Terms Not Relevant to the Actual Loss — s.11 of the Insurance Act 2015

A small manufacturing client in South Bristol recently had a fire in its plant. The damage was contained, the business interruption was modest, and the claim ought to have been settled in weeks. Instead the insurer raised a defence: an unrelated condition in the policy required the insured to keep a written log of nightly perimeter security checks. The log was not up to date. The insurer argued the claim was refused in its entirety. The cause of the fire was an electrical fault in a machine on the other side of the building. The perimeter log had nothing to do with it.

That kind of defence — refusing a loss for breach of a term that could not possibly have caused or worsened the loss — is precisely the mischief that section 11 of the Insurance Act 2015 was designed to address. For commercial insureds whose policies remain stuffed with conditions, warranties and survey requirements, s.11 is one of the most practically important protections in the Act. This article explains its scope, its limits, the burden of proof, and how brokers should use it both at placement and at notification.

The mischief: claims refused for unrelated breaches

Before the 2015 Act, the strict warranty regime under section 33 of the Marine Insurance Act 1906 (analysed in the previous article in this series) meant that insurers could refuse a claim entirely for breach of a warranty, even where the breach had nothing whatever to do with the loss. The position with conditions precedent was similar where the precedent went to the whole of cover.

The result was a steady stream of cases — and a much larger stream of disputed claims that never reached court — in which a fire claim was refused because of an unrelated alarm breach, a theft claim was refused because of an unrelated fire-suppression condition, or a liability claim was refused because of an unrelated record-keeping requirement. The Law Commissions consulted on this for nearly a decade and concluded that the principle of causation should be reintroduced into the law of warranties and risk-specific terms.

The scope of s.11(1)

Section 11(1) defines the terms to which the section applies. It applies to a term of a contract of insurance (other than a term defining the risk as a whole) compliance with which would tend to reduce the risk of one or more of the following:

• (a) loss of a particular kind;
• (b) loss at a particular location;
• (c) loss at a particular time.

The breadth of that definition is deliberate. It captures most operational warranties and conditions a commercial insured will encounter in practice: alarm warranties, sprinkler conditions, hot work permits, lone working procedures, lock specifications, key-holding arrangements, vehicle parking conditions, refrigeration monitoring conditions, and so on. Each of those terms exists to reduce the risk of a particular kind of loss, or at a particular location, or at a particular time.

The provision applies whether the term is described as a warranty, condition precedent, exclusion or condition — substance, not form, governs. This is one of the most important features of the section.

The s.11(3) protection

Section 11(3) is the operative protection. It provides that the insurer may not rely on non-compliance with a term to which s.11 applies to exclude, limit or discharge its liability “if the insured shows that the non-compliance with the term could not have increased the risk of the loss which actually occurred in the circumstances in which it occurred”.

Two features of that wording deserve emphasis.

First, the test is could not have increased the risk of the actual loss in the actual circumstances. The insured does not need to show that the breach in fact made no difference; the insured needs to show that, on the facts of the loss as it occurred, the breach could not as a matter of logic have increased the risk. The test is hypothetical and is judged at the point of loss.

Second, the burden is on the insured. The insurer does not need to prove that the breach caused or aggravated the loss; the insured must prove the negative — that it could not have done so. In practice that often turns on expert evidence (fire engineering, security consultancy, claims forensics) which the broker should be ready to help marshal.

If s.11(3) bites, the insurer cannot rely on the breach at all. The claim is paid as if the term had been complied with.

The s.11(4) carve-out: terms defining the risk as a whole

Section 11(4) preserves the position for terms which define the risk as a whole. The boundary between a risk-defining term and a risk-reducing term is the single most contested aspect of s.11 in practice and the area where drafting practice has shifted most since 2016.

In broad terms:

• A risk-defining term tells you what risk is being underwritten — for example, a clause that the cover applies only to vehicles used for private use, or only to premises in the United Kingdom, or only to professional services of a particular kind. Breach of such a term takes the loss outside the cover entirely; the insurer is not relying on the breach to limit liability, the loss is simply not within the insured risk.
• A risk-reducing term sits within the cover but operates to reduce the probability or severity of a covered loss — alarm warranties, sprinkler conditions, hot work permits and so on. These fall squarely within s.11(1).

The distinction is not always clean. A condition that the insured “shall only carry out hot work in accordance with a permit system” might be characterised as defining the risk (the insurer never agreed to insure hot work without a permit) or as risk-reducing (the underlying activity is insured, but the permit system reduces the risk of fire). The Court will look at the term in its policy context, including any descriptive language about the nature of the cover, the operative clause, the warranties and conditions section, and the underwriting rationale.

A useful test: if the term is removed from the policy, does the cover still describe an identifiable insured risk? If yes, the term is risk-reducing. If no, it may be risk-defining.

Case-law illustration

Direct UK appellate authority on s.11 remains relatively thin, but the principles have been considered in the Commercial Court and have been applied with a clear bias towards protecting insureds against unrelated technical defences.

The Singapore International Commercial Court decision in Bayswater Carriers Pte Ltd v QBE Asia (Singapore) Pte Ltd [2024] SGHC(I) 6 is illustrative of the causal analysis even though it sits outside the UK statutory framework. The court applied a careful counterfactual analysis to ask whether the alleged breach could have increased the risk of the actual loss in the actual circumstances. That kind of structured counterfactual is precisely the analysis a UK court will undertake under s.11(3).

UK courts have applied a similar approach in dealing with arguments about the boundary between risk-defining and risk-reducing terms, generally resisting attempts to recharacterise plainly operational warranties as risk-definers in order to escape s.11. Brokers should be prepared for the insurer to argue the s.11(4) point in any significant dispute.

Worked examples

Example 1 — Fire claim, unrelated burglary log condition. The insured’s policy contains a condition requiring it to maintain a written log of nightly perimeter security checks. The log has not been kept up to date. A fire breaks out due to an electrical fault in a machine, well away from the perimeter, during working hours. The log could not have made any difference to the cause, location or progression of the fire. Section 11 protection is likely to be available; the insured must prove that the absence of the log could not have increased the risk of this fire in these circumstances. Expert evidence on cause and origin will usually establish that.

Example 2 — Theft claim, unrelated fire-suppression warranty breach. The insured warrants that its sprinkler system will be tested monthly. A test has been missed. A burglary occurs and stock is stolen. The sprinkler test could not have prevented the burglary or reduced the loss. Section 11 protection is again likely to be available.

Example 3 — Theft claim, alarm warranty breach. The insured warrants that its intruder alarm will be set whenever the premises are unattended. The alarm was not set on the night of a burglary. The breach is plainly capable of having increased the risk of this loss in these circumstances. Section 11 will not assist. The analysis falls back to s.10 (suspension of cover during the breach period) and any other applicable defences.

Example 4 — Liability claim, undisclosed trade. The insured’s policy describes the business as “office-based IT consultancy”. The insured has, in fact, been performing on-site electrical installations. A claim arises out of the electrical work. The “office-based IT consultancy” wording is risk-defining; the loss is outside the cover. Section 11 does not assist — this is s.11(4) territory and the claim falls to be analysed instead under the duty of fair presentation and the operative clause.

Example 5 — Property claim, missing hot work permit. The insured is required by warranty to operate a hot work permit system. A welding operation has been carried out without a permit. A fire results from that welding operation. The breach directly relates to the cause of the loss; s.11 will not assist.

These examples illustrate the central point: s.11 protects against unrelated breaches, not against breaches that go to the cause of the loss. The closer the breach is to the cause, location or time of the loss, the harder it will be for the insured to discharge the s.11(3) burden.

Drafting trends since 2016

Three drafting trends are worth flagging to clients.

1. Reframing as risk-defining language. Some insurers have rewritten what were operational warranties as descriptions of the risk — for example, by inserting clauses such as “this policy insures the premises only while the intruder alarm is set”. Whether such wording succeeds in taking the term outside s.11 will depend on how it sits in the policy as a whole and whether it genuinely describes the risk or simply repackages an operational requirement.
2. Bundled warranties. Some markets now use composite warranty wording that bundles risk-defining elements with risk-reducing elements. The court will sever and analyse each element on its own merits; the broker should do the same at placement.
3. Contracting out. Section 11 is within the scope of Part 5 of the Act. In non-consumer insurance, insurers may contract out of s.11 provided the transparency requirements are met. Contracting-out wording should always be flagged to the client, explained in writing, and weighed against the price and the cover. See our article on contracting out.

Interaction with s.10 and Part 5

Section 11 does not replace s.10; the two sit alongside each other. When a warranty is breached and a loss occurs, the analysis proceeds in stages:

1. Was the loss within the period of suspension under s.10? If no, s.10 does not bite.
2. If yes, is the term within s.11(1) — i.e. risk-reducing rather than risk-defining? If yes, the insured may invoke s.11(3).
3. Has the insured discharged the burden of showing the breach could not have increased the risk of the actual loss?
4. Is there valid contracting-out wording under Part 5? If yes, the Act’s protections may have been disapplied.

A broker advising on a notified breach should walk through each of those stages in writing and keep a clear record on the file.

Practical broker checklist at placement

• Audit every term that could constitute a risk-reducing requirement: alarm warranties, sprinkler conditions, hot work permits, lone-worker policies, record-keeping requirements, key-holding arrangements, vehicle conditions.
• Identify any wording that purports to be risk-defining and ask whether it really is, or whether it is operational wording dressed up to evade s.11.
• Push back on bundling of risk-defining and risk-reducing elements in the same clause; clarity helps the insured if there is a future dispute.
• Flag contracting-out wording to the client in writing, explain its effect and consider whether it should be removed, narrowed or repriced.
• Document the policy’s protections at placement, so that if a breach is later notified the file already records the broker’s view of the s.11 position.

Practical broker checklist at notification

When a breach is notified to the broker:

• Establish exactly what the breach was, when it began, and whether it has been remedied.
• Identify whether the term is risk-reducing (s.11(1)) or risk-defining (s.11(4)).
• Identify the cause, location and time of the loss and consider whether the breach could have increased the risk of that loss in those circumstances.
• Consider what expert evidence the insured will need (fire engineering, security consultancy, forensic accounting) to discharge the s.11(3) burden.
• Put the insurer on notice in writing that the insured intends to rely on s.11 where appropriate.
• Run the s.10 analysis in parallel — has the loss occurred during a suspension period?
• Document the analysis on the file; the broker’s contemporaneous record may itself become an exhibit in any later coverage dispute.

FAQs

1. What does s.11 of the Insurance Act 2015 protect against? It prevents insurers from relying on breach of a risk-reducing term to refuse a claim where the breach could not have increased the risk of the loss that actually occurred. It is aimed at unrelated technical defences.

2. Which terms does s.11 apply to? Terms (other than those defining the risk as a whole) compliance with which would tend to reduce the risk of loss of a particular kind, at a particular location, or at a particular time. That captures most operational warranties, conditions and conditions precedent.

3. Who bears the burden of proof under s.11(3)? The insured. The insured must show that non-compliance with the term could not have increased the risk of the loss which actually occurred in the circumstances in which it occurred.

4. Can insurers avoid s.11 by relabelling warranties? Not by labels alone. The court will look at the substance of the term in its policy context. However, insurers can legitimately draft terms as risk-defining where the substance supports that characterisation, and the s.11(4) carve-out will then apply.

5. Can the parties contract out of s.11? Yes, in non-consumer insurance only, and only if the transparency requirements of Part 5 are satisfied. Contracting-out should always be flagged and explained to the client.

6. Does s.11 apply to conditions precedent as well as warranties? Yes. The section applies to any term within s.11(1), regardless of how it is labelled in the policy.

7. Does s.11 help if the breach did cause the loss? No. Section 11 only assists where the breach could not have increased the risk of the actual loss. Breaches that go to cause, location or time of the loss fall outside the protection.

8. How does s.11 interact with s.10? Section 10 controls the mechanism of warranty breach (suspension of cover); s.11 controls the relevance of breach to a particular loss. Both should be analysed in any case where a warranty has been broken and a loss has occurred.

Related articles in this series

1. The Insurance Act 2015 — a broker’s overview
2. The duty of fair presentation
3. The material circumstance test
4. Warranties reform under the Insurance Act 2015
5. Terms not relevant to the actual loss — s.11 (this article)
6. Remedies for breach of the duty of fair presentation
7. Contracting out under Part 5
8. Condition precedent vs warranty post-Insurance Act
9. Late payment of claims under the Enterprise Act 2016
10. The impact of the Insurance Act 2015 on PI claims

---

Reviewed by the Apex Insurance Brokers technical team, May 2026. Apex Insurance Brokers Ltd is authorised and regulated by the Financial Conduct Authority (firm reference 724952) and is registered in England and Wales (Companies House 07014570). This article is general guidance and is not legal advice; insureds should take advice on specific facts.