Kyndryl UK Limited v Jaguar Land Rover Limited

This is an application by the Defendant for summary judgment and/or to strike out the claims against it. The application is made by application notice dated 20 February 2024 on the five grounds set out in the schedule to the notice. There is also a cross-application by the Claimant by notice dated 12 June 2024 to amend its Particulars of Claim.

The applications were argued over two days, with Alex Charlton KC representing the Defendant and Benjamin Pilling KC, leading Gideon Shirazi, representing the Claimant. I am grateful for their clear and helpful written and oral submissions. The Defendant’s application was supported by statements from Sian Dyer. The Claimant relied on witness statements from Martin Graves, Martin Walsh and Karen Berry.

The Claimant is a member of the Kyndryl group, which is in business providing managed IT infrastructure services. The Kyndryl group was spun out of the IBM group in November 2021, taking over IBM’s global technology services (GTS) business. In anticipation of the spin-out, that part of the GTS business that was operated by IBM United Kingdom Ltd was transferred to the Claimant by a sale and purchase agreement dated 1 September 2021 (the SPA). The Claimant’s case is that it is entitled to bring its claims in the present proceedings as assignee of IBM United Kingdom Ltd pursuant to the SPA.

The Defendant is a member of the Tata Motors group, which is in business manufacturing automobiles, including under the Jaguar and Land Rover brands. It was formerly a customer of IBM United Kingdom Ltd, which provided it with IT infrastructure services under a long-term Data Centre and Hosting Agreement (the DCHA). The DCHA terminated on 29 October 2021.

In this judgment, I shall refer to the Claimant as “Kyndryl”, the Defendant as “JLR”, and IBM United Kingdom Ltd as “IBM”.

It is not necessary for present purposes to set out the parties’ claims and defences in the full detail in which they are pleaded. I take the following summary of Kyndryl’s claims from its Particulars of Claim, noting that much of it is disputed by JLR.

By 2013, JLR’s IT infrastructure—including software, servers, network devices, storage and backup devices—had become outdated and increasingly expensive for IBM to host, manage and maintain. In February 2013, IBM presented to JLR two possible ways forward for their commercial relationship. First, JLR could continue using its existing systems, but costs would increase substantially over time. Alternatively, those systems, or parts of them, could be migrated to a virtual multi-customer environment hosted by IBM known as “Flex”. That would result in a substantial cost saving, for both IBM and JLR, by reducing the personnel and work required for maintenance of the systems.

JLR agreed to move forward in accordance with the second proposal. On 30 September 2013, IBM and JLR executed a Change Control Notice (CCN025) under the change control procedure in the DCHA, which I describe further below. CCN025 attached an amended and restated version of the DCHA, including, at schedule A, a statement of the services to be provided by IBM and, at schedule C, a statement of the charges to be paid by JLR. Kyndryl says that those charges were calculated on the assumption that the systems would be migrated to Flex as JLR had agreed.

JLR engaged Tata Consulting Services, a technology services business within the Tata group, to provide the software services necessary for the migration. However, in about mid-2015, JLR informed IBM that it did not have funding available for Tata Consulting Services to undertake or complete the work. In the event, the migration to Flex did not proceed and IBM continued to maintain the existing, outdated, systems. Anticipated savings from Flex were not achieved; on the contrary, supporting JLR’s legacy infrastructure became more burdensome to IBM as time went on.

JLR’s systems included mid-range and mainframe servers. The position in relation to the maintenance of mainframe servers was eventually resolved by a further change to the DCHA, recorded in CCN034 dated 30 September 2016. However, a separate resolution was required for midrange servers. Kyndryl says that there was a common understanding that the DCHA would require a further change to reflect the additional cost and lack of savings in relation to midrange servers. However, the parties failed to reach an agreement as to the form this change should take.

In late 2019, IBM and JLR entered into discussions aimed at finding a “re-solutioning” for the midrange servers. Those discussions, which were referred to as “Project Defender”, proceeded on the common understanding that, if such a solution could not be found, then JLR would pay additional charges reflecting the difference between the services specified in schedule A (i.e. as agreed under CCN025) and those actually being provided by IBM. The difference is referred to by Kyndryl as the “Services Delta”.

By December 2019, IBM and JLR had agreed the commercial principles for Project Defender. There was to be a variation to the DCHA, which would extend its period, replace the existing services schedule, and update the schedule of charges so that the contractual services and charges reflected what was actually being provided by IBM. Following negotiations in early 2020, on 21 April 2020, IBM and JLR agreed a revised schedule A, referred to as the “Red Schedule”, as a record of the services that IBM was in fact providing.

On 14 September 2020, IBM issued to JLR a draft Change Control Notice, CCN050. Under the heading “brief description of change”, the draft stated:

JLR declined to approve or execute CCN050. On 6 November 2020, JLR gave notice to terminate the DCHA for convenience, and the DCHA therefore came to an end in accordance with its terms on 29 October 2021. There is no issue about the validity of that termination.

On those facts, as elaborated in the Particulars of Claim, Kyndryl advances three principal claims.

The value of the Legacy Environment Claim is approximately £19.5 million, reflecting the cost or value of the Services Delta.

In the summer of 2017, JLR requested IBM to provide a new data storage solution. IBM was to build a new hardware environment and IBM and JLR were to cooperate to migrate the existing data into that environment. IBM and JLR agreed three connected contractual Statements of Work dated 18 August 2017 covering the work to be carried out. IBM purchased the relevant hardware and software, but JLR failed to comply with its own obligations including the requirement under Statement of Work 894 to complete the remediation of applications on their servers. As a result, IBM was required at JLR’s request to maintain and manage both the old and new environments. That resulted in additional work and costs in a total amount of approximately £2.9 million between February 2019 and October 2021.

On those facts, Kyndryl claims: (a) that it is entitled under the terms of the DCHA to a variation of the DCHA reflecting those additional costs; (b) alternatively, damages for breach of the terms of Statement of Work 894; (c) alternatively, restitution on the ground that JLR was unjustly enriched by the additional work at the expense of IBM.

Under CPR 3.4(2), the court may strike out the Particulars of Claim if it discloses no reasonable grounds for bringing the claim. To strike out a claim, “the court must be certain that the claim is bound to fail”: Hughes v Colin Richards & Co [2004] EWCA Civ 266 at [22], emphasis in original. Moreover, if the court finds that there is a defect in a pleading, it is normal for the court to refrain from striking out that pleading (or refusing an amendment) unless the party concerned has been given an opportunity of putting right the defect, provided that there is reason to believe that it will be in a position to do that: Kim v Park [2011] EWHC 1781 (QB) at [40]-[41]; ACS v Efacec [2021] EWHC 915 (TCC) at [56].

Under CPR 24, the court may give summary judgment against a claimant on the whole of a claim or on an issue if it considers that the claimant has no real prospect of succeeding on the claim or issue and there is no other compelling reason why the case or issue should be disposed of at a trial. The principles applicable to CPR Part 24 are well known and were summarised by Lewison J in Easyair Ltd v Opal Telecom Ltd [2009] EWHC 339 (Ch) at [15]:

A particular feature of the present application is that some of the grounds relate only to specific pleaded issues rather than the claim as a whole. The court’s power to give summary judgment under CPR 24.2 on an issue was considered in Anan Kasei Co v Neo Chemicals and Oxides [2021] EWHC 1035 (Ch). Bacon J said, at [82]

JLR’s first ground for its application is that there was no valid assignment by IBM of all or some of the claims now made by Kyndryl.

Kyndryl says that it has a right to claim as assignee under clause 2.1 of the SPA, which provides:

Paragraph 2 of schedule 1 to the SPA defines Transferred Assets as including any contract listed in schedule 1-2 to the SPA, and paragraph 6 extends the definition to include:

Schedule 1-2 comprises a table in four columns. The copy of the schedule adduced by Kyndryl in evidence is almost entirely redacted. The single unredacted line reads as follows:

Kyndryl says that that line in the schedule is referring to the DCHA. It relies on a witness statement of Martin Graves, its general counsel and company secretary. Mr Graves was the legal lead for the GTS business unit within IBM and had responsibility for the JLR relationship. He says in his statement that the spin-out of Kyndryl was a memorable episode for him. He specifically confirms that the line in schedule 1-2 refers to the DCHA. He says that JLR was an indirect subsidiary of Tata Motors Ltd at the date of the SPA and, having made enquiries, he is not aware of any other contract associated with Tata Motors to which the schedule might be referring. He says that there are other documents showing that the contract ID code in the schedule relates to JLR and the DCHA, and, in support of that point, exhibits two invoices issued by IBM under the DCHA that bear the same reference, SOM015, appearing in the schedule.

Kyndryl says that its claims under the DCHA were therefore Transferred Assets assigned to it by clause 2.1. As for its claims under the Variation Agreement and in unjust enrichment, it says that they were assigned by paragraph 6 of schedule 1 of the SPA because they are rights “exclusively arising from the ownership” of the DCHA. Notice of assignment was given by Kyndryl to JLR on 13 July 2023.

During his oral submissions in response to the application, Mr Pilling sought to raise a further, new argument. He said that Kyndryl wished also to rely on paragraph 1 of schedule 1 of the SPA, which brings within the definition of Transferred Assets all “Assets… of [IBM] that relate exclusively to the Business”, where Business is defined as “the managed IT infrastructure services business” and Assets are defined broadly as including “all assets, properties, and rights of every kind” including “Contracts and rights arising thereunder”. He produced a draft amendment to that effect. Mr Charlton, who had no prior notice of the point, said that he was not ready to address it in his reply oral submissions, and I did not think it fair to require him to do so. I have therefore not taken it into account in this judgment.

JLR argues that there was no valid assignment of the claims for essentially three reasons. First, schedule 1-2 does not refer to the DCHA. Second, schedule 1-2 does not refer to the claim under the Variation Agreement, which is a separate contract from the DCHA, or to the claim in unjust enrichment, neither of which can therefore be Transferred Assets. Third, any assignment would be void because, under clause 1.11.7 of the DCHA, JLR’s prior written consent was required but was not given.

JLR says that the relevant line in schedule 1-2 does not mention JLR or the DCHA specifically, nor any contract date, and that the contract ID codes have not been properly explained. It says that Kyndryl pleads in its Reply that there was a common understanding between Kyndryl and IBM that the reference to a contract with Tata Motors Ltd included the DCHA, but that Kyndryl has adduced no evidence of such an understanding in response to the application.

However, in the light of Mr Graves’ evidence, as set out above, I am satisfied that there is, at least, a real prospect of Kyndryl proving at trial that the schedule is referring to the DCHA. Two facts strongly support Kyndryl’s case: (a) the contract ID code matches that used for the DCHA; and (b) Mr Graves could find no other contract that was associated with Tata Motors. I do not think that it is necessary for present purposes for Kyndryl to adduce specific evidence of a common understanding beyond the material in Mr Graves’ statement.

JLR argues that the (unamended) Particulars of Claim are defective in that, while pleading the assignment of the DCHA, they do not explain the basis on which Kyndryl is also entitled to claim under the separate Variation Agreement as assignee of IBM or to bring the unjust enrichment claims. Kyndryl seeks to address that defect by a proposed amendment to paragraph 17 of the Particulars of Claim. The amendment clarifies that the assignment extends to the Variation Agreement and unjust enrichments claims because of the effect of paragraph 6 of schedule 1. JLR does not object to the amendment as such, but maintains that, even as amended, the assignment case has no real prospect of success.

On the assumption that the DCHA itself was assigned by the SPA, as discussed in the previous section, whether the Variation Agreement and unjust enrichment claims were also assigned depends on whether they are “claims or rights” or “rights of recovery” “whether in tort, contract or otherwise” “exclusively arising from the ownership of” the DCHA, using the relevant words of paragraph 6. The interpretation of paragraph 6 is a matter of Delaware law, the law governing the SPA, but neither party argues that there is any relevant difference from English law.

IBM can reasonably be regarded as having “ownership” of the DCHA at the date of the SPA in the sense that it held the contractual rights under it. JLR argues, however, that claims or rights under the Variation Agreement cannot be regarded as “exclusively arising from” that ownership of the DCHA for the purpose of paragraph 6. That is because the Variation Agreement is a separate contract from the DCHA; indeed, Kyndryl itself pleads in its Reply that they are separate in seeking to show that the contractual limitation period in the DCHA does not apply to the Variation Agreement. JLR argues that “exclusively” means that the claims or rights “cannot arise in any other way other than by way of ownership” and that a claim for breach of contract separate from the DCHA does not meet that test. As for the unjust enrichment claims, JLR argues that they arise from requests or instructions by JLR for work outside the DCHA and so do not arise exclusively from the DCHA.

The Variation Agreement is, as pleaded, an agreement between IBM and JLR to vary the DCHA. Claims under the Variation Agreement are therefore, in my view, claims “arising from” IBM’s ownership of the DCHA, bearing in mind the broad and inclusive meaning usually given to that expression; see Fiona Trust v Privalov [2007] UKHL 40, [2007] 4 All ER 951 at [11]-[13]. The qualification that they must arise “exclusively” from that ownership is a little more difficult to interpret. However, it is hard to see how such claims could be said to arise out of the ownership of any asset other than the DCHA and I therefore consider that there is at least a real prospect that Kyndryl is right to say that, for the purpose of paragraph 6, the claims arise exclusively from IBM’s ownership of the DCHA. I take a similar view about the claims in unjust enrichment. They are not claims under the DCHA. However, paragraph 6 is not limited to contractual claims but extends to “rights of recovery” of all kinds, and so is capable of embracing a restitutionary claim. A claim for the value of services provided outside but in connection with the DCHA can fairly be described as a right of recovery exclusively arising from the DCHA.

Moreover, I agree with Kyndryl that the commercial context for the SPA strongly suggests a broader rather than narrower interpretation of the scope of the assignment. IBM was transferring to Kyndryl its entire GTS business, including (according to Mr Graves’ evidence) about 16,000 contracts and associated personnel. It is unlikely that the parties would have intended for IBM to assign to Kyndryl rights under an identified and significant contract but for IBM to retain associated rights, such as rights under a collateral contract or rights of variation or unjust enrichment claims. That would have involved splitting up the GTS business by leaving behind potentially valuable assets.

I also accept Kyndryl’s argument that there are reasonable grounds for supposing that documentation produced in connection with the SPA and the spin-out generally might be relevant to the commercial context or factual matrix and so to the interpretation of paragraph 6 of schedule 1. Mr Graves’ evidence was that there was a “universe” of such documentation. Mindful of the need to take into account all the evidence that might reasonably be expected to be available at trial, that argument provides a further reason against summary disposal of this issue.

JLR next argues that any purported assignment under the SPA was void because JLR did not give its prior written consent. It relies on clause 1.11.7 of the DCHA, which provides:

The effect of this clause is to prohibit and make void any purported assignment of rights under the DCHA without consent unless the assignment is either (a) a “transfer of [the DCHA]… within the legal entity of which either party is a part” or (b) a “transfer… to a successor organisation by merger or acquisition”.

Kyndryl argues that both parts of that exception apply. It says that “the legal entity of which either party is a part” means, or includes, the corporate group of which a party is a member, and that the DCHA was transferred to Kyndryl when Kyndryl was still within the IBM group. (It is common ground that at the relevant time, i.e. at the date of the SPA, both IBM and Kyndryl were subsidiaries of the same ultimate parent, IBM Corp.) It also says that it was a “successor organisation by… acquisition” because it acquired the GTS business from IBM.

JLR’s starting point is that only transfers are permitted by clause 1.11.7 without consent, and there was no transfer of the DCHA. It says that an assignment is something distinct from a transfer and that, in referring to a transfer rather than an assignment, the parties intended that the exception should apply only to a transfer of the whole of the DCHA, i.e. the burden as well as the benefit, and not to an assignment of rights only.

I do not agree with that argument. The words “assign, or otherwise transfer” (my emphasis) clearly imply that an assignment is a kind of transfer, not something distinct. In ordinary usage, transfer is a more general term; assignment is more specific. Kyndryl suggested in oral argument some examples of possible transfers otherwise than by assignment, such as by declaration of trust, subrogation, or novation, which illustrate the point.

JLR next argues that a permitted transfer must be “within the legal entity of which [IBM] was a part”, which means an internal transfer within IBM (i.e. within the IBM United Kingdom Ltd company), such as a change in management or administrative responsibility for the contract from one internal department or division of IBM to another. In support of that argument, it relies on the following:

In considering this question of interpretation, I have followed the well-established approach in Rainy Sky v Kookmin Bank [2011] UKSC 50; 1 WLR 2900; Arnold v Britton [2015] UKSC 36, [2015] AC 1619 and Wood v Capita [2017] UKSC 24, [2017] AC 1173. I remind myself that interpretation is a unitary exercise involving an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated. In striking a balance between the indications given by the language and the implications of the competing constructions the court must consider the quality of drafting of the clause; and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest. Similarly, the court must not lose sight of the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms.

The quality of drafting of clause 1.11.7 is poor. As a result, both parties’ interpretations involve a significant departure, in different ways, from the express wording. On balance, however, I prefer Kyndryl’s interpretation.

I do not attach much weight to JLR’s argument about the risk that the DCHA might be assigned to a company in the IBM group that could not perform the obligations. As a matter of general law, IBM could not unilaterally assign or transfer its liabilities or obligations under the DCHA, and clause 1.11.7 does not purport to permit that. It is concerned with assignments of rights.

JLR’s other arguments have greater force, particularly the argument that the expression “legal entity” generally means a legal person. I accept that; and if the wording of clause 1.11.7 had gone no further, then I would have been reluctant to stretch the interpretation of “legal entity” to cover a group of companies or some form of organisation that lacked unitary legal personality.

However, the clause must be interpreted as a whole, including the important phrase “of which either party is a part”. Kyndryl’s interpretation makes much better sense of those words. I agree with Kyndryl that they imply that “legal entity” is referring to some larger organisation of which the assigning “party” (here, IBM) is a part. The larger organisation can only be, in this case, the IBM group. On JLR’s interpretation, that phrase would be redundant or meaningless. In argument, Mr Charlton was not able to ascribe any real meaning to it, submitting instead that the language was clumsy and inelegant and ought to yield to the commercial context. In the circumstances, although I accept that it is not the usual meaning of “legal entity”, it is reasonable to interpret it as used in clause 1.11.7 as meaning “legal organisation” and so as referring to the IBM group.

It is also significant that the second sentence of clause 1.11.7 provides that any attempt to transfer the DCHA is void if not within the exception. That strongly suggests that the clause is concerned with transfers that would otherwise have legal effect. However, a change in the administration of a contract from one division of a company to another, not involving a change in parties, has no legal effect. Such an internal change is therefore generally no concern of the counterparty, and one would not expect the parties to make express provision for it in their contract, either by prohibiting it or permitting it. Indeed, I would not regard such an internal change as properly described as a “transfer” at all.

As to the second part of the exception, JLR argues that since Kyndryl neither merged with nor acquired IBM it cannot be regarded as a “successor organisation”. To be a successor, there needs to be “something that occurs at the corporate share level”. It would be uncommercial to interpret the exception in a way that would permit IBM to assign the DCHA to its own subsidiaries or to any company within the IBM group. Kyndryl responds that it is sufficient for it to be a successor organisation that it acquired the relevant part of the business of IBM; it was not necessary that it should acquire the shares in IBM.

Again, I prefer Kyndryl’s interpretation. In my view, the expression “a successor organisation by… acquisition” is capable of describing an acquisition of the business and assets of a predecessor organisation as well as an acquisition of its shares. There might be room for debate about whether any particular transaction amounts to an acquisition in that sense, but in the present case I am confident that the sale of the entirety of IBM’s GTS business to Kyndryl constituted Kyndryl a successor to IBM in respect of that business for the purpose of the clause. JLR’s attempt to distinguish a share sale from an asset or business sale seems uncommercial in that the right to assign under the DCHA should not depend on precisely how an acquisition is structured.

Finally, Kyndryl argues that the prohibition on assignment in the first sentence of clause 1.11.7 is in any case limited to the DCHA and rights under it and does not prohibit an assignment of the Variation Agreement or unjust enrichment claim (although an assignment of the Variation Agreement divorced from the DCHA might be of no practical value to Kyndryl).

JLR’s response is that the wording of clause 1.11.7 is sufficiently broad to include all the claims made by Kyndryl, not just the claims for breach of the DCHA. I note the argument that it might seem strange if IBM were prohibited from assigning rights under the DCHA but permitted to assign connected rights, such as claims in unjust enrichment. There is some equivalence with the point about the scope of paragraph 6 of schedule 1 of the SPA, discussed in paragraph 36 above. However, clause 1.11.7 is drafted narrowly. It is in terms restricted to “[the DCHA] or rights under [the DCHA]”; broader “arising from” or similar language is absent. On that drafting, I see no reason to extend the scope of the prohibition to a restitutionary claim not made under the DCHA.

For those reasons, I decline to give summary judgment, or to strike out the Particulars of Claim, on ground 1 of the application.

JLR urged me to “grasp the nettle” by making a final decision on some of the issues of interpretation discussed above (although presumably in the hope that I would take a more favourable view of its arguments than I have). Kyndryl was more cautious, suggesting that I should simply dispose of the application by reference to the summary judgment test.

The issues on the validity of any assignment depend entirely on the interpretation and application of clause 1.11.7. They were argued in detail, both in writing and orally, and I consider that a future trial judge will be in no better position than I am to resolve them. Given that I prefer Kyndryl’s case on clause 1.11.7, the only reason to defer a final decision would be to allow JLR to adduce any further relevant evidence or to develop its arguments in the light of the facts as may be found at trial. However, as I have said, JLR urged me to make a final decision now; it did not suggest that there might be any disputed factual or contextual issues that might result in a different outcome at trial. I will therefore grasp the nettle to the extent of deciding that, if there was an assignment of any of Kyndryl’s claims by the SPA, then that assignment was not avoided by clause 1.11.7 through lack of consent by JLR. To be clear, however, I grasp it no further, and the prior question of whether there was an assignment will be a matter for further consideration at trial in the light of the evidence at that time.

JLR argues that Kyndryl acquired no right to vary the DCHA because:

Under this ground, JLR also argues that Kyndryl’s secondary case, that IBM was entitled to a variation pursuant to an express or implied term of the DCHA, fails because there was no such term.

It is convenient to address first the debate about uncertainty.

The terms of the Variation Agreement, and the breach of those terms, are pleaded as follows:

The starting point for the law on uncertainty is Viscount Dunedin’s statement in May & Butcher Ltd v R [1934] 2 KB 17 at p21 that “to be a good contract there must be a concluded bargain, and a concluded contract is one which settles everything that is necessary to be settled and leaves nothing to be settled by agreement”.

In RTS Flexible Systems v Molkerei Alios Müller [2010] UKSC 14, [2010] 1 WLR 753 at [49], the Supreme Court set out the principles, of which the following are particularly relevant to the present case:

I was also referred to Mamidoil-Jetoil Greek Petroleum Company SA v Okta Crude Oil Refinery AD [2001] EWCA Civ 406, where Rix LJ said at [69]:

JLR argues that the Variation Agreement was uncertain or incomplete in three respects: first, no agreement is pleaded or had been made as to the date on which CCN050 was to come into effect; second, the draft schedule A (services) and schedule C (charges) had not been agreed; third, those schedules differed from the documents produced as part of Project Defender.

Kyndryl explained in its written and oral arguments that its case is that the agreement reached was that CCN050 would “cover all additional services since JLR’s failure to take the steps necessary to migrate to Flex”, i.e. since January 2017. It is fair for JLR to complain that such a term is not pleaded, or not with any clarity, but that complaint could properly and easily be addressed through a request for further information or a short further amendment. It would not be right to dismiss the claim in its entirety until Kyndryl has had an opportunity of responding to such a request.

JLR also disputes that there was in fact any such agreement about the effective date. It points to internal inconsistencies in the documentation about the date: draft CCN050 refers (in exhibit C-1 to schedule C) to charges being applicable from 1 April 2020 and the Project Defender materials refer to charges starting in July 2020. However, that is a dispute that will turn on the evidence at trial about what was in fact agreed; it is plainly unsuitable for summary determination. I note that Kyndryl’s case on the amount payable by JLR appear to assume that the increase in charges contemplated by CCN050 was to take effect retrospectively. However, there is no reason why the parties could not or would not have agreed that.

JLR argues that the charges to be paid under the Variation Agreement were never agreed and that the pleaded entitlement to a “difference in price reflecting the Services Delta” is insufficiently certain to found a contract.

Kyndryl’s position is that the charges could be calculated “mechanically” from the parameters already agreed under the DCHA. It relies on evidence from Ms Karen Berry, who at the relevant time was global account director for IBM with overall responsibility for the agreements and services the subject of the dispute. She explains what had been discussed and agreed between IBM and JLR about pricing. She says that “the price had not been fully calculated in the sense that a precise pounds and pence figure had not been quantified prior to the issue of CCN050”. However, “the process of calculating the price was clear… [it] would involve calculating the additional cost associated with the additional work which IBM had been, and would be, providing through the Services Delta”. She says the calculation of the final price was “simply a mechanical exercise” and not a “negotiation where JLR could choose the price that it would pay”.

In oral argument, Mr Charlton took me at some length through some of the contemporaneous documents referred to by Ms Berry in her statement in an attempt to demonstrate that her account should not be accepted. Mr Pilling criticised that exercise as akin to a cross-examination of Ms Berry in her absence. While I am not bound to accept the written evidence at face value, I agree with Mr Pilling that points of detail of the kind taken by JLR are indeed only suitable for determination after cross-examination rather than on a summary basis. Nothing I was shown persuaded me that there was no real prospect of the trial judge accepting Ms Berry’s evidence. I would also expect that at trial there will be further and more detailed evidence from the parties about their discussions.

In any case, as the Court of Appeal held in Mamidoil, it is not fatal to the enforcement of a contract that the parties may not have reached a complete and express agreement on the price or other terms. In an appropriate case, the court can imply a term, such as a term to pay a reasonable price. Whether this is such a case, and whether there was sufficient agreement between the parties on the essential aspects of the price to make the Variation Agreement workable, are matters for trial.

For completeness, I note that Kyndryl also argues that, if any terms of CCN050 remained uncertain or not agreed, then they could be ascertained through the dispute resolution procedure in schedule G of the DCHA. Paragraph 5.1(f) of the DCHA provides:

JLR argues that the Red Schedule, which recorded the services being provided and to be provided by IBM, was not agreed. It relies on internal inconsistencies in the Particulars of Claim about the date on which the Red Schedule was finalised. It also relies on passages in the statement of Ms Berry where she says that the Red Schedule was only partially agreed by 21 April 2020 and (only) 95% complete by 16 June 2020. Again, I do not see that arguments of that kind are anywhere near sufficient to justify dismissing the contractual claims on grounds of uncertainty. It is not necessary for the parties to have reached express agreement on every detail. Kyndryl’s position is that there was at least substantial agreement on the services to be provided and that any outstanding points were non-substantive. That position is supported by Ms Berry’s evidence, including her account of the “near-daily in-person and telephone meetings” to discuss the Red Schedule. Whether at the end of those meetings there remained essential matters to be agreed sufficient to make the Variation Agreement unworkable is, again, a matter for trial.

JLR argues that schedules A and C attached to the draft CCN050 differ materially from the Red Schedule and the other documents prepared as part of Project Defender. The differences between the Red Schedule and schedule A were identified in a comparison document that accompanied the draft CCN050. Examples of differences in the charges were given in JLR’s witness evidence on the application.

Kyndryl disputes that there were any material differences. It says that they are either minimal or reflect the different context in which the Red Schedule and schedule A were used. It says that the agreement was that CCN050 would reflect the Red Schedule rather than reproduce it precisely.

Any determination of materiality would require the court to consider the differences in the context of the project as a whole, an exercise not suitable for a summary procedure. Moreover, as Kyndryl argues, even if there were material differences, that would not necessarily lead to the conclusion that the parties were not in agreement; an alternative possible conclusion would be that the schedules had been incorrectly drafted and should have been adjusted to reflect the true agreement.

If the Variation Agreement would otherwise be enforceable, JLR argues that it is nevertheless invalid because it was not made in accordance with the contractual change control procedure. The relevant provisions of the DCHA addressing change control are as follows:

JLR relies on Rock Advertising Ltd v MWB Business Exchange Centres Ltd [2018] UKSC 24, [2019] AC 119 as authority that the court will uphold contractual provisions denying validity to variations or changes that are not agreed in accordance with a contractually mandated process. Rock concerned a licence agreement containing a “no oral modification” clause, that is, a contractual term that the licence agreement could not be amended other than in writing signed on behalf of the parties. The issue before the court was whether that clause rendered contractually ineffective a purported oral agreement to vary the licence. Lord Sumption JSC (with whom the majority of the court agreed) said, at [10], that “the law should and does give effect to a contractual provision requiring specified formalities to be observed for a variation”. He continued, at [15]:

The Variation Agreement was not the subject of any CCN. Therefore, JLR argues, it was contractually ineffective. Indeed, IBM acknowledged that a CCN was a precondition to any variation by preparing and sending CCN050 for acceptance by JLR.

Kyndryl argues that there is a material distinction between the present case and the situation in Rock. By entering into the Variation Agreement, the parties were not agreeing to dispense with the formal change procedure but, rather, were agreeing that it should be operated. JLR’s obligation was to execute a CCN in a form that reflected the changes that had been informally agreed in the Variation Agreement. The draft CCN50 prepared by IBM did reflect those changes and so JLR was contractually bound to accept it.

I was not shown any authorities in which there had been consideration of that distinction, i.e. between, on the one hand, an informal agreement for variation without compliance with the prescribed formalities and, on the other, an informal agreement to take the prescribed formal steps. It might be forcefully argued that, at least in some cases, the distinction is too fine to make any difference and, perhaps, would have the effect of allowing a party to circumvent the decision and reasoning in Rock and negate the purpose of a “no oral modification” or similar clause. That said, absent authority, I do not see why as a matter of principle the court could not in an appropriate case enforce an informal agreement to take a formal step. In Rock, Lord Sumption said that the natural inference from the parties’ failure to observe the formalities was that they had overlooked them rather than that they had agreed to dispense with them (or, if they had the formalities in mind, that they were “courting invalidity”). However, the position in the present case may be different. On Kyndryl’s case, the parties neither overlooked the need for a CCN nor agreed to dispense with one; instead, they recognised the need for a CCN and agreed to prepare and execute one. I cannot say that an argument along those lines is certain to fail or that the rule in Rock is necessarily an answer to it. Moreover, the precise ambit and application of that rule seems to me to raise precisely the kind of question in a developing area of law that ought to be answered by reference to the complete facts at trial, particularly as the rival arguments may depend on the details of what was agreed between the parties.

Kyndryl has another potential answer to the preclusionary rule in Rock based on estoppel. I return to that below when discussing ground 3 of the application.

For those reasons, I consider that Kyndryl has at least a real prospect of proving that there was an enforceable Variation Agreement as pleaded.

As an alternative to its claim under the Variation Agreement, Kyndryl contends that IBM had a right to a variation under the DCHA itself. It pleads that case as follows:

It is unnecessary for present purposes to dwell on the first proposed term, because no claim is made for breach of it. As for the second term, Kyndryl says in its skeleton that it is express “not in the sense that the pleaded words appear in the contract, but in the sense that on the true construction of the contract it is the legal effect of [other] express terms”. I understand that to mean that the pleaded term is said to be derived by a process of interpretation from existing express terms.

Kyndryl relied on the following terms of the DCHA as giving rise to the proposed term:

However, none of those terms says anything at all about the contractual consequences of JLR failing to carry out the (alleged) agreed transformation or migration, or of JLR requesting IBM to provide additional services in relation to the legacy estate. Certainly there is nothing in any of them to suggest that either of those events would trigger a mandatory variation of the DCHA.

As Sir Thomas Bingham MR said in Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472, 481, “the courts’ usual role in contractual interpretation is, by resolving ambiguities or reconciling apparent inconsistencies, to attribute the true meaning to the language in which the parties themselves have expressed their contract”. I find it impossible to see how that kind of process applied to the existing terms of the DCHA could produce an express term in the form pleaded.

More generally, and contrary to Kyndryl’s position, I cannot see that the DCHA provides for any circumstances (or any circumstances relevant to the present case) in which JLR would be obliged to vary the DCHA by approving a CCN submitted by IBM through the change control process. On the contrary, paragraph 5.1.c.xi of schedule G to the DCHA expressly gives JLR the right to reject a CCN if it wishes. Although that paragraph is expressed in mandatory terms (“[JLR] shall… either reject… require… request… or accept the CCN”), that means, in my view, that JLR is obliged to take one of the specified actions but has a free choice as to which one. The furthest the DCHA goes in imposing any relevant obligation is, in schedule C paragraph 3 above, to “agree to discuss in good faith” any changes to charges via the Change Control Procedure.

It is not surprising that the DCHA does not provide for a mandatory variation if JLR requests additional services. IBM was not obliged to provide additional services beyond those specified in the DCHA. If there was such a request then IBM had a choice: it could have refused to provide the services or it could have proposed a variation through the Change Control Procedure, including any variation to charges. JLR could then have either accepted or rejected the change. If IBM chose to provide additional services without a formal change and without any other agreement, then its right to be paid for those services, if any, would be expected to lie in restitution rather than contract.

Kyndryl relies on the change control dispute resolution procedure in support of its argument that JLR can in principle be obliged to accept a CCN. I refer to that procedure in paragraph 71 above. As I have already said, I do not think that the effect of the procedure is that the court can impose upon the parties contractual terms or changes to which they have not agreed. In particular, IBM cannot be obliged to provide additional services absent agreement, nor can JLR be obliged to pay for them.

Kyndryl argues that there is a relevant comparison to be made with construction contracts where an engineer has the power to certify variations. The cases, including Henry Boot v Alstom [2005] EWCA Civ 813, [2005] 1 WLR 3850 and Grove v S&T [2018] EWHC 123, [2018] BLR 173 show that the court can decide the true value of a certificate and if appropriate open it up. However, that is a different situation from the present where Kyndryl is seeking (in this part of the case) to impose on JLR a variation that JLR has not accepted.

As for the implication of the proposed term, the principles are well known and were set out in Marks & Spencer plc v BNP Paribas Securities Services [2015] UKSC 72; 2016 AC 742. For present purposes, I focus on the requirements that a term can only be implied if it is necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it, or so obvious as to go without saying.

JLR argues that the proposed term is unnecessary. In summary, if JLR was obliged to migrate its systems, as Kyndryl alleges, but failed to do so, then there were other contractual rights and remedies available to IBM. First, IBM could claim damages for breach of that obligation. Second, since IBM was not obliged to provide services outside those specified in CCN025, it could have refused to continue supporting legacy systems. Third, if JLR requested additional services and IBM was prepared to provide them, then any required change to the contract could be effected through the change control procedure (and JLR adds that the implication of the proposed term would be inconsistent with that procedure).

Kyndryl responds that the parties should not be thrown into a position where one has to claim damages against the other, and it would be more sensible and practical, and also necessary, to imply the proposed term. It says that the argument that IBM could have refused to provide services outside the contract until JLR agreed a variation fails to take any account of the realities of a long-term IT contract and does not reflect the way in which any commercial party in a long-term business relationship would be expect to behave, particularly where critical systems were involved. It says that the existence of the change control procedure does not mean that there cannot also be an implied term providing that a change should be agreed if certain conditions are satisfied.

I do not consider this to be an adequate answer to JLR’s argument on necessity. In a large, long-term IT contract it is not unusual for requirements to change and for the customer to request service changes. That is what the change control procedure is for. I appreciate that IBM may have been reluctant for commercial or relationship reasons to take too firm a line on what services it was obliged to provide or to press its customer for another formal change. However, as a contractual matter, IBM was entitled to refuse to provide additional services until a change had been agreed. That being so, the contract was fully efficacious without the need for the additional proposed term.

Kyndryl relies on Davy Offshore and Davy Offshore v Emerald Field (1991) 55 BLR 1, in which Judge Forbes QC held (pages 61, 62) that an engineer or employer who has a permissive power to order a variation may be required to exercise it if the contractor would otherwise be unable to perform the contract or where the contract would not otherwise work. I do not agree that that has any application to the present case. First, JLR did not have a power to order variations, rather it had the right to accept or reject CCNs proposed by IBM. Second, there is no question of the contract being unworkable or incapable of performance. As I have already said, IBM could have refused to provide additional services unless and until a formal change was agreed.

Kyndryl argues that the question of implication should not be decided now because it would be informed by the circumstances in which CCN025 came to be negotiated, and they were not fully in evidence. However, it did not identify any particular circumstances, beyond those discussed above, that might make the proposed term necessary.

JLR also argues that the term should not be implied because it is incapable of clear expression. For completeness, I should say that, if I had regarded the term as necessary or obvious, then I would not have regarded that further objection as sufficient in itself to reject the claim summarily.

I conclude that Kyndryl has no real prospect of establishing that there was an express or implied term of the DCHA as pleaded in paragraph 15.2 of the Particulars of Claim.

In the alternative to its primary case that IBM had a contractual right to a variation, Kyndryl pleads, at paragraph 46 of the Particulars of Claim, estoppels in the following terms:

“…[JLR] is estopped by convention or representation or promissory estoppel from denying that that [IBM] was entitled to a variation of the contract reflecting the cost of the Services Delta because… [JLR] represented to [IBM], expressly and/or by its conduct in the meetings and communications set out above) that if a re-solutioning was not agreed in Project Defender then it would agree a Change which compensated [IBM] for the additional work it had carried out constituting the Services Delta. There was also a convention at all material times between [IBM] and [JLR] to the same effect.”

The conventions, representations and promises are pleaded in paragraphs 29, 31 and 34 and JLR accepts for present purposes that Kyndryl will prove them. JLR argues, however, that that they are not sufficiently clear or unequivocal to found the pleaded estoppels. It argues that they “suffer from the same uncertainty as the Variation [Agreement] and the entitlements to variations: they are all species of agreement-to-agree”.

JLR also argues that the estoppels do not properly address the need for compliance with the contractual change process. In Rock, Lord Sumption said, at [16], that a person could in principle be estopped from relying on non-compliance with formalities as a reason to invalidate a variation, but “the scope of estoppel cannot be so broad as to destroy the whole advantage of certainty for which the parties stipulated… [a]t the very least, (i) there would have to be some words or conduct unequivocally representing that the variation was valid notwithstanding its informality; and (ii) something more would be required for this purpose than the informal promise itself”.

As is apparent from the way they are framed, JLR’s complaints about the uncertainty of the estoppels overlap significantly with those already discussed above about whether the Variation Agreement is unenforceable for uncertainty or as an agreement to agree. The same conduct and communications are relied on by Kyndryl both to prove the Variation Agreement and, in the alternative, to prove the estoppels. As I see it, contract-by-conduct and estoppel are presented as different legal analyses of the same facts, either one leading to the conclusion that JLR was obliged to execute a CCN. In those circumstances, I do not think that it would be right to entertain an application for strike-out or summary judgment on the estoppel case as an issue separate from the contractual case, even if there might be some difference in the applicable test. Unless JLR can dispose summarily of both the contract and estoppel cases, the matter will anyway proceed to trial. As Bacon J said in the passage quoted at paragraph 21 above, summary determination of an issue is not appropriate if it would have no consequences except that there would be one fewer issue for trial.

I have already held that Kyndryl has a real prospect of success on its primary case to enforce the Variation Agreement. That will go to trial; and Kyndryl should therefore be permitted to pursue its alternative case on estoppel on the same facts.

In reaching that decision, I take into account that the test for proving an estoppel is not quite the same as that for proving the terms of a contract. As Lord Denning MR explained in Woodhouse AC Israel Cocoa Ltd SA v Nigerian Produce Marketing Co Ltd [1971] 2 QB 23 (page 61), once a contract is made, it is the duty of the court to give effect to it if it possibly can, and it does so by resolving ambiguities, no matter how difficult that may be. “But when a man is estopped, he has not agreed to anything. Quite the reverse. He is stopped from telling the truth. He should not be stopped on an ambiguity. To work an estoppel, the representation must be clear and equivocal.” It is therefore possible, as JLR argues, that, even if the Variation Agreement is enforceable, the estoppel case could still fail for lack of certainty (although it is hard to see in that case why the estoppel case would be relevant). However, in order to draw that distinction for the purposes of a summary determination, I would need to be persuaded that there are particular, clearly identified, aspects in which the estoppel case is defective or objectionable, over and the above the complaints about the uncertainty in the Variation Agreement that I have already discussed. None was identified by JLR in its written argument or oral arguments.

Kyndryl’s unjust enrichment claim is pleaded at paragraph 50 of the Particulars of Claim:

Kyndryl seeks to introduce the underlined words by amendment in order to pursue an unjust enrichment claim based on the principle of free acceptance. That principle is set out in Goff & Jones on Unjust Enrichment, 10th edition, paragraph 17–03 as follows:

JLR consents to that amendment and does not seek summary dismissal of the unjust enrichment claim insofar as it is based on free acceptance. However, insofar as the claim is also based on requests by JLR for additional services, JLR objects that no requests have been pleaded and none was made.

In response, Kyndryl applies to amend the Legacy Environment Claim to introduce the following paragraphs:

JLR opposes those amendments. It argues that they are inadequate, and do not answer its objection, because they do not properly particularise the requests, contrary to the requirements of CPR 16.4(1)(a), 16PD7 paragraphs 7.4 and 7.5, and the overriding objective. It says that Kyndryl should at least identify when the alleged requests were made, by whom, and in what terms. There is force to this complaint. Kyndryl says that its case is that there was a consistent course of conduct giving rise to the requests, and that to plead every individual request would be burdensome and unnecessary. But in my view, although the amendments set out the general nature of the case, JLR is at least entitled to know whether Kyndryl relies on any specific requests or whether its case is confined to an implication from a course of conduct. If the former, then JLR is entitled to particulars of the requests; if the latter, it is entitled to confirmation that the case is limited in that way and to an explanation of how the implication nevertheless arises. To say only that Kyndryl will rely on the course of conduct over many years and on the entirety of interactions between the parties is not satisfactory.

I am not persuaded, however, that JLR’s criticisms of the existing pleading and the proposed amendments, are such that it would be right to refuse permission to amend and to dismiss the claim summarily now, as JLR asks. As JLR accepted in argument, this is a pure pleading point—whether there were in fact requests for additional services sufficient to give rise to an unjust enrichment claim is an evidential matter for trial. I therefore have in mind what Tugendhat J said in Kim v Park [2011] EWHC 1781 (QB) at [40]:

I am fortified in my view that that is the correct approach in the present case by the consideration that Kyndryl’s unjust enrichment claim will proceed to trial in any event on the basis of free acceptance, so the court will in any event have to examine the parties’ conduct in order to decide whether JLR was unjustly enriched by accepting services provided by IBM outside the contract.

JLR objects to the amendments on the further ground that they are inconsistent with the existing pleading. It says that the existing case (at paragraphs 1.1, 9.3, 23.1, 23.2 and appendix 3 of the Particulars of Claim) is that IBM suffered loss by failing to achieve expected savings on the migration to Flex, whereas the proposed amendments advance the case that IBM incurred additional costs by performing additional requested services. I do not think that that objection takes matters any further for present purposes. There is no reason in principle why Kyndryl could not pursue both claims together or in the alternative.

I will therefore permit the amendments to paragraphs 42A, 42B, 69A and 69B. JLR may make a request for further information, if so advised, and is at liberty to seek further appropriate relief if it considers that the request has not been properly answered or that the claims remain insufficiently particularised.

Clause 1.11.5 of the DCHA provides:

JLR’s position when the application was issued was that all Kyndryl’s claims were barred by that clause because the relevant causes of action arose, or Kyndryl (or IBM) was aware or ought reasonable to have been aware of the breach, before 20 July 2020, three years before proceedings were issued. However, at the hearing, JLR confined its application to the claim under the DCHA, that is, the claim for breach of the express or implied term pleaded in paragraph 15.2 of the Particulars of Claim. Since I have decided that that claim has no real prospect of success, the question of limitation does not arise, but I consider it for completeness.

In paragraph 45 of the Particulars of Claim, Kyndryl pleads that JLR breached the DCHA by failing “to execute CCN050 or to pay the sums due in respect of the Services Delta”. That cause of action could not arise until the draft CCN050 was presented to JLR for approval, which was after 20 July 2020. Paragraph 49 of the Particulars of Claim pleads a further unspecified breach of the DCHA “which entitled IBM to a variation”. It was initially unclear to me whether that paragraph was asserting a different and possibly earlier breach from the one in paragraph 45. In oral argument, however, Kyndryl explained that paragraph 49 was addressed to the possibility that the draft CCN050 might have incorrectly recorded the parties’ agreement, but that JLR nevertheless had an obligation to execute a corrected change notice. The breach would also occur after 20 July 2020, since, again, JLR could not be obliged to execute any change notice until IBM had presented one. It follows that time bar under clause 1.11.5 is not in itself an answer to the pleaded claims under the DCHA.

Kyndryl has no real prospect of proving that there was an express or implied term of the DCHA that, if JLR did not carry out the transformation and migration project and instead requested that IBM provide additional services to host and maintain the legacy estate, IBM would be entitled to a variation of the DCHA requiring JLR to pay IBM for that additional work. I therefore grant summary judgment to JLR on that part of ground 2 of the application and dismiss Kyndryl’s claim for breach of that alleged term of the DCHA.

In all other respects, JLR’s application for summary judgment and to strike out the claim fails.

I grant Kyndryl permission to amend its Particulars of Claim to introduce paragraphs 42A, 42B, 69A and 69B in the form provided to me.

I will hear the parties on the form of the order consequential on this judgment.