I dissent.
Plaintiffs filed a simple seven-page complaint alleging a single cause of action for breach of warranty after the defendant had already acknowledged its marketing mistake and was taking steps to correct it, and while the Santa Cruz County District Attorney and the California Attorney General were investigating the matter and preparing to take appropriate action. The complaint constituted plaintiffs’ entire legal effort regarding the underlying lawsuit. They obtained no judicial ruling of any kind in their favor. Nevertheless, *585to date, plaintiffs have parlayed this complaint into an award of attorney fees of $762,830, most of it for work unrelated to the underlying lawsuit. Now the majority remands the matter for yet more litigation. I disagree for several reasons.
This court has never awarded attorney fees to a party with no judicial ruling in its favor. We should not start now. Relying solely on federal cases that have been overruled and California cases that either denied attorney fees or involved a plaintiff with a judicial ruling in its favor, the majority permits an award of attorney fees to the plaintiffs as the “prevailing” or “successful” party. To do so, it adopts the so-called catalyst theory, a theory that was once prevalent in federal courts, but that the United States Supreme Court has now repudiated. We should not resurrect it.
Moreover, plaintiffs do not qualify for attorney fees even under the majority’s catalyst theory. Their lawsuit was unnecessary when filed, it was moot within days of its filing, and it conferred no substantial public benefit. Plaintiffs have also failed to show their suit had any merit in light of the corrective steps defendant had already taken. The majority implicitly recognizes that plaintiffs failed to justify their award of attorney fees, but it inexplicably remands the matter for yet more litigation, which will undoubtedly increase plaintiffs’ attorney fee demand to a truly astronomic amount. I disagree here also. No reason appears to give plaintiffs a second chance to try to prove what they failed to prove the first time. Courts should seek to resolve litigation, not perpetuate it.
Finally, the majority permits qualifying plaintiffs to receive not only (1) attorney fees for litigating the underlying lawsuit, but also (2) a multiplier on those fees, and also (3) attorney fees for litigating their entitlement to attorney fees, and also (4) a multiplier on the fees for litigating entitlement to fees. I disagree on the final point. Surely, awarding fees for the underlying litigation, with a potential multiplier, plus fees for litigating entitlement to fees, is sufficient. A multiplier for litigating fees on fees is excessive and can only lead to outrageously inflated awards like the one here, where a simple complaint is transformed into an award of over three-quarters of a million dollars.
The majority today goes further than this court has ever gone before— indeed, so far as I can tell, further than any other court has ever gone—in permitting plaintiffs to win large attorney fee awards. I cannot agree. Lest California truly become a mecca for plaintiffs and plaintiffs’ attorneys throughout the country, we need to be at least somewhat in step with the rest of the country.
*586I. The Facts and Procedural History
DaimlerChrysler Corporation (DaimlerChrysler) incorrectly marketed its 1998 and 1999 Dakota R/T trucks as having a 6,400-pound towing capacity when they actually could tow only 2,000 pounds. The error occurred because the Dakota R/T was a sporty version of an existing truck model, which could tow 6,400 pounds. However, to obtain a sporty design, DaimlerChrysler lowered the suspension on the Dakota R/T, thus reducing its towing capacity. During these two .years, DaimlerChrysler sold or leased fewer than 7,000 of the Dakota R/T’s nationwide, including fewer than 1,000 in California.
DaimlerChrysler became aware of the mistake by early 1999. By February 1999, it had set up a response team to address the problem. By June 1999, DaimlerChrysler had replaced the incorrect marketing materials, owners manuals, and engine and door labels for not-yet-sold Dakota R/T’s. DaimlerChrysler had also notified existing buyers of the error, told them not to attempt to tow more than 2,000 pounds, and provided them with the same modified materials. It told buyers who wanted to tow more than 2,000 pounds they could do so only if their Dakota R/T was modified with a trailer hitch costing $300. DaimlerChrysler also began to address remedial measures for customers who had bought or leased their Dakota R/T’s under the incorrect marketing program. Many R/T buyers never intended to tow more than 2,000 pounds. When informed by DaimlerChrysler of the error, most of those customers were satisfied with DaimlerChrysler’s offers of cash and merchandise. Initially, DaimlerChrysler offered buyers who had bought the hitches refunds of the $300 cost. By the summer 1999, DaimlerChrysler authorized dealers to repurchase or replace Dakota R/T’s on a case-by-case basis for customers who demanded such a remedy.
On July 29, 1999, the Santa Cruz County District Attorney contacted DaimlerChrysler about the problem, threatened legal action, and requested DaimlerChrysler’s input before acting. On August 10, 1999, the California Attorney General notified DaimlerChrysler that it had joined the Santa Cruz County District Attorney. The public agencies requested a response by the end of August 1999.
On August 23, 1999, plaintiffs filed the seven-page complaint underlying this appeal. They alleged that they had bought 1999 Dakota R/T’s from various DaimlerChrysler dealers. One of the plaintiffs lived and bought his truck in California. Plaintiffs alleged a single cause of action for breach of express warranty based on the mistake regarding the trucks’ towing capacity. They alleged that DaimlerChrysler acknowledged the error by letter to all purchasers dated June 16, 1999. They alleged that they had previously notified DaimlerChrysler of their trucks’ failure to comply with the warranted *587towing capacity and that they were revoking their acceptance of their trucks. They sought, but never obtained, class certification for all buyers of Dakota R/T’s nationwide. They also sought return of their purchase or lease payments, compensatory damages, and attorney fees. Nothing in the complaint referred to any threat to public safety or requested any remedy related to public safety.
The day the lawsuit was filed, the Detroit News contacted DaimlerChrysler’s legal counsel about the lawsuit. DaimlerChrysler’s counsel claimed DaimlerChrysler had responded appropriately to the marketing error, including offering buybacks to customers who requested them. Plaintiffs faxed their complaint to DaimlerChrysler the same day. The next day, August 24, 1999, DaimlerChrysler’s employee newsletter ran an article on the plaintiffs’ case. DaimlerChrysler’s response team met throughout August 1999. The team knew about both public agency inquiries and the response deadline. DaimlerChrysler wrote to the public agencies that its internal approval process prohibited a response by August 31, but promised a response by September 8, 1999. On September 10, 1999, DaimlerChrysler informed all buyers of Dakota R/T’s that, among other options, DaimlerChrysler would repurchase or assist in replacing their 1998 or 1999 Dakota R/T’s. Evidence showed that the response team was aware of plaintiffs’ lawsuit before September 10, 1999.
DaimlerChrysler demurred to the complaint. Plaintiffs filed an amended complaint, acknowledging DaimlerChrysler’s offer of, among other remedies, repurchase or replacement of the trucks for all previous buyers. The trial court sustained the demurrer without leave to amend and dismissed the case as moot because DaimlerChrysler had already offered all purchasers the relief plaintiffs sought.
The public agency investigation continued. That investigation revealed that some brochures containing the error were distributed as late as August 1999. In late 2000, DaimlerChrysler settled the public agency investigation by paying a $75,000 fine and agreeing to continue to assure that the marketing error did not reoccur.
Although the court dismissed plaintiffs’ case, the parties continued to litigate plaintiffs’ entitlement to attorney fees. As the Court of Appeal described it, “Over a year of hotly-contested discovery and other motions occurred to clarify the facts described above.” The trial court held three contested hearings on the fee request. On October 18, 2000, the court held a lengthy evidentiary hearing and made factual findings rejecting DaimlerChrysler’s claim that it had at least decided to offer all buyers repurchase or buybacks before plaintiffs filed their case. The court found *588plaintiffs’ case was a catalyst for DaimlerChrysler’s eventual offer. It found that this action was necessary despite the public agency investigation because the public agencies had not yet commenced any actual proceeding against plaintiffs, and they “were only concerned with DaimlerChrysler’s false advertising materials and never sought any remedies on behalf of the consumers . . . .” It also found that plaintiffs’ action enforced “consumer rights, including highway safety,” and conferred a significant public benefit, including pecuniary benefits for consumers and “enhanced safety.” It found an additional benefit “if DaimlerChrysler and/or other manufacturers are deterred from similar conduct in the future.”
The court found the “lodestar” fee amount (i.e., the number of hours of work multiplied by a reasonable hourly compensation; see Ketchum v. Moses (2001) 24 Cal.4th 1122, 1131-1132 [104 Cal.Rptr.2d 377, 17 P.3d 735]) was $329,620 through the October 18, 2000, hearing. It awarded a 2.25 multiplier for the fees until the October 18, 2000, hearing to account for risk and success. Ultimately, it awarded a total of $762,830 in attorney fees. It did not distinguish how much of this total was due to the underlying litigation and how much of it to litigating entitlement to attorney fees. However, DaimlerChrysler states and, at oral argument, plaintiffs agreed that roughly 90 percent of this award was for fees plaintiffs generated while seeking fees.
DaimlerChrysler appealed limited to the question of attorney fees. The Court of Appeal affirmed the judgment, and we granted DaimlerChrysler’s petition for review.
II. Discussion
A. California should not adopt the catalyst theory.
“California follows what is commonly referred to as the American rule, which provides that each party to a lawsuit must ordinarily pay his own attorney fees. [Citations.] The Legislature codified the American rule in 1872 when it enacted Code of Civil Procedure section 1021, which states in pertinent part that ‘Except as attorney’s fees are specifically provided for by statute, the measure and mode of compensation of attorneys and counselors at law is left to the agreement, express or implied, of the parties. . . .’ ” (Trope v. Katz (1995) 11 Cal.4th 274, 278-279 [45 Cal.Rptr.2d 241, 902 P.2d 259].)
Code of Civil Procedure section 1021.5, enacted in 1977, provides an exception to this American rule. As relevant, it states that, “[u]pon motion, a court may award attorneys’ fees to a successful party against one or more opposing parties in any action which has resulted in the enforcement of an *589important right affecting the public interest if’ certain requirements are met.1 Although not at issue here, Government Code section 12965, subdivision (b), part of the California Fair Employment and Housing Act, is similar. That section provides as relevant: “In actions brought under this section, the court, in its discretion, may award to the prevailing party reasonable attorney’s fees and costs, including expert witness fees, except where the action is filed by a public agency or a public official, acting in an official capacity.” (Ibid.)
The issue here is what it takes to be a “successful” or “prevailing” party within the meaning of these statutes. (I agree with the majority that these terms are synonymous for these purposes.) (Maj. opn., ante, at p. 570.) Although plaintiffs did not obtain any judicial ruling in their favor, they claim entitlement to attorney fees as the successful party because their lawsuit was a “catalyst” that caused DaimlerChrysler to offer the relief they sought. We have never awarded attorney fees predicated on the catalyst theory, but we have discussed it. As we explained in Westside Community for Independent Living, Inc. v. Obledo (1983) 33 Cal.3d 348 [188 Cal.Rptr. 873, 657 P.2d 365] (Westside Community) (a case that reversed an award of attorney fees), “Numerous federal decisions have . . . [held] that attorney fees may be proper whenever an action results in relief for the plaintiff, whether the relief is obtained through a ‘voluntary’ change in the defendant’s conduct, through a settlement, or otherwise. [Citations.] [j[] Thus, an award of attorney fees may be appropriate where ‘plaintiffs’ lawsuit was a catalyst motivating defendants to provide the primary relief sought . . . .’ [Citation.] A plaintiff will be considered a ‘successful party’ where an important right is vindicated ‘by activating defendants to modify their behavior.’ [Citation.]” (Id. at pp. 352-353.)
Although, as we explained in Westside Community, supra, 33 Cal.3d 348, lower federal courts had generally recognized the validity of the catalyst theory, the United States Supreme Court recently rejected it as a basis for *590awarding attorney fees to a “prevailing party.” (Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources (2001) 532 U.S. 598 [149 L.Ed.2d 855, 121 S.Ct. 1835] (Buckhannon).) In that case, Buckhannon Board and Care Home, Inc. (Buckhannon), operated care homes that provided assisted living to their residents. It failed a state inspection because some of the residents were incapable of “self-preservation” as required under state law. (Id. at p. 600.) After receiving cease-and-desist orders requiring closure of its facilities, it brought suit in federal court against the State of West Virginia and others claiming that the “self-preservation” requirement violated the Fair Housing Amendments Act of 1988 (42 U.S.C. § 3601 et seq.) and the Americans with Disabilities Act of 1990 (42 U.S.C. § 12101 et seq.). The defendants agreed to stay enforcement of the cease-and-desist orders pending resolution of the case and the parties began discovery. In the meantime, the West Virginia Legislature enacted legislation eliminating the “self-preservation” requirement. The district court then dismissed the case as moot. Buckhannon requested attorney fees under two statutes that permitted the court to award attorney fees to the “prevailing party.” (Buckhannon, supra, at pp. 600-601; see 42 U.S.C. § 3613(c)(2) [“[T]he court, in its discretion, may allow the prevailing party ... a reasonable attorney’s fee and costs”]; 42 U.S.C. § 12205 [“[T]he court. . . , in its discretion, may allow the prevailing party ... a reasonable attorney’s fee, including litigation expenses, and costs”].) Buckhannon argued, as plaintiffs argue here, “that they were entitled to attorney’s fees under the ‘catalyst theory,’ which posits that a plaintiff is a ‘prevailing party’ if it achieves the desired result because the lawsuit brought about a voluntary change in the defendant’s conduct.” (Buckhannon, supra, at p. 601.)
The high court began its analysis by noting that in the United States parties ordinarily must bear their own attorney fees, but Congress has authorized the award of such fees to the “prevailing party” in numerous statutes. (Buckhannon, supra, 532 U.S. at pp. 602-603.) “In designating those parties eligible for an award of litigation costs, Congress employed the term ‘prevailing party,’ a legal term of art. Black’s Law Dictionary 1145 (7th ed. 1999) defines ‘prevailing party’ as ‘[a] party in whose favor a judgment is rendered, regardless of the amount of damages awarded <in certain cases, the court will award attorney’s fees to the prevailing partyx—Also termed successful party.’ This view that a ‘prevailing party’ is one who has been awarded some relief by the court can be distilled from our prior cases.” (Id. at p. 603.) “In addition to judgments on the merits, we have held that settlement agreements enforced through a consent decree may serve as the basis for an award of attorney’s fees. [Citation.] Although a consent decree does not always include an admission of liability by the defendant [citation], it nonetheless is a court-ordered ‘chang[e] [in] the legal relationship between [the plaintiff] and the defendant.’ [Citations.] These decisions, taken together, *591establish that enforceable judgments on the merits and court-ordered consent decrees create the ‘material alteration of the legal relationship of the parties’ necessary to permit an award of attorney’s fees. [Citations.]” (Id. at p. 604, fn. omitted.)
The court recognized that some of its cases contain dicta supporting the catalyst theory but noted that its holdings have never applied it; its cases awarding attorney fees involved a judgment on the merits or at least a consent decree. (Buckhannon, supra, 532 U.S. at pp. 603-604 & fns. 5, 7.) It concluded that “the ‘catalyst theory’ falls on the other side of the line from these examples. It allows an award where there is no judicially sanctioned change in the legal relationship of the parties. ... A defendant’s voluntary change in conduct, although perhaps accomplishing what the plaintiff sought to achieve by the lawsuit, lacks the necessary judicial imprimatur on the change. Our precedents thus counsel against holding that the term ‘prevailing party’ authorizes an award of attorney’s fees without a corresponding alteration in the legal relationship of the parties.” (Id. at p. 605.) In response to the dissent’s suggestion that it suffices if the plaintiff shows that the lawsuit stated a “colorable” and not “groundless” claim (id. at p. 627 (dis. opn. of Ginsburg, J.)), the court disagreed “that the term ‘prevailing party’ authorizes federal courts to award attorney’s fees to a plaintiff who, by simply filing a nonftivolous but nonetheless potentially meritless lawsuit (it will never be determined), has reached the ‘sought-after destination’ without obtaining any judicial relief.” (Id. at p. 606.)
In response to the policy arguments that the catalyst theory was necessary to prevent defendants generally from unilaterally mooting actions before judgment to avoid paying attorney fees and to not deter those plaintiffs with meritorious but expensive cases from bringing suit, the court cited contrary policy arguments. It noted “the disincentive that the ‘catalyst theory’ may have upon a defendant’s decision to voluntarily change its conduct, conduct that may not be illegal.” (Buckhannon, supra, 532 U.S. at p. 608.) It also noted “that ‘[a] request for attorney’s fees should not result in a second major litigation,’ [citation], and [the court has] accordingly avoided an interpretation of the fee-shifting statutes that would have ‘spawn[ed] a second litigation of significant dimension,’ [citation]. Among other things, a ‘catalyst theory’ hearing would require analysis of the defendant’s subjective motivations in changing its conduct, an analysis that ‘will likely depend on a highly factbound inquiry and may turn on reasonable inferences from the nature and timing of the defendant’s change in conduct.’ [Citation.] Although we do not doubt the ability of district courts to perform the nuanced ‘three thresholds’ test required by the ‘catalyst theory’—whether the claim was colorable rather than groundless; whether the lawsuit was a substantial rather than an insubstantial cause of the defendant’s change in conduct; whether the defendant’s change in conduct was motivated by the plaintiff’s threat of victory rather *592than threat of expense [citation to the dissenting opinion]—it is clearly not a formula for ‘ready administrability.’ [Citation.]” (Id. at pp. 609-610.) Ultimately, “[g]iven the clear meaning of ‘prevailing party’ in the fee-shifting statutes,” the court did not “determine which way these various policy arguments cut.” (Id. at p. 610.) It concluded that “the ‘catalyst theory’ is not a permissible basis for the award of attorney’s fees under” these statutes. (Ibid.)
I agree with the majority that we are not required to follow the high court’s interpretation of these federal statutes in interpreting California’s statutes. (Maj. opn., ante, at p. 568.) But federal decisions have persuasive value. “Since both this court and the Legislature have relied on federal cases in framing the private attorney general theory, we regard the federal precedent in this area as persuasive.” (Maria P. v. Riles (1987) 43 Cal.3d 1281, 1290 [240 Cal.Rptr. 872, 743 P.2d 932].) Because this court has never awarded attorney fees on a catalyst theory, but only recognized the existence of the federal rule, I see no reason suddenly to go an independent route for the first time after the federal courts have abandoned that theory.
In the companion case of Tipton-Whittingham v. City of Los Angeles (2004) 34 Cal.4th 604 [21 Cal.Rptr.3d 371] (Tipton-Whittingham), the United States Court of Appeals for the Ninth Circuit certified to this court questions similar to the one presented in this case. (See Tipton-Whittingham v. City of Los Angeles (9th Cir. 2003) 316 F.3d 1058.) In its certification order, it summarized our cases in this regard: “California cases preceding Buckhannon, while containing dicta that endorses the catalyst theory for the award of prevailing-party attorneys’ fees, have involved circumstances where there has been a judicially enforceable change in the legal relationship between the parties. See Maria P. v. Riles[, supra,] 43 Cal.3d 1281, 1290-91 . . . [240 Cal.Rptr. 872, 743 P.2d 932] (1987) (determining plaintiffs qualified as prevailing parties under [Code of Civil Procedure section] 1021.5 on the basis of their preliminary injunction against defendants); In re Head 42 Cal.3d 223, 225 [228 Cal.Rptr. 184, 721 P.2d 65] (1986) (awarding attorneys’ fees after petitioners prevailed on their habeas corpus claims); Folsom v. Butte County [Assn, of Governments], 32 Cal.3d 668, 675-76 [186 Cal.Rptr. 589, 652 P.2d 437] (1982) (awarding attorneys’ fees where the court entered partial summary judgment and an injunction against one defendant); Northington v. Davis, 23 Cal.3d 955, 960 [154 Cal.Rptr. 524, 593 P.2d 221] (1979) (upholding plaintiffs’ fee award where the trial court granted summary judgment against the defendants).” (Id. at p. 1062.)
The majority says we “endorsed” the catalyst theory in Westside Community, supra, 33 Cal.3d 348. (Maj. opn., ante, at p. 566.) But, as the Ninth Circuit recognized, any such endorsement was dictum, because we denied attorney fees in that case. (Westside Community, supra, at p. 355.) *593Moreover, our dictum did not endorse the rule so much as merely recognize what “federal decisions” (since overruled) had done. (Westside Community, supra, at p. 352.) Indeed, as I explain in my separate dissent in Tipton-Whittingham, supra, 34 Cal.4th at p. 612 (which, like Westside Community, involves a governmental entity as defendant), Westside Community reversed an award of attorney fees in part for reasons that argue against the catalyst theory as a whole, not merely its application in that case. Now that we have occasion to examine the question ourselves, we should not make the catalyst theory our own.
In Buckhannon, supra, 532 U.S. 598, the high court relied on the plain meaning of the word “prevailing” to reject the catalyst theory. Here, the language of Code of Civil Procedure section 1021.5 militates much more strongly against the catalyst theory. The federal statutes simply give trial courts discretion to allow the “prevailing party” attorney fees. (See Buckhannon, supra, at p. 601.) Code of Civil Procedure section 1021.5, however, permits an award only to a “successful” (which is synonymous with “prevailing”) party in an action “which has resulted in the enforcement of an important right affecting the public interest . . . .” (Italics added.) The italicized words mean that the plaintiffs must have compelled the defendant’s conduct to protect some “right.” (See Black’s Law Dict., supra, at p. 549 [defining “enforcement” as “[t]he act or process of compelling compliance with a law, mandate, or command”].)
But voluntary action is not compelled action. Without some judicially enforceable order, there is no way to know whether the action was voluntary or compelled. Persons and entities act voluntarily in response to a lawsuit for many reasons, some unrelated to the lawsuit’s merits: to avoid the expense of litigation or bad publicity, to foster good public relations, to make an improvement or take other useful action not required by law, perhaps simply to put the litigation behind and move on. The pressure to yield voluntarily to a lawsuit’s demands, even if not legally required, is exacerbated by the circumstance that historically attorney fee awards have not gone in both directions. Although the statutes do not prohibit awards to prevailing defendants, the private attorney general doctrine has generally resulted only in attorney fee awards to the prevailing plaintiffs and not also to the prevailing defendants. Thus, unlike the plaintiffs who can hope to be reimbursed for their attorney fees, the defendants generally cannot expect to receive compensation from the plaintiffs for their attorney fees. Those defendants who choose to fight a lawsuit lose even when they win; they must pay their attorneys themselves, which can be very expensive even for the victor. This circumstance places the defendants under great pressure to settle a lawsuit, even if unmeritorious, as soon as possible.
*594A “judicial imprimatur” (Buckhannon, supra, 532 U.S. at p. 605) on a defendant’s change in conduct is thus necessary to show that the plaintiff actually enforced a legal right. Merely eliciting a voluntary action is not enforcing a legal right. But the catalyst theory simply assumes the defendant’s action was required to right a legal wrong; it assumes the defendant had acted unlawfully. This assumption is contrary to the requirements of Code of Civil Procedure section 1021.5.
The majority, as well as plaintiffs and supporting amici curiae, argue that not adopting the catalyst theory might discourage lawsuits like this one, and lawsuits like this one are so beneficial to society that courts must not do anything that might discourage them. They- claim the catalyst theory is necessary to provide plaintiffs a full incentive to undertake the cost of public interest litigation. (E.g., maj. opn., ante, at p. 574.) I agree that the private attorney general doctrine serves a valuable purpose. (Woodland Hills Residents Assn., Inc. v. City Council (1979) 23 Cal.3d 917, 933 [154 Cal.Rptr. 503, 593 P.2d 200].) But it can also impose a substantial cost in a litigious world, especially as extended by the catalyst theory. The majority confidently asserts that the catalyst theory requires only “relatively economical, straightforward inquiries.” (Maj. opn., ante, at p. 573.) It bases this assertion partly on “our own judicial experience,” although it does not identify what that judicial experience might be. (Ibid.) Our only judicial experience with the catalyst theory consists of this case and Westside Community, supra, 33 Cal.3d 348. Our experience in this case is far from comforting and does not support the majority’s confident assertion. Here, plaintiff filed a seven-page complaint stating a single cause of action. Then, after a year of “hotly-contested discovery,” various contested hearings, and a lengthy evidentiary hearing, the trial court awarded plaintiffs $762,830 in attorney fees, about 90 percent of which was for litigating the catalyst theory. And we are not done yet, as the majority remands the case for yet more litigation. Our experience in Westside Community, supra, 33 Cal.3d 348, is also not very comforting. There we reversed a grant of attorney fees predicated on the catalyst theory in our own hotly contested four-to-three decision, which also hardly suggests the doctrine is as easy to apply as the majority asserts.
In Tipton-Whittingham, supra, 34 Cal.4th at p. 608, the majority summarizes its catalyst theory requirements: “In order to obtain attorney fees without such a judicially recognized change in the legal relationship between the parties, a plaintiff must establish that (1) the lawsuit was a catalyst motivating the defendants to provide the primary relief sought; (2) that the lawsuit had merit and achieved its catalytic effect by threat of victory, not by dint of nuisance and threat of expense . . . ; and (3) that the plaintiffs reasonably attempted to settle the litigation prior to filing the lawsuit.” These requirements can be complex, not straightforward.
*595The first of these requirements—causation—can itself be difficult to establish. The mere coincidence of lawsuit followed by action is not enough under the majority’s catalyst theory. “[I]n order to justify a fee award, there must be a causal connection between the lawsuit and the relief obtained.” (Westside Community, supra, 33 Cal.3d at p. 353 [reversing the award of attorney fees for want of causation].) This requirement generally forces an inquiry into the motivation behind the defendant’s actions, actions often undertaken by public or corporate officials. (See Folsom v. Butte County Assn, of Governments (1982) 32 Cal.3d 668, 686 [186 Cal.Rptr. 589, 652 P.2d 437] [phrasing the question as “ ‘whether or not the local politicians would have done what they have done absent the lawsuit’ ”].) The Attorney General persuasively argues that the catalyst theory should never be based on a change in legislation because of the difficulty and impropriety of delving into legislators’ subjective motivation in enacting legislation. (See County of Los Angeles v. Superior Court (1975) 13 Cal.3d 721, 726-727 & fn. 5 [119 Cal.Rptr. 631, 532 P.2d 495].) But similar concerns apply to actions of public officials in the executive branch or even corporate decision makers and other persons. “Obviously it can be difficult to prove causation where as here plaintiff seeks to recover on a catalyst theory.” (Californians for Responsible Toxics Management v. Kizer (1989) 211 Cal.App.3d 961, 968 [259 Cal.Rptr. 599].) In this case, for example, to show causation, plaintiff had to establish that DaimlerChrysler adopted its policy, announced on September 10, 1999, due to this lawsuit and not due to the ongoing efforts of the response team it had already created to address the problem or the investigations of the Santa Cruz County District Attorney and California Attorney General that had begun before the lawsuit.
The second of these requirements forces a court that has entered no judicial ruling in the plaintiff’s favor (otherwise the catalyst theory would not come into play) to make some sort of ruling regarding the merits of the underlying lawsuit. It is not clear to me exactly what the majority means in this regard, or how the trial court is supposed to go about making this determination, but here, after more than a year of litigating the catalyst theory, no court has yet made the ruling the majority demands. Future courts will have to struggle mightily to decide how to determine whether a moot lawsuit had merit when filed. Finally, the majority requires the plaintiffs to establish that they attempted to settle the litigation without a lawsuit (a requirement that, as I explain below, has long existed). This, too, is a factual question of some complexity, as today’s remand for yet more litigation demonstrates.
Thus, permitting attorney fees on a catalyst theory, with no objective manifestation, in the form of judicial action, that the lawsuit vindicated a legal right, may, as here, “ ‘result in a second major litigation.’ ” (Buckhannon, supra, 532 U.S. at p. 609.) “[T]he catalyst theory of fee recovery engenders confusion and unnecessary litigation. . . . Too frequently, *596legal battles over attorneys’ fees merely add another round of protracted litigation to what already has been protracted litigation on the merits of a claim. . . . This collateral litigation over attorneys’ fees is often more heated, more arcane, and over far higher monetary stakes than the underlying lawsuit. The relationship of all of this activity to the larger public good is becoming increasingly difficult to discern.” (S-l By and Through P-1 v. State Bd. of Educ. (4th Cir. 1993) 6 F.3d 160, 171 (dis. opn. of Wilkinson, J.).)2
I can perceive of few things less useful to society than generating great amounts of attorney fees litigating the catalyst theory. In another attorney fee case, we stated that “scarce judicial resources should not be used to try the merits of voluntarily dismissed actions merely to determine which party would or should have prevailed had the action not been dismissed.” (Santisas v. Goodin (1998) 17 Cal.4th 599, 621 [71 Cal.Rptr.2d 830, 951 P.2d 399].) In this case, scarce judicial resources should not be used to litigate the various requirements of the catalyst theory.3
The majority argues the catalyst theory is needed to eliminate risk in public interest litigation. (Maj. opn., ante, at p. 574.) But there will always be risk. Indeed, one of the requirements for any plaintiff seeking attorney fees is that the plaintiff must have attempted to settle the dispute without litigation. (Grimsley v. Board of Supervisors (1985) 169 Cal.App.3d 960, 966 [213 Cal.Rptr. 108]; see maj. opn., ante, at pp. 576-577.) Carried to its logical conclusion, however, the majority’s catalyst rationale should extend to attorney fees expended in seeking relief short of litigation. If the threat of litigation causes the prospective defendant to provide the relief sought, why should attorney fees be denied merely because no lawsuit was needed? Denying attorney fees when the desired result is obtained without a lawsuit can deter those plaintiffs who will have to expend attorney fees that they *597may never recover. Yet even the majority is forced to admit that no one can be deemed to be a successful party without a lawsuit. (Maj. opn., ante, at p. 570.)
The private attorney general doctrine inherently contains both a risk and a cost. A line must be drawn somewhere to balance this risk and this cost. I would hold that the statute here draws the necessary line by requiring some kind of a judicial imprimatur before a plaintiff can be considered to be a successful or prevailing party that enforced an important public right.
The potential for awards of this kind can also greatly increase the possibility of undue pressure to settle meritless claims. If DaimlerChrysler had simply paid the requested fees at the outset rather than litigate the question, it could have spared itself most of the award (as well as its own attorney fees, which are no doubt substantial). But surely plaintiffs’ entitlement to attorney fees was, and is, not so clear that DaimlerChrysler could not, and cannot, reasonably litigate it. The threat of a huge award of attorney fees generated while litigating the catalyst theory permits the plaintiffs to extort attorney fees from businesses no matter how weak their entitlement to them may be. With this case as a warning, future defendants may surrender to attorney fee demands, no matter how unmeritorious, rather than risk a substantial award of attorney fees down the road.
Indeed, the private attorney general doctrine, even without the catalyst theory and multipliers on fees on fees (see pt. II.C, post), gives the plaintiffs a great advantage in settlement negotiations. The defendants generally have to pay their own attorney fees. Thus, those defendants who litigate rather than sell out as cheaply as possible as soon as possible face not the risk, but the near certainty, that they will incur attorney fees they will not recover. They also risk incurring a potentially substantial award for the opponents’ attorney fees. The plaintiffs, by contrast, merely face the possibility they will not be compensated for their own attorney fees; they run little risk of having to pay their opponents’ attorney fees. And to compensate for even this possibility, the private attorney general doctrine permits courts to add a multiplier to the plaintiffs’ attorney fees, which can be very rewarding, as this case illustrates. The plaintiffs thus have relatively little incentive to settle, defendants a very strong need to settle. I see no need for the catalyst theory to provide yet more incentive to plaintiffs.
For all of these reasons, I would not adopt the catalyst theory as a basis for awarding attorney fees. I would conclude that before a party can be considered to be a successful or prevailing party under Code of Civil Procedure section 1021.5 or Government Code section 12965, subdivision (b), there must be some court-ordered change in the legal relationship between the plaintiff and the defendant in the plaintiff’s favor.
*598B. Plaintiffs have not established entitlement to attorney fees even under the majority’s catalyst theory.
Even accepting the majority’s catalyst theory, plaintiffs have failed to establish entitlement to attorney fees for several reasons.
For any plaintiff (including those who actually win their lawsuit) to receive attorney fees, the action must have “resulted in the enforcement of an important right affecting the public interest. . . .” (Code Civ. Proc., § 1021.5.) “A decision which has as its primary effect the vindication of the litigant’s personal rights is not one which brings into play the attorney fees provisions of [Code of Civil Procedure] section 1021.5.” (In re Head (1986) 42 Cal.3d 223, 228 [228 Cal.Rptr. 184, 721 P.2d 65].) Plaintiffs’ complaint was solely for breach of warranty. It sought only class certification (which plaintiffs never obtained), an award of “compensatory damages for breach of warranty,” and attorney fees. This action was, at most, a vindication of personal rights, not an important right affecting the general public.
In reaching the opposite conclusion, the trial court and the majority of this court claim that the lawsuit “implicated an issue of public safety, and that the lawsuit benefited thousands of consumers and potentially thousands more by acting as a deterrent to discourage lax responses to known safety hazards.” (Maj. opn., ante, at p. 578.) Neither the trial court nor the majority gets more specific, but they must be referring to the incorrect advertising, not any failure to fully compensate the consumers for their damages; whether the consumers were made whole does not implicate public safety. I agree there is some evidence that DaimlerChrysler’s mistake regarding the towing capacity implicated public safety at one time. (See id. at p. 561 [“The reduced towing capacity was a potential risk factor”].) I also agree that the public agency investigation revealed that brochures containing the mistake were distributed as late as August 1999. (Ibid.) But entirely missing is any relationship between public safety concerns and this lawsuit. The plaintiffs expressly alleged that in June 1999, DaimlerChrysler admitted its error in a letter sent to owners of the affected trucks. They alleged nothing regarding any continuing misrepresentations or any other public safety concerns, whether in the past or present. The only remedies the lawsuit sought were individual damages and attorney fees. No evidence whatever supports the conclusion that this lawsuit affected any public safety concerns. All that this lawsuit implicated was the truck owners’ parochial financial interests. Maximizing plaintiffs’ pecuniary gain does nothing to enhance public safety.
In trying to distinguish this lawsuit from the public agency investigation, and thus respond to DaimlerChrysler’s argument that this was an unnecessary “tagalong” lawsuit, the trial court said that the public agencies “were only *599concerned with DaimlerChrysler’s false advertising materials and never sought any remedies on behalf of the consumers who acquired these vehicles while they were being misrepresented. Private enforcement was needed.” But it was the false advertising, not plaintiffs’ ability to maximize their monetary recovery, that implicated public safety. Plaintiffs (and the majority here) cannot have it both ways. They cannot assert that this lawsuit was more than a tagalong lawsuit because the public agencies were solely interested in public safety, and then also claim that plaintiffs conferred a substantial public benefit in enhancing public safety. The public agency investigation took care of public safety. The private attorney general doctrine is not necessary when the real Attorney General was protecting the public interest.4
The trial court also said that the Santa Cruz County District Attorney and the Attorney General “had only made an inquiry and had not commenced any proceeding when plaintiffs filed this action.” But the private attorney general doctrine should not reward someone merely for winning the race to the courthouse, especially given the long-standing requirement that the plaintiff must have attempted to settle the matter before filing the lawsuit, which the public agencies were doing.
The trial court and majority also suggest the attorney fee award was appropriate because this action served as a deterrent to others who might otherwise have a lax response to safety concerns. This suggestion fails for two reasons, one legal, one factual. First, “Carried to its logical conclusion, the reasoning adopted by the trial court and espoused by plaintiff would make the private attorney general doctrine applicable in every case in which a plaintiff successfully sued a public agency [or, as here, a large business] for some wrongful conduct, because every such lawsuit would communicate a message to the losing party. Such an expansive reading of the statutory requirement is untenable.” (Flannery v. California Highway Patrol (1998) 61 Cal.App.4th 629, 636 [71 Cal.Rptr.2d 632].) Second, even if the deterrence *600rationale could be used in some cases, this is not one of those cases. The public agencies, not plaintiffs, took steps to ensure that this mistake will not recur. The public agencies, not plaintiffs, forced DaimlerChrysler to pay a $75,000 fine. For plaintiffs to seek credit for what the public agencies did proves the truth of DaimlerChrysler’s claim that, for purposes of the private attorney general doctrine, plaintiffs’ lawsuit was, indeed, merely a tagalong action.
In addition to erroneously seeking and obtaining credit for what the public agencies did, plaintiffs have failed to satisfy two other requirements: (1) they have failed to show that the lawsuit had any merit; and (2) they have failed to show that they reasonably attempted to settle the matter short of litigation.5 The majority implicitly recognizes this failure. (Maj. opn., ante, at pp. 575-576.) But, determined to reward these plaintiffs no matter how unwarranted the reward may be, it remands the matter for yet more litigation. It does so by a clever bit of judicial sleight of hand. It says that “these limitations on the catalyst theory are to some degree new.” (Maj. opn., ante, at p. 561, italics added.) Implicit is the argument that it would be unfair to deny the plaintiffs the opportunity to prove newly minted requirements.
I agree that, because the majority adopts the catalyst theory for the first time today, it has just invented some of the rules—in particular, the rule that a court that has never ruled on the merits should do so as part of the attorney fee litigation. Accordingly, to some degree, the limitations are new. But one critical requirement—that plaintiffs show the lawsuit was actually necessary—is not new. The majority tries to obfuscate this circumstance by saying the “Attorney General proposes” this rule. (Maj. opn., ante, at p. 577.) It hopes, no doubt, that the reader will infer that the Attorney General is proposing something new. But the Attorney General is not proposing something new. Rather, he is merely citing a requirement that has long existed. “[Attorney fees under Code of Civil Procedure section 1021.5, will not be awarded unless the plaintiff seeking such fees had reasonably endeavored to enforce the ‘important right affecting the public interest,’ without litigation and its attendant expense(Grimsley v. Board of Supervisors, supra, 169 Cal.App.3d at p. 966 [denying attorney fees for failure to satisfy this requirement even though the plaintiff had won a final judgment].) This language is quite clear, and it was written in 1985, long before the events of this case. Accordingly, plaintiffs have always been on notice of this requirement. I see no reason, and the majority supplies none, to permit plaintiff to relitigate this question.
*601Even in the course of the proceedings in this court, plaintiffs have not attempted to show their action had any merit. They have not shown that DaimlerChrysler was legally required to offer a full refund in addition to the steps it had already taken regarding plaintiffs, which included full disclosure, prospective correction, and offers to pay for a hitch that, so far as this lawsuit demonstrates, would have cured all harm. The majority suggests that the “precise remedy chosen” need not be legally required and hypothesizes the existence of some other remedy that plaintiffs sought and that DaimlerChrysler was legally required to provide, and for which the actual remedy of a full refund was a “compromise.” (Maj. opn., ante, at p. 576, fn. 7.) I cannot imagine what that remedy might be, and neither plaintiffs nor the majority suggests any, but I suppose plaintiffs can attempt to prove one on remand if they choose. But for the lawsuit to have any merit there must be some “primary relief sought” (Tipton-Whittingham, supra, 34 Cal.4th at p. 608) that DaimlerChrysler was required to provide. Plaintiffs will have to make this showing on remand, and the trial court will have to make this determination.
The court will also have to determine whether plaintiffs can show that they attempted to settle the matter short of litigation. Because at least waiting until DaimlerChrysler had responded to the public agencies’ inquiry before filing a complaint would have been eminently reasonable, plaintiffs will not be able to make this showing, which is no doubt why they have not yet tried to do so despite the long-standing existence of Grimsley v. Board of Supervisors, supra, 169 Cal.App.3d 960. I also hope that on remand, the court will reconsider its contradiction in (1) finding this lawsuit different from the public agency investigation and (2) predicating the actual award of attorney fees on what the public agencies had accomplished. The court should look instead to what this lawsuit accomplished, which had nothing to do with public safety.
I can only hope that future courts apply the catalyst theory with more care than the majority does its own creation.
C. Plaintiffs should not receive a multiplier for litigating fees on fees.
The majority also holds that a plaintiff may recover, as attorney fees, not only its fees incurred prosecuting the underlying litigation, with a multiplier, and its fees incurred litigating its entitlement to attorney fees (i.e., fees on fees), but also a multiplier on fees on fees. I appreciate the majority’s attempt to limit the size of such multipliers. The majority’s efforts might help reduce the instances of the tail wagging the dog like here, where the fee for litigating fees on fees is nine times greater than the fee for litigating the underlying lawsuit. But I would hold that a multiplier is never appropriate for litigating fees on fees. The majority disagrees with courts from other states that have considered this question and, tellingly, cites no out-of-state cases *602supporting its conclusion. (Maj. opn., ante, at p. 580.) If, as the majority claims, the private attorney general doctrine is intended to encourage societally useful lawsuits (like the majority finds this one to be), and not merely to swell attorneys’ coffers, permitting fees for work expended on the actual lawsuit plus a multiplier, and permitting attorneys to be paid for their efforts in obtaining those fees plus that multiplier, is a sufficient incentive. A multiplier on fees generated litigating fees, which, as here, can make the overall reward truly absurd compared to the effort regarding the underlying litigation, is not necessary.
Permitting this second multiplier further stacks settlement leverage in the plaintiffs’ favor. Not only must a defendant be concerned about paying its own attorney fees, and about having to pay for the plaintiffs’ attorney fees incurred in the underlying litigation, with a potential multiplier, and about having to pay attorney fees the plaintiff incurred in seeking fees, it must also worry about paying a multiplier on that amount. All this greatly increases the pressure on the defendants to buy their way out of lawsuits as cheaply as possible no matter how meritless they may be.
I must also comment on the irony, no doubt unintended, of the majority’s statements that a multiplier often takes into account the attorney’s “exceptional skill,” and that litigating fees on fees “is for the most part simpler than litigation on the merits.” (Maj. opn., ante, at p. 582.) Plaintiffs exhibited no exceptional skill in litigating the underlying lawsuit. Because DaimlerChrysler had long since voluntarily informed plaintiffs of its mistake, plaintiffs had to undertake little or no investigation. Plaintiffs’ attorneys merely filed a simple seven-page complaint alleging a single cause of action and containing largely boilerplate language. Ironically, these attorneys’ best lawyering came when litigating their entitlement to attorney fees, including their ability to convince the trial court both to find that their action was distinct from the public agency investigation and to credit them with what the public agencies had accomplished. Although I hesitate to suggest this lest the court on remand take, me seriously, in a perverse way, under the majority’s analysis, plaintiffs’ effort while litigating their entitlement to fees might be entitled to a larger multiplier than their effort regarding the underlying lawsuit.
Thus is the topsy-turvy world of catalyst theory and fees plus multipliers plus fees on fees plus more multipliers for fees on fees.
III. Conclusion
At a time when Californians are increasingly concerned about extortionate lawsuits against businesses, large and small, and worried that the legal climate in California is so unfriendly to businesses that many are leaving the *603state and others are deterred from coming here in the first place,6 today’s ruling goes in exactly the wrong direction. And it goes further in that direction than this court has ever gone before. We should interpret and apply California’s private attorney general statutes sensibly to encourage responsible litigation while also keeping attorney fee judgments within reasonable bounds and maintaining some semblance of balance between the litigation positions of the plaintiffs and the defendants.
Because the majority does not do so, I dissent. I would reverse the judgment of the Court of Appeal.
Baxter, J., and Brown, J., concurred.
On January 12, 2005, the opinion was modified to read as printed above.
In its entirety, Code of Civil Procedure section 1021.5 provides today: “Upon motion, a court may award attorneys’ fees to a successful party against one or more opposing parties in any action which has resulted in the enforcement of an important right affecting the public interest if: (a) a significant benefit, whether pecuniary or nonpecuniary, has been conferred on the general public or a large class of persons, (b) the necessity and financial burden of private enforcement, or of enforcement by one public entity against another public entity, are such as to make the award appropriate, and (c) such fees should not in the interest of justice be paid out of the recovery, if any. With respect to actions involving public entities, this section applies to allowances against, but not in favor of, public entities, and no claim shall be required to be filed therefor, unless one or more successful parties and one or more opposing parties are public entities, in which case no claim shall be required to be filed therefor under Part 3 (commencing with Section 900) of Division 3.6 of Title 1 of the Government Code.
“Attorneys’ fees awarded to a public entity pursuant to this section shall not be increased or decreased by a multiplier based upon extrinsic circumstances, as discussed in Serrano v. Priest [(1977)] 20 Cal.3d 25, 49 [141 Cal.Rptr. 315, 569 P.2d 1303].”
The Fourth Circuit adopted this dissenting opinion after in bank review. (S-l and S-2 v. State Bd. of Educ. of N.C. (4th Cir. 1994) 21 F.3d 49, 51 (in bank).) The high court later cited the in bank decision with approval. (Buckhannon, supra, 532 U.S. at pp. 602, 608.)
The majority suggests that Santisas v. Goodin, supra, 17 Cal.4th 599, supports adoption of the catalyst theory. (Maj. opn., ante, at pp. 571-572.) That case does not do so. It involved the interplay of several statutes and certain contractual language not relevant here. The issue was under what circumstances, if any, a defendant might be considered a prevailing party when the plaintiff voluntarily dismisses the action. When a plaintiff voluntarily dismisses an action, the court ultimately issues the order, which is a judicial action favorable to the defendant. We had to decide whether it was the type of favorable action that supported an award of attorney fees. To simplify a complex analysis (and one irrelevant here), we held that it might be so in some circumstances. We relied in part on a statute that defines “prevailing party” as including “a defendant in whose favor a dismissal is entered.” (Code Civ. Proc., § 1032, subd. (a)(4); see Santisas, supra, at p. 621.) No equivalent statute exists for a plaintiff in whose favor no order of any kind is entered.
The majority says I question the rule of Beasley v. Wells Fargo Bank (1991) 235 Cal.App.3d 1407, 1417-1418 [1 Cal.Rptr.2d 459], that Code of Civil Procedure section 1021.5 requires a finding that the lawsuit conferred a significant benefit on a substantial number of people and that the action’s subject matter implicated the public interest. (Maj. opn., ante, at p. 578, fn. 9.) I do not question that rule. Instead I question the majority’s assertion that this lawsuit implicated public safety, the only ground it provides for awarding plaintiffs attorney fees.
The majority also accuses me of “reweighing and recharacterizing]” the evidence. (Maj. opn., ante, at p. 578, fn. 9.) However, no evidence exists that this lawsuit implicated public safety that can be reweighed or recharacterized. The majority has not even attempted to identify any such evidence. It merely refers the reader to unspecified “facts reviewed in the first part of this opinion.” (Id. at p. 578.) But the majority’s factual recitation shows that the public agencies, not plaintiffs, addressed public safety concerns. (See id. at p. 562 [the “public agency investigation revealed that brochures misrepresenting the trucks’ towing capacity were still being distributed as of August 1999”].)
Indeed, as noted, the trial court awarded plaintiffs attorney fees in part because they filed their lawsuit while the public agencies were trying to settle the matter short of litigation.
On November 2, 2004, for example, the voters approved Proposition 64, which places limitations on private enforcement of California’s unfair competition law. The supporting ballot argument urged a yes vote to “protect small businesses from frivolous [shakedown] lawsuits” that “make businesses want to move to other states where lawyers don’t have a legal extortion loophole. When businesses leave, taxpayers who remain pick up the burden.” (Ballot Pamp., General Elec. (Nov. 2, 2004) argument in favor of Prop. 64, p. 40.)