Doug SUTTON and Prescott Nottingham, on behalf of themselves and all others similarly situated, Plaintiffs, v. Robert F. BERNARD, et al., Defendants. Appeal of: Milberg Weiss Bershad & Schulman LLP, lead counsel for plaintiffs, on behalf of itself and all other counsel for plaintiffs, 504 F.3d 688

WILLIAMS, Circuit Judge. The legal claims in this multimillion dollar securities class action have been settled. The issue of attorneys’ fees for class counsel, however, has not. Having saved the calculation of the fee award for the end of the litigation, the district court concluded that the “degree of success” obtained for the class was the controlling factor in its decision. As we explain more fully below, the district court failed to approximate what the market would have paid the lawyers for their services had they negotiated their fee at the beginning of the case, an approach that our precedent requires.

Therefore, we vacate the district court’s decision and remand for a recalculation of the award. I. BACKGROUND On February 27, 2001, the district court appointed the law firm of Milberg Weiss Bershad & Schulman LLP (“Counsel”) as the lead attorneys for the plaintiffs in a securities fraud action against three former executive officers of the Internet consulting company Marchfirst, Inc.

The complaint alleged that the defendants had made false and misleading public statements that artificially inflated the price of the company’s stock, resulting in damages to those who had purchased or acquired Marchfirst stock between March and November of 2000. Marchfirst, which filed for bankruptcy shortly before the complaint was filed, was not named in the suit. In the district court’s February 27, 2001, order appointing Counsel, it reserved for later determination the issue of how attorneys’ fees would be awarded.

On December 17, 2001, the district court granted the plaintiffs’ motion for certification of the class, and discovery began. During this time, the Trustee in the Marchfirst bankruptcy proceeding filed an action against former directors and officers of the company, including the three defendants to the class action, alleging a breach of their fiduciary duties. Claiming priority over the class action, the Trustee succeeded in obtaining an injunction to stay those proceedings, except for discovery, pending the resolution of its action.

The class members found themselves competing with the Trustee for Marchfirst’s assets, so they, along with the defendants and the defendants’ insurer, took

part in a sixteen-month-long mediation. As a result of this process, an agreement was finalized on December 20, 2004, which provided for an $18,000,000 settlement for the class and a $6,000,000 settlement of the Trustee’s action. After receiving no objections to the settlement from the class, on March 24, 2005, Counsel petitioned the district court for compensation equal to 28% of the gross settlement amount, or $5,040,000, in fees.

Counsel explained its request by first noting that according to a study by National Economic Research Associates, the $18,000,000 settlement exceeded the median expected settlement by 50%. In addition, Counsel argued that the quality of its legal services was demonstrated by its ability to secure this amount in spite of the legal and factual complexities presented by the case.

Counsel further contended that its fee request was fair and reasonable because it represented 88% of its cumulative lodestar of $5,693,884 (the equivalent of Counsel’s hours spent on the case multiplied by its hourly rate), and was lower than fee percentage awards of 33% approved by other judges in the Northern District of Illinois. Counsel also told the court that it risked nonpayment at the outset of the litigation since its compensation depended on the success of the suit.

On July 12, 2006, before addressing Counsel’s fee award, the district court granted final approval of the settlement amount, declaring it “reasonable, considering the uncertain outcome of the legal and factual questions that would be involved in a trial of the case.... ” Sutton v. Bernard, 446 F.Supp.2d 814, 816 (N.D.Ill.2006). The court then stated that Counsel’s fee would be awarded from the net settlement amount of $17,360,491.44, a figure that excluded expenses, as opposed to the total recovery. Id. at 819-22 (citing Montgomery v. Aetna Plywood, Inc., 231 F.3d 399, 408 (7th Cir.2000)).

The court began its calculation of the appropriate fee percentage by looking to the “degree of success” Counsel had obtained for the class, a factor it considered crucial to its determination of an appropriate fee percentage. Id. at 820. Next, it cited Counsel’s representation that the settlement equaled a recovery of nineteen cents per share, which the court calculated would yield a distribution of $78 to a class member with the median number of shares, 400. Id. at 821.

Although the court regarded the settlement as the best Counsel could obtain “under the circumstances,” it did not believe that the small recovery could justify Counsel’s requested fee percentage of 28%, which it considered “excessive.” Id. at 821, 822. Citing its duty to render a fair judgment, the court decided that a fee of 15% of the net settlement amount, or $2,605,000, was the highest it could award. Id. at 822.

A few days after issuing its decision, the district court filed a supplemental opinion and stated that the Private Securities Litigation Reform Act of 1995, 15 U.S.C. § 78u-4(a)(6) (2007), “contemplate[s] a fee ... that is measured by the results that have actually been achieved in the case.” See Sutton v. Bernard (“Sutton II ”), 446 F.Supp.2d 823, 824 (N.D.Ill.2006). Counsel appeals, challenging the district court’s approach in determining that the appropriate attorneys’ fee was 15% of the settlement fund, barely half of Counsel’s fee request. II.

ANALYSIS We review the district court’s methodology de novo “to determine whether it reflects procedure approved for calculating awards.” Harman v. Lyphomed, Inc., 945 F.2d 969, 973 (7th Cir.1991). Our review of the final award is for an abuse of discretion, which “occurs when the court reaches an erroneous conclusion of law, fails to explain a reduction or reaches a conclusion that no evidence in the record supports as rational.” Id.

Once a settlement has been reached in a class action, the attorneys for the class petition the court for compensation from the settlement or common fund created for the class’s benefit. This method of payment is an exception to the “American Rule,” which requires each party to bear its own expenses. Florin v. Nationsbank of Ga., N.A., 34 F.3d 560, 562-63 (7th Cir.1994); Skelton v. Gen. Motors Corp., 860 F.2d 250, 252 (7th Cir.1988).

Also known as the common fund doctrine, this payment scheme “is based on the equitable notion that those who have benefited from litigation should share in its costs.” Skel- ton, 860 F.2d at 252 (internal quotations omitted). In deciding fee levels in common fund cases, we have consistently directed district courts to “do their best to award counsel the market price for legal services, in light of the risk of nonpayment and the normal rate of compensation in the market at the time.” In re Synthroid Mktg. Litig., 264 F.3d 712, 718 (7th Cir.2001) (collecting cases).

In Synthroid, we vacated the district court’s award after finding that it failed to follow our market-based approach when it capped counsel’s fee percentage at 10% on the basis that there was a large recovery involved. Id. at 717-18. Because the attorneys’ fee had not been determined up front in that case, we instructed the district court on remand to undertake an analysis of the terms to which the private plaintiffs and their attorneys would have contracted at the outset of the litigation when the risk of loss still existed. Id. at 718-19; see also In re Cont’l Ill. Sec.

Litig., 962 F.2d 566, 572 (7th Cir.1992) (the goal in awarding a reasonable attorneys’ fee is to give the lawyer what he would have received in an arms-length negotiation). A similar disposition is required in this case. Like the court in Synthroid, the district court here employed a methodology that does not reflect the market-based approach for determining fee awards that is required by our precedent. As described earlier, the district court’s analysis centered solely on the results achieved by Counsel to justify awarding a fee percentage of 15% of the common fund.

The court attributed its emphasis on this standard to Hensley v. Eckerhart, 461 U.S. 424, 436, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983), which involved the appeal of an award of attorneys’ fees to a group of plaintiffs who prevailed on five of their six constitutional challenges to the defendants’ practices.

In deciding the issue of whether a partially prevailing plaintiff could recover fees for legal services on unsuccessful claims, the Court held “that the extent of a plaintiffs success is a crucial factor in determining the proper amount of an award of attorney’s fees under 42 U.S.C. § 1988,” the applicable fee-shifting provision. Id. at 440, 103 S.Ct. 1933. The Court continued, “where the plaintiff achieved only limited success, the district court should award only that amount of fees that is reasonable in relation to the results obtained.” Id.

Based on this language, the district court decided that Counsel was not entitled to a fee percentage of 28% because it had achieved what the court considered to be only a small net recovery for the class. See Sutton, 446 F.Supp.2d at 820-21. In reaching this conclusion, however, the district court misapplied the principles that govern fee shifting cases to the common fund case before it.

As indicated above, Hensley involved an award of attorneys’ fees under 42 U.S.C. § 1988(b), which provides that in certain proceedings “the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney’s fee as part of the costs....” By contrast, in this case, the district court was required to determine a reasonable attorneys’ fee for Counsel to be paid out of the common fund. In Skelton, we highlighted a key difference between these two fee recovery schemes, namely the source of the plaintiffs’ attorneys’ fees. See 860 F.2d at 252.

As we observed, in statutory fee-shifting cases, the prevailing party, usually the plaintiff, recovers the attorneys’ fee from the defendant, whereas in a common fund case, the court awards the plaintiffs’ attorneys’ fees from the fund. Id. In Hensley, the Court emphasized the significance of the results obtained to avoid awarding excessive fees in cases where the plaintiff had only achieved limited success. See 461 U.S. at 436, 103 S.Ct. 1933.

In a common fund case, where the defendant’s liability is limited to the amount paid into the fund, there is no danger of unduly burdening that party with payment of the plaintiffs’ attorneys’ fee for time spent on unsuccessful claims. This is especially so considering that compensating Counsel from the common fund is premised on the idea that the plaintiffs and their counsel share in the litigation’s costs. See Skelton, 860 F.2d at 252.

We therefore find that the district court erred in holding that the “degree of success” standard articulated in Hensley applied to its calculation of a fee percentage in a common fund ease. Cf. Hensley, 461 U.S. at 433 n. 7, 103 S.Ct. 1933 (applying standard to cases where Congress has authorized a fee award to the prevailing party). The district court also sought to justify its methodology by citing to the

section of the Private Securities Litigation Reform Act that governs the award of attorneys’ fees and expenses in securities class actions. See Sutton II, 446 F.Supp.2d at 823. Under that provision, “[tjotal attorneys’ fees and expenses awarded by the court to counsel for the plaintiff class shall not exceed a reasonable percentage of the amount of any damages and prejudgment interest actually paid to the class.” 15 U.S.C. § 78u-4(a)(6). According to the district court, this language requires an ex post determination of the fee percentage that is measured by the results achieved for the class.

See generally Sutton II, 446 F.Supp.2d 823. We emphasize that we do not take issue with the timing of the district court’s fee determination, but rather, with the method used. Because the court chose to wait until the end of the litigation, it was required to set the fee by estimating what the parties would have agreed to had negotiations occurred at the outset. Synthroid, 264 F.3d at 719.

The trouble we have with the district court’s methodology is that the fee determination began and ended with the amount actually recovered for the class; the court did not consult the market for legal services for guidance in what constituted, as an abstract matter, a “reasonable percentage.” Cf. Missouri v. Jenkins, 491 U.S. 274, 285, 109 S.Ct. 2463, 105 L.Ed.2d 229 (1989) (in determining attorneys’ fees, “we have consistently looked to the marketplace as our guide to what is ‘reasonable’ ”).

By looking to the marketplace, the district court avoids assigning a value “based on nothing more than a subjective judgment regarding [Counsel’s] work.” Montgomery, 231 F.3d at 408; see also In re Cont’l, 962 F.2d at 568 (in failing to follow the market, district court judge “thought he knew the value of the class lawyers’ legal services better than the market did”).

We have said that the market price for legal fees “depends in part on the risk of nonpayment a firm agrees to bear, in part on the quality of its performance, in part on the amount of work necessary to resolve the litigation, and in part on the stakes of the case.” Synthroid, 264 F.3d at 721. Here, the district court dismissed Counsel’s contention that it faced a risk of nonpayment for its services by declaring, “[t]his is a common argument in support of large fees in class actions, but it has no relationship to reality,” and adding that class actions rarely go to trial and that they all settle.

Sutton, 446 F.Supp.2d at 822. In Continental, we rejected this rationale as an inadequate basis for the district court’s refusal to account for the risk of loss in its fee award. 962 F.2d at 569-70. We recognized that there is generally some degree of risk that attorneys will receive no fee (or at least not the fee that reflects their efforts) when representing a class because their fee is linked to the success of the suit. Id. Because the district court failed to provide for the risk of loss, the possibility exists that Counsel, whose only source of a fee was a contingent one, was undercompensated.

See id. at 569. A review of the shortcomings in the district court’s methodology leads us to conclude that the coui't abused its discretion in awarding Counsel’s fee percentage. The district court erroneously concluded that its fee determination was controlled by the “degree of success” achieved for the class. In addition, it did not factor into its assessment the value that the market would have placed on Counsel’s legal services had its fee been arranged at the outset. Because the district court’s fee award of 15% of the common fund is based on an erroneous conclusion of law, it must be vacated. III.

CONCLUSION The judgment of the district court is Vacated and the case is Remanded for further proceedings consistent with this opinion. On remand, Circuit Rule 36 shall apply. . Counsel also sought reimbursement of $757,274.23 in expenses it incurred during the litigation. The district court approved expenses of $639,508.56, which Counsel does not appeal. .

We also pointed the district court to the following benchmarks to aid in its determination on remand of what might have occurred ex ante: (1) actual fee agreements; (2) data from large common fund cases where the parties negotiated the fees privately, and (3) bids and results from class counsel auction cases for insight into the fee levels attorneys in competition were willing to accept. Synthroid, 264 F.3d at 719-21.