is far from clear that Plaintiffs’ counsel would have settled this case for a fixed fund where the maximum amount that would be paid to the class was less than $450,000.00 (the amount currently slated to be paid to the Class). Moreover, as discussed below, it is not clear that a fixed fund in that amount would be a fair, reasonable and adequate settlement. However, this is the de facto result of the current settlement for the Class and Class Representatives. In this Court’s assessment, this end result does not reflect the product of three separate law firms fulfilling their fiduciary duties to a class of 39,769 clients. 3. Reaction of Class All Class Members who could be located have received three separate mailed notices regarding the proposed settlement: the first notice was mailed on or around September-October 2004, the second amended notice was mailed on or around February 2005 and a final reminder postcard mailed in March 2005. The Court has received no written objections and no objectors have appeared at any of the hearings. In addition, only 173 valid opt-out forms were received (representing well less that 1 percent of the total class). Certainly, a complete lack of objections and a minimal number of opt-outs weighs in favor of approving the settlement. However, in considering the. “reaction of the class,” the Court also considers the relatively small percentage of class members (19.7 percent) who have expressed implicit support for the proposed settlement by returning claim forms as well as the vast majority that have remained silent and essentially expressed a reaction of utter indifference to the settlement. Given that the large majority of Class Members fall into this latter category, the Court finds the reaction of the Class is a factor that weighs neither in favor of approving or rejecting the proposed settlement. 4. Prospects of the Case & Likely Result of Litigation Although the Court must be careful to not engage in a trial on the merits, 19 many *50 courts faced with proposed class actions settlements do compare the proposed settlement with the likely result of litigation, which implicitly requires the court to consider the prospects of the case, including risk, complexity, expense and duration. This particular inquiry focuses on whether the proposed settlement in adequate in light of “the net expected value of continued litigation to the class,” to the extent that it can reasonably quantified. Reynolds, 288 F.3d at 284-85. In its most recent submission to the Court, Plaintiffs’ Counsel states: “It is unlikely that Class Members could have done better if the case had proceeded to trial. In fact, continuing this litigation would entail substantial risk that the case might not reach trial stage, and that, if it did the Class might recover nothing.” (Pis.’ Second Supp. Mem. (Docket # 81) at 4.) Similarly, the mediator reports that “a number of potential weaknesses in the Plaintiffs’ case” were discussed during the mediation, including “the age of the claims, statute of limitations, the difficultly] of establishing RICO claims and the proof problems of marshaling data from various computer systems going back to the 1980’s.” (Groff Aff. ¶¶ 11 & 15.) Defendants’ own submission goes into even greater detail explaining various defenses it would pursue if this litigation were to proceed. (See Defs.’ Mem. in Support of Approval of Am. Settlement (Docket # 65) at 4-17.) In the same vein, Defendants have also detailed the hurdles Plaintiffs would face in even attempting to obtain class certification in this case. 20 (See id. at 18-20.) Undoubtedly, the se defenses and obstacles to certifying the proposed class support “a meaningful risk discount,” as suggested by Defendants. (Id. at 4.) Nonetheless, a court may not approve a settlement simply based on a simple finding that “the prospects for the class if litigation continue[s][are] uncertain” or even poor. Reynolds, 288 F.3d at 284-85; see also In re Compact Disc MAP Litigation, 216 F.R.D. at 221 (refusing to approve a settlement of “dubious value” to the class despite finding that continued litigation might well “generate absolutely no benefit for the class”). Rather, the Court must determine whether the settlement represents a reasonable effort to “give[] [the class] the expected value of their claim if it went to trial, net the costs of trial.” Mars Steel Corp. v. Continental Illinois Nat’l Bank & Trust Co., 834 F.2d 677, 682 (7th Cir.1987). In Mars Steel, Judge Posner provides the following illus *51 tration of how a court might perform such an analysis: Suppose the claim of the class in this matter is worth $750 million, but the probability that the class would prevail if the case were tried is only one percent and if it did prevail it would have to pay a contingent fee of $10 million. Then assuming risk neutrality (meaning indifference between a sum certain and its uncertain equivalent — e.g., between $2 and a 10 percent chance of $20), the class would be better off settling for any amount greater than $7.4 million than taking its chances on a trial. And if the members of the class were risk averse they might consider themselves better off even if the settlement were much smaller. Id. (quoted by Defendants’ in their Memorandum in Support of Approval of- Settlement (Docket # 65) at 3). A conservative starting point for such an analysis in this case is that the Plaintiffs’ claims are worth $2.3 million (the estimate of Defendants’ expert). 21 Based on the latest estimates of the parties, it appears that they propose to settle the claims of the class for a total payout of $409,159.81. Assuming risk neutrality, the Class is “better off settling” for this amount if the probability that the Class will prevail to a judgment of $2.3 million is approximately 17 percent or less: Mars Steel, 834 F.2d at 682. While the proponents of this settlement hate described significant evidentiary and legal hurdles that Plaintiffs may not be able to cross as well as pointing to viable defenses that CIGNA may pursue, they have not shown that the probability of success is less than 17 percent. 22 Thus, the Court declines Defendants’ invitation to declare this settlement fair based on a Mars Steel analysis of its reasonableness. 23 Admirably, Defendants’ submission in support of settlement also attempts to establish the adequacy of the settlement through quantifying the net expected value of continued litigation to the class by laying out various recovery scenarios. (See Defs.’ Mem. in Supp. of Approval of Settlement at 20-22.) While the Court has considered this analysis in assessing the settlement here, the analysis does not support *52 the conclusion that it is reasonable for the Class to settle their claims for a net payout of $449,159.81. 24 Presenting their analysis in terms of the net payout to each class member, Defendants’ analysis would arguably find that per-class-member payouts in the $20-$30 range adequately reflected the net expected value of continued litigation to the class once a “meaningful risk discount” is factored in. However, dividing the proposed net payout to the class by the 39,769 Class Members who have not opted out of this settlement yields a per class member payout of $11.29. No proponent of this settlement has shown the Court that this amount adequately reflects the net expected value to the Class if this case was litigated. Thus, despite the parties’ attempts to candidly disclose the risks associated with continuing this litigation, the Court simply cannot find that the proposed settlement is adequate even when viewed in light of these real litigation risks. 5. Proposed Allocation and Distribution of the Settlement Fund 25 The Court has already described in great detail how the parties propose to distribute the Settlement Fund. To summarize, the parties proposed to distribute approximately 13 percent of the fund ($449,159.81) to approximately 19 percent of the Class. The more than 80 percent of the Class that did not return claim forms will receive nothing. Class Counsel claims it is entitled to 36 percent ($1,231,238.25) of the fund to cover its fees and expenses. What remains after these disbursements and the payment of the settlement administration expenses amounts to more than 48 percent of the fund, which would revert to CIGNA. This distribution and allocation of the settlement fund is unfair, unreasonable and inadequate on its face. The Court has spent a great deal of time attempting to understand how a proposed settlement that appeared quite viable when it was first presented to the Court could have resulted in distribution that this Court simply cannot approve. Stripped to its essence, the Court believes that three factors combined to create this untenable distribution scenario: (1) a claims made settlement premised on a 100 percent response rate, (2) a reverter clause, and (3) a clear sailing provision. Unfortunately, when this matter first came before the Court from preliminary approval in July 2004, the Court was not independently aware (nor did any proponent of the settlement bring to the Court’s attention) what the parties and the Settlement Administrator already knew — namely, that “claims made” settlements regularly yield response rates of 10 percent or less. Having achieved a response rate that is above this average, the amount being distributed to the class is still woefully inadequate in large part because the maximum payout to each class member of $51.97 is premised on a 100 percent response rate. In short, the parties agreed to cap the payout to each individual class member based on a 100 percent response rate when they fully expected a much lower response rate. Of course, this cap and the parties’ expectations of a low response rate gives the reverter clause real value for Defendants. Likewise, the clear sailing provision, which *53 guarantees that Defendants will not object to Plaintiffs’ Counsel essentially seeking 30 percent of the entire settlement fund— regardless of how much is disbursed to the class, gives this settlement real value for Plaintiffs’ Counsel. The Court does not mean to suggest that each of these features (i.e., claims made settlements, payout caps based on 100 percent response rates, reverter clauses, or clear sailing provisions) is per se unacceptable. However, this case provides a telling example of how these features can work in concert to produce a settlement that is unfair, inadequate and unreasonable and that in practice yields comparably little value for the Class. The Court has fully considered the impact of its decision to deny approval of the settlement at this stage, especially in light of the fact that Class Members have already received repeated notification of this settlement and arguably some 7,875 Class Members are eagerly awaiting the receipt of checks for $51.97. The Court is aware that this denial may seem unfair following in the footsteps of the Court’s preliminary approval given to both the original Settlement Agreement and the Amended Settlement Agreement. However, as the Court has learned in this case, the inherent nature of a claims-made reversionary fund settlement is that the settlement can have the potential to provide fair and adequate compensation to the Class and yet in practice yield an actual settlement that is inadequate. This Court is simply not willing to approve such a settlement and thereby disregard the small amount actually paid to Class Members as a collective action problem that cannot be solved in the context of this litigation. The parties have represented to the Court that approximately 39,942 known subscribers were allegedly harmed by Defendants conduct and that the damages incurred total somewhere between $2.3 million and $2.9 million. This case is not an extraordinarily large consumer class action with millions of unknown members. Rather, the settlement administrator has been able to locate approximately 32,592 Class Members (approximately 81.6 percent of the Class). Nonetheless, the parties now ask the Court to approve a settlement that will extinguish the claims of the entire class by paying $51.97 to a mere 7,875 Class Members. In the Court’s assessment, the proponents of this settlement have failed to show that the proposed distribution of the settlement fund is fair, reasonable and adequate. Having considered all relevant factors, the Court finds that it cannot approve the proposed settlement in accordance with Federal Rule of Civil Procedure 23(e)(1)(C). III. CONCLUSION For the reasons just explained, the motions seeking approval of the proposed settlement (Docket # s 55, 66 & 81) are hereby DENIED. Having determined that the proposed settlement cannot be approved, Class Counsel’s Supplemental Memorandum in Support of Application for Award of Attorneys’ Fees (Docket # 67) and Class Counsel’s Second Supplemental Memorandum in Support of Application for Award of Attorneys’ Fees (Docket #83) are DENIED as MOOT. Per the Court’s Amended Order Granting Preliminary Approval of Amended Settlement (Docket #80), the parties shall revert to their respective positions as of March 4, 2004. SO ORDERED. Footnotes [1] Notably, the Class as alleged (and as previously approved for purposes of settlement) specifically excludes "(1) subscribers or dependents whose claims have already been resolved in the Court-approved class action settlements in Lecza v. Healthsource, Inc. et al., Civil Action No. C-95-382-JRM, U.S. District Court, District of New Hampshire or Garvin v. CIGNA Healthsource of South Carolina, Inc., Civil Action No. 94-CP-40-3472, South Carolina Court of Common Pleas, Richland County; and (2) those individuals who hold or have held any executive position at Health-source, and their immediate families.” (Am. Compl-¶ 114.) [2] In their application for attorneys’ fees, Class Counsel represents that when the case began, they believed that this case was "possibly a $50 million case.” (Class Counsel’s Second. Supp. Mem. in Support of Application for Award of Attorneys' Fees (Docket # 83) at 2.) [3] The relevant provision of the Settlement Agreement defines each class member's "Share of the Settlement” as “the lesser of (a) the predetermined amount of forty dollars ($40), which represents that average amount of alleged damaged recoverable by the Class Members pursuant to this Settlement, or (b) the Residual Corpus divided by the number of Valid Claims Forms Timely Signed and Returned.” (Settlement Agreement ¶ 1.20.) [4] The parties have supplied the Court with an affidavit from the mediator that also indicates that the parties had informed him "that they deliberately would not discuss the issue of fees at all until an agreement had been reached independently on a settlement of the litigation.” (Groff Aff. ¶ 14.) [5] The Court did express concern that Class Members would either not receive notice of the settlement or not return a claim form and hypothetically noted that, in light of the re-verter provision, Defendants might "get back $300,000.” (July 27, 2004 Hearing Tr. at 20.) In hindsight, the Court's hypothetical was a significant underestimation of the amount that would revert to Defendants. (As the Court explains later is this Order, a much larger sum would revert to CIGNA if the Court were to approve the proposed settle *41 ment in its current form.) However, at this hearing, none of the proponents of the settlement made any attempt to disclose their actual expectations with respect to the amount that would ultimately revert to CIGNA as a result of the anticipated claims rate. Nonetheless, at later hearings, it became clear that both sides had anticipated from the beginning that more than $300,000 would be returned to CIGNA pursuant to the terms of the Settlement Agreement. [6] Plaintiffs' counsel represented at the December 6, 2004 hearing that it had only learned of the substantial shrinkage in the number of identifiable Class Members while talking to the settlement administrator in November 2004. By comparison, Defendants knew of the actual number of Class Members shortly before September 15, 2004 (at the very latest). (See Dec. 6, 2004 Hearing Tr. at 43 (statement by Mr. Armando).) With the benefit and clarity of hindsight, it is clear to see that Defendants should have promptly alerted both Plaintiffs' counsel and this Court that Defense counsel's previous representations regarding the size of the class were incorrect. [7] In fact, Attorney McCutchen essentially agreed with the Court that the Settlement Agreement with the $40 maximum payout was not fair to the 39,942 Class Members and should not be approved by the Court. (See Dec. 6, 2004 Hearing Transcript (Docket # 64) at 28.) [8] Although this was the opt-out number originally supplied to the Court, the Proposed Final Judgment and Order of Dismissal (Docket # 88) lists only 173 individuals who have filed timely and valid opt-out forms. [9] This 28,259 figure amounts to a non-response rate of approximately 88 percent of the Class Members who apparently received the mailed notices. Suffice it to say, the Court considered this non-response rates both surprising and troubling. [10] (See Aff. of L. Stephens Tilghman, dated Nov. 17, 2004 (Ex. 1 to Docket #55) ¶ 8.) [11] Notably, in spite of the clear sailing clause quoted earlier in this memorandum, Defendants did object to Class Counsel’s Application and asked the Court to limit its award to 30 percent "of the settlement fund net of expenses.” (Defs.’ Mem in Opp. to Pis. Counsel's Request for Attorney Fees (Docket # 57) at 4.) The Court understood this to be a limited objection to Class Counsel’s request for reimbursement of expenses and costs in addition to attorneys’ fees totaling 30 percent of the settlement fund. [12] Notably, Mr. Tilghman is President of a company that has administered over 175 class action settlements, including the settlement in Garvin v. Healthsource Inc., Civil Action No. 94-CP-40-3472, South Carolina Court of Common Pleas, Richland County (a case pressing substantially similar claims to the claims presented here albeit for a smaller class of subscribers). At the July 27, 2004 Preliminary Approval Hearing, Attorney Hansel represented that both sides had previous experience working with Mr. Tilghman. (See July 27, 2004 Hearing Transcript at 6.) [13] In its colloquy with Prof. Sherman, the Court specifically referenced Judge Posner's discussion of the importance of deterrent effect of class actions. See Richard A. Posner, Economic Analysis of Law 626-27 (5th ed. 1998) ("[T]he most important point from an economic standpoint is that the violator be confronted with the costs of his violation— this achieves the allocative purpose of the suit — not that he pays them to his victims.”) Although Professor Sherman went on to explain his basis for believing that the proposed settlement achieves this deterrent effect, in this Court’s assessment, the proposed settlement falls more closely within the problem that Judge Posner went on to describe in which, “[t]he lawyer for the class will be tempted to offer to settle with the defendant for a small judgment and a large fee, and such an offer will be attractive to the defendant, provided the sum of the two figures is less that the defendant’s net expected loss from going to trial.” Id. at 627. [14] Justice O’Connor also explained that rever-sionary fund settlements that allow attorneys' fees to be based upon the total fund may "potentially undermine the underlying purposes of class actions by providing defendants with a powerful means to enticing class counsel to settle lawsuits in a manner detrimental to the class" and, in turn, “could encourage the filing of needless lawsuits.” International Precious Metals Corp., 530 U.S. at 1223, 120 S.Ct. 2237. Justice O'Connor’s statements in International Precious Metals were actually made in the context of a request to review an actual attorney fee award on a $40 million reversionary fund. In that case, the District Court approved an attorney fee award of $13,333,333.00 even though only $6,485,362.15 was disbursed to the Class with the remainder reverting to the defendants. See id. In denying review, Justice O'Connor noted that petitioners had waived their right to challenge the reasonableness of this award because of a clear sailing agreement in which they agreed not to oppose a fee request for up to one-third of the reversionary fund. See id. Thus, Justice O'Connor's statement simply highlights why the District Court must give particular scrutiny to reversionary fund agreements with clear sailing clauses at this stage. In short, International Precious Metals appears to stand for the proposition that once a reversionary fund settlement with a clear sailing provision is approved, it may well be impossible for other courts to address any apparent unfairness in the misallocation of the total payout between attorneys’ fees and actual payments to the class. [15] The Court believes that its declaration of a "presumption of unfairness" in this case is a logical extension of the "presumption of fairness” rule. However, the Court also recognizes that a "presumption of unfairness” is not a rule that has yet been adopted or even proposed in the First Circuit. Nonetheless, as discussed in the previous footnote, it appears unlikely that any proposed settlement containing these provisions would be the subject of a properly preserved appeal. In the absence of any precedent that declares a "presumption of unfairness” for settlement agreements containing reverter and clear sailing provisions, the Court’s declaration of a presumption of unfairness in this case serves as both an honest explanation of the basis for the Court's concern and a notice to future litigants that proposed class settlements with these provisions will be subjected to close scrutiny by this Court. [16] The Court has deducted an additional $75,000.00 in settlement administration expenses to arrive at this amount. [17] An additional example that less than complete discovery was completed prior to the proposed settlement being reached can be found in Defendants' Memorandum in Support of Approval of Settlement (Docket # 65). In that submission, Defendants discuss the risks Plaintiffs face in establishing their claims. The discussion cites to a subscriber agreement that has not yet been provided to Plaintiffs in discovery. (See Defs.' Mem. in Supp. of Approval of Amended Settlement at 5 & n.3.) Regardless of what this particular document shows regarding Plaintiffs' chance of success, it is fair to say that Plaintiffs did not have a chance to review or consider this document if it was not produced prior to the Settlement Agreement being finalized. [18] As discussed in other sections of this Order, Plaintiffs’ Application for Award of Attorneys' Fees actually seeks fees and expenses that total $1,231,238.25, which equals 36 percent of the total settlement fund. Defendants have filed a limited objection to this request that seeks to limit the Court's award of expenses and fees to 30 percent of the settlement fund. (See Defs.' Mem in Opp. to Pis. Counsel’s Request for Attorney Fees (Docket # 47) at 4.) [19] See, e.g., In re Compact Disc MAP Litigation, 216 F.R.D. at 211 (noting that in assess *50 ing a settlement a judge should not "prejudge the merits of the case” since the judge may be called on "to decide the merits if the settlement fails.”) [20] Without expressing any opinion on the merits of a yet-unfiled motion for class certification, Defendants do raise a valid question regarding whether a single class can consist of both subscribers to plans governed by ERISA and plans that are not subject to ERISA. To be sure, it is apparently unclear how many Class Members were subscribers to plans governed by ERISA and how many Class Members were subscribers to non-ERISA plans. Putting aside for a moment this sorting issue, to the extent some, but not all, Class Members subscribed to plans governed by ERISA ("ERISA Plaintiffs”), these ERISA Plaintiffs would be limited to the remedies provided under ERISA and could not pursue state law claims. See, e.g., Aetna Health Inc. v. Davila, 542 U.S. 200, 124 S.Ct. 2488, 2495-96, 159 L.Ed.2d 312 (2004). In contrast, subscribers to non-ERISA plans may well be able to pursue the other common law claims asserted in the Amended Complaint. There is at least a basis for arguing that these two categories of Plaintiffs with two different types of claims cannot — when combined into one class — satisfy the Rule 23(a)’s requirement of typicality or the Rule 23(b)(3)'s requirement of predominance of common questions. [21] Plaintiffs could be entitled to recover reasonable attorneys' fees in addition to this amount pursuant to the fee-shifting provisions in both ERISA, 29 U.S.C. § 1132(g), and' RICO, 18 U.S.C. § 1964(c). Thus, for purposes of this analysis the Court will not deduct a contingency fee from the $2.3 million damage estimate. Nonetheless, the Court acknowledges the arguments put forth by Defendants suggesting that Plaintiffs might have to pay attorneys’ fees out of any award they recovered. (See Defs.' Mem. in Support of Approval of Am. Settlement (Docket # 65) at 18.) In addition, for purposes of this analysis the Court has not considered Plaintiffs' requests for interest, punitive damages and treble damages, but notes that it is aware that Plaintiffs' Amended Complaint seeks these additional amounts. [22] In fact, Defendants’s own evaluation of the settlement and attempt to justify the current settlement as reasonable and fair assumes the chances of a defense verdict at somewhere between 45 percent and 60 percent. (See Defs.’ Mem. in Support of Approval of Am. Settlement (Docket # 65) at 22.) [23] In addition, the Court notes that even if the parties were to certify that the probability of success was less than 17 percent, the Court would be reluctant to approve the settlement as fair. As discussed later in this order, the $409,159.81 that the Class will recover under the current proposed settlement will be distributed to less than 20 percent of the Class. Thus, approximately 80 percent of the Class will receive nothing in exchange for their claims being extinguished. Thus, for this large majority of the class, the probability of success would need to be zero percent in order to justify finding that the settlement was fair under a Mars Steel analysis. [24] This amount includes the incentive payments to the Class Representatives. [25] See Wright, Miller & Kane, Federal Practice & Procedure: Civil 3d § 1797.1 & nn. 10 & 11 (2005) (collecting cases in which the allocation and distribution contemplated under the settlement were considered in deciding whether to approve settlement).