OPINION

REARDON, J. —

I. INTRODUCTION

This case is a companion to Beasley v. Wells Fargo Bank, A048490, ante, page 1383 [ 1 Cal.Rptr.2d 446], in which we affirm a $5,227,617 judgment in a class action which challenged Wells Fargo Bank's assessment of fees against credit card customers who failed to make timely payments or exceeded their credit limits. In the present action we affirm a subsequent judgment requiring the bank to pay the plaintiffs their attorney fees, costs and expenses in the total sum of $1,958,509, based on California's "private attorney general" statute. (Code Civ. Proc., § 1021.5)

We hold as follows: (1) the award was not precluded by the fact the litigation resulted in a common fund recovery from which attorney fees could have been paid; (2) this consumer protection action was in the public interest for purposes of a private attorney general award; (3) the trial judge did not abuse his discretion in applying a lodestar multiplier to the fee award; and (4) expert witness fees and other nonrecoverable expenses may be awarded under the private attorney general statute.

II. BACKGROUND

After judgment on the merits, the plaintiffs moved for an award of $1,133,570 in "lodestar" attorney fees with a 2.0 multiplier, $11,157 in recoverable costs (Code Civ. Proc., § 1033.5), $198,679 in expenses (primarily expert witness fees) that were nonrecoverable as ordinary costs ( id., § 1033.5, subd. (b)), and $51,433 in additional attorney fees for preparing the fee request. The judge granted the motion, citing as alternative grounds for recovery the private attorney general statute ( id., § 1021.5) and the reciprocal contractual attorney fees statute (Civ. Code, § 1717). The judge's order on the motion deviated from the plaintiffs' request in three respects: it applied a multiplier of 1.5 rather than 2.0 to the lodestar attorney fees (for an increased attorney fee award of $1,700,355) due to the plaintiffs' partial lack of success on the merits, it awarded $196,997 rather than $198,679 in nonrecoverable expenses, and it awarded $50,000 rather than $51,433 for preparation of the fee request. The court rendered a judgment incorporating the order, and Wells Fargo filed a timely notice of appeal.

III. DISCUSSION

A. Code of Civil Procedure Section 1021.5

Wells Fargo first challenges the award of attorney fees to the extent it was based on Code of Civil Procedure section 1021.5 The private attorney general statute permits a fee award when the following criteria are met: (1) the action "has resulted in the enforcement of an important right affecting the public interest," (2) "a significant benefit" has been "conferred on the general public or a large class of persons," (3) "the necessity and financial burden of private enforcement are such as to make the award appropriate," and (4) the fees "should not in the interest of justice be paid out of the recovery, if any." (§ 1021.5; see Los Angeles Police Protective League v. City of Los Angeles (1986) 188 Cal.App.3d 1, 6 [ 232 Cal.Rptr. 697].) Wells Fargo contends the "financial burden," "interest of justice" and "public interest" criteria are absent here.

1. The financial burden and interest of justice criteria.

(1a), (2) (See fn. 2.) Wells Fargo argues that in class action litigation that yields a common fund recovery, attorney fees should not be awarded under section 1021.5 because they may be paid from the common fund, so that there is no financial burden from private enforcement beyond the usual American practice of requiring litigants to bear their own attorney fees, and hence the interests of justice do not preclude payment from the fund.

(1b) This is essentially an issue of first impression. There is only one published opinion on point, Bank of America v. Cory (1985) 164 Cal.App.3d 66 [ 210 Cal.Rptr. 351], upon which Wells Fargo places great reliance. The court in that case, however, dealt with the issue only summarily. The trial judge had awarded fees under both the common fund theory and section 1021.5 The appellate court modified the judgment to deduct the portion of the award based on section 1021.5, because "[t]hose benefiting from the recovery of the fund . . . must bear their share of the cost of litigation." (164 Cal.App.3d at pp. 90-91.) The court provided no meaningful analysis of the underlying question, whether the provisions of section 1021.5 authorize a fee award despite the existence of a common fund. That task is left to us.

(3) It is generally said that the financial burden criterion of section 1021.5 is satisfied when the cost of litigation is disproportionate to the plaintiff's "individual" stake in the matter. (E.g., Baggett v. Gates (1982) 32 Cal.3d 128, 142 [ 185 Cal.Rptr. 232, 649 P.2d 874]; Woodland Hills Residents Assn., Inc. v. City Council (1979) 23 Cal.3d 917, 941 [ 154 Cal.Rptr. 503, 593 P.2d 200]; Friends of "B" Street v. City of Hayward (1980) 106 Cal.App.3d 988, 994 [ 165 Cal.Rptr. 514].) Wells Fargo argues that in common fund class actions the cost of litigation should be compared to the entire common fund that is recovered, not merely to each plaintiff's individual stake, because the common fund doctrine makes the entire fund available to pay attorney fees.

The problem with this argument is that although it is persuasive to the extent we should not limit our inquiry to each plaintiff's individual stake, it is based on a fundamental misconception that the focal point is the common fund that is actually recovered. As amici curiae point out, in comparing the cost of litigation to the plaintiffs' stake in the matter, we do not look at the plaintiffs' actual recovery after trial, but instead we consider "the estimated value of the case at the time the vital litigation decisions were being made. . . ." ( Los Angeles Police Protective League v. City of Los Angeles, supra, 188 Cal.App.3d at pp. 9-10, italics added.) In other words, the inquiry looks forward from the outset of counsel's vital litigation decisions, rather than backward after judgment. This is because the purpose of section 1021.5 is to encourage public interest litigation by offering the "bounty" of a court-ordered fee (188 Cal.App.3d at p. 10), and the focus of that incentive is on the point in time when vital litigation decisions are being considered.

(4) The opinion in Los Angeles Police Protective League prescribes a two-step process for determining the "estimated value" of the case. First, the court must determine the monetary value of the "gains actually attained" by the successful litigants. Second, the court "must discount these total benefits by some estimate of the probability of success at the time the vital litigation decisions were made. . . ." (188 Cal.App.3d at p. 9.) This "discount" is essential because of the purpose of encouraging public interest litigation as of the time vital litigation decisions are being considered. Thus, for example, if the plaintiffs had only a one-third probability of ultimate victory, the estimated value of the case was only one-third the actual recovery. ( Ibid.)

Once estimated value is determined, it is compared to actual litigation costs. In general, a fee award will be appropriate unless estimated value "exceeds by a substantial margin the actual litigation costs." (188 Cal.App.3d at p. 10.) In other words, unless the estimated value of the case was sufficient to absorb actual litigation costs and still provide an incentive to litigate, fees may be awarded under section 1021.5

Also, the financial burden criterion is interrelated with the other pertinent criteria under section 1021.5 If public benefits are very significant, it is more important to encourage litigation, and thus it may be appropriate to award fees under section 1021.5 "even in situations where the litigant's own expected benefits exceed its actual costs by a substantial margin." (188 Cal.App.3d at p. 10.) In contrast, if public benefits are modest, "the courts should award fees only where the litigant's own expected benefits do not exceed its costs by very much (or possibly are even less than the costs of the litigation)." ( Ibid.)

Thus, the following rule may be extrapolated from the Los Angeles Police Protective League opinion: If the estimated value of a class action common fund recovery, determined as of the time the vital litigation decisions were being made, does not exceed actual litigation costs by a substantial margin, the financial burden of private enforcement is such as to make it appropriate to award attorney fees under section 1021.5 In contrast, if estimated value is substantially more than actual litigation costs, there should be no award under section 1021.5 unless public benefits are very significant.

(1c) The record in the present case does not indicate whether the trial judge followed the procedures set forth in Los Angeles Police Protective League for assessing the financial burden criterion. That omission, however, is inconsequential. (5) When public interest litigation has resulted in a decision by an appellate court, and there is a subsequent appeal from a fee award, the appellate court is in just as good a position as the trial judge to determine the issue of estimated value, and may do so independently. (188 Cal.App.3d at p. 11.) The situation is no different where, as here, the appellate court considers the underlying litigation simultaneously with the fee proceedings.

(1d) We must first determine the estimated value of this case, i.e., actual recovery discounted by probability of success at the time of the vital litigation decisions. We view the amount of the actual recovery as being the full $5,227,617 in the common fund — not merely each plaintiff's individual stake — since the entire fund is available to pay attorney fees under the common fund doctrine.

Wells Fargo claims that any calculation of estimated value should be made by reference to the full amount sought by the plaintiffs at the outset of the litigation (purportedly more than $23 million) rather than the actual recovery. The opinion in Los Angeles Police Protective League leaves some room for this sort of argument. Although the court said "gains actually attained" must be discounted by probability of success, the court also said "the litigants must discount any monetary benefits they hope to achieve by the probability of success." (188 Cal.App.3d at p. 9, italics added.) This point would normally be important when there is substantial disparity between actual recovery and the amount the plaintiffs had "hoped" to recover. In the present case, however, the point is inconsequential. The amount initially "sought" from Wells Fargo could hardly be the touchstone, as it is not necessarily the same as what the plaintiffs reasonably hoped to achieve. Indeed, few plaintiffs ever expect to recover the amount they first demand. Several expert witness declarations stated that the jury verdict for $5,227,617 was an "excellent result." This being the case, that amount would logically be within the range of what the plaintiffs had realistically hoped to achieve. In any event, under the present circumstances the only fair approach is to calculate estimated value by reference to the lesser actual recovery, since the other side of the equation — actual litigation costs — was similarly reduced by the court's use of a lesser multiplier due to the plaintiffs' partial lack of success.

Quantifying the probability of success is more difficult. However, in view of the complexity of the case (which in large measure turned on arcane accounting disputes), and the limited litigation resources available to the plaintiffs as compared with the virtually unlimited resources available to Wells Fargo, it is safe to say that the plaintiffs' likelihood of success was no greater than 50 percent, and probably much less.

Thus, the estimated value of the case was no greater, and probably much less, than half the amount of the recovery, or $2,613,808. It remains for us to compare this sum with actual litigation costs. We view those costs as being the full amount awarded, less the 1.5 lodestar multiplier, or $1,391,724. We do not include the multiplier for purposes of the financial burden criterion because the risk in pursuing the case, for which a multiplier is utilized, has been considered as a factor in quantifying the probability of success.

Comparing estimated value of $2,613,808 or less with actual litigation costs of $1,391,724, we find it a close issue whether estimated value "exceeds by a substantial margin the actual litigation costs." ( Los Angeles Police Protective League v. City of Los Angeles, supra, 188 Cal.App.3d at p. 10, italics added.) But, as further discussed below ( post, pp. 1417-1418), we view this litigation as providing a very significant public benefit. This consumer protection action has directly benefitted hundreds of thousands of Wells Fargo customers. Moreover, it 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.

We therefore conclude that even if the estimated value of this case is viewed as exceeding actual litigation costs by a substantial margin, the public benefits from the litigation are so significant that an award of fees under section 1021.5 is appropriate. (188 Cal.App.3d at p. 10.) Thus, Wells Fargo's interest of justice argument fails along with the financial burden argument on which the former is dependent.

2. The public interest criterion.

(6a) Wells Fargo also contends this litigation did not result "in the enforcement of an important right affecting the public interest" (§ 1021.5) because it vindicated only the private rights of Wells Fargo cardholders, rather than benefitting the public as a whole, and thus there was no public interest at stake.

This argument confuses the question whether there was an important public interest at stake with the question whether a "significant benefit" has been "conferred on the general public or a large class of persons. . . ." (§ 1021.5, subd. (a).) (7) The significant benefit criterion calls for an examination whether the litigation has had a beneficial impact on the public as a whole or on a group of private parties which is sufficiently large to justify a fee award. This criterion thereby implements the general requirement that the benefit provided by the litigation inures primarily to the public. ( Marini v. Municipal Court (1979) 99 Cal.App.3d 829, 836 [ 160 Cal.Rptr. 465] .) In contrast, the question whether there was an important public interest at stake merely calls for an examination of the subject matter of the action — i.e., whether the right involved was of sufficient societal importance. ( Ibid.)

For example, in Angelheart v. City of Burbank (1991) 232 Cal.App.3d 460 [ 285 Cal.Rptr. 463], the plaintiffs successfully challenged the application of a city ordinance to a large family day-care home. The court held that although the public interest criterion of section 1021.5 was satisfied in light of the Legislature's policies underlying its preemptive regulation of such day care homes, the significant benefit element was absent because there was no evidence that the litigation had affected more than the one day care home at issue in the case. The number of persons benefitted, and hence whether the benefit was "public" or "private," was crucial to the significant benefit criterion but had nothing to do with whether there was an important public interest at stake. What mattered in the latter inquiry was whether the subject matter of the action implicated the public interest.

(6b) The present action to recover excessive fee assessments on behalf of hundreds of thousands of Wells Fargo customers, on the ground the fees were not valid as liquidated damages (Civ. Code, § 1671, subd. (d)), is appropriately characterized as a consumer protection action. (See Cal. Law Revision Com. com., 9 West's Ann. Civ. Code, § 1671 (1985) p. 497 [Deering's Ann. Civ. Code, § 1671 (1991 pocket supp.) p. 30].) Such actions have long been judicially recognized to be vital to the public interest. (E.g., Vasquez v. Superior Court (1971) 4 Cal.3d 800, 808 [ 94 Cal.Rptr. 796, 484 P.2d 964, 53 A.L.R.3d 513].) Accordingly, there was an important public interest at stake in this action. The question whether a significantly large number of "private" persons was benefitted so as to justify a fee award is pertinent only to the significant benefit criterion, which Wells Fargo expressly concedes is satisfied by virtue of the huge size of the class.

We conclude the financial burden, interest of justice and public interest criteria were satisfied in this case, and thus the plaintiffs were properly awarded their attorney fees under section 1021.5

B. The Lodestar Multiplier

(8) Next, Wells Fargo challenges the court's decision to apply the 1.5 lodestar multiplier to the award of attorney fees. The standard of review is abuse of discretion: the trial judge was in the best position to determine the value of the professional services rendered by plaintiffs' counsel, and we may not disturb the judge's decision on this point unless we are convinced it was clearly wrong. ( Serrano v. Priest, supra, 20 Cal.3d 25, 49.)

Wells Fargo contends the multiplier was inappropriate because the plaintiffs' success was "limited" and "not exceptional," in that four of their causes of action were unsuccessful and they achieved only a partial recovery on the liquidated damages theory. (See Sokolow v. County of San Mateo (1989) 213 Cal.App.3d 231, 248-249 [ 261 Cal.Rptr. 520] [trial judge must consider degree of success in determining amount of award under section 1021.5].) But the trial judge expressly considered the lack of success on the four causes of action. The order awarding attorney fees cited the "plaintiffs' failure to prevail on many issues" as a reason for applying only a 1.5 multiplier instead of the 2.0 multiplier requested by the plaintiffs. Admittedly, the plaintiffs recovered only half the damages they requested from the jury on the liquidated damages theory, but a 50 percent recovery of $5,227,617 can hardly be called insubstantial by any standard. The jury verdict was an "excellent result" given the complexity of the case and the disparity in resources available to the opposing parties. The plaintiffs' lack of success on four causes of action but achievement of a monetary recovery which expert witnesses characterized as excellent supported the judge's decision to apply a multiplier of 1.5 instead of 2.0. We are in no position to second-guess that decision.

(9) Wells Fargo also challenges the trial judge's reliance on the "contingency risk factor" as a basis for the multiplier. The judge said this factor, as evidenced in part by the expert witness declarations, demonstrated "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." The bank argues that plaintiffs' counsel did not establish they had failed to recover full compensation in other consumer class actions, and thus failed to show justification for an enhancement in this case as compensation for such losses. This argument betrays an unduly narrow construction of the contingency risk factor. A multiplier might benefit counsel by making up for specific past losses, but that is not the primary purpose of the enhancement. The purpose is to compensate for the risk of loss generally in contingency cases as a class. ( Fadhl v. City and County of San Francisco (9th Cir. 1988) 859 F.2d 649, 650.) A contingency fee is not intended to pay for previous contingency cases that the lawyer has lost. (Newberg, Attorney Fee Awards (1986) § 1.08, p. 13.) Consequently, the purpose of the multiplier does not require proof of such losses.

C. Expert Witness Fees and Other Nonrecoverable Expenses

(10) Finally, Wells Fargo challenges the award of counsel's expenses in the sum of $196,997 that were nonrecoverable as ordinary costs. (§ 1033.5, subd. (b).) Most of these expenses ($178,589) were expert witness fees; the remainder were for photocopying, postage, travel and the like.

Wells Fargo contends such expenses should not be awarded in a private attorney general action because they are not expressly authorized by section 1021.5 and are not "attorney fees" within the meaning of the statute. (See § 1033.5, subd. (b) [such expenses are not recoverable as costs "except when expressly authorized by law"].) The bank relies on West Va. Univ. Hospitals v. Casey (1991) 499 U.S. ___ [113 L.Ed.2d 68, 111 S.Ct. 1138], in which the United States Supreme Court held that expert witness fees may not be awarded in federal civil rights actions under 42 United States Code section 1988. Casey explained that because dozens of statutes in the United States Code explicitly authorize awards of attorney fees and expert witness fees in certain cases, the authorization in 42 United States Code section 1988 for recovery of "a reasonable attorney's fee" cannot be construed as including expert witness fees; otherwise, the statutes referring separately to attorney fees and expert witness fees would "become an inexplicable exercise in redundancy." (499 U.S. at p. ___ [113 L.Ed.2d at p. 79].) The California codes similarly contain numerous provisions that expressly authorize awards of both attorney fees and expert witness fees in certain cases. Wells Fargo contends the reasoning in Casey should apply equally to section 1021.5, since it, like 42 United States Code section 1988, only mentions attorney fees.

There is a crucial difference, however, with regard to section 1021.5 California's private attorney general statute was an explicit reaction to Alyeska Pipeline Co. v. Wilderness Society (1975) 421 U.S. 240 [44 L.Ed.2d 141, 95 S.Ct. 1612], which held that federal courts could no longer award attorney fees in private attorney general actions without specific statutory authorization. ( Woodland Hills Residents Assn., Inc. v. City Council, supra, 23 Cal.3d 917, 934; County of Fresno v. Lehman (1991) 229 Cal.App.3d 340, 347-348 [ 280 Cal.Rptr. 310] .) The California Supreme Court has observed that, "Because in framing the provisions of section 1021.5 the Legislature drew heavily upon the pre- Alyeska federal decisions, we believe that such authorities — while no longer viable in the federal realm — will often be helpful and reliable guides in interpreting the various provisions of the California statutes." ( Woodland Hills Residents Assn., Inc. v. City Council, supra, 23 Cal.3d at p. 934, fn. omitted.) Before Alyeska, many federal decisions that awarded attorney fees on a private attorney general theory also awarded expert witness fees and similar expenses, in accordance with the policy of encouraging private citizens to bring such suits. (E.g., Fairley v. Patterson (5th Cir. 1974) 493 F.2d 598, 606-607 fn. 17; Pyramid Lake Paiute Tribe of Indians v. Morton (D.D.C. 1973) 360 F. Supp. 669, 672, revd. (D.C. Cir. 1974) 499 F.2d 1095, 1096-1097 [163 App.D.C. 90]; La Raza Unida v. Volpe (N.D.Cal. 1972) 57 F.R.D. 94, 102; Bradley v. School Board of City of Richmond, Virginia (E.D. Va. 1971) 53 F.R.D. 28, 44, revd. (4th Cir. 1972) 472 F.2d 318, 327-331, vacated (1974) 416 U.S. 696, 724 [40 L.Ed.2d 476, 94 S.Ct. 2006]; see West Va. Univ. Hospitals v. Casey, supra, 499 U.S. at p. ___ fn. 8 [113 L.Ed.2d at pp. 89-90] (dis. opn. of Stevens, J.).)

Because our Legislature relied heavily on the pre- Alyeska federal decisions in enacting section 1021.5, we must assume the Legislature intended to adopt the pre- Alyeska federal practice of awarding expert witness fees and similar expenses when attorney fees were recoverable in private attorney general cases, consistent with the guiding purpose of encouraging public interest litigation. This indicator of legislative intent outweighs any inference we might otherwise draw from statutory comparisons of the sort conducted in Casey.

42 United States Code section 1988 was similarly enacted in response to Alyeska, but the majority and dissenting opinions in Casey disagreed regarding the extent to which that statute had been intended to return the federal courts to the pre- Alyeska law. The majority concluded that because the statute is both broader and narrower than the pre- Alyeska "regime," it had not been intended to return the federal courts "precisely" to the former law, and thus there was no reason to depart from the "normal import" of the statute's text. ( West V.A. Univ. Hospitals v. Casey, supra, 499 U.S. at p. ___ [113 L.Ed.2d at p. 82].) The dissenters believed that Congress had indeed intended a return to the pre- Alyeska practice in which federal courts could award expert witness fees. ( Id., at p. ___ [113 L.Ed.2d at pp. 89-91] (dis. opn. of Stevens, J.).) Here, in contrast, our Legislature's heavy reliance on the pre- Alyeska federal decisions is not open to similar dispute, and those decisions are considered to be "helpful and reliable guides" in interpreting section 1021.5 ( Woodland Hills Residents Assn., Inc. v. City Council, supra, 23 Cal.3d at p. 934.)

We conclude that, consistent with the pre- Alyeska federal practice underlying the enactment of section 1021.5, expert witness fees and other nonrecoverable expenses incurred by counsel may be awarded under section 1021.5 if, as is undisputed here, they represent expenses ordinarily billed to a client and are not included in the overhead component of counsel's hourly rate. (See Bussey v. Affleck (1990) 225 Cal.App.3d 1162, 1166 [ 275 Cal.Rptr. 646] [same rule for award of expert witness fees under Civ. Code, § 1717].)

D. Attorney Fees on Appeal

(11) The plaintiffs request an award of their attorney fees for successfully defending both appeals. We conclude that, as at the trial level, they are entitled to such an award under section 1021.5 ( Laurel Heights Improvement Assn. v. Regents of University of California (1988) 47 Cal.3d 376, 425-427 [ 253 Cal.Rptr. 426, 764 P.2d 278]; Serrano v. Unruh (1982) 32 Cal.3d 621, 637-639 [ 186 Cal.Rptr. 754, 652 P.2d 985].) The appropriate amount of the award is to be determined by the trial court. ( City of Oakland v. Oakland Raiders (1988) 203 Cal.App.3d 78, 86 [ 249 Cal.Rptr. 606]; see Phipps v. Saddleback Valley Unified School Dist. (1988) 204 Cal.App.3d 1110, 1123, fn. 10 [ 251 Cal.Rptr. 720].)

IV. DISPOSITION

The judgment is affirmed.

Perley, Acting P.J., and Stein, J., concurred.

Appellant's petition for review by the Supreme Court was denied March 12, 1992. Baxter, J., did not participate therein. Lucas, C.J., and Panelli, J., were of the opinion that the petition should be granted.