On Application for Rehearing
The opinion of June 30, 2000, is withdrawn, and the following is substituted therefor.
Union Fidelity Life Insurance Company ("Union Fidelity") appeals from the trial court's award of $915,000 in attorney fees to class counsel based on a class-action settlement in a case in which the evidence indicated that an extremely small percentage of the class would eventually claim the benefits of the settlement.
In January 1996, Norman McCurdy, individually and on behalf of a class of all persons similarly situated, sued Union Fidelity *Page 188 and others,1 alleging fraud in the procurement of credit-life insurance by failing to disclose that premiums exceeded amounts allowed under Alabama law. In November 1996, this Court, in an unrelated proceeding, established the defendant's liability for such conduct. See McCullar v. Universal Underwriters LifeIns. Co., 687 So.2d 156 (Ala. 1996).
After initial proceedings on class certification, the parties entered into a settlement agreement purportedly representing a class of nearly 104,000 individuals with claims averaging approximately $43 and providing a potential class recovery of $4.5 million.2 Under the terms of the settlement, Union Fidelity agreed to pay reasonable attorney fees over and above the $4.5 million available to the class. The settlement provided that class counsel's attorney fees would be fixed by the trial court at a hearing subsequent to final approval of the settlement. Each party reserved the right to appeal the trial court's award of attorney fees.
On December 9, 1999, the trial court preliminarily approved the settlement, before notice was given to members of the class. With regard to notice, Union Fidelity claimed that, because of the nature of its relationship with the class members, it did not have readily available each member's last known address. Union Fidelity asserted that developing a list of such addresses, while not impossible, would take "thousands of hours." We can find nothing in the record indicating that class counsel challenged this assertion. On January 20, 1999, some 4,080 individual notices of settlement were sent to class members who had previously given their addresses to Union Fidelity for unrelated reasons. On January 31 and February 7, 1999, the remaining approximately 100,000 class members were given notice of the pendency of the action and preliminary approval of the settlement through publication in 12 Alabama newspapers. Class members were given until February 23, 1999, to opt out of the settlement, and, if they did not, they had until June 14, 1999, to file a claim accompanied by the required documentation.
On March 24, 1999, the trial court conducted a final fairness hearing on the proposed settlement and also heard evidence from both parties on the attorney-fee issue. During the hearing affidavits were filed indicating that, as of the date of the hearing, only 113 claims had been made, resulting in potential disbursements in the amount of $4,520.16. Nevertheless, with no objections having been presented and no one having opted out, the trial court gave the settlement final approval on April 1, 1999. The class members were notified that they could opt out of the settlement if they so chose, but that, in order to participate, they would have to submit a claim form along with supporting documentation.
On April 15, 1999, the trial court conducted a hearing on determining a reasonable attorney fee and awarded class counsel a fee of $915,000. Union Fidelity appeals only the attorney-fee award.
Abuse of discretion is the applicable standard for reviewing the trial court's award of an attorney fee. See Ex parte Horn, 718 So.2d 694, 702 (Ala. 1998); and Edelman Combs v. Law, 663 So.2d 957, 958 (Ala. 1995).
The issue before us is whether the trial court abused its discretion in awarding *Page 189 $915,000 in attorney fees based primarily upon the common-fund theory, when attorney fees were not to be taken as a percentage of the class recovery, class notice was given primarily by publication, and only a minor percentage of the class had filed claims.
Alabama recognizes the American Rule, which generally requires each party to pay its own attorney fees. See Horn, 718 So.2d at 702. This rule applies to class actions when "the plaintiff's successful litigation confers `a substantial benefit on the members of an ascertainable class, and where the court's jurisdiction over the subject matter of the suit makes possible an award that will operate to spread the costs proportionately among them.'" Hall v. Cole, 412 U.S. 1, 5 (1973) (quoting Mills v. Electric Auto-Lite Co., 396 U.S. 375, 393-94 (1970)). This Court has recognized that attorneys who recover an award for the class are entitled to a reasonable fee for their services.Edelman Combs, 663 So.2d at 958. When a class benefits through the use of Rule 23, Ala.R.Civ.P., the class generally bears the costs associated with the litigation, out of the proceeds collected through the litigation. Id. Where class counsel has provided a benefit to the class and an award of an attorney fee is appropriate, some courts have used a percentage of the common fund generated as the starting point for determining an appropriate attorney fee, see Edelman Combs, 663 So.2d at 959, while other courts have worked from a benchmark based on the number of hours expended, the so-called "lodestar," see Lindy Bros.Builders, Inc. v. American Radiator Standard Sanitary Corp., 487 F.2d 161 (3d Cir. 1973).
As we have said before, Alabama will look to federal law in interpreting this most complex area of litigation. Adams v. Robertson,676 So.2d 1265, 1268 (Ala. 1995). However, we note that federal law is merely persuasive and a starting point for our evaluation.
The trial court's award was based primarily on the common-fund theory, by which attorney fees in a class action are to be based on a reasonable percentage of the fund generated for the benefit of the class. Edelman Combs, 663 So.2d at 959. Normally, attorney fees paid under the common- fund theory are deducted from the amount recovered and are not an additional sum due from the defendant.3 See Camden I CondominiumAss'n, Inc. v. Dunkle, 946 F.2d 768, 774 (11th Cir. 1991) ("[A] key element of the common fund case is that fees are not assessed against the unsuccessful litigant (fee shifting), but rather, are taken from the fund or damages recovery (fee spreading).").
While this Court has held that the common-fund approach is thepreferred method for calculating attorney fees in class actions, seeEdelman Combs, supra, the decision whether to apply the common-fund approach or the lodestar approach is within the trial court's discretion. See, e.g., Goldberger v. Integrated Resources, Inc., 209 F.3d 43 (2d Cir. 2000). We must determine whether the common-fund *Page 190 approach was appropriate in this case.
Union Fidelity agreed to pay reasonable attorney fees separate from any amount paid to the class.4 The $915,000 award constitutes just over 20% of what class counsel called the "common fund." This amount is well within the general guidelines set forth by this Court. See Edelman Combs, 663 So.2d at 961 (citing Reynolds v. First Alabama Bank ofMontgomery, N.A., 471 So.2d 1238 (Ala. 1985), for the general rule that sets attorney fees between 20% and 25% of the common fund). Union Fidelity argues that the trial court abused its discretion in applying the common-fund approach because, it says, no actual fund was created, and, therefore, it says the trial court should have used the lodestar method in determining the fees in this case.
We do not find the existence of a separate fund determinative of the relevance of the common-fund approach. It is clear that Union Fidelity admitted liability for $4.5 million. A solvent company that agrees to pay reasonable attorney fees over and above the amount of an admitted liability should not, by failing to create an escrow account, forfeit the application, if only by analogy, of standards pertaining to computation of reasonable attorney fees based upon recovery of a common fund. See In re General Motors Corp. Pick-Up Truck Fuel Tank Prods.Liability Litigation, 55 F.3d 768, 821 (3d Cir. 1995) ("Courts have relied on `common fund' principles and the inherent management powers of the court to award fees to lead counsel in cases that do not actually generate a common fund.").
Class counsel argue that they created a $4.5-million common fund for the benefit of the class. Specifically, they argue that Union Fidelity created an internal fund through which it intended to pay all claims made against the settlement. While provisions for establishing such a fund are not included in the settlement agreement, class counsel cite an interrogatory response in which Union Fidelity stated that a fund was created for this purpose. Union Fidelity, on the other hand, insists that its interrogatory response was a typographical error, and it filed affidavits indicating that no such fund was, in fact, created. Class counsel further argue that, under the common-fund approach, the attorney fees should be computed in light of the size of the potential recovery to the class, without regard to sums actually recovered by the class. Class counsel cite Boeing Co. v. Van Gemert, 444 U.S. 472 (1980), andCity of Ozark v. Trawick, 604 So.2d 360, 364 (Ala. 1992), for the proposition that attorney fees should not be contingent upon the actual benefits paid to the class.
In Boeing, class notice was primarily obtained by means other than publication.5 In Trawick, there is no indication that publication was the primary means of notice.6 Boeing and Trawick both turn on the conclusion that, once informed, each class member is free to make a conscious decision to forgo the benefits bestowed upon him or her as a member of the class. *Page 191 See Trawick, 604 So.2d at 364. We agree that an informed decision not to participate should not diminish the award to class counsel, because the benefit in such circumstances has actually been provided, and the class member has simply chosen not to reap the rewards due him or her. However, this case is very different from both Boeing and Trawick. Here, roughly 4% of the just over 100,000 class members received notice of the settlement by mail. The parties concluded that determining addresses for notice by mail to the other 96% of the class was too difficult and time-consuming. Therefore, notice to the remaining class members was effected through two-day publication in several Alabama newspapers.
While notice by publication may have been the best notice practicable, its effect in this case produced results dramatically different from those obtained in Boeing and Trawick. Twenty percent of the admitted liability was disbursed to the class in Boeing. While the opinion in Trawick does not state the total amounts paid to the class, nearly 90% of the class members participated in the Trawick settlement. At the hearing on attorney fees in this case, the parties contemplated payment of less than 1% of the admitted liability.7 Because of the extremely low percentage of participation, we simply do not have in this case the same level of confidence we had in Trawick that the absence of claims implies that class members made informed decisions not to participate in the settlement.8 Thus, because of the extremely low percentage of participation, we cannot follow the general rule that normally applies in cases such as Boeing and Trawick, where the absence of participation was held not to affect the computation of a reasonable fee under the common-fund doctrine.9 We therefore hold that, because the percentage of class participation was low enough to undermine the assumption of an informed decision, the trial court abused its discretion in basing attorney fees primarily on the common-fund approach. We further hold that, when a trial court is faced with such circumstances as are present in this case, it shall employ the lodestar analysis to determine appropriate attorney fees.
However, because the trial court also buttressed its attorney-fee determination with the conclusion that it would arrive at the same fee based upon the lodestar analysis, we must review the trial court's consideration of an appropriate multiplier to be applied to the lodestar.
The lodestar method relies heavily upon the time expended by class counsel. This method was adopted by the United States Court of Appeals for the Third Circuit as the preferred method for calculating attorney fees in class actions. Lindy Bros., supra. Under the lodestar approach, *Page 192 the trial court must first determine the number of hours reasonably spent by class counsel on the matter, and then multiply those hours by a reasonable hourly rate. Goldberger, 209 F.3d at 47. Once this lodestar figure has been determined, the court can adjust it upward or downward, depending on a variety of factors, eventually arriving at a final fee.Id. This adjustment is called a multiplier. Id. In this case, the trial court added a multiplier of two under the lodestar method in order to justify the fee awarded. We review the reasonableness of such a multiplier under the circumstances of this case.
In Peebles v. Miley, 439 So.2d 137 (Ala. 1983), this Court established several factors to be used in determining the reasonableness of an attorney fee. These same factors should be used to determine whether a multiplier is warranted when a court is calculating attorney fees in the class-action setting. See Edelman Combs, 663 So.2d at 960 (applying the Peebles factors generally to class actions). In Edelman Combs, this Court listed the Peebles factors:
"(1) `the nature and value of the subject matter of the employment';
"(2) `the learning, skill, and labor requisite to its proper discharge';
"(3) `the time consumed';
"(4) `the professional experience and reputation of the attorney';
"(5) `the weight of [the attorney's] responsibilities';
"(6) `the measure of success achieved';
"(7) `the reasonable expenses incurred by the attorney';
"(8) `[w]hether the fee is fixed or contingent';
"(9) `[t]he nature and length of a professional relationship';
"(10) `[t]he fee customarily charged in the locality for similar legal services';
"(11) `[t]he likelihood that a particular employment may preclude other employment'; and
"(12) `[t]he time limitations imposed by the client or by the circumstances.'"
663 So.2d at 959-60 (quoting Peebles, 439 So.2d at 140-41).
The actual benefit to the class falls under Peebles factor number 6 ("measure of success achieved") and is an extremely important factor that must be examined with close scrutiny when the benefit to the class is extremely low. The record before us simply does not give sufficient justification for the multiplier used by the trial court. We therefore reverse the order awarding the fee and remand with instructions and guidance concerning further proceedings.
We have no information before us concerning the benefit to the class other than the text of the settlement and the fact that only 113 claims had been submitted by class members, totaling $4,520.16, as of the date of the hearing on attorney fees. We do not have before us the total number of claims submitted or dollars paid from April 15 to June 14, 1999, the end of the claims period. However, the record indicates that the trial court was aware that claims were likely to only reach about $10,000, as previously noted, less than 1% of the potential liability. Also, we have no data supporting the allegations that the difficulty in determining addresses of class members outweighed the benefit that notice by first-class mail would have rendered to the vast majority of the class members who were entitled to recover their portion of the $4.5- million liability.
We do know that each claimant is required to submit various documents in order to participate in the recovery, and we know that the average claim is approximately $43. In Goodrich v. E.F. Hutton *Page 193 Group, Inc., 681 A.2d 1039 (Del. 1996), the Delaware Supreme Court affirmed the trial court's attorney-fee award. The Court distinguished Boeing and expressed the importance of considering the actual benefits to the class when setting attorney fees when small claims are involved and recovery is dependent upon the submission of substantial documentation:
Goodrich, 681 A.2d at 1049."[I]n this instance . . . the particular facts of the case raise a very significant risk that some portion and perhaps a large portion of the total fund available for payments to the class will in fact not be distributed to the class members. This is so for several reasons. First the class is very large and the `losses' that class members may have suffered, if the claims made are assumed to be valid, are quite small. Moreover the transaction costs that class members will necessarily encounter in making a claim are relatively high. Members of the class will need to have complete brokerage records going back over the entire period to make a full claim and will need to locate and search such records. It is quite likely that many and perhaps most class members will not be able to or motivated to make such a claim. In fact, . . . for some class members with smaller transaction totals, if one assumes even a modest opportunity cost in searching out old records, it will be economically irrational to make a claim. Thus, the order limited the fee awarded to one-third of the benefits actually delivered to the class members."
In order to file a valid claim in this present case, the settlement required each class member to
"complete and mail the proof of claim along with a copy of [their] certificate of insurance and loan documents, including the retail sales contract, evidencing the purchase of credit life insurance issued by any of the defendants, and documents evidencing the net amount of the loan, the interest rate of the loan, and the total of payments scheduled on the loan. . . ."
We are not saying that, under the circumstances of this case, the benefit to the class should be discounted because of unwarranted complexity of the claims process. Nor are we holding that the benefit to the class should be viewed solely in terms of the amounts actually claimed. We simply express our concern that Union Fidelity may have had copies of some or all of such records with respect to each class member and needed only an address and perhaps less documentation in order to disburse the settlement. If this was the case, then Union Fidelity could be vulnerable to charges that it negotiated successfully for the inclusion of unduly onerous requirements so as to keep down the number of class members who would be willing or able to submit well-documented claims for comparatively small sums. We merely direct the trial court's attention to the specifics of the claims process in order to fully identify the actual benefits to the class and weigh them appropriately under the Peebles factors listed above.
To aid the trial court in its further analysis of the Peebles factors, specifically in its determining if a multiplier is warranted under the lodestar method of setting an attorney fee, we call to the trial court's attention Strong v. BellSouth Telecommunications, Inc., 173 F.R.D. 167 (W.D.La. 1997), aff'd, 137 F.3d 844 (5th Cir. 1998). On account of the relatively minimal benefits bestowed upon the class, the trial court inStrong denied a recommendation by both parties to increase attorney fees for class counsel by $1.5 million. The court looked at the actual benefits bestowed upon the class, in comparison to the "potential *Page 194 maximum value of the settlement," and stated:
Strong, 173 F.R.D. at 172. As in this present case, on appeal class counsel in Strong cited Boeing and argued that the district court abused its discretion by establishing the attorney fees based on the actual benefits received by class members, rather than on the entire common fund. 137 F.3d at 844. In Strong, the United States Court of Appeals for the Fifth Circuit distinguished Boeing, determining that a common fund had not been created. Id. at 851. The Fifth Circuit also affirmed the trial court's decision not to increase the fee, holding that the trial court had not abused its discretion in weighing the actual benefit to the class when determining whether to apply a multiplier to its initial lodestar figure. Id. Again, we do not suggest that the pejorative labels placed on class counsel in Strong are appropriate in this case, because we do not have an adequate record from which to draw positive or negative conclusions. But, on remand, development of the record is essential because, on its face, the amount of attorney fees in this case appears to be grossly disproportionate to the actual benefits bestowed upon the class, and, therefore, a multiplier of two would appear not to be justified."In this case, $66 million represented the potential maximum value of the settlement if all class members were eligible for a credit and returned a claim. In fact, the value of the actual credit requests submitted was $1,718,594.40. As we have noted that there may be intangible benefits to the class, we will adjust this valuation figure up to $2 million.
"A request for $6 million in attorneys' fees where counsel has provided no more than $2 million in benefits to the class is astonishing. It is a sad day when lawyers transmogrify from counselors into grifters. Suffice it to say that we find the request unreasonable."
Class counsel argue that their fee is justified because no one objected to the settlement and because the trial court approved the settlement as fair, reasonable, and just.10 We note that, by setting fees over and above any recovery paid to the class instead of setting fees as a percentage to be deducted from the recovery, each class member has the potential to recover 100% of his or her damages. While this methodology provides the highest recovery for each class member and commendably puts the attorney fees to one side while the issue of fairness of the settlement is resolved, it nonetheless makes possible a scenario in which there will be no objectors from the class membership at the hearing on attorney fees. When the defendant pays the attorney fees, separate and unrelated to the benefits available to the class, the class members have no incentive to challenge the fee because it will not diminish their recovery. Therefore, only the settling defendant is in a position to oppose the attorney-fee award; yet, it may have a conflict of interest in terms of offering candid evidence concerning actual benefits to the class. That same party will have recently joined hands with class counsel in extolling the virtues of the settlement, in order to gain the trial court's approval at the fairness hearing mandated by Rule 23(e), Ala.R.Civ.P. A defendant would be reluctant to denigrate the adequacy of notice or benefit to the class, *Page 195 because doing so may undermine the settlement altogether and thereby defeat its attempt to buy its peace, a consideration perhaps more important than concerns over paying excessive attorney fees.
The trial court is, therefore, forced to determine the appropriate fee award without the benefit of a fully adversarial process. We decline to leave the defendant to suffer the consequences of an arrangement that requires the defendant to pay an inflated attorney fee. Rule 23(e) requires judicial approval of an award. Fees awarded by the court in class-action settlements should be fair and reasonable when viewed in light of the actual benefits to the class. Attorney fees in class actions should not expose the judicial system to contempt and ridicule in an already skeptical era.11
To assure the requisite adverseness and impartiality in determining whether a multiplier is warranted in light of the actual benefits provided, the trial court should appoint an expert, through the use of Alabama Rule of Evidence 706, and/or a special master pursuant to Alabama Rule of Civil Procedure 53, to evaluate the settlement, to point out any deficiencies, and to recommend a reasonable attorney fee based on the actual benefit to the class. See Edelman Combs, 663 So.2d at 961 (appointing an "attorney ad litem" to represent the interests of the class as to the settlement, including the amount of attorney fees); andGoldberger, 209 F.3d at 56 (using special masters to examine the benefits to the class and the reasonableness of attorney fees). Such an appointment is appropriate in this proceeding.
Summary Although the common-fund approach remains the preferred method for calculating attorney fees in class actions in Alabama, when the percentage of the class members responding to the class notice is so low as to undermine the assumption of an informed decision, we distinguishTrawick and hold that the trial court must use the lodestar method to calculate appropriate attorney fees. In using that method the court is to consider the benefits of the settlement, along with the remainingPeebles factors, in determining whether a multiplier is warranted under the lodestar analysis, whether it be positive, negative, or neutral. Under such circumstances as are here presented, where we have concerns that the failure of class members to participate in the claims process may not be attributable to an informed decision by each class member, the court must closely scrutinize the fairness of the settlement and the true benefit to the class, in order to set appropriate attorney fees. An independent party vested with authority under Rule 706, Ala. R. Evid., and/or Rule 53, Ala.R.Civ.P., would best assist the court in being certain that its decision is properly informed.
APPLICATION FOR REHEARING OVERRULED; OPINION OF JUNE 30, 2000, WITHDRAWN; OPINION SUBSTITUTED; REVERSED AND REMANDED.
Hooper, C.J., and Maddox, Houston, See, and Brown, JJ., concur. *Page 196
England, J., dissents from the order overruling the application for rehearing; and concurs in part and dissents in part.
Cook and Johnstone, JJ., dissent.