Summary. Any business that makes, imports, distributes, or sells a physical product can be sued when that product hurts someone, and the claim usually does not require proof that anyone was careless. This article explains the three theories that carry product cases and the three defect categories courts recognize: strict liability's origins in Greenman and § 402A, the manufacturing defect claim that is nearly self-proving, the design defect claim and the divide between consumer expectations and risk-utility, and the failure-to-warn claim that generates most modern litigation. It covers the parallel negligence and warranty theories and the disclaimers that do and do not work. A long section covers the defenses that decide cases: preemption after Riegel and Wyeth, state of the art and compliance, misuse and alteration, comparative fault, learned intermediary and sophisticated user, statutes of repose, and sealed container statutes. It closes with recall obligations, insurance, supply chain risk allocation, a worked example, checklists, an FAQ, and related reading.

A hardware company sells a $190 cordless drill. It designed nothing, manufactured nothing, and imported nothing — it bought the drill from a domestic distributor and put it on a shelf.

A customer is injured when the trigger lock fails. The manufacturer is in Shenzhen and cannot practically be sued. The distributor has dissolved.

In most states, the retailer is strictly liable. It placed the product in the stream of commerce, and the doctrine does not require fault. Its remedies are indemnity from parties that no longer exist and an insurance policy it may or may not have read.

Product liability is the area of law where a business's exposure is least connected to its own conduct, and where the protections that actually work are commercial rather than doctrinal: choosing suppliers who can stand behind their goods, contracting for indemnity from someone solvent, and buying the right insurance.

The short answer

Three theories, usually pleaded together:

  1. Strict liability in tort — the product was defective and unreasonably dangerous when it left the defendant's control, and the defect caused the injury. No proof of negligence required.
  2. Negligence — the defendant failed to exercise reasonable care in design, manufacture, inspection, or warning.
  3. Breach of warranty — express, implied warranty of merchantability (UCC § 2-314), or implied warranty of fitness for a particular purpose (UCC § 2-315).

Three defect categories:

  • Manufacturing defect — the product departed from its intended design. The easiest claim to prove and the hardest to defend.
  • Design defect — the design itself is unreasonably dangerous. Tested by consumer expectations, risk-utility, or both, depending on the state.
  • Failure to warn (marketing defect) — inadequate warnings or instructions about non-obvious risks.

Who is liable: everyone in the distribution chain — manufacturer, component maker, assembler, importer, distributor, wholesaler, and retailer — subject to statutory protections for innocent sellers in many states.

The most important practical point: liability attaches without fault, so risk management happens through supplier selection, contracts, insurance, and documentation, not through being careful alone.

Part I: How strict liability arose

The doctrine's development is short and worth knowing, because the reasoning still drives outcomes.

Privity fell first. MacPherson v. Buick Motor Co., 217 N.Y. 382 (1916), Judge Cardozo, held that a manufacturer owes a duty of care to the ultimate user even without a contractual relationship. Before it, an injured consumer could sue only the retailer it bought from.

Justice Traynor proposed strict liability in his concurrence in Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453 (1944), arguing that "public policy demands that responsibility be fixed wherever it will most effectively reduce the hazards to life and health inherent in defective products that reach the market," and that the manufacturer can distribute the risk as a cost of doing business.

The California Supreme Court adopted it in Greenman v. Yuba Power Products, Inc., 59 Cal. 2d 57 (1963), holding that "[a] manufacturer is strictly liable in tort when an article he places on the market, knowing that it is to be used without inspection for defects, proves to have a defect that causes injury to a human being."

The Restatement codified it at § 402A of the Restatement (Second) of Torts (1965), which nearly every state adopted in some form. It imposes liability on one who sells "any product in a defective condition unreasonably dangerous to the user or consumer," even though "the seller has exercised all possible care."

The Restatement (Third) of Torts: Products Liability (1998) reorganized the field into the three defect categories and, controversially, requires proof of a reasonable alternative design for most design defect claims. States have adopted it unevenly; many retain § 402A and their own case law, so the operative framework is state-specific.

Part II: The three defects

Manufacturing defect

The product departed from its own intended design, however careful the quality control. A single bolt with a void, a batch with the wrong resin, a missing weld.

  • Proof is straightforward: compare the unit to the specification or to other units.
  • The malfunction theory (sometimes called res ipsa for products) allows the plaintiff to prove a defect circumstantially where the product failed in a manner that ordinarily does not occur absent a defect, and other causes are excluded — useful where the product was destroyed in the incident.
  • There is essentially no defense on the merits beyond causation, alteration, and misuse. The Restatement (Third) makes liability for manufacturing defects flatly strict.

Consequence for manufacturers: quality records, lot traceability, and retained samples are the difference between defending one unit and defending a production run.

Design defect

The entire product line is alleged to be unreasonably dangerous as designed. These are the expensive cases, because a design defect finding implicates every unit ever sold.

Two tests, and the state's choice matters enormously:

Consumer expectations. The product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner. Plaintiff-friendly, requires no expert on alternative design, and works best for simple products where consumers have expectations. It fits poorly where the product is complex and the ordinary consumer has no basis for an expectation about, say, the crashworthiness of a suspension geometry.

Risk-utility. The risk of the design outweighs its benefits, usually assessed on factors including the gravity and likelihood of harm, the feasibility and cost of a safer alternative design, the adverse consequences of that alternative to the product and the consumer, and the user's ability to avoid the danger. The Restatement (Third) makes a reasonable alternative design essentially an element, which raises the cost of bringing a case and has been criticized for foreclosing claims against products that are unreasonably dangerous but have no safer alternative.

California's hybrid. Barker v. Lull Engineering Co., 20 Cal. 3d 413 (1978), permits the plaintiff to prove a design defect under either test, and — importantly — shifts the burden to the defendant on the risk-utility prong once the plaintiff shows the design proximately caused the injury. Several states follow variants; others use one test exclusively; and some apply consumer expectations only to products within ordinary experience.

Practical consequence: the applicable test should be identified before the first expert is retained, because it determines whether an alternative design analysis is required and who bears the burden.

Failure to warn

The largest category of modern product litigation.

The duty is to warn of non-obvious dangers that the seller knows or should know about, arising from foreseeable uses and foreseeable misuses. The warning must be adequate in content (identifying the risk, its severity, and how to avoid it), form (conspicuous, comprehensible, appropriately placed), and reach (delivered to the person who needs it).

Points that decide cases:

  • Open and obvious dangers generally need no warning, though the doctrine is narrower than defendants hope.
  • Post-sale duty to warn. Many states recognize a continuing duty where the seller learns of a danger after sale and can practicably identify and reach users. Registration cards, service networks, and connected products all make identification practicable, which paradoxically increases the duty.
  • Heeding presumption. Many states presume a plaintiff would have followed an adequate warning, shifting the burden on causation to the defendant.
  • Learned intermediary. For prescription drugs and devices, the manufacturer's duty runs to the prescribing physician, not the patient. Widely but not universally accepted, and eroded in some states for direct-to-consumer advertised products.
  • Sophisticated user / bulk supplier. No duty to warn a user who already knows the risk by virtue of training or position, or where the supplier reasonably relies on an intermediary to warn.
  • Warnings cannot cure a defective design. In most states a manufacturer cannot warn its way out of a design that should have been made safer.

Part III: Negligence and warranty

Negligence

Requires proof of a breach of reasonable care. It is pleaded alongside strict liability because:

  • It reaches conduct claims that strict liability does not: negligent inspection, negligent testing, negligent recall, negligent entrustment.
  • It supports punitive damages more readily, where the conduct was reckless or where the defendant knew of the danger and did nothing.
  • Some states restrict strict liability against non-manufacturing sellers but allow negligence claims.

Negligence per se may apply where the product violated a safety statute or regulation.

Warranty

A contract theory with different rules and, importantly, different defendants and different damages.

  • Express warranty, UCC § 2-313: any affirmation of fact or promise relating to the goods that becomes part of the basis of the bargain. Marketing statements, spec sheets, and demonstrations all qualify. Puffery does not.
  • Implied warranty of merchantability, UCC § 2-314: goods sold by a merchant must be fit for the ordinary purposes for which such goods are used. Automatic unless disclaimed.
  • Implied warranty of fitness for a particular purpose, UCC § 2-315: arises where the seller knows the buyer's particular purpose and that the buyer is relying on the seller's skill or judgment.

Disclaimers, UCC § 2-316: merchantability may be disclaimed only by language mentioning merchantability and, if written, conspicuous; fitness only in writing and conspicuously; and both may be excluded by "as is," "with all faults," or similar language, or by the buyer's examination of the goods.

Limitation of remedy, UCC § 2-719: a seller may limit remedies to repair or replacement, but if that limited remedy fails of its essential purpose, other remedies revive; and limitation of consequential damages for personal injury in the case of consumer goods is prima facie unconscionable.

The Magnuson-Moss Warranty Act, 15 U.S.C. §§ 2301-2312, adds a federal layer for consumer products: a written warranty must be designated full or limited, must be available before sale, and — critically — a supplier that gives a written warranty may not disclaim or modify implied warranties, only limit their duration to the written warranty's duration for a limited warranty. It also provides a private right of action with attorney's fees, which is why warranty claims accompany most consumer product suits.

Privity still matters for some warranty claims in some states, particularly for economic loss. Which brings up the doctrine that separates the tort and contract worlds.

The economic loss rule

Where a defective product damages only itself and causes purely economic loss (repair cost, lost profits, diminished value), most states bar recovery in tort and relegate the buyer to contract and warranty remedies. Damage to other property and personal injury remain in tort.

This is why a commercial buyer of a failed machine sues for breach of warranty, and why the limitation of liability clause in the sales contract — rather than tort doctrine — determines the recovery. See Indemnification and Limitation of Liability.

Part IV: The defenses that decide cases

Federal preemption

The most powerful defense where it applies, because it defeats the claim entirely rather than contesting it.

Express preemption. Riegel v. Medtronic, Inc., 552 U.S. 312 (2008), held that the Medical Device Amendments' express preemption clause bars state common-law claims challenging the safety or effectiveness of a device that received premarket approval from the FDA, because such claims would impose requirements "different from, or in addition to" the federal ones. PMA devices are therefore largely immune from design and warning claims, subject to "parallel claims" that allege violation of the federal requirements themselves. Devices cleared through the less rigorous 510(k) pathway are not preempted.

Implied conflict preemption. Wyeth v. Levine, 555 U.S. 555 (2009), held that a failure-to-warn claim against a brand-name drug manufacturer was not preempted, because the manufacturer could have strengthened its label unilaterally under the changes being effected regulation, and because Congress had not intended FDA approval to be a ceiling. The Court framed the test as impossibility: preemption requires "clear evidence that the FDA would not have approved" the warning the state law would require. Albrecht later clarified that whether that showing is made is a question for the judge, not the jury.

Generic drugs are different. Because federal law requires a generic's label to match the brand's, a generic manufacturer cannot unilaterally change it, and failure-to-warn claims against generics are preempted. That distinction produces the widely criticized result that a patient's remedy depends on which version of an identical drug the pharmacist dispensed.

Other preemptive regimes: motor vehicle safety standards (preemption is narrow and fact-specific), pesticides under FIFRA, boat safety, and pharmaceutical design claims in some formulations.

State of the art and regulatory compliance

State of the art — the design conformed to the best available technology and scientific knowledge at the time of manufacture. In many states this is a defense or at least strong evidence, particularly on design and warning claims; in a few it is not a defense to strict liability at all.

Regulatory compliance — the product met all applicable government standards. Almost everywhere this is evidence of non-defectiveness but not a complete defense, on the reasoning that regulatory standards are minimums. Several states have enacted a rebuttable presumption of non-defectiveness for compliant products, and a few provide near-immunity for FDA-approved pharmaceuticals absent fraud on the agency.

Misuse, alteration, and assumption of risk

  • Unforeseeable misuse breaks the causal chain. Foreseeable misuse does not, and the duty to warn extends to it, so "the user did something stupid" is only a defense if the stupidity was genuinely unforeseeable.
  • Substantial alteration after sale, including removal of a guard or defeat of an interlock, is a strong defense — though a manufacturer whose guard is routinely removed in the field may be found to have designed a guard that invited removal.
  • Assumption of risk survives in some states as a complete or partial defense where the plaintiff knew of and voluntarily encountered the specific danger.

Comparative fault

Most states apply comparative fault to product claims, reducing the award by the plaintiff's share. Modified comparative fault jurisdictions bar recovery entirely at 50 or 51 percent. A minority retain pure comparative fault, and a very small number still use contributory negligence as a complete bar.

Allocation among defendants is governed by state joint and several liability rules, which have been modified extensively by statute: some states retain joint liability, some apportion severally, and many use hybrid rules keyed to the defendant's percentage of fault.

Statutes of limitations and repose

  • Limitations run from injury or discovery, typically two to four years.
  • Statutes of repose run from the date of sale or first use and extinguish the claim regardless of when the injury occurs, commonly after ten to fifteen years. Roughly a third of states have them for products; several apply only to specified product categories, and constitutional challenges have succeeded in some states.
  • The General Aviation Revitalization Act, 49 U.S.C. § 40101 note, is a federal repose statute barring most claims against manufacturers of small aircraft more than eighteen years after delivery.

Repose is a first-week defense to check, because it is dispositive.

Innocent seller protections

Because strict liability reaches everyone in the chain, most states have enacted sealed container or innocent seller statutes protecting non-manufacturing sellers who did not alter the product and had no knowledge of the defect. The protections vary: some dismiss the seller once the manufacturer is identified and subject to jurisdiction; some preserve claims where the manufacturer is insolvent or beyond reach; some do not protect a seller that exercised control over the design or held the product out as its own.

The "apparent manufacturer" doctrine removes the protection where the seller put its own brand on the product — which is precisely what private-label and white-label programs do. A retailer selling under its house brand is treated as the manufacturer in most states.

The marketplace question. Whether an online marketplace that does not take title is a "seller" for strict liability purposes has divided courts, with several state high courts and legislatures moving toward liability where the platform controls the transaction, and other courts holding the platform is merely a service provider. This is an active area and the answer is jurisdiction-specific.

Damages, and what drives the number

Compensatory damages in a product case include medical expenses (past and future), lost earnings and impaired earning capacity, pain and suffering, disfigurement, and, in a death case, the wrongful death and survival damages the state allows. Property damage and, where the economic loss rule permits, consequential losses may be added.

Punitive damages are the variable that makes product cases existential rather than merely expensive. They generally require clear and convincing evidence of conduct beyond negligence — variously described as malice, oppression, fraud, or conscious disregard of a known risk. The evidence that supports them is almost always internal documents: a risk assessment that quantified the hazard and recommended a fix that was deferred for cost, a complaint trend report that circulated for two years without action, an engineer's memo that was overruled. Constitutional limits from the due process line of cases constrain the ratio to compensatory damages in most circumstances, and many states cap them by statute, but the exposure remains material.

The practical implication is uncomfortable and worth stating plainly: the documents that make a company safer — hazard analyses, complaint trending, failure mode reviews — are the same documents a plaintiff will use if the company identified a risk and did nothing. The answer is not to stop creating them. It is to close them out: every identified hazard should have a documented disposition, and a decision not to act should record the engineering basis for that decision rather than the cost.

Settlement dynamics in product cases are driven less by the individual claim than by the line exposure. A design defect finding implicates every unit sold, which is why defendants settle strong individual cases and try weak ones, and why a plaintiff's first case against a product line is often the most valuable one.

Part V: Compliance, recalls, and risk transfer

Reporting and recall obligations

Under the Consumer Product Safety Act, 15 U.S.C. § 2064(b), a manufacturer, importer, distributor, or retailer that obtains information reasonably supporting the conclusion that a product fails to comply with a consumer product safety rule, contains a defect that could create a substantial product hazard, or creates an unreasonable risk of serious injury or death must immediately inform the Consumer Product Safety Commission — in practice within 24 hours of obtaining reportable information, per the Commission's regulations.

Companies also must report certain settled or adjudicated lawsuits alleging death or grievous bodily injury under § 2084 in defined circumstances.

Penalties for late reporting are substantial, and the Commission has pursued them vigorously. The most common error is an internal decision to "investigate first," which delays a report past the deadline. The correct approach is a defined reportability review process with a short clock, run by counsel, that documents the analysis whether or not a report follows.

Other regulators have their own regimes: NHTSA for vehicles and equipment (with its own defect reporting and recall obligations), FDA for food, drugs, devices, and cosmetics, and USDA for meat and poultry.

A recall is a project, not an announcement: scope determination, corrective action plan, notification to the regulator and the public, retailer and distributor communication, remedy logistics (repair, replace, refund), monitoring of completion rates, and record retention. It should be planned before it is needed, with a designated team and a rehearsed playbook.

Insurance

Product liability coverage sits within CGL Coverage A (bodily injury and property damage). Points to confirm:

  • Products-completed operations hazard is included, not excluded, and the aggregate for it is adequate.
  • Occurrence form, so injury during the policy period is covered whenever the claim arrives — important because product claims surface years later.
  • Vendors endorsement naming distributors and retailers as additional insureds, which is what your customers will require of you.
  • Recall coverage is generally excluded from CGL and must be bought separately (product recall / product contamination insurance), covering recall expenses, and sometimes lost profits and rehabilitation costs.
  • Tail considerations on any sale of the business, since occurrence coverage responds by year of injury and the entity may no longer exist.

See Business Insurance and Coverage Disputes.

Contractual risk allocation

For a distributor or retailer, this is the practical defense.

Up the chain (from your supplier), require:

  • An indemnity covering third-party product claims, including defense, with no cap for bodily injury.
  • Additional insured status on the supplier's CGL with a vendors endorsement, primary and non-contributory, and waiver of subrogation — evidenced by the actual endorsements, not a certificate.
  • Minimum limits appropriate to the product's risk, with products-completed operations aggregate confirmed.
  • Warranties of compliance with applicable safety standards and of adequate warnings and instructions.
  • Notification obligations for defects, incidents, and regulatory contacts.
  • Recall cooperation and cost allocation.
  • For imported goods, a domestic entity with assets, or a resident agent, because an indemnity from an offshore manufacturer with no U.S. presence is unenforceable in practice.

Down the chain (to your customers): conspicuous warranty disclaimers where permitted, limitation of remedies, and a limitation of liability that respects UCC § 2-719(3) for personal injury in consumer goods.

The imported-goods problem deserves emphasis. A U.S. importer is treated as the manufacturer for many purposes, is the party CPSC will pursue, and cannot practically enforce an indemnity against a foreign supplier. If your business imports, your insurance and your supplier due diligence are the real controls, and both should be priced into the landed cost.

A worked example

Thistle Home Goods (fictional) sells a private-label $79 electric kettle manufactured in Vietnam and imported by Thistle directly. Twenty-two units have failed with the base overheating; one caused a kitchen fire with a burn injury.

Thistle's position. It is the importer and the apparent manufacturer (its brand is on the product). Innocent seller statutes do not protect it. It is the entity CPSC will look to. The Vietnamese manufacturer has no U.S. assets.

Immediate obligations:

  1. Reportability review within 24 hours of the information reasonably supporting a conclusion of a defect creating a substantial product hazard. Given twenty-two failures and a fire with injury, a report is almost certainly required. Document the analysis and report.
  2. Preserve the failed units, all quality records, the incident file, and communications with the manufacturer. Issue a litigation hold. See Litigation Hold and Evidence Preservation Checklist.
  3. Tender to the CGL carrier immediately, and separately assess recall coverage.
  4. Root cause analysis through a qualified engineer, under privilege where possible, coordinated with the regulatory workstream.

The claims to expect: manufacturing defect (if the failures trace to a component lot), design defect (if the thermal cutoff design is inadequate across the line), and failure to warn. Warranty claims under the UCC and Magnuson-Moss will accompany them, and Thistle's written warranty means it cannot disclaim implied warranties, only limit their duration.

Thistle's realistic risk transfer: minimal from the manufacturer. What it should have had, and should build now: a domestic entity or agent in the supply chain, additional insured status with a vendors endorsement, a compliance certificate for each production lot, retained samples, lot traceability, and recall insurance.

The recall itself: determine scope by lot and date range from traceability records; agree a corrective action plan with CPSC; notify retailers and consumers; provide a remedy; monitor completion; and retain records. A traceability system that can identify affected lots turns a full-line recall into a partial one, which is the single highest-return investment a consumer products company makes.

Checklists

Design and pre-launch

  • Documented design review with hazard analysis and consideration of alternative designs.
  • Compliance with applicable mandatory and voluntary standards, documented.
  • Testing records retained.
  • Warnings and instructions drafted against foreseeable use and foreseeable misuse; reviewed against ANSI Z535 formatting conventions where applicable.
  • Labeling reviewed for regulatory requirements.
  • Lot traceability designed into manufacturing and distribution.
  • Retained samples program.

Supply chain

  • Supplier qualification, including financial capacity and insurance.
  • Indemnity, additional insured with vendors endorsement, primary and non-contributory, waiver of subrogation — endorsements obtained.
  • Compliance and testing certificates per lot.
  • Domestic entity or agent for imported goods.
  • Change control: no unapproved component substitutions.

Post-sale

  • Complaint intake and trending, with a defined escalation threshold.
  • Reportability review process with a 24-hour clock.
  • Recall playbook with a named team.
  • Post-sale duty to warn assessed as information develops.
  • Registration or contact data to enable notification.

Insurance

  • Products-completed operations included with adequate aggregate.
  • Occurrence form; historical policies located and catalogued.
  • Recall coverage evaluated separately.
  • Vendors endorsements issued to customers as required.

Frequently asked questions

We only sell the product. Are we liable? Potentially, under strict liability. Many states protect innocent sellers by statute, but the protection disappears if you sold under your own brand, altered the product, or the manufacturer cannot be reached.

Do we have to prove the manufacturer was careless? No. Strict liability requires a defect and causation, not fault. That is the doctrine's central feature.

Is compliance with government standards a defense? It is evidence, and in some states a rebuttable presumption, but almost never a complete defense — standards are treated as minimums. Preemption is different and can be complete, as with PMA medical devices.

Can we warn our way out of a design problem? Generally no. Most states hold that an adequate warning does not cure an unreasonably dangerous design.

How long are we exposed? Limitations run from injury or discovery, typically two to four years. Statutes of repose in about a third of states cut off claims ten to fifteen years after sale regardless of injury date. Check both early.

What is the learned intermediary doctrine? For prescription drugs and devices, the manufacturer's warning duty runs to the prescriber rather than the patient. Widely adopted, with erosion in some states for directly advertised products.

When must we report to CPSC? Immediately — in practice within 24 hours — of obtaining information reasonably supporting a conclusion that the product contains a defect that could create a substantial product hazard, creates an unreasonable risk of serious injury or death, or fails to comply with a safety rule. Do not wait to complete an investigation.

Does our general liability policy cover a recall? The cost of the recall itself, generally no; that requires separate product recall coverage. Liability for injuries the product caused is covered under Coverage A.

We import. What is our biggest exposure? Being treated as the manufacturer, with an indemnity you cannot enforce against a supplier you cannot reach. The controls are insurance, supplier qualification, traceability, and, where possible, a solvent domestic party in the chain.

What is the single best investment in reducing product risk? Lot traceability. It limits recall scope, supports the manufacturing-defect defense for unaffected units, and shortens every investigation.

Closing thought

Product liability is the clearest example in commercial law of risk that cannot be eliminated by care alone. A company can design well, test thoroughly, warn adequately, and still be strictly liable for the one unit that left the line wrong.

What careful companies actually control is the consequence: whether they can identify which units are affected, whether the record shows a documented design and testing process, whether someone solvent stands behind the goods, and whether the insurance responds. Those four things determine whether an incident is a manageable claim or an existential one.

And for the retailer or distributor who did nothing but put a box on a shelf, the entire defense is contractual and commercial. Choose suppliers who can be sued, get the endorsements rather than the certificate, and read the innocent seller statute in every state you ship to — before the fire, not after.

Related articles

This article is provided for general informational purposes and does not constitute legal advice. Product liability doctrine, defenses, and repose periods vary substantially by state, and regulatory reporting obligations are time-sensitive. Consult qualified counsel immediately upon learning of a potential product hazard.