FEHA Fee Haircuts and the Scrutiny Split

The problem: the across-the-board cut

California fee law starts from two settled propositions. First, the lodestar governs: “a trial court must calculate a lodestar figure, which is defined as the time spent representing the party on the issues on which it prevailed multiplied by a ‘reasonable hourly compensation’ for that time,” and the court “must carefully review attorney documentation of hours expended” to avoid “‘padding.’” (Cash, supra, 111 Cal.App.5th at p. 747, rev. granted, citing Ketchum v. Moses (2001) 24 Cal.4th 1122, 1131–1132 and Chavez v. City of Los Angeles (2010) 47 Cal.4th 970, 985.) Second, review is deferential: fee awards are reviewed for abuse of discretion, “based on the notion that ‘[t]he “experienced trial judge is the best judge of the value of professional services rendered in [their] court.”’” (Id. at p. 747, quoting Serrano v. Priest (1977) 20 Cal.3d 25, 49.)

The friction point is a device trial courts reach for when a fee application is voluminous: instead of ruling entry by entry, the court applies an across-the-board percentage reduction — functionally a negative multiplier — to the hours or to the lodestar itself. The device is legitimate in every camp; what divides the courts is what the trial judge must say to justify it. One line of authority would “infer findings and defer to a trial court’s ‘general observation that an attorney overlitigated a case.’” (Cash, supra, 111 Cal.App.5th at pp. 743–744, rev. granted, quoting Karton v. Ari Design & Construction, Inc. (2021) 61 Cal.App.5th 734, 744; see California Common Cause v. Duffy (1987) 200 Cal.App.3d 730, 754–755.) The other demands that the court “articulate ‘case-specific reasons for [any] percentage reduction,’ including a ‘clear[]’ ‘expla[nation of] its reasons for choosing the particular negative multiplier [or percentage] that it chose.’” (Cash, at p. 744, quoting Warren v. Kia Motors America, Inc. (2018) 30 Cal.App.5th 24, 41, 37.) The stakes are not academic: in the four decisions that frame the split, the contested cuts ran from 20 percent to two-thirds of the fees a statutory fee-shifting provision was enacted to guarantee.

Warren and Morris: the split opens in lemon-law territory

Warren: the clear-explanation rule arrives. In Warren v. Kia Motors America, Inc., supra, 30 Cal.App.5th 24, a jury awarded the plaintiff $17,455.57 under the Song-Beverly Act; her three law firms sought a $351,055.26 lodestar, and the trial court — troubled by the “disconnect” between verdict and request — awarded 33 percent of the lodestar, $115,848.24. (Id. at pp. 28, 33–34.) Division Two of the Fourth District reversed. Tying a fee award to a proportion of the damages was error: “when a trial court applies a substantial negative multiplier to a presumptively accurate lodestar attorney fee amount, the court must clearly explain its case-specific reasons for the percentage reduction,” and “[i]f, as occurred here, the reasons for the reduction include tying the fee award to some proportion of the buyer’s damages recovery, the court abuses its discretion.” (Id. at p. 37.) The court was explicit about its method: “We have effectively applied ‘heightened scrutiny’ to the court’s selection of the 33 percent negative multiplier … . This was appropriate.” (Id. at p. 41.) Across-the-board cuts remain available for voluminous applications, “[b]ut the court must clearly explain its reasons for choosing the particular negative multiplier that it chose; otherwise, the reviewing court is unable to determine that the court had valid, specific reasons for its across-the-board percentage reduction.” (Ibid., citing Kerkeles v. City of San Jose (2015) 243 Cal.App.4th 88, 102–104 [“‘We can’t defer to reasoning that we can’t review … .’”].) The genealogy matters to everything that followed: Kerkeles was a federal civil rights fee case under 42 U.S.C. § 1988, and Warren reasoned that “[i]n consumer law cases, as in civil rights cases,” the same safeguard should apply. (Id. at p. 41.)

Morris: the first rejection. A year later, Division Seven of the Second District affirmed a fee order that had cut a requested $127,792.50 Song-Beverly lodestar to $73,864 — disallowing all 83.5 hours billed by six of its eleven attorneys and reducing the attorneys’ hourly rates (id. at pp. 28, 31–32) — and, in a footnote that became the other pole of the split, declined to follow Warren: “We disagree with the court in Warren that such a heightened standard is appropriate for appellate review of fee awards under the Song-Beverly Act.” (Morris v. Hyundai Motor America (2019) 41 Cal.App.5th 24, 37, fn. 6, review den. Nov. 13, 2019.) Kerkeles, the Morris court explained, applied a “more stringent federal standard” peculiar to section 1988; California instead presumes the award correct and infers from a substantial reduction that the court found the request inflated. (Ibid.) Morris also supplied the defense bar’s favorite evidentiary rule in this area: where the final written order states valid reasons, a reviewing court “will not speculate, based on a stray remark the court made at the hearing, that it had other, prohibited reasons that would require reversal” — even a remark musing that a $192,000 request in an $85,000 settlement was “a little much.” (Id. at pp. 35, 37–38.)

Snoeck: the clear-explanation rule comes to FEHA

Snoeck v. ExakTime Innovations, Inc., supra, 96 Cal.App.5th 908 moved the fight into employment law. Snoeck won one of six claims — failure to engage in the interactive process — and $130,088; his counsel sought a $1,193,870 lodestar with a 1.75 multiplier. (Id. at pp. 910–913.) The trial court found the rates reasonable, applied a 20 percent across-the-board cut to the hours for overstaffing and duplicative billing, added a 1.2 positive multiplier for the contingency and delay, and then — the move that made the case famous — applied a 0.4 negative multiplier, a $457,863 reduction, “to account for [p]laintiff’s counsel’s … lack of civility throughout the entire course of this litigation.” (Id. at pp. 910–911, 914–915.) Division Three affirmed, holding that civility is an aspect of the “skill” a court may weigh under Ketchum: “Civility is not just a moral good. ‘Attorney skill is a traditional touchstone for deciding whether to adjust a lodestar. [Citation.] Civility is an aspect of skill.’” (Id. at p. 915, quoting Karton, supra, 61 Cal.App.5th at p. 747.)

For present purposes, what matters is the standard Snoeck applied on the way to affirming. It quoted Warren’s rule verbatim — “the court must clearly explain its reasons for choosing the particular negative multiplier that it chose” (Snoeck, supra, 96 Cal.App.5th at p. 921, quoting Warren, supra, 30 Cal.App.5th at p. 41) — and it upheld the 40 percent incivility haircut precisely because the trial court had satisfied that obligation: “when a trial court applies a substantial negative multiplier to a presumptively accurate lodestar attorney fee amount, the court must clearly explain its case-specific reasons for the percentage reduction. … That way, a reviewing court can determine if the trial court reduced the fee award for valid reasons. … The trial court did so here.” (Id. at p. 928, quoting Warren, at p. 37.) Snoeck thus stands on the Warren side of the methodological line even though the plaintiff lost the appeal — a FEHA decision demanding, and finding, a reasoned explanation for the particular percentage chosen. Review was denied January 24, 2024.

Cash: the split ripens — and review is granted

Inside Cash: the majority’s three moves and Baker’s answer

Move one: the genealogy critique. The Cash majority traces heightened scrutiny to federal section 1988 doctrine — Kerkeles quoting Moreno v. City of Sacramento (9th Cir. 2008) 534 F.3d 1106 — and objects that “Warren decoupled Kerkeles from its moorings” by transplanting the safeguard into California consumer law. (Cash, supra, 111 Cal.App.5th at pp. 747–748, rev. granted.) The federal rationale — erring “on the side of overinclusive fee awards to incentivize lawyers to litigate federal civil rights cases” — “does not justify the spread of heightened scrutiny to every fee award for every employment, consumer protection, or other civil case arising in California.” (Id. at p. 748.)

Move two: the workability objection. Faithfully applied, the majority argues, a duty to justify “the particular negative multiplier” would “all but eliminate any across-the-board percentage reductions because trial courts would be hard pressed to justify a ‘particular’ percentage — why 30 percent instead of 29 or 31 percent?” (Cash, supra, 111 Cal.App.5th at p. 748, rev. granted.)

Move three: the deference principle. The bottom line is institutional: “heightened scrutiny is fundamentally inconsistent with the deference that California courts have granted to trial courts as the percipient witnesses to the quality of representation and hence the amount of fees that representation justifies.” (Cash, supra, 111 Cal.App.5th at p. 748, rev. granted.) On that standard the affirmance was easy: padding and duplication are justifiable reasons, and “the unusually high number of billed hours … provided substantial evidence of such excess.” (Id. at p. 749.) Notably, the majority added that “[s]pecific findings regarding a lodestar reduction are not required, at least where they were never requested.” (Ibid., italics added — a practice cue discussed below.)

Baker, J., concurring and dissenting: the meat-cleaver principle. Justice Baker agreed the case was “a good candidate for a grant of Supreme Court review” and reframed what Warren and Snoeck actually hold. (Cash, supra, 111 Cal.App.5th at p. 751, rev. granted (conc. & dis. opn. of Baker, J.).) The “heightened scrutiny” label, he wrote, misleads; the “core insight” is “easily explained and quite sensible: if a court is going to take a meat cleaver rather than a slicing knife to an attorney fees request in a case where the Legislature has authorized recovery of fees for good public policy reasons, the court should have a correspondingly greater obligation to justify its choice of that blunt tool and thereby enable meaningful appellate review of its award.” (Id. at p. 753.) On his reading, Warren and Snoeck import only “the common sense logic found in federal authority and Kerkeles (to wit, strong medicine requires stronger justification) not federal substantive law foreign to California jurisprudence.” (Id. at p. 755.) And the Cash record showed why the safeguard matters: the County never tethered its proposed 30 percent to any estimate of reasonable time — it appears to have picked the number “simply because the court in Morris … opined a 30 percent across-the-board reduction would have been proper in that case” — and the trial court adopted it wholesale in a four-sentence tentative (id. at p. 752), a cut equal to more than 407 hours, “the equivalent of wiping out the entirety of the amount billed for the 20-day trial in this case still with over 100 cut hours left to spare,” justified concretely only by witness-examination habits worth “maybe 15 hours at most if you really push the limits of imagination.” (Id. at pp. 755–756.) His warning is the dissent’s headline: if a bare incantation that counsel “overlitigated” suffices, “we will have reduced appellate review to abdication.” (Id. at pp. 756–757.)

The cross-statute seam: discretionary FEHA fees, mandatory Song-Beverly fees

A structural wrinkle runs beneath the split, and it should shape how the Supreme Court’s eventual opinion is read. The four cases arise under two very different fee statutes. FEHA fees are discretionary: “the court, in its discretion, may award to the prevailing party … reasonable attorney’s fees.” (Gov. Code, § 12965, subd. (c)(6), quoted in Cash, supra, 111 Cal.App.5th at pp. 746–747, rev. granted.) Song-Beverly fees are mandatory: a prevailing buyer “shall be allowed by the court to recover” fees “based on actual time expended,” reasonably incurred. (Civ. Code, § 1794, subd. (d), quoted in Warren, supra, 30 Cal.App.5th at p. 35.) Warren and Morris are Song-Beverly cases; Snoeck and Cash are FEHA cases; the four opinions cite one another across the statutory boundary as if the review standard were a single trans-substantive rule.

Two observations follow. First, the Cash majority’s genealogy critique — heightened scrutiny belongs to federal civil rights policy — sits awkwardly in a FEHA case, because FEHA is California’s civil rights statute, with its own version of the incentive rationale: fee awards “promote the important public policy in favor of eliminating discrimination in the workplace,” and a prevailing FEHA plaintiff “should ordinarily recover an attorney’s fee unless special circumstances would render such an award unjust.” (Snoeck, supra, 96 Cal.App.5th at p. 920, internal quotation marks omitted, quoting Vines v. O’Reilly Auto Enterprises, LLC (2022) 74 Cal.App.5th 174, 182.) Whatever force the anti-importation argument has in lemon-law cases, the Moreno logic was never foreign to FEHA. Second, even the deferential camp concedes outer limits: Cash itself distinguishes — rather than disapproves — Mountjoy v. Bank of America, N.A. (2016) 245 Cal.App.4th 266, 280–281 and Vines, supra, 74 Cal.App.5th at pp. 185–186, where across-the-board cuts were reversed because the percentages were “arbitrarily keyed to another metric unrelated to the reasonableness of the fees.” (Cash, supra, 111 Cal.App.5th at p. 749, rev. granted.) So the real doctrinal distance between the camps is narrower than the labels suggest — deference with an arbitrariness backstop, versus explanation calibrated to the size of the cut — which is exactly why Justice Baker’s formulation, closer review “designed to ensure there exists a roughly reasonable fit between the magnitude of a trial court’s concerns with fees billed and the choice of an across-the-board reduction remedy” (id. at p. 755, fn. 4 (conc. & dis. opn. of Baker, J.)), may prove the most cite-able synthesis on offer.

Key quotes

“[W]hen a trial court applies a substantial negative multiplier to a presumptively accurate lodestar attorney fee amount, the court must clearly explain its case-specific reasons for the percentage reduction.” (Warren, supra, 30 Cal.App.5th at p. 37; accord, Snoeck, supra, 96 Cal.App.5th at p. 928.)

“As between Warren and Snoeck on the one hand, and Morris on the other, we agree with Morris. … [H]eightened scrutiny is fundamentally inconsistent with the deference that California courts have granted to trial courts as the percipient witnesses to the quality of representation and hence the amount of fees that representation justifies.” (Cash, supra, 111 Cal.App.5th at p. 748, rev. granted Aug. 20, 2025, S291827.)

“[I]f a court is going to take a meat cleaver rather than a slicing knife to an attorney fees request in a case where the Legislature has authorized recovery of fees for good public policy reasons, the court should have a correspondingly greater obligation to justify its choice of that blunt tool and thereby enable meaningful appellate review of its award.” (Cash, supra, 111 Cal.App.5th at p. 753, rev. granted Aug. 20, 2025, S291827 (conc. & dis. opn. of Baker, J.).)

Read the Cash opinion (CourtListener) →

Practice pointer

Open questions

Start with what the Supreme Court has agreed to decide. Per the Court’s Issues Pending compilation, the questions presented in Cash are: “(1) Did the trial court’s across-the-board reduction of the fees requested by plaintiff’s counsel trigger heightened scrutiny of its fee order on appeal? (2) Did the trial court commit reversible error in reducing the fee request on an across-the-board basis?” How broadly the answer sweeps is the first open question: the court could resolve it for all statutory fee awards, or narrowly — for substantial across-the-board reductions only, per Justice Baker’s formulation limited to fee requests “supported by detailed billing statements unchallenged as to form” (Cash, supra, 111 Cal.App.5th at p. 753, fn. 1, rev. granted (conc. & dis. opn. of Baker, J.)). Second, the cross-statute seam: does the answer differ where the Legislature made fees mandatory (Civ. Code, § 1794, subd. (d)) rather than discretionary (Gov. Code, § 12965, subd. (c)(6)) — and if explanation duties exist anywhere, should they be at their strongest under a civil rights statute whose fee provision exists to attract counsel? Third, calibration: if clear explanation is required, how clear — Warren demands reasons for “the particular negative multiplier,” Cash mocks that demand as an impossible 29-versus-31-percent exercise, and Baker’s “roughly reasonable fit” (id. at p. 755, fn. 4) splits the difference without yet telling trial judges what to write. Fourth, whatever the answer, its application to fees-on-fees and implied findings (the Cash majority inferred a purposeful denial of a supplemental request from a general padding finding (id. at p. 750)) will determine how much of the ruling can be enforced in practice. This analysis will be updated when the Supreme Court rules.