National House Building Council v Peabody Trust

Introduction

The principal issue in this appeal concerns the proper construction of an NHBC insurance policy. The policy provides insurance cover to an employer (“Peabody”) when they “have to pay more” than they would otherwise have done due to the contractor’s insolvency or fraud before practical completion. Here the contractor became insolvent. The NHBC argues that the cause of action under the policy accrued on the insolvency of the contractor in 2016, and that in consequence, Peabody’s cause of action in these proceedings is statute-barred. Peabody argues that the cause of action accrued in 2021 when they had to pay more to complete the building of the homes as a result of the insolvency, and that therefore the claim is not statute-barred. Mr Andrew Mitchell KC, sitting as a Deputy High Court Judge (“the judge”), found against the NHBC’s construction, but left it open as to whether, on Peabody’s construction, the claim was statute-barred. The NHBC has permission to appeal that decision.

In Section 2, I set out the relevant terms of the insurance policy. In Section 3, I set out the background facts. In Section 4, I identify the application to strike out and/or for reverse summary judgment made by the NHBC. In Section 5, I set out the relevant parts of the judge’s judgment. In Section 6, I identify the principles relevant to the construction of a contract of insurance, although there was little between the parties on the law. In Section 7, I set out the issues on appeal. Thereafter, at Sections 8-12, I analyse and answer those issues, albeit in a slightly different sequence to the numbered grounds. I am very grateful to leading and junior counsel on both sides for the clarity and economy of their arguments.

The Insurance Policy

The policy can be seen as a laudable attempt to convey in user-friendly language what is covered and when claims can be made. But it is not unfair to say that some of the language is rather muddled, perhaps because it is trying to cover the relationship between three different parties, namely the NHBC, the employer and the contractor. It is also designed to dovetail in with the Buildmark policy, which is designed to protect a different party altogether, namely the homeowner that buys the completed home. Perhaps it is unsurprising that, in looking in so many different directions at once, the language of the policy can be rather opaque.

One of the areas of cover in the Policy is property damage, covered in ‘Part C – Home damage cover’. That applied “if there is physical damage to the home(s) because the contractor failed to build certain identified structural elements of the homes in accordance with NHBC requirements”. The employer (in this case Peabody) could make a claim under this section if they thought “there is physical damage to the homes”. In those circumstances, the NHBC promise:

“We will take responsibility for having the work done to put right the physical damage to the homes, if the cost to us is above the excess.

Alternatively, if we choose to, we will pay you what it will cost us to have the work done, less the excess amount applicable.”

Option 1 is the section of the policy with which we are principally concerned. I adopt the judge’s use of the lettering [A]-[E] to divide up the relevant parts because, although they do not appear in the original, they make the different elements of Option 1 easier to identify:

“[A] Option 1 – Insolvency cover before practical completion

[B] When this section applies

This section applies if you lose the amount paid to the contractor in accordance with the building contract or have to pay more to complete the building of the home(s), because the contractor is insolvent or commits fraud.

[C] When you can claim

You can only claim under this section up to the date of the Buildmark Choice certificate. Contact us and tell us if you have lost the amount you paid to the contractor or the contractor has not completed the home(s).

[D] What we will do

We will pay you the reasonable extra cost above the contract price including professional fees, for work necessary to complete the home(s) to the NHBC requirements; or

We will reimburse the amount paid to the contractor in accordance with the building contract which cannot be recovered from them.

[D1] In addition, we will pay the cost of reasonable precautions to secure the work defined in the building contract against unauthorised entry, theft and vandalism until work resumes.

[E] Conditions and limitations

This option will only apply if included on the quotation and the additional premium has been paid to and accepted by us.

There are limits to how much we will pay (as explained on pages 14 & 15)

Some things are not NHBC's responsibility under Buildmark Choice (as explained on page 16).”

Limits as to how much the NHBC would pay out are set out at pages 14-15 of the policy. In respect of Option 1, it is said that “the most we will pay for all claims under Option 1 is 10% of the original contract price. We deal with all claims in the order they are made. When the overall limit is reached, we will not pay further claims.” Another limitation is that the NHBC say that “we will not be responsible for costs resulting from your unreasonable delay in pursuing a claim.”

Amongst the relevant definitions are:

Catalyst Housing Limited (“Catalyst”) engaged Vantage Design and Build Limited (“Vantage”), pursuant to a JCT Design and Build Contract dated 20 November 2015, to build 175 new dwellings at the former RAF Stanbridge site at Bedfordshire. This included 88 social housing units. The total contract sum was almost £23.8 million, of which the cost attributable to the 88 social housing units was £10.3 million.

The insurance policy with the NHBC was dated 2 March 2016. It related solely to the 88 social housing units.

Vantage commenced work on or around 14 December 2015. They ceased work on 17 June 2016 and administrators were appointed on 29 June 2016. It is common ground that the appointment of the administrators meant that, pursuant to the policy of insurance, Vantage were insolvent within the relevant definition. Squatters moved onto the site and the documents suggest that over the autumn and winter of 2016/2017, Catalyst spent some £56,350 on site security.

Following the administration of Vantage, in late 2016/early 2017, Catalyst’s subsidiary CHL Developments Limited engaged Stack London Limited (“Stack”) to perform the role of construction manager and to procure the individual trade contracts so as to complete the works. The trade contracts would be between CHL and the trade contractors.

At the same time, on 7 December 2016, Catalyst informed the NHBC that “Catalyst intend to claim via the insolvency cover for additional costs in relation to the first registered units.” No figures were given. However, on 10 February 2017, Catalyst wrote again to the NHBC identifying the costs forecast from the Stack management contract and saying:

“So the additional costs on the insured units is approximately £1,358,000, more than the max 10% claim. However this is based on build cost only and we also have a lot of additional fees, abortive works costs, legal costs, payments to Vantage post administration etc.”

The project cost plan referred to in the construction management contract between CHL and Stack identified a forecast final cost of £25.6 million, about £1.8 million higher than the Vantage contract sum. In addition, Stack were entitled to a construction management fee of over £1 million, which may also have been additional to the sums that would have been payable to Vantage (there is a dispute about that).

Practical Completion of the relevant units occurred on 19 January 2021. The final cost of the 88 units was some £11.3 million.

Thereafter, there was a good deal of correspondence between the NHBC and Catalyst in respect of this claim. In June 2023, Catalyst merged with Peabody, who took over the management of the claim. In Peabody’s skeleton argument, it is noted that, during this correspondence, the NHBC did not take the point as to the accrual of the cause of action until more than six years from the date of the insolvency (which was 29 June 2022), and there is a hint that NHBC may have misled Peabody during this period by regularly seeking further information as to the additional costs incurred. However, although a plea of estoppel was originally put forward by Peabody, that was expressly abandoned in advance of the hearing before the judge. In those circumstances, it seems to me that, save for one tangential point identified in paragraph 60 below, the correspondence between the parties is of no relevance to the construction issue at the heart of this appeal.

The NHBC’s Application

On 24 July 2023, Peabody commenced these proceedings against the NHBC under the policy. It was alleged that the total additional costs of the 88 units was £913,555.36. That claim was less than 10% of the Vantage contract sum, which was the cap set out in the policy in respect of Option 1.

On 18 January 2024, NHBC applied to strike out the claim on the ground that it was statute barred, alternatively they sought reverse summary judgment on the same basis. The application was in general terms. The reason for the application was, however, explained at paragraph 16 of the accompanying witness statement of Ms Ibukunoluwa Alabi dated 18 January 2024. She said:

“In the course of my review of the matter, it became evident to me that Catalyst’s claim had become statute-barred on 29 June 2022 pursuant to section 5 of the Limitation Act 1980, by virtue of more than six years having passed since Vantage went into administration on 29 June 2016.”

This proposition was denied by Peabody, who served a statement from Mr Mark London, Peabody’s solicitor. That statement was dated four months later, in May 2024. It said that the NHBC were wrong to say that the cause of action arose on Vantage’s insolvency ([3]). Although the statement did not spell out in terms when Peabody said that the cause of action accrued, from [17] onwards, Mr London spent a good deal of time explaining “when Peabody incurred additional costs of completing the homes”. He said at [20] that “the date when Peabody incurred more than it would have done with Vantage is relevant to the accrual of a cause of action, which can be calculated.”

In response to this, there was a second statement from Ms Alabi dated 31 May 2024, which said at [7.3] that “whilst I do not intend to argue NHBC’s position in this witness statement, I do think it may be helpful to the court to note that, even on Peabody’s position…, the facts clearly point to Peabody’s claim having been issued out of time as I explain further below.” The statement then contains a detailed explanation as to why Ms Alabi said that “there was never any prospect of Peabody ‘paying less’ to complete the insured homes following Vantage’s insolvency…the risk that the insured would have to ‘pay more’ must, therefore, have arisen either on or shortly after Vantage’s insolvency. But it undoubtedly arose at the latest by January or February 2017 and Catalyst realised this by then.” Given that 6 years before the commencement of the proceedings was 23 July 2017, if Ms Alabi was right, it would mean that, even on Peabody’s construction of Option 1, the claim was still statute-barred.

Ms Alabi’s second statement was provided about a week before the hearing of the application, which took place on 7 June 2024. The judgment dismissing NHBC’s application was dated 6 August 2024.

The Judgment

Having set out the background facts and the policy, the judge’s judgment ([2024] EWHC 2063 (TCC)) next contained a lengthy section entitled ‘The Scope of the Application’. The judge called NHBC’s argument that the cause of action accrued on 29 June 2016 “the insolvency point”. He described it as a discrete point of law and construction. The judge said that the half day estimate given by NHBC to deal with the application on that basis was “demanding but not unreasonable”.

The complication that the judge went on to identify was that Mr London had included in his witness statement a good deal of elaboration on Peabody’s case that the claim did not accrue until Peabody incurred the extra costs of completing the homes. As the judge pointed out at [20], Mr London put forward three alternative ways of ascertaining when that was. The judge described them as follows:

“20…Mr London put forward three alternative ways, in principle, of ascertaining the point in time when Peabody "had to pay" extra costs. The first would be to conduct a deductive exercise from the dates of final accounts (the earliest of which is July 2017); Peabody alleges that this has always been NHBC's position, that it could not finalise what it had to pay until final accounts were prepared. The second would be to analyse cumulative costs over time as the work progressed. The third would be to instruct an independent expert quantity surveyor to retrospectively analyse the letting of the individual works contracts packages, and individual interim applications for payment, and compare the outputs of those analyses with a hypothetical analysis where Vantage did not enter insolvency.

21.

Mr London's essential point is that, at the time of insolvency, only some £1.5m had been paid to Vantage, and that there was approximately £8.8m left in the tank, as it were, before the Contract Price for the units (c. £10.3m) would be exceeded. And that it is at the least realistically arguable that, depending on what approach is taken, the moment of "having to pay more" did not arise before March 2020, alternatively June 2020, alternatively certainly not prior to July 2017 (being six years prior to the Claim Form).”

The judge noted at [22] that, in response to this evidence, the NHBC had developed an alternative case that, even if they were wrong on the insolvency point, the claim was still statute-barred. That was based on Ms Alabi’s second statement: that, even if the relevant cause of action accrued when Peabody ‘have to pay more’ as a result of the insolvency, that was either coterminous with the insolvency or, in the alternative, when the reasonable extra cost of completing the homes was capable of assessment. The NHBC said that was when the work went out to competitive tender, so that by February 2017 at the latest, the claim was still statute-barred.

The judge concluded that this alternative case was not capable of being determined during the half day hearing, and that he would therefore only address the insolvency point. He said at [32] that he was not prepared to determine either the “complex issues raised by the alternative argument”, or the question as to precisely when Peabody “have to pay more”, whether in principle or on the facts. He indicated at [33] that “the correct answer to the alternative case does not appear obvious”.

Starting at [41], therefore, the judge addressed the insolvency point. He concluded that, for a variety of reasons arising out of the language of Option 1, NHBC were wrong to say that time ran from the insolvency of Vantage. The kernel of the judge’s construction can be found at [57] in these terms:

“57.

In my judgment, Option 1 cover does not apply (i.e. is not triggered) if the insured did not "have to pay more to complete" the units, or if the insured did not lose any money paid to the contractor, despite the contractor going insolvent. The event insured against is not the insolvency (or fraud of the contractor) per se, but rather the insured being required to pay more above the contract price to complete. The requirement to pay more must have been caused by the insolvency (or fraud) of the contractor, but the insolvency (or fraud) itself is not the risk which is covered. Peabody is correct therefore in its submission that the insured losses (the extra costs, or lost payments) are an essential and definitional part of the insuring clause itself [B], and not matters which simply go to the delineation of quantum [D]. It would have been easy enough to draft [B] to make clear that the claim arose on insolvency itself, regardless of whether any extra expense (or lost payment) was caused by it, if that had been intended.”

Separately, the judge dealt with the site security costs issue from [63] onwards. NHBC’s argument had been that, since the claim for those costs was statute-barred, the entire claim under Option 1 was also statute-barred because there was just one unitary cause of action. The judge rejected that submission on the basis that the claim for site security costs under D1 of the policy was additional to, and comprised separate cover from, the main Option 1 claims: see [65]. He declined to decide when that cause of action arose in the absence of any specific evidence on the point.

For these reasons, the judge dismissed NHBC’s application to strike out at [69], finding that time did not start running on the insolvency of Vantage on 29 June 2016 but at a time (to be determined at trial) when Peabody “have to pay more to complete the units, as a result of that insolvency”.

The Law

It is trite law that the cause of action under an insurance policy accrues on the happening of the event insured against. In Callaghan & Anr v Dominion Insurance Co. Limited & Ors [1997] 2 Lloyds Rep 541, Sir Peter Webster said:

“It seems to me that the best way to define an indemnity insurance is that it is an agreement by the insurer to confer on the insured a contractual right which, prima facie, comes into existence immediately when loss is suffered by the happening of an event insured against, to be put by the insurer into the same position in which the insured would have been had the event not occurred, but in no better position.”

Many insurance policies are in respect of property damage. In those cases, the cause of action will accrue on the happening of the fire or flood, or whatever the insured event may be. Thus, in Kelly v Norwich Union Fire Insurance Limited [1990] 1 WLR 139, there had been a burst pipe before the inception of the policy, which subsequently resulted in heave and property damage during the period of the policy. The Court of Appeal rejected the claim on the basis that the word “event” in the policy was a reference to the event which brought about liability, and that was the leak prior to the inception of the policy, not the subsequent damage.

Bingham LJ (as he then was) at page 148 of the report, gave five reasons for concluding that the insured’s right to indemnity was dependent upon his showing that the specified event in question occurred during the term of the policy. One of those reasons was that, if asked, the insured would have said that he had insured against “fire, explosion, lightening, earthquake, storm, flood” not “loss or damage caused by fire, explosion, lightening, earthquake, storm, flood…” He said that this was a case where “the colloquial response accurately reflects the legal reality.”

A more recent example of this line of cases is Griffiths v Liberty Syndicate 4472 [2020] EWHC 9480 (TCC); [2020] Lloyd’s Law Reports 312. Under the terms of the policy in question, HHJ Pelling KC, sitting as a High Court Judge, rejected the submission that, under the terms of the policy, the cause of action did not accrue until the cost of carrying out the repair work was incurred by the claimant. The cause of action accrued when a loss was caused by the event insured against, which in that case was the manifestation of the defect. Points were also made in the judgment in Griffiths about how, if the opposite construction were right and it was only on actual expenditure that the cause of action accrued, it would allow the claimant to control when time started to run: Legal Services Commission v Henthorn [2012] 1 WLR 1173. It would also defeat the purpose of the insurance in a situation where the work was too expensive for the claimant to carry out at all, which would see the insurer avoiding liability altogether: Manchikalapati v Zurich Insurance PLC [2019] EWCA Civ 2163; [2020] Lloyd’s Rep. IR 77.

In the absence of wording to the contrary, this then is the standard approach: that a claim arises under an insurance policy as soon as the event which directly results in the loss has occurred, and not when the loss is manifested (as confirmed by Colinvaux & Merkin’s Insurance Contract Law at C-0236). But as this commentary makes clear, everything turns on the precise words of the policy. There is no principle that prohibits the parties from agreeing that the relevant event is either the incurring by the insured of a liability to pay, or the actual expenditure of monies that the insured would otherwise not have had to have paid.

In British Credit Trust Holdings v UK Insurance Limited [2003] EWHC 2404 (Comm); [2004] 1 All ER) Comm) 444, the assured financed hire purchase agreements to the purchasers of motor cars. The losses sustained by the assured in respect of their non-prime lending was insured with the defendants. As customers defaulted on their hire purchase commitments, so claims continued to arise and were notified to the insurers on a monthly basis.

Clause 5 of the policy made plain that no legal action could be brought against the insurer unless it was commenced within one year after any loss incurred in respect of the hire purchase agreement. Morison J found that time did not start to run under clause 5(n) on the date of termination of the hire purchase agreement: that was merely a trigger for potential recoverable loss. Instead, he found that time could not start to run against the insured unless it knew whether there was a loss at all, and identified what that loss was in order to recover it under the policy.

Other authorities were cited to us on different points and I address those in the relevant sections of the judgment below.

The Issues on Appeal

Ground 1 of the appeal encapsulates NHBCs case that, on a proper construction of Option 1, Peabody’s cause of action under the policy of insurance accrued on the date of Vantage’s insolvency. Ground 2 is a complaint that, although Peabody failed to discharge the burden of demonstrating that their claim was in time, the claim was not struck out. Ground 3 is linked, because it suggests that the judge was wrong not to find that, in any event, the cause of action had accrued more than six years before the issue of the claim form. Ground 4 focusses on the judge’s case management, submitting that he erred in keeping to the two hour time limit (which he regarded as inadequate to determine all the issues which had now arisen), rather than adjourning the application, and so only decided the insolvency point. Ground 5 is concerned with the potentially separate issue about the site security costs.

The Judge’s Case Management Decision (Ground 4)

Although it may appear counter-intuitive, it is appropriate to start with the judge’s case management decision that, in the circumstances which he describes in his judgment, he was only going to deal with the insolvency point. He explained that the original application to strike out was based on the proposition that the cause of action accrued on Vantage’s insolvency, without any qualification or elaboration. He decided that that issue was capable of being dealt with in the estimated timeslot of two hours. He decided that the issues which would arise if the insolvency point was wrong, such as what precisely “if you…have to pay more” meant, and whether that required actual or merely foreseeable extra cost to trigger the cause of action, could not be dealt with in the time and might require further evidence.

In my view, the judge was entitled to make that case management decision. Although Mr Grant KC argued that the application for reverse summary judgment/strike out was in general terms, it was plainly narrowed by Ms Alabi’s first statement. There was no mention of any alternative case based on the “if…you have to pay more” part of Option 1 until her statement in response to Mr London’s statement, just a week before the hearing. In their submissions to the judge, Mr Casey KC and Mr Mesfin at [39] and [40] made plain that the issue for the judge was whether Vantage’s insolvency was the point when the cause of action accrued and that, for the purposes of the strike out application, the court did not need to go any further. In particular, if the court rejected the insolvency point, they said that it did not need to determine when in fact the cause of action actually accrued. In my view, as a matter of case management, the judge was entitled to conclude that, in all the circumstances, that was the best course.

I should say that I am not sure whether I would have made the same case management decision. Had I been the judge, I may have wanted to deal with all the arguments relating to limitation together, and in advance of the trial. I may not have wanted to decide only one aspect of that argument, and thereby at least run the risk of hobbling the remainder of the debate. I would not have regarded the pre-hearing events and the changes in the parties’ positions as anyone’s fault so, if necessary, I may have adjourned the hearing part-heard and allowed further evidence to be submitted on the alternative case. But of course what another judge might have done is irrelevant, if what the case-managing judge did was open to him or her. Here, dealing just with the insolvency point was plainly a case management decision that the judge was entitled to make.

In those circumstances, ground 4 of the appeal must fail (a result expressly foreshadowed when I granted permission to appeal). It follows that the scope of this appeal is considerably narrower than the NHBC would wish. That is best explained by reference to the three options as to the accrual of the cause of action which Mr Grant helpfully identified in his oral submissions.

The three options were:

Mr Grant was very clear about these three options. He said that if the answer was option i), then he won. He accepted that if the answer was option iii), he lost. He urged the court to decide option ii) because, he said, it was an issue that was before the court and did not require any further fact-finding.

But as a result of the judge’s case management decision, option ii) was not decided below. Because I have concluded that ground 4 must fail, such that that case management decision cannot be revisited, it seems to me that the inevitable consequence is that option ii) cannot be properly determined on this appeal. Moreover, I am not persuaded that option ii) can properly be considered without at least the possibility of some further evidence concerning when it was possible/likely/probable/foreseeable that Peabody would “have to pay more”. That was a point which the judge made. It was also inherent in Mr Grant’s submission that, under option ii), what mattered was “the factual reality that, in consequence of the insolvency, Peabody would have to pay more to complete the homes”.

Consistent with what I have said at paragraph 39 above, I would add that, if option i) fails, then the issue as to whether the answer is option ii) or iii) should be determined at an early stage, perhaps by way of a preliminary issue, and certainly in advance of the trial. It would be unsatisfactory for the matter to go to trial in circumstances where the entire claim might fail on limitation grounds.

Having addressed the procedural position, I therefore turn to the critical issue for the purposes of this appeal, namely the insolvency point.

The Proper Construction of Option 1 (Ground 1)

The Words Used

At first sight, the argument that Peabody’s cause of action accrued on the date of Vantage’s insolvency has some force. [A], the heading of Option 1, is “Insolvency Cover”, which may be an indication that that was the event/risk being insured against. But that is then immediately undercut by the words at [B], “if you…have to pay more to complete the building of the home(s), because the contractor is insolvent or commits fraud”. That on its own demonstrates that [A] is misleading, because Option 1 also covers fraud; it is not limited to insolvency. Moreover, the words strongly suggest that what matters most is Peabody having to pay more to complete the homes.

In short, the provision at [B] identifies two criteria which must be fulfilled before liability under the policy is triggered. The critical first criterion is that the insured must “have to pay more” to complete the building of the homes. That is significant because the policy is only engaged “if you…have to pay more to complete the building of the homes”. It is a condition without which Option 1 cannot apply. The second criterion is that the payment of “more” must be because of the insolvency (or fraud) of the contractor. Both criteria must be fulfilled, but the words “if you…have to pay more” make plain that what is being insured against is a particular financial loss and that, without that trigger, there can be no claim.

I shall come on to the other words of Option 1 in a moment. But on the basis of the words in [B] alone, it seems to me clear that this is not a conventional property damage clause of the kind described in Callaghan and Kelly v Norwich. It is also very different to Part C (paragraph 4 above) of this same policy which is much more akin to a conventional property damage clause. Indeed, in Harrison v Shepherd Homes Limited & Ors [2010] EWHC 1398 (TCC), a case about this very provision, Ramsey J found that the cause of action against the NHBC ran from the moment of physical damage to the property, that being the relevant event under Part C. By contrast, on the plain meaning of the words, [B] of Option 1 only triggers a liability “if you…have to pay more”.

In my view, Peabody were right to say that NHBC’s insolvency point ignores these critical words of [B]. It concentrates instead solely on the second criterion of insolvency. It makes no attempt to explain, let alone get round, the words “if you…have to pay more”.

Going on to the rest of Option 1, I consider that this interpretation is supported by the text under the heading at [C] “When you can claim”. That provides that the NHBC should be contacted “if…the contractor has not completed the homes(s)”. I agree with the judge’s description of this at [60] as a notification provision, but it is also another indication that the relevant trigger is paying more (or the risk of paying more) because of the non-completion of the homes by the original contractor. That is inconsistent with the NHBC’s construction, and consistent with Peabody’s case that non-completion is inextricably linked to having to pay more to complete the homes.

The argument that the cause of action did not accrue simply on insolvency is also supported by the text at [D], under the heading “What we will do.” That talks about the NHBC paying Peabody “the reasonable extra cost above the contract price…for work necessary to complete the homes”. That again is consistent with the construction that a critical requirement before a claim can be made concerns the payment out of the “reasonable extra cost”, which is necessitated by Peabody having to pay more to complete the homes in the first place. Again, as with the section at [C], this provision is all about extra cost, non-completion, and payment. There is no reference in either section to insolvency.

In the submissions before the judge, there was a good deal of argument about whether the fact of insolvency necessarily meant that an insured would have to pay more to complete the homes than under the original contract. That argument rightly loomed much less large at the appeal hearing. I can see no basis for saying that the insolvency of contractor A automatically means that a subsequent contract with contractor B will be more expensive. It will always depend on the facts. Moreover, there is nothing about the construction of sections [B] – [D] of Option 1 that would justify an interpretation of insolvency as automatically requiring the employer to pay more. It is all conditional (hence the use of the word “if”). To that extent, therefore, I agree with the judgment below at [55]-[56].

For all those reasons, it seems to me that section [B] (in particular), but also sections [C] and [D] of Option 1, are contrary to NHBC’s argument that the cause of action accrued immediately on Vantage’s insolvency, and provide strong support for the view that this is a policy where the payment (whether actual or foreseeable) of sums additional to that which would have been payable to Vantage was required before a cause of action accrued.

Fraud

There is an entirely separate point, unconnected with insolvency which, in my view, also answers the insolvency point decisively against the NHBC. As I have noted, Section [B] is concerned not only with insolvency, but also provides cover where the contractor commits fraud. In such circumstances, it could not sensibly be argued that Peabody’s cause of action accrued at the moment that the contractor committed the fraud (which, by parity of reasoning with the insolvency argument, must be NHBC’s case).

An employer in the position of Peabody will almost certainly not know if or when the contractor commits fraud. Indeed, that may not become apparent to anyone for months, if not years, after the relevant acts or omissions. If the cause of action accrued at the moment the contractor committed fraud, Peabody’s time for commencing proceedings under the policy may be severely truncated, if not completely extinguished, before it even knew that a fraud had occurred. So, in the context of fraud, it makes perfect sense for the cause of action under the policy not to accrue until the principal requirement is fulfilled, namely Peabody having to pay more to complete the building of the homes as a result of that fraud.

Commerciality

It is trite law that, in a dispute about the meaning of a term in a contract, where there are two alternative constructions, the court must test both to see how commercially sensible they may be. That is clear from [12] of Wood v Capita Insurance Services Limited [2017] UKSC 24; [2017] A.C. 1173 and Lord Hodge’s reference to the iterative process “by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated”.

I consider that Mr Casey KC was right to say that NHBC’s construction – the insolvency point - is essentially uncommercial. The NHBC must argue that the cause of action accrued on the insolvency, at a time when Peabody simply did not know whether or not the eventual costs would be more or less than the amount they would have had to have paid Vantage. Depending on the facts, they may not know if they would “have to pay more” for months or years after the insolvency, given the projected time for the completion of the project as a whole. It would mean that the cause of action accrued to Peabody at a time when they may have had no way of knowing whether they would actually suffer a loss at all, much less the extent of that loss. That would be a very curious outcome in circumstances where Option 1 was designed to protect Peabody against the particular financial loss they suffered, entitling them to rather more than a back-of-an-envelope guess.

If it were right that a cause of action accrued immediately on insolvency, what would be the consequences? Peabody would be obliged to claim, and the NHBC would have to pay out, on a hypothetical basis when nobody knew whether there had been or would be a loss at all, let alone how much that loss might be. On this basis, the NHBC would probably not be entitled to defer paying out until the actual costs were known, because that would mean that the cause of action had accrued to Peabody but they could do nothing meaningful about it. If your cause of action has accrued, and is established in principle, you are entitled to be paid out promptly: you do not expect to be fobbed off for years whilst the figures are endlessly pored over.

Accordingly, it seems to me to be very unlikely from a commercial perspective that a cause of action accrued at a time when the consequences of the insolvency were entirely unknown. Again, that can be contrasted with the occurrence of a flood or a fire. The need to rebuild the factory after a fire, or to replace all the ground-floor fixtures and fittings in a house after a flood, will be readily apparent immediately after the event. But where the event concerns the payment of a sum over and above another sum, it is inevitably much more uncertain.

Moreover the certainty inherent in a typical property damage clause can be contrasted with what actually happened here. We know that five years after the insolvency, in 2021, the NHBC were saying to Peabody that “with respect to the position on the overall claim in general…without the final cost to the work we are unable to accurately establish and ascertain the true value of any claim” (see the NHBC email of 15 January 2021). Whilst I accept Mr Grant’s submission that that could be consistent with the ascertainment of loss as opposed to the accrual of the cause of action, it does I think highlight the uncommerciality of the suggestion that the cause of action had actually accrued to Peabody five years before this email was even sent, in circumstances where the NHBC were still refusing to pay out. The concept of the cause of action accruing in 2016, whilst the claim itself remained unpaid because of the uncertainty about the figures in 2021, strikes me as commercially unrealistic.

What about the commerciality of Peabody’s case, that it was necessary to fulfil the “if you…have to pay more” requirement before the cause of action accrued? In one sense, that may depend on what precisely that requirement entails, and for the reasons I have already explained, that is not a matter for this appeal. But if Mr Grant’s option ii) is right, and the second requirement is fulfilled when it was possible, likely, probable or foreseeable that extra costs would be incurred, then that ought to be capable of relatively swift determination on the evidence.

I accept that, if it is option iii) – the actual payment of more than the sum that would have been payable to Vantage - the position is more complex, for all the reasons explained by the judge. However, I am not persuaded that these potential difficulties mean that Peabody’s construction of the provisions is uncommercial. The mere fact that it may be difficult to work out the precise date in practice does not mean that a construction which is otherwise right as a matter of language should then be rejected, in favour of a construction which is not the natural meaning of the clause as a whole. The principal question for now is whether or not, on Peabody’s construction (option iii), an accrual date can be calculated. It is plain that it can. So that is not a reason to reject that construction.

Furthermore, complexity surrounding the date of accrual of a cause of action is not uncommon in the construction field. BDW v URS [2025] UKSC 21 is a recent case, decided by a seven-strong Supreme Court, concerned with when a cause of action accrued in respect of defective design. Another example arises out of the regular arguments in the TCC that limitation periods should be extended because of deliberate concealment by the contractor of defective materials or workmanship. It is a fact of life that the accrual of a cause of action may not always be straightforward to determine: that does not mean that the construction of a clause which has that effect should be rejected.

Finally, the proof of this pudding is in the eating. Here, if option (iii) is correct, it makes no different which of Mr London’s three alternatives finds favour with the court. That is because, as Mr Grant fairly accepts, whichever is right, the claim is not statute-barred.

The Colloquial Response

Mr Grant and Mr Smith made much in their skeleton argument of Bingham LJ’s ‘colloquial response’ test in Kelly (see paragraph 30 above). Here, for the reasons that I have set out, I am clear that, if asked what they had insured against, Peabody would have said: “We insured against the risk that we might have to pay more to a new contractor if Vantage went bust or were found to be on the fiddle.”

Summary on Ground 1

For all those reasons, therefore, it seems to me that the proper construction of Option 1 is that the cause of action accrued if Peabody “have to pay more” as a consequence of the insolvency of Vantage. The insolvency point is not the natural meaning of the words used and is not a commercial outcome. I would therefore reject Ground 1 of the appeal. That then brings me to Ground 2.

The Limitation Argument (Ground 2)

Mr Grant complained that, in consequence of the events at the hearing, when the judge declined to deal with anything other than the insolvency point, Peabody avoided the burden that they otherwise had of proving that their claim was within time. Therefore, as he put it, Peabody “won on a technicality”. On analysis, I do not think that that submission can be sustained.

I accept Mr Grant’s first proposition of law that, once a defendant has raised a limitation defence, the claimant has to show that his claim is not statute barred: see Cartledge & Ors v E Jopling & Sons Limited [1962] 1QB 189 at page 202; and London Congregation Union Inc v Harriss and Harriss (a firm) [1988] 1All ER 15 at page 78, b-j. I also accept that the general approach of the law, as a matter of policy, is to accelerate, rather than retard, the accrual of a cause of action: see Battley & Anr v Faulkner & Anr [1820] 3B & ALD 286; Nykredit Mortgage Bank PLC v Edward Erdman Group Limited [1997] 1W.L.R.1627 at page 152C-E.

However, I do not consider that what has happened here is in breach of either of those principles. The judge did not decide that Peabody had discharged the burden of proving that their claim was not statute barred. As noted above, the only thing that the judge decided was that the cause of action did not accrue automatically on insolvency, so he dismissed the NHBC’s application to strike out. The judge was not deciding that Peabody had discharged the burden of proving that their claim was not statute barred; he deliberately left open that question to the trial. They may discharge the burden of proof then, but they may not; at the moment, the issue remains moot.

Similarly, although the law generally wants to see claims advanced speedily rather than otherwise, in a case like this, accrual will depend on the provisions of the contract or policy. General policy cannot override the precise words of the agreement between the parties.

For those reasons, I consider that (aside from the specific argument as to the insolvency point, which the NHBC rightly lost), the limitation arguments remain to be decided. I would therefore reject Ground 2 of the appeal.

The Alternative Case (Ground 3)

Ground 3 is the complaint that the judge should have found that the cause of action accrued more than 6 years before the issue of a claim form. That is the NHBC’s alternative case, now encapsulated as option ii) in paragraph 41 above. For the reasons that I have explained, the judge declined to decide that alternative case, as he was entitled to do. Since he did not decide it, it is not appropriate for this court to decide it de novo. For that simple reason, Ground 3 must fail.

Site Security Costs (Ground 5)

The argument that arises here is that, as the NHBC submitted to the judge, the claim for security costs clearly arose in 2016 and, because this is a unitary cause of action, that meant that the whole claim was statute-barred. The judge rejected that on the basis that this was an additional and separate claim under D1. He went on to say that:

“66.

As to whether the [D1] claim is time-barred, on the basis that the costs were incurred more than six years before the Claim Form, the Court is faced with the difficulty that, no doubt because this point did not appear as part of the Application or the evidence, Peabody did not address it either in its evidence or its skeleton argument. There is limited evidence about the costs in question. For example, the letter of 18 January 2018 relied upon by NHBC stated that the costs were incurred to secure the site and to remove squatters. But the detail of the costs was set out in Section E of the claim document, which is not in evidence.”

On appeal, Mr Smith argued cogently that it was plain that these costs were incurred in 2016; that they were covered by the Policy at D1; and that liability did not depend on the requirement that Peabody “have to pay more” than they would have had to have paid to Vantage. Mr Smith said that pursuant to D1 this was an additional cost, and therefore recoverable as such. In consequence he said that, since this was a unitary cause of action, it meant that the whole claim was statute barred. In support of that last submission he relied on Bann Carraig Limited v Great Lakes Reinsurance(UK) PLC [2021] NIQB 63 and the passage at Colinvaux & Merkin at C-0242.

In response, Mr Casey argued that this was covered by the “if…you have to pay more” provision and so could not be determined by this court. In any event he said that, even if he was wrong about that, the only claim that would be statute barred was the separate free-standing claim for site security costs, not the whole claim.

In my view, the judge was right to say that this was a separate claim. He was also right to say at [65] that this separate claim was not triggered by the “if…you have to pay more” requirement. This was a different claim, to which different considerations applied.

If, however, this was a separate and free-standing claim, as the judge said it was, then I do not accept that it was part of a unitary cause of action. It was distinct from the main cause of action. It therefore falls to be considered separately. Bann Carraig is not on point since that was a claim under a property policy and for the reasons that I have explained, that is not this case.

I am bound to say, that on the material presented by Mr Smith, a prima facie case was made out that the separate claim for site security costs was potentially statute barred. However, I have set out [66] of the judgment below, where the judge declined to reach that conclusion, including the absence of specific evidence. In those circumstances I would again be reluctant to interfere with that decision: it might be said that it was a matter of evaluation for the judge. That means that this point too remains to be decided.

For those reasons, therefore, whilst I decline to allow ground 5, I would suggest that Peabody look very closely at this individual head of loss. I reiterate that, had I been considering this matter at first instance, I may have concluded that, absent any compelling evidence that led to a different view, it was statute-barred.

Conclusions

For the reasons that I have set out, if my Lords agree, I would dismiss this appeal.

I agree with both judgments.

I agree that the appeal should be dismissed for the reasons given by Coulson LJ. But in deference to the forceful submissions made by Mr Grant KC in support of the appeal, I add a short judgment of my own on the interpretation of the policy. Coulson LJ has set out the relevant terms of the policy; and I will use his lettering.

In Callaghan v Dominion Insurance Co Ltd [1997] 2 Lloyd’s Rep 541, 544 Sir Peter Webster said:

“It seems to me that the best way to define an indemnity insurance is that it is an agreement by the insurer to confer on the insured a contractual right which, prima facie, comes into existence immediately when loss is suffered by the happening of an event insured against, to be put by the insurer into the same position in which the insured would have been had the event not occurred, but in no better position.”

As Colinvaux’s Law of Insurance (13^th ed) puts it at para 10-133:

“There is a consistent line of authority for the proposition that the date on which the assured’s action accrues is the date on which the insured peril occurs and not on the later dates when the loss is manifested, the assured incurs expenditure or the insurers deny liability, the principle being that the insurer has agreed to hold the assured harmless against the occurrence of an insured event so that when the event takes place the insurers are in immediate and automatic breach of contract and are liable for unliquidated damages.”

The critical question, then, is what is the “insured event”? The answer to that question is, in my view, a question of the interpretation of the policy, which is to be interpreted in accordance with the normal principles applicable to the interpretation of contracts. As the Supreme Court put it in Financial Conduct Authority v Arch Insurance (UK) Ltd [2021] UKSC 1, [2021] AC 649:

“The core principle is that an insurance policy, like any other contract, must be interpreted objectively by asking what a reasonable person, with all the background knowledge which would reasonably have been available to the parties when they entered into the contract, would have understood the language of the contract to mean. Evidence about what the parties subjectively intended or understood the contract to mean is not relevant to the court’s task … In the case of an insurance policy of the present kind, sold principally to SMEs, the person to whom the document should be taken to be addressed is not a pedantic lawyer who will subject the entire policy wording to a minute textual analysis (cf Jumbo King Ltd v Faithful Properties Ltd (1999) 2 HKCFAR 279, para 59). It is an ordinary policyholder who, on entering into the contract, is taken to have read through the policy conscientiously in order to understand what cover they were getting.”

In many cases the event insured against (or peril) will be defined in terms of some physical event (e.g. fire, storm, flood etc). If a peril is so defined, then as soon as a fire happens or a flood takes place, the insured’s cause of action arises.

But in my view that is not how the insured events are defined in Option 1. The first defined event is “if you lose the amount paid to the contractor in accordance with the building contract.” That is a financial event, namely the loss of a payment already made. The second defined event is “[if you] … have to pay more to complete the building of the home(s).” That, too is a financial event. Both these events are qualified by a causation requirement, namely that the loss (or additional payment) is “because the contractor is insolvent or commits fraud.” This cuts down what would otherwise be the insured event. It is necessary for it to do so in order to avoid the risk that NHBC might be liable on the policy if, for example, the cost of works rises because of variations in the specification. Thus, on the facts that are relevant to this appeal, Option 1 will apply if (a) the employer has to pay more to complete the building of the homes and (b) the reason why it has to do so is the insolvency of the contractor. The second criterion cannot be divorced from the first.

NHBC placed some reliance on the title of Option 1. I consider that there is no real help to be derived from that. In the first place Option 1 covers not only loss caused by insolvency, it also covers loss as a result of the contractor’s fraud. Second, for unexplained reasons, the definition of “insolvent” includes the death of the contractor. It does not matter for this purpose whether the contractor died solvent or insolvent.

That the cause of action does not arise immediately on the contractor’s insolvency is, in my view, confirmed by [C] of Option 1. That box does not instruct the insured to notify NHBC of the contractor’s fraud or insolvency. Rather in the case of the first specified event it requires the insured to notify NHBC “if you have lost the amount you paid to the contractor”; that is to say on the happening of the financial event. In relation to the second specified event, the insured is required to notify NHBC if “the contractor has not completed the home(s)”. Once again, this does not require the insured to notify NHBC of the contractor’s insolvency. NHBC argue that non-completion of the homes will occur on the contractor’s insolvency, not on the employer having to pay more. That, to my mind, is no more than an assertion. As the judge said at [9] and [56], some of the defined insolvency events might have no impact on the eventual construction costs. Or the work might be completely abandoned and not restarted. In addition, bearing in mind that “insolvent” includes a case where the contractor dies solvent, that assertion does not tally with the definitions in the policy.

NHBC next argue that Option 1 applies “if you have to pay more”, not “if you have paid more”. I agree; but I do not see how it advances the case that insolvency alone is the insured event. NHBC assert that the necessity of paying more arises on the contractor’s insolvency. But that, too, is an untested factual assertion. The employer might find an alternative contractor who is actually cheaper than the insolvent one.

NHBC next argue that under the heading “What we will do” they will pay the reasonable extra cost of completing the homes, which is calculable from the moment of insolvency. Again, I do not regard this as self-evidently correct. I would accept that at the date of the insolvency it may well be possible to calculate the reasonable cost of completing the homes, but NHBC’s liability is limited to the reasonable extra cost of completion. In order to calculate the extra costs it would be necessary to have a comparator. That may not be known at the date of the insolvency or, indeed, until the employer had entered into alternative contractual arrangements. Moreover the commercial consequences of this interpretation seem to me to be surprising. If the argument is correct, NHBC would be immediately liable to pay the reasonable costs of completing the homes even if the homes were never completed; or if they were in fact completed at no extra cost.

There are also difficulties in reconciling Option 1 with other parts of the policy. Part C of the policy, for example, (headed “Home damage cover”) states:

“This section applies if there is physical damage to the home(s) because the contractor failed to build the following parts of the home(s) to comply with the NHBC requirements [they are then set out]”

But in the Box “Conditions and limitations” the policy provides:

“You cannot claim for something under this section if you can claim for it under … Option 1, or if you could have done when you first knew about it.”

If the insured event is, as NHBC argues, the contractor’s insolvency, I find it difficult to understand how a claim under Option 1 would interact with a claim under Part C.

NHBC relied on a number of consequences of the judge’s interpretation which they said were contrary to commercial common sense. Commercial common sense does have a role to play in the interpretation of contracts, but its importance must not be overstated. In Arnold v Britton [2015] UKSC 36, [2015] AC 1619 Lord Neuberger warned against over-reliance on commercial common sense. At [17] he said:

“First, the reliance placed in some cases on commercial common sense and surrounding circumstances … should not be invoked to undervalue the importance of the language of the provision which is to be construed. The exercise of interpreting a provision involves identifying what the parties meant through the eyes of a reasonable reader, and, save perhaps in a very unusual case, that meaning is most obviously to be gleaned from the language of the provision. Unlike commercial common sense and the surrounding circumstances, the parties have control over the language they use in a contract. And, again save perhaps in a very unusual case, the parties must have been specifically focussing on the issue covered by the provision when agreeing the wording of that provision.”

The point is all the stronger in a case like the present, where the Buildmark Choice policy is a non-negotiable standard form of policy drafted entirely by NHBC.

At [19] he said that commercial common sense should not be invoked retrospectively; and at [20] he said:

“Fourthly, 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, even ignoring the benefit of wisdom of hindsight. The purpose of interpretation is to identify what the parties have agreed, not what the court thinks that they should have agreed. Experience shows that it is by no means unknown for people to enter into arrangements which are ill-advised, even ignoring the benefit of wisdom of hindsight, and it is not the function of a court when interpreting an agreement to relieve a party from the consequences of his imprudence or poor advice. Accordingly, when interpreting a contract a judge should avoid re-writing it in an attempt to assist an unwise party or to penalise an astute party.”

There are, as Coulson LJ has shown, uncommercial consequences of NHBC’s interpretation of the policy which balance, if they do not outweigh, the uncommercial consequence suggested by NHBC. I prefer, therefore, to rely on the words of the policy.

Ultimately, NHBC say that the event insured against is contractor insolvency. No doubt it could have been defined in that way (although bearing in mind that insolvency includes the death of the contractor it would have been some form of hybrid life insurance policy as well). But the fact is that it was not. I agree, therefore, that ground 1 must fail.