Why Product Liability Limits Quietly Fail Manufacturers

The Product Liability Illusion: Why Many Manufacturers Are Carrying Limits That Will Not Hold in a Serious Defect Claim

For a long time, mid-sized manufacturers treated Product Liability insurance as a routine line item. The purchase process was mechanical: review the prior year’s premium, adjust the sales figures, confirm the standard $1 million or $2 million structure, issue certificates to distributors, and move on. That approach worked reasonably well in a softer casualty market where defense costs were manageable, plaintiff litigation was less capitalized, and carriers still competed on broad wording.

That environment has changed more quickly than many insureds realize.

The modern Product Liability placement is no longer just a question of whether a company has coverage. The more important question is whether the wording, attachment structure, and contractual assumptions behind that coverage are remotely adequate once a systemic defect event begins to unfold. In many cases, the answer is no. Companies are increasing premiums, carrying what appear to be respectable aggregate limits, and still discovering—usually too late—that the policy was never built for the way current claims are actually adjusted.

One of the most misunderstood issues remains the relationship between the per-occurrence limit and the aggregate limit. Buyers routinely focus on the aggregate because it appears larger and therefore psychologically safer. But in a batch failure scenario, that aggregate number can become almost irrelevant.

Most standard Commercial General Liability placements still allow the carrier broad discretion to consolidate multiple losses arising from a common manufacturing defect into one occurrence. The exact language differs by carrier manuscript form, but the practical outcome is familiar to anyone who has watched a defective component claim develop over twelve or eighteen months: dozens, sometimes hundreds, of individual incidents are treated as a single originating event.

Take a manufacturer supplying pressure valves into regional HVAC contractors. A machining inconsistency in one production run causes gradual seal failure after installation. The failures do not occur in one week; they surface over time, across multiple buildings, often in different states, and each one appears at first to be a separate service problem. By the time the pattern becomes obvious, there may be hundreds of water-damage claims, labor invoices, emergency replacements, and property complaints circulating simultaneously.

This is where many insureds discover that a $2 million per-occurrence / $4 million aggregate structure does not function the way they assumed. The carrier may recognize the entire chain as one batched occurrence tied to one defective run. The result is simple: one $2 million ceiling against a gross loss that may be several times that number.

That is not merely a limit issue; it is a wording issue. And many brokers still present it as if buying a larger aggregate somehow solves the problem. It does not if the occurrence definition remains restrictive.

The pressure on those occurrence limits is being compounded by something the casualty market has been talking about for several renewal cycles: severity inflation. AM Best’s commercial casualty commentary has repeatedly reflected profitability strain in liability lines, and that strain is now visible in underwriting behavior. Primary carriers are no longer assuming that a seven-figure bodily injury or property damage claim is unusual. They are pricing with the expectation that once plaintiff counsel, forensic experts, and venue-sensitive litigation strategy enter the file, the first million dollars can erode faster than the insured expects.

This is particularly dangerous under policies where defense costs sit inside the limits or where reimbursement mechanics create cash flow drag. Many insureds do not notice this during placement because the declarations page still shows the same familiar numbers. The problem emerges after the reservation of rights letter arrives and panel counsel begins billing against available limits month after month.

A manufacturer may think it purchased $1 million in indemnity. In practice, after metallurgical testing, engineering reports, deposition preparation, expert witness review, and eighteen months of procedural delay, the usable indemnity can look very different. By the time settlement discussions become serious, a substantial portion of the policy may already have been consumed by defending whether the claim should be covered in the first place.

That distinction rarely appears in broker marketing summaries, but it matters enormously once litigation becomes technical.

Another recurring source of confusion is the persistent assumption that Product Liability coverage protects the manufacturer against the cost of pulling defective goods out of circulation. Standard forms generally do not work that way.

The liability policy is designed to respond to damage caused by the product, not to the economic burden of replacing the defective product itself. That sounds obvious when stated plainly, yet it remains one of the most common post-loss misunderstandings in this segment.

If an industrial sensor fails and damages a customer’s facility, the resulting property damage may trigger the liability form. But the cost of replacing every identical sensor already distributed to warehouses, contractors, or downstream clients is a different financial event. Without dedicated Product Withdrawal or Recall language, the insured is often funding that remediation directly.

This is where technical insolvency begins for many mid-market firms. The actual lawsuit may still be manageable, but the uninsured liquidity drain associated with freight, field labor, disposal, customer communication, emergency sourcing, and replacement inventory starts long before the carrier has even accepted or denied the liability position. By the time coverage counsel and adjusters finish arguing over causation, the insured may already be carrying six figures or seven figures of unplanned corrective expense.

Underwriters know this, which is why modern submissions are being reviewed with far more attention to supplier contracts, quality assurance documentation, and indemnity recovery rights than they were a few years ago.

A clean five-year loss run no longer creates automatic comfort.

Carriers want to know who manufactured the critical component, where it was sourced, what recourse exists if that supplier failed, and whether the insured has contractually preserved any realistic path of subrogation. If the answer is vague, premiums move accordingly. Worse, some carriers will quietly manuscript endorsements that narrow how supplier-related defect chains are interpreted after the fact.

This is one of the less glamorous realities of the current market: many insureds are negotiating price while the carrier is negotiating claims posture.

Those are not the same discussion.

The contractual side creates another layer of underinsurance that is frequently missed even when companies buy higher umbrellas. Large customers increasingly impose broad indemnity language through Master Service Agreements, vendor manuals, and supply contracts. Manufacturers sign these documents to keep the account, forward them to the broker for certificate issuance, and assume the liability transfer has been insured.

Sometimes it has not.

A policy may contain healthy total limits and still fail to track the contractual liability the manufacturer has agreed to assume, particularly where the agreement extends beyond ordinary tort responsibility into broader hold harmless commitments. This gap tends not to reveal itself during underwriting because the declarations page still looks robust. It reveals itself when the downstream customer tenders a claim and expects defense and indemnity under wording the policy never fully contemplated.

At that point, increasing the umbrella after the fact is irrelevant. The insured did not purchase the right legal response in the first place.

This is why the common executive question—why am I paying more every year and still feeling exposed—has become so difficult to answer with simple limit charts.

The exposure is no longer driven solely by the amount of insurance purchased. It is driven by whether the buyer has actually audited the mechanism through which the policy will respond under a multi-claim defect event, a supplier recovery failure, a contractual indemnity tender, or a prolonged defended lawsuit.

Many have not.

They are renewing familiar numbers attached to unfamiliar restrictions.

For manufacturers entering 2026 renewals, the practical review should be less about shopping premium and more about interrogating three specific areas: the batching language attached to occurrence treatment, the treatment of defense expenses relative to available indemnity, and the extent to which customer and supplier contracts are actually mirrored by insured contract wording. Those three points do more to determine survivability in a serious product defect claim than the declarations page headline limit that most buyers spend their time negotiating.

In the present casualty environment, a standard Product Liability policy can still be useful. But usefulness should not be confused with sufficiency.

A large number of insureds are not carrying defective insurance because they failed to buy coverage.

They are carrying defective insurance because they bought coverage that still assumes claims will behave the way they did ten years ago.