KLINE, P.J.

Appellants Richard U. Lealao and his wife Sese Lealao, who commenced this successful consumer class action, claim that the attorney fees awarded class counsel were unreasonably low. The questions before us are whether, under the circumstances of this case, the trial court had discretion to award a fee based solely on a percentage of the class benefit or, in the alternative, to measure an award calculated under the lodestar methodology by a percentage-of-the-benefit yardstick and to adjust the lodestar upward or downward on that basis. Our answers are no to the first question and yes to the second.

FACTS AND PROCEDURAL HISTORY

Appellants filed their class action complaint in the San Francisco Superior Court on October 3, 1995. The complaint claimed that respondent, Beneficial California, Inc., a major lender which then had 130 offices in this state, wrongfully charged a penalty when persons with whom it entered into certain credit line account agreements prepaid their open-end loans, as the imposition of such a penalty was neither provided for in the agreement nor otherwise authorized. For example, in August 1993, appellants opened a credit line account with respondent that was secured by their home. When they sold the home in March 1995, respondent imposed a prepayment penalty in the amount of $8,048.20.

The original complaint alleged causes of action for (1) breach of contract, (2) breach of the duty of good faith and fair dealing, (3) money had and received, (4) unjust enrichment, (5) conversion, and (6) unfair business practice. The first amended complaint substituted a claim under the Consumer Finance Lender's Law (Fin. Code, § 24500 et seq.) for that alleging breach of the duty of good faith and fair dealing. The trial court granted respondent's demurrer without leave to amend as to the fifth cause of action for conversion, which was thereupon dismissed. On June 3, 1996, after a subsequent demurrer was overruled in its entirety, respondent filed its answer, which raised the defenses of laches, waiver and estoppel, unclean hands, comparative fault, lack of standing, that violation of the Consumer Finance Lender's Law was unintentional, and the statute of limitations. At the same time, respondent filed a cross-complaint against appellants alleging that because they had been told there would be a prepayment penalty respondent was entitled to reform the parties' credit line agreement on the ground that it failed to reflect the true intent of the parties.

In February 1997, appellants moved for class certification. Respondent resisted the motion chiefly on the ground that issues particular to individual class members predominated and that a fact based inquiry needed to be made with respect to particular loan agreements. Respondent also maintained that appellants' claims were not typical and they were therefore not competent to represent the class. On August 13, 1997, the trial court rejected respondent's arguments and issued an order granting class certification. The class consisted of borrowers on 6,698 loans opened by respondent between October 1991 and November 1994 and closed on or before February 28, 1998, all of whom were charged prepayment penalties. Each loan was secured by real property, generally the borrower's home. The total amount of prepayment penalties imposed on the class was approximately $19,200,000; the average penalty was $2,866.53.

The settlement ultimately reached by the parties provided that respondent would pay members of the class who filed claims 77 percent of the amount they paid as prepayment penalties. Since prepayment penalties of approximately $19,200,000 had been paid by the class, respondent effectively agreed to pay $14,784,000 if every member of the class filed a valid claim.

The settlement agreement did not require that class counsel's fee be deducted from the refunds received by members of the class, nor did it specify the amount of the fee class counsel would receive. With respect to attorney fees, the agreement provided only that respondent would pay class counsel "such reasonable attorneys' fees and costs as determined by the Court."

Class counsel emphasize that even though their fee was not to be paid directly out of the class recovery, they always contemplated it would be paid out of the unclaimed residual amount of funds available to the class, as may be done in cases in which a common fund is established. (See, e.g., Boeing Co. v. Van Gemert (1980) 444 U.S. 472, 480-481; Williams v. MGM Pathe Communications Co. (9th Cir. 1997) 129 F.3d 1026, 1027.) This understanding was communicated to members of the class in the notice of settlement approved by the court. The notice explained that respondent agreed to pay reasonable attorney fees as determined by the court, and that class counsel would seek fees in the amount of $3.5 million, which was approximately 24 percent of the "fund" created for the benefit of the class. As stated in the notice, "[c]lass counsel contends that the attorneys' fees will come out of the alleged class settlement fund of $14,784,000.00 because it is anticipated that this amount will be sufficient to pay class members who file valid claim forms and the attorneys' fees as determined by the court." The notice went on to state, however, that respondent "does not agree that a common fund has been created and contends that any attorneys' fees must be based on the amount of time spent by plaintiffs' counsel on the case."

Appellants advanced dual theories in support of their initial request for an award of $3.5 million. The first, which was consistent with the view of counsel set forth in the notice to the class, was that appellants' attorneys succeeded in creating a fund of $14,784,000 for the benefit of the class and were entitled to approximately 24 percent of that amount or $3.5 million. Counsel alternatively maintained this amount was also justified under the lodestar method of calculating fees, because the size of a class recovery may be used as a factor to enhance a lodestar award. Declarations filed in support of the application for fees stated that the reasonable hourly fees of appellants' attorneys ranged from $350 to $225, and that applying these rates to the more than 1,450 hours they expended on the case produced a "base lodestar" of $418,343.25. Pointing out that trial courts have broad discretion to enhance this lodestar upward to take into account various risks undertaken by counsel (see Serrano v. Priest (1977) 20 Cal.3d 25, 49 ( Serrano III)), class counsel argued that a multiplier of 8, which would roughly justify the $3.5 million award they sought, was appropriate.

In its October 30, 1998, order approving the settlement and granting attorney fees and costs, the trial court determined that no common fund had been established and declined to utilize any multiplier to enhance fees under a lodestar calculation. Looking only to the hourly rates claimed by appellants' attorneys and paralegals, and the time they expended on the case, the court awarded reasonable attorney fees in the amount of $425,000.

On December 21, 1998, appellants filed a motion for new trial limited to the issue of attorney fees. Based on new information that the valid claims then actually filed by members of the class totaled $7.35 million, counsel used this amount rather than the amount of potential claims respondent agreed to pay ($14,748,000) as the basis of a percentage fee. Instead of the $3.5 million fee they originally sought, class counsel requested 24 percent of $7.35 million, or $1.76 million. Counsel urged the court to reach this result by adjusting the lodestar figure ($425,000) upward to $1,487,500 by a multiplier of 3.5 "based among other things on the result obtained by counsel as shown by actual payments to be made to the class."

The order granting fees of $425,000 solely on the basis of the hours counsel expended on the case, and the order denying a new trial as to attorney fees, do not set forth the court's reasoning. However, the court's orders and comments from the bench indicate its belief that in a class action case such as this in which the class benefits are not in the form of a separate fund out of which fees are to be paid — a court has no discretion to award a percentage fee. Concluding it was compelled to employ the lodestar formula for calculating a reasonable attorney fee, the court apparently agreed with respondent that it could use a multiplier to adjust the lodestar upward only in cases in which the litigation resulted in a great public benefit or the litigation was particularly complex and difficult, and that this was not such a case.

DISCUSSION I.

A trial court's determination of reasonable attorney fees is reviewed under the abuse of discretion standard. ( Westside Community for Independent Living, Inc. v. Obledo (1983) 33 Cal.3d 348, 355.) "The scope of discretion always resides in the particular law being applied, i.e., in the `legal principles governing the subject of [the] action . . . .' Action that transgresses the confines of the applicable principles of law is outside the scope of discretion and we call such action an `abuse' of discretion." ( City of Sacramento v. Drew (1989) 207 Cal.App.3d 1287, 1297.) Accordingly, the question before us is whether the trial court's refusal to either award a percentage fee or consider the percentage-of-the-benefit approach in connection with its lodestar calculation is consistent with the applicable law.

II.

The parties' dispute centers on the validity and continuing jurisprudential viability of the distinction between "fee shifting" and "fee spreading."

In so-called "fee shifting" cases, in which the responsibility to pay attorney fees is statutorily or otherwise transferred from the prevailing plaintiff or class to the defendant, the primary method for establishing the amount of "reasonable" attorney fees is the lodestar method. The lodestar (or touchstone) is produced by multiplying the number of hours reasonably expended by counsel by a reasonable hourly rate. Once the court has fixed the lodestar, it may increase or decrease that amount by applying a positive or negative "multiplier" to take into account a variety of other factors, including the quality of the representation, the novelty and complexity of the issues, the results obtained, and the contingent risk presented. (See Pearl, California Attorney Fee Awards (2d ed. 1998) §§ 13.1-13.7).

Fee spreading occurs when a settlement or adjudication results in the establishment of a separate or so-called common fund for the benefit of the class. Because the fee awarded class counsel comes from this fund, it is said that the expense is borne by the beneficiaries. Percentage fees have traditionally been allowed in such common fund cases, although, as will be seen, the lodestar methodology may also be utilized in this context.

Respondent maintains that, because the settlement in this case did not result in the establishment of a traditional common fund, the percentage-of-the-benefit approach cannot be utilized, even in connection with the lodestar formulation. Appellant, conceding a conventional common fund does not exist, answers with federal cases suggesting that the distinction between "fee shifting" and "fee spreading" is an illusory jurisprudential construct, and that unless the fee award is in some fashion considered as a percentage of the monetary benefit received by the class the law will create counterproductive economic incentives and disincentives that should no longer be tolerated.

The primacy of the lodestar method in California was established in 1977 in Serrano III, supra, 20 Cal.3d 25. Adopting the view at that time of the Second and Third Circuits, our Supreme Court declared: "`The starting point of every fee award, once it is recognized that court's role in equity is to provide just compensation for the attorney, must be a calculation of the attorney's services in terms of the time he has expended on the case. Anchoring the analysis to this concept is the only way of approaching the problem that can claim objectivity, a claim which is obviously vital to the prestige of the bar and the courts.'" ( Id. at p. 48, fn. 23, quoting City of Detroit v. Grinnell Corp. (2d Cir. 1974) 495 F.2d 448, 470, abrogated in Goldberger v. Integrated Resources, Inc. (2d Cir. 2000) 209 F.3d 43, and citing Lindy Bros. Builders of Phila., Inc. v. American Radiator Standard Sanitary Corp. (3d Cir. 1973) 487 F.2d 161, 167-169 ( Lindy I).)

Despite its primacy, the lodestar method is not necessarily utilized in common fund cases. The common fund or "fund-in-court" doctrine, first articulated by the United States Supreme Court in Internal Imp. Fund Trustees v. Greenough (1881) 105 U.S. 527, is a venerable exception to the general American rule disfavoring attorney fees in the absence of statutory or contractual authorization. (Code Civ. Proc., § 1021) The exception "is grounded in `the historic power of equity to permit the trustee of a fund or property, or a party preserving or recovering a fund for the benefit of others in addition to himself, to recover his costs, including his attorneys' fees, from the fund of property itself or directly from the other parties enjoying the benefit.'" ( Serrano III, supra, 20 Cal.3d at p. 35, quoting Alyeska Pipeline Co. v. Wilderness Society (1975) 421 U.S. 240, 257, italics added.)

Because the common fund doctrine "rest[s] squarely on the principle of avoiding unjust enrichment" ( Serrano v. Unruh (1982) 32 Cal.3d 621, 632 ( Serrano IV); Boeing Co. v. Van Gemert, supra, 444 U.S. at p. 478), attorney fees awarded under this doctrine are not assessed directly against the losing party (fee shifting), but come out of the fund established by the litigation, so that the beneficiaries of the litigation, not the defendant, bear this cost (fee spreading). Under federal law, the amount of fees awarded in a common fund case may be determined under either the lodestar method or the percentage-of-the-benefit approach ( In re Washington Public Power Supply System Litigation (9th Cir. 1994) 19 F.3d 1291, 1296), although, about a decade ago, as the Ninth Circuit then noted, there commenced a "ground swell of support for mandating the percentage-of-the-fund approach in common fund cases." ( State of Fla. v. Dunne (9th Cir. 1990) 915 F.2d 542, 545.) Prior to 1977, when the California Supreme Court decided Serrano III, supra, 20 Cal.3d 25, California courts could award a percentage fee in a common fund case. (See, e.g., Melendres v. City of Los Angeles (1975) 45 Cal.App.3d 267, 284.) After Serrano III, it is not clear whether this may still be done. (See, Dunk v. Ford Motor Co. (1996) 48 Cal.App.4th 1794, 1809 ["The award of attorney fees based on a percentage of a `common fund' recovery is of questionable validity in California."].)

The trial court refused to award a percentage fee in this case, or even to incorporate a percentage-of-the-benefit approach into its lodestar calculation, apparently because it concluded that, as a matter of law, this approach could only be employed in cases in which a separate fund was created for the benefit of the class, and that no such fund resulted from the settlement of this case. Conceding, as we have said, that this is not a traditional common fund case, class counsel describe it instead as a "hybrid case." They maintain that, because the benefit to the class can easily be monetized, it should be thought of as having created a "constructive common fund." ( In re General Motors Corp. Pick-Up Truck Fuel Tank Products Liability Litigation (3d Cir. 1995) 55 F.3d 768, 820; see also, Wing v. Asarco, Inc. (9th Cir. 1997) 114 F.3d 986, 989.)

III.

Class counsel's argument rests heavily on federal cases. During the nearly quarter of a century since Serrano III, many federal courts, heavily burdened with the class and derivative actions that give rise to the need to adjudicate fee issues, became disillusioned with the lodestar method. This shift is perhaps most dramatically exemplified by the Third Circuit, whose 1973 opinion in Lindy I, supra, 487 F.2d 161, which was relied upon in Serrano III ( 20 Cal.3d at p. 49, fn. 23), pioneered adoption of the lodestar methodology. (See also, Lindy Bros. Builders of Phila. v. American Radiator Standard Sanitary Corp. (3d Cir. 1976) 540 F.2d 102 ( Lindy II).) The so-called " Lindy lodestar" technique quickly gained wide acceptance among the federal circuits, sometimes in modified fashion (see, e.g., Johnson v. Georgia Highway Exp., Inc. (5th Cir. 1974) 488 F.2d 714; Norman v. Housing Authority of City of Montgomery (11th Cir. 1988) 836 F.2d 1292), and was eventually approved by the United States Supreme Court. ( Hensley v. Eckerhart (1983) 461 U.S. 424, 433; see also Blum v. Stenson (1984) 465 U.S. 886, 897; City of Riverside v. Rivera (1986) 477 U.S. 561; Pennsylvania v. Delaware Valley Citizens' Council for Clean Air (1986) 478 U.S. 546; Pennsylvania v. Delaware Valley Citizens' Council for Clean Air (1987) 483 U.S. 711.)

However, in 1985, concerned about increasing criticism of the lodestar method, the Third Circuit reexamined the concept it is credited with inventing in Lindy I, supra, and concluded it was seriously deficient and subject to abuse when applied in cases resulting in the creation of a fund. A task force commissioned by the Third Circuit concluded that the lodestar approach (1) "increases the workload of an already overtaxed judicial system," (2) is "insufficiently objective and produce[s] results that are far from homogenous," (3) "creates a sense of mathematical precision that is unwarranted in terms of the realities of the practice of law," (4) "is subject to manipulation by judges who prefer to calibrate fees in terms of percentages of the settlement fund or the amounts recovered by the plaintiffs or of an overall dollar amount," (5) "encourages lawyers to expend excessive hours, and . . . engage in duplicative and unjustified work," (6) "creates a disincentive for the early settlement of cases," (7) deprives trial courts of "flexibility to reward or deter lawyers so that desirable social objectives, such as early settlement, will be fostered," (8) "works to the particular disadvantage of the public interest bar," and (9) results in "confusion and lack of predictability." ( Court Awarded Attorney Fees, Report of the Third Circuit Task Force (1985) 108 F.R.D. 237, 246-249 (Report of the Third Circuit Task Force). Due to these perceived deficiencies, the Report of the Third Circuit Task Force concluded that the lodestar technique is a "cumbersome, enervating, and often surrealistic process of preparing and evaluating fee petitions that now plagues the Bench and Bar" ( id. at p. 255), and recommended a return to the percentage of the recovery fee formula in cases involving a settlement fund.

The Report of the Third Circuit Task Force has been as influential as were the Third Circuit opinions it reevaluates and questions, even with respect to class action cases not involving a conventional common fund. The criticisms of the lodestar approach set forth in this Report are now echoed by many authorities, who have been most vocal about the manner in which it exacerbates the problem of "cheap settlements" and burdens already overworked trial judges. As one commentator has stated, "by severing the fee award from the settlement's size, [the lodestar] formula facilitates the ability of defendant's and the plaintiff's attorneys to arrange collusive settlements that exchange a low recovery for a high fee award. In addition, the lodestar formula essentially places the court in the position of a public utility commission that regulates the `fair' return the attorney receives by both determining the attorney's normal billing rate and assessing whether the attorney's time was reasonably expended. At a minimum, such an undertaking imposes a substantial burden on the already overloaded judicial system, thus increasing the . . . disparity between the social and private costs of litigation." (Coffee, Understanding the Plaintiff's Attorney: The Implications of Economic Theory for Private Enforcement of Law Through Class and Derivative Actions (1986) 86 Colum. L.Rev. 669, 691, fns. omitted; see also, Issacharoff, Class Action Conflicts (1997) 30 U.C. Davis L.Rev. 805; Silver, Unloading the Lodestar: Toward a New Fee Award Procedure (1992) 70 Tex. L. Rev. 865; Macey Miller, The Plaintiffs' Attorney's Role in Class Action and Derivative Litigation: Economic Analysis and Recommendations for Reform (1991) 58 U.Chi. L.Rev. 1; Leubsdorf, The Contingency Factor in Attorney Fee Awards (1981) 90 Yale L.J. 473.)

This now widely shared view has stimulated greater judicial willingness to evaluate a fee award as a percentage of the recovery. (See, e.g., In re Thirteen Appeals Arising out of San Juan Dupont Fire Litigation (1st Cir. 1995) 56 F.3d 295, 305-306; Gottlieb v. Barry (10th Cir. 1994) 43 F.3d 474, 482; Rawlings v. Prudential-Bache Properties, Inc. (6th Cir. 1993) 9 F.3d 513, 517; Matter of Continental Illinois Securities Litigation (7th Cir. 1992) 962 F.2d 566, 572; Chun v. Board of Trustees of the Employees' Retirement System of the State of Hawai'i (2000) 92 Haw. 432, 992 P.2d 127.) Although, as will be seen, it is an overstatement, one commentator has observed that "[b]asically, all courts except the Florida Supreme Court and, to some extent, the Fifth Circuit, have abandoned the failed lodestar experiment," which he describes as a "welcome development." (Issacharoff, Class Action Conflicts, supra, 30 U.C. Davis L. Rev. 805 at p. 827; see also, Grady, Reasonable Fees: A Suggested Value-Based Analysis for Judges, supra, 184 F.R.D. 131 ["lodestar method . . . has been so unsatisfactory that it is now being abandoned in most Circuits, at least in class actions, in favor of a modified percentage-of-recovery approach."].) The Ninth Circuit, responding perhaps to the growing complaint of trial judges in the circuit and elsewhere that lodestar analyses "consume an undue amount of time with little resulting advantage to anyone, but [operate] to the detriment of the class members" ( In re Activision Securities Litigation (N.D.Cal. 1989) 723 F. Supp. 1373, 1375; see also, Mashburn v. Nat'l Healthcare, Inc. (M.D.Ala. 1988) 684 F. Supp. 679, 689-692), has made the percentage-of-the-benefit approach the preferred method for determining fees in common fund cases. (See, e.g., In re Pacific Enterprises Securities Litigation (9th Cir. 1995) 47 F.3d 373, 379; Paul, Johnson, Alston Hunt v. Graulty, supra, 886 F.2d 268.) While most federal appellate courts allow trial judges a range of discretion in this area (see, e.g., Cook v. Niddert (7th Cir. 1998) 142 F.3d 1004, 1010-1011; Johnston v. Comerica Mortg. Corp. (8th Cir. 1996) 83 F.3d 241, 246; In re General Motors Corp. Pick-Up Truck Products Liability Litigation, supra, 55 F.3d 768, 819-822), in the District of Columbia Circuit and Eleventh Circuit, the percentage-of-the-benefit approach now appears compulsory in common fund cases. (See Swedish Hosp. Corp. v. Shalala (D.C. Cir. 1993) 1 F.3d 1261, 1271; Camden I Condominium Ass'n, Inc. v. Dunkle (11th Cir. 1991) 946 F.2d 768, 774.)

It is for present purposes significant, as class counsel are at great pains to emphasize, that the federal trend in favor of the percentage-of-the-benefit approach has been extended to cases in which (1) the "fund" that results from an adjudication or settlement is not deposited in a separate account; (2) the value of the "fund" depends on the number of valid claims presented or is imprecise for other reasons; and (3) attorney fees are not deducted from monies made available to the class, but are paid by the defendant directly. All that has been required in many such cases is that the benefits received by the class, or the range thereof, can be monetized without undue speculation. Johnston v. Comerica Mortg. Corp., supra, 83 F.3d 241, the case class counsel rely most heavily upon, provides a good example. Johnston involved consolidated fee applications in successful class actions alleging that lenders improperly maintained escrow accounts for taxes and insurance on residential mortgages they serviced. The plaintiffs claimed the defendants failed to properly refund or credit surplus funds and by their practice were violating federal law and the terms of the mortgage agreements. ( Id. at p. 243.) The terms of the settlement agreements provided members of the class cash "rebates" representing damages for lost interest on past retained overages totaling at least $123,000 in one action and $29,000 in the other. The settlements also provided injunctive relief changing the defendants' future mortgaging practices. In the view of class counsel, this injunctive relief "constituted the real heart of the settlements." ( Ibid.) Although the settlements did not require the establishment of a fund or funds for the benefit of the class, they did provide that each defendant would establish a fund from which attorney fees could be paid. They also contained "clear sailing" provisions, whereby the defendants agreed not to oppose requests for attorney fees not to exceed $57,000 in one case and $100,000 in the other. Class counsel justified their requests for the full amounts set aside "solely upon a `percentage of the benefit' approach." ( Id. at p. 244.)

A magistrate judge recommended that fees be denied on the grounds that class counsel "failed to produce any information which would shed light upon the reasonableness of the fee applications and . . . the total amount of benefit to the classes could not be accurately calculated and amounted to speculation." ( Ibid. at p. 244.) Class counsel objected to the recommendation and asked the district court to grant their applications, arguing that the class recoveries were monetarily substantial and that fees were justified on the basis of a percentage of those recoveries. The district court rejected the requests and thereafter also denied motions for reconsideration and for leave to submit time records. The Court of Appeals reversed, vacated the order denying fees, and remanded the cases to the district court, authorizing it to utilize either the lodestar or the percentage-of-the-benefit method.

After discussing the critique of the lodestar method in the Report of the Third Circuit Task Force, and its recommendation "that the percentage of the benefit method be employed in common fund situations" ( id. at p. 245), the Johnston court addressed the determination of the district court — identical to that of the trial court here — that "because the attorney fees were to be paid by the defendants separate and apart from the settlement funds, the fees did not come from a `common fund' belonging to the plaintiffs, and thus the percentage of the benefit approach was inappropriate." ( Id. at pp. 245-246) The Court of Appeals rejected this conclusion, stating as follows: "Although under the terms of each settlement agreement, attorney fees technically derive from the defendant rather than out of the class' recovery, in essence the entire settlement amount comes from the same source. The award to the class and the agreement on attorney fees represent a package deal. Even if the fees are paid directly to the attorneys, those fees are still best viewed as an aspect of the class' recovery. [Citation.] [¶] Accordingly, the direct payment of attorney fees by defendants should not be a barrier to the use of the percentage of the benefit analysis in the cases." ( Id. at p. 246, italics added.)

In support of this conclusion, the Johnston court relied on the earlier Third Circuit opinion In re General Motors Corp. Pick-Up Truck Fuel Tank Products Liability Litigation, supra, 55 F.3d 768. There the court explained that the lodestar method and the percentage-of-the-benefit approach each have "distinct advantages for certain kinds of actions, which will make one of the methods more appropriate as a primary basis for determining the fee." ( Id. at p. 820.) The lodestar method is most useful in statutory fee shifting cases, the court indicated. First, by decoupling the fee award from the class recovery, "the lodestar assures counsel undertaking socially beneficial litigation (as legislatively identified by the statutory fee shifting provision) an adequate fee irrespective of the monetary value of the final relief achieved for the class." ( Id. at p. 821.) Asserting that the lodestar rationale has the additional advantage of eliminating the need for subjective judgments where the value of the settlement cannot be precisely evaluated, the court noted that, on the other hand, it may provide class counsel "the incentive to delay settlement in order to run up fees while still failing to align the interests of the class and its counsel, and for not rewarding counsel incrementally for undertaking the risk of going to trial. [Citation.]" ( Ibid.) The lodestar method is less useful, the court suggested, where the monetary value of the relief is monetarily ascertainable; and in such cases there is therefore less need to suffer the disadvantages of that method of fee assessment.

As in the present case, the settlement in In re General Motors Corp. Pick-Up Truck Fuel Tank Products Litigation, supra, did not create a common fund for the benefit of the class; instead, it provided members of the class a $1,000 or $500 certificate redeemable only if they purchased specified models of General Motors pick-up trucks. The Court of Appeals nevertheless concluded that, although the ultimate choice of methodology rests with the sound discretion of the trial court, it "should probably use the percentage-of-recovery rather than the lodestar method as the primary determinant." ( 55 F.3d at p. 821.) The court observed that the percentage of recovery method is used in common fund cases to avoid unjust enrichment of the class. "Because these cases are not presumed to serve the public interest (as evidenced by the lack of a fee statute), there is no social policy reason that demands an adequate fee. Instead the court apportions the fund between the class and its counsel in a manner that rewards counsel for success and penalizes it for failure." ( Ibid.) Noting that "[c]ourts 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," and pointing to the Fifth Circuit's opinion in In re Air Crash Disaster at Florida Everglades (5th Cir. 1977) 549 F.2d 1006 as an example, the General Motors court concluded "that this case presents a situation more closely aligned with the common fund paradigm than the statutory fee paradigm. Although class counsel and GM contend (and the district court believed) that the fee was a separate agreement, thus superficially resembling the separate awards in statutory fee cases, private agreements to structure artificially separate fee and settlement arrangements cannot transform what is in economic reality a common fund situation into a statutory fee shifting case. Certainly the court may select the lodestar method in some non-statutory fee cases where it can calculate the relevant parameters (hours expended and hourly rate) more easily than it can determine a suitable percentage to award. But the court must vigilantly guard against the lodestar's potential to exacerbate the misalignment of the attorneys' and the class's interests." ( Ibid., citing Coffee, Understanding the Plaintiff's Attorney, supra, 86 Colum.L.Rev at 717.)

The General Motors court realized that the size of the class recovery was influenced by and therefore related to the size of the fee even though the fee was paid directly by the defendant and not out of the class recovery, stating: "this court has recognized that `a defendant is interested only in disposing of the total claim asserted against it; . . . the allocation between the class payment and the attorneys' fees is of little or no interest to the defense." ( In re General Motors Corp. Pick-Up Truck Fuel Tank Products Litigation, supra, at pp. 820-821, quoting Prandini v. National Tea Co. (3d Cir. 1977) 557 F.2d 1015, 1020.) The court therefore took the position that the fee the defendant separately agreed to pay directly to class counsel "is, for practical purposes, a constructive common fund." ( Id. at p. 820.) The fact that the fee could not fairly be "de-coupled" from the class recovery was among the reasons the Court of Appeals concluded that the trial court "should probably use the percentage-of-recovery rather than the lodestar as the primary determinant [of the fee award]." ( Id. at p. 821.)

Johnston and General Motors are not the only cases reflecting greater federal reliance on the percentage-of the-benefit approach; use of that approach is also illustrated by two recent opinions of the Ninth Circuit. Wing v. Asarco, Inc. , supra, 114 F.3d 986 was a class action for alleged contamination of residential soils by toxic substances. The $67.5 million settlement included cash payments to the class of $60.5 million, the amount the defendant expected to receive from its insurer, which was not guaranteed. The $60.5 million also included attorney fees and expenses as determined by the court, which the defendant agreed to pay directly to counsel rather than detract from the class's recovery. A "supplemental agreement" regarding attorney fees included a stipulation that class counsel's lodestar was equal to or less than $4 million, with expenses slightly exceeding $1.5 million, and that the defendant would not litigate these amounts. The parties also stipulated that they could litigate the propriety of using a multiplier to adjust the lodestar upward and appeal the amount of any multiplier awarded. After the settlement was approved, class counsel petitioned the trial court, alternatively, for fees and expenses equal to 25 percent of the $67.5 million ($16,875,000) or for fees equal to the lodestar times a multiplier of 3.8 plus expenses ($16,782,951.30).

After considering both a lodestar calculation plus a multiplier and a percentage-based figure, the trial court determined that $8 million was a reasonable fee and also awarded $1.6 million in expenses. ( Wing v. Asarco, Inc. , supra, 114 F.3d at p. 988.) "Cross-checking the $8 million fee against the stated value of the Settlement Agreement, the court determined that the fee award constituted only 12% of the class' recovery." ( Id. at p. 990.) On appeal, the defendant disputed class counsel's contention that the case presented a common fund situation, pointing out that the cash fund was not guaranteed. The Ninth Circuit acknowledged at the outset "that this situation is neither fish nor fowl nor fair weather game" ( id. at p. 989), i.e., it was not strictly a fee shifting case because fees were not statutorily mandated, nor was it a conventional common fund case, because fees would come from the defendant directly, not out of the class recovery. The court additionally acknowledged, as did the trial court, that the cash fund contemplated by the parties was not guaranteed, that other aspects of the settlement were "open-ended" and "difficult to value," and that, indeed, it was "impossible to determine the `actual' value of the recovery for percentage purposes." ( Id. at 990.) Nevertheless, the Court of Appeals concluded that the trial court was able to estimate the total value of the settlement and properly considered the percentage of the benefit when evaluating the propriety of the amount of the fee award. "Particularly in light of the parties' agreement as to value, we cannot say that the district court abused its discretion by using an estimated value equal to the settlement value that was communicated to the class . . . ." ( Ibid.; accord, In re Prudential Ins. Co. of America Sales Practices Litigation (3d Cir. 1998) 148 F.3d 283, 336 [though case created no cash fund and attorney fees were paid directly to counsel, "the settlement was most closely aligned to the common fund paradigm and . . . a percentage-of-recovery calculation was the appropriate measure of the attorneys' fee award."].)

More recently, in Hanlon v. Chrysler Corp., supra, 150 F.3d 1011, the Ninth Circuit again indicated that even where use of the lodestar approach is justified by the difficulty of placing a value on the settlement, the correctness of the amount awarded under the lodestar method should, where possible, be measured by inquiring whether it represents a reasonable percentage of the benefit to the class claimed by the parties: "Although no class member is entitled to a cash recovery — making valuation of the settlement agreement more difficult — Chrysler and class counsel valued the settlement at $115 million. This is the amount Chrysler charged against its earnings in order to account for its voluntary service action and the production and installation of the replacement latches. The fee award of $5.2 million represents roughly 4.5% of this `common fund,' significantly less than the 25% commonly used under Six Mexican Workers [ supra, 904 F.2d at p. 1311]." ( Id. at p. 1029; see also Paul, Johnson, Alston Hunt v. Graulty, supra, 886 F.2d 268, 271.)

IV.

The cases just described reflect the growing willingness of federal courts to disregard the strict theoretical distinction between fee shifting and fee spreading in cases in which fees are not authorized by statute, no separate fund is established and fees are paid directly by the defendant — provided, of course, that the monetary value of the class recovery is reasonably ascertainable. Respondent does not address the policies that inform this trend; instead, it simply argues that (1) the federal cases class counsel rely upon are inapposite, because the defendants in those cases did not object to the amount of fees sought; (2) in any case, the California Supreme Court has rejected pure percentage fees; and (3) that a lodestar calculation cannot be enhanced on the basis of a percentage-of-the-benefit analysis, or for any other reason, absent showings, which cannot be made in this case, that the litigation conferred a substantial benefit on the public or a large number of persons and was particularly complex or novel.

A.

B.

Respondent's contention that pure percentage fees have been rejected by the California Supreme Court, at least in cases such as this in which there is not a conventional common fund, and that the Court's view is binding, is persuasive.

The statement of some courts that "California does not follow the approach to fee awards adopted by the federal courts" ( Weeks v. Baker McKenzie (1998) 63 Cal.App.4th 1128, 1173; Flannery v. California Highway Patrol (1998) 61 Cal.App.4th 629, 644), is correct only insofar as there is California precedent regarding the question at issue. Thus in Serrano IV, where the question related to the construction of a state statute authorizing fees on the private attorney general theory, the California Supreme Court followed the lead of federal courts construing similar statutes because it believed such federal authority has "analogous precedential value" in construing the state statute. ( Id. at p. 639, fn. 29; see also Feminist Women's Health Center v. Blythe (1995) 32 Cal.App.4th 1641, 1674, fn. 8; Sokolow v. County of San Mateo (1989) 213 Cal.App.3d 231, 249.) Accordingly, when there is no relevant California precedent on point, federal precedent should be consulted. (Pearl, California Attorney Fee Awards, supra, § 13.4 and cases there cited.)

With respect to the propriety of a pure percentage fee award, Serrano III, supra, 20 Cal.3d 25 provides California precedent. That case stemmed from the class action victory in Serrano v. Priest (1976) 18 Cal.3d 728 ( Serrano I), holding, in effect, that the then-existing California public school financing system was invalid as in violation of state constitutional provisions guaranteeing equal protection of the laws, and that the system must be brought into constitutional compliance within a period of six years. Plaintiffs' counsel moved for reasonable attorney fees, resting not on statute but on the inherent equitable powers of the court. In support of their claim they relied on three theories: the common fund, substantial benefit, and private attorney general exceptions to the general rule disfavoring fees. Though the court affirmed the award of fees under the private attorney general theory (later codified in Code Civ. Proc., § 1021.5), it affirmed the denial of fees on either the common fund or substantial benefit theories. The trial court found the common fund exception inapplicable, reasoning that whatever additional monies Serrano I made available for public education resulted "from legislative implementation of the judgment, not from the judgment itself," and because the judgment "does not require any particular level of expenditure." ( Id. at p. 36.) The Supreme Court agreed in Serrano III, and went on to note that "even if it were determined that the monies to become available for education in the wake of Serrano [ I] should be considered a `fund' for these purposes, plaintiffs and their attorneys nowhere suggest that payment should be made to them out of such monies. Instead, they seem to indicate, with perhaps intentional vagueness, that their fees should be paid by `the State.'" ( Id. at p. 37, fn. omitted.)

The high court then went on to disagree with Brewer v. School Board of City of Norfolk, Virginia (4th Cir. 1972) 456 F.2d 943 (cert. den. (1972) 406 U.S. 933), the primary authority relied upon by plaintiffs' attorneys. In Brewer "the Court of Appeals ordered the award of reasonable attorneys fees against a school district after determining that its desegregation plan was inadequate insofar as it failed to provide a practical method of free transportation for students assigned to schools beyond normal walking distance from their homes. There the court, stating that this was a case for `at least a quasi-application of the "common fund" doctrine' [citation], reasoned that whereas each of the students involved had secured a right worth approximately $60 per year to each of them, it would `defeat the very purpose of the relief provided' to impose a charge against them for a proportionate share of the attorneys fees [citation]. `The only feasible solution in this particular situation,' the [ Brewer] court held, `would seem to be in requiring the school district itself to supplement its provision of free transportation with payment of an appropriate attorney's fee to plaintiffs' attorneys for securing the addition of such a provision to the plan of desegregation.' [Citation.]" ( Serrano III, supra, 20 Cal.3d at p. 37.) Our Supreme Court concluded that Brewer "represents an improper application of the `common fund' theory . . . [that] is not consistent with the law of this state. We hold that where plaintiffs' efforts have not effected the creation or preservation of an identifiable `fund' of money out of which they seek to recover their attorneys fees, the `common fund' exception is inapplicable." ( Id. at pp. 37-38.)

Respondent rests not just on the foregoing language in Serrano III but also on the statement elsewhere in that opinion that "[t]he starting point of every fee award . . . must be a calculation of the attorney's services in terms of the time he has expended on the case." ( Serrano III, supra, 20 Cal.3d at p. 48, fn. 23, italics added.) The reason the fee analysis must be "anchored" to the time spent on the case and a reasonable hourly rate, the court declared, is that "this concept is the only way of approaching the problem that can claim objectivity, a claim which is obviously vital to the prestige of the bar and the courts." ( Ibid.) This statement, which arguably renders it questionable whether a pure percentage fee can be awarded even in a conventional common fund case (see Dunk v. Ford Motor Co., supra, 48 Cal.App.4th at p. 1809), certainly precludes such an award in a case such as this. Even if the ascertainable amount of money respondent has actually paid to satisfy valid claims were deemed a "fund," class counsel has never suggested that their fee should come from this source.

Accordingly, we hold that refusal of the trial court to award class counsel a fee calculated purely as a percentage of the class recovery was not an abuse of discretion.

C.

Respondent's final argument is that the lodestar cannot be enhanced on the basis of a percentage-of-the-benefit analysis, or for any reason, because the settlement did not greatly benefit the public and the litigation was neither complex nor novel, all of which must assertedly be shown in order to justify any upward adjustment of the lodestar. We are not persuaded by this argument.

As indicated, the lodestar formula does not limit consideration to hours expended and hourly rate, though that is the foundation of the calculation. The base amount produced by multiplying hours spent on the case by a reasonable hourly rate "may then be increased or reduced by application of a `multiplier' after the trial court has considered other factors concerning the lawsuit." ( Press v. Lucky Stores, Inc. (1983) 34 Cal.3d 311, 322.) In Serrano III, the Supreme Court held that the seven factors the trial court took into consideration to determine whether to augment or diminish the lodestar amount were appropriate in light of the circumstances of the case, deferring to the discretion of the trial judge. "The `experienced trial judge is the best judge of the value of professional services rendered in his court, and while his judgment is of course subject to review, it will not be disturbed unless the appellate court is convinced that it is clearly wrong.'" ( Serrano III, supra, 20 Cal.3d at p. 49, quoting Harrison v. Bloomfield Building Industries, Inc. (6th Cir. 1970) 435 F.2d 1192, 1196 and citing Mandel v. Hodges (1975)[1976] 54 Cal.App.3d 596, 624.)

However, neither in Serrano III nor in any other opinion has our Supreme Court carved the factors used in that case into concrete or barred consideration of other relevant and non-duplicative factors; nor have the courts of appeal sought to do so. On the contrary, in Press v. Lucky Stores, Inc., supra, 34 Cal.3d 311 the Supreme Court described the factors employed in Serrano III as among those "the trial court may consider in adjusting the lodestar figure." ( Press v. Lucky Stores, Inc., supra, 34 Cal.3d at p. 322, fn. 12, italics added.) The courts of appeal have been similarly deferential to trial court discretion as to the factors that may be used to increase or decrease the base lodestar figure in a given case. As one such court recently noted, there is "no mechanical formula that dictate[s] how the [trial] court should evaluate all these factors. Instead, it ha[s] wide latitude in assessing the value of the attorney's services, and its decision [is] not to be disturbed on appeal absent a manifest abuse of discretion. [Citations.]" ( Flannery v. California Highway Patrol, supra, 61 Cal.App.4th at p. 639.)

Respondent's contention that a court may enhance the lodestar through the use of a multiplier only in cases "where the public benefit is great and the litigation complex and difficult" is based on Beasley v. Wells Fargo Bank (1991) 235 Cal.App.3d 1407 and Weeks v. Baker McKenzie, supra, 63 Cal.App.4th 1128. Those cases cannot be construed so broadly. Beasley was a successful class action challenging a bank's assessment of fees against credit card customers who failed to make timely payments or exceeded their credit limits. The judgment was for more than $5 million and the trial court awarded attorney fees of nearly $2 million based on the "private attorney general" doctrine. (Code Civ. Proc., § 1021.5) In affirming the fee award, the appellate court stated, in language seized upon by respondent, that the action "has undoubtedly been closely watched by lending institutions nationwide that do business in California, consequently providing an indirect benefit to non-Wells Fargo California customers who could easily number in the millions." ( Id. at p. 1417.) This factor was relevant in Beasley because the fee in that case, unlike the one here, was awarded under a statute that specifically required a "substantial benefit" to have been "conferred on the general public or a large class of persons. . . ." (Code Civ. Proc., § 1021.5) Moreover, and more importantly, the Beasley court also justified enhancement of the lodestar because it agreed with the trial judge "`that this kind of consumer class action litigation would not be pursued by counsel but for the expectation of receiving enhanced fee awards in successful cases.'" ( Id. at p. 1419.) Beasley certainly does not stand for the proposition that a lodestar can be enhanced in a case such as the one before us only if it benefits a significant number of persons beyond the class.

Nor does Weeks v. Baker McKenzie, supra, 63 Cal.App.4th 1128 stand for any such principle. Weeks was a sexual harassment action brought by a legal secretary against a partner in a law firm and the firm itself under the Fair Employment and Housing Act (FEHA). (Gov. Code, § 12900 et seq.) The jury awarded the plaintiff $50,000 in compensatory damages, and punitive damages of $225,000 against the partner and $6.9 million against the law firm, though the trial court reduced the latter figure to $3.5 million. The trial court also awarded the plaintiff approximately $1.8 million in attorney fees, a figure it reached by enhancing the base lodestar amount of $921,565 by a multiplier of 1.7. The Court of Appeal affirmed the damage awards but found the trial court had failed to articulate a "sufficient public or private reason" to justify the use of a 1.7 multiplier, and remanded the matter for reconsideration of the fee. ( Id. at p. 1176.) Respondent isolates the statements in Weeks that "[t]he purpose of a fee enhancement is not to reward attorneys for litigating certain kinds of cases, but to fix a reasonable fee in a particular action" ( Weeks v. Baker McKenzie, supra, 63 Cal.App.4th pp. 1171-1172), and that "[w]hen the public value of the case is great and the risk of loss results from the complexity of the litigation or the uncertainty of the state of the law, fee enhancement may be proper." ( Id. at p. 1175) Respondent takes these statements out of context. First, as noted in Weeks, fees in that case were authorized under an attorney fee statute that made no mention of enhancements. (Gov. Code, § 12965, subd. (b).) "Nonetheless," the Weeks court went on to observe, "it is recognized that some form of fee enhancement may be appropriate and necessary to attract competent representation of cases meriting legal assistance." ( Id. at p. 1172.) The reason enhancement of the lodestar amount was questioned in Weeks, which was not a class action , was the absence of any reason to think the incentive of an enhanced fee was necessary to induce lawyers to represent persons whose injuries individually entitled them to substantial damages, as legal representation can invariably be obtained by such persons. The court pointed out that the availability of statutory fees attracts counsel as does the fact that a fee statute "does not limit fees to a percentage of the plaintiff's recovery, . . ." ( Id. at p. 1175.) Weeks is in this critical respect very different from the present case, in which fees are not authorized by statute and counsel are unable to negotiate a percentage fee with a client.

Respondent also overlooks the acknowledgment by the Weeks court that California trial courts have considerably wider latitude than their federal counterparts in the selection of factors that may be used to adjust the lodestar and that, therefore, "an upward or downward adjustment from the lodestar figure will be far more common under California law than under federal law." ( Id. at p. 1173.) For example, certain factors California courts can use to enhance a lodestar award, such as the novelty and complexity of the issues, the special skill and experience of counsel, and the contingent nature of the recovery cannot be used to increase an award in federal courts because they are presumed to be already incorporated into the lodestar figure. ( City of Burlington v. Dague (1992) 505 U.S. 557, 562-563; Pearl, California Attorney Fee Awards, supra, § 11.4.) Flannery v. California Highway Patrol, supra, 61 Cal.App.4th 629 provides a useful example of California's relatively permissive attitude on the use of multipliers. That case was an action by a former officer against the California Highway Patrol alleging harassment and wrongful termination in violation of the FEHA. After judgment was entered in favor of the plaintiff, the trial court awarded him attorney fees and expenses. The Court of Appeal reversed the fee award and remanded the matter for reconsideration, because some of the factors the trial court relied upon to calculate the reasonable hourly rate component of the lodestar were duplicative of other factors it relied upon to justify enhancing the lodestar. Significantly, however, the appellate court rejected the defendant's main contention that no multiplier should be used to increase the lodestar because the United States Supreme Court rejected the use of multipliers to enhance fees under federal fee-shifting statutes analogous to the FEHA.

Acknowledging that the United States Supreme Court limited the factors that could be considered in determining whether to adjust the lodestar amount ( Blum v. Stenson, supra, 465 U.S. 886, 889-900; accord, Pennsylvania v. Delaware Valley Citizens' Council for Clean Air, supra, 478 U.S. 546, 565-566), and specifically barred the use of a contingent risk enhancement ( City of Burlington v. Dague, supra, 505 U.S. 557), the Flannery court concluded that California courts are not so bound. The federal cases, the court reasoned, "are based in substantial part on federal legislative history, for which there is no California parallel. Defendant has not demonstrated that the California Legislature intended or intends federal standards to apply to limit the trial court's exercise of discretion in calculating the amount of reasonable attorney fees under California fee-shifting statutes generally or under the FEHA provision in particular. . . . [¶] In sum, given the state of California law, we cannot say that it is an abuse of discretion for a trial court awarding fees under the FEHA ever to apply a multiplier in determining reasonable attorney fees." ( Flannery, supra, 61 Cal.App.4th at p. 646.)

In support of its contention that Serrano III bars California trial courts from using the percentage-of-the-benefit approach to adjust the lodestar, respondent relies primarily on two cases: Jutkowitz v. Bourns, Inc. (1981) 118 Cal.App.3d 102 and Dunk v. Ford Motor Co., supra, 48 Cal.App.4th 1794.

In Jutkowitz, as part of approval of a negotiated settlement of a minority shareholder class action, the trial court awarded attorney fees to the plaintiff under a lodestar formula. The lodestar was augmented by 50 percent as a result of the complexity of the litigation, the results achieved and the contingency of the award. The corporate defendant did not oppose allowance of that fee, which was in the amount of $90,000. The trial court refused, however, to increase this amount on the basis of benefits assertedly obtained for other minority shareholders, who, as a result of the litigation, obtained more favorable settlements of their claims. Rejecting plaintiff's argument that the attorney fees were inadequate, the Court of Appeal characterized his argument as "an attempt to engraft a `contingent fee' concept on to the equitable common fund doctrine. [¶] In our opinion, the clear thrust of the holding in Serrano, supra, and the cases upon which that holding relied, is a rejection of any `contingent fee' principle in cases involving equitable compensation for lawyers in class actions or other types of representative suits." ( Jutkowitz, supra, at p. 110.) The main point of the Jutkowitz opinion, and the proposition for which it has been cited by other courts of appeal, is that the lodestar methodology must be strictly adhered to because "favorable public perception and the prestige of the legal profession and our system of justice, requires a formula for computation which can be objectively measured." ( Id. at p. 111, quoted with approval in People ex rel. Dept. of Transportation v. Yuki (1995) 31 Cal.App.4th 1754, 1769 and Salton Bay Marina, Inc. v. Imperial Irrigation Dist. (1985) 172 Cal.App.3d 914, 954, both of which are statutory fee cases and not class actions.)

In Dunk, supra, 48 Cal.App.4th 1794, the settlement of a class action against an automobile manufacturer relating to defects in the doors of certain cars provided that each member of the class would receive a coupon redeemable for $400 off the price of any new Ford car or light truck purchased within a year. The defendant also agreed to pay attorney fees and costs not to exceed $1.5 million. The trial court approved the settlement and awarded counsel attorney fees of $985,000. The judgment was affirmed except as to attorney fees, which were challenged by an objecting member of the class. Counsel argued that "the nearly $1 million attorney fees were properly determined to be a reasonable percentage of the common fund, as they were only a tiny percentage of the potential settlement value of over $26 million." ( Id. at p. 1809.) The Court of Appeal found two flaws in this argument: "(1) The award of attorney fees based on a percentage of a `common fund' recovery is of questionable validity in California; and (2) even if it is valid, the true value of the fund must be easily calculated." ( Ibid.)

The trial court here appears to have agreed with respondent that Jutkowitz and Dunk, which relied upon Serrano III, barred any adjustment of the lodestar by evaluating the fee award as a percentage of the class recovery. We do not share this view of the cases. The fees disapproved in Jutkowitz and Dunk could not be squared with Serrano III because they were anchored in a percentage of the recovery rather than a lodestar. The gravamen of Jutkowitz is that "the correct amount of compensation cannot be arrived at objectively by simply taking a percentage of that fund." ( Jutkowitz, supra, 118 Cal.App.3d at p. 111, italics added.) The central points of Dunk are that fee awards based on a percentage of the recovery are questionable after Serrano III (an issue the court did not dispositively resolve), but even if they are valid they are appropriate only where the class benefit consists of a "`certain or easily calculable sum of money'" ( Dunk, supra, 48 Cal.App.4th at p. 1809, quoting Serrano III at p. 35), and the recovery in Dunk could not be so monetized. Neither the Jutkowitz nor the Dunk court felt called upon to address the question whether an award anchored in a lodestar calculation could be adjusted to reflect the amount of a monetizable recovery. Nor did the Supreme Court directly address this question in Serrano III. All it said was that "the starting point" [of every equitable fee award] "must be a calculation of the attorney's services in terms of the time he has expended on the case." ( 20 Cal.3d at p. 48, fn. 23, italics added.) As earlier explained, the Court explicitly acknowledged that the lodestar could be adjusted by the consideration of relevant factors justifying an increase or decrease, but it did not mandate the use of any particular factor or prohibit the use of others.

Moreover, intermediate appellate courts in this state have, in effect, adopted the common federal practice of "cross-checking" the lodestar against the value of the class recovery (which is not duplicative because the amount or value of the recovery is not reflected in the basic lodestar), because the award is still "anchored" in the time spent by counsel on the case, and the practice is therefore consistent with the mandate of Serrano III. Thus, California courts often use "the amount at stake, and the result obtained by counsel" as relevant factors justifying enhancement of a lodestar fee through use of a multiplier (see, e.g., City of Oakland v. Oakland Raiders (1988) 203 Cal.App.3d 78, 83), as do their federal counterparts. ( Hensley v. Eckerhart, supra, 461 U.S. 424, 434 [the "product of reasonable hours times a reasonable rate does not end the inquiry. There remain other considerations that may lead the district court to adjust the fee upward or downward, including the important factor of the `results obtained.'"]; Hanlon v. Chrysler, supra, 150 F.3d 1011, 1029 [the lodestar "may be adjusted upward or downward to account for several factors, including . . . the benefit obtained for the class"]; Johnson v. Georgia Highway Exp., Inc., supra, 488 F.2d 714, 718.)

An adjustment reflecting the amount of the class recovery is not significantly different from an adjustment reflecting a percentage of that amount; and California courts have evaluated a lodestar as a percentage of the benefit. Glendora Community Redevelopment Agency v. Demeter (1984) 155 Cal.App.3d 465, is a useful illustration. After the jury found that the property that was the subject of this eminent domain proceeding was worth $3.9 million, the trial court awarded the defending property owners attorney fees in the amount of $656,028.50, on the basis of a fee-shifting statute. (Code Civ. Proc., § 1268.610) The amount awarded was exactly that called for by the contingent fee agreement between the defendants and their counsel, which was 25 percent of all amounts recovered in excess of the plaintiff's initial offer of $1.3 million. The Court of Appeal affirmed, concluding that a fee award under the statute may be made on the basis of a contingent fee agreement where the trial court considered other factors in determining that the fee was reasonable. ( Id. at pp. 479-480.) The trial court's finding that the 25 percent figure was reasonable relied in part on the California Rules of Professional Conduct. At that time, rule 2-107 set forth nine factors that could be considered in determining the reasonableness of a fee, one of which was "[t]he amount involved and the result obtained[.]" ( Id. at pp. 474-475.)

With respect to this factor, the trial court noted that the defendant property owners "sought to prevent appellant from taking the property at considerably less than its value and by the attorney agreement sought to convey a `proportionate share of the value of the property which exceeded the offer in an effort to protect the whole.' The [trial] court concluded that: `[t]he agreement geared the amount involved to the result obtained.'" ( Glendora Community Redevelopment Agency v. Demeter, supra, 155 Cal.App.3d at p. 476.) The Court of Appeal found this reasoning proper, and agreed that "[t]wenty-five percent of the value of the property, which exceeded $1.3 million, is not unreasonable under all the circumstances." ( Id. at p. 479.)

The present case differs from Glendora Community Redevelopment Agency because this is a class action, and there was therefore no negotiated fee agreement. As many courts have noted, however, the amount of attorney fees typically negotiated in comparable litigation should be considered in the assessment of a reasonable fee in representative actions in which a fee agreement is impossible. Given the unique reliance of our legal system on private litigants to enforce substantive provisions of law through class and derivative actions, attorneys providing the essential enforcement services must be provided incentives roughly comparable to those negotiated in the private bargaining that takes place in the legal marketplace, as it will otherwise be economic for defendants to increase injurious behavior. (See Deposit Guaranty Nat. Bank, Jackson, Miss. v. Roper (1980) 445 U.S. 326, 338 , rehg. denied, 446 U.S. 947; Beasley v. Wells Fargo Bank, supra, 235 Cal.App.3d 1407, 1419; Menell, A Note on Private Versus Social Incentives to Sue in a Costly Legal System (1983) 12 J. Legal Studies 41 (Menell).) It has therefore been urged (most persistently by Judge Richard Posner) that in defining a "reasonable fee" in such representative actions the law should "mimic the market." (See, e.g., Gaskill v. Gordon (7th Cir. 1998) 160 F.3d 361, 363 ["When a fee is set by a court rather than by contract, the object is to set it at a level that will approximate what the market would set. . . . The judge, in other words, is trying to mimic the market in legal services."]; Cooper v. Casey (7th Cir. 1996) 97 F.3d 914, 920 ["the reasonable fee is capped at the prevailing market rate for lawyers engaged in the type of litigation in which the fee is being sought."]; Matter of Continental Illinois Securities Litigation (7th Cir. 1993) 985 F.2d 867, 868; Henry v. Webermaier (7th Cir. 1984) 738 F.2d 188, 193.)

"In the class action context, that would mean attempting to award the fee that informed private bargaining, if it were truly possible, might have reached. The simplest way for the law to duplicate the bargain that informed parties would reach if agency costs were low is to look to fee award levels in actions brought by sophisticated private parties under the same or comparable statutes. . . . To be sure, a variety of factors distinguish class actions from private suits. For example, in class actions, the plaintiff's attorney typically assumes more risk because the fee is contingent, but may also have lower search costs (if the attorney is piggybacking on a prior governmental action). Still, if courts were to ask what fee structure an informed, sophisticated client would use to compensate his attorney when close monitoring is not feasible, they would at least have focused on the correct question. Moreover, . . . the most logical answer to this problem of premature settlement would be to base fees on a graduated, increasing percentage of the recovery formula — one that operates, much like the Internal Revenue Code, to award the plaintiff's attorney a marginally greater percentage of each defined increment of the recovery. While this approach cannot be said to eliminate the inevitable tension between the interest of plaintiffs' attorneys and their clients in class actions, it can at least partially counteract the tendency for premature settlements." (Coffee, Understanding the Plaintiff's Attorney, supra, 86 Colum.L.Rev. 669 at pp. 696-697, fns. omitted.) Courts agree that, because the percentage-of-the-benefit approach "is result-oriented rather than process oriented, it better approximates the workings of the marketplace" than the lodestar approach ( In re Thirteen Appeals Arising out of San Juan Dupont Plaza Hotel Fire Litigation, supra, 56 F.3d 295, 307); though they do not necessarily agree the percentage should increase with the size of the recovery. (See, e.g., In re Prudential Ins. Co. of America Sales Practices Litigation (3d Cir. 1998) 148 F.3d 283, 339; Shaw v. Toshiba America Information Systems, Inc., supra, 91 F. Supp.2d 942, 972; Report of the Third Circuit Task Force, supra, 108 F.R.D. at p. 256 ["absent unusual circumstances, the percentage will decrease as the size of the fund increases."].)

It is in large part because it provides a credible measure of the market value of the legal services provided that some federal courts use a percentage-of-the-benefit analysis to "cross-check" the propriety of a lodestar fee award. ( In re General Motors Corp. Pick-Up Truck Fuel Products Liability Litigation, supra, 55 F.3d at p. 820 ["it is sensible for a court to use a second method of fee approval to cross check its conclusion under the first method"]; Wing v. Asarco, Inc., supra, 114 F.3d at p. 990; In re Chrysler Motors Corp. Overnight Evaluation Program Litigation (E.D.Mo. 1990) 736 F. Supp. 1007, 1015; accord, In re Activision Securities Litigation, supra, 723 F. Supp. 1373, 1375; Report of the Third Circuit Task Force, supra, 108 F.R.D. at p. 247; 3 Newberg on Class Actions (3d ed.) § 14.03, pp. 14-4-14-5, fns. omitted.) As we have said, the California Supreme Court has never prohibited adjustment of the lodestar on this basis. Due to the complexity of the judicial responsibility to structure a class action fee award providing counsel the optimum mix of incentives and disincentives, and the variety of different and sometimes competing considerations that may come into play (see Hay, Asymmetric Rewards: Why Class Actions (May) Settle for Too Little (1997) 48 Hastings L.J. 479), trial judges need the flexibility Serrano III provides, as it enables them to relate fee awards to the economic realities that determine the efficacy of the private enforcement contemplated by our civil justice system.

Accordingly, we hold that, in cases in which the value of the class recovery can be monetized with a reasonable degree of certainty and it is not otherwise inappropriate, a trial court has discretion to adjust the basic lodestar through the application of a positive or negative multiplier where necessary to insure that the fee awarded is within the range of fees freely negotiated in the legal marketplace in comparable litigation. This approach is in our view not only consistent with the mandate of our Supreme Court in Serrano III, but ameliorates those aspects the lodestar approach that have come under criticism while avoiding the pitfalls of pure percentage fees. As the Supreme Court of Hawaii recently observed, "[t]he percentage method is not without flaws of its own," and "the lodestar approach entails sufficient benefits that it should not be altogether abandoned at the present time." ( Chun v. Board of Trustees of the Employees' Retirement System of the State of Hawai'i, supra, 92 Haw. 432, 443, 444, accord, Goldberger v. Integrated Resources, Inc., supra, 43, 50 ["we question the wisdom of abandoning the lodestar entirely.".)

V.

The record reveals nothing about this case which would make it manifestly inappropriate to evaluate the lodestar as a percentage of the recovery and adjust it accordingly if it can be determined that the lodestar is significantly different from the range of percentage fees freely negotiated in comparable litigation.

Though the settlement did not create a common fund out of which fees are to be paid, the monetary value of the benefit to the class is much less speculative than that of some traditional common funds. (See, e.g., discussion of Shaw v. Toshiba America Information Systems, Inc., supra, 91 F. Supp.2d 942, ante, at p. 12, fn. 6.) Respondent does not dispute that the settlement exposes it to claims in the collective amount of $14,748,000, and class counsel were able to provide the court the exact number of valid claims submitted at the time of the renewed fee application, and the collective value of those claims ($7.35 million), which is also undisputed. As has been noted, basing the fee "on the number of claims actually remediated" serves the important goal of encouraging class counsel's continued participation after settlement has been reached ( In re Prudential Ins. Co. of America Sales Practices Litigation, supra, 148 F.3d 283 at p. 334, fn. 110), and the record shows that plaintiffs' counsel advised members of the class of their rights under the settlement and urged the submission of claims.

The benefit to the class in this case is similar in most important respects to a "reversionary" common fund, in which unclaimed amounts revert to the defendant. The amount of claims actually paid, and the basis for the percentage analysis class counsel urge, is approximately $7.4 million less than the amount respondent agreed to pay members of the class. Thus, if the lodestar amount were reasonably adjusted upward to reflect an appropriate percentage of the claims paid, the total amount of respondent's out-of-pocket expense arising from this litigation would still be considerably less than the amount to which it was exposed under the settlement agreement. Moreover, the fee awarded in this manner would be less than that which could be awarded as a pure percentage fee in a traditional common fund case, which may be calculated on the basis of the total fund made available rather than the actual payments made to the class. ( Boeing Co. v. Van Gemert, supra, 444 U.S. 472; Waters v. Intern. Precious Metals Corp. (11th Cir. 1999) 190 F.3d 1291; Williams v. MGM-Pathe Communications Co. , supra, 129 F.3d 1026 [attorney fee award should have been based on a total recovery fund of $4.5 million, even though actual payout only totaled about $10,000].)

Adjusting the lodestar to reflect the monetary value of the benefit received by the class is also justified by the fact that the class was notified its counsel would seek a fee representing approximately 24 percent of the recovery; no member of the class objected and only two members opted out.

It is also noteworthy that, as counsel for respondent pointed out at the hearing on the motion for fees, appellants "could have had a deal at $5 million of the pot and they could have made their [attorney fee] claim against it, or $6 million," in which case there "would have been a common fund." In that event, respondent's counsel implied, a percentage fee (which would have been well in excess of $1 million if a 25% benchmark were adopted) would have been available; but that was not the deal struck, he emphasized, and the claims made settlement precluded a fee based on a percentage of the benefit. In a declaration submitted in support of the fee application, lead counsel for the class stated that he knew he could have insisted on a settlement establishing a common fund, and that he actually "rejected an offer in which class members would have received 75 percent of the prepayment penalty if plaintiffs agreed to cap their attorneys' fees at $2,500,000.00" explaining that "I rejected any offer which included any specific amount of attorneys' fees to avoid any conflict of interest with the class." (See Folsom v. Butte County Ass'n. of Governments (1982) 32 Cal.3d 668, 681; Prandini v. National Tea Co., supra, 557 F.2d 1015.) As he also explained in the declaration, a claims made settlement "allows me to contact class members and urge them to file a claim form without any conflict based upon the fact that the more claim forms which are filed the lower will be the payment made to each class member filing these claims. I realize that this approach makes it more difficult to establish the use of [the] common fund approach when determining attorneys' fees, but it would have been contrary to the interest of the class and, therefore, a conflict of interest for me to object to a claims made settlement which provided a guaranteed 77 percent to those class members who file valid claims."

The $6 million common fund respondent's counsel adverted to was much less than the $7.35 million which, at the time of the hearing below, had been received by the class under the claims made settlement the parties actually entered into, although a $6 million common fund would arguably have justified a fee award more than three times greater than the base lodestar amount. A fee award that fails to take into account such commendable conduct by counsel, which subordinates his economic interest to that of the class, encourages attorneys to accept settlement proposals more favorable to them than to the class.

Although the trial court apparently refused to evaluate the lodestar as a percentage of the benefit primarily because it felt legally prevented from doing so, it indicated that this was also justified by the fact that the case settled relatively quickly. But the prompt settlement, which accounted for the relatively small amount of time spent on the case, was the major reason the basic lodestar was so low relative to the monetary value of the recovery. Moreover, the promptness of settlement cannot be used to justify the refusal to apply a multiplier to reflect the size of the class recovery without exacerbating the disincentive to settle promptly inherent in the lodestar methodology. Considering that our Supreme Court has placed an extraordinarily high value on settlement (see, e.g., Neary v. Regents of University of California (1992) 3 Cal.4th 273, 277-280, citing, inter alia , McClure v. McClure (1893) 100 Cal. 339, 343), it would seem counsel should be rewarded, not punished, for helping to achieve that goal, as in federal courts. ( Merola v. Atlantic Richfield Co. (3d Cir. 1975) 515 F.2d 165, 168 [lodestar approach "permits the court to recognize and reward achievements of a particularly resourceful attorney who secures a substantial benefit for his clients with a minimum of time invested"]; Bowling v. Pfizer, Inc., supra, 922 F. Supp. 1261, 1282-1283 [case settled "in swift and efficient fashion"]; Arenson v. Board of Trade of City of Chicago (N.D.Ill. 1974) 372 F. Supp. 1349, 1358-1359 [awarding a fee four times the normal hourly rate on ground that, if the case had not settled and gone to verdict, "there is no doubt that the number of hours of lawyer's time expended would be more than quadruple the number of hours expended to date"].) While, as our Supreme Court recognized in Serrano III, supra ( 20 Cal.3d at p. 48, fn. 23), class action fee awards that are unjustifiably large create problems for the bench and bar, awards that are too small can also be problematic, as they chill the private enforcement essential to the vindication of many legal rights and obstruct the representative actions that often relieve the courts of the need to separately adjudicate numerous claims.

Finally, as earlier noted, permitting the amount of the recovery to influence the fee is most justified where the amount of the recovery is not due primarily to the size of the class. This case is not one in which the individual recoveries of class members were de minimis. As a result of the settlement, named plaintiffs recovered over $6,000; and the average recovery was over $2,000. Thus, whether the benefit to the class resulting from the settlement be considered the $14,780,000 that would be paid if all members presented valid claims, or the $7,350,000 of valid claims processed at the time of the renewed fee application, the amount would in either case be determined in significant measure by the size of the individual recoveries, not just by the size of the class. The large recovery in this case is thus a more authentic indication of the value of counsel's contribution than might otherwise be true.

What constitutes a reasonable fee in a representative action has been shown to be a far more complex question than the judiciary once thought it to be. There are no easy answers. The lodestar methodology originated as an alternative to percentage recoveries, which often resulted in exorbitant fee awards clearly unjustified by the contributions of counsel, which in turn undermined public confidence in the bench and bar. (See Report of the Third Circuit Task Force, supra, 108 F.R.D. 237, 242; see also, Serrano III, supra, 20 Cal.3d at p. 48, fn. 23.) Considering the fee only as a percentage of the benefit would simply resurrect that problem. Refusing ever to take that consideration into account, however, would, as we have indicated, create equally pernicious problems. The federal judicial experience teaches that the "reasonableness" of a fee in a representative action will often require some consideration of the amount to be awarded as a percentage of the class recovery. How much weight that factor should receive may well be, as an experienced trial judge has said, "the most difficult question in present-day jurisprudence concerning attorney's fees." (Grady, Reasonable Fees: A Suggested Value-Based Analysis for Judges, supra, 184 F.R.D. at p. 141.) However difficult this question, it cannot be avoided; and the ability of California courts to intelligently address it would not be enhanced by diminishing the tools with which they have to work.

VI.

We concur:

HAERLE, J.

LAMBDEN, J.