FL-2022-000014 - [2025] EWHC 2631 (Ch)

DETERMINATION OF THE ELEMENTS OF THE CLAIM

The first issue is the identification of the transaction or transactions. The claimants allege that the CEA and the TA taken together constitute the relevant transaction for the purposes of the claim under section 423.

The claimants relied on the following features:

The SBDs did not seek to suggest that these two agreements should not be treated as single transaction. However they alleged that the relevant transaction, which they called the “Greensill/Katerra Transaction”, comprised fifteen agreements or arrangements, including the understanding that the Greensill Group would apply the $440 million injected under the CLNs in purchasing or redeeming the Fairymead Notes. These fifteen agreements or arrangements were entered into in November and December 2020. They included (a) the Omnibus Deed between GPCL, GCUK, SBG, Mr Greensill and SVF II Holdings dated 10 November 2020; (b) the $440 million CLN between SVF II Wyatt and GCPL dated 10 November 2020; (c) various ancillary agreements entered on 10 November 2020; (d) the Secured Promissory Note between Katerra Inc. and SVF Abode dated 1 December 2020; (e) the Amendment Deed dated 8 December 2020 between LG Trustee, Mr Greensill and SBG; (f) the Amended Omnibus Deed dated 23 December 2020; (g) the Amended CLN dated 23 December 2020 between SVF II Wyatt and GCPL; (h) the CEA between GL, Katerra Cayman and Katerra Delaware dated 30 December 2020; (i) the TA between SVF Abode and GL; (j) the PSPA dated 30 December 2020 between SVF Abode and Katerra Cayman; and (k) the parties’ understanding that the Greensill Group would apply the $440 million injected under the CLNs in purchasing or redeeming the Fairymead Notes.

The SBDs relied on a number of features of the history in support of their contention that there was a single transaction made up of these various agreements and understandings.

The claimants disputed this analysis. They contended first that the various transactions identified by the SBDs were not linked factually.

I have already found as a fact that the various agreements that took place between November and December 2020 were seen by the SBDs and Mr Greensill as part of an overall package, having various purposes. They evolved over time (and some were superseded by others), but the genesis and essential purpose of the package was to address the potentially disastrous consequences of Katerra going into bankruptcy. That was a real and immediate possibility unless it was recapitalised. The threat was amplified by the concern that the Credit Suisse SCF funds, the chief provider of liquidity for Greensill, would probably withdraw their support if any particular Vision Fund portfolio company collapsed. I have found that the commercial genesis of the suite of agreements was the realisation that a Katerra collapse would undermine GCPL’s pre-IPO fundraising. That would seriously have damaged the interests of the SBDs and Greensill. I have also found that the package of measures that was put in place was designed to address these risks. As explained above, in broad terms this involved injecting $440 million, as the amount of money representing the outstanding amount of the Fairymead Notes into Greensill to enable it to buy or redeem them and thereby internalising the risk; and restructuring Katerra’s balance sheet and capital by a combination of fresh equity and the compromise of its liabilities. There were changes in the deal after the New Money Consortium dropped out. I have earlier set out my findings that these aspects of the commercial deal were negotiated and were seen by the SBDs and Greensill as a package or suite of agreements.

In reaching this conclusion, I have considered a number of submissions to the contrary advanced by the claimants.

First, the claimants submitted that the Greensill Group was not involved in SoftBank’s discussions with Katerra about its recapitalisation and Katerra was not involved in the discussions between SoftBank and the Greensill Group. This may be so, but it does not follow that the various agreements should not be considered to be factually linked in the sense that the SBDs would not have entered into any of the agreements relating to the relevant phase of the negotiations unless they had anticipated that the other agreements relating to that phase would also be executed. I find that they were inter-linked in that sense.

Second, the claimants observed that the agreements were between different corporate groups. The answer to this point is essentially the same as to the first. I find that the SBDs would not have entered into the various agreements (relating to the same phase of negotiations) unless they had anticipated that each of the other agreements relevant to that phase would be executed.

Third, the claimants pointed out that the Omnibus Deed assumed that the New Money Consortium would inject $176 million into Katerra; but there was no certainty that that would happen as at 10 November 2020. That is so, but it does not follow that the various agreements were not considered to be linked by the SBDs and Mr Greensill in the above sense.

Fourth, the claimants submitted that Mr Romeih said in evidence that the Katerra and Greensill transactions were not predicated on one another. The passage they relied on consisted of one answer which was not easy to follow. The question he was asked related to the position as at 5 November 2020 and was essentially whether the Vision Fund would not have been investing $200 million into Katerra unless it had expected to receive a recovery. It appeared to me that the “recovery” in the question meant a recovery in the future from the shareholding (presumably as a dividend). Mr Romeih appears to have understood that the question was concerned instead with the $176 million that was to be accounted for under the deal as understood at that time. He explained that the $176 million was the recovery that SVF2 would receive in connection with the injection of the $440 million into Greensill. He said that this was separate from the injection of $200 million into Katerra, which he said was to keep it as a going concern. He then said that while they were concurrent transactions one was not predicated on the other. He appeared to be referring to the two expected injections (of $200 million and $176 million), and the relationship between them and the payment of $176 million. At other parts of his evidence he made it clear that he thought of the various arrangements with the Greensill companies as a rescue package. He was not seriously challenged about this. I do not think that the single answer which the claimants now seek to focus on bears anything like the weight they sought to give it. I do not consider that Mr Romeih intended to say that the CEA and TA were entirely independent of the injection of $440 million.

Moreover, his answer has to be seen in the light of the large amount of contemporaneous documentation summarised earlier in this judgment which shows that the SBDs were considering the various transactions as part of a broader package. Indeed, those documents referred to “the Greensill/Katerra transaction”.

Fifth, the claimants noted that within SBIA there were two internal workstreams, with the deal teams having a Chinese Wall between them. That may be so, but again it does not run counter to the conclusion that the various documents constitute an overall rescue package in the sense described above.

Sixth, the claimants submitted that the terms changed after 18 November 2020 when the New Money Consortium walked away. There had to be a new phase of negotiations and a further round of approvals. The arrangements had to be re-worked. Indeed the proposal that 5% of the equity in Katerra would be transferred to SVF2 via GL only arose as part of the re-cut deal. Again these points are correct in point of fact, but they do not to my mind affect the conclusion that the SBDs would not have entered into the CEA and TA otherwise than as part of the wider rescue package made up of interrelated agreements. What it shows is that the arrangements evolved.

For these reasons, I conclude that the SBDs would not have entered into the various transactions listed above (as at each phase of the negotiations – i.e., pre- and post- 18 November 2020) unless they had anticipated that all the remaining transactions for that phase would be entered into.

Moreover, for the reasons given above, I have specifically found that the PSPA would not have been entered into unless the CEA had been entered into.

However, as the claimants submitted and as the earlier summary of the relevant principles shows, the identification of the “transaction” for the purposes of the claim under section 423 is not determined merely by the factual question whether the agreements were negotiated or entered as linked dealings.

I have concluded that the claimants’ position is to be preferred. The relevant “transaction” for the purposes of the claim is limited to the Impugned Transactions and does not include the various other agreements or understandings pleaded by the SBDs. My reasons for this conclusion follow.

First, the Impugned Transactions were the only agreements to which GL was a party or under which it acquired any rights or liabilities. There was no suggestion that the transactions were artificially divided. Moreover, even leaving aside the position of GL, the parties to the various agreements varied, some involving Katerra and others not, with some (but not all) involving the SBDs.

Based on the case law summarised above, the court will not generally treat agreements to which the debtor is not a party as part of a single transaction for the purposes of section 423. The fact that the agreements may have been conceived or negotiated as part of an overall package is not sufficient for them to constitute a single transaction (see e.g. Phillips CA, National Westminster v Jones).

Second, the agreements were entered into over a period of some two months. Some of the later agreements superseded earlier ones (e.g. the Amended Omnibus Deed and the Amended CLN). This makes it still harder to treat the later ones as part of the same “transaction” as the agreements they replaced.

Third, the various agreements had different purposes. Some of the agreements were concerned with making an investment into Greensill group companies. Others were concerned with making investments into Katerra. The CEA was concerned with the release of the debts and obligations of Katerra under the RPA.

Fourth, it was common ground that section 423 requires that the relevant transaction must be “entered into” by the debtor company. GL did not enter any of the agreements other than the Impugned Transactions.

The SBDs submitted that GL can nevertheless be treated as entering the Greensill/Katerra Transaction by entering the CEA and TA. I am unable to accept this. There may be cases (e.g. Feakins) where a party who has instigated or arranged or procured others to enter into agreements or arrangements is held to have entered into such agreements or arrangements, while not being a party to them. However, in the present case GL did not instigate, arrange, or procure the other agreements. On the contrary, its only role in relation to the agreements was at the end of December 2020, when it entered the final agreements. I am therefore not satisfied that GL “entered into” any agreements or arrangements other than those to which it was party.

Fifth, the cases show that when considering the scope of the transaction under section 423 the court must have regard to the purposes of the section. This requires the court to consider which arrangements were capable of causing prejudice to victims. In this regard, it is important in my judgment to bear in mind that GL was a special purpose vehicle whose only function was to facilitate the Fairymead Note programme. The securitisation structure (with the various assignments) was set up to ensure that the Noteholders would be protected in the event of insolvency of the Greensill group. GL was not therefore a normal trading company within the Greensill group. Moreover the only class of potential victims of its dealings were the Fairymead Noteholders; and the only obvious way in which it could cause them prejudice was by varying or releasing the terms of the RPA. In my judgment, the case of National Westminster v Jones shows that the court should properly focus on the particular transactions alleged to have been entered into for the relevant purpose, as this tends to promote the purposes of the statute.

I therefore conclude that the CEA and the TA together constitute the relevant “transaction” for the purposes of these proceedings. The various other agreements or understandings identified by the SBDs cannot properly be seen as a single transaction for the purposes of the claim. They had different parties. Some concerned relations between the SoftBank Group and the Greensill Group and Mr Greensill himself. Some elements were also concerned with the historical dealings between those groups, including the settlement of the dispute about the CEP and the arrangements in June 2020 when the SBG had injected $1.5 billion into the Credit Suisse funds. At least one of the arrangements was between parties within the SoftBank Group itself.

The claimants contended that the transaction was at an undervalue.

Their case is that, by entering the CEA, GL gave up its rights against Katerra under the RPA. The value of this included the realisable value of the Purchased Receivables as at 30 December 2020. The only consideration given to GL by Katerra under the CEA was the Katerra Shares issued to GL. But that was not consideration to GL since GL immediately transferred those shares under the TA to SVF2. In any event the agreed value of such shares, $11.3 million, was significantly less on either side’s case than the value of what GL gave up.

I have addressed this issue above and concluded that the value of the Purchased Receivables and secured collections was $86.2 million as at 30 December 2020 (see [572] above).

The SBDs’ case about undervalue was two-fold. First, they contended that the release of the obligations under the RPA had the consequence (by reason of the terms of the Participation Agreement) that GL was released from any obligations to pay equivalent amounts to GCUK. Hence the release of Katerra’s RPA obligations effected by the CEA had no impact on GL’s net position. This was called the “pass-through point”.

Second, they contended that the relevant transaction was a body of arrangements constituting the “the Greensill/Katerra Transaction” and that under that broader transaction GL received valuable consideration.

It is convenient to start with the second point as I have already determined that the transaction for the purposes of the claim is the CEA and the TA taken together. Hence the SBDs’ case about consideration under the wider Greensill/Katerra Transaction does not arise. But in case I am wrong about the scope of the transaction, I should address it. For the reasons given below I am unable to accept the SBD’s case.

First, as already explained, consideration is to be assessed objectively by reference to what the debtor receives and it must enure to the benefit of the debtor. It must also be capable of being measured in money or money’s worth.

Second, even on the wider view of the scope of the transaction advanced by the SBDs, the payment of $440 million under the CLN was not a payment to GL and it did not enure to its benefit.

There may be some cases where a payment to one company in a corporate group may be said to benefit other members of the group (for instance by enabling them to provide liquidity or support to one another). However, the court would require evidence of the value of such benefits. There was no such evidence here.

Moreover, as already noted, GL was an SPV, designed exclusively to function as part of the securitisation structure underlying the Fairymead Note Programme. It had no employees and no assets or liabilities other than those arising in connection with that note programme. The securitisation structure of which GL was part was designed to be isolated from the Greensill Group in the event of insolvency.

Third, I am unable to accept the argument of the SBDs that the alleged arrangement that the $440 million payment to the Greensill Group would be used to buy or redeem the Fairymead Notes itself constituted consideration to GL. In my judgment, even if a Greensill company had bought the Notes, that would merely have changed the identity of the Noteholders. As to a possible redemption of the Notes, unless and until that actually happened, GL would have derived no benefit from the arrangement. There was no contractual obligation owed to it by another Greensill company as to the use of the proceeds, so it was not able legally to require the redemption to take place.

The SBDs also submitted that GL received a benefit from the payment of the $440 million to GCUK. They said that the payment somehow enabled GL to repay the Notes and therefore reduce or avoid liabilities under the Fairymead Notes. However, the $440 million was received by GCUK on behalf of GCPL. GL did not receive it and did not have any legal right to receive it. I consider that GL received no benefit from the payment itself. Moreover, the $440 million was provided before GL itself entered into the CEA and none of the $440 million had been used by then. The issue of undervalue is an objective one. Looking at things objectively, GL received nothing in money or money’s worth for giving up its rights under the RPA.

In short, value to the debtor has to be assessed objectively from the debtor’s perspective and there was no evidence that GL obtained value in money or money’s worth from the payment of $440 million to GCUK or GCPL under the CLN.

I turn to the pass-through point. This falls to be addressed on the basis of the conclusion I have reached that the transaction consists of the combined CEA and TA. The pass-through point had the following essential steps:

I am unable to accept this contention for the following reasons.

First, as already explained the issue under section 423 requires a comparison to be made between the value obtained by the debtor for the transaction and the value of consideration provided by the debtor. The issue is not whether the debtor was left worse off.

The attempt to reframe the issue in this way is not supported by the statutory wording. Moreover it would be at odds with that purpose, which is to protect the victims of debtor companies from prejudicial transactions. Here, the SBDs accept that GL operated as an SPV, functioning as part of the Fairymead Note Programme. The efficacy of the structure depended on GL acting as a form of conduit. By giving up its rights against Katerra under the RPA, GL gave up valuable assets, being its rights over the Purchased Receivables and related security. Expressed in the terms of section 423, GL provided consideration under the transaction equal to the value (in money or money’s worth) of the assets released. This transaction was potentially prejudicial to the claimants (who are victims for this purpose) as it cut off, at source, valuable rights over which they would otherwise have been able to exercise security.

Hence, applying the first side of the comparison required by the section, by releasing its claims under the RPA the debtor, GL, provided consideration to Katerra.

The other side of the comparison requires one to ask what consideration the debtor has obtained, in money or money’s worth. In my judgment, GL cannot be said to have received consideration in money or money’s worth by reason of not having to pay participation amounts to GCUK equivalent to the amounts it would have received had the RPA not been released. In my judgment that is not consideration obtained by GL as a result of the transaction. It is simply the consequence of the existing terms of the Participation Agreement, which (in broad terms) only required GL to pass on what it received. The CEA did not operate as a variation or release of the terms of the Participation Agreement; the only difference it brought about was that (as a matter of fact) there would be no incoming sums requiring an outgoing payment pursuant to the Participation Agreement. GL was required both before and after the transaction to comply with the terms of that agreement. In my judgment, while the release of Katerra’s obligations under the RPA may have had the consequence that there would be no further amounts for GL to have to pay as Participation Obligations, that consequence cannot be regarded as consideration obtained by GL.

In short, under the CEA, GL gave up substantial assets under the RPA, the value of which would otherwise have passed to the victims of the transaction.

In reaching this conclusion I have also been guided by the consideration that it would undermine the statutory purpose of section 423 to conclude that GL obtained “valuable consideration” under the transaction.

There is a second, independent, reason for rejecting the SBDs’ pass-through point. GL owed an obligation under clause 8.1(C) of the Participation Agreement to Participation Holders (which include the Note Trustee, Hoffman and GCUK) that it would:

“not agree or permit any amendment, modification, waiver, variation or novation, or with respect to any Participated Payment Obligation or any Transaction Document without the prior written consent of the Participation Holder”.

It was common ground that no such prior written consent was given.

The claimants submitted that any benefit alleged to have been received by GL (from not having to pay the Participation Obligations) would have been offset by a liability for breach of clause 8.1(C). The claimants said that GL therefore received no such benefit.

The SBDs answered this argument by an amendment to their Defence, made in April 2025. They contended that upon entering the CEA to cancel the RPA, GL acquired a right under clause 7 of the Participation Agreement against GCUK to be indemnified against any cost, loss, liability or expenses incurred by GL in connection with any claim against GL regarding the entry into the CEA and cancellation of the RPA, including any claim for breach of clause 8.1(C) of the Participation Agreement. As a result any liability of GL to GCUK and/or the Note Trustee would be subject to set off against GL’s indemnity against GCUK.

Clause 7 provides that:

“[GCUK] hereby agrees to indemnify [GL] against any cost, loss, liability, or expense incurred by [GL] in connection with any claim made against [GL] (i) in its role as grantor of a Participation, (ii) under or in connection with the Transaction Documents or (ii) [sic] otherwise in connection with Clause 8.3 (Cooperation). For the avoidance of doubt, [GCUK]’s obligations under this Clause 7 shall continue to apply notwithstanding the transfer, assignment and/or novation of any corresponding Participation (or portion thereof) by [GCUK] to any third party.”

I am unable to accept the SBDs’ case based on clause 7. I agree with the claimants that the indemnity under clause 7 would not be engaged by a breach by GL of its obligations under clause 8.1(C) owed to GCUK.

As Lewison on the Interpretation of Contracts (8th edn.) states at para 12.142, it is inherently improbable that one party to a contract would wish to absolve the other from liability for breach of contract. An indemnity clause falls to be interpreted in the context of the network of rights and obligations contained in the contract as a whole.

In my judgment, the SBDs’ reading of clause 7 would negate the protection provided by the undertakings in clause 8, including that in 8.1(C). The undertaking in clause 8.1(C) was given in order to protect the integrity of the securitisation structure. It seems to me that clear words would be required before a reasonable reader could conclude that clause 7 would allow GL to claim indemnification from GCUK in respect of a breach of the contractual obligations it owed to GCUK under clause 8.1(C). There are no such clear words.

Moreover, clause 7 is capable of being given abundant context by covering liabilities of GL to third parties.

The consequence of the SBDs’ interpretation of clause 7 is that it would operate as a form of exclusion clause, but there is nothing in the wording to suggest that that was the intention. In my judgment, if that had been the intention, the parties would have used one of the well-known techniques of excluding or limiting liability.

In addition, the Participation Agreement is an element in the securitisation structure designed to protect the Noteholders (as the parties economically interested) and it is accordingly expressed to be for the benefit of third party Participation Holders (see clause 1.3). It would go entirely against the grain of the agreement if their rights (which include the crucial protections given by clause 8.1) were to be treated as defeated by clause 7, which places indemnity obligations under GCUK.

For these reasons, I agree with the claimants that any putative benefit to GL from being factually relieved of its obligations to pay Participation Obligations would have been off-set by a liability of GL to GCUK under clause 8.1(C) of the Participation Agreement.

I therefore conclude that the transaction was at an undervalue.

It was common ground that Mr Greensill’s knowledge, intention and purpose are to be attributed to GL. Mr Greensill accepted in evidence that he authorised GL to enter the CEA and the TA.

The claimants’ case about the relevant purpose may be summarised as follows. First, Mr Greensill knew that GL held rights against Katerra in respect of the underlying receivables and security; that the rights GL had under the RPA and related security were its only assets; that the Noteholders had the ultimate economic interest and the ultimate source of value under the securitisation structure was the receivables; that the CEA released all the rights and security of GL under the RPA; and that this changed the rights of the ultimate creditors from secured to unsecured.

Second, Mr Greensill admitted that the effect of the CEA was to forgive the debt owed by Katerra to Greensill and that in order to forgive the debt GL had to put assets out of the reach of creditors.

Third, Mr Greensill admitted that he had the relevant purpose. They relied on several passages in his evidence, culminating in this exchange:

“Q. Well I completely understand that as of November you say that you were trying to do that. But by December, you obviously are in communication, as we have seen, with Credit Suisse over the concerns about reducing their exposure and so on. And so, at that point, you don’t tell them what’s happening, you just carry on?

A. Yes.

Q. And you know, therefore, the purpose of the agreement [the CEA] is to put those assets – as you say, to put the receivables and the security to be discharged so it means that it’s out of the reach of Credit Suisse, but you still don’t tell them?

A.

That’s right, I don’t. That has been my evidence all along.”

They also submitted that Mr Greensill accepted that the transaction clearly prejudiced the creditors to the extent that they no longer had access to the assets and that this was not what they bargained for. They relied too on passages in his evidence where he accepted that he had not sought consent from Credit Suisse despite Mr Lane having advised him to do this.

Fourth, the court should draw the appropriate inference from the obvious effect of the CEA and TA, which was to remove assets from GL’s creditors or otherwise prejudice their interests, by removing the source of the assets from which they were to be paid and doing so without their consent. These consequences were not only obvious but they were intended by GL.

Fifth, Mr Greensill gave evidence that he was “between a rock and a hard place”. The rock was giving up the rights under the RPA without the consent of Credit Suisse. The hard place was the need to obtain the continued support of the SoftBank Group.

The SBDs submitted, in summary, that Mr Greensill’s essential purpose was to seek to further the interests of the Greensill Group and that he had no wish or desire to damage the interests of the Fairymead Noteholders. Indeed, he appreciated that damaging their interests would very probably lead to the withdrawal from the Greensill Group of its principal source of liquidity. They also submitted that his state of mind was to be assessed when the decision was taken to enter the various transactions and that the latest date on which he made any active decision was 3 December 2020. They also submitted that on 30 December 2020 he was bound to approve the execution of the CEA and TA. The SBDs also cautioned against the conflation of the outcome of the transactions with Mr Greensill’s purposes in entering into them.

The issue is GL’s (as the debtor’s) purpose in entering into the transaction. As I have said, it is common ground that Mr Greensill’s state of mind falls to be attributed to GL. The inquiry concerns his intentions or aims in authorising GL to enter the CEA and the TA. It does not, however, follow that his state of mind in relation to the other dealings involving the SBDs and Katerra is not relevant to the factual issue of his aims in causing GL to enter the Impugned Transactions.

In my judgment, it is helpful to start by considering what Mr Greensill knew about the purpose and functions of GL. He knew that GL was a special purpose vehicle set up and operated as part of the securitisation structure under which the SCF Subfund had the economic interest. GL was a link in the securitisation chain under which (in commercial terms) resources deriving from the SCF Subfund would be advanced to Katerra and payments made by Katerra would be paid back to the SCF Subfund. Katerra was a secured borrower from GL and GL owed obligations to pay amounts (participations) equal to receipts from Katerra to GCUK which was in turn required to pay the amounts on to other entities in the chain. The obligations of GL to GCUK were assigned to Hoffman so that in any insolvency of the Greensill Group those obligations would fall outside GL’s insolvency estate. Hence GL was insolvency remote. GL did not have any assets other than its claims against Katerra. It did not have any employees. Its only function was as a link in the securitisation chain.

Mr Greensill, a trained solicitor, appreciated all these facts about the functions of GL.

Mr Greensill also understood that the CEA would break the securitisation chain. He accepted in evidence that Mr Lane had told him at the time that by writing off the debt GL was effectively killing off the assets that underpinned the Notes, and that was effectively an asset held by CSV. Mr Lane told him that he could not kill the asset without speaking to the person who owned it beneficially. Mr Greensill said in evidence that this conversation had occurred after the $440 million had been received.

Mr Greensill also accepted in oral evidence that he understood that by releasing the claims of GL against Katerra under the RPA he was shifting CSV’s credit risk from its (indirect) claims against the Katerra receivables to its claims against companies in the Greensill Group.

As noted above, the SBDs submitted that Mr Greensill’s state of mind was properly to be assessed by reference to the dates when he decided that GL should enter into the CEA and TA. They observed that the history had several phases. First, there were discussions in late October. These led to the original commercial agreement on about 2 November 2020. The parties agreed that the lawyers should draft the agreements to give effect to the commercial deal. The documents were executed on 10 November 2020. The second phase started on about 18 November 2020 when the New Money Consortium dropped out. That phase culminated on 4 December 2020, when the GCPL Board met and approved the revised arrangements. At that meeting the Board of GCPL authorised the new arrangements by entry into the Amended Omnibus Deed and the Amended CLN, and any other document contemplated by or ancillary to those documents. Thereafter the lawyers worked on new draft documents and these were ultimately executed on a number of dates, including on 23 December 2020 for the Amended Omnibus Deed and Amended CLN, and culminating with the CEA and TA (and indeed the PSPA) on 30 December 2020.

The SBDs submitted that, on the evidence, the meeting of 4 December 2020 was the last date on which Mr Greensill made a decision about the various transactions (including the Impugned Transactions). After that date there was a process of drafting and finalisation of the agreements, but Mr Greensill’s understanding of the purposes of the various agreements did not materially change. They relied on Mr Greensill’s evidence to the effect that the decision was made much earlier than 30 December 2020. They also relied on the documents to show that Mr Greensill was not involved in considering or commented on the detailed terms of the transactions after the 4 December 2020 meeting.

The SBDs therefore submitted that since the focus of the inquiry was Mr Greensill’s aims when he decided to enter the transactions, the inquiry should essentially be concerned with Mr Greensill’s purposes and aims in the period up to 4 December 2020. They submitted that during that period before 4 December 2020 the financial position, including the liquidity, of the Greensill Group, remained reasonably healthy. Although Mr Greensill had apparently decided not to use the $440 million injection to buy back or redeem the Fairymead Notes immediately, he intended to do one or other of those things once the Katerra recapitalisation closed. In early December 2020 he believed that the Greensill Group would have sufficient liquid funds to achieve that once the recapitalisation had occurred. He had caused $250 million of the money to be placed in Greensill Bank, effectively as a deposit. The group was well advanced in its pre-IPO fundraising efforts, and Mr Greensill expected substantial further funds to flow into the group.

The SBDs contended in short that when he decided on 4 December 2020 to cause his companies to enter the various transactions he intended that the SCF Subfund would be held whole by the purchase or redemption in full of the Fairymead Notes. Mr Greensill consistently explained in his evidence that that was the purpose of the $440 million injection by SVF2 and at the latest actual decision date, 4 December 2020, he had every reason for thinking a Greensill company was in a position to complete the purchase or redemption. In those circumstances, far from having the relevant purpose, by causing GL to be a party to the CEA and the TA, Mr Greensill aimed and, indeed, positively intended to place the claimants into a better position than they would otherwise have been in.

I am unable to accept this analysis for several reasons.

First, on a plain reading section 423 requires the debtor’s purpose to be ascertained at the date on which the transaction is entered into.

Second, Mr Greensill accepted in his evidence that he authorised GL to enter into the CEA and TA on 30 December 2020.

Third, the CEA and TA only came into existence after 4 December 2020. There were drafts of the Amended Omnibus Deed and Amended CLN in existence at that time but there was no draft of the CEA or TA. Moreover the 4 December 2020 meeting was a meeting of GCPL, not of GL. Further thought must have been given to the terms of the CEA and the TA after the 4 December 2020 meeting, and Mr Greensill accepted in his evidence that he was the person who authorised GL to enter into these further agreements. I find that he was aware of their terms at the time they were entered into.

Fourth, the financial position of the Greensill Group changed materially in the course of December 2020. Among other things, as already explained:

Mr Greensill accepted in his evidence that the position at the end of December 2020 was materially different from the position in early November 2020. He referred to the business as being in peril by the end of December 2020.

He also said in his evidence that at the date when he agreed the Secondary Trade with Credit Suisse, which he put at 31 December 2020, the Greensill Group did not have sufficient liquid assets to complete the transaction at once. He said in evidence that he explained this to Mr Degen and that he agreed with Mr Degen that the Secondary Trade would take place on a best endeavours basis. I am unable to accept that he actually agreed those things with Mr Degen. The documents exchanged at the time do not contain any trace of an understanding that the trade was only to be on a best endeavours basis. Specifically, Mr Greensill’s email of 4 January 2021 explained that the close of the equity raise (the IPO process) had run into January 2021 and therefore asked to renegotiate the position by cancelling the Secondary Trade. There would have been no need for that proposal if the deal had only been to use best efforts or endeavours.

But for present purposes what matters is Mr Greensill gave evidence that Greensill was not in a financial position on 31 December 2020 to carry out the purchase of the Fairymead Notes.

On the previous day, 30 December 2020, there was, at best, some possibility that SoftBank might be prepared to advance the bridging loan, but this was speculative at best. Without that loan, there was no realistic prospect that the Greensill Group could carry out the buy-back or redemption.

In my judgment the serious deterioration in the financial position of the Greensill Group in the course of December 2020 has a material bearing on the assessment of Mr Greensill’s state of mind as at the date of the transactions of 30 December 2020. By that date Mr Greensill no longer believed that the Greensill Group had sufficient liquid funds to buy or redeem the Fairymead Notes in full. At most there was a chance or hope of obtaining the bridging loan, but that was speculative.

It follows that on 30 December 2020 the claimants’ interests under the Fairymead Note Programme could not have been “internalised” or “adopted” by an immediate purchase or redemption from the Greensill Group’s own resources.

Mr Greensill therefore understood that by releasing Katerra from its liabilities under the RPA, GL was cutting off the originating link in the securitisation chain and was transforming the SCF Subfund from being (effectively) a secured creditor of Katerra to being an unsecured creditor of the Greensill Group. In my judgment, it is necessary, when assessing the purposes of Mr Greensill at the date when GL entered into the transactions, to do so in the light of his knowledge and understanding of the Greensill Group’s financial position at that date.

Mr Greensill evidence to the effect that he was seeking to protect the position of the Noteholders (by holding them harmless) and indeed improving their position (as they would not recover fully in a Katerra bankruptcy) all appeared to me to assume that the Greensill Group had sufficient liquid assets to buy or redeem the Fairymead Notes in full. His evidence was to the effect that writing off the RPA was justified because the Noteholders were always going to be repaid in full, from the $440 million funding provided by SVF2.

But once, in light of the various events that occurred in December 2020, the Greensill Group lost the ability (the liquidity) to buy or redeem the Fairymead Notes, that justification was no longer available. As Mr Lane put it, the effect of the CEA was to kill off the rights of the Noteholders and Mr Greensill knew that the CEA would do that. He knew that the effect of the CEA was that the Noteholders were left with no more than an unsecured claim against GCUK. GCUK’s ability to meet that claim depended on the outcome of the then speculative prospects of a bridging loan from SoftBank or a revised investment proposal from TDR.

In my judgment, the financial predicament of the Greensill Group in late December 2020 assists in explaining Mr Greensill’s motives when authorising GL to enter the CEA and TA. By 30 December 2020 there had already been a series of transactions, including the Amended Omnibus Deed and the Amended CLN. Mr Greensill knew that the CEA and TA were further steps that were to be entered into. He knew that it was important for the SBDs that the remaining transactions (which essentially effected the recapitalisation of Katerra) should complete. BaFin’s requirements for rapid action, communicated on 8 December 2020, followed by TDR’s decision of 23 December 2020 to withdraw its existing investment proposal, had placed the Greensill Group in need of a huge injection of liquidity. This explains why Mr Greensill approached SoftBank for a $1.5 billion loan. As at 30 December 2020, Mr Greensill was doing all he could to persuade SoftBank to make this advance. He needed to proceed with the CEA and the TA to keep SoftBank happy even those agreements would prejudice the rights or interests of the SCF Subfund and the Noteholders. He had an immediate and pressing need to find a large injection of liquidity to deal with the requirements imposed by BaFin. That was required to keep the Greensill Group alive.

I also find that Mr Greensill benefited or suffered from the optimism of many founders of successful businesses. He probably thought that there still was a reasonable chance on 30 December 2020 that he would be able to persuade SoftBank to advance the funds or that TDR would come back with a different investment proposal. I find that at least part of Mr Greensill’s thinking in entering the CEA and TA was that he would have stood a real chance of being able to find a way of raising the funds to pay everyone off in the end. But he also knew by authorising the CEA, the rights and interests of the SCF Subfund would be prejudiced in the sense that their secured claims against Katerra would be replaced by an uncertain unsecured claim against the Greensill Group, which was itself in financial difficulties.

I also take into account the need, when considering the relevant purpose of the debtor, to focus on the position of the creditors of the debtor. I agree with the submission of the claimants that from the perspective of Mr Greensill the obvious consequence of the CEA was to remove the entirety of the security backing the Notes. Mr Greensill was a solicitor and he clearly understood the effect of the CEA. He accepted in evidence that the effect of the CEA was to remove the rights for which the claimants had bargained. He also accepted that he had done this without seeking the claimants’ consent. He also accepted that he understood that GL was a special purpose vehicle created for the purposes of the Fairymead Note Programme and that its only assets were its rights under the RPA. He did not contest the suggestions made to him in cross-examination that the CEA prejudiced the claimants by removing their rights. In my judgment this was powerful material from which to draw the inference, absent some countervailing reason or explanation, that Mr Greensill had the relevant purpose.

As I understood his evidence, the additional reasons he gave were first that he considered himself bound to enter the Impugned Transactions, and, second, that he thought it in the interests of the claimants for him to be able to continue making efforts to raise the resources to be able to pay off the Notes. As at the transaction date, 30 December 2020, this included seeking to raise money from the SoftBank Group.

I was unpersuaded by Mr Greensill’s evidence where he sought to justify entry into the CEA and TA on the basis that he considered that GL was already contractually bound to do so. First, GL was not a party to any earlier agreement which required it to enter the CEA. Second, Mr Greensill accepted in some answers in the LG transcript that he knew that GL was not bound to enter the CEA or TA. Third, some of his answers referred to being bound by the Omnibus Deed. But that agreement was superseded by the Amended Omnibus Deed. Neither agreement placed any obligation on GL to enter the CEA or TA.

I was also unable to accept Mr Greensill’s suggestion that it was in the best interests of the claimants for GL to enter the CEA in order that the Greensill Group could continue to raise resources for at least two reasons. First, I have already explained that by 30 December 2020 the prospects of the Greensill Group being able to buy back or redeem the Fairymead Notes had become speculative. Second, this was, in my judgment, an attempt to explain the means used by reference to the end being pursued. The CEA and TA, which were prejudicial to the claimants (unless the Notes had been redeemed or bought-in or there had been some other arrangement to secure that result), was a means to raising further resources.

In this regard, Mr Greensill said in some of his answers that the CEA and TA were part of a package of agreements under which the Greensill Group had raised the $440 million. That might have been a material answer had the funds indeed been used at an earlier stage to acquire the Fairymead Notes (as had been anticipated by SBDs). But by the date of the CEA and TA the money had not been used for that purpose.

Indeed, Mr Greensill’s own evidence was that the Greensill companies were not in a position to buy-back the Notes on 31 December 2020 when they entered the trade. As already explained, it is necessary to assess the issue of GL’s purpose at the date when the transactions were entered into. Mr Greensill cannot have believed that the earlier injection of $440 million, which had been used for other purposes and was not available, justified GL in entering into the CEA and TA or in any way protected the interests of the claimants (who had not in fact been protected by the use of the $440 million).

In this regard, proper regard must again be had to the special purposes and functions of GL. As explained above, it functioned as a chain in a securitisation structure, under which the ultimate economic interest was in the claimants. It was deliberately structurally isolated from the Greensill Group – so that in the event of insolvency, the claimants would be able to claim via the secured lending to Katerra without being affected by the creditors of other Greensill companies. GL was not therefore an ordinary trading entity in the wider Greensill Group. The special status is material when it comes to the assessment of Mr Greensill’s state of mind. Some of the answers Mr Greensill gave to explain the rationale for the CEA and TA were ultimately about promoting the financial health of the Greensill Group companies, which he said was in the interests of all creditors including the SCF Subfund.

But, as explained above, he appreciated the special status of GL and understood that it was not akin to the trading entities in the group. This understanding was brought home to him by the conversation in which Mr Lane said that writing off the RPA was a way of killing Credit Suisse’s beneficial interests. In short, even if Mr Greensill was thinking of the wider Greensill Group and his creditors, he also knew at the time that the CEA had the effect of cutting the securitisation chain under which the specific interests of the claimants arose.

In assessing Mr Greensill’s state of mind, it is also material that, as I have found, he misled Credit Suisse about the position. The exchange with Mr Degen on 30 December 2020 was opportunistic and misleading (see [424] above). Mr Greensill gave the impression that the release of the RPA would only happen after the buy back of the Fairymead Notes. Subsequent emails from Mr Greensill gave a similarly misleading impression. This evidence led to me to treat some of Mr Greensill’s evidence about his intentions in entering into the CEA and TA with considerable caution.

But this evidence has further significance. It shows that by 30 December 2020 Mr Greensill was willing to take considerable risks to try to keep the Greensill Group alive. He was prepared to mislead Credit Suisse in the hope that he would be able to raise liquidity from SoftBank or the pre-IPO fundraising which would enable him to find a way of buying in or redeeming the Fairymead Notes, even though there was an obvious risk that the truth would emerge.

I also find that Mr Greensill’s concealment of what had actually happened supports the inference that Mr Greensill knew that Credit Suisse would not have consented to the CEA, and he knew that that it would not have consented because its interests as a securitised creditor were prejudiced by the transaction.

In this regard, I was unable to accept the evidence of Mr Greensill that he was confident that Credit Suisse would have consented to the release of the RPA, as it was in the interests of the fund. His evidence is undermined by the steps he actually took to conceal the true position from Credit Suisse.

Mr Greensill said in evidence that he wanted to tell Credit Suisse about the release of the RPA but was prevented from doing so by the SBDs, who insisted on the confidentiality provisions in the Omnibus Deed. I do not accept that. There are several reasons for this conclusion. First, there is nothing in the documents to show that Mr Greensill sought such consent. The documents in which consent was sought concerned disclosure of the terms of the Omnibus Deed to Credit Suisse in relation to its pre-IPO fundraising activities. Second, the SBD witnesses were cross-examined on this same limited basis (see above) and they were not squarely confronted with a wider case. Third, Mr Greensill accepted the evidence of Mr Lane about their killing the securitised rights of the SCF Subfund. Mr Lane said to Mr Greensill that he could not do this without informing Credit Suisse as the beneficiary. In Mr Lane’s evidence to the liquidators about the conversation there was no suggestion that Mr Greensill had said that he wanted to speak to Credit Suisse but was prevented from doing so. Rather, Mr Greensill said that the money would be deposited in Greensill Bank but would in due course be used to repay the Notes. Mr Lane did not give evidence that Mr Greensill wanted to tell Credit Suisse but was prevented from doing so by SoftBank.

Drawing the threads together, I return to the question whether GL, through Mr Greensill, had a relevant purpose in entering into the CEA and TA on 30 December 2020. I have concluded that GL did have that purpose. The CEA had only one function, which was to release the RPA. That brought down the securitisation structure, and wrote off GL’s only asset. It thereby prejudiced the interest of the claimants. The claimants lost their interests under the securitisation structure and were left with an unsecured claim against the Greensill companies. Mr Greensill knew that those companies lacked the existing resources to meet the claimants’ claims. Mr Greensill knew that the whole purpose of the CEA was to release the RPA. He knew that that was necessary as part of the refinancing and recapitalisation of Katerra. In my judgment, he caused GL to enter the transactions on 30 December 2020 with the purpose of releasing the only asset of GL. He had to do that in order to keep SoftBank on side. That was because he was seeking a loan of $1.5 billion from SoftBank in order to deal with the demands of BaFin concerning Greensill Bank. Mr Greensill knew that Credit Suisse would not consent to the release of the RPA. That is why he did not ask for its consent. He then concealed the position from Credit Suisse in the hope that his continuing efforts to raise money from SoftBank, TDR and others would work out.

This conclusion is not to conflate consequences with purpose. The consequences or effects of a transaction are part of the material from which the court may draw inferences. The sole and obvious consequence of the CEA was to release the RPA and, given Mr Greensill’s understanding of the role of the RPA in the securitisation structure, he knew that by giving up the RPA, that structure would be brought down to the prejudice of the claimants. Giving up those rights was necessary because the recapitalisation of Katerra would not otherwise have occurred. In circumstances where the Greensill Group did not have available resources to pay off the Notes in full, and did not put in place an alternative means of achieving that, the intended effect of the CEA cannot be regarded as a mere consequence. The position would doubtless have been different if, simultaneously with the CEA, the Greensill companies had effected a secured (i.e. asset-backed) transaction to pay off the Fairymead Notes. But absent such a transaction, the purpose of what was done was to remove assets from the reach of the claimants or otherwise prejudice them.

I accept that Mr Greensill did not positively wish to prejudice the creditors of GL. On the contrary, he positively wanted to find a way if possible to hold them whole. That was indeed why he approached the SBDs in October 2020 and the reason why the SBDs invested $440 million into GCPL on 10 November 2020. Mr Greensill’s reasons for wanting to fully pay the Fairymead Notes indeed extended beyond the Fairymead Programme. The Credit Suisse funds were, overall, an essential source of liquidity for the Greensill group and a default on the Fairymead Notes would be likely to lead to a wider withdrawal of funding. Mr Greensill wanted if possible to pay the claimants in full and he would clearly have preferred not to prejudice them.

I am nonetheless satisfied that his aim or intention in entering the CEA was to prejudice their interest. The concepts of means and ends are material here. As explained by Lords Hoffmann and Nicholls in OBG Ltd v Allan [2007] UKHL 21, a person who intends to bring about an end also intends the means used to achieve that end. It may be that Mr Greensill’s ultimate end was to keep the Greensill Group afloat by raising further investments (including potential bridge finance from SoftBank and equity funding from TDR and others); and he intended also to use such further investments ultimately to repay the Fairymead Noteholders. But, as Mr Greensill knew, the CEA was a necessary means to achieve that end. He also knew that its entire purpose was to write off the RPA and enable Katerra to recapitalise. At that date Mr Greensill was trying to persuade SoftBank to prop up his companies and those efforts would have been ruined if GL had not completed the CEA and TA.

In any event, the claimants do not have to establish that the relevant purpose was the only or even the dominant purpose of Mr Greensill in authorising the CEA and TA on behalf of GL. I am satisfied that at least one of his purposes was to bring about precisely what the CEA and TA achieved namely the outright termination of the only assets that secured the rights of the claimants under the securitisation structure, which was prejudicial to their claims.

For completeness, I do not find that Mr Greensill actually admitted in evidence that he had the relevant purpose when authorising the CEA and TA. He said different things at different times. I have set out the passage most relied on in para [713] above. However, the first question was about Mr Greensill not informing Credit Suisse about the CEA. The second question rolled up two points: the purpose of the CEA and not telling Credit Suisse about it. Mr Greensill’s answer concerned the second point – “That’s right, I don’t” is an answer to the question “but you still don’t tell them?” The reason he said that it had been his evidence all along was that he had accepted throughout that he had not told Credit Suisse. Nor did I conclude that he admitted the relevant purpose in the other passages relied on by the claimants.

However, the exercise for the court is to assess the totality of the evidence and draw the appropriate inferences. I have concluded that GL had the relevant purpose in entering the Impugned Transactions.

This element was not separately in issue.

The legal principles are set out in paras [622] to [649] above.

The claimants in closing submitted that “in circumstances where the SBDs received benefits from the transactions (s425(1)(d)) through the write off of the $440 million debt and the receipt of the Katerra Shares, the just order is to require them to make payment to the [claimants] as envisaged under the statute.” They also contended that the SBDs were the cause as well as the beneficiaries of the transaction: in their pleading they contended that the SBDs had orchestrated the transaction in order to obtain Katerra free of debt.

It is helpful to start by considering the benefits obtained by the SBDs from the transaction. The claimants submitted that the CEA and TA provided the SBDs with benefits through the release of the debt under the RPA, and the receipt of the Katerra Shares by SVF II Abode. The CEA and TA enabled the SBDs to achieve a dominant equity position in the Katerra Group and the release of $440 million by the CEA strengthened the balance sheet of each of the Katerra companies, thereby improving their value to the Vision Funds. This in turn benefited SBG through its 100% ownership of SVF2 and its 33.6% investment in SVF1.

The pleaded case in the Re-Amended Particulars of Claim was that SVF2 received the Katerra Shares with “a value of US$11.3 million, alternatively US$14.4 million (or such other amount at the Court considers fit).”

As to the value of the c. 11.4 million shares in Katerra transferred to SVF1 under the PSPA, the claimants pleaded at para 80.8.2 of the RAPOC that given that they were acquired by SVF1 for $200 million (under the PSPA), a fair value for Katerra Shares (i.e. the 760,000 odd received by SVF2) would be $13.3 million odd; this however fell to be discounted to c. $11.3 million to reflect a minority discount and the fact that there was to be dilution under a Management Incentive Plan.

Hence on the pleadings the claimants did not advance a case that the value of the 11.4 million shares transferred to SVF1 under the PSPA for $200 million represented other than a fair value for those shares.

As for the valuation evidence, Mr Brown placed a value on the Katerra Shares at $14.4 million based on a discounted cashflow (“DCF”). Mr Farrell valued them at $11.3 million. Mr Farrell placed considerable weight on the PSPA as giving a contemporaneous statement of the value. He said that there was no reason to look beyond to demonstrate their value.

There was negligible cross-examination of the experts on this aspect of the case. I consider that Mr Farrell’s position is to be preferred. The PSPA constitutes directly contemporaneous evidence of value. The claimants did not take issue with this value in their pleading. Mr Brown accepted that the PSPA was a source of evidence for the valuation of the shares. He did not explain in his reports why his own DCF should be preferred to the valuation evidence provided by the contemporary PSPA. DCF valuations are necessarily subjective and dependent on numerous highly sensitive assumptions. And Mr Brown did not suggest that the PSPA did not represent an arm’s length valuation.

Despite the pleadings and their own expert evidence, the claimants sought to contend in closing that the value of the 760,000 odd shares obtained by SVF2 was $21 million. I do not accept that it was open to them to do that in light of the pleadings, their own expert evidence, and the lack of challenge to the evidence of Mr Farrell that the PSPA was the best evidence of value of the shares in Katerra Cayman as at 30 December 2020.

The claimants also contended in closing that the SBDs obtained a total benefit of $420 million. They based this on an internal SBIA presentation dated 11 January 2021 for the quarter ended 30 December 2020, which showed that the Vision Funds valued the 5% Katerra Shares at $21 million as their fair value. The claimants said that this implied a total value of $420 million for the shares in Katerra and that the shares had this value as a result of the writing off of the debt of $440 million.

I do not accept that it was open to the claimants to run this case in closing. In light of the pleadings, it is not open to the claimants to contend for a value greater than $11.3 million for the Katerra Shares. Moreover, the only intelligible pleaded case about the value of the 11.4 million shares issued to SVF1 was that $200 million represented a fair value. I also note, in any case, that the internal SBIA valuation dated 11 January 2021 did not value SVF1’s stake at $440 million. Rather it valued it at $317 million.

I find on the basis of Mr Farrell’s evidence and the claimants’ own pleadings that the value of the 11.4 million shares obtained by SVF1 under the PSPA was the $200 million price contained in that agreement.

The SBDs contended that the claimants had ignored the fact that SVF1 had to pay $200 million under the PSPA to acquire shares with a value of $200 million. In other words, the value of any benefit they obtained was matched by the amount they paid to acquire it.

The SBDs also observed that the claimants’ case ignored the fact that as part of the series of transactions on 30 December 2020 SVF1’s pre-existing shareholding in Katerra was written down effectively to nil (through dilution) so that the remaining entire value resided in the new shares issued by Katerra under the PSPA and the CEA.

As to this the SBDs submitted that:

The SBDs contended that the claimants had ignored another element of the 30 December 2020 transactions, by which SVF1 released debts of $300 million owed to it by Katerra. This was another element of the restructuring of Katerra’s balance sheet effected on 30 December 2020. The claimants did not take issue with this point.

The SBDs also contended that the benefits received by SVF2 as transferees of the 760,000 odd shares under the TA have to be seen in the light of the payment of $200 million made by SVF1 and the release of SVF1’s promissory notes of $300 million. The claimants disputed this and contended that the benefits obtained by SVF1 and SVF2 had to be considered separately.

The conclusions I reach on the benefits received by the SBDs from the Impugned Transactions on 30 December 2020 are as follows:

The claimants contended that SVF1 received a benefit from or by reason of the CEA, as Katerra’s liabilities to GL under the RPA were released. As a shareholder SVF1 therefore held an interest in a company which was $440 million better off than it would otherwise have been. However, as just noted, SVF1’s pre-existing stake was diluted to zero and SVF1 paid full value for the further stake it received on 30 December 2020 under the PSPA. Hence I am unable to accept that SVF1 received a material benefit from or by reason of the Impugned Transactions.

As already explained, the claimants contended that, in shaping the appropriate relief, the court should also take account of the culpability of the SBDs. The claimants contended, in summary, that the SBDs actions were central to the restructuring that led to the Impugned Transactions. The SBDs indeed drove the transactions and their terms in order to protect their investments in Greensill and Katerra. As Mr Greensill explained in his evidence, it was the SBDs who directed the writing-off of the RPA liabilities under the CEA, and the TA.

The claimants contended that they did not need to advance any case of dishonesty against the SBDs. They said that dishonesty is not a necessary element of the cause of action and that they did not have to show that the SBDs knew that GL had the relevant improper purpose, let alone that they shared it.

The claimants nevertheless invited the court to conclude as a matter of fact that the SDBs were aware that Mr Greensill had the improper purpose. As to this they contended that:

My conclusions in relation to these submissions are as follows. First, the SBDs of course participated in the Impugned Transactions, and they acted in their own commercial interests. As detailed above, the Impugned Transactions were part of a series of factually linked agreements which were entered into by various entities in the Greensill Group, the SoftBank Group and the Katerra Group. There were extensive negotiations running from October to December 2020. The commercial and legal terms of the transactions evolved. A key aim of the various agreements was to restructure and recapitalise Katerra so as to stave off bankruptcy. A broader aim was to ensure that the pre-IPO fundraising being undertaken by Greensill would not be derailed. The SBDs had extensive investments in both the Greensill and Katerra Groups and they wished to protect and, indeed, salvage their positions. Mr Cheung readily accepted that the steps taken by the SBDs from October 2020 onwards were designed to serve the commercial interests of the SBDs. That is to be expected. They had a keen interest in the outcome.

Second, as to the claimants’ contention that the SBDs were driving the transactions, the metaphor requires careful analysis. The SBDs were certainly anxious to seek to the bring about the series of agreements detailed above, as they considered these were needed to protect their investments. Katerra needed to be restructured and recapitalised if it was to survive. This involved new money (ultimately provided under the PSPA) and the removal of debt from its balance sheet (the release of the RPA liabilities via the CEA, and the write off of $300 million of promissory notes). Each of Greensill, Katerra and SoftBank employed lawyers to structure the deals and their commercial teams were liaising to seek to consummate them. These were commercial transactions, and each of the three parties may be considered to have been “driving” their parts of them to obtain a commercially favourable outcome. I do not consider there is any basis for suggesting that SoftBank orchestrated the transactions if by that there is a suggestion that SoftBank imposed the terms through commercial coercion or control. The claimants’ sought to rely on Mr Greensill’s evidence that the SBDs had imposed very tough terms on him. However that evidence was concerned with the personal obligations Mr Greensill had had to concede relating to private jets and a personal guarantee.

The claimants also relied on evidence about the SBDs having a substantial level of influence over both Greensill and Katerra. SVF1 was by 2020 a very substantial investor in both Greensill and Katerra. It had the right to appoint observers at Board meetings of GCPL and Mr Cheung and others attended such meetings. Mr Romeih accepted that SVF1 had a degree of influence over Katerra. The minutes of a meeting of the SVF1 Investment Committee on 14 April 2020 indeed said that the Fund was driving the major decisions of Katerra. However, ultimately the evidence shows that in November and December 2020 each of the three groups was negotiating the commercial and legal terms of the transactions with a view to its own commercial interests. No doubt SVF1 had a position of considerable influence as a key investor in Katerra. But the documents read as a whole show that the terms of the transactions represented the outcome of trilateral commercial negotiations involving external lawyers. I do not consider that the SBDs were in a position unilaterally to impose terms on either the Greensill Group or the Katerra Group.

The claimants also relied on the email dated 3 November 2020 (see [106] above) in which Mr Greensill recorded that on confirmation from Mr Son that the terms were agreed, he would inform the CEO of Katerra that they were prepared to accept such haircut on the Greensill facility as the SVFs should direct. The claimants contended that this showed that the SBDs were driving the deal. However, I find that this message represented the outcome of the relatively complex commercial negotiation that had taken place up to that stage. The deal at that stage involved the New Money Consortium investing in Katerra and an amount closely similar to that amount being received by and accounted for by GL to SVF2. That was part of the overall deal for the injection of $440 million. The amount that had to be paid, the $176 million, was part of the overall negotiated commercial package. From the perspective of Mr Greensill the amount did not really matter as it was to be passed on to SVF2. However, it was the outcome of the overall commercial package that had been negotiated at that stage (including with the New Money Consortium). It was not an arbitrary number simply imposed on GL.

It was also in this context that Mr Greensill referred in his evidence to SoftBank having their pound of flesh. But, as already explained, that was about the terms required of the Greensill Group and on Mr Greensill personally, which Mr Greensill regarded as a tough bargain. In an email to his team dated 3 November 2020 he described the terms as painful.

The claimants relied on Mr Greensill’s evidence that there was a call on an unidentified Sunday night where he was presented by the SBDs with terms that he was required to accept. I do not accept that evidence. The documents revealed there was no realistic candidate for such a call, at least in anything like the stark take-it-or-leave it terms suggested by Mr Greensill. Mr Greensill appeared to suggest that the call was at a fairly early stage in the process. The claimants by contrast suggested that it was at about the time of the message dated 23 November between Mr Greensill and Mr Romeih in which Mr Romeih said that he could lay out the various pieces that will need to be completed near-contemporaneously. That was after the New Money Consortium dropped out. The message said no more than that there would have to be package of agreements. That was hardly surprising.

More generally, the documents detailed above show that there was a continuing negotiation from October 2020 onwards. Mr Greensill did think at times that he was being required personally to give up a pound of flesh, but I do not accept that he was compelled to enter the transactions as a pre-ordained scheme imposed on him by the SBDs. Indeed, some of the documents show that the SBDs were reluctant or frustrated participants. This was particularly the case after the withdrawal of the New Money Consortium which resulted in the Vision Funds making a greater overall economic commitment. As explained above, the SBDs were doubtless negotiating to protect their own commercial interests. So was Mr Greensill on behalf of his group of companies. I do not accept the claimants’ submission that SBDs unilaterally imposed terms on Greensill. Nor do I think that the SBDs can be said to have orchestrated an improper transaction at an undervalue.

Third, I have also found as a fact that the SBDs did not know that the Fairymead Notes had not been repaid at the date of the CEA and TA. On the contrary, I have concluded that the SBDs positively believed that Mr Greensill must already have bought back or redeemed the Notes or entered a transaction whereby the Noteholders would be fully secured, as they did not think he would otherwise have been able to release the RPA. The SBDs thought that Mr Greensill had an urgent need in October 2020 to address the Katerra crisis and this is why he had sought the $440 million (the face value of the outstanding Notes). I find that they believed by the end of December 2020 that he had used the money for the purpose for which he had asked for it. I do not consider that the SBDs were culpable for failing to inquire whether he had done so. They reasonably assumed that the money had been used for the purpose for which it had been sought.

I do not accept either that the SBDs knowingly exposed the claimants to the risk that the Impugned Transactions might complete without the Fairymead Notes having been acquired or repaid. The SBDs agreed to a linked series of agreements which were designed to salvage Katerra from bankruptcy and protect the pre-IPO fundraising of Greensill. This was to be done by (among other things) injecting the amount of $440 million (the face value of the Fairymead Notes) in the clear expectation that it would be used for that purpose. That was why Mr Greensill had asked for the injection, and the documents show that that was what the SBDs expected would happen. The documents (in the parties’ jargon) referred to this as “internalising the risk”. I have already explained that the SBDs indeed thought that this had to happen before Katerra’s debts under the RPA could be compromised. I find that it did not occur to the SBDs to suppose that Mr Greensill had not entered into a secure arrangement to internalise the risk attaching to the Notes. The SBDs thought, and had reasonable grounds for thinking, that the claimants indeed would be held whole (i.e. that the risk was transferred to Greensill).

I reach this conclusion notwithstanding that there were no contractual requirements about the use of the $440 million. As I have explained, as the SBDs saw things, Mr Greensill had come to them seeking $440 million for a quite specific purpose. The amount to be injected was the face value of the Fairymead Notes. He needed the money in order to buy or redeem the Notes. The money was paid. That there were no terms as to the mechanism or timing for its use does not mean that the SBDs did not understand that the money would be used for the purpose for which it was sought. In any event, as already noted, I accept the evidence of the SBDs that they thought that Mr Greensill must have addressed the purchase of the Notes by 30 December 2020 as otherwise the RPA could not be compromised.

Fourth, I have also found as a fact that the claimants were unaware of the Impugned Transactions. As explained above, Mr Lane told Mr Greensill that he could not kill off the rights of the claimants without their consent. Mr Greensill decided to go ahead without telling them. The claimants invited the court to conclude that the SBDs refused to allow Mr Greensill to inform the claimants about the Omnibus Deed or the Impugned Transactions and this was why Mr Greensill did not seek consent from the claimants. I have rejected that submission. As already explained, the discussions between the Greensill companies and the SBDs concerned disclosure of the Omnibus Deed to potential pre-IPO investors. I have rejected the suggestion made in Mr Greensill’s evidence that he was prevented by the SBDs from disclosing the Impugned Transactions to the claimants. That did not happen. Instead Mr Greensill hoped that he could keep the show on the road by entering the Impugned Transactions and then finding additional money to pay off the claimants. I therefore reject the claimants’ submissions that the SBDs were instrumental by preventing Mr Greensill from telling the claimants about the transactions and in robbing them of their contractual right to decide whether to consent.

Fifth, I find that by 30 December 2020 the SBDs knew about the Greensill Group’s liquidity difficulties. I find that the SBDs did not however believe that these problems placed Greensill on the verge of insolvency. Mr Greensill was seeking to raise further funds from SoftBank and other investors and there was nothing to suggest to the SBDs that the Greensill Group was insolvent or on the verge of insolvency. In any event, as just explained, the SBDs thought that Mr Greensill must already have arranged the repayment or redemption of the Fairymead Notes. They thought that it would be unlawful for the RPA to be released without the repayment or purchase of the Notes.

For these reasons I also reject the claimants’ (conclusory) contention that the SBDs sought to shore up their own position even if that was at the expense of the creditors of GL. On the contrary, I repeat that I am satisfied that the SBDs believed (and had reasonable grounds to believe) that the $440 million they had injected to enable the risk on the Fairymead Notes to be internalised (i.e. removed from the claimants) had been used for the purpose for which it had been sought.

I conclude that the SBDs did not know that Mr Greensill and his companies had not used the $440 million of liquidity provided by SVF2 under the CLNs to acquire or redeem the Fairymead Notes at the date of the CEA and TA, or to make an arrangement which would secure this outcome. On the contrary, I find that they believed and had reasonable grounds to believe that by that date he had taken steps to secure the position of the claimants in full. Hence the SBDs did not know or have grounds to suspect that GL had the relevant purpose when entering those transactions.

In reaching this conclusion I have considered the claimants’ reliance on various statements made by individuals within the SoftBank Group about Mr Greensill’s representations. The claimants relied on these to support a submission that the SBDs did not trust him and, therefore, that they had good reason to be sceptical about the use he would make of the $440 million.

Specifically, in the email dated 13 June 2020 (in connection with Mr Greensill’s request for an extension of the redemption of the $1.5 billion invested by SBG into the CSV Subfund) Mr Misra said, “Lex is slippery and prone to lying so the penalty has to be high.” In a second email of the same date Mr Misra sent an email to Mr Son, saying of a statement by Mr Greensill, that “the regulator approval is a lie. We need EY to ASAP go in and audit his financials. If not we sue him and David Cameron for securities fraud.”

These first of these emails was written in response to Mr Greensill’s repeated requests for extensions in relation to the redemption of the $1.5 billion injection made by SBG. The comments had to be seen in that context. I find that Mr Misra was frustrated to the point of exasperation. This was some months before the arrangements which form the subject matter of these proceedings. I do not consider (taking the evidence as a whole) that there was a consistent or settled view within the SoftBank Group that Mr Greensill was not to be trusted. Comments of this kind are often made in particular moments of annoyance or irritation. Mr Cheung, who dealt with Mr Greensill in relation to the subsequent injection of $440 million, thought of Mr Greensill as a “happy founder,” and saw him as optimistic and visionary. Moreover it was not suggested to Mr Cheung that any mistrust expressed by Mr Misra in the first email caused the SBDs to doubt that the $440 million injection would be used for the purpose anticipated by the SBDs of buying-in or redeeming the Fairymead Notes. The second email was not put to Mr Cheung in evidence. What was suggested to Mr Cheung and Mr Romeih was that by November 2020 Mr Son and others in SBG thought that where something was to be agreed with Mr Greensill it had to be included in the contract as clearly as possible because he could not otherwise be trusted to do what he had said he would do. But that is different the submission that Mr Son or others in SBG thought that Mr Greensill was dishonest. Nor was it suggested by the claimants in their cross-examination of Mr Greensill that he had been dishonest in the events of June 2020.

The claimants also relied on the email from Mr Son to Mr Greensill dated 6 August 2020 (see para [88] above) after the redemption of SBG’s investment in the SCF Subfund saying “you need to lead towards discipline after the recent near death experience. There is no need to be so aggressive in the short term until you are sure of liquidity.” This was a suggestion that he temper his aggression. It did not suggest that he was generally not to be trusted.

For these various reasons, I reject the claimants’ case that the SBDs orchestrated the Impugned Transactions with a view to obtaining a majority stake in the Katerra group free of the RPA indebtedness. The reality is that there was a series of negotiations and agreements over more than three months under which the SBDs had the aims of salvaging Greensill’s fundraising efforts and recapitalising Katerra. The Impugned Transactions were steps in these processes, but the SBDs did not orchestrate the transactions in order to obtain Katerra shares free of debt.

The SBDs submitted that in shaping any relief the court should also have regard to the fact that their shareholdings in Katerra were cancelled in the subsequent bankruptcy of Katerra. They said that this is equivalent to the recipient of a benefit under a transaction obtaining property which loses value after the transaction date. They relied on the authorities referred to in paras [‎631] et seq. above, which show that, in shaping relief, the court may have regard to post- transaction events. They submitted that (in addition to the other points addressed above) there could be no proper basis here for granting monetary relief by reference to the value the Katerra Shares may have had at earlier dates.

The claimants contended that as a matter of law and standard valuation practice the benefits to the SBDs were to be assessed at the date on which they were acquired and that later events were irrelevant. They relied on the cases of Stanley v TMK Finance and Phillips HL. In my judgment that submission conflates two separate issues. The first is the value of the benefit received by relevant defendants. The second is what relief (if any) the court is to order against such defendants. As to that the authorities discussed in paras [‎631] et seq. above establish that relief may properly be shaped in the light of post-transaction events. This also accords with principle. Were it otherwise the court would not, for example, be able to take account of subsequent dealings with assets transferred under the relevant transaction or, indeed, improvements carried out by transferees to property they have received.

For the same reasons, principles of valuation practice cannot control the court’s broad discretion as to the appropriate relief to grant in the light of all the circumstances, which may properly include events after the transaction date.

Drawing the threads together, I have reached the conclusion that there should be no order against the SBDs. It may assist if I summarise the principal reasons.

First, for the reasons given above, the SBDs did not know or suspect that the CEA and TA were being entered into by GL for the purpose of prejudicing the claimants. The SBDs believed in good faith that the $440 million injection of liquidity would be used to pay the Fairymead Noteholders and internalise the debt. They thought that Mr Greensill must have arranged this before the CEA and the TA were completed. They did not share in the improper purpose of GL, nor did they know of it. I have also found that the SBDs did not orchestrate the transactions. They were parties to commercially negotiated agreements.

Second, SVF2 received the Katerra Shares worth $11.3 million under the TA. That was a benefit arising from or by reason of the Impugned Transactions.

Third, SVF1 received a block of shares under the PSPA. It received no benefit in relation to its pre-transaction investment in Katerra, which was diluted down to nothing. Under the PSPA SVF1 received new equity in Katerra worth $200 million in return for a cash payment of $200 million. Any benefits received by SVF1 were at least matched by the amounts it paid for them. (I say “at least” because it also gave up $300 million of debt as part of the recapitalisation of Katerra.) In my judgment it would not be appropriate in the exercise of the court’s restorative jurisdiction to make an order against SVF1 in respect of its pre-existing investment in Katerra as this was written off as part of the recapitalisation. SVF1 therefore received no benefit in respect of that investment. As to the new equity received by SVF1 under the PSPA, I do not consider that the court should order any relief in the exercise of the court’s restorative jurisdiction in respect of any value received on 30 December 2020, since SVF1 paid in full for those shares as part of the transaction under which it obtained them. The fact that SVF1 received no relevant benefits from or by reason of the Impugned Transactions would have led me to decline to order relief against it.

Fourth, it follows that I would only have been prepared to award relief in respect of the Katerra Shares received by SVF2, capped at $11.3 million.

Fifth, in any event, Katerra entered bankruptcy in June 2021 and the shares were cancelled. I do not consider that it would be appropriate in these circumstances to order any of the SBDs to make a payment in respect of the value of the shares at an earlier date. Had the SBDs continued to hold Katerra shares and had they continued to have value it might well have been appropriate to make a transfer or monetary payment order in respect of the block of shares held by SVF2: see, for instance, 4Eng at [14(2)]. But, for present purposes, the conclusion I have reached is that it would be inappropriate, where the Vision Funds simply held the shares until they lost their value, to require them to make any payment based on a value the shares may have had on earlier dates. As 4Eng shows, it is hard to see why a transferee (or other beneficiary under a transaction) who has no knowledge of the debtor’s wrongful purpose should be required to protect victims against fluctuations in the value of the assets thereafter. That would be to require such persons to become guarantors against market fluctuations. It appears to me that the bankruptcy of Katerra is an extreme case of a market fluctuation of the kind described in 4Eng.

This point would also have applied in respect of the shares received by SVF1 had I concluded (contrary to the above) that SVF1 had received a benefit from or by reason of the Impugned Transactions in December 2020.

For these reasons, I do not consider that the court should make any order for relief against the SBDs.

It might be suggested that the court should not decline to award a remedy where it has found there to be a transaction at an undervalue. Indeed, as already explained, the court said in Sequana that it is likely to be an exceptional case where relief will not be ordered. There are a number of reasons why on the particular facts of this case an order against these particular defendants would not be appropriate under the court’s restorative jurisdiction.

First, the transaction at an undervalue in this case is not the usual one of a transfer of an asset by a debtor for no or inadequate consideration. The prejudice to the victims arising from the transaction at an undervalue in this case is the release by GL of the debt and related security under the RPA. The obvious and indeed primary remedy in such a case would have been an order for the reinstatement of the debt and related security. In the case of a solvent debtor this would be straightforward. The bankruptcy of Katerra means such an order would have been pointless. But the fact remains that the obvious order the court would make to reverse the effects of a debt-release transaction falling within section 423 is not practically available. It appears to me that the cases stating that it will only be in an exceptional case that relief will not be ordered have been concerned with the usual, straightforward, case of a transfer by the debtor of its assets to another party. This is not such a case.

Second, decisions under section 423 are highly fact-specific. I have considered the benefits alleged to have been received by SVF1 and SVF2, their states of mind and culpability, and the fact that there was a supervening bankruptcy of Katerra. As explained above, had it not been for the bankruptcy of Katerra, some form of transfer or payment order might have been appropriate against SVF2. But the supervening bankruptcy of Katerra amounts, on the analogy of market fluctuations, to the extreme case of the value of an asset falling to nil. Such cases, where the value of an asset transferred as part of a transaction at an undervalue falls to nothing through no fault or action of the transferee, are likely to be unusual.

In the light of my earlier conclusions there is no independent basis for seeking relief against SBG.

It is also unnecessary to consider the SBDs’ contention that any relief awarded by the court should take account of the fault of the claimants. However I shall very briefly state my conclusions.

As already explained, in my judgment the defence of contributory negligence does not apply to sections 423 or 425. There is no requirement of causation and the jurisdiction is restorative, rather than being concerned with damage for a statutory tort. Moreover a claim under section 423 does not require a finding of negligence or fault on the part of the defendant. That is not the nature of the claim.

In any case, the defence would have failed on the facts. The SBDs’ pleaded case was that the claimants caused or contributed to their own losses by agreeing to the cancellation of the Secondary Trade. However, I have found that the Greensill Group lacked the necessary resources to perform the Secondary Trade either as at 31 December 2020 when it was entered or when it was cancelled or indeed on 14 January 2021, the settlement date. Mr Greensill accepted that there were insufficient resources as at 31 December 2020. He said indeed that he agreed with Mr Degen that the trade should be on a best efforts basis for this reason. I do not accept that part of his evidence but it shows vividly that he knew that Greensill did not have the wherewithal to settle the Secondary Trade. There was no significant improvement in the fortunes of the Greensill Group between then and the date of the cancellation of the trade, or 14 January 2021. Hence the cancellation of the trade did not worsen the position of the claimants.

Moreover, the claimants were not negligent or at fault. I have found that Mr Greensill’s communications concealed the true position from them. He informed the claimants that the release of the RPA would take effect on the buy-back and I find that they acted on that understanding. I reject the SBDs’ case that the claimants were aware of the release of the RPA at the time when they agreed the cancellation of the Secondary Trade.

I would therefore not have made any adjustment by relation to the conduct of the claimants in the period after the date of the transactions.