Penalties in the Paperwork: Why UK MGAs Keep Misreading Contractual Intangibles as Physical Loss

There is a particular kind of claim that arrives at an MGA looking familiar and turns out not to be. The project was late, the plant underperformed, the service fell below the agreed standard — and the insured's loss is not the cost of putting something right, but the sum a counterparty is now entitled to deduct under a contract. The paperwork looks like a physical damage claim. The mechanics are entirely different, and so is the wording needed to respond to it.

The loss that is a number in a contract

Physical damage has the virtue of being verifiable. A contractor's exposure under a liquidated damages clause does not: it exists because two commercial parties agreed, at the outset, that delay would be compensated at a fixed rate per day. Nothing has been damaged. Something has been promised, and the promise has not been met.

The difficulty for underwriters is that these exposures do not announce themselves. They are embedded in contract terms that sit outside the policy, in employer's requirements, in schedules of rates and in service level agreements that may never be seen by the underwriter at all. The insured does not necessarily regard them as insurable, so they are not disclosed. The broker may not regard them as material, so they are not flagged. And yet when a project runs into difficulty, they are frequently the largest number in the conversation.

Four exposures that travel together

Liquidated damages. A pre-agreed sum payable for delay, most often encountered in construction and engineering. The commercial logic is to avoid the need to prove loss; the underwriting consequence is that liability can crystallise without any damage occurring at all.

Performance penalties. Common where plant, equipment or systems must meet specified output, efficiency or availability thresholds. The penalty is triggered by measured underperformance, which may be marginal, disputed and technically complex.

Service credits. The recurring mechanism in outsourced and technology arrangements, where the remedy for missed service levels is a financial credit against fees. Individually modest, cumulatively significant, and often capable of escalating over a contract term.

Termination payments. The tail risk. Where a breach or persistent underperformance entitles a customer to terminate, the resulting liability can dwarf the value of the contract itself, particularly where transition, repatriation or replacement costs are recoverable.

These four travel together because they share a root cause: a contractual allocation of consequence that operates independently of physical loss. An underwriter who addresses one without the others has usually addressed none of them.

Why construction and engineering keep the door open

UK construction contracts are unpromising territory for underwriters seeking certainty. Under the JCT suite, liquidated damages provisions are standard; under NEC4, secondary option X7 performs the same function with a different drafting discipline, including a requirement that the rate be a genuine pre-estimate of loss. The point of legal interest — whether a clause is a penalty and therefore unenforceable — is precisely the point that makes these exposures awkward to underwrite: the insured may be contractually bound to a sum that is later contested, producing a claim that is really a dispute with a lengthy runway.

Engineering policies have historically been drafted around damage to the works and to third-party property. Extensions for delay in start-up or advanced loss of profits go some way towards the consequential gap, but they are usually written around an insured event rather than around a contractual promise. The result is that a contractor facing liquidated damages after a subcontractor failure may find the exposure sits between the covers.

There is also the professional indemnity interface. Many construction professionals now provide design and advisory services under appointments that carry their own liability caps, net contribution clauses and fitness-for-purpose obligations. A fitness-for-purpose promise is, in substance, a guarantee, and it can generate liability without any allegation of negligence.

Professional lines: the duty that quietly includes a promise

The same pattern appears in professional lines, particularly in technology, outsourcing and consultancy. A service provider that agrees to a set of service levels and a credit mechanism has accepted an exposure that is contractual rather than tortious. If the customer later alleges that the provider failed to exercise reasonable skill and care in designing the service such that the levels could never be met, the claim is framed as negligence — but the measure of loss is the contract.

This matters for wording. Where cover is limited to civil liability for breach of professional duty, the question of whether a contractual penalty is a loss flowing from that breach, and whether it is a direct or consequential loss, will determine whether the policy responds at all. UK courts have not been generous to attempts to recover sums that look like agreed remedies rather than proven damage, and the resulting disputes are as likely to be about coverage as about liability.

Quantifying what is not yet a claim

Quantification is where the discipline either holds or collapses. Three practical approaches are worth applying at the point of underwriting.

First, obtain the contract. Not a summary, not the appointment letter — the operative clauses dealing with delay, performance, service levels, liability caps, net contribution and termination. If the insured will not provide them, that is itself an underwriting signal.

Second, compute a maximum contractual exposure. Take the liquidated damages rate, multiply by the longest credible delay, net of any cap, and compare that with the policy limit. Do the same for service credits over the full term. The exercise is crude and it is far better than no number at all.

Third, identify the trigger mechanism. Does the exposure arise automatically on a measured shortfall, or only on a determination by a certifier, adjudicator or court? Automatic triggers accumulate faster and are harder to dispute.

Wording the perimeter

Several drafting choices will materially change the nature of the cover.

• Treat contractual penalties, liquidated damages and service credits as expressly excluded unless cover is intended, rather than leaving them to be argued through the definition of loss.
• Where the insured requires cover, define the sub-limit by reference to the specific contractual mechanism, and set it well below the limit for physical damage.
• Address the relationship with any delay in start-up or advanced loss of profits extension explicitly, so that the two do not double up.
• Consider whether defence costs for contractual disputes should sit inside or outside the limit; disputes of this kind can consume a sub-limit in legal fees alone.
• Require notification of material contract variations, since it is the variation — not the original contract — that most often moves the exposure.

The underwriting point

Contractual intangibles are not exotic. They are in the paperwork of a large proportion of the commercial arrangements that UK MGAs underwrite, and they are frequently the largest single exposure on a project that presents as a property or engineering risk. The failure is not one of imagination; it is one of documentation. Underwriters see the schedule of insurance and not the contract, and the contract is where the liability lives.

For the 2026 cycle, the practical ask is modest: read the operative clauses, compute the maximum contractual number, and say in the file what the policy does with it. Where that number cannot be quantified because the contract is unavailable, the honest position is a restrictive one. It is far cheaper to narrow cover at inception than to argue about the definition of loss after the liquidated damages clock has started running.