Kohli v Lit & Ors

This is the judgment of the Court, prepared by Mann J.

This is an appeal from an order of HH Judge Purle QC given in 4 separate judgments. The first was one in which he found that the respondent in this matter, (Ms Kohli) a petitioner in a Companies Act section 994 petition, had been unfairly prejudiced and in which he ordered that her interest in the subject company (Sunrise Radio Ltd – “Sunrise Radio”) be bought out by the majority shareholders. The principal shareholder was a Dr Lit, and although he was not the only respondent to the petition, for ease of presentation this judgment will assume that he was. Having so determined in the first of his judgments ([2009] EWHC 2893 (Ch)), Judge Purle then held further hearings in order to determine the price to be paid. He decided various elements of the price across the remaining 3 hearings ([2011] EWHC 3821 (Ch), [2012] EWHC 1337 (Ch) and [2012] EWHC 1338 (Ch)). When it is necessary to distinguish between those judgments and the hearings they related to we shall refer to them as the liability judgment and the first, second and third quantum hearings and judgments respectively. The end result was an order that the price paid should be £560,000. The appellants in this appeal (the respondents to the petition) do not seek to appeal the finding that they should buy out Ms Kohli. Nor do they dispute much of the valuation technique adopted by the judge. What they do seek is to challenge his treatment of 4 particular elements of his assessment, saying that they were dealt with inappropriately or inconsistently with other elements, resulting in a price that was too high. This appeal is brought with the permission of Lewison LJ given on 11 October 2012 (who refused permission to appeal on a fifth ground).

It is unnecessary to recount much of the background to the valuation exercise, which appears in the liability judgment. The following short points will suffice.

At the material time the majority shareholders in Sunrise Radio were a Dr Lit and a company controlled by him, Asian Broadcasting Corporation Ltd. By the time of the trial they held almost 89% of the share capital, having increased their shareholding from 78% as a result of a disputed rights issue. Ms Kohli was a minority shareholder. There were some other small shareholdings that do not matter.

Sunrise Radio had its own trading income, but it also had a number of subsidiaries. Some of them had their own trading income. They included Club Concorde Ltd and Sunrise TV Limited. A further subsidiary, Hayes Gate House Ltd, held a valuable property. Other subsidiaries were loss-making.

In his main judgment on liability the judge had determined that the value of Ms Kohli's shareholding should be taken to be her rateable proportion of the value of the Sunrise Radio shares as a whole without any discount for her being a minority shareholder. His order of 25th May 2010 embodied the effects of his determination of unfair prejudice and provided for the share buy-out. It also provided:

Global Radio Services Ltd and Asian Broadcasting Corporation Ltd were not subsidiaries of Sunrise Radio. For these purposes the debts were therefore not intra-group debts. They had the nature of external indebtedness, albeit to other Lit entities. They seem to have been specifically referred to because one of the issues that arose in the case was the propriety of invoices raised by the two external companies, including invoices which gave rise to the two debts just referred to. That propriety was challenged by Ms Kohli. Having considered the propriety of the alleged underlying transactions, the judge ruled that not all of the Global Radio Services debt was good and that:

The debt due from Club Concorde to Asian Broadcasting Corporation seems to have been dealt with in paragraph 203:

It is necessary to set out that piece of detail from the liability trial because of the use that Mr Peter Griffiths, who appeared for the appellants before us, sought to make of the reference to the debts in the order. As will appear, one of his complaints is that the valuation exercise was conducted without reference to some external debt, including the 2 debts referred to. He pointed to the oddity of that, bearing in mind that the judge had made particular findings in relation to the existence of the debt. However, it is useful to record at this point in the judgment that such a submission has to be treated with care once one sees the context in which the judge made the relevant findings. The paragraphs which we have cited appear in a long section of the judgment entitled “Remuneration, consultancy fees and accounting”. The section is dealing with complaints about the amount which Dr Lit took out of the company by way of consultancy fees or service charges. The judge had to deal with an enormous amount of detail, and the two conclusions to which we have referred are among many others that he reached in considering the propriety of various charges. He was really making findings on matters of complaint by the petitioner as to how the business was conducted. So far as he was making findings that might impact on a subsequent valuation exercise, his findings clearly have to be read subject to a qualification of relevance. One cannot simply point to the provisions of the order without more (as Mr Griffiths did) and complain that the judge's specific determinations were not reflected in his findings on valuation. He was providing factual assumptions for the purposes of the valuation, not indicating how those factual assumptions were to be applied in that exercise.

When it came to arguing the price to be paid, each party had expert evidence. Miss Kohli instructed a Mr Alan Thompson; Dr Lit instructed a Ms Moira Hindson. It is plain from his first quantum judgment that the judge had significant misgivings about important aspects of the evidence of both experts (more in relation to Mr Thompson than in relation to Ms Hindson) and while he proceeded from the starting point of both experts he then took what seems to have been his own route, guided by Mr Hanke (then counsel for the petitioner), while from time to time treading the same path as one or other of the experts.

So far as the trading side of Sunrise Radio was concerned, the experts agreed that the appropriate method of valuation was to ascertain maintainable earnings and to apply an appropriate multiplier to them. There were disputes as to the appropriate period across which to ascertain the net earnings, and as to the multiplier, on which the judge made findings which are not the subject of this appeal. The judge took the earnings for the years 2006 to 2009 and he chose an appropriate multiplier.

His valuation technique does not appear, in terms, in any of his judgments. It emerges in the course of all three of them, though mainly in his first. It can be seen from a table that he sets out at the end of his third quantum judgment. He takes the value of the trading side of Sunrise Radio’s activity, valued as just referred to (applying a multiplier to net earnings), and adds the value of Club Concorde similarly determined. He then brings in an agreed value of Sunrise TV Ltd. He ignores all other subsidiaries on the footing that they are loss making (or, in the case of one, on the footing that it had a zero value). Then he adds to that the value of the property held by Hayes Gate House Limited at a valuation, and takes away from the aggregate of those positive sums the amount of a bank facility afforded to the group by Allied Irish Bank (AIB) together with a redemption fee that would be chargeable on redemption. That gave him what he regarded as the value of the holding company, to which he then applied Ms Kohli’s proportionate shareholding to arrive at a rounded figure of £560,000.

The dispute on this appeal turns on what are said by the respondents to the petition to be the following errors. Three of them are complaints about the manner in which he arrived at net earnings (complaints that the judge wrongly failed to add items on the expenditure side, or that he added items to the income side which ought not to have been allowed, with the result that the net earnings were too high), and one relates to the next stage of the operation when it is said that he failed to take out particular debts from the overall calculation of value. The complaints are:

Ms Kohli had also sought permission to challenge a finding that a purchaser would be likely to allow loss-making subsidiaries to go into liquidation, but permission to challenge this was refused.

Under this head Mr Griffiths challenges that part of the decision of the judge which deducts liabilities from the overall valuation reached by applying the earnings/multiplier technique. He does not complain about the deduction of bank borrowing as a liability. His point is that it is illogical and contrary to the evidence to stop there and not to deduct all the net liabilities of the group (and particularly those on which the judge had made specific findings - see above), though if one takes out the loss-making subsidiaries (which one has to do in the light of the ground of appeal on which he did not get permission) then the liabilities attributable to them should not be brought into the picture. Mr Griffiths points out that Mr Thompson had not taken account of any of the subsidiaries in his valuation, and this does not seem to have found favour with the judge. Ms Hindson’s evidence did look to the subsidiaries, including their liabilities, but she did not single out the bank liability for special treatment. She made a deduction for net liabilities across the group.

The judge’s findings on this point appear across two of his judgments. In the first quantum judgment he rejected Mr Thompson’s methodology in paragraph 14, and in paragraph 15 he criticised part of Ms Hindson’s methodology insofar as it related to the bank debt, though not the deduction of the bank debt as a liability. He said:

That does not deal with the treatment of the liabilities of the group as a whole. He then went on in paragraph 16 as follows:

He then moved on to other matters relating to maintainable profits. His ultimate methodology follows the pattern of those paragraphs in that it does deduct the bank debt from the values which he has established for the trading side of companies’ activities and adds the value of the property. He does not, however, explain why it is that he brings only the bank debt into account. This is despite the fact that Ms Hindson did bring other debts into account in her report.

However, as a result of reading Mr Griffiths’ skeleton argument supporting his application for permission to appeal the judge gave some further reasoning. In the second quantum judgment he said:

Thus his reasoning in paragraph 3, which takes a bit of unpicking, seems to be this:

It is not easy to follow the logic of this part of the judgment, and not easy to reconcile it properly with some of the expert evidence given. It is right that the shared assumption of the valuers was that an earnings/multiplier technique was the appropriate one to adopt. It is also right that net assets or net liabilities determinations would by and large have no part to play in such a technique. But the judge applied a modified version of that in deducting the bank debt. He justified that by reference to his finding that a purchaser would want to pay it off on completion. That is a finding which is not challenged on this appeal. Its effect is said by the judge to justify taking out any other element of net liabilities from the end calculation (step (iii) and step (x)), but we do not see why that consequence follows. Nor is it fully consistent with the expert evidence on the point, and the judge’s technique was not properly pursued with any of the experts.

The judge’s reasoning pre-supposes that there are aspects of assets and liabilities which can be brought to bear in an earnings/multiplier based calculation of value. That is the assumption underpinning his allowance of the AIB debt as a deduction in his figures. Mr Thompson did not deal with this. Ms Hindson did. In her report she said (first in relation to a valuation in 2005, necessary for considering the effects of the disputed rights issue):

She then went on to value the profit-generating subsidiaries on an earnings basis, but said that loss-making subsidiaries should be valued on an asset basis “which recognises the overall value of their net assets or liabilities”. Having produced her positive values for the holding company and the three profit-making subsidiaries, she said:

She then carried out an exercise of consolidating the activities of the group to produce a group balance sheet, eliminating such things as inter-company debt and arrived at a net liability for the group. The total net liability owed by the group as at 31 December 2005 was £2,090,000, which she said should be deducted from the value of the business of the company to arrive at a value for its entire issued share capital. However, she then carried out an adjustment to remove the £150,000 which the judge had ordered be taken out of the debt owed by Sunrise Radio to Global Radio Services. That demonstrates that she was treating the larger debt as being part of the external indebtedness which, in her view, needed to be taken into account.

Thus her valuation exercise was not one conducted by taking into account an earnings/multiplier calculation by itself. She indicated it would be appropriate to take at least some liabilities into account. Her key concept seems to have been “assets and liabilities… that are surplus to its trading activities”.

It is not apparent from the cross-examination of Miss Hindson that the whole principle of taking some liabilities into account was properly challenged. The point was touched on in cross-examination, in a passage starting from something said in the agreed experts’ statement:

This cross examination only goes so far in dealing with the point now under consideration because it focuses on large liabilities, and in the last two questions she was asked to assume a sale of Sunrise Radio only (ignoring the subsidiaries and therefore debts in the subsidiaries). She seems to have been interrupted when trying to deal with some of the points. However, it does not seem that she abandoned her concept of taking into account assets and liabilities which were surplus to trading activities, and she was certainly not accepting that one automatically ignores all debts. This reference to the evidence also demonstrates that the reason that Ms Hindson took liabilities into account was not because of some “inter-dependency”, but because of something else – her concept of a liability (or an asset) being surplus to trading activities.

Mr Thompson’s report does not go into the point. Nor does his cross-examination contain anything on it.

This does not provide support for the finding that it was appropriate to take the bank liability into account because of some element of inter-dependency (stages (ii) and (iii) of the judge’s reasoning). It would be appropriate to take it into account on the footing that it is a sort of surplus liability, in the sense of one not being part of the normal trading liabilities of the company (which we can see might be left out of account in an earnings-based calculation of a business), particularly since a very large part of Sunrise Radio’s liability arose under a guarantee and was not its primary liability. But it is not apparent why this inter-dependency (if that is what the judge meant) meant that it, and only it, had to be taken into account, and that was not what Ms Hindson said. Accordingly, so far as the judge’s reasoning justifies the deduction of the bank debt by reference to an inter-dependency, which is thereby removed, making further deductions inappropriate, we do not think that it can be correct. It is, of course, correct that the judge did not accept Ms Hindson’s evidence in its entirety, and that is something that he was entitled to do, but to the extent that he does have resort to it his reasoning does not seem to work.

Nor was it fair, in the circumstances, to base his decision on a failure by the Lit parties to adduce evidence as to the appropriateness of the deduction of other particular debts (his steps (iv) and (vi)). Ms Hindson and Mr Thompson had not approached the matter in that particular way. As we understand it the idea of taking a price/multiplier figure and deducting a specific debt emerged in Mr Hanke’s final speech. We do not understand at what point the Lit parties should have addressed this point evidentially, bearing in mind the manner in which and time at which it came up. The passage in Ms Hindson’s cross-examination cited above shows that a start was made in putting such a case to her, confined to the AIB debt, but no more.

What we can accept is the judge’s unchallenged finding that a purchaser would want to pay off the bank debt, and we can see that he would be likely to want a discount from the price in that amount, but we do not understand why that means that all other liabilities (reflected in net liabilities) have to be ignored. Presumably if this is to have an effect on the earnings/multiplier element on the valuation it must be because it is a liability surplus to its trading activities, in Ms Hindson’s analysis, or has a “significant impact” in the judge’s terminology (step (xi)). If other liabilities were sufficiently serious, and had the quality of being “surplus” to the business activities, then a purchaser may well want to deduct those as well, because the company would still be subject to them. This is particularly likely to be the case where the debt is in a subsidiary which has a positive value on a multiplier-based calculation, which is the case with Club Concorde.

We therefore consider that the reason for drawing a distinction between bank debt and other debt is flawed. The judge held that it would be right to discard loss-making subsidiaries, so the portion of the net indebtedness attributable to them would not cause a purchaser to make a deduction from the price (save for any bank debt caught by the guarantee). The deemed purchaser would let the companies go, so that they would have a nil value to the purchaser, not a negative value. To that extent the logic of the judge is sustainable. His “significant impact” point (point (xi)) may be a reasonable way of capturing the “surplus” point referred to by Ms Hindson, or her “if big enough” point made in the cross-examination passage cited above. However, it is unfair to say that there was no evidence of any such debts bearing in mind the way the cross-examination went, the fact that the point was not investigated with Ms Hindson, and the fact that this sort of analysis only emerged in final submissions (if then - it may have been the judge’s own).

All this means that while the judge was correct to acknowledge the principle of some debts being deductible for the purposes of carrying out the valuation, he was not necessarily right in confining those debts to the bank debt. If he was going to go about the matter in the way in which he did then he ought to have considered other candidates.

The next question is what this court should do about that. It would be open to us to remit the valuation question for further hearing below. That has an obvious lack of attraction. It would also be probably inconsistent with the limited basis on which this appeal has been run, because it would be likely to re-open wider aspects of the valuation exercise in respect of which permission to appeal has not been sought.

To remedy the above shortcomings on this appeal it is not possible to re-invoke the idea of deducting group net indebtedness as a whole, as Ms Hindson did, because that indebtedness included the debts of loss-making subsidiaries which have to be taken out of the picture because of the refusal of permission to appeal on that point. We are not in a position to survey all other debts in the profit-making companies. However, there are two other debts which have to be considered as candidates, and in fact Mr Griffiths confined his submissions to bringing in those two debts. They are the two debts which were the express subjects of the judge’s original findings – see paragraphs 6, 7 and 8 above. It seems to us that if significant debts, which a purchaser would expect to have to arrange the repayment of, and which have a strong historic element to them so that they are no longer the sort of normal day to day trading debts of the company which would normally be ignored on an earnings-multiplier calculation, have to be deducted, then those two debts should fall into that category. We so determine.

We appreciate that this has an element of the “do it yourself” about it, which may be of questionable worth in a valuation exercise, but, to be frank, that is partly what seems to have happened in final submissions before the judge. We think that doing it on this appeal is an appropriately proportionate response to the problem with which we are faced, and the result is fair and in accordance with the evidence.

The profit of the company, reflected in its accounts, was calculated after bank loan interest was deducted in the sums of £131,002 for 2006, £193,725 for 2007 and £245,568 for 2008. In his first quantum judgment the judge found that it would be appropriate to take bank interest out of the expenditure figures, with the result that the net maintainable earnings figure would be greater. He did so because he apparently took the view that the bank debt was not a borrowing that was necessary to fund the conduct of the business; it was some form of historic debt which a purchaser of the company would be likely to eliminate, and thus eliminate the interest burden. As to this, he said:

On this appeal Mr Griffiths contended that that determination was wrong. The basic method of valuation was by reference to maintainable earnings, to which the deduction of the principal amount of a debt due is irrelevant. There is a measure of earnings which does leave interest out of account (EBITDA – Earnings before Interest Tax, Depreciation and Amortisation), but that was not the measure taken by the experts and, if it had been, a different multiplier would have been applicable to it. Furthermore, as at the valuation date (and therefore as at the date at which a notional purchaser would be assessing the value of the company) the liability to the bank existed, as did the ongoing interest costs. The company did not have the funds to repay the loan. This was the company that the hypothetical buyer was buying. If that notional purchaser decided to “clean up” the bank debt then there would still be a cost in relation to that which would have to be funded; but that is nothing to do with the vendor.

Mr Butler for the respondent in this appeal says that the judgment is correct. He draws attention to the fact that the judge found that the debt was not necessary to support the company’s day-to-day business and arose only because of historical indebtedness. There is no appeal from that finding in relation to the nature of the indebtedness, and it was a finding which the judge was entitled to reach. Mr Thompson had opined that, because each subsidiary had been incurring losses with the exception of one of them, it would appear that bank loans incurred by Sunrise Radio had been used to fund the substantial losses in subsidiaries or to finance the cost of acquiring a substantial property held by Hayes Gate House Ltd (one of the subsidiaries).

In our view the answer to this point lies in pursuing the logic of the judge’s hybrid approach to valuation which, as such, is not the subject of this appeal. The bank debt was, however one looks at it, a cost to the business. A purchaser would, on any footing, pay less for the business than he would have paid if it were not there. The interest was a cost on the profit and loss side; and the amount of the debt was a liability on the balance sheet side. If the basis of valuation had been a pure earnings/multiplier valuation then its adverse effect would have been reflected by leaving the interest in the expenditure side, with the effect of depressing the earnings (to which the multiplier is then applied). If it had been on an assets basis, then the amount of the liability would have been taken out in arriving at the net value. Either way, a purchaser would take it into account in determining what he is going to pay for the business.

The judge’s hybrid approach takes out the liability as such. This would be on the hypothesis that the putative purchaser would pay less for the business because it was there. But if one is going to do that one has to consider what one is going to deduct it from – the earnings/multiplier figure. If that latter figure has the bank interest reflected in the net earnings then the resulting figure will be lower than if it is omitted. The burden of the debt is already reflected at this stage in the calculation. It would be double counting to allow the purchaser a further deduction from the price to reflect the amount of the liability itself. He has already received a benefit because the multiplier calculation has yielded a lower result. No vendor would allow the purchaser to benefit again from the existence of the liability. One reflects the liability either by allowing the interest burden to reduce the multiplier-based figure, or by allowing a deduction from the price of the amount of the liability itself. In this latter event double counting would be avoided by removing the interest burden from the multiplier-based calculation.

The judge seems to have adopted the latter technique. It follows, logically, that he is right to remove the bank interest from the earnings calculation, thus swelling the multiplier calculation itself. On this footing the judge was right.

However, one still has to consider the multiplier point raised by Mr Griffiths. He took the point that the multiplier which was applied was one which assumed interest was built into the sum to which it was applied (it was not an EBITDA multiplier). If he is right about that then one would have to consider recalculating the multiplier calculation using a different multiplier.

That is a little troubling for this court, because we have not seen any material in the court below which demonstrates that the point was considered or debated, much less what the right answer might be. However, we think that ultimately there is nothing in the point. The multiplier used by the experts was derived from some (limited) comparables in which p/e ratios were deduced from comparable businesses. Comparable businesses presumably have similar types of income and expenditure. They might or might not be supposed to have some sort of interest burden, but in the present case the bank debt had been found by the judge to be not one referable to the day to day business of the company. In those circumstances it might be thought to be appropriate to adopt the technique of leaving the interest out of the expenditure side of the multiplier calculation (swelling the resulting figure) because one is then leaving out of account a sum which is atypical of the general run of such businesses and enhancing comparability with the comparables. One is then justified in taking the capital sum out as a liability to be deducted from the multiplier-based figure. One avoids double counting in that way. There is nothing in the evidence which suggests that the multiplier would be any different if that were done.

Accordingly this ground of appeal fails.

The income figures for Sunrise Radio from which the judge was working in order to produce maintainable net earnings included income from subsidiaries paid by way of management charges. In his first judgment the judge below said nothing about these charges, with the effect that they remained within the income figures and thus swelled the maintainable income. He revisited the point in his third quantum judgment when, according to him, he was asked to clarify the point. He gave the following explanation, in support of a conclusion that strictly speaking the income ought to come out, but it would be matched by an equivalent cost deduction, leaving the overall net figure the same:

Mr Griffiths submitted that the judge was wrong in holding that there was no evidence that there would be no costs saving once Sunrise Radio was not responsible for managing the subsidiaries which it is deemed would have disappeared on a purchase. He relies on the very paragraph which the judge himself cites as being evidence to that effect, albeit hearsay. Ms Hindson was entitled to gather material for the purposes of the report, and to rely on whatever she thought was appropriate. That being the case, and since the judge had himself acknowledged that it would be logical to deduct the management charges collected by Sunrise Radio, the position on the evidence was that the charges should have been deducted (as Ms Hindson said in the paragraph preceding the one cited) and there should be no counter-figure to reflect cost saving. This would have the effect of reducing net maintainable earnings figure.

The essence of the respondents’ case on this point is that the point in issue is an evidential one on which the appellant did not adduce any, or any sufficient, evidence. There was no factual evidence to support the assertion about directors’ remuneration made by Ms Hindson on instructions, and no evidence that there would be no other cost savings.

The trial judge recognised that logically the management charges fell to be taken out of the maintainable earnings (paragraph 8) and that his calculation had not done so (paragraph 5). In his further consideration of the point he justified the numbers by finding that there would be a commensurate reduction in cost. This part of the appeal therefore really has to focus on whether he was justified in reaching that latter decision.

A review of the transcript (to which we were not referred in the appeal) reveals that there was limited cross-examination of Ms Hindson on the point by Mr Hanke - just a page and a half (pages 62-3). He pointed out to her the weakness of the material on which she relied (a statement by a Mr Dattani) and she acknowledged that one might deduct costs “if one could identify a few”. She also expressed her experience as being that management charges were:

There was no re-examination on the point. Mr Thompson was cross-examined briefly on the charges (20th July, pages 40-41), but no factual case was put to him as to the whether there would be any saving of management time.

The judge’s finding is a factual one. He found that management effort spent in providing services for the subsidiaries would be saved or used profitably elsewhere. There was really virtually no positive evidence on the point (or none that we have been directed to) so the question becomes whether that was a decision that was open to the judge. With some hesitation, we find that it was. In order to justify not adding back in the costs of providing services the case of the respondents to the petition must have been that, despite the service charges (which seem to have been substantial), no cost was incurred in providing those services, and no future expenditure would be saved by not providing them. That requires evidence, and the only evidence was the single sentence about directors appearing in Ms Hindson’s report and, perhaps, her expression of her general experience of such matters (which is not strong in these circumstances). It is true that the petitioner did not seek to adduce any evidence of her own on the point, but that is less surprising (and less significant) in the circumstances than the failure of the respondents to the petition to address it more fully. In the end the judge had to arrive at a decision, and we consider that his decision was one that was open to him.

This ground of appeal therefore fails.

As part of the process of valuing the subsidiaries the judge had to reach a conclusion as to the value of one of them, Club Concorde. He dealt with this in his first quantum judgment at paragraph 28. In doing so he had to deal with consultancy fees payable by Club Concorde in respect of the services of Dr Lit. The effect of the first and second quantum judgments was that he left in the calculation a third of the amounts said to be payable to Dr Lit (£60,000 as opposed to £180,000) as a deduction in arriving at net earnings.

In the first quantum judgment he said:

In the second quantum judgment he added the following:

Mr Griffiths’ point about this is one of inconsistency. When considering the position of Sunrise the judge had provided for all consultancy fees of Dr Lit paid by Sunrise Radio and not reduced them. He submitted that there was no reason for treating the consultancy fees charged to Club Concorde any differently on the evidence that the judge had. This had a significant knock-on effect in relation to the valuation of Club Concorde, increasing it by several hundred thousand pounds (according to Mr Griffiths).

Mr Butler submitted that the judge’s findings were essentially evidence-based and should not be interfered with on appeal.

In relation to Sunrise Radio the judge found that there was no evidence that the amount of the fees would be saved by a purchaser, but that there was evidence to show that in Club Concorde it would not be necessary to incur the full amount of the fees.

There is no point of principle involved in this ground of appeal other than the principle that a finding must be supported by evidence. It seems to us that Mr Butler’s position, which is that the judge reached evidence-based conclusions in relation to both items, is correct. In the case of Sunrise Radio, the judge started with an apparent need for consultancy (management) services which, on the evidence, he found would continue. Since there was no evidence that it would be obtained any more cheaply by a purchaser he left in the full cost of the services as a debit item. So far as Club Concorde is concerned, he found, on the evidence, that the same level of management would not be required after a purchase, and he came to the best conclusion that he could (apparently not having been well served by the experts) as to the reduced level of management. That is a different conclusion to that reached in relation to Sunrise Radio, but that is because the evidence was different. It cannot be said that he was not entitled to reach that view on the evidence.

Accordingly, this ground of appeal fails.

It follows that this appeal succeeds in relation to the external liabilities point, but fails on the other points. The sum payable to the petitioner falls to be adjusted accordingly, and the parties should agree the calculation for the purposes of the order that we make on this appeal.