Drax Energy Solutions Limited (formerly Haven Power Limited) v Wipro Limited

This is the trial of two preliminary issues concerning the interpretation of a limitation of liability clause (“the Clause”) contained in a Master Services Agreement made between the Claimant, Drax Energy Solutions Ltd (“Drax”) and the Defendant, Wipro Ltd (“Wipro”) dated 20 January 2017 (“the MSA”).

Drax issued proceedings against Wipro on 2 September 2021, and the main trial is due to commence on 1 October 2024, with a time estimate of 36 days.

The MSA concerned the provision of software services by Wipro for Drax, an energy supplier. The new IT system to be provided, based on Oracle software, included customer relationship management, billing and smart metering facilities for Drax’s business.

Under Schedule 10 of the MSA, the core services to be provided by Wipro under numbered “statements of work” (“SOW”) were as follows:

There were then agreed go-live dates for the various SOWs.

The agreed charges for SOW1-4 were as follows (rounded up):

In addition, there was a separate Insights SOW to be provided by 30 September 2017. £304,000 was to be paid in the first year and £152,000 per annum for the next 3 years.

It is common ground that the total SOW charges payable in the first 12 months of the MSA were £7,671,118. Wipro contends that the total charges for each of the succeeding 4 years would be:

On any view, the project was not a success though the reasons for this are very much in dispute. Milestones were missed, rearranged and missed again and Drax alleges that it had to spend very large sums of money to render acceptable the deliverables provided by Wipro. In the end, Drax terminated the MSA on 7 August 2019 for what it says were repudiatory breaches on the part of Wipro.

Before turning to the Clause itself and the Preliminary Issues, I need to set out the nature and structure of Drax’s claim and, critically, the amounts claimed. I should say at the outset that Drax's total quantified claim is for some £31 million. There is then a counterclaim by Wipro of around £10 million. The counterclaim includes £5.5 million by way of damages for wrongful termination, £1.28 million for prolongation costs, unpaid invoices of £1.5 million and termination claims of £2.4 million.

The total charges made by Wipro under the MSA and which were paid by Drax amounted to £4.9 million.

The claim made by Drax breaks down into 4 categories:

Here, Drax alleges that, but for the representations which were false, it would not have entered into the MSA and the SOWs at all. It has therefore lost its entire net expenditure on the project being £31.7 million.

The losses claimed here are set out in the Schedule to the Particulars of Claim (“POC”) under two headings - “expenditure caused by pre-termination breaches” and “Balance of expenditure wasted by reason of the termination”. As those titles suggest, only the first heading describes losses said to be attributable directly to the Quality breaches of the contract themselves.

These losses are made up of costs incurred by Drax over and above what it would have to have incurred in any event, in relation to dealing with Wipro’s work product. In other words, it spent more than it should have done to make the work product contractually acceptable and useful. Those losses come to some £9.8 million.

A similar approach is taken to the description of losses here. The losses caused by the various delays, in breach of contract, are put at £9.7 million.

The maximum claim made in respect of the damages claimed is £31.6 million. This comprises all of the expenditure on the MSA and SOWs. However, there is a further item, claimed separately, in the sum of £126,195. This relates to exit planning costs in excess of what Drax would reasonably have incurred had the MSA been properly performed. This item is included in the £31.7 million claimed in respect of the Misrepresentation Claim.

If the Quality and Delay Claims are made out in full, then they would amount to some £19.5 million which would be deducted from the £31.6 million referred to above leaving a figure of just over £12 million. This was itself made up of the £4.4 million of charges claimed by Wipro and paid by Drax, some other costs that might be described as “reliance expenditure” and then the costs listed under the second description of Quality and Delay Claim losses in the Schedule to the POC. These were the expected costs, assuming that the MSA had run its course but were now wasted due to its wrongful termination by Wipro. These amounted to £3.7 million in respect of quality and £827,278 in respect of Delay, making a total of just over £4.5 million.

If any of the first category of Quality and Delay Claim costs were not recovered by reference to such alleged breaches, then they are claimed as a further element of the Termination Claims, effectively as wasted expenditure.

It follows from the above that on the basis of Drax’s primary case on its claims, they are as follows, using round figures:

I set them out in this order because if one excludes the Misrepresentation Claim, there is no overlap between any of the other 3 claims. They each seek different and separate losses. The way the Misrepresentation Claim is framed, however, means that different parts of it do overlap with each of the other 3 claims.

Of course, these are Drax’s maximum quantified figures and any or all of them could be less, perhaps significantly less, even if the underlying liability is established.

I should add that in the alternative to almost all of the Termination Claim, there is an alternative claim set out at paragraph 79 of the POC. This alternative claim has not yet been quantified and does not form part of the analysis proffered by both sides for the purposes of determination of the Preliminary Issues. I therefore say no more about it.

The Clause itself is clause 33.2 of the MSA. However, it is necessary to set it out in context, which means reciting the first 6 sub-paragraphs of clause 33, headed “Liability”, as follows, with bold added for the Clause.

“33.

LIABILITY

33.1

Subject to clauses 33.5 and 33.6, the Supplier’s liability to the Customer, whether in contract, tort (including negligence) for breach of statutory duty or otherwise, for loss or damage to tangible property arising out of or in connection with this Agreement (including all Statements of Work) shall be limited to £20 million per event or series of connected events.

33.2

Subject to clauses 33.1, 33.3, 33.5 and 33.6, the Supplier’s total liability to the Customer, whether in contract, tort (including negligence), for breach of statutory duty or otherwise, arising out of or in connection with this Agreement (including all Statements of Work) shall be limited to an amount equivalent to 150% of the Charges paid or payable in the preceding twelve months from the date the claim first arose. If the claim arises in the first Contract Year then the amount shall be calculated as 150% of an estimate of the Charges paid and payable for a full twelve months.

33.3

The Supplier’s total aggregate liability arising out of or in relation to this Agreement for any and all claims related to breach of any provision of clause 21 whether arising in contract (including under an indemnity), tort (including negligence), breach of statutory duty, laws or otherwise, shall in no event exceed 200% of the Charges paid or payable in the preceding twelve months from the date the claim first arose or £20m (whichever is greater).

33.4

Subject to clauses 33.5 and 33.6, the Customer’s total liability to the Supplier, whether in contract, tort (including negligence), for breach of statutory duty or otherwise, arising out of or in connection with this Agreement (including all Statements of Work) shall be limited to the Charges payable under this Agreement (including all Statements of Work) in respect of Services properly performed by the Supplier at the point the claim is made.

33.5

Subject to clause 33.6, neither party will be liable to the other party for any indirect, consequential or special loss including but not limited to any loss of profits or loss of goodwill arising out of, or in connection with, this Agreement or any Statement of Work.

33.6

Nothing in this Agreement or any Statement of Work shall exclude or limit:

33.6.1

either party’s liability for death or personal injury caused by its (or its agent’s or sub-contractor’s) negligence or for fraud or fraudulent misrepresentation;

33.6.2

the Supplier’s liability, whether categorised as direct or indirect losses, to the Customer arising out of a breach of clause 4.2.4 (Compliance with Laws), clause and/or where the Supplier has indemnified the Customer in clause 19 (IPR Indemnity), and/or arising out of a breach of clause 26 (Confidentiality and Announcements) and/or clause 37 (Anti corruption);

33.6.3

the Supplier’s liability for wilful misconduct or abandonment; or

33.6.4

either party’s liability that cannot, as a matter of law, be limited or excluded.”

These are as follows:

Drax says that if these preliminary issues are determined in its favour, the effect of the cap is to reduce Wipro’s maximum possible liability in these proceedings down from £31.7m (see paragraphs 13 and 17 above) to around £23m. This is because, while the claims in respect of Quality and Delay would fall below the relevant caps for each, and those claims are worth £19.5m, there would be an applicable cap of £3.78m on the Termination Claims of £12 million (see paragraph 18 above).

On the other hand, if the preliminary issues are determined in Wipro’s favour, its maximum liability would be limited to around £11.5m if ‘the claim’ is held to have arisen in the first contract year, but much less so if it arose subsequently.

For present purposes, both parties agree that (a) the different limits set out in clauses 33.1 and 33.3 do not apply to any of the claims here and (b) the Misrepresentation, Quality and Delay Claims arose in the first year. The latter means that the single overall cap contended for by Wipro would be around £11.5 million. The cap would only be reduced if none of those claims succeeded but the Termination Claims succeeded, at least on Wipro’s case.

In its Defence, Wipro invokes the Clause against Drax, in summary, as follows:

For its part, Drax contends, first, that the £11.5 million figure is a limit which applies to each and every separate claim, assuming they all arose in the first year. It is not a single maximum applied to all claims.

As for what constitutes a “claim”, for these purposes, Drax’s primary position is that a claim means “cause of action”. On that footing, a relevant claim for present purposes could be each and every item in the Schedule to the POC which has a particular loss ascribed to it. That would entail 14 claims in respect of quality, 19 claims in respect of Delay and 23 claims in respect of Termination (assuming no overlap with the Quality and Delay Claims). The Misrepresentation Claims would be more than one since, on any view, different Representations are relied upon, which are said to be false.

In fact, Drax did not go quite that far. In paragraph 5 (a) to (p) of its Further Information, dated 8 July 2022, it itemised:

However, Drax has an alternative position on the meaning of “claim” which was the real focus of oral argument. As set out in paragraph 5 (q) – (t) of the Further Information, this is on the basis that the relevant claims were 4 in number, being the Misrepresentation Claim, the Quality Claims, the Delay Claims and the Termination Claims.

This would mean that, if established, the Quality and Delay claims would not be affected by the Clause since (a) they arose in the first year and (b) were for less than the £11.5 million applicable to each of them. However, the Misrepresentation claim, which clearly arose in the first year, would be affected by the Clause because Drax could only recover £11.5 million out of a maximum sum claimed here of £31 million. And as for the Termination Claims, these would also be affected by the Clause with an applicable cap of £3.78 million, as against a claim of around £12 million.

There are two distinct areas of law to consider. The first concerns the proper approach to contractual interpretation generally and provisions like the Clause in particular. The second concerns the proper approach to the interpretation of the expression “claims” in the context of the Clause. I deal here with the first area. The second is more conveniently dealt with in context, when I come to Preliminary Issue 2.

I consider that the effect of the well-known series of cases in the House of Lords and Supreme Court culminating in Wood v Capita [2017] UKSC 24 has been helpfully summarised by the then Chancellor, Sir Geoffrey Vos in Lamesa v Cynergy [2020] EWCA Civ 821, as follows:

“i)

The court construes the relevant words of a contract in their documentary, factual and commercial context, assessed in the light of (i) the natural and ordinary meaning of the provision being construed, (ii) any other relevant provisions of the contract being construed, (iii) the overall purpose of the provision being construed and the contract or order in which it is contained, (iv) the facts and circumstances known or assumed by the parties at the time that the document was executed, and (v) commercial common sense, but (vi) disregarding subjective evidence of any party’s intentions – see Arnold v. Britton [2015] UKSC 36 [2015] AC 1619 per Lord Neuberger PSC at paragraph 15 and the earlier cases he refers to in that paragraph;

ii)

A court can only consider facts or circumstances known or reasonably available to both parties that existed at the time that the contract or order was made - see Arnold v. Britton (ibid.) per Lord Neuberger PSC at paragraph 20;

iii)

In arriving at the true meaning and effect of a contract or order, the departure point in most cases will be the language used by the parties because (a) the parties have control over the language they use in a contract or consent order and (b) the parties must have been specifically focussing on the issue covered by the disputed clause or clauses when agreeing the wording of that provision – see Arnold v. Britton (ibid.) per Lord Neuberger PSC at paragraph 17;

iv)

Where the parties have used unambiguous language, the court must apply it – see Rainy Sky SA v. Kookmin Bank [2011] UKSC 50 [2011] 1 WLR 2900 per Lord Clarke JSC at paragraph 23;

v)

Where the language used by the parties is unclear the court can properly depart from its natural meaning where the context suggests that an alternative meaning more accurately reflects what a reasonable person with the parties’ actual and presumed knowledge would conclude the parties had meant by the language they used but that does not justify the court searching for drafting infelicities in order to facilitate a departure from the natural meaning of the language used – see Arnold v. Britton (ibid.) per Lord Neuberger PSC at paragraph 18;

vi)

If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other – see Rainy Sky SA v. Kookmin Bank (ibid.) per Lord Clarke JSC at paragraph 2 - but commercial common sense is relevant only to the extent of how matters would have been perceived by reasonable people in the position of the parties, as at the date that the contract was made – see Arnold v. Britton (ibid.) per Lord Neuberger PSC at paragraph 19;

vii)

In striking a balance between the indications given by the language and those arising contextually, the court must consider the quality of drafting of the clause and the agreement in which it appears – see Wood v. Capita Insurance Services Limited [2017] UKSC 24 per Lord Hodge JSC at paragraph 11. Sophisticated, complex agreements drafted by skilled professionals are likely to be interpreted principally by textual analysis unless a provision lacks clarity or is apparently illogical or incoherent– see Wood v. Capita Insurance Services Limited (ibid.) per Lord Hodge JSC at paragraph 13; and

viii)

A court should not 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, because it is not the function of a court when interpreting an agreement to relieve a party from a bad bargain - see Arnold v. Britton (ibid.) per Lord Neuberger PSC at paragraph 20 and Wood v. Capita Insurance Services Limited (ibid.) per Lord Hodge JSC at paragraph 11.”

The leading decision on this subject now is that of the Supreme Court in Triple Point Technology v PTT [2021] AC 1148. One of the points at issue here concerned a clause which limited the liability of the contractor to the contract price. However, that limitation did not apply to any liability of the contractor resulting from “fraud, gross negligence, negligence or wilful conduct…” This was in a context where the underlying cause of action could only be breach of contract. The question was whether the word “negligence” here meant a breach of the contractual duty to take reasonable care and skill, as the claimant contended, or whether it was to be interpreted more narrowly so as to mean some other breach of duty of care in tort which did not give rise to a coterminous breach of the contractual duty. This was the position of the contractor. If correct, the claim for £14 million would have a cap of about £1 million.

Lord Leggatt, with whom the other Justices agreed, considered that in context, the word “negligence” here clearly had the former wider meaning and therefore the claim was not subject to the limitation. He did not consider that the construction contended for by the contractor was even a possible interpretation. However, he went on to say that even if it had been:

“106…a further reason for giving the word “negligence” its straightforward and ordinary legal meaning is that clear words are necessary before the court will hold that a contract has taken away valuable rights or remedies which one of the parties to it would have had at common law (or pursuant to statute).

107 The approach of the courts to the interpretation of exclusion clauses (including clauses limiting liability) in commercial contracts has changed markedly in the last 50 years. Two forces have been at work. One has been the impact of the Unfair Contract Terms Act 1977, which provided a direct means of controlling unreasonable exclusion clauses and removed the need for courts to resort to artificial rules of interpretation to get around them:… The second force has been the development of the modern approach in English law to contractual interpretation, with its emphasis on context and objective meaning and deprecation of special rules of interpretation encapsulated by Lord Hofmann’s announcement in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896, 912 that “almost all the old intellectual baggage of legal interpretation has been discarded”.

108.

The modern view is accordingly to recognise that commercial parties are free to make their own bargains and allocate risks as they think fit, and that the task of the court is to interpret the words used fairly applying the ordinary methods of contractual interpretation. It also remains necessary, however, to recognise that a vital part of the setting in which parties contract is a framework of rights and obligations established by the common law (and often now codified in statute). These comprise duties imposed by the law of tort and also norms of commerce which have come to be recognised as ordinary incidents of particular types of contract or relationship and which often take the form of terms implied in the contract by law. Although its strength will vary according to the circumstances of the case, the court in construing the contract starts from the assumption that in the absence of clear words the parties did not intend the contract to derogate from these normal rights and obligations.”

Accordingly, while the modern approach is not to employ some special rule of interpretation when dealing with exclusion or limitation clauses, the fact that this is what the relevant provision does may still have a contextual role to play when considering what the parties are to be taken to have intended objectively.

On the facts of Triple Point, Lord Leggatt found that the operation of the limitation clause in question, if “negligence” was construed according to the contractor’s interpretation, would be a “substantial departure” from what would otherwise be the normal position; that is, because of the effect of the clause reducing the claim from £14 million to £1 million.

A similar approach had been taken earlier, by Briggs LJ (as he then was) in Nobahar-Cookson v Hut Group [2016] EWCA Civ 128. Here, the Court of Appeal adopted a narrow interpretation of a clause which required a claim made under the share sale agreement to be notified within 20 days of the buyer “becoming aware of the matter”. This was in circumstances where there was an ambiguity as to the correct interpretation of that phrase. He said this:

“18.

In my judgment the underlying rationale for the principle that, if necessary to resolve ambiguity, exclusion clauses should be narrowly construed has nothing to do with the identification of the proferens, either of the document as a whole or of the clause in question. Nor is it a principle derived from an identification of the person seeking to rely upon it. Ambiguity in an exclusion clause may have to be resolved by a narrow construction because an exclusion clause cuts down or detracts from the ambit of some important obligation in a contract, or a remedy conferred by the general law such as (in the present case) an obligation to give effect to a contractual warranty by paying compensation for breach of it. The parties are not lightly to be taken to have intended to cut down the remedies which the law provides for breach of important contractual obligations without using clear words having that effect:…

19.

This approach to exclusion clauses is not now regarded as a presumption, still less as a special rule justifying the giving of a strained meaning to a provision merely because it is an exclusion clause. Commercial parties are entitled to allocate between them the risks of something going wrong in their contractual relationship in any way they choose. Nor is it simply to be mechanistically applied wherever an ambiguity is identified in an exclusion clause. The court must still use all its tools of linguistic, contextual, purposive and common-sense analysis to discern what the clause really means. In the Seadrill Management case Moore-Bick [LJ] described the principle as, ‘essentially one of common sense; parties do not normally give up valuable rights without making it clear that they intend to do so…

21.

For those legal reasons I approach the issue as to the construction of clause 5.1 upon the basis that there remains a principle that an ambiguity in its meaning may have to be resolved by a preference for the narrower construction, if linguistic, contextual and purposive analysis do not disclose an answer to the question with sufficient clarity.”

At paragraph 29, Briggs LJ said that as a matter of language, the provision before him was not clear and

“The natural meaning of the language is by no means so clear as to preclude serious consideration of the commerciality or otherwise of rival interpretations or, for that matter, to preclude recourse to the principle that ambiguous exclusion clauses should be construed narrowly.”

He therefore found that the narrow interpretation, which meant that notice had been given in time, should be preferred.

Finally, I turned to Royal Devon and Exeter NHS Trust v ATOS [2017] EWCA Civ 2196. This case is potentially relevant since on the facts, there are at least some similarities with the case before me. It concerns the interpretation of a limitation of liability clause in the context of claims brought by the employer against the contractor for breach of contract for the provision of a software system for the Trust.

Clause 8.1.2 (b) of the contract provided that:

“(b)

the aggregate liability of either party under the Contract for all Defaults, other than those

governed by sub-clause 8.1.2 (a) above, shall not exceed the amount stated in schedule G to be the

limit of such liability.”

Paragraph 9.2 of schedule G then provided that:

“…The aggregate liability of the Contractor in accordance with sub-clause 8.1.2 paragraph (b) shall not exceed:

9.2.1

for any claim arising in the first 12 months of the term of the Contract, the Total Contract Price as set out in section 1.1; or

9.2.2

for claims arising after the first 12 months of the Contract, the total Contract Charges paid in the 12 months prior to the date of that claim.”

The question then was whether paragraph 9.2 imposed one cap, or two. At first instance, O’Farrell J held that there was a single cap, focusing in particular on the use of the expression “aggregate liability” at the beginning of paragraph 9.2 and the later use of the word “or”. She then held that the relevant single cap would be determined by whenever the first claim arose. If in the first 12 months, it would be the higher amount specified by paragraph 9.2.1 - if later, it would be the lower cap set out in paragraph 9.2.2.

The Court of Appeal disagreed and held that paragraph 9.2 provided a system of two separate caps, one for any and all claims which arose in the first year, and a second for all claims arising subsequently. Jackson LJ considered that while it was right to have a higher cap for Year 1, which was when most of the value was to be provided (as in this case), this did not mean that the contractor should have a “free ride” in subsequent years such that there was no further margin for claims arising then, which would be the case if there was only one cap.

The case was obviously not straightforward because one can see an argument that the expression “aggregate liability” in both Clause 8.1.2 (b) and then in paragraph 9.2, suggests one overall cap for all claims. Jackson LJ however considered that this was not determinative and the use of the expression in 9.2 itself could simply refer to the aggregate of both caps. He did this in a context where he observed that paragraph 9.2 was itself a “home-made clause” (unlike the underlying contract which was in standard form), and which, on any interpretation, would yield some odd results. He said that the natural meaning, and the one which yielded “the least bizarre consequences” was the one which he adopted.

It is of some interest that it was not suggested to the Court of Appeal (though it had been at first instance) that paragraph 9.2 was in fact meaningless and therefore could not be enforced at all. Nor was it contended for the claimant on appeal that there was in fact a separate cap for each separate claim whenever it arose, not one cap for group of Year 1 claims and another for those arising subsequently. The claimant had so contended at first instance but O’Farrell J rejected this argument at paragraph 86 of her judgment. She made the point that if there was a separate cap for each claim, the potential, very large, total cap would render the clause devoid of any real purpose.

On analysis, I do not think that, in terms of the law, Royal Devon adds very much, save that the Court will sometimes have to choose between two interpretations neither of which make complete sense or are free from odd or unusual consequences. To use Jackson LJ’s graphic words, it is then a question of adopting the interpretation which produces the “least bizarre” results. And this may be in circumstances where it is not clear that the ultimate interpretation is what either party actually had in mind. And further that it is legitimate to see whether one party’s interpretation would deprive the relevant limitation clause of any real purpose because the operative cap would be so high.

Although there are two separate Preliminary Issues they are obviously connected. Each therefore needs to be considered in the context of the other. So, for example, if Drax was correct that a “claim” meant any cause of action (or at least those 16 enumerated in its primary case as set out in the Further Information) that would potentially allow for a larger amount to be claimed, taking the cap into account, as opposed to the position if the narrower and alternative definition of “claim” was correct.

That said, because even the wider interpretation contended for by Drax treats the entire Misrepresentation and Termination Claims as each constituting only one claim (excluding the Exit Plan Claim for the moment) there is in practice not much difference between the effect of the two formulations on the facts of the case here. But since Drax has not abandoned its primary case on “claim” I will need to deal with it below.

However, I should say now that my conclusion on Drax’s wider interpretation of “claim” is that it is not correct. In particular, the expression “claim” cannot here be simply equated with “cause of action”. See paragraphs 98 to 107 below.

This means that the only interpretation of “claim” proffered by Drax is the narrower one whereby there are 4 relevant claims only. Although, in its written submissions, Drax invited me to consider some other interpretation of “claim” which could operate if I was against Drax on the narrow interpretation too, I decline to do so. It seems to me that I should proceed on the basis of the actual position of the parties. Wipro’s position, of course, is that (a) there is only one claim anyway but (b) even if not, the Clause imposes one single cap in relation to all claims, however described and however numerous - unless, of course, any of them fall within 33.1 or 33.3.

I start first with the language of the Clause itself. Taken in isolation, if one examines the first 3 lines of the Clause and the words “limited to” and stops there, the language strongly suggests that this is a cap for all claims. The phrase “total liability” supports that reading, as well as the absence of words like “for each claim” after the word “liability”. So if there had been here a specific sum stated after the words “limited to”, for example £10 million or £20 million, rather than the formulae which in fact follow them, the language would clearly indicate a single cap.

However, one has to have regard to the actual words used after the words “limited to”. The formulae used refer to when “the claim” first arose, or arises. If “claim” is to be interpreted as meaning each claim that has arisen, then this would suggest that the relevant formula was to be applied to each such claim, in which case there would be a number of separate caps. However, one might then expect language which speaks of when “each claim” arose. Indeed, the absence of such wording is what Wipro submits is fatal to Drax’s position on the meaning of the Clause (along with other matters). There is clearly force in that point.

Wipro then adds to this argument by saying that “claim” here means no more than “liability”. In fact, as already noted, and for the reasons given in paragraphs 98 to 107 below, I reject that interpretation. However, that is not fatal to Wipro’s position because of the alternative way of looking at “claim” which is that the reference to when “the claim first arose” really means when the first [of the various] claim[s] arose. On that interpretation, although there could be several claims, the cap is still for the total of those claims and the timing of the first of those claims simply sets the appropriate limit. Of course, Drax then makes the point that those actual words were not used and they could have been. There is some force in this point, too.

Nonetheless if one paused the analysis here, I think that on balance, the language overall favours Wipro’s interpretation over Drax’s.

One then turns to the language in the other relevant provisions of the MSA and in particular, other parts of Clause 33 to see if that assists.

The most important of the other clauses to consider is clause 33.3. It is common ground that this imposes a single cap for all claims relating to a breach of clause 21 of the MSA. That provision deals with data protection.

Two particular features of this provision stand out. First, there is again the reference to the date when “the claim first arose”. This is in circumstances where it is accepted that, while there could be more than one claim (as contemplated by the words “… any and all claims…”) there is a single cap for all claims. But if so, the expression “the claim first arose” must mean when the first of the claims (first) arose, even though that is not explicitly stated. This would then suggest that the same expression in the Clause should be interpreted in the same way (see paragraph 58 above), since the same words are used in each provision. That would therefore provide added support for view of the Clause as imposing a single cap.

On the other hand, one has here the words “aggregate” and “for all and any claims” which are absent from the Clause. That may suggest that absent those words, it is at least not clear that a single cap for all claims is contemplated by the Clause.

Of course, one needs to add a note of caution here. On any view, I do not consider that Clauses 33.2 and 33.3 are well-drafted. It may also be that they were not drafted at the same time so as to form a discrete collection of provisions within clause 33. I note for example, that there is the same basic language employed in Clauses 33.1, 33.2 and 33.4 whereas the language used in clause 33.3 is somewhat different. So what might be said to be the logical implication of the language used in clause 33.3, for the proper interpretation of clause 33.2, does not necessarily follow.

There is another aspect to the relevance of clause 33.3, however, and it is this. It clearly imposes a maximum of £20 million for all clause 21 claims. The Clause is said to be “subject to” clause 33.3, along with clauses 33.1, 33.5 and 33.6. But if the Clause itself imposed a single maximum cap of £11.5 million, it does not make much sense to say that it is “subject to” clause 33.3 which allows for a much greater cap for only one group of claims, the liability for which “shall in no event exceed” the £20 million. One could in fact make the same point about clause 33.1, even though it preceded the Clause.

On the other hand, if the Clause imposed a separate cap for each claim, and on the assumption that a clause 21 claim would prima facie fall within the Clause to begin with, clause 33.3 may indeed cut down what was otherwise recoverable under clause 33.2, depending, of course, on how a “claim” is defined. On any view, clause 33.3 could be more easily seen as a theoretical limiting factor, to which the Clause was indeed “subject”.

Drax says that the above features of clause 33.2 and 33.3 support its position that the Clause does indeed provide for separate caps. It adds that clause 33.3 itself does not explicitly state that it is subject to, for example, clauses 33.5 and 33.6. I have little doubt that in fact it is, and the terms of those clauses are quite explicit. But Drax’s point is that there would be no need to make an express reference in clause 33.3 to it being subject to the other clauses if, in effect, it should be looked at as a proviso to or, in effect, a sub-clause of, the Clause even if not expressed as such. All of that would then support the notion that Clause 33.3 does indeed operate as a limiting factor to what might otherwise be recoverable under the Clause, with its separate caps.

Having said that, there of course remains the fact of the similar language in the Clause and clause 33.3 concerned with when the claim “first arose”. On Drax’s interpretation, that expression would have to be construed differently in each provision. The linguistic tension resulting from that is heightened if Clause 33.3 is itself to be construed simply as a proviso to the Clause.

Further, in argument, Drax accepted that clause 33.1 can be seen as freestanding and not as some other proviso to the Clause. On that basis, although the words “shall be limited to” are used, in substance, the meaning here is that the relevant liability is “up to” £20 million per event. But if so, there is no reason not to view clause 33.3 in the same light. In other words, its wording does not necessarily imply that it is simply a proviso to the Clause on the footing that the clause 21 liabilities would be theoretically higher under the latter.

I also do not think that the absence of the specific expression “subject to” the various other clauses, in clause 33.3 is really material. It is subject to them, anyway.

At the end of the day, so far as clause 33.3 is concerned, and notwithstanding possible drafting infelicities and applying the caution I referred to in paragraph 64 above, I do think that considerable weight should be given to the use of the same expression about when the claim “first arose”. This militates against Drax’s interpretation of the Clause, especially where there is no explicit “per claim” form of words used therein.

On the other hand, while some weight should be given to the fact of the use of different words like “aggregate” and “any and all claims” in clause 33.3, I have concluded that while the language is more emphatic here, it is not really a significant pointer to the Clause not being a single cap This is especially so where the word “total” is used before the word “liability”. On that latter point, I do not regard the use of the word “total” as effectively just unnecessary boilerplate language as Drax has suggested.

I think it is worth adding, on clause 33.1, that it does show that where it was intended to create a number of caps each applying to something, the parties were quite capable of using explicit language like “per event”. They could have done so in respect of the Clause and individual claims, but they did not.

In my judgment, overall, the considerations arising from clause 33.3 constitute a clear indicator that the Clause is to be interpreted as imposing a single, not multiple, caps.

Finally, I should here address a separate point made by Wipro which refers to the insurance provisions set out at clause 35. This states that Wipro will at its own cost be solely responsible for taking out and maintaining policies of insurance covering its liabilities arising out of its acts or omissions in connection with the MSA. Such policies had to include:

“35.1.1

employers liability insurance for a maximum amount of cover of £10 million on a per occurrence and an annual aggregate basis;

35.1.2

professional indemnity insurance for a minimum amount of cover of £20 million on a per occurrence and an annual aggregate basis;

35.1.3

public liability insurance for a minimum amount of cover of £30 million on a per occurrence and an annual aggregate basis; and

35.1.4

product liability insurance for a minimum amount of cover of £30 million on a per occurrence and an annual aggregate basis.”

The relevant insurance for present purposes is the professional indemnity insurance. Here, Wipro contends that the fact that all of the required policies provided for a minimum amount of cover in specific sums indicates that the parties were not contemplating a separate cap for each individual claim (however defined, but especially if it meant a cause of action) on which depended Wipro’s overall liability, as opposed to caps with a single value for example £11.5 million.

I doubt whether this inference can really be drawn. Indeed, I suspect that the particular liability contemplated by clause 35.1.2 was that dealt with by clause 33.1.

Accordingly, the references to insurance do not take the matter any further.

Overall, therefore, on the question of the language used, both in and around the Clause, as it were, I consider that despite some linguistic quirks, if I can put it like that, the correct interpretation is that there is a single cap in the Clause, as Wipro contends, and not separate caps for each claim. This is also in the context of where both parties are large corporations which obviously had professional advice and assistance in the making of the MSA. I reach that conclusion on the basis that the number of claims is 4 and not 16.

That said, I would accept that Drax’s claimed interpretation is at least possible, though in my view wrong. I do not consider that the language is, ultimately, ambiguous, however. Compare the position in Nobahar where the issue turns on the meaning to be ascribed to the single word “becoming aware of the matter”. That seems to me to be a different sort of interpretive problem than that arising here.

There may be interesting questions as to whether the existence of a different possible interpretation is the same as “ambiguity” but they do not need to be resolved here.

That is because, in my judgment, and following Triple Point, I should in any event deal with two further matters, being (a) commercial (or “business common sense”) considerations and (b) what I shall refer to as Triple Point considerations, as part of the overall exercise of interpretation.

Drax submits that Wipro’s interpretation would lead to some surprising results. In particular, Drax points out that the MSA contemplated not just the provision of the initial “Planets” suite of software over 5 years by Wipro under SOWs 1-5, but also potential further works which could give rise to further claims. That is, in theory, true because, as Schedule 10 made clear, while SOWs 1-5 fall under the Planets Programme, the parties also intended that the MSA could be used to order further SOWs not related to that programme. Indeed, one such SOW was the separate Insights SOW agreed on 18 September 2017, referred to above, and which was concerned with cyber-security and the dark web. All of this is reflected in the fact that under clause 2, there is in theory no limit to the length of the MSA, albeit that the Planets Programme itself was to be delivered and maintained over a 5 year period.

Further, the MSA provides that other companies in the same group as Drax could themselves execute SOWs and benefit from the services provided by Wipro under that SOW. See, for example, the reference to Drax Retail in connection with SOW4 .1.

Drax also says that, given the possibility of many claims going into the future if other projects are agreed, a gradually reducing single cap would not reflect the fact that later SOWs may be for much greater value, yet they could be “stuck with” a cap based on charges made in earlier years, if the first claim arose then. The maximum that could be claimed would be the £11.5 million but less if the claim arose later.

On that footing, Drax contends that there is really no business sense to Wipro’s interpretation because a single cap would be inapposite where there could be (looking at the date when the MSA was made) so many potential claims over an indefinite period of time by more than one Drax entity.

In theory, all of that is true, but one has to be realistic. If, as Drax alleges here, the project was proving or threatening to be a disaster within the first year, it was hardly likely to commission yet further work and indeed at some point, it would surely terminate. That, of course, is exactly what Drax did in 2019. It should also be pointed out that clause 29.3 provides a right of partial termination and clause 31 provides for step-in rights. So I do not think that these arguments raised by Drax show that Wipro’s interpretation makes no business sense, or was commercially absurd or anything like that.

So far as the figures are concerned, Drax then points out that even on the basis that there are only 4 relevant claims, this means that on Wipro’s interpretation they are still capped at one third of their potential value i.e. £11.5 million as opposed to £31 million. As a matter of arithmetic that is obviously true but it can hardly be said that £11.5 million is insignificant.

It may be that Drax did not, in the end, protect itself in terms of claims to be made as much as it could or should have done. But that is not a reason for preferring its interpretation. And it is quite different from saying that the Clause makes no commercial sense.

I should add that for its part, Wipro makes the point that on Drax’s interpretation, and on the basis of the claims actually made, there could in theory be an overall cap of £132 million. This is based on the application of an £11.5 million limit for each of the claims set out in paragraph 5 (a)-(l) of the Further Information. And that would be absurdly high given the actual charges payable. I see that, but of course, I have rejected Drax’s primary case on the meaning of “claim”.

However, Wipro contends that even on Drax’s secondary case, the overall cap for the claims other than termination would be £34.5 million being £11.5 million x 3. I am not persuaded that this would fall into the category of being absurd or making no business sense. Nor do I think that it can be said that this interpretation would mean that the cap would be so high as to be devoid of any real purpose (pace the judgment of O’Farrell J in Royal Devon at paragraph 86). But it does not matter. That is because there is no “business common-sense” point which can assist Drax.

Wipro also relies on Royal Devon where it makes the point that an effective reduction in the cap for claims arising after Year 1, makes sense because of the greatly reduced value of the work to be done later on. In reality this is common ground. But it does not really affect the debate before me which is whether there is a single total cap, or one for each claim.

Overall, I do not consider that there are any commercial considerations which weigh against the interpretive outcome suggested by the language.

I accept that one should have regard to the notion that objectively, parties do not easily give away rights which they would otherwise have under the general law. But as Lord Leggatt pointed out, the force of that consideration must vary from case to case. Unlike Triple Point, for example, there was not, in my view, the same kind of disparity between a claim put at £14 million, and a cap of £1 million. The fact is that for the majority of the claims here, there is a cap at £11.5 million. The points made in paragraphs 88 and 89 apply equally, here.

Overall, I do not consider that the Triple Point notion carries much weight here. It is certainly not such as to affect my overall view as to which interpretation is the correct one in this case.

Accordingly, and for all the reasons given above, the answer to the question in Preliminary Issue 1 is: Clause 33.2 of the MSA provides for a single aggregate cap which applies to the Defendant’s liability for the Claimant’s claim.

In one sense, the answer to this Preliminary Issue does not matter, since there is a single cap for all of the claims. However, I need still to deal with it because I am formally required to do so, my conclusions on it have some relevance to how I determined Preliminary Issue 1, and in any event this matter has been fully argued.

As already noted, Drax’s primary position is that “claim” here means “cause of action”. I quite accept that a claim must certainly include at least one cause of action. But that does not answer the question here which is whether it is precisely coterminous with “cause of action”.

I also accept that the classic definition of a cause of action is that given by Diplock LJ (as he then was) in Letang v Cooper [1965] 1 QB 232 at 242, being a factual situation, the existence of which entitles one party to obtain a remedy against another.

Most of the case law which discusses claims and causes of action is concerned with questions of limitation and in particular what is now section 35 of the Limitation Act 1980 (“section 35”). The point there, of course, was that if a claim was sought to be introduced by way of amendment but involved a new cause of action, as opposed to for example, expanding or amplifying an existing cause of action already pleaded, the court then had to consider if it would now have been time-barred. And if so, whether it arose from the same or substantially the same facts and matters in issue on any claim already made.

To that extent, the word “claim” was being used in a highly specific context although even here, distinguished from a cause of action. Thus, in Lloyds v Rogers (1999) EG 187, Auld LJ stated that:

“It is important to note that what makes a “a new claim” as defined in Section 35(2) is not the newness of the claim according to the type or quantum of remedy sought, but the newness of the cause of action which it involves. The formula employed in Section 35(2)(a) and (5) is “a claim involving … the addition or substitution of a new cause of action”. And Order 20, Rule 5(5) refers not to a claim but to“[a]n amendment the effect of which is to add or substitute “a new cause of action”. Diplock LJ's widely accepted definition of a cause of action in Letang v. Cooper [1965] 1 QB 232 , CA, at 242–3, as “simply a factual situation the existence of which entitles one party to obtain from the court a remedy against another person”, as distinct from “a form of action … used as a convenient and succinct description of a particular category of factual situation”, is of importance. It makes plain that a claim and a cause of action are not the same thing. It follows, as Mr Croally argued, that an originally pleaded “factual situation” may disclose more than one cause of action, although one of them may not be individually categorized as such or the subject of a claim for a separate remedy. However, as Mr Browne-Wilkinson submitted, it does not follow that a claim so categorizing it and/or seeking a remedy for it made for the first time by amendment is the addition of a new cause of action so as to render it a new claim.”

But even where the expressions “claim” and “cause of action” are used interchangeably, this tells us nothing about their equivalence or otherwise in the entirely different context of limitation clauses such as that before me. Or, to take a different example, the proper interpretation of the word “claim” in the context of insurance law.

So I do not accept that there is any decided case which establishes that “claim” is to be regarded as equal to “cause of action” for the purposes of a limitation clause like that before me, or at least, I have not been shown any such case.

Indeed, the consequences would be very odd if there were such equivalents. On Drax’s primary case, and as already noted, there would be a total cap of £132 million for the first 12 claims and then a further cap for the remainder. In fact, and on a strict application of the notion that “claim” means “cause of action” that is probably an understatement. For example, on the Misrepresentation Claim, there are 13 separate statements relied upon and they are said to give rise to 3 separate representations, with 6 different pleaded falsities. I have little doubt that, certainly for the purposes of amendment and section 35, a court would be likely to hold that there was clearly more than one cause of action involved and in my view, at least 3. That would give rise to a cap for the Misrepresentation Claims alone of £34.5 million.

Moreover, given the multitude of facts pleaded in respect of the Quality and Delay breaches, I think it likely that they would be more causes of action than the 10 and 4 claims respectively identified by Drax.

But on that footing, and if Drax was correct on its interpretation of the Clause, this really would be a consequence which would make the Clause, if not devoid of all utility, then devoid of much of it.

For all those reasons, I reject Drax’s primary contention on the meaning of “claim”.

However, that does not mean that the only alternative interpretation is that proffered by Wipro which is that here, “claim” simply means “liability”. On that footing there could never be more than one operative claim. True it is that this approach avoids the need to engage with the task of equating claims to causes of action. But that is not sufficient to say that it is correct. In my view, it is not. “Claim” means something different from “liability” and I see no reason, whether from the language of the Clause or otherwise, to treat them as the same. Further, this artificial approach to the interpretation of the word “claim” is not even necessary for Wipro to be correct on its interpretation of the Clause as setting one overall cap. All that is required is that “claim” in this context means the first of all of the claims made. See paragraph 58 above.

In truth, I think there is a middle ground. This involves construing a claim in the context of and for the purposes of the operation of the Clause. Here, I see nothing wrong with Drax’s alternative description of the claims as being the 4 addressed above and indeed this description appears appropriate to me. The 4 claims do not (save in the case of the Misrepresentation Claim) overlap in terms of loss and they represent how Drax has pleaded out its claim. Further, and in a broad (not section 35) sense, they correspond to different causes of action or groups thereof, because they are plainly different, and rely on different sets of facts, and they correspond to a common-sense view of what claims are being made in this case.

Wipro suggests that the selection of these 4 claims is arbitrary. But if one rejects the alignment of “claim” with “cause of action” and “liability”, then some other meaning must be given, on the hypothesis that Drax was correct and there was a separate cap for each claim. This exercise of identification cannot simply be avoided.

Of course, it might be possible to select some other group of claims, but none has been suggested. Moreover, if there was a separate cap for each of the 4 claims, it does not seem an odd outcome that there would be a cap of £11.5 million for each of the Misrepresentation, Quality and Delay claims and then £3.8 million for the Termination Claims.

Accordingly, and for the reasons given, the answer to Preliminary Issue 2 is: there are not multiple caps provided by Clause 33.2 but if there had been, the Claimant’s claims to which those caps applied would be

The Preliminary Issues are therefore to be answered as set out in paragraphs 96 and 112 above. I am grateful to counsel for their most helpful oral and written submissions.