Category: Specialist underwriting · Reviewed by Al Jabbar, Broker · Specialist Risks · Last reviewed May 2026
A consulting engineer practice of around 35 staff renewed its professional indemnity (PI) cover in early 2025 having signed off the proposal form in the usual way. Buried in the policy schedule was a condition precedent requiring the firm to disclose, at each renewal, any circumstances coming to its attention during the policy period which might give rise to a claim. Six months into the new policy period a partner remembered a heated exchange with a former client at the previous renewal date which had, in retrospect, all the hallmarks of a circumstance. The firm had not disclosed it. Was cover for any resulting claim now lost? Under the pre-2016 regime the answer would have been straightforward and unforgiving. Under the Insurance Act 2015 the analysis is considerably more nuanced — and a competent broker should be able to walk a partner through it without reaching for a textbook. What follows sets out that framework.
For more than a century before the 2015 Act, the law of insurance warranties was governed largely by the Marine Insurance Act 1906 as applied (sometimes uncomfortably) to non-marine business. A warranty was a strict condition. If breached, cover was automatically discharged from the moment of breach. The insurer did not need to show the breach caused the loss. The breach did not need to be material. Once a warranty was broken, the policy was effectively void from that point — and where the warranty was a “basis of the contract” clause, the breach could void cover from inception.
This produced harsh results. A standard proposal form might contain dozens of statements which the proposer “warranted” to be true. A trivial inaccuracy — a typing error in the address, an outdated employee count, an honest misremembering of a five-year-old claim — could be deployed by an insurer to walk away from an entirely unrelated loss. The Law Commission’s reform work, building on earlier reports, identified this as one of the principal injustices of UK insurance law.
The Insurance Act 2015 came into force on 12 August 2016 and applies to non-consumer insurance contracts and to variations of those contracts on or after that date. For PI specifically, four sections of the Act bear directly on warranties and conditions precedent:
These four sections together rewrite the practical operation of warranties and conditions precedent. The strict regime has not gone — warranties remain strict in the sense that fault on the insurer’s side is not required to enforce them — but the remedies and the carve-outs have changed materially.
Under section 10, breach of warranty no longer voids cover. It suspends it. The insurer is not on risk while the breach subsists. The moment the breach is remedied — for example, by reinstating a security measure that had been allowed to lapse — cover resumes. Crucially, losses occurring before the breach, and losses occurring after it is remedied, are unaffected.
This is a markedly different proposition from the pre-Act position. It means brokers and insureds must think about warranties as conditions to be monitored and remediated, not as one-time tripwires. It also means that disputes will increasingly turn on questions of when a breach occurred, when it was remedied, and whether the relevant loss occurred during the suspension window.
Section 11 then layers on a further protection. Even if a breach exists and a loss occurs during the suspension window, the insurer cannot rely on the breach to refuse the claim if the term was designed to reduce a particular kind of risk and the insured can show the breach could not have increased the risk of the loss which actually occurred. Take an example: a warranty that a firm maintains a written conflict-checking procedure. A loss arises from a cyber breach unrelated to conflicts. The warranty, being directed at the risk of conflicted instructions and not at cyber risk, falls within section 11. The insurer should not be able to defeat the cyber claim by reference to a conflict-checking lapse.
The burden of proof on section 11 sits with the insured. The insured must show that the term related to a particular risk and that the breach could not have increased the risk of the actual loss. This is not always easy. But it transforms what was a binary outcome under the old law into a substantive argument about risk relevance.
Conditions precedent (CPs) operate differently from warranties. A warranty goes to the policy as a whole; breach (subject to sections 10 and 11) suspends cover until remedied. A condition precedent typically goes to a specific obligation — usually the duty to notify, the duty to provide information, the duty to cooperate. Breach of a CP gives the insurer the right to decline the specific claim affected by the breach, without disturbing cover under the policy more broadly.
The section 11 carve-out applies equally to CPs. Where the condition precedent is directed at a particular kind of risk and the breach could not have increased the risk of the loss which actually occurred, the insurer cannot rely on the breach. The architectural distinction between warranty and CP remains, but the analytical framework for relief is shared.
Common CPs in PI policies include:
Each of these is a discrete obligation, and breach of any of them potentially gives the insurer a defence to a specific claim — subject always to the section 11 analysis.
The condition precedent prohibiting settlement or admission without insurer consent is among the most commercially sensitive provisions in any PI policy. It is what gives the insurer control of the defence. But it is paired in most PI wordings with a “QC clause” (or its modern variant, sometimes called a senior-counsel clause) which provides a mechanism for resolving disputes between insurer and insured about whether a claim should be contested or settled.
The classic QC clause provides that, where insurer and insured disagree on whether to fight or settle, the matter is referred to a King’s Counsel agreed between them, whose opinion binds the question of whether the claim should be contested. If counsel advises that a defence has reasonable prospects, the insurer must indemnify continued defence costs. If counsel advises against contest, the insurer is entitled to settle within the relevant authority.
Wording variants matter here. Some clauses bind only the insured to counsel’s opinion; some bind only the insurer; some bind both. Some require a particular eminence of counsel; some are silent. Brokers reviewing PI wordings for senior decision-makers should treat the QC clause as a discrete object of review rather than a piece of boilerplate.
Not every obligation in a PI policy is a warranty or a condition precedent. A clause may be drafted as either, or as neither — a “mere term”, breach of which gives the insurer a right to damages (typically nominal) but no right to refuse a claim. The labelling matters.
Three drafting variants commonly appear:
Where a clause does not, on its true construction, qualify as a warranty or CP, the insurer’s remedy for breach is typically damages, not refusal of cover. That is a profoundly different commercial outcome. A PI broker reviewing a policy should be able to identify each obligation in the wording and classify it. Where the drafting is ambiguous, the insured may benefit from that ambiguity, but it is better to negotiate clarity at placement than to litigate it after a loss.
The principles underpinning contractual construction in this area, as restated in cases such as Wood v Capita Insurance Services Ltd [2017] UKSC 24 and in earlier insurance-specific authority, emphasise the natural meaning of words in their commercial context. Insurance clauses, like other commercial terms, are read in context, not as isolated phrases.
Section 16 of the Insurance Act permits insurers to contract out of the default regime for non-consumer business. But the freedom is constrained. To rely on a contracting-out term that places the insured in a worse position than under the Act, the insurer must satisfy the “transparency requirements” under section 17:
In practice, courts have shown a willingness to scrutinise contracting-out clauses closely. A clause buried in a schedule, or expressed in dense and ambiguous language, is unlikely to meet the transparency requirement. Insurers seeking to displace section 11, in particular, must do so by clear and prominent drafting.
A broker reviewing a PI wording should identify any contracting-out language, assess whether it meets section 16/17 requirements, and discuss the commercial implications with the insured. Section 11 protection is one of the more valuable features of the Act for the insured; clauses purporting to remove it should not be accepted without consideration.
Although the topic is adjacent to warranties and CPs, it is worth noting the Supreme Court’s decision in Versloot Dredging BV v HDI Gerling Industrie Versicherung AG [2016] UKSC 45, which considered the use of “fraudulent devices” — lies told in support of a genuine claim. The court held by majority that telling a lie in support of a genuine claim does not entitle the insurer to forfeit the entire claim, distinguishing between fraudulent claims (which remain forfeit) and fraudulent devices in support of otherwise genuine claims.
For PI practice the case is a reminder that strict insurance-law rules have been moderated by both legislation and judicial decision over recent years. The combination of the Insurance Act 2015 and Versloot Dredging reflects a wider rebalancing in favour of policyholders where strict rules produced disproportionate outcomes — though the Enterprise Act 2016, which amended the Insurance Act to introduce remedies for late payment of claims, sits within the same direction of travel.
Three patterns of disputed CP issues recur in PI practice:
The late-notified circumstance. The firm becomes aware of a potential issue, does not notify the insurer at the time, and notifies later. The insurer asserts breach of the prior-notification CP. The analysis turns on whether the firm was “aware” within the meaning of the policy, when “as soon as practicable” started running, and — under section 11 — whether the delay could have increased the risk of the loss which actually occurred.
The consent breach. The firm makes an offer of settlement to a claimant without first obtaining insurer consent. The insurer asserts breach of the consent CP. The analysis turns on whether the communication amounted to an “offer” or merely a discussion, whether the insurer was contemporaneously informed in substance, and again under section 11 whether the absence of formal consent could have increased the risk of the loss.
The cooperation breach. The firm fails to provide information requested by the insurer or fails to attend interviews. The analysis turns on the reasonableness of the request, the materiality of the information sought, and the section 11 question of whether non-cooperation could have increased the risk of the actual loss.
In each scenario, the Insurance Act has transformed what would have been a strict insurer victory under pre-2016 law into a substantive argument. Insureds do not always win these arguments. But they have arguments to make, and the arguments turn on facts and risk-relevance rather than mere proof of technical breach.
When reviewing a PI policy at placement or renewal, the following exercise is worth doing properly:
This is not a five-minute exercise. But for any firm with a meaningful PI exposure, it is the foundation on which a defensible claims position is built.
In one sense yes — the insurer does not need to show fault on the insured’s part to enforce a warranty. In another sense no — section 10 means breach suspends rather than discharges cover, and section 11 carves out terms not relevant to the actual loss. The strict character of the warranty has been retained in form but materially softened in operation.
A warranty goes to the policy as a whole; breach suspends cover until remedied. A condition precedent typically goes to a specific obligation in relation to a specific claim; breach gives the insurer a defence to that claim without affecting the rest of the policy. Both are subject to the section 11 carve-out for terms not relevant to the actual loss.
The provision applies where a term is designed to reduce the risk of a particular kind of loss, a particular cause of loss, or a loss at a particular time or place. If the insured can show that the breach could not have increased the risk of the loss which actually occurred in the circumstances in which it occurred, the insurer cannot rely on the breach to refuse the claim. The burden of proof sits with the insured.
No. Section 9 of the Insurance Act 2015 abolished basis clauses for non-consumer insurance. Representations made in connection with a proposed insurance contract cannot be converted into warranties by any term of the contract. Any wording that attempts to do so is of no effect to that extent.
A QC clause (or senior-counsel clause) provides a dispute-resolution mechanism within a PI policy: where the insurer and insured disagree on whether to fight or settle a claim, the matter is referred to King’s Counsel agreed between them. Counsel’s opinion binds the relevant question. The clause is the practical limit on the insurer’s settlement authority and the insured’s exposure to a forced settlement.
In principle yes, under section 16, but only if they satisfy the section 17 transparency requirements. The disadvantageous term must be sufficiently drawn to the insured’s attention, must be clear and unambiguous as to its effect, and the character of the insured and the circumstances of the transaction are relevant to whether transparency has been achieved. Courts have shown willingness to scrutinise such clauses closely.
It bears on the treatment of fraudulent devices — lies told in support of an otherwise genuine claim. The court held that such conduct does not automatically forfeit the genuine claim, distinguishing it from a wholly fraudulent claim. In PI practice the relevance is indirect but illustrative of the wider judicial shift away from disproportionately strict outcomes in insurance law.
The practical answer is to maintain a register of every warranty and CP in the policy, with the operational owner for compliance, the monitoring frequency, and the trigger for escalation. This is rarely done in practice but is the single most useful internal control a firm can maintain. A broker should be able to produce a draft register on request.
Apex Insurance Brokers Ltd is an independent UK insurance broker based in Bristol, advising professional services firms on professional indemnity insurance and related covers. The firm is authorised and regulated by the Financial Conduct Authority (firm reference 724952) and registered at Companies House (company number 07014570).
This commentary reflects market conditions as at May 2026 and is provided for general information. Insurance market conditions, policy wordings and regulatory positions change frequently; firms should obtain advice specific to their circumstances rather than rely on general commentary.
Apex Insurance Brokers serves UK professional services firms and commercial businesses. Call 0117 325 0027, email hello@apexinsurancebrokers.co.uk, or request a quotation.
Get a quote