The scope-of-duty question — how much of a loss is legally recoverable from a negligent professional — sits at the heart of most engineer PI claims. Since the Supreme Court's twin judgments in Manchester Building Society v Grant Thornton UK LLP [2021] UKSC 20 and Khan v Meadows [2021] UKSC 21, the analytical framework has been reformulated. For civil, structural, mechanical and electrical engineers, whose calculations and specifications are relied on by contractors, architects, and ultimately building owners, understanding how the counterfactual test now operates is essential to reading a PI claim correctly.
An engineer's contribution to a construction project is unusually concentrated in a few discrete outputs: structural calculations, member schedules and specifications, mechanical and electrical services layouts, load assessments, and on-site technical advice during construction. Each output is relied on by another party — the contractor building to the specification, the architect coordinating the design, or the employer signing off variations. When something later fails, the engineer's alleged breach is rarely diffuse. It is typically a specific technical error in a specific document.
Historically that discreteness pushed engineer claims towards the older South Australia Asset Management Corp v York Montague Ltd [1997] AC 191 (SAAMCO) framework. If the engineer was providing information rather than advice on the whole transaction, recoverable loss was limited to the consequences of the information being wrong — not every loss flowing from the decision it informed.
In George Fischer Holding Ltd v Multi Design Consultants Ltd (1998) 61 Con LR 85, a structural engineer was held liable for the consequences of specifying a deficient roof structure. The court examined which losses fell within the scope of duty and which were consequential to the client's wider commercial decisions. Similar reasoning appears throughout the design-and-build case law traceable back to Independent Broadcasting Authority v EMI Electronics Ltd (1980) 14 BLR 1, where the House of Lords confirmed that a designer of a specialist structure owes a duty to design something fit for its purpose — but the recoverable measure still depends on what the designer's task actually was.
In Manchester Building Society and Khan v Meadows, the Supreme Court replaced the older information/advice dichotomy with a six-stage analysis. The critical stage for engineers is the counterfactual: what would the client (or the contractor, or the architect coordinating the design) have done differently had the engineer discharged the duty properly? The answer defines the loss that falls within scope.
For an engineer, the counterfactual is often technically precise. If the correct calculation would have produced a different specification, the counterfactual is that the contractor would have built to that different specification — usually with a modest incremental cost. The recoverable loss is the gap between the outcome that occurred and the outcome that would have occurred.
Where an engineer is engaged by a design-and-build contractor, the contractor typically warrants fitness for purpose to the employer. The engineer's own duty to the contractor may still be one of reasonable skill and care under the sub-consultancy appointment. When the contractor is sued on the fitness warranty and passes the claim down the chain, the engineer's scope of duty is examined against the sub-consultancy terms — not the head contract. This split has been material in several recent construction PI disputes and reinforces the value of collateral warranties being drawn narrowly.
Worked example. A structural engineer specifies a steel section in 2020 for a warehouse extension. The contractor builds accordingly. In 2023 a partial failure occurs. Investigation confirms the specified section was undersized for the actual load path. Remedial works cost £900,000. The occupier's business interruption during the reinstatement adds £300,000. Total loss: £1.2 million.
Applying the counterfactual: had the engineer specified the correct section in 2020, the contractor would have used a larger steel member at an additional build cost of £40,000. The engineer's scope of duty covers the difference between the outcome that occurred and the outcome that would have occurred with a correct specification. The recoverable loss is £1.2 million less £40,000 — that is, £1.16 million. The engineer's PI policy responds to that measure, subject to the policy limit, the excess, and any relevant aggregation provisions.
Three points bear emphasis. First, the counterfactual has to be capable of proof — expert evidence on what a competent engineer would have specified is central. Second, the causation chain is not automatically broken by an intervening contractor or architect; the question is whether the engineer's error remained an effective cause. Third, engineers should keep contemporaneous records of the assumptions underlying calculations, because those assumptions frame the scope of the duty owed.
For related concepts, see the Apex guide to engineers' PI insurance, the wiki entries on advice versus information in scope of duty and standard of care versus strict liability, and the Apex guide to architects' PI insurance for the coordinating-designer perspective.
Apex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Firm reference number 724952. This entry is general information, not advice on any particular policy.