Rule 23(c)(1)(A) Governs Pleading-Stage Motions to Deny Class Certification; Pre-Discovery Denial Allowed Only for Facial Rule 23 Failure

1. Introduction

Nine minority applicants for residential mortgage products sued Navy Federal Credit Union, alleging systematic discrimination in mortgage lending. The plaintiffs (eight Black and one Latino) sought to proceed on behalf of a broad putative class of “All minority residential loan applicants from 2018 through the present” whose applications were denied, approved on worse terms, or processed more slowly than similarly situated non-minority applicants.

The key procedural issue was whether, and under what standards, a district court may deny class certification and strike class allegations before any discovery—i.e., at the pleading stage—when the defendant argues the proposed class cannot satisfy Rule 23. Substantively, the class theory hinged on the allegation that Navy Federal used a proprietary (semi-)automated underwriting algorithm applied across products, producing discriminatory disparate impacts.

The Fourth Circuit’s decision addresses (i) the proper rule-based source of authority for early class-certification denials, (ii) the correct standard of review and decision framework when such a request is made before discovery, and (iii) the differing fit of the complaint’s allegations with Rule 23(b)(3) (damages) versus Rule 23(b)(2) (injunctive/declaratory relief).

2. Summary of the Opinion

Majority (Heytens, joined by Moon)

• Authority: The mandatory certification determination—grant or denial—derives from Rule 23(c)(1)(A), not Rule 12(f) and not solely Rule 23(d)(1)(D).
• Procedure: A defendant should file a “motion to deny class certification under 23(c)(1)(A), coupled with a motion to strike under 23(d)(1)(D) should the motion to deny class certification be granted.”
• Standard at the pleading stage: Reaffirming Goodman v. Schlesinger, a court may deny certification pre-discovery only if the complaint’s class allegations fail Rule 23(a) and (b) as a matter of law, evaluating only the face of the complaint and whether it makes a prima facie showing.
• Application:
  • Rule 23(b)(3) damages class: Affirmed denial/striking because predominance and/or superiority problems were “readily apparent” from the complaint given multi-state, multi-product, multi-outcome variability and individualized damages issues.
  • Rule 23(b)(2) injunctive/declaratory class: Vacated denial as premature because the complaint plausibly alleged a single underwriting algorithm that could generate common answers (commonality) and thus could support (b)(2) certification after discovery.

Separate Opinion (Richardson, concurring in the judgment in part and dissenting in part)

Judge Richardson would have affirmed the striking of all class allegations (both (b)(3) and (b)(2)). He emphasized district courts’ case-management discretion under Rule 23(d)(1)(D) and argued the pleadings (including incorporated reports) revealed fatal commonality defects: differences across mortgage products, minority groups, and alleged adverse outcomes.

3. Analysis

3.1 Precedents Cited

The Fourth Circuit’s framework is built primarily on older circuit precedent and modern Supreme Court class-action doctrine, woven together into a pleading-stage standard. Key authorities include:

• Goodman v. Schlesinger, 584 F.2d 1325 (4th Cir. 1978)
  The cornerstone of the majority’s holding. Goodman reversed a pre-discovery denial of class treatment as “premature,” stating class allegations should not be dismissed at the pleading stage unless they show “non-compliance with FRCP 23 as a matter of law.” The majority treats Goodman as continuing circuit law that “narrows” district-court discretion at the pleading stage: no weighing of facts, no “rigorous analysis” on a thin record, and no early denial unless facially inevitable.
• International Woodworkers of Am. v. Chesapeake Bay Plywood Corp., 659 F.2d 1259 (4th Cir. 1981)
  Reinforces Goodman’s caution that certification issues should “seldom” be resolved prior to discovery, underscoring a default preference for developing an evidentiary record—while preserving the possibility of rare facial denials.
• General Tel. Co. of Sw. v. Falcon, 457 U.S. 147 (1981)
  Cited for the proposition that courts may sometimes decide certification based on pleadings, and for the conceptual baseline that class treatment requires more than generalized allegations. Both the majority and dissent use Falcon as a touchstone on the representational risks of class litigation.
• Wal‑Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011)
  Central to the commonality analysis. The majority relies on Wal‑Mart’s formulation: commonality requires common questions capable of classwide resolution producing “common answers” in “one stroke.” Importantly, the majority distinguishes this case from “30,000 foot” commonality defects (invoking Stafford v. Bojangles' Rests., Inc.) because plaintiffs alleged a single algorithmic decision system rather than a vague “systemic” failure.
• Microsoft Corp. v. Baker, 582 U.S. 23 (2017)
  Used to confirm that orders “striking class allegations” are functionally equivalent to denials for interlocutory review purposes. The majority leverages Microsoft to explain why Rule 23(d)(1)(D) is not itself the “engine” of certification denial, but a follow-on “housekeeping” mechanism once the Rule 23(c)(1)(A) determination is made.
• Eisen v. Carlisle & Jacquelin, 417 U.S. 156 (1974)
  Supports the basic principle that class propriety turns on whether Rule 23’s requirements are met—reinforcing the mismatch between Rule 12(f) and Rule 23 standards.
• Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442 (2016) and Amchem Prods., Inc. v. Windsor, 521 U.S. 591 (1997)
  Both inform the predominance/superiority discussion for Rule 23(b)(3). They frame (b)(3) as an efficiency-and-fairness device requiring aggregation-enabling common issues to outweigh individualized ones and requiring the class method to be “superior.”
• Combs v. Bakker, 886 F.2d 673 (4th Cir. 1989)
  Not a class case, but used as an analogy: when courts decide threshold issues without evidentiary development, they ask only whether the plaintiff has made a “sufficient prima facie showing” on the pleadings, and act “prematurely” when they require more at that stage.
• Mills v. Foremost Insurance Co., 511 F.3d 1300 (11th Cir. 2008) and Manning v. Boston Medical Center Corp., 725 F.3d 34 (1st Cir. 2013)
  Out-of-circuit support for the “prematurity” doctrine: striking/denying class allegations pre-discovery is appropriate only when the defect is apparent from the complaint alone.
• Reyes v. Waples Mobile Home Park Ltd. P'Ship, 903 F.3d 415 (4th Cir. 2018)
  Used to emphasize that disparate-impact claims require a “robust causal connection” between the challenged policy and the disparate impact—here, the alleged algorithm.
• Stafford v. Bojangles' Rests., Inc., 123 F.4th 671 (4th Cir. 2024)
  Warns against overgeneralized commonality. The majority uses it to show what the plaintiffs did not do: rely on vague “systemic failures,” instead alleging a concrete algorithmic mechanism.

3.2 Legal Reasoning

(a) Identifying the proper procedural vehicle: Rule 23(c)(1)(A) as the source of authority

The majority’s first doctrinal move is structural: certification decisions must come from Rule 23(c)(1)(A), which commands the court to “determine by order whether to certify.” In the majority’s view, neither Rule 12(f) nor Rule 23(d)(1)(D) provides the substantive authority to decide certification.

• Rule 12(f): The majority rejects using “impertinent” as a backdoor for Rule 23 analysis. Because Rule 12(f) does not track Rule 23 requirements, it cannot be the tool for assessing certification propriety.
• Rule 23(d)(1)(D): The majority reads this as a post-decision mechanism—once the court denies certification under Rule 23(c)(1)(A), it may then require pleadings to be amended to remove representation allegations so the case proceeds individually.

Practical upshot: defendants challenging class allegations early should frame the request as a Rule 23(c)(1)(A) motion to deny certification, with a contingent Rule 23(d)(1)(D) request to strike class allegations if denial is granted.

(b) The pleading-stage standard: “as a matter of law” / prima facie on the face of the complaint

The opinion’s core rule is a constraint on timing-driven decisionmaking: if a district court elects to decide certification before discovery, it must treat the record like a pleadings-only posture. That means no factual resolution and no weighing; the question becomes whether the complaint itself shows Rule 23 cannot be satisfied.

This reconciles two class-action axioms the majority sees as simultaneously true: (i) district courts have broad discretion over class management and timing (“early practicable time”), and (ii) discretion is bounded by legal standards that vary with procedural posture.

(c) Applying Rule 23(b)(3) vs Rule 23(b)(2): why the outcome splits

The majority interprets the district court’s reasoning as aimed at Rule 23(b)(3)’s predominance and superiority—requirements absent from Rule 23(b)(2). On the face of this complaint, individualized variables and product differences rendered a (b)(3) damages class facially unworkable: multiple states, multiple loan products, varying applicant profiles and outcomes, and individualized damages categories (“restitutionary,” “compensatory,” “statutory,” and “punitive”). For (b)(3), those features plausibly make individual issues “aggregation-defeating.”

By contrast, the (b)(2) request for declaratory/injunctive relief did not require predominance or superiority. The majority focused on whether commonality is at least plausibly alleged: the complaint asserted that every applicant submits standardized data that is run through a single “proprietary underwriting algorithm” producing a “uniquely discriminatory result.” That allegation, if developed, could yield common answers (existence of one algorithm; whether it causes disparate impact; whether it is justified). Because those are classic “central to validity” questions for all class members, the majority held denial of (b)(2) certification was premature.

3.3 Impact

• Clarifies the doctrinal “hook” for early class-certification denials in the Fourth Circuit: litigants and district courts should treat Rule 23(c)(1)(A) as the operative authorization for granting/denying certification, rather than relying on Rule 12(f) rhetoric.
• Reaffirms and modernizes the Fourth Circuit’s “prematurity” constraint from Goodman v. Schlesinger: if a court denies certification pre-discovery, the denial must be justified by facial legal insufficiency, not anticipated factual failure. This will tend to preserve class discovery (at least targeted) where the key information is within the defendant’s control (e.g., a proprietary algorithm).
• Encourages bifurcated certification thinking: a complaint may be facially insufficient for a Rule 23(b)(3) damages class while remaining plausibly viable for Rule 23(b)(2) injunctive relief, especially in policy-or-process challenges.
• Algorithmic/process-based discrimination claims: the opinion signals that alleging a single, uniform decision system can satisfy commonality at the pleading stage, even where class members sought different products—so long as the alleged mechanism plausibly spans those decisions. Future defendants may respond by showing (through early evidentiary proffers or targeted discovery) that underwriting is not uniform across products or is materially discretionary.
• Interlocutory appeal posture: by treating striking as functionally equivalent to denial (per Microsoft Corp. v. Baker), the decision reinforces that early “strike” orders can quickly become appellate flashpoints, shaping litigation strategy and settlement dynamics.

4. Complex Concepts Simplified

• Class certification (Rule 23): The court’s gatekeeping decision whether named plaintiffs may litigate for absent class members.
• Rule 23(a) requirements:
  • Numerosity: too many class members for individual joinder to be practical.
  • Commonality: shared questions that can be answered in a way that advances all class members’ claims together (not merely similar facts).
  • Typicality: the named plaintiffs’ claims resemble the class’s claims in relevant ways.
  • Adequacy: the named plaintiffs and counsel will fairly represent absent members.
• Rule 23(b)(3) (damages class): requires predominance (common issues outweigh individual ones) and superiority (class treatment is the best method compared to alternatives). These are often difficult in individualized lending decisions involving damages.
• Rule 23(b)(2) (injunction/declaration class): designed for uniform forward-looking relief (stop a policy; declare it unlawful). It does not require predominance/superiority.
• “On the face of the complaint” / “prima facie showing”: At the pleading stage, the court assumes the complaint’s well-pleaded factual allegations are true and asks whether, even so, Rule 23 cannot be met (or, conversely, whether Rule 23 is at least plausibly satisfied).
• Strike vs deny:
  • Deny certification (Rule 23(c)(1)(A)): the substantive ruling that the case will not proceed as a class.
  • Strike class allegations (Rule 23(d)(1)(D)): the pleadings-cleanup step removing representation allegations so the case proceeds as an individual action.

5. Conclusion

The Fourth Circuit’s published decision establishes a clear pleading-stage rule: when a defendant seeks to defeat class treatment before discovery, the district court’s authority to deny certification flows from Rule 23(c)(1)(A), and the denial is proper only when Rule 23 failure is apparent from the complaint as a matter of law. Applying that framework, the court upheld an early rejection of a Rule 23(b)(3) damages class due to facial predominance/superiority problems, but revived the Rule 23(b)(2) injunctive/declaratory class because the complaint plausibly alleged a single underwriting algorithm capable of generating common answers.

The decision’s broader significance lies in its procedural discipline: it preserves district-court discretion over timing while constraining early merits-like factfinding, and it provides a roadmap for how algorithmic or proprietary-process discrimination claims may survive (or fail) at the class pleading stage.