A Missouri-specific framework where the common law decides what the jury hears, and where the recall question has no square answer

Few pieces of evidence are as coveted by a products plaintiff, or as feared by the defense, as a manufacturer’s recall. A recall looks like an admission that something was wrong, and juries hear it that way. Its admissibility, however, is neither automatic nor governed by a tidy rule, and in Missouri it turns on a framework that diverges in a consequential way from federal practice. Missouri is an uncodified, common-law evidence jurisdiction: it has never adopted a comprehensive, numbered code of evidence modeled on the Federal Rules, and there is no Missouri Rule 407. The treatment of recalls and other post-sale changes is therefore governed by case law, discrete Chapter 490 statutes, and ordinary relevance and hearsay principles. That structural fact is what makes Missouri, in this corner of the law, more receptive to such evidence than the federal courts—and it is why the single most important strategic decision in a recall case is often the choice of forum.

One preliminary, because it recurs in commentary: as of this writing Missouri has not adopted a codified evidence code. The one modern, codification-adjacent development is the 2017 rewrite of section 490.065 RSMo to install a Daubert-style expert standard, which governs expert admissibility only and does not touch subsequent-remedial-measure or recall evidence. Any source asserting that Missouri has enacted rules of evidence “mirroring the Federal Rules” should be treated with suspicion and independently verified against the current legislative session before it is relied upon.

I. Subsequent Remedial Measures: The Pollard Framework

A recall is, functionally, a species of subsequent remedial measure (SRM)—a post-sale step to address a hazard—so the SRM doctrine is the analytical starting point. The controlling Missouri authority is Pollard v. Ashby, 793 S.W.2d 394 (Mo. App. E.D. 1990) (en banc), arising from a stroke allegedly caused by a chymopapain injection, in which the plaintiffs recovered $700,000 against the manufacturer on a strict-liability failure-to-warn theory. Pollard draws the line by theory of liability. In negligence, the traditional rule excludes SRM evidence. In strict liability, the court held, the rule is different, because the strict-liability inquiry asks whether the product was defective—not whether the manufacturer was at fault—so a post-injury corrective change can be probative of the defect. The court declined to import Federal Rule of Evidence 407’s strict-liability bar into Missouri common law. (Practitioner sources place the SRM discussion around 793 S.W.2d at 401–403; reconcile the exact pin against the bound reporter before quoting.)

The rule is not confined to a single district. In Stinson v. E.I. DuPont de Nemours & Co., 904 S.W.2d 428, 432 (Mo. App. W.D. 1995)—a different appellate district, which meaningfully reduces the risk that Pollard is an outlier—the Western District reversed a directed verdict for the manufacturer and held that the SRM exclusionary rule is inapplicable in strict-liability cases “because (1) the culpability of the defendant is irrelevant and (2) the purposes of the rule are not served,” finding the exclusion of DuPont’s later, stronger warning labels to be error. Stinson is the more useful of the two authorities for a recall case, because a later warning or label change is the closest analog to a recall. Two negligence-side decisions round out the doctrine: Dick v. Children’s Mercy Hospital, 140 S.W.3d 131 (Mo. App. W.D. 2004), states the general bar and its recognized exceptions—“ownership, control, or feasibility of precautionary measures, if controverted, or impeachment”; and Boggs ex rel. Boggs v. Lay, 164 S.W.3d 4, 21 (Mo. App. E.D. 2005), adds a transferable escape hatch, admitting post-accident measures where they were “planned, provided for, or undertaken prior to the accident,” because the exclusionary policy does not apply to changes the defendant was already making. That prior-planning principle is powerful in a recall case, since a recall frequently reflects a defect the manufacturer already knew about and was already addressing before the plaintiff’s injury.

Distilled, the Missouri rule is this: in negligence claims, SRM evidence is excluded as proof of fault and admissible only for a permitted purpose (ownership or control, feasibility if controverted, impeachment, or the Boggs prior-planning situation); in strict-liability design and warning claims, the exclusionary rule does not apply at all, and post-injury design changes and improved warnings are admissible as probative of the defect (Pollard; Stinson). Even so, admission is not automatic—the evidence must be relevant and survive a probative-value-versus-prejudice weighing under Missouri’s common-law analog to the familiar balancing test.

II. Why the Missouri Rule Diverges From Federal Practice

This is a genuine and consequential state-federal divergence. Federal Rule of Evidence 407, as amended in 1997, bars subsequent-remedial-measure evidence offered to prove “negligence,” “culpable conduct,” “a defect in a product or its design,” or “a need for a warning or instruction.” The 1997 amendment resolved a prior circuit split against plaintiffs by adding the two product-specific prohibited purposes, so the federal rule now bars in strict-liability design and warning cases exactly the evidence that Pollard and Stinson treat as admissible. A plaintiff frequently has a materially better chance of admitting a post-accident design change—or a recall—in Missouri state court than in federal court or a state that has codified the federal rule.

The corollary is an Erie problem the practitioner must anticipate. Because Missouri’s strict-liability SRM rule is one of admissibility, federal courts sitting in diversity have generally treated FRE 407 as controlling and applied it even to claims arising under Missouri substantive law—meaning the plaintiff loses the Pollard advantage the moment the case is in federal court. There is a respectable minority argument that SRM exclusion is so bound up with state substantive tort policy that state law should govern in diversity, and it is worth preserving in a brief, but the prevailing view is that FRE 407 controls admissibility in federal court. The lesson is blunt: forum can decide the case. A plaintiff who wants the recall in should think hard about defeating removal; a defendant who wants it out should think hard about achieving it.

III. Recall Evidence Specifically

Now the candor point that must anchor any honest treatment: no Missouri appellate decision squarely governs the admissibility of a product recall as such. Recall admissibility in Missouri is an inference from the general Pollard SRM framework plus ordinary relevance and hearsay doctrine—not a holding. Any statement that “Missouri law holds recalls admissible” (or inadmissible) is synthesis, and should be presented as such.

Within that framework, recall evidence is most defensible when offered to show notice or knowledge of a hazard (central to failure-to-warn and to punitive-damages state of mind), the feasibility of a safer design or better warning where the defense controverts feasibility, impeachment (to rebut testimony that the product was the safest available), and—in a strict-liability posture under Pollard—as circumstantial evidence of the defect itself. It is most vulnerable, and commonly excluded, when offered as conclusive proof that the plaintiff’s particular unit was defective: the regulatory threshold that triggers a recall is not the tort standard for “unreasonably dangerous,” and a recall issued out of an abundance of caution does not establish the tort element as a matter of law.

The relevance-matching requirement is the defense’s strongest lever. The recalled defect must match the defect and failure mode the plaintiff alleges, and most recall evidence is won or lost here—often under a Rule 403-type prejudice balance rather than the SRM rule. The controlling illustration under Missouri law is Bizzle v. McKesson Corp., 961 F.2d 719 (8th Cir. 1992), in which a walking cane allegedly broke and the plaintiffs sought to introduce that the manufacturer recalled the model after learning of the lawsuit. The Eighth Circuit affirmed exclusion, and its reasoning is the crucial nuance: it recognized that Rule 407 does not of its own force bar recall evidence in a strict-products case, but affirmed exclusion under Rule 403 because the evidence tying the plaintiff’s specific cane to the recalled defect was too thin—probative value substantially outweighed by unfair prejudice. Bizzle is the single most citable federal-Missouri recall decision, and it teaches that even where SRM doctrine does not bar the recall, relevance-matching under 403 will. The same lesson runs through persuasive out-of-circuit authority—Calhoun v. Honda Motor Co., 738 F.2d 126 (6th Cir. 1984) (recall letter irrelevant absent proof the recalled condition existed in this motorcycle), and HDM Flugservice GmbH v. Parker Hannifin Corp., 332 F.3d 1025 (6th Cir. 2003) (leading authority on the voluntary-versus-mandatory recall distinction under Rule 407)—which are not Missouri law but frame the relevance and voluntariness fights.

IV. The Federal Statutory Overlay

Counsel sometimes assume a federal statute bars recall evidence. Generally, none does. On motor vehicles, the National Traffic and Motor Vehicle Safety Act governs the administrative recall process but contains no evidentiary bar; to the contrary, its savings clause, 49 U.S.C. § 30103(e), provides that “[c]ompliance with a motor vehicle safety standard prescribed under this chapter does not exempt a person from liability at common law,” preserving state claims. The Act’s express-preemption provision, § 30103(b), limits conflicting state standards; it is not an evidentiary rule. On consumer products, the Consumer Product Safety Act’s savings clause, 15 U.S.C. § 2074(a), preserves common-law liability, and the only pertinent bar, § 2074(b), excludes evidence of the Commission’s failure to act—“[t]he failure of the Commission to take any action or commence a proceeding . . . shall not be admissible.” Read precisely, that provision bars evidence of CPSC inaction, not evidence of an actual recall; defense counsel occasionally over-read it, and the statute does not reach that far.

Preemption doctrine likewise does not create an evidentiary bar. Geier v. American Honda Motor Co., 529 U.S. 861 (2000), held a state “no-airbag” design claim impliedly preempted by a federal safety standard, and Williamson v. Mazda Motor of America, Inc., 562 U.S. 323 (2011), later narrowed Geier’s reach—but both are substantive conflict-preemption holdings that can bar certain claims; neither bars the admission of recall or SRM evidence where a claim survives. In short, no statute affirmatively bars an NHTSA or CPSC recall; the only statutory evidentiary bar in the space is the narrow one on CPSC inaction.

V. Hearsay and Foundation

A manufacturer’s own recall notice, offered against the manufacturer, is ordinarily not hearsay—it is a statement of a party opponent, and internal recall memoranda and dealer communications qualify as party or authorized admissions. This is the cleanest foundation and the reason recalls are usually admissible against the maker once they clear the relevance-match and SRM hurdles. The harder case is the recall offered against a downstream distributor or retailer: the notice is the upstream manufacturer’s statement, not the seller’s, so it faces a stronger hearsay objection. Plaintiff’s routes against a non-manufacturer seller are to show the seller adopted or republished the notice (adoptive admission), to offer it not for its truth but for the notice or knowledge the seller received (a non-hearsay purpose, with a limiting instruction), or to reach agency-generated records through a public-records or business-records theory—in Missouri, via the Uniform Business Records as Evidence Act, section 490.680 RSMo, and the official-records provisions of Chapter 490. Foundation is laid by authenticating the document (self-authentication for certified agency records), establishing the non-hearsay or exception basis and matching it to the specific defendant, tying the recalled defect to the alleged failure mode with independent evidence, and tendering a limiting instruction where the evidence is admissible only for notice, impeachment, or feasibility.

VI. Litigation Strategy

Plaintiff—getting it in. Frame the claim in strict liability and lean on Pollard and Stinson; keep the recall tethered to the defect or warning theory, where the SRM bar does not apply at all. Prove the defect match independently first—expert testimony that this unit shared the recalled condition and failure mode—so the recall corroborates rather than carries the case; that is the antidote to Bizzle. Where the case sounds in negligence, drive the defense to controvert feasibility or oversell the product, then offer the recall for feasibility or impeachment, or invoke Boggs where the remedial plan predated the injury. Above all, stay in state court if possible: removal forfeits the Pollard advantage, because FRE 407 bars the very defect-and-warning use Missouri permits.

Defense—keeping it out. Remove to federal court where possible to obtain Rule 407’s product-liability bar. In state court, attack relevance and prejudice hard under the 403 analog (Bizzle; Calhoun): different model, year, component, or failure mode; recall issued from an abundance of caution; recall threshold not the tort standard. Stipulate to ownership, control, and feasibility to close those exception doors (Dick). Argue voluntariness under HDM Flugservice if Rule 407 is in play federally, and keep witnesses from superlatives that open impeachment. Press the hearsay objection when a manufacturer’s recall is offered against a non-manufacturer client, and do not over-rely on section 2074(b) or on Geier, neither of which bars recall evidence.

VII. Analytical Tensions

Three cautions deserve emphasis. First, Pollard is an intermediate en banc decision, not a Supreme Court of Missouri holding, and it drew a partial dissent; it should be KeyCited before it is relied upon, though Stinson’s agreement from a second district reduces outlier risk. Second, the entire recall analysis is inferential: because no Missouri appellate case squarely governs recall admissibility, the practitioner builds from SRM, relevance, and hearsay analogies plus the Eighth Circuit’s Bizzle gloss on Missouri law, and should say so rather than overstate the authority. Third, the framework is vulnerable to a single legislative act: if Missouri ever adopts a numbered evidence code with a Rule 407 analog tracking the post-1997 federal rule, the Pollard strict-liability advantage could evaporate by codification. That has not happened, and nothing located suggests it is imminent—but it is the development most worth monitoring, because it would quietly rewrite the strategic calculus this article describes.

Related in the Products Liability Series

• Missouri Product Liability Law: The Complete Guide
• Design Defect Claims Under Missouri Law — the undefined “unreasonably dangerous” standard and how design cases are submitted.
• Manufacturing Defect Claims Under Missouri Law — proving departure from design by circumstantial inference and expert proof.
• Failure-to-Warn Claims Under Missouri Law — the two causation prongs and the read-and-heed presumption.
• Safer Alternative Design Requirements — why Missouri treats alternative design as evidence, not an element.
• The Learned Intermediary Doctrine in Missouri Products Cases — the physician as the object of the duty to warn, and the preemption overlay.
• Investigating the Products Case: Preservation, Spoliation, and Cost.