Citable answer: Concurrent delay is a period of project overrun caused by two or more effective causes of delay — one the employer's risk, one the contractor's — of approximately equal causative potency. Under the usual English law approach, the contractor gets the extension of time for that period but generally cannot recover its prolongation costs for it.
Almost every serious delay dispute reaches the same moment: the contractor points at the late design information, the employer points at the slow progress that was happening anyway, and both are right. What happens next is decided by three things — how the contract defines the entitlement, which legal approach governs it, and whose records can actually prove what drove the critical path. Here is the ground, plainly.
What concurrent delay actually is — and is not
The working definition in English law comes from John Marrin KC, in a paper for the Society of Construction Law: concurrent delay is "a period of project overrun which is caused by two or more effective causes of delay which are of approximately equal causative potency." The Commercial Court adopted that definition as a useful working test in Adyard Abu Dhabi v SD Marine Services [2011] EWHC 848 (Comm), and it has anchored the English cases since (Pinsent Masons).
Two words in it carry the weight:
- Effective. Each event must actually be driving Delay to Completion — affecting the critical path — not merely existing at the same time. A contractor delay on a non-critical activity that happens to overlap with a critical employer delay is not concurrency; it is background noise.
- Equal causative potency. Where one cause would have delayed completion anyway and the other would not, there is no true concurrency — there is one operative cause and a bystander.
The SCL Delay and Disruption Protocol (2nd edition, February 2017) is blunter still: true concurrent delay — two events of equal potency starting and biting at the same time — is rare. Most alleged concurrency is sequential or overlapping delay dressed up, which is why the analysis usually collapses into a fight about facts and programme logic rather than law (SCL Protocol, 2nd ed.).
The English default: time yes, money no
Where genuine concurrency is established, English law's starting position was set in Henry Boot Construction (UK) Ltd v Malmaison Hotel (Manchester) Ltd (1999) 70 Con LR 32 — the "Malmaison approach." If one of the concurrent causes is an employer risk event, the contractor receives a full extension of time for the period of delay it causes, even though the contractor's own delay was running alongside it (CMS).
The Technology and Construction Court confirmed the approach in Walter Lilly & Co Ltd v Mackay [2012] EWHC 1773 (TCC): where two or more events cause delay and one is a relevant event under the contract, the contractor is entitled to the full extension — and English law does not apportion the time between the parties. That is a deliberate divergence from Scotland, where City Inn Ltd v Shepherd Construction Ltd [2010] CSIH 68 allows the decision-maker to apportion delay between competing causes (Simmons & Simmons).
Money is the other half, and it runs the other way. The SCL Protocol's position — widely followed in practice — is that an extension of time does not automatically carry prolongation costs with it. For the concurrent period, the contractor would have been in delay at its own cost anyway, so it recovers compensation only to the extent it can separate out loss actually caused by the employer delay (SCL Protocol, 2nd ed.).
That asymmetry — time yes, money no — is the single most useful sentence in this topic. The extension protects the contractor from delay damages; the concurrency defence protects the employer from paying twice for a period the contractor was late on its own account.
Your contract can change all of it
None of the above is mandatory law. It is the default that applies when the contract is silent — and increasingly, the contract is not silent.
In North Midland Building Ltd v Cyden Homes Ltd [2018] EWCA Civ 1744, the parties had amended a JCT Design and Build Contract so that any delay concurrent with a contractor-caused delay would not count towards the extension of time. The contractor argued the clause offended the prevention principle — the doctrine that an employer cannot hold a contractor to a date the employer itself prevented it from meeting. The Court of Appeal disagreed: the prevention principle is not an overriding rule of law, and parties are free to allocate the risk of concurrent delay to the contractor by clear words (Herbert Smith Freehills Kramer).
The practical consequence is sharp. Since North Midland, concurrency-exclusion clauses have become a standard employer amendment. A contractor that prices a heavily amended contract on the assumption that Malmaison will save it may find its extension entitlement was signed away on day one. The first question on any project is therefore not "what does the law say about concurrency?" but "what does this contract say?" — the same discipline that applies to notice periods and time bars, which amendments routinely shorten.
How FIDIC 2017 and NEC4 handle it
FIDIC 2017 confronts concurrency directly — by declining to decide it. The final paragraph of Sub-Clause 8.5 provides that where a delay caused by the employer is concurrent with a delay caused by the contractor, the contractor's entitlement to extension of time is assessed "in accordance with the rules and procedures stated in the Special Provisions" — or, if none are stated, "as appropriate taking due regard of all relevant circumstances." FIDIC's own guidance points parties towards the SCL Protocol as the reference increasingly adopted internationally (Fenwick Elliott, Jeremy Glover). In other words: if your Particular Conditions are silent, you have an open-textured test and an argument waiting to happen. If they are not silent, the drafted rule — whatever it is — governs.
NEC4 contains no express concurrent delay provision at all. Compensation events are assessed prospectively under clause 63 against the Accepted Programme as it stood at the dividing date — a forecast of the event's effect on planned Completion, rather than a retrospective inquest into what actually caused the overrun. That structure changes how concurrency plays out in practice: the assessment asks what the compensation event would do to the programme, and employer amendments adding an express concurrency clause are common for exactly that reason. The contrast with FIDIC's regime is part of a broader difference in philosophy we cover in FIDIC notice vs NEC4 early warning.
Concurrency disputes are won on records, not rhetoric
Look back at the definition: effective causes, equal causative potency, critical path. Every element is a question of fact, and the tribunal answers it from the documents that existed at the time — the accepted and updated programmes, progress records, site diaries, correspondence dating each delay's start and bite. The SCL Protocol's first core principle is that programmes should be updated and records kept precisely so that delay can be analysed properly when it happens, not reconstructed years later (SCL Protocol, 2nd ed.).
In practice, the party that can show which activity drove the critical path in which week usually wins the concurrency argument — because it can demonstrate that the other side's "concurrent" event had no causative potency at all. The party with a stale baseline and a gap in its progress records is left asserting concurrency it cannot prove, or defending against concurrency it cannot disprove. And the argument starts earlier than most teams think: the moment a programme slips is the moment the contractual machinery engages, as we set out in when a programme slip becomes a contractual notice.
Where a governed AI layer helps
Concurrency is decided by humans — a contract administrator, an adjudicator, a tribunal. What a contract-aware system can do is make sure the humans arrive with the record intact. A layer that has read your particular contract — including the concurrency clause the amendments added — knows whether your project runs on Malmaison's default or on words that displaced it. It can watch the live programme against the baseline, date each slip as it emerges, connect the slip to the notice the contract requires, and keep the chain of programmes, notices and records aligned so causation can be demonstrated rather than asserted. It flags and drafts, with every flag cited back to the clause; the people still decide what to claim and what to concede. The machine's contribution is narrower: when the concurrency argument comes, nobody is reconstructing the critical path from memory.
This is general information on how concurrent delay is treated, not legal advice. Approaches differ by contract, amendment and governing law — the sources below are cited so you can verify, and you should take advice on your own contract.
Sources & further reading
- Society of Construction Law — Delay and Disruption Protocol, 2nd edition (February 2017). The definition of true concurrent delay, the extension-of-time and compensation principles, and the record-keeping core principles: https://www.scl.org.uk/sites/default/files/documents/SCL_Delay_Protocol_2nd_Edition_Final.pdf
- Pinsent Masons (Out-Law) — Concurrent delay in UK construction projects. The Marrin definition and its adoption in Adyard Abu Dhabi v SD Marine Services [2011] EWHC 848 (Comm), and the consequences of concurrency for time and money: https://www.pinsentmasons.com/out-law/guides/consequences-of-concurrent-delay-on-construction-projects
- CMS — Concurrent delay: the English approach. The Malmaison approach from Henry Boot v Malmaison (1999) 70 Con LR 32 and the English cases applying it: https://cms.law/en/gbr/legal-updates/concurrent-delay-the-english-approach
- Simmons & Simmons — Concurrent Delay in Construction Contracts. Walter Lilly v Mackay [2012] EWHC 1773 (TCC), the rejection of apportionment in England, and the Scottish contrast in City Inn v Shepherd Construction [2010] CSIH 68: https://www.simmons-simmons.com/publications/ck0bazxjoo1b40b33yz7j85k9/100818-concurrent-delay-in-construction-contracts
- Herbert Smith Freehills Kramer — The prevention principle is not an overriding principle of law. North Midland Building v Cyden Homes [2018] EWCA Civ 1744 and the validity of concurrency-exclusion clauses: https://www.hsfkramer.com/notes/construction/2018-08/the-prevention-principle-is-not-an-overriding-principle-of-law-express-terms-allocating-risk-of-concurrent-delay-still-prevail
- Fenwick Elliott (Jeremy Glover) — Some thoughts on how the 2017 FIDIC contract deals with time. Sub-Clause 8.5, the Special Provisions route for concurrency, and FIDIC's reference to the SCL Protocol: https://www.fenwickelliott.com/sites/default/files/jg_-_how_the_2017_fidic_contract_deals_with_time.pdf