Opinion
BIRD, C. J.
This court granted a hearing in this case primarily to consider the appropriate methods for distributing damages in consumer class actions. That issue arose as an aspect of intervener’s appeal from the superior court’s approval of a class action settlement. As will appear below, it is now clear that as a result of the passage of time and changes in the position of intervener no practical purpose would be served by invalidating the settlement of this six-year-old class action.
However, a sizable residue has developed in the intervening period, consisting of damages due to claimants who can no longer be located and interest on those damages. This residue remains unallocated under the settlement approved by the trial court. Today’s ruling will serve as a source of guidance both for the trial court on remand in allocating the residue and for other courts in confronting the largely uncharted area of fluid recovery. Therefore, it is appropriate to address the damage distribution issue on which hearing was granted.
I.
Intervener appeals from the superior court’s approval of a class action settlement. The settlement purports to resolve the widely publicized antitrust action brought by the Attorney General on behalf of the state and consumers against Levi Strauss & Co., a clothing manufacturer. The action sought monetary relief for millions of consumers who were assertedly overcharged on jeans purchased during the early 1970’s.
As finally approved, the settlement yielded between 35 and 40 cents per pair of jeans, an average individual recovery of $2.60-$3. The Attorney General’s office was awarded $1.2 million in attorney fees, and the costs of administration were estimated at $1.8-$2.2 million. The total amount allocated to consumers was approximately $9.3 million. This amount, plus ac[465]*465cumulated interest, is currently deposited with the state pending the outcome of this appeal.
Intervener, along with 19 objectors,1 challenged both the fairness of the process which led to the settlement and the fairness and reasonableness of the outcome. First, she questioned the settlement’s plan for distributing the money to individual consumers. She argued that the court erred in approving the distribution of class funds on the basis of claims nearly 95 percent of which were unsworn and unverified. She also suggested that the individual recoveries are trivial and where—as here—there is no feasible method for verifying the overwhelming majority of claims, an aggregate remedy such as the establishment of a consumer trust fund should be preferred.
Second, intervener maintained that the trial court failed to provide adequate means for dissenting class members to participate and express their preferences. Specifically, she argued that the court denied her attempts to discover the factual and economic basis of the settlement, placed obstacles in the way of participation by objectors, failed to file or respond to the over 300 objections that were submitted, denied her a modest sum to conduct a poll of class preferences, and approved a nonneutral class notice that promoted the settlement and failed to provide information essential to an intelligent choice.
The facts are set forth in detail below.
On May 5, 1976, after an extensive investigation, the Federal Trade Commission (the Commission) issued an administrative complaint against Levi Strauss & Co. The complaint charged the company with pressuring retailers to maintain high prices in violation of federal antitrust law.
In late 1977, the Commission and the company entered into a consent decree. By the terms of the decree, the company agreed to cease all efforts to control retail pricing.
Some time later, the then Attorney General, Evelle Younger, obtained the results of the Commission’s investigation and began his own inquiry into the impact of the company’s activities on California consumers. On the basis of the information thus acquired, he tentatively decided to bring suit alleging [466]*466violations of the Cartwright Act (Bus. & Prof. Code, § 16720 et seq.),2 California’s antitrust law. He informed the company of his intent and began negotiations for a settlement.
Between March and June of 1978, the Attorney General and Levi Strauss negotiated their first proposed settlement in the amount of $3.5 million. The settlement provided that the money be distributed to individuals on a prorata basis according to the number of pairs of men’s and boys’ jeans purchased during the period 1972-1975.
In June of 1978, the state, represented by the Attorney General, commenced this action in the superior court, both as a class action on behalf of itself and similarly situated consumers (Code Civ. Proc., § 382) and as a suit in parens patriae on behalf of its residents (§ 16760). At the same time, the Attorney General notified the court of the proposed settlement.
Shortly afterward, a number of prospective interveners including Hannah Kerner, the present named intervener, learned of the proposed settlement and filed a declaration in opposition. It argued that individual distribution would produce only small individual returns and that, without prohibitively expensive security measures, a large percentage of the fund would be distributed to persons who did not in fact purchase jeans during the relevant period. It also averred that the requirement of written claim forms would discourage low income and minority members of the class from claiming and receiving their fair share of the recovery.
As an alternative to individual distribution, interveners’ declaration proposed that the settlement money be used to establish a nonprofit corporation which would administer a “consumer trust fund.” This corporation would engage in consumer protection projects, including research and litigation. It would be controlled by a board of nine directors, five appointed by the Governor and four by the Attorney General. The corporation would be organized so that it could receive funds from future class action settlements.
The court granted leave to intervene.
In the early fall of 1978, the Attorney General and the company had a dispute over the accuracy of the sales figures used as a basis for the proposed settlement. As a result, the proposed settlement was abandoned and the Attorney General commenced discovery.
In January of 1980, then Attorney General George Deukmejian moved to certify a class consisting of all current California residents who purchased [467]*467Levi Strauss products at retail during the 1972-1975 period. At various times, the Attorney General estimated that this class consisted of from several to 7 million households.
While the class certification motion was under consideration by the court, the Attorney General and the company negotiated what might be called a “gamble” agreement. This agreement provided that if the requested class or a specified variant were certified, the company would pay $12.5 million. However, if the class were not certified, the company would pay $5.5 million.
In July of 1980, the court certified the class as requested. Shortly thereafter, the Attorney General and the company informed the court of the second proposed settlement, which—like the first—provided for individual distribution. Costs of distribution were projected to be as high as $2.2 million. In addition, the Attorney General’s office was to receive an attorney’s fee of $1.2 million.
The court determined that the new proposed settlement merited consideration and granted the Attorney General $75,000 from the settlement fund to prepare a plan of distribution.
Intervener reasserted her objections to an individual distribution plan and averred that a substantial percentage, if not a majority, of the class shared her views. She moved for $12,000 to enable her to conduct a survey of class preferences and to study ways to prevent fraud and to encourage participation by minority and low income class members. This motion was denied pending the development of the Attorney General’s plan.
In November of 1980, the Attorney General submitted his plan of distribution. It provided for individual notices, printed in English and Spanish, to be mailed to 8.6 million households. An additional 3.4 million notices were to be made available at various pickup centers around the state.
The legal notice of settlement, which set forth the terms of the agreement and the options available to class members, was sandwiched between an introductory letter, signed by the Attorney General, and a claim form for those class members who desired to participate in the settlement. The Attorney General’s letter, labeled “Cash Refund Notice,” urged eligible consumers to claim their cash refunds by returning the enclosed claim form. It stated that the Attorney General “has settled” the action and that purchasers “are now eligible for a cash refund of up to $2.00 per pair.” It did not mention the legal requirement that the court approve.
[468]*468The notice was to be mailed in a folded cover instead of an envelope. The proposed design of this cover is depicted below.3
The plan provided for a two-week campaign of television and radio advertising to be aired in conjunction with the mailing of notices. This campaign, which featured personal appearances by the Attorney General, was to be organized around the pocket-and-bills logo so that recipients of the mailed notice could distinguish it from commercial bulk mailings.
The plan gave consumers six weeks from the start of the mailing process to return their claim forms and/or objections or to opt out of the settlement. No verification of purchase was required, but claims were to be analyzed according to unspecified “fraud tests.”
Intervener reasserted her objections to the individual distribution plan. In addition, she charged that the notice and media campaign improperly provided free political exposure for then Attorney General Deukmejian.
A short time thereafter, the Attorney General proposed revisions to the notice, including deletion of the “Pocket Money” slogan on the cover. At the same time, he informed the court that he would seek an entertainment industry figure to replace him in the media campaign.
In January of 1981, the court approved the distribution plan, including the notice as revised by the Attorney General. A hearing on the fairness of [469]*469the settlement was set for April 27, 1981. The Attorney General was granted $1.2 million out of the settlement fund to cover preliminary expenses.
Beginning February 2l, notices were mailed to 8.6 million households. An additional 1.5 million forms were made available at post offices. The media campaign ran simultaneously with Michael Landon replacing the Attorney General in the television commercials.
As finally approved, the notice was labeled “Notice of Cash Refund” on its address cover. The back cover contained the language: “State of California, Attorney General’s Office; Consumer Cash Refund; You May Be Eligible For A Cash Refund From The State of California; Please Read The Attached Notice Carefully.” The record does not reveal the final graphic design of the cover. The introductory letter was unchanged from the original proposed notice.4
By April 6, the deadline for responding to the notice, 1.4 million forms claiming 37 million pairs of jeans had been received. Between 345 and 510 objections were submitted,5 and 111 exclusions were filed. The objections were never made part of the record.
According to the Attorney General, 24 percent of the claims alleged 10 or more purchases per individual during the 5-year period. Twelve percent claimed more than 15 purchases, 5 percent more than 21, and 1 percent more than 35.
The Attorney General recommended that the top 1 percent of the claims be treated as suspect.6 Each person submitting a suspect claim would be required to submit a notarized confirmation that the information contained [470]*470in the claim was accurate. Failure to do so would result in denial of compensation.
The fairness hearing was held as scheduled on April 27. Plaintiff, defendant, and intervener each argued their positions concerning the fairness of the settlement. An undetermined number of objectors also attended but did not testify. From the record, it appears that the court neither prohibited nor solicited their participation.
Prior to the hearing, both plaintiff and intervener had moved for attorney fees, plaintiff for $1.2 million and intervener for $120,000. However, these motions were not discussed at the hearing. Instead, by agreement of the parties and the court, the question of attorney fees was postponed to a second hearing.
On May 15, the trial court approved the settlement. It ordered that the top 5 percent of claims (more than 21 pairs of jeans per individual) be treated as suspect. It also granted the Attorney General’s motion for attorney fees and denied intervener’s. The promised hearing on the question of attorney fees was never held.
In early November, the state issued a progress report on claims processing. Pursuant to the court’s order designating the top 5 percent of claims as suspect, over 75,000 claims had been returned with instructions to resubmit with a notarized statement confirming their veracity. Over 80 percent of these claims were not resubmitted. A similar percentage of claims returned as suspect because of excessive household membership (over 12 individual claimants) were not resubmitted. In all, over $1.6 million in suspect claims were eliminated.
Pursuant to intervener’s request for findings of fact and conclusions of law, the state submitted proposed findings. Intervener responded with numerous objections. On November 16, 1981, the court entered its final judgment, adopting verbatim plaintiff’s proposed findings.
II.
The course of events described above presents this court with a difficult problem. Nearly $1.5 million of the settlement fund has already been spent on the present distribution plan. Intervener criticizes this expenditure, but does not contend that reversal would somehow restore the funds. Further, over 1 million household claims have been received and processed, thereby inducing legitimate expectations of compensation among class members. Intervener remains critical of this method of distribution to individual con[471]*471sumers on the basis of largely unverified claims. Nonetheless, at oral argument, intervener joined amici Consumers Union et al. in urging that— with some additional security precautions—these claims be honored. However, this court can preserve the present claims only by upholding the settlement. (See Trotsky v. Los Angeles Fed. Sav. & Loan Assn. (1975) 48 Cal.App.3d 134, 153 [121 Cal.Rptr. 637], [a settlement must be approved or rejected as a whole since it is impossible to determine whether the parties considered the different portions of the agreement to be interdependent].) In view of the concessions made by intervener at oral argument, it would serve no practical purpose to invalidate the settlement in this six-year-old class action.
In a sense, this court is confronted with a fait accompli. The only effective means to prevent further expenditures of class funds and to vindicate the claimants’ current expectations—ends desired by all the parties—is to affirm the settlement.
Indeed, it appears that intervener and amici are principally concerned not with reversing the settlement, but with advocating a particular use for the residue remaining after individual distribution. They estimate that this residue will be substantial since the settlement fund has accumulated considerable interest and since many claimants can no longer be located. Intervener and amici have devoted their principal efforts to arguing that this sum should be used to establish a “consumer trust fund.” Their proposal is entirely consistent with affirmance.
ra.
Since the pathbreaking case of Vasquez v. Superior Court (1971) 4 Cal.3d 800 [94 Cal.Rptr. 796, 484 P.2d 964, 53 A.L.R.3d 513], the California courts have recognized that the consumer class action is an essential tool for the protection of consumers against exploitative business practices. The Vasquez decision challenged the trial courts to develop “pragmatic procedural devices” to “simplify the potentially complex litigation while at the same time protecting the rights of all the parties.” (Id., at p. 820; cf. Sindell v. Abbott Laboratories (1980) 26 Cal.3d 588, 610 [163 Cal.Rptr. 132, 607 P.2d 924, 2 A.L.R.4th 1061], cert. den. 449 U.S. 912 [66 L.Ed.2d 140, 101 S.Ct. 285, 101 S.Ct. 286] [courts must fashion consumer class action remedies to meet the changing needs of modern industrial society].)
Damage distribution, the crux of the present case, poses special problems in consumer class actions. Often, proof of individual damages by competent evidence is not feasible. Each individual’s recovery may be too small to [472]*472make traditional methods of proof worthwhile. In addition, consumers are not likely to retain records of small purchases for long periods of time.
In response to these problems, the courts have turned to the equitable doctrine of cy pres. This doctrine originated in the law of charitable trusts. Where compliance with the literal terms of a charitable trust became impossible, the funds would be put to “the next best use,” in accord with the dominant charitable purposes of the donor. (See Estate of Tarrant (1951) 38 Cal.2d 42, 49 [237 P.2d 505, 28 A.L.R.2d 419], and cases cited.) In the class action context, the cy pres doctrine is generally denominated “fluid recovery.”
The propriety of fluid recovery in a particular case depends upon its usefulness in fulfilling the purposes of the underlying cause of action. (Bruno v. Superior Court (1981) 127 Cal.App.3d 120, 129 [179 Cal.Rptr. 342] [hereafter Bruno], citing Simer v. Rios (7th Cir. 1981) 661 F.2d 655, 676; Blue Chip Stamps v. Superior Court (1976) 18 Cal.3d 381, 389 [134 Cal.Rptr. 393, 556 P.2d 755] (cone. opn. of Tobriner, J.).) Fluid recovery may be essential to ensure that the policies of disgorgement or deterrence are realized. (Simer v. Rios, supra, 661 F.2d at p. 676.) Without fluid recovery, defendants may be permitted to retain ill gotten gains simply because their conduct harmed large numbers of people in small amounts instead of small numbers of people in large amounts.
The parties do not dispute that fluid recovery methods may be employed in appropriate Cartwright Act cases. (See Bruno, supra, 127 Cal.App.3d at p. 132.) As the Bruno court pointed out, the purposes of the private damages action for violations of the Cartwright Act include disgorgement and deterrence as well as compensation. (Id., at pp. 132-133.)7
The implementation of fluid recovery involves three steps. (See generally, Malina, Fluid Class Recovery as a Consumer Remedy in Antitrust Cases (1972) 47 N.Y.U. L.Rev. 477, 482.) First, the defendant’s total damage liability is paid over to a class fund. Second, individual class members are afforded an opportunity to collect their individual shares by proving their particular damages, usually according to a lowered standard of proof. Third, any residue remaining after individual claims have been paid is dis[473]*473tributed by one of several practical procedures that have been developed by the courts.
In order to assess the propriety of the fluid recovery method selected by the trial court in the present case, it is necessary to review briefly the available alternatives. The principal methods include a rollback of the defendant’s prices, escheat to a governmental body for either specified or general purposes, establishment of a consumer trust fund, and claimant fund sharing. All of these methods promote the policies of disgorgement and deterrence by ensuring that the residue of the recovery does not revert to the wrongdoer. However, they differ substantially in their compensatory effect and in their suitability for particular cases.
In determining which method to employ, the courts should consider: (1) the amount of compensation provided to class members, including non-claiming (or “silent”) members; (2) the proportion of class members sharing in the recovery; (3) the extent to which benefits will “spill over” to nonclass members and the degree to which the spillover benefits will effectuate the purposes of the underlying substantive law; and (4) the costs of administration. (See generally, Shepherd, Damage Distribution in Class Actions: The Cy Pres Remedy (1972) 39 U.Chi. L.Rev. 448, 464 [hereafter Cy Pres Remedy].)
Under the price rollback approach, the uncollected portion of the fund is distributed through the market by lowering the price of the defendant’s product for a specified period.8 For example, the class action in Daar v. Yellow Cab (1967) 67 Cal.2d 695 [63 Cal.Rptr. 724, 433 P.2d 732] was resolved by “refunding” taxicab overcharges through a court-ordered reduction in fares. (See Blue Chip Stamps v. Superior Court, supra, 18 Cal.3d at p. 388, fn. 1 (conc. opn. of Tobriner, J.).)
Price rollbacks are particularly effective for remedying overcharges on items which are repeatedly purchased by the same individuals. In cases involving such “repeat items,” the individuals who were overcharged are likely to benefit from the reduced prices. (See 3 Newberg, Class Actions (1977) § 4620, p. 85; see also Cy Pres Remedy, supra, 39 U.Chi. L.Rev. at pp. 461-462.)
None of the parties has proposed a price rollback in the present action. This method is not appropriate in nonmonopoly markets like the jeans market since it compels consumers to collect their refunds by making further [474]*474purchases of the defendant’s products, to the detriment of the defendant’s competitors. (See 3 Newberg, Class Actions, supra, § 4620, p. 85; CyPres Remedy, supra, 39 U.Chi. L.Rev. at pp. 462-463.) Further, the price of jeans to the consumer is fixed not by the defendant, but by independent retailers. Hence, a price rollback would pose difficult if not insuperable management problems.
A second form of fluid distribution is “earmarked escheat.” Under this approach, the uncollected funds are disbursed to a responsible governmental organization for use on projects that benefit noncollecting class members and promote the purposes of the underlying cause of action.
Market St. Ry. Co. v. Railroad Commission (1946) 28 Cal.2d 363 [171 P.2d 875], though not a class action, provides an example of this method. In that case, this court was faced with the problem of how to refund over $700,000 in streetcar overcharges. Individual riders had been offered an opportunity to collect refunds by submitting sworn applications, but less than $15,000 had been claimed. This court awarded the remaining funds to the City and County of San Francisco, which had purchased the railway subsequent to the overcharges, for the improvement of street car services. The court reasoned that the people of San Francisco, who had paid the overcharges in the first place, would benefit from the improvements. (Id., at pp. 366, 372-373.)9
Earmarked escheat provides indirect compensation to silent class members. The recipient governmental body may use class funds to ameliorate the effects of past harm and to reduce the risk of future harm. Administrative costs are minimized by utilizing already extant governmental bodies to administer the fund.10
[475]*475Since earmarked escheat depends for its success upon the active cooperation of the government, it should not be employed where—as in the present case—the relevant governmental body opposes its use.
Another alternative is general escheat: payment of the recovery to the general fund of a governmental body. Of all the fluid recovery devices, general escheat provides the least focused compensation to the class. The benefits of the recovery are spread among all taxpayers, and there is no attempt to ensure that the spillover is used to effectuate the purposes of the substantive law. (See generally, Economic Analysis, supra, 34 Stan.L.Rev. at p. 180.) The only advantage of general escheat is ease of administration. Hence, it is usually regarded as a last resort for use where a more precise remedy cannot be found. (See, e.g., Developments in the Law—Class Actions (1976) 89 Harv.L.Rev. 1318, 1523, fn. 353.)
Intervener advocates a relatively new and increasingly popular form of fluid recovery—the establishment of a “consumer trust fund.” Under this approach, the court appoints a board of directors to administer the recovery in the interests of the class. (See generally, Nash, Collecting Overcharges From the Oil Companies: The Department of Energy’s Restitutionary Obligation (1980) 32 Stan.L.Rev. 1039, 1057-1058.)
Like earmarked escheat, the consumer trust fund device uses the residue to further the purposes of the substantive law and provide indirect compensation to class members. Unlike earmarked escheat, there is no danger that the recovery will be submerged in the state’s general fund. However, the consumer trust fund device does entail the establishment of a new organization with its own administrative expenses. To avoid this additional cost, some courts have allocated the funds directly to responsible private organizations. (For examples of this approach, see cases cited in In re Folding Carton Antitrust Litigation (N.D.Ill. 1983) 557 F.Supp. 1091, 1109, fn. 10, affd. in part and revd. in part (7th Cir. 1984) 744 F.2d 1252.)11
Finally, the residue, or the fund itself, may be divided among the individual claimants. As initially approved, the present settlement incorpo[476]*476rated this “claimant fund sharing” approach. (See generally, Economic Analysis, supra, 34 Stan.L.Rev. at pp. 177-178.)12
Unlike other methods of fluid recovery, claimant fund sharing uses the entire class recovery to provide monetary compensation to individual class members. Hence, in an appropriate case, it may yield a greater and more direct compensatory impact.
However, claimant fund sharing provides no benefits to silent class members. Further, if there is a windfall, it goes not to further the purposes of the substantive law, but to overcompensate legitimate claimants or to pay erroneous or fraudulent claims. Hence, the advantages of claimant fund sharing can only be realized where a large proportion of class members participate and submit accurate claims. (See generally, Economic Analysis, supra, 34 Stan.L.Rev. at p. 178.)
Intervener argues that the trial court abused its discretion in approving the present claimant fund sharing plan. First, she contends that the individual recoveries were so small that only a minority of class members bothered to file claims. She points out that although over 1 million households did come forward, that number must be measured against the total size of the class, estimated by the Attorney General at between 3 and 7 million households. If the fund had been distributed as planned, 60 to 80 percent of the class would have received no compensation—direct or indirect.
In addition, she asserts that a large percentage of the recovery went to pay inflated or baseless claims. The accuracy of the distribution hinged on class members’ ability to recall small purchases made more than six years earlier. An informal poll conducted by intervener suggested that a large percentage of consumers could not remember the number of Levi’s jeans they had purchased during the relevant period.
This problem was, according to intervener, exacerbated by widespread fraud. The settlement provided no safeguard against fraud for 95 percent of the claims. The 5 percent of claimants who alleged the largest number of jeans were required to resubmit their claims with a sworn statement. Over 80 percent of these claimants preferred to abandon their refunds rather than confirm their claims on pain of perjury. The magnitude of this negative [477]*477response casts doubt on the accuracy of claims beyond the top 5 percent as well.
Further, there appears to be no reported decision in any jurisdiction approving a plan so lacking in safeguards against fraud. The Attorney General cites In re Coordinated Pretrial Proceedings, etc. (D.Minn. 1975) 410 F.Supp. 706. However, the distribution plan in that case required proof for all above-average claims. (Id., at p. 713.) The opinion does not reveal whether notarization or other sworn verification was required.
Finally, intervener suggests that administrative expenses and attorney fees were grossly disproportionate to the compensation received by the average claimant. Administrative expenses were projected at over $2 million. The Attorney General was awarded $1.2 million in attorney fees. By contrast, the average individual’s recovery was projected at the time of the court’s ruling on the settlement to total between $2.60 and $3, or 30-40 cents per pair of jeans. Hence, it appears that the only substantial beneficiaries of the settlement were a few large claimants—including the state—and the Attorney General, who gained not only the fee award for his office, but also substantial free publicity for his then upcoming gubernatorial campaign.
The dissent’s response to intervener’s arguments is twofold. First, the dissent suggests that trial courts should turn to the parens patriae statute (§ 16760) for guidance on how to distribute damages in private consumer class actions. (See cone. & dis. opn., post, at p. 490.) However, that statute was not intended to limit the equitable discretion of the courts in managing private consumer class actions. Nor should the courts turn to its narrow provisions for “guidance” on how to exercise their discretion.
The parens patriae statute empowers the Attorney General to sue on behalf of persons injured by violations of the Cartwright Act. (§ 16760, subd. (a)(1).) Its damage distribution provisions are narrowly drawn. Subdivision (e)(1) requires that monetary relief recovered pursuant to the parens patriae power be distributed “[i]n such manner as the superior court ... in its discretion may authorize to insure, to the extent possible, that each person be afforded a reasonable opportunity to secure his appropriate portion . . . .” Subdivision (e)(3) provides that the residue “shall be treated . . . as unclaimed property,” and eventually escheat to the state’s general fund.
The concurring and dissenting opinion’s reliance on the parens patriae statute finds no justification in law or policy. Far from being intended to supersede or restrict the private consumer class action, the parens patriae action is designed to provide a means of redress where a private class action is not viable. (See generally, Jones, Perspectives in Consumer Advocacy: [478]*478Antitrust Parens Patriae Suits Pursuant to the Hart-Scott-Rodino Antitrust Improvements Act—A Solution for Wrongs Without Redress (1977) 5 Pepperdine L.Rev. 77, 78.)13
The Court of Appeal’s well-reasoned opinion in Bruno, supra, 127 Cal.App.3d 120 is instructive on the relation between the parens patriae power and private class actions. In that case, the plaintiff class proposed that damages be distributed by a price rollback or earmarked escheat. The defendant sought to strike these methods from the complaint, arguing that the parens patriae statute prohibited them in private consumer class actions. (Id., at p. 133.) The court held that the proposed methods should not be stricken from the complaint and stated: “The fact that the law empowers the Attorney General ... to bring parens patriae antitrust lawsuits and specifies the methods for distributing the damages they recover evidences no legislative intent to prohibit fluid class recovery in private antitrust class actions.” (Ibid.)
In short, as the Attorney General himself convincingly argued in his memorandum in support of certification of the present class, the parens patriae statute was not intended to impose any restrictions on the conduct of private consumer class actions.14
Nor should the courts seek guidance from the parens patriae statute. The two-pronged approach of the statute—individual distribution or general es-cheat—is overly restrictive in the class action context. It would disfavor or preclude two of the most widely used and effective methods of fluid recovery—the earmarked escheat and the consumer trust fund. Moreover, general escheat, the residue distribution method mandated by the statute, provides the least focused compensatory effect of any of the various methods of fluid recovery. (See ante, at p. 475.) To compel the use of this method would be to cripple the “substantial compensatory function” of the private class action (Bruno, supra, 127 Cal.App.3d at p. 134).15
[479]*479Aside from invoking the parens patriae statute, the concurring and dissenting opinion asserts that the distribution plan reasonably compensated class members since about 1.2 million families may ultimately recover between $10 and $12 each. Though the 1.2 million figure seems impressive at first glance, it is meaningful only when compared with the total class size. As the concurring and dissenting opinion recognizes, it is the “ ‘correlation’ or ‘overlap’ between the injured class of persons and the class to be benefitted” (cone. & dis. opn., post, at p. 490)—not the absolute number of persons recovering—that provides the principal criterion for assessing the compensatory effectiveness of a distribution plan.
The trial court made no finding as to the size of the class. Further, the court denied intervener’s attempt to discover that information. The Attorney General estimated the class size at between 3 and 7 million households. With only this broad estimate to go by, the 1.2 million figure—between 20 and 40 percent of the class—does not support the concurring and dissenting opinion’s summary assertion that the settlement as initially approved reasonably compensated class members. This conclusion is strengthened by the likely high incidence of erroneous or fraudulent claims.
TV.
Due to the practical considerations set forth earlier in this opinion, this court has concluded that the judgment should be affirmed. But far more is at stake than the decision whether to affirm or reverse. Today’s ruling will serve as a source of guidance for both the trial court on remand and for other courts in confronting the largely uncharted area of fluid recovery in consumer class actions.
In both contexts, the sound approach of Vasquez continues to provide the proper framework. The trial courts should have the full range of alternatives at their disposal. In choosing the appropriate method, they should consider the amount of compensation provided to class members, the proportion of class members sharing in the recovery, the size and effect of the spillover to nonclass members, and the costs of administration.
The disposition of the residue on remand is a matter within the discretion of the trial court. However, it should be noted that under the criteria set forth here, amici’s suggestion that the residue should be used to establish a consumer trust fund has considerable merit. It is estimated that up to one-half of the original group of claimants may have moved since 1981. Claimants who have moved may be difficult to locate. Hence, further attempts to distribute the entire fund to individual claimants would result in an even lower rate of participation than did the original plan. Amici’s consumer [480]*480trust fund plan—or a plan employing earmarked escheat in place of the consumer trust fund—would provide some indirect compensation for silent class members, while ensuring that the residue of the recovery is used on projects designed to effectuate the aims of the Cartwright Act.
The judgment is affirmed. The cause is remanded to the trial court for further proceedings consistent with the views expressed in this opinion.
Broussard, J., Reynoso, J., and Sutter (John), J.,
Judge, Alameda County Superior Court, assigned by the Chairperson of the Judicial Council.