The issue on this appeal is whether the repayment by Comet Group plc (“Comet”) of £115.4 million of unsecured intra-group debt to Kesa International Ltd (“KIL”) on the occasion of the disposal of Comet to companies controlled by OpCapita amounted to a preference within the meaning of section 239 of the Insolvency Act 1986. The application under section 239 was brought by the liquidator of Comet (under its new name). In a judgment handed down on 17 November 2022, Falk J decided that it did. Her judgment is at [2022] EWHC 2873 (Ch), [2023] BPIR 305. The repayment took place pursuant to a sale and purchase agreement (“the SPA”) dated 9 November 2011; and repayment of the debt was formally approved by the board of Comet on 3 February 2012. The transaction was completed on the same day. As part of the completion arrangements Comet agreed to repay the KIL debt. Comet entered administration on 2 November 2012, which was converted into a creditors’ voluntary liquidation on 3 October 2013.

This appeal is brought by Darty Holdings SAS, as successor to KIL, with the permission of Falk J. Darty sought permission to appeal on additional grounds, which I refused.

The judge set out the narrative extensively; but for the purposes of this appeal the following summary will suffice. At the relevant time, Comet was owned by Kesa Electricals plc (“Kesa” or “KEP”) through an intermediate holding company, Kesa Holdings Limited (“KHL”). KHL additionally owned Triptych Insurance NV (“Triptych”), a Curacao-incorporated but UK tax-resident captive insurer. Triptych provided extended warranties to Comet customers. KIL was the group treasury company. Comet was financed through the provision by KIL of a £300m revolving capital facility (the “KIL RCF”). The interest rate on the KIL RCF was equal to KIL’s costs of funds plus 1.27%. Conversely, at a similar time the cash-rich Triptych had entered into an intra-group loan facility with KIL, agreeing to lend it up to £70 million. Both loans were repayable on demand.

Comet ran into financial difficulties with increased competition and declining footfall. It made a £3.8 million loss on ordinary activities in the year to April 2010 (FY2010), rising to £31.8 million in the year to April 2011 (FY2011). A concern developed that Comet was becoming a “drag” on KEP’s earnings and share price, and an activist shareholder started agitating for Comet to be demerged. At the same time Comet’s defined benefit pension scheme (the “DB Scheme”) was in financial difficulty. As at 31 March 2010 the DB Scheme had an estimated deficit of around £307 million. In early 2011 a recovery plan had been agreed to address the deficit, under which Comet had agreed to contribute £6.1 million per annum.

The upshot was that Kesa began looking for ways to dispose of Comet. A number of offers were made by various potential buyers; but in the end the Kesa board decided to proceed with OpCapita. Comet’s prospects had worsened since 2011, with tougher market conditions and an increased concern that it would not achieve sustainable profitability. The board minutes for 14 September 2011 also refer to an adverse change in management’s view about the longer term sustainability of Comet, in part driven by the deteriorating economic environment.

Heads of terms were agreed with OpCapita on 13 October 2011. In summary, these envisaged a sale of Comet and Triptych for £1. OpCapita would ensure that the purchasing vehicle (Newco) was capitalised with at least £30 million of share capital and a committed £40 million ABL facility. Kesa would retain the DB Scheme and provide £50 million of share capital to Newco. A form of “locked box” mechanism was envisaged by reference to the 30 April 2011 balance sheet, with forecast net debt owed by Comet of £26.9 million, the calculation of which included £42.5 million owed to Kesa. The target date for completion was 3 February 2012. The debt figures in the heads of terms reflected the way in which Comet’s figures were presented internally, namely on a consolidated basis with its sister entity Triptych.

The judge explained the structure of the transaction at [37]. She dealt first with OpCapita’s structure. A three tier structure was established to make the acquisition. At the top was a limited partnership, Hailey 2 LP (“H2L”). H2L owned Hailey Holdings Limited (“HHL”), and HHL in turn owned Hailey Acquisitions Limited (“HAL”). Kesa’s investment was to be at the top of the structure, in H2L, alongside OpCapita, or more accurately an investor or investors procured by it.

She went on to set out a summary of the terms of the SPA. So far as relevant they were as follows. The parties to the SPA were KHL as seller of the shares, KEP, and HHL and HAL as the purchasing entities. Comet was not a party to the SPA.

The SPA provided for the sale of the shares of Comet to HAL and Triptych to HHL, in each case for £1, subject to the satisfaction of certain conditions, the major ones being KEP shareholder approval and the removal of Comet from exposure to the pension scheme.

Clauses 7 and 8 dealt with the arrangements for payment of inter-company loans. Clause 8 dealt with setting up the machinery for making the payments, and clause 7 dealt with the payments themselves. In other words, the chronology of events covered by clauses 7 and 8 work in reverse order. I therefore deal with them in reverse order.

Under clause 8, Kesa was required to capitalise debt owed by Comet insofar as net debt would otherwise exceed £32.275 million, being the target of £26.9 million plus an additional amount to which the purchasing group agreed to be exposed. In the event the amount owing to KIL was approximately £129 million, and just under £13.6 million was capitalised, leaving £115,415,524 owed by Comet to KIL. It is the latter sum that the liquidator says was a preference.

There was specific provision for Kesa to procure that a board meeting of Comet would be held at which all directors other than Mr Darke would resign and the purchasers’ nominees would be appointed. Clause 8.3 provided that the newly constituted board “shall review the financial position of the Company” [i.e. Comet] in the light of a business plan for a minimum of 18 months, the ABL facility and the availability terms and conditions of the revolving credit facility to be provided by HAL (the “HAL RCF”).

The KIL RCF was dealt with in three tranches, Tranche A, Tranche B and Tranche C, representing three tranches of the proposed HAL RCF. Tranche A of the HAL RCF was £35 million, corresponding to capital injections of that amount by the investors into H2L, and then via HHL to HAL. Tranche B was equal to the amount owed by KIL to Triptych (the “Triptych Amount”). Tranche C covered the balancing amount owed by Comet to KIL plus additional headroom.

Clauses 8.9 and 8.10 provided for the Tranche A element of £35 million to be drawn down by Comet under the HAL RCF (i.e. Tranche A) and a corresponding amount being demanded by the Kesa group (in practice KIL), whereupon Comet “shall agree to repay such amount”. Prior to completion, therefore, the SPA envisaged a draw down, a demand, and an agreement to pay; but not an actual payment. The remainder of the clause dealt with Tranche B and Tranche C in a broadly similar way.

Clauses 8.11-8.15 contemplated Triptych being repaid the Triptych Amount by KIL, Triptych lending the same amount to HHL and HHL lending it on to HAL. Comet would then draw an amount equal to the Triptych Amount under the HAL RCF (Tranche B), with a further demand from the Kesa group in that amount and Comet again agreeing to repay it.

Clauses 8.16-8.21 dealt with Tranche C. It provided for KIL to make a £50 million capital contribution to H2L, together with an agreed additional amount of £22.66 million plus a further pensions related amount of £5.8 million, a total of £78.46 million. This amount would be passed down to HAL via HHL. There would then be a further demand for the balance owed to Kesa, with Comet again agreeing to repay it, using funds drawn from Tranche C.

Thus, in relation to each tranche, the position immediately before completion would be that the money would be demanded and available, but not actually paid. Clauses 7.5 and 7.6 then dealt with the actual payments.

Clause 7.5 provided that the purchasers (HAL and HHL) would procure that intercompany balances owed by Comet and Triptych, as determined as at 20 April 2011, would be repaid at completion, and the seller (KHL) would procure that intercompany balances owed by the retained Kesa group would be paid to Comet and Triptych.

Clause 9 provided for completion. Clause 9.2 required the seller and purchaser respectively to do the things listed in Schedule 2. Among the seller’s obligations under that Schedule were an obligation to procure a board meeting of Comet at which it would be resolved that the transfers relating to Comet’s shares should be approved. The purchasers’ obligations under that Schedule included making the payments required by clauses 7 and 8.

Clause 9.4 provided that if the obligations under Schedule 2 had not been complied with then Kesa could (a) defer completion; (b) proceed to completion or (c) terminate the SPA.

Clause 11.6 replaced an undertaking that had been included in the heads of terms to run Comet as a going concern with a much weaker statement of intent under which HHL and HAL confirmed that, on the basis of their current business plans, projections and related assumptions, it was their “current intention” to conduct the business as a going concern for at least 18 months from completion, and to consult with KHL if it was proposed to commence insolvency proceedings within that period.

Further detail about the mechanics of payment and repayment were set out in a completion agreement, whose terms the judge described at [46]. In short, the completion agreement mirrored the relevant terms of the SPA.

Shareholder consent to the proposed transaction was obtained on 15 December 2011; and it was not until after that that Comet sought its own external legal advice. Mr Goldring of SJ Berwin was the solicitor who advised Comet.

As envisaged by the SPA the board of Comet met on 3 February 2012. As also envisaged by the SPA the directors of Comet (apart from Mr Darke) resigned, and new directors were appointed. The meeting was presented with the completion agreement. Section 16 of the minutes of that meeting record the new board’s evaluation of Comet’s solvency. It records their conclusion that there was no ground on which Comet could be found to be unable to pay its debts. Section 16.6 of the minutes went on to record that if the proposed transactions were entered into, and Comet entered into insolvency proceedings within two years “no remedy would be available to the insolvency office holder” under section 239.

Section 17 of the minutes recorded the new board’s consideration of the completion and finance documents. Paragraph 17.3.1 recorded that the board had been advised that a number of provisions of the original drafts, would, as between a commercial lender and a corporate borrower, be subject to further negotiation. It had been possible to negotiate an improvement in some of the terms, but certain revisions were not accepted. It was also recorded that attempts to obtain finance from HSBC, Barclays and Burdale had been unsuccessful. Paragraph 17.6 recorded that the board considered the appropriateness of the company executing the documents. They reminded themselves of their fiduciary duties to the company; and paragraph 17.7 noted that without additional funding the company would inevitably run out of cash in the foreseeable future.

Section 18 of the minutes approved entry into and execution of the completion agreement, which would be “for the commercial benefit of the Company and was most likely to promote the success of the Company for the benefit of its members and creditors as a whole”. Section 19 approved the repayment of the KIL RCF, as set out in the completion agreement.

It is not suggested (and the judge did not find) that the minutes are anything other than an accurate summary of the board’s deliberations on 3 February 2012.

The completion agreement was executed on 3 February 2012, the same day as the board meeting. Comet was a party to this agreement.

Section 239 of the Insolvency Act 1986 relevantly provides:

The “relevant time” referred to in section 239 (2) is defined by section 240 (1). It is concerned with identifying the time at which a company actually gives a preference (rather than the time at which the company decides to give a preference). In the case of connected persons (which Comet and KIL were) that time is a time within the period of two years ending with the onset of insolvency; in this case the period beginning on 2 November 2010 and ending when Comet entered administration on 2 November 2012. But a time within that period is not a relevant time unless at that time the company:

The insolvency code is directed towards achieving a pari passu distribution of the insolvent estate among unsecured creditors. The rules as to preference are designed to preserve the sanctity of the pari passu principle by which creditors in a winding-up share rateably in the assets available for distribution: Goode Principles of Corporate Insolvency para 13-75. Accordingly, the justification for setting aside a disposition of assets made shortly before the onset of insolvency is that, by depleting the estate, the disposition unfairly prejudices creditors; and even where the disposition is in satisfaction of a debt lawfully owing by the insolvent company, by altering the distribution of its estate it makes a pari passu distribution among all the unsecured creditors impossible: see Invest Bank PSC v El-Husseini [2023] EWCA Civ 555.

As a result of the statutory provisions, there is a presumption in play on the facts of this case, which is rebuttable. The presumption is that in deciding to repay the debt Comet desired to put KIL into a position which, in the event of its going into insolvent liquidation, would be better than the position it would have been in if the debt had not been repaid. The second presumption, referred to in section 240, that at the relevant time Comet was insolvent, does not apply because it only applies to transactions at an undervalue.

Falk J did not find it necessary to rely on the presumption. The main building blocks which led her to her conclusion were her findings that:

Although the various grounds of appeal assert that the judge was “wrong in law and/or in fact” in reaching her conclusions, I found it hard to discern any clear-cut points of law on which it is said that she went wrong. Mr Smith KC relied on the decision of the House of Lords in Benmax v Austin Motor Co Ltd [1955] AC 370 for the proposition that an appeal court is in as good a position as a trial judge to draw appropriate inferences from findings of primary facts. That was a case in which the issue was whether an alleged invention disclosed by a patent was obvious. It is open to question whether the question of obviousness is one of inference, rather than an evaluative judgment: see Actavis Group PTC EHF v Icos Corpn [2019] UKSC 15, [2019] Bus LR 1318 at [78] to [81]. Moreover, Benmax was a case decided at a time when all appeals were appeals “by way of rehearing”. But in my opinion the approach of an appeal court has changed markedly since the 1950s; more especially since the introduction of the CPR. That process began with Biogen Inc v Medeva plc [1997] RPC 1, 45 (another case of obviousness) in which Lord Hoffmann qualified the effect of the Benmax case which he said was really about an evaluation of facts. It continued in Piglowska v Piglowski [1999] 1 WLR 1360, 1372D in which he said that appellate caution applied equally to the evaluation of facts. More recently, in Henderson v Foxworth Investments Ltd [2014] UKSC 41, [2014] 1 WLR 2600 Lord Reed said at [67]:

The general approach which therefore applies to this appeal is the current approach that an appellate court adopts in relation to appeals on fact. There is no need for me to set out the principles (yet again). I have done so in FAGE UK Ltd v Chobani (UK) Ltd [2014] EWCA Civ 5, [2014] FSR 29 at [114]; Volpi v Volpi [2022] EWCA Civ 464, [2022] 4 WLR 48 at [2] and McCarthy v Jones [2023] EWCA Civ 589 at [18] and [19]. Those principles have since been applied by this court in other cases: Kynaston-Mainwaring v GVE London Ltd [2022] EWCA Civ 1339; Deutsche Bank AG v Sebastian Holdings Inc [2023] EWCA Civ 191; Re T (Fact-Finding: Second Appeal) [2023] EWCA Civ 475. The only point that is worth repeating is one that I made in FAGE:

That observation has also been approved and applied in subsequent cases: Walter Lilly & Co Ltd v Clin [2021] EWCA Civ 136, [2021] 1 WLR 2753 at [83]; Kynaston-Mainwaring at [30]; Deutsche Bank at [50].

In the present case, however, there are two additional points to be made. First, the contemporaneous documentary evidence (in the shape of emails and other informal communications) was incomplete, because of wholesale deletion of emails at some stage after the disposal. The judge therefore had to do the best she could to fill in the gaps. Second, the judge heard the evidence of six witnesses of fact and two experts. Of the six factual witnesses, the most important was Mr Enoch. He was Kesa’s Group General Counsel, as well as being a director of Comet (until he resigned on 3 February 2012), and the company secretary of other companies within the Kesa group. The judge was critical of his evidence. She described him as very keen to defend Darty’s position and his own actions, which led to “a somewhat combative approach and, unfortunately, an impression that questions were not always answered in a wholly open and straightforward manner.” Nor did she accept that Mr Enoch’s recollection of a lack of involvement in the debt repayment mechanics reflected the full picture of what occurred.

But the judge also had the evidence of Mr Darke (Comet’s CEO) and Mr Platt (Kesa’s Chief Financial Officer) both of whom she found to be impressive witnesses.

The liquidator’s original case was that the impugned decision was taken by the (new) Comet board on 3 February 2012, when the resolutions were passed. But that case was, for practical purposes, abandoned. The case that the liquidator pursued at trial was that the relevant decision was taken on or around 9 November 2011 by Mr Enoch on behalf of Comet. But the statement of case was not amended to plead that case until part way through the trial.

The statutory question is whether the impugned decision is influenced by the desire to prefer. That decision may have been made earlier than the actual giving of the preference: Re MC Bacon Ltd [1990] BCC 78, 88. As mentioned, the repayment of the KIL RCF took place on 3 February 2012 following a meeting of the Comet board. It is now accepted that none of those who participated in the approval of the transaction on that day were influenced by that desire. If that was the date of the decision, then this appeal succeeds. But the liquidator argued, and the judge accepted, that Comet, acting through Mr Enoch, made the “real” or “substantive” decision on 9 November 2011; and that he was then influenced by a desire to prefer.

The question under this head is: when did Comet decide to repay the KIL RCF? That, to my mind, is the key issue in this appeal. As David Richards J correctly said in Re Stealth Construction Ltd [2011] EWHC 1305 (Ch), [2012] 1 BCLC 297 at [63] it is a question of fact to be determined in the particular circumstances of each case. A contractual obligation to make the repayment is neither a necessary nor a sufficient condition.

In Wills v Corfe Joinery Ltd [1997] BCC 511 the company owed money to two of its directors. In January 1994 they gave notice to the company that it was their intention to call in the loans in January 1995 and not before. That was approved and recorded in a board minute. The loans were in fact repaid by cheques drawn in February 1995, a few days before the company went into creditors’ voluntary liquidation. Lloyd J held that the date of the relevant decision was made at or immediately before the cheques were drawn. Even if the company had accepted an obligation in 1994 to repay the loans in January 1995, “it was necessary for the board to review at that time whether to honour that obligation.” In my judgment, the key to that decision was that a lot of debts were repayable by the company in January 1995 and the company had to decide which ones to repay.

In Stealth Construction the relevant parties made an oral agreement in October 2007 under which monies would be lent to the company secured by a second charge on property. That agreement was not legally binding, because of a lack of formality. Instructions were given to solicitors to draft the necessary documentation between October and December 2008. David Richards J held that the relevant decision was taken in about November 2008. As he explained, where there had been an interval of 12 months or more between the oral agreement and the execution of the charge, that would “necessarily involve a decision to proceed with the grant of the charge”. The key to that case, to my mind, is that in the absence of any enforceable obligation to grant the charge, the decision to grant must have been taken at the earliest when the solicitor was instructed to prepare the necessary documentation.

I do not regard either of these cases as laying down any point of principle. Each is a decision on its own facts. But I agree with Mr Gledhill KC that in both those cases there was more than one decision, and the question for the judge in each case was which was the operative decision for the purposes of section 239.

The judge found that Mr Enoch, who was a director of Comet as well as being Kesa’s General Counsel, led the team agreeing the terms of the transaction. He took the key role in negotiating the detailed terms of the transaction on the Kesa side. He was clearly involved in, and agreed the fundamentals, and in particular, that the various intragroup debts would be repaid. He was acting very much in an executive capacity. His job was to get the deal done. The overriding objective was to achieve a “clean break” which capped Kesa’s exposure to Comet. The judge found that the remainder of the board of Comet were content to leave the details to him, although Mr Platt was also part of the “core deal team”. Mr Enoch saw no conflict between his role as director of Comet and his role as Kesa’s General Counsel. The judge said at [244]:

The board to which the judge was referring in this paragraph was the old board, all of whom (including Mr Enoch but excluding Mr Darke) were to be replaced before completion.

In the present case, Comet had no enforceable contractual obligation to make the repayment until it entered into the agreements formally approved by the board on 3 February 2012. There was no direct evidence that the board of Comet had made the decision to repay the KIL RCF at the time of the SPA. But the judge’s inference was that such a decision must have been made, otherwise the SPA would not have been drafted as it was. She decided that the SPA was explicit about what Comet would be required to do on or before completion of the SPA. This included entering into new secured borrowings, capitalising debt and repaying all remaining intercompany balances. The SPA (to which Comet was not a party) provided that Comet “shall repay” the intra-group debt “prior to” completion, although the cash with which it would make that repayment would not be transferred until completion itself. Although the SPA contemplated that the board would “review the financial position of the Company” prior to completion, no provision was included to cover the possibility that it might find it to be unsatisfactory. Given the care usually taken in detailed documentation to cover risks of anticipated events not occurring, she inferred that the possibility of the board not falling into line was not considered to be a real risk. In reality the manner in which intragroup debts was to be dealt with was prescribed in detail in clause 8 of the SPA, and required the full participation of both Kesa and Comet. Clause 8 also required the relevant steps to be taken prior to completion, whereas clause 7.5 imposed obligations at completion. In practice it would never actually operate.

She placed particular weight on her assessment of the contemporaneous documents (in particular, the terms of the SPA) and the commercial realities at the time, together with Mr Enoch’s group-wide role and his failure to identify any difference of interest between Kesa and Comet.

The judge found that by the time that it got to 3 February 2012 the board’s “hands were tied”. At best there was a binary decision whether to go ahead with the proposed transaction as a whole or to refuse to do so. Although it was theoretically possible for them to refuse to pass the necessary resolutions, if they had done so they would have been sacked and replaced by directors who would pass them.

The judge concluded at [245]:

She continued at [248]:

The judge considered whether the fact that it had not been put to Mr Enoch that he was acting for Comet on 9 November 2011 made any difference. She decided that it did not. She said at [252]:

Having then considered the minutes of the board meeting on 3 February 2012 she finally concluded at [274]:

Mr Smith argued that the decision to make the repayment of the KIL RCF was made by Comet’s board at the meeting on 3 February 2012, as the board minutes record. Until that time, there had been no decision to repay. Although it may have been likely, or even very likely, that the board would decide to make the repayment, that is not the relevant question. The relevant question is when the decision was actually made. The judge herself found at [256] that, at least formally, the board took the decision on that date; and that the directors at that time believed that they had a decision to make. She also concluded that, at least in theory, the board could have refused to pass the necessary resolutions approving the transactions. On the basis of those findings, the relevant decision was the board’s decision on 3 February 2012. As at that date none of the directors of Comet had a desire to prefer KIL.

Mr Smith also argued that, accepting all the limitations on the interference by an appeal court with findings of fact (including inferential findings) made by the trial judge, the judge’s conclusion that the “real” or “substantive” decision was made earlier was wrong. There was simply no evidence to support it. First, Comet was not a party to the SPA and therefore execution of the SPA could not itself be, or be evidence of, a decision made by Comet. Second, there was no contemporaneous evidence of any decision made by Comet at that time. Given the magnitude of the obligations that the SPA envisaged that Comet would take, and the fact that Comet was a substantial public company, it is not plausible that a decision of that nature would leave no documentary trace. Third, the structure of the contemplated transaction did not require any decision to be made by Comet when the SPA was made. It was no more than an agreement by the shareholders of Comet to sell their shares. On the contrary, the terms of the SPA envisaged that Comet would make the decision on completion. Fourth, the judge was wrong to say at [239] that no provision was made to cover the possibility that Comet might find the transaction to be unsatisfactory. Clause 9.4, which gave Kesa the right to terminate the agreement if the obligations in Schedule 2 were not fulfilled, did precisely that. Fifth, although Mr Smith did not challenge the judge’s evaluative conclusion that the parties to the SPA did not consider there to be a real risk that the board of Comet would not fall into line, that did not amount to a decision by Comet. An expectation (even a strong expectation) by A that B will do something, is not a decision by B. Sixth, clauses 7.5, 8.10 and 8.15 all contain obligations to “procure” that things shall be done. Those are forward-looking obligations. If Comet had already made the decision to approve the transaction and make the payments, obligations of that kind would have been unnecessary. Seventh, Comet did not engage with the transaction at all until after shareholders’ approval had been obtained on 15 December 2011, over a month after the SPA was signed. Eighth, the judge was wrong to regard the payment of the KIL RCF as the only relevant part of the transaction. The KIL RCF would not have been paid unless the HAL RCF was accepted. It is clear that at the board meeting of 3 February 2012, the board considered whether to enter into the HAL RCF; and equally clear from the minutes that Comet had sought alternative means of finance before accepting the terms of the RCF. They had, in addition, proposed changes to the terms of the HAL RCF which had been accepted. Ninth, the judge did not identify what decision Mr Enoch made, or how he made it. Did he decide that Comet would pay the KIL RCF; or merely that the SPA should take the form that it did? There is a qualitative difference between authorising a single director to negotiate the terms of a proposed transaction on a “take it or leave it” basis, and a decision actually to take it. How did Mr Enoch make the decision? No act was ever identified. Tenth, if Mr Enoch had made any decision on behalf of Comet, it was not a decision that he communicated to anybody. Eleventh, the judge found that on 3 February 2012 the board took decisions and believed that they had a decision to make. Although the judge characterised it as a “binary decision”, a binary decision is still a decision. She also accepted that the board could have refused to approve the transaction and/or refused to repay the KIL RCF. Even if the judge was right to say that had they refused, they would have been replaced by compliant board members, that does not detract from the fact that the decision was in the hands of the board on 3 February 2012. The judge said that the transactional documents were “carefully choreographed” but she did not say that the board meeting of 3 February 2012 was a sham or merely a charade. On the contrary, she accepted the evidence of Mr Darke that there was a difficult decision to be made.

Mr Gledhill stressed the fact that the judge’s decision about when the “real decisions” or the “substantive decision” were taken was an inferential finding of fact. In both Wills and Stealth the judges made findings about the date of the relevant decision in each case on the basis of inference. It is perfectly possible for an operative decision to be a conditional one. In this case the judge found that the operative decision was made in November 2011, even though it may have been conditional on approval by Comet’s board.

He also stressed the point that the factors that led the judge to the inferential finding that she made should not be picked off one by one. They operated cumulatively. To borrow the words of Parke B directing the jury in R v Exall (1866) 4 F & F 922:

The features that he relied on were:

I accept Mr Gledhill’s point that an appeal court must be restrained in reversing a finding of fact, even an inferential finding. I accept also that, at least in principle, an operative decision may be a conditional one. But much will depend on the nature of the condition. If, for instance A agrees to buy Blackacre subject to the grant of planning permission, that can properly be characterised as an operative conditional decision. But if he agrees to buy Blackacre “subject to contract” that is not an operative decision at all. In my judgment a decision which is conditional on board approval (or ratification) does not amount to an operative decision. A further decision by the board is necessary to make the operative decision: compare Goodwood Investments Holdings Inc v Thyssenkrupp Industrial Solutions AG [2018] EWHC 1056 (Comm).

In the course of his oral submissions, Mr Gledhill said that the old board effectively made a decision about, at the very least, whom the new board was going to be asked to pay at completion. But even if that submission is correct, it does not amount to a decision by the old board that the new board would do what they were asked to do.

The judge placed particular weight on the terms of the SPA and the “commercial realities”. The main points that impressed her about the terms of the SPA were the prescriptive description of what Comet “would be required to do” on or before completion of the SPA; and the lack of any specific provision covering a failure by Comet to do “what was expected of it”. The latter phrase is telling. Comet was, no doubt, “expected” to do what Schedule 2 of the SPA envisaged, but as Mr Smith submitted, an expectation by A that B will do something is not a decision by B. An expectation by A that B will act in a certain way in the future, is incompatible with B’s having already decided to do that thing. Second, the SPA did not actually require Comet to do anything. It required Kesa to procure that Comet would do various things, but that was in the future. Third, in saying that the SPA made no provision for a failure by Comet to do what was expected, the judge overlooked the terms of clause 9.4 of the SPA which dealt with that eventuality. Fourth, the terms of the SPA specifically provided for a future decision of the (new) Comet board. The judge was also mistaken in saying that because of the steps envisaged by clause 8 of the SPA, clause 7.5 would not come into operation. The two clauses were dealing (albeit in reverse order) with what would happen before completion in setting up the mechanics for payment, and the making of the actual payments on completion. The judge placed some reliance on the fact that the SPA made no reference to the fiduciary duties owed to Comet by its directors. But that is only significant if Comet had made a decision to repay the KIL RCF at the date of the SPA, which is the very question to be decided. If Comet had made no such decision (but was expected to in due course) then a reference to fiduciary duties was unnecessary.

So far as the commercial realities are concerned, the judge may well have been right to find that by the time of the board meeting on 3 February 2012, the board had little choice but to decide to approve the transaction. As Mr Gledhill submitted, the overall transaction required Comet to enter into the HAL RCF out of which the KIL RCF was to be paid. Mr Gledhill accepted, however, that there was no reason to suppose that the (new) Comet board perceived their hands to be tied as a result of some earlier decision taken by Comet. Indeed, the board minutes record that the board had approached other potential sources of funds before deciding to enter into the completion agreement; and it is accepted that the board minutes are accurate. Those approaches are themselves inconsistent with the conclusion that the decision to repay the KIL RCF out of the HAL RCF had already been made. Moreover, as Newey LJ said in argument, if the commercial reality were that in due course Comet would have no choice but to agree, then there would have been no need for Comet to agree at the date of the SPA.

There is also an inconsistency between the judge’s conclusion that, on the one hand, there was a decision to be made on 3 February 2012 and that the board could have refused to approve the transaction and, on the other, that the decision had already been taken.

Mr Gledhill relied, as he had before the judge, on the decision of this court in Re Drabble Bothers [1930] 2 Ch 211. But that was a very different case. Drabble Brothers were a partnership of two brothers, George and Frederick Drabble carrying on business as builders. Frederick Drabble signed a cheque in favour of Swan & Co Ltd, a creditor in relation to a contract relating to a development in Stoke. The cheque was held to be a fraudulent preference, even though Mr Drabble himself had no intention to prefer the payee. But on the facts of that case, Frederick Drabble was not the decision-maker. The decision-maker was a man called Tiley, who did have an intention to prefer. That is clear from the judgment of Lord Hanbury MR. I quote a few extracts:

In other words, on the facts of that case, decisions about whom to pay were Tiley’s decisions. The questions in this case were whether Mr Enoch made a decision at all, and if so, what that decision was.

Mr Gledhill also took us to the facts of MC Bacon. In that case there were three directors of the company, Mr Creal, Mr Glover and Mr Martin Creal. The impugned decision was a decision to grant a debenture to the bank. Mr Glover conducted the negotiations with the bank, but resigned as a director on 3 April 1987 shortly before the debenture was granted. Nevertheless, he continued to be involved in the financial affairs of the company. The debenture was granted on 20 May. Millett J found that the decision to grant the debenture was made by all three men: MC Bacon at 86F. Although only Messrs Creal actually executed the debenture, they were influenced by a recommendation to do so from Mr Glover. Thus, Millett J considered whether Mr Glover was influenced by a desire to prefer because if he was, then the company was similarly influenced, even though Mr Glover did not communicate any such desire to Messrs Creal. In the event, Millett J found that Mr Glover was not so influenced. But the point is that Mr Glover was one of the decision-makers, so his desire was relevant. Mr Gledhill’s reliance on MC Bacon overlooked that critical finding of fact.

On the question whether Mr Enoch did make a decision, it was not put to him in the course of his evidence that he made any decision on behalf of Comet. Mr Gledhill rightly said that Mr Enoch gave no evidence about it in his witness statements, but that was because at the time when he made his statements it was no part of the liquidator’s case that he had made a decision on behalf of Comet on or about 9 November 2011. The judge accepted that it had not been put to Mr Enoch that he had been acting on Comet’s behalf and said that she had assumed that had the point been put, he would have denied it. But that, to my mind, is no answer to the question what he was actually doing on behalf of Comet. As Mr Gledhill accepted, he could not have been acting on behalf of Comet when he actually signed the SPA which he did on behalf of the Kesa parties to the SPA (who did not include Comet). Thus, the judge’s conclusion at [274] that the decision was made “when the SPA was signed” cannot be sustained.

Assuming, however, that Mr Enoch did make a decision on behalf of Comet, what was the decision? Was it a decision that Comet would repay the KIL RCF; or was it no more than a decision that the terms of the SPA would be the terms that would be presented to Comet for approval? That question, too, was not explored in evidence, and the judge did not, I think, grapple with it. Mr Darke accepted in the course of his cross-examination that he played no part in the negotiation of the SPA, so the judge was probably entitled to find (as she did at [244]) that the Comet board left it to Mr Enoch to negotiate the terms of the SPA. But it was never put to Mr Darke (or anyone else) that Mr Enoch was authorised to decide that Comet would repay the KIL RCF. So, there was no evidence that the (old) Comet board were content for Mr Enoch to decide that Comet would enter into the transaction, especially since the SPA provided for the (new) Comet board to take the decision. There was, therefore, an evidential void which was not capable of supporting the judge’s inferential finding in [244] that the board was content to enter into a transaction decided upon by Mr Enoch.

Chen v Ng [2017] UKPC 27 demonstrates the dangers inherent in a trial judge making findings about matters that have not been put.

In addition, if Mr Enoch made any decision on Comet’s behalf at the time of the SPA, it was not a decision that he communicated to anybody. Nor was it suggested to him (or indeed to any of the witnesses) that he did. Mr Goldring, the solicitor advising Comet in the early part of 2012, said (and the judge accepted at [250]) that he got the impression the Comet was “expected” to comply with the wishes of the other transaction parties. If a decision had already been made, he knew nothing of it. If neither the other existing board members of Comet at the date of the SPA, nor the new board members who approved the transaction on 3 February 2012, nor the solicitor advising them knew that a decision had already been made, it is difficult to see how any earlier decision by Mr Enoch could have been an operative decision by Comet; or, indeed, to have had any influence outside Mr Enoch’s own mind. It would have been a decision without any impact on anything.

Accepting all the points made by the judge, they do not, to my mind justify the conclusion that a decision by Comet to repay the KIL RCF had been made at the time of entry into the SPA. No doubt Kesa and OpCapita were arranging matters so that Comet would have little, if any, choice but to accept the terms on offer. But the fact that a creditor puts pressure on a debtor to repay a debt does not mean that the debtor has decided to repay it. Even if the new board had little choice but to accept the terms on offer on 3 February 2012, it does not follow that Comet had already made a decision in November 2011.

I agree, therefore, with Mr Smith that there is no basis in the evidence for the judge’s inferential finding that on or before the making of the SPA Mr Enoch made an operative decision on Comet’s behalf that Comet would repay the KIL RCF. The only operative decision was the decision of the board on 3 February 2012; and it is accepted that the decision made on that date was not influenced by a desire to prefer.

It follows, in my judgment, that this ground of appeal succeeds; and the remaining grounds of appeal do not arise.

I would allow the appeal.

I agree.

I also agree.