Delay claims, particularly in major construction and engineering projects, are usually factually and legally complex. Contractors submit hefty claims, supported by lengthy claims documents and a delay or programming analysis. For such claims to have the best chance of persuading the employer (or a court or arbitrator) of their merits, they need to reflect both what actually happened on site and the way the contract requires that extensions of time are dealt with.

In Carillion Construction Ltd v Emcor Engineering Services Ltd [2017] EWCA Civ 65 (10 February 2017), the Court of Appeal has recently considered whether the ‘orthodox’ or conventional approach to extending time can really be right when it can lead to some anomalous results which (the contractor argued) offended commercial common sense. In this article we consider whether there may be some wider implications to this most recent approach.

Delays in Court

Carillion was the main contractor for the redevelopment of the High Court’s Rolls Building, which houses both the Commercial Court and the Technology and Construction Court. Emcor was Carillion’s subcontractor for mechanical and electrical works. Carillion’s main contract was signed in June 2007, and required the work to be completed by early 2011. Specifically, the main contract provided for sectional completion of two packages (including the court fit-out) by 28 January 2011. Emcor’s subcontract reflected that completion date for the relevant subcontract works.

Completion was delayed by 182 days, until 29 July 2011. As a result, the parties got to spend some more time in the Rolls Building as they litigated over who was responsible for the delay, and in particular how any extension of time for Emcor should be computed. In a nutshell, the competing arguments were that:

- either, the period for which Emcor was entitled to an extension of time would simply be added to the completion date (a ‘contiguous’ extension), which is what Emcor said should happen; or

- as Carillion argued, further specific periods were to be fixed during which the works could be carried out, but these periods were not necessarily going to be contiguous. Instead, these periods would reflect when the contractor was actually delayed by the relevant events.

Carillion argued for the second approach because awarding non-contiguous extensions would lead to a fair and commercially sensible result, while a ‘contiguous’ approach might end up relieving the subcontractor of liability for delay in circumstances where the work was in fact delayed because of something for which the subcontractor was responsible.

Contiguous or non-contiguous?

Carillion illustrated this with an example. Assume that subcontract works are meant to be completed within 100 days, coinciding with the agreed completion date of the main contract works. The subcontract works are, however, delayed in such a way that completion under the main contract is also delayed. This is all the subcontractor’s fault. At this point, the employer would be entitled to liquidated damages against the main contractor, and the main contractor would also be incurring prolongation costs. One would expect that all this loss could be passed down (at least to some extent) by the main contractor to the subcontractor.

On day 150, the main contractor then instructs the subcontractor to carry out a significant variation. These new works should take 50 days. Obviously, the subcontractor is entitled to an extension of time. If a ‘contiguous’ extension of time is granted (adding 50 days to the original completion date so that it becomes day 150), that would mean that the subcontractor would not be in breach of contract on day 150 - even though there has in fact been culpable delay of 50 days. If the completion date were simply extended in that manner without having regard to the actual progress (or lack thereof) of the works, the contractor (as employer to the subcontractor) might lose any entitlement to liquidated damages. It would then be necessary to wait and see when the subcontract works would be completed - perhaps by day 200 - and then determine what the cause of critical delay for the overrun period was.

Instead of awarding a contiguous extension of time, Carillion argued that the subcontractor in the example should be absolved from any liability for delay only from day 150 to day 200, on the basis that the additional works would require an additional 50 days of work. That would reflect what was actually happening on site. Awarding a contiguous extension of time could have the unsatisfactory result of putting the subcontractor in default just as it would embark on additional work, whilst ignoring responsibility for prior delays in completing the original scope of work.

The extension of time clause in the subcontract

In the Court of Appeal, Jackson LJ described Carillion’s argument as novel and indeed unprecedented, never having come across it at the Bar or during his time on the bench. His Lordship also considered a number of prior cases dealing with extensions of time (well known to construction lawyers, including Henry Boot v Malmaison and Walter Lilly v Mackay), and found that these all supported the award of a contiguous extension of time, based on the actual delay caused by the relevant event. In none of these cases had the Courts been asked to determine a period of culpable delay followed by an extension of time starting on a date other than the original completion date.

Ultimately, however, none of these cases were directly relevant as the question was one of contractual interpretation: was there any support in the wording of the relevant extension of time clause for the non-contiguous approach? Emcor had been engaged under a standard form construction contract for domestic use, the JCT’s “DOM/2” agreement.

Clause 11.3, the relevant extension of time provision, stated:

“11.3 If on receipt of any notice, particulars and estimate under clause 11.2 the Contractor properly considers that:

      1. any of the causes of the delay is an act, omission or default of the Contractor, his servants or agents or his sub-contractors, their servants or agents (other than the Sub-Contractor, his servants or agents) or is the occurrence of a Relevant Event; and
      2. the completion of the Sub-Contract Works is likely to be delayed thereby beyond the period or periods stated in the Appendix, part 4, or any revised such period or periods,

then the Contractor shall, in writing, give an extension of time to the Sub-Contractor by fixing such revised or further revised period or periods for the completion of the Sub-Contract Works as the Contractor then estimates to be reasonable.”

Jackson LJ started by considering the natural meaning of these words. He felt that the expression ‘extension of time’ simply connoted that the period for completing the work would be made longer. References to ‘revised periods’ had the same meaning, since they would be read in that context. Other parts of the contract referred to the “… expiry of an extended period previously fixed under Clause 11.” This suggested that by granting an extension of time, the employer would have extended the date for completing the work, rather than providing for (potentially several) periods within which the work, or some of it, could be done. All this supported the contiguous approach.

Commercial common sense

Carillion had argued that Jackson LJ’s reading of the contract did not make commercial common sense, and should not therefore be adopted. In dealing with this argument, the Court of Appeal restated the principles of contractual interpretation, whilst making some interesting comments on the relevance of ‘commercial common sense.’

Everyone agreed that the latest principles on contractual interpretation were set out in the decision of the Supreme Court in Arnold v Britton [2015] UKSC 36. In Arnold, the Supreme Court stressed that commercial common sense could not be invoked retrospectively, and could only be relevant for something that the parties could have foreseen at the time of entering into the contract. In other words, it is not open to a party to advance an argument based on commercial common sense with the benefit of hindsight, in order to mitigate the consequences of some unforeseeable event that has turned out to that party’s detriment.

This rule against relying on the benefit of hindsight would not have prevented Carillion’s argument here. When entering into a construction contract for any reasonably complex project, the parties may well be taken to have been alive to the possibility that their project might be delayed for any number of reasons, with each party accepting the risk of some of these potential delay events happening further down the line. The parties might also reasonably have expected that the cost of the work could vary, depending on the stage of the project: delay events might therefore give rise to different costs, depending on when they might happen. So, at the time of contracting, the parties may have had the commercial consequences of granting contiguous, or non-contiguous, extensions of time in mind.

Despite the fact that Jackson LJ described the argument for a non-contiguous extension of time as novel, it does not seem unusual in any way for an employer to think that once the original completion date has come and gone, and the contractor is responsible for the delay, the employer would be able to levy liquidated damages. It seems equally sensible to think that if something else happens after the date when liquidated damages start to become due, the contractor might be relieved from having to pay them - but only from that time on.

In Arnold, the Supreme Court elaborated that:

… while commercial common sense is a very important factor to take into account when interpreting a contract, a court should be very slow to reject the natural meaning of a provision as correct simply because it appears to be a very imprudent term for one of the parties to have agreed …

One could ask how precisely commercial common sense can be “a very important factor in interpreting a contract” while the court or tribunal is also told that it must be “very slow” to reject the natural meaning of the contract because of it. The natural meaning is a matter for ordinary language, always bearing in mind that under English law, the contract is interpreted objectively by a reader who knows the factual - and commercial - background to the transaction just as well as the parties themselves did at the time of entering into the agreement. Commercial common sense will not dictate the natural meaning of the words. Instead, commercial common sense is used to resolve ambiguities: where the words in the contract can bear different, alternative meanings, the more commercially sensible is to be preferred (see Rainy Sky SA v Kookmin Bank [2011] UKSC 50). The crucial question remains: when exactly is the contractual wording ‘ambiguous’ to allow such a choice?

A brief detour

In an earlier case, Balfour Beatty Regional Construction Ltd v Grove Developments Ltd [2016] EWCA Civ 990, Jackson LJ had to consider a similar argument based on commercial sense in the construction context. In that case, the Court of Appeal was split as to whether the contract was sufficiently ambiguous to bring the principle into play.

In Balfour Beatty, the majority of the Court of Appeal found that the parties had agreed that the contractor would receive interim payments but only for a period of 23 months, regardless of how long the work would actually take. The contract was based on the JCT Design and Build Contract form, with some amendments. The parties had indicated that they wanted ‘stage payments’ (as opposed to ‘monthly periodic payments’) by ticking the relevant box on the standard form. They did not, however, specify what stages would trigger payments, and instead noted that this was ‘to be agreed.’ They did subsequently agree a schedule. Even though they had ticked ‘stage payments,’ the schedule that they then agreed looked a lot like ‘monthly payments.’

The schedule was headed:

“Interim Valuation/Payment Dates 2013 – 2015

Valuation Application on Third Thursday of the month.”

It consisted of a table of 23 valuations and payments, giving 23 dates when the contractor was to make an application for each such payment, when the value of relevant work would be certified, and when payment would be made. Those dates stopped just short of the original completion date in July 2015. The work was delayed, ultimately taking about a year longer. A dispute arose as to whether Balfour Beatty was entitled to an extension of time. The employer refused to make any further interim payments after the 23rd payment.

In the Court of Appeal, Balfour Beatty argued that any agreement that the contractor would not receive any further stage payments beyond the original completion date no matter how much longer it would have to be mobilised would be ‘commercial nonsense.’ Jackson LJ rejected this argument. He found that the parties’ schedule was clear and unambiguous: it was expressly described as the “agreed schedule of valuation / payment dates for this project,” and it said nothing about dates for applications, valuations, relevant notices or indeed payments beyond the last date (close to completion). His Lordship held that:

“There is no ambiguity in the present case which enables the court to reinterpret the parties’ contract in accordance with “commercial common sense”.”

Absent any ambiguity, Jackson LJ stated that commercial common sense could only rescue a party “… if it is clear in all the circumstances what the parties intended, or would have intended, in the circumstances which subsequently arose.” He felt that the parties’ schedule was anything but clear as to how and when payments would be due after the dates had run out. Jackson LJ dismissed an argument for an implied term for very similar reasons. In his view, there was nothing commercially unreasonable about Balfour Beatty having to wait until the final date for payment (following completion of the project) before they would be paid for some of their work. Longmore LJ agreed with that conclusion, so Balfour Beatty’s appeal was dismissed.

Vos LJ, however, dissented. He found that the contract was sufficiently ambiguous such that the principle of commercial common sense could operate. His Lordship felt that a schedule which was headed “Interim Valuation/Payment Dates 2013-2015” could not be taken as meaning that the parties had agreed that interim payments would stop if the work took longer than the original completion date. The parties’ agreement was silent as to what would happen after the original date. Vos LJ thought that “… clear words would be required …” to support a construction which in reality would mean that Balfour Beatty would have to wait for two to three years until after the last interim payment date on the schedule before it would be paid substantial sums. Vos LJ’s preferred construction of the agreement was to insert the words “etcetera” at the end of the schedule, such that monthly payments would continue beyond the date when the parties had initially expected the work to complete. In that, however, he was in the minority.

Common sense does not help Carillion

Whether a contract is ambiguous so that you can have regard to commercial common sense, and choose the more commercially sound alternative meaning, may not be an easy question to answer: everything will depend on the contract wording in issue. Returning to Carillion v Emcor, Clause 11.3 referred to “… fixing such revised or further revised period or periods for the completion of the Sub-Contract Works.” The starting point may have been just one completion date for the work (set out in Part 4 of the Appendix). However, the extension of time provision in Clause 11.3 did refer to fixing ‘further periods’ (plural), which might be thought to introduce some ambiguity - perhaps enough to support Carillion’s case that there could be more than one period for completing relevant parts of the work.

Jackson LJ disagreed. As to the relevance of commercial common sense, His Lordship noted Arnold and simply said that:

“Recent case law establishes that only in exceptional circumstances can considerations of commercial common sense drive the court to depart from the natural meaning of contractual provisions.”

Gone from this statement is any reference to the ‘importance’ of commercial common sense (compare the dictum in Arnold). The principle seems firmly relegated to being only of ‘exceptional’ relevance. The express reference to “Recent case law” establishing this also suggests that Jackson LJ considered that the law has in fact changed.

Turning to Carillion’s substantive complaint, Jackson LJ could see no answer to Carillion’s argument that a contiguous extension of time might excuse the subcontractor during periods of culpable delay, while making it liable during periods when the subcontractor was actually performing additional work following a variation, and where the time allowed for that new work had not yet been exceeded. The Court of Appeal also accepted that under some subcontracts, such as Emcor and Carillion’s, this could make a real difference. That was because Emcor’s liability to Carillion was not expressed in terms of liquidated damages, but by reference to the actual loss that Emcor’s delay had caused Carillion to suffer under the main contract: Carillion’s claim was for loss and expense caused by Emcor, which might include Carillion’s exposure to liquidated damages but also irrecoverable prolongation costs under the main contract.

Returning to the example given earlier, the main contractor’s loss suffered from day 100 to day 150 (the delay in completing the original work) may not be the same as that suffered from day 150 to day 200 (the period when the scope of work has expanded to include the late variation). For example, the main contractor may be scaling down its site presence towards the end of the period, perhaps while waiting for the delayed subcontractor to complete. Similar problems may arise for the subcontractor: if the delay starts at day 150 (as it would on the contiguous approach), the subcontractor will be in breach of contract just as the scope of work expands, which might require additional resources. Applying the general principle that a subcontractor is not entitled to be paid its prolongation costs during periods of delay, the subcontractor might end up being penalised in that scenario.

Jackson LJ accepted that the contiguous approach could lead to the situation where “… one or other party will gain a windfall benefit” because the main contractor’s loss is unlikely to be the same in these different periods. There was no answer to the logic behind that argument. However, His Lordship noted that:

“On the other hand, as Oliver Wendell Holmes famously observed in his lectures on The Common Law, ‘the life of the law has not been logic: it has been experience.’ In practice the system of awarding extensions of time contiguously has worked satisfactorily, even though it is open to the criticisms which Mr Reed advances. It appears that no contractor or sub-contractor in a reported case has ever before felt the need to argue that awards of time should be non-contiguous.”

Ultimately, Jackson LJ agreed with the judge at first instance that these anomalies that could result from granting contiguous extensions of time were not sufficient to displace the natural meaning of the words in Clause 11.3.

Where does that leave delay claims?

The immediate consequence of the Court of Appeal’s decision in Carillion is that an employer would do well to apply liquidated damages as soon as the original time for completing the works has expired, if the contractor is at fault. If the employer waits, perhaps in the hope of ‘sorting out’ matters towards the end of the project, difficulties can arise if the project is further delayed because of something that is not the responsibility of the contractor. A contiguous extension of time might then wipe out at least some of the period during which (at the time) the employer could have claimed liquidated damages - so the employer will have to wait and see if there is further culpable delay during the remaining period of the work (as contiguously extended). The only really safe option for the employer would be to enforce that right as soon as it arises, and while the contractor is still in breach, by operating the relevant contractual machinery. Some employers may hesitate to make such a claim during the works, perhaps out of concerns that the project might suffer further if the relationship with the contractor becomes strained. However, the only way to be sure that liquidated damages are actually due is to cash them in.

Beyond that issue, Carillion (and perhaps also Balfour Beatty) reflect a continuation of the more restrictive approach to interpreting contracts heralded by Arnold v Britten, with commercial common sense taking a backseat. Could such a more literal approach have further implications on delay claims?

The example deployed in Carillion was concerned with a situation where there are several causes of delay: the subcontractor had been guilty of inexcusable delay in completing the original scope, but was then asked to perform additional work, which would push out the completion date further. This brings to mind concurrent delay, although the situation is not quite analogous (concurrent delay strictly speaking means two causes affecting completion of (the same) work at the same time). The received wisdom in English construction law is that where there are two concurrent causes of delay, one for which the contract is responsible, and one caused by the employer, then the contractor is entitled to an extension of time.

This is frequently referred to as the Malmaison approach (a reference to Henry Boot Construction (UK) Ltd v Malmaison Hotel (Manchester) Ltd (1999) 70 Con LR 32):

“… it is agreed that if there are two concurrent causes of delay, one of which is a relevant event, and the other is not, then the Contractor is entitled to an extension of time for the period of delay caused by the relevant event notwithstanding the concurrent effect of the other event …”

So where the contractor is prevented from working by extremely bad weather, but also lacks necessary labour or materials such that it could not have made any progress in any event quite regardless of the weather, the usual result is that the contractor is entitled to an extension of time. The contractor is not liable for liquidated damages, but cannot recover prolongation costs (‘time but no money’).

The quote from Malmaison set out above actually refers to the position as it was agreed between the parties. Subsequent cases, however, have treated the Malmaison approach almost as a legal principle. In De Beers UK Ltd v Atos Origin It Services UK Ltd [2010] EWHC 3276 (TCC), the Court described it as a “… the general rule in construction and engineering cases.

It should be recalled that a contractor’s entitlement to an extension of time - including in circumstances where there is concurrent delay - comes down to interpreting the relevant clauses in the contract. Consistent with the Malmaison approach, extension of time clauses have often been construed as entitling the contractor to more time upon the occurrence of a particular, specified event that is referred to in the contract. So long as that event has occurred, and it has caused delay to completion, then the contractual entitlement to an extension of time arises, and it does not matter that some other event has also, concurrently, caused delay. In Steria Limited v Sigma Wireless Communications Limited [2008] B.L.R. 79, the Court described this reasoning as follows:

“The rationale for such an approach is that where the parties have expressly provided in their contract for an extension of time caused by certain events, the parties must be taken to have contemplated that there could be more than one effective cause of delay (one of which would not qualify for an extension of time) but nevertheless by their express words agreed that in such circumstances the contractor is entitled to an extension of time for an effective cause of delay falling within the relevant contractual provision.”

That rationale depends on assuming that the parties must have contemplated something that they did not spell out in the contract. However, in both Carillion and Balfour Beatty, the Court of Appeal refused to look beyond the natural meaning of the words that the parties had included in their contract.

A second reason for adopting the Malmaison approach that one finds in the case law is that many delaying events which are expressly referred to in extension of time clauses would otherwise be ‘acts of prevention’ by the employer. At common law, where the employer by any act (which does not necessarily have to be a breach of contract) prevents the contractor from completing the work on time, the employer can no longer insist on the original completion date and time is then ‘at large’ - meaning the contractor has a reasonable time to complete the work. The Courts have suggested that it would be unfair not to give the contractor an extension of time under the contract if something happens that would otherwise be an act of prevention at common law: see Walter Lilly v Mackay [2012] EWHC 1773 (TCC), at 370.

The prevention principle is displaced in virtually any sophisticated construction contract, by the inclusion of an extension of time clause. Ten years ago, Jackson J (as he then was) reviewed the authorities relating to the prevention principle in Multiplex v Honeywell [2007] EWHC 236 (TCC). In so doing, he also identified the proposition that any extension of time clause (which displaces the prevention principle) that is ‘ambiguous’ should be construed in favour of the contractor. As to that proposition, he noted that:

“… it must be treated with care. It seems to me that, in so far as an extension of time clause is ambiguous, the court should lean in favour of a construction which permits the contractor to recover appropriate extensions of time in respect of events causing delay.”

That appears consistent with the explanation given in Walter Lilly as to why, in line with the Malmaison approach, extension of time clauses tend to be construed so as to bite on the occurrence of a specified event, even though there is another effective cause of delay for which the contractor is responsible. As regards establishing causation, there has been some debate as to the correct test to be applied. The Commercial Court in Adyard Abu Dhabi v SD Marine Services [2011] EWHC 848 Comm noted that true concurrency arose where delay was “… caused by two or more effective causes of delay which are of approximately equal causative potency.”

In the words of Jackson LJ, ‘let us now draw the threads together’ and ask whether after Arnold v Britten, Carillion and Balfour Beatty, this approach to construing extension of time clauses - to the extent that it depends on ambiguity in the wording - should still be the default position. An extension of time clause might say that “If Event A occurs and causes delay to Completion, the Completion Date shall be extended to reflect the delay so caused.” Would the Court of Appeal today read this as meaning “If Event A occurs and causes delay to Completion, the Completion Date shall be extended to reflect the delay so caused, even if such delay would also have happened without Event A by reason of something for which the Contractor is responsible.”?

The answer may be ‘Yes’, because the Malmaison approach has been the status quo for so long now. However, as this boils down to a point of contractual interpretation, one cannot be sure of the outcome of any particular case (and Jackson LJ has yet to decide a case raising this point). An (appropriately worded) extension of time clause could end up being read as meaning that the delay in question should not have happened without the occurrence of Event A, otherwise there is no extension of time. That result may be more likely where the concurrent event for which the contractor cannot be blamed was not caused by the employer. In that situation, the prevention principle would not be relevant.

In conclusion, there have been two Court of Appeal cases that apply a more restrictive approach to interpreting construction contracts. Whether this has more extensive implications, perhaps for concurrent delay, remains to be seen.