Ramirez v. Charter Communications, Inc.

Facts

Charter Communications, a telecommunications employer with nearly 100,000 employees, requires job applicants to agree to its alternative dispute resolution program, Solution Channel. Prospective employees complete a computerized onboarding process and electronically sign a Mutual Arbitration Agreement, which incorporates the program’s Solution Channel Guidelines. (Ramirez v. Charter Communications, Inc., supra, 16 Cal.5th at pp. 489–490.) Charter hired Angelica Ramirez in July 2019; she accepted the Agreement through onboarding. Charter fired her in May 2020, and in July 2020 she sued for discrimination, harassment, and retaliation under FEHA, plus wrongful discharge in violation of public policy. (Id. at p. 490.)

Four features of the paperwork mattered. Section A committed both parties to arbitrate “any dispute arising out of or relating to” Ramirez’s application, employment, or termination, “except as specifically excluded below.” Sections B and C then specified covered and excluded claims: wrongful termination, discrimination, harassment, retaliation, wage and hour, leave, whistleblower, and background-check claims were directed into arbitration, while claims for workers’ compensation and unemployment benefits, ERISA and NLRA matters, claims under severance or noncompete agreements, injunctive relief “related to unfair competition and the taking, use or unauthorized disclosure of trade secrets or confidential information,” theft or embezzlement, previously adjudicated claims, and intellectual property claims were excluded. Charter’s own Guidelines labeled eight covered categories “employee claims” and three “Charter claims.” (Ramirez, supra, 16 Cal.5th at pp. 491, 496–497 & fn. 2.) Section E required any claim that must first go to an administrative agency to be filed with Solution Channel “within the time period by which the charge, complaint or other similar document would have had to be filed with the agency” — for a 2019 FEHA claim, one year, though the full period to reach court could have run three years. (Id. at pp. 500–501 & fn. 4.) The Guidelines allowed 90 days for discovery, with each side “permitted to take up to four (4) depositions,” 20 interrogatories, and 15 document requests, and provided that “[a]ny disagreements regarding the exchange of information or depositions will be resolved by the arbitrator to allow a full and equal opportunity to all parties to present evidence.” (Id. at pp. 503–504.) And section K, after requiring Charter to pay arbitral fees, provided that a party resisting arbitration “will be required to pay to the other party all costs, fees and expenses that they incur in compelling arbitration, including, without limitation, reasonable attorneys’ fees.” (Id. at pp. 491, 507.)

Procedural history

Charter moved to compel arbitration and sought its fees for the motion. The trial court found the Agreement adhesive and held it substantively unconscionable in three respects: the shortened filing period; a fee clause that violated FEHA by failing to limit Charter’s recovery to frivolous or bad-faith claims; and the interim award of fees to a party that successfully compels arbitration. It rejected challenges to the discovery limits and the covered-claims structure. Finding the Agreement “permeated with unconscionability,” it refused enforcement and denied the motion. (Ramirez, supra, 16 Cal.5th at p. 490.)

The Court of Appeal affirmed, but went further: it also found the covered/excluded claims structure non-mutual and the four-deposition limit unconscionable as applied to Ramirez, who had estimated — without dispute from Charter — that she needed at least seven depositions. It disagreed with Patterson v. Superior Court (2021) 70 Cal.App.5th 473, which had construed the same Charter fee clause to impliedly incorporate FEHA’s asymmetric fee rule. (Ramirez, supra, 16 Cal.5th at pp. 490–491, 505, 509–510.) The Supreme Court granted review to resolve the Patterson conflict, to review the mutuality, filing-deadline, and discovery rulings, and to decide whether the refusal to sever violated the Federal Arbitration Act. (Id. at p. 491.) It reversed and remanded. (Id. at p. 519.)

Issue

Four questions: (1) Are the Agreement’s covered/excluded claims provisions, filing deadline, discovery limits, and compel-arbitration fee clause substantively unconscionable? (2) In assessing a discovery clause, may a court rely on a particular plaintiff’s post-contracting litigation needs? (3) When a contract contains multiple unconscionable terms, must — or may — a court refuse enforcement outright rather than sever? (4) Does refusing to sever and enforce violate the FAA? (See Ramirez, supra, 16 Cal.5th at pp. 489, 491.)

Holding

Reasoning

The framework: adhesion plus a sliding scale, measured at formation. The court restated settled doctrine. A contract is unconscionable if one party “lacked a meaningful choice” and the terms are “unreasonably favorable to the other party”; procedural and substantive elements must both be present, on a sliding scale. (Ramirez, supra, 16 Cal.5th at pp. 492–493, quoting OTO, L.L.C. v. Kho (2019) 8 Cal.5th 111, 125, and citing Armendariz v. Foundation Health Psychcare Services, Inc. (2000) 24 Cal.4th 83, 114.) Adhesion alone establishes procedural unconscionability, though “generally … only a low degree”; still, “the potential for overreaching in the employment context warrants close scrutiny of the contract’s terms.” (Ramirez, at p. 494.) Critically, “whether a contract is fair or works unconscionable hardship is determined with reference to the time when the contract was made and cannot be resolved by hindsight.” (Id. at p. 493, quoting Yeng Sue Chow v. Levi Strauss & Co. (1975) 49 Cal.App.3d 315, 325.) That formation-time principle became the fulcrum of the discovery holding.

Non-mutual coverage: one-sidedness plus no timely justification. Following Armendariz’s “modicum of bilaterality” requirement, the court held the coverage structure unconscionable because the Agreement “tends to exempt claims likely to be made by Charter while directing Ramirez’s likely claims into arbitration.” (Ramirez, supra, 16 Cal.5th at p. 498; see id. at pp. 495–496, 498, citing Armendariz, supra, 24 Cal.4th at pp. 117–118, and Mercuro v. Superior Court (2002) 96 Cal.App.4th 167, 176.) The exclusions for intellectual property, severance and noncompete disputes, and trade secret injunctions favored the employer; several “employee” exclusions — workers’ compensation, unemployment, time-barred claims — were “illusory” because those claims are excluded from arbitration or from court by law. (Ramirez, at p. 498.) A “modicum of bilaterality will not save a clause that is, in practical effect, unjustifiably one sided.” (Id. at p. 499, quoting Cook v. University of Southern California (2024) 102 Cal.App.5th 312, 327.) The trade secret carve-out also “exceeds the protections” of Code of Civil Procedure section 1281.8, so Baltazar’s safe harbor for clauses that merely restate statutory rights did not apply. (Ramirez, at p. 499, citing Carbajal v. CWPSC, Inc. (2016) 245 Cal.App.4th 227, 249–250.) Armendariz makes unconscionability turn on “both a one-sided result and the absence of justification for it” (Ramirez, at p. 496) — and Charter offered its “business realities” justification (a nationwide agreement requiring quick judicial relief in some states) for the first time at oral argument. The court declined to consider it: “In the absence of a cognizable justification, properly asserted, we ‘must assume’ the Agreement’s lack of mutuality is unconscionable.” (Id. at p. 500, quoting Armendariz, supra, 24 Cal.4th at p. 120.)

The filing deadline: contracting around FEHA’s limitations scheme is unconscionable. Parties may shorten limitations periods, but the shortened period must be reasonable. (Ramirez, supra, 16 Cal.5th at p. 501, citing Ellis v. U.S. Security Associates (2014) 224 Cal.App.4th 1213, 1222.) Tracking Baxter v. Genworth North America Corp. (2017) 16 Cal.App.5th 713 — “[t]he relevant circumstances here are identical to those in Baxter” — the court held that requiring a FEHA claim to be arbitrated within the one-year administrative filing window “truncates the period the Legislature has determined employees need” and “potentially deprives Charter employees of meaningful DFEH participation.” (Ramirez, at p. 502.) Charter’s saving construction failed because the clause is “plain and unambiguous,” and courts may not “rewrite or alter by construction” unambiguous terms. (Id. at pp. 502–503, quoting Crow v. P.E.G. Construction Co., Inc. (1957) 156 Cal.App.2d 271, 278.) Nor did it matter that Ramirez herself had waived a DFEH investigation and obtained an immediate right-to-sue letter: unconscionability is judged at formation, and FEHA’s protections “are for the benefit of the entire public,” incapable of abrogation by private agreement under Civil Code section 3513. (Ramirez, at p. 503, quoting Wherry v. Award, Inc. (2011) 192 Cal.App.4th 1242, 1249.)

Discovery limits: valid — and the as-applied line of cases is disapproved. This is the opinion’s principal pro-employer move. Armendariz requires that mandatory employment arbitration permit “discovery sufficient to adequately arbitrate” statutory claims, with adequacy “determined by the arbitrator(s).” (Ramirez, supra, 16 Cal.5th at p. 505, quoting Armendariz, supra, 24 Cal.4th at p. 106.) The Court of Appeal had faulted the four-deposition cap because this plaintiff needed seven — an approach the Supreme Court identified in Baxter, Davis v. Kozak, De Leon, and similar cases, and flatly rejected: “This approach … clearly looks to postcontract formation circumstances. We disapprove this line of reasoning.” (Ramirez, at pp. 505–506.) The proper assessment “should focus on general factors that can be examined without relying on subsequent developments”: the claims covered, the amount of discovery allowed, the gap from conventional litigation, asymmetries between the parties, and “the arbitrator’s authority to order additional discovery.” (Id. at p. 506.) The second flaw was interpretive: empowering the arbitrator to resolve “all discovery disputes” so as to allow “a full and equal opportunity” to present material evidence is best read to authorize expanded discovery; at minimum the clause is ambiguous, and courts must adopt the validating construction. (Id. at pp. 506–507, citing Civ. Code, §§ 1643, 3541.) So construed, the provision “eliminates any unconscionability.” (Id. at p. 507.)

Section K: an unqualified fee shift for compelling arbitration violates FEHA, and Patterson cannot rescue it. FEHA permits a prevailing defendant to recover fees and costs only on a finding that the action was “frivolous, unreasonable, or groundless” (Gov. Code, § 12965, subd. (c)(6)), and Armendariz forbids mandatory arbitration from imposing on employees “any type of expense that the employee would not be required to bear” in court. (Ramirez, supra, 16 Cal.5th at p. 508, quoting Armendariz, supra, 24 Cal.4th at pp. 110–111.) Section K “unambiguously requires an award of attorney fees” to the party that compels arbitration, with no frivolousness finding — and could even be invoked against an employee who successfully proved parts of the agreement unconscionable but saw them severed, a possibility that “could chill an employee’s right to challenge the enforceability of an arbitration agreement.” (Ramirez, at pp. 508–509.) Patterson’s implied-incorporation construction was rejected root and branch: the validating-construction canons apply only to genuinely ambiguous terms; Armendariz filled a contractual silence rather than rewriting unambiguous language; and the policy favoring arbitration makes agreements “‘as enforceable as other contracts, … not more so.’” (Id. at p. 512, quoting Morgan v. Sundance, Inc. (2022) 596 U.S. 411, 418; see id. at pp. 511–512.) “Section K is unconscionable because it unambiguously violates FEHA. The policy favoring arbitration cannot save it.” (Ramirez, at p. 512.)

Severance: a qualitative, not arithmetic, inquiry. Civil Code section 1670.5 gives courts three options — refuse enforcement, sever, or limit — and the “strong legislative and judicial preference is to sever the offending term and enforce the balance of the agreement.” (Ramirez, supra, 16 Cal.5th at p. 513, quoting Roman v. Superior Court (2009) 172 Cal.App.4th 1462, 1477.) Both lower courts had leaned on the number of unconscionable provisions. Surveying a split — Carmona, Ontiveros, and Murphy treating the question quantitatively; Lange and Bolter rejecting any per se rule — the court sided with the latter: “no bright-line rule requires a court to refuse enforcement if a contract has more than one unconscionable term,” nor must a single term always be severed; “the appropriate inquiry is qualitative and accounts for each factor Armendariz identified.” (Id. at pp. 515–516.) The sequence: ask whether “the central purpose of the contract is tainted with illegality”; if not, whether the defects can be cured by severance or restriction, or only by impermissible “reformation by augmentation”; and whether severance serves “the interests of justice” — including whether it would condone an illegal scheme and whether the defects show “a systematic effort to impose arbitration on the weaker party not simply as an alternative to litigation, but to secure a forum that works to the stronger party’s advantage,” with deterrence a legitimate consideration under Mills v. Facility Solutions Group, Inc. (2022) 84 Cal.App.5th 1035, 1045. (Ramirez, at pp. 516–517.) A severance clause like Charter’s section Q “should [be] take[n] … into account as an expression of the parties’ intent,” though it cannot divest the court’s discretion. (Id. at p. 517.) Because the Court of Appeal had called the deposition cap the one non-severable defect (now held valid), never discussed section Q, and focused on counting, the judgment was reversed for a fresh severance analysis: a court “cannot refuse to enforce an agreement simply by finding that two or more collateral provisions are unconscionable as written and eschewing any further inquiry.” (Id. at pp. 517–518.)

No FAA problem. Applying generally applicable unconscionability law, and retaining severance as a discretionary option, places arbitration agreements on the same footing as other contracts; the FAA demands no more. (Ramirez, supra, 16 Cal.5th at pp. 518–519, citing McGill v. Citibank, N.A. (2017) 2 Cal.5th 945, 961–962.)

Significance

Ramirez is now the controlling California authority on the remedy side of unconscionability doctrine, and it deliberately pulls in both directions. For employees, it entrenches three substantive rules with bite: coverage clauses that channel employee claims into arbitration while carving out employer claims are presumptively unconscionable absent a justification the employer must raise on time; filing deadlines keyed to the administrative-charge window are invalid for FEHA claims under the Baxter/Ellis line, now given Supreme Court endorsement; and any fee clause that lets an employer recover fees without FEHA’s frivolousness finding is void as written — Patterson’s judicial rewrite is disapproved. (Ramirez, supra, 16 Cal.5th at pp. 498–503, 507–512.) For employers, the decision is equally consequential: discovery caps survive so long as the arbitrator can expand discovery to meet Armendariz’s adequacy floor, and the whole family of as-applied, this-plaintiff-needs-more-depositions attacks — Baxter, Davis, De Leon, Ontiveros, Sanchez v. Carmax, Torrecillas on this point — is disapproved. (Id. at pp. 505–507.)

The severance holding recalibrates two decades of practice under Armendariz, whose “multiple defects” language many trial courts had applied as a near-automatic rule that two or more bad terms meant no enforcement. After Ramirez, the count still matters — “the greater the number of unconscionable provisions … the less likely it is that severance will be the appropriate remedy” (Ramirez, supra, 16 Cal.5th at p. 517) — but it is one factor in a qualitative inquiry that asks whether the drafter engaged in a systematic effort to stack the forum, whether severance would condone the scheme, and whether cure requires forbidden augmentation. The remand itself illustrates the shift: even with three unconscionable terms, refusal to enforce must be earned by analysis, not assumed from arithmetic. Plaintiff-side counsel should also note the court’s reliance on Mills’s deterrence rationale, which preserves the argument that severing everything would invite employers to overreach and litigate backward. Our companion analyses of Armendariz, OTO, L.L.C. v. Kho, and Civil Code section 1670.5 trace the doctrine Ramirez now governs.

Key quotes

“Here, we clarify that no bright-line rule requires a court to refuse enforcement if a contract has more than one unconscionable term. Likewise, a court is not required to sever or restrict an unconscionable term if an agreement has only a single such term. Instead, the appropriate inquiry is qualitative and accounts for each factor Armendariz identified.” (Ramirez, supra, 16 Cal.5th at p. 516.)

“Section K is unconscionable because it unambiguously violates FEHA. The policy favoring arbitration cannot save it.” (Ramirez, supra, 16 Cal.5th at p. 512.)

“The assessment of whether a discovery clause is unconscionable should focus on general factors that can be examined without relying on subsequent developments.” (Ramirez, supra, 16 Cal.5th at p. 506.)

Read the full opinion (Justia) →

Practice pointer

Open questions

The court left the endgame unresolved: whether this agreement should be severed and enforced was remanded, with the Court of Appeal “free on remand to entertain supplemental briefing.” (Ramirez, supra, 16 Cal.5th at pp. 512–513, fn. 13, 518.) It expressly took “no view” on the validity of several provisions never challenged by Ramirez but raised by an amicus — the each-party-bears-its-own-fees sentence of section K that troubled the Court of Appeal, and sections D (filing capacity), L (jury waiver), and Q (severance). (Id. at pp. 512–513, fn. 13.) The court also disclaimed any exhaustive test for discovery clauses: arbitrator authority to expand discovery is “one way the adequacy concern can be addressed,” and “[w]e do not foreclose other formulations.” (Id. at p. 506.) How the qualitative severance inquiry cashes out in practice — in particular, when multiple defects show a “systematic effort” justifying outright refusal — will be worked out case by case; the first data point is the Second District’s 2025 decision on remand in this very case, which practitioners should read alongside this opinion. And because the court decided the fee issue under FEHA’s asymmetric rule, the enforceability of compel-arbitration fee clauses attached to non-FEHA statutory claims with different fee regimes remains open.