Goldstein v Levy Gee (a Firm)

In 1971 Mr David Goldstein and Mr Dudley Leigh set up in business as estate agents. In the following year they began to invest and deal in property on their own account, using various companies for that purpose. In 1974 they incorporated Marchday Group plc as a commercial investment and investment company. On 1 August 1988 Mr Goldstein ceased to be a director of the company, although he remained employed by it. Until 23 August 1993 Mr Goldstein, Mr Leigh and their respective families were equal shareholders in the company. On that day either Mr Leigh or Marchday itself acquired some of Mr Goldstein’s and his family’s shares. This left Mr Goldstein with 60,000 shares owned by him personally, and 50,000 shares owned by the trustees of his children’s trust. Mr Goldstein’s personal holding amounted to some 26 per cent of the company’s share capital and the trustees’ shareholding amounted to a further 21 per cent, making about 47 per cent in all.

In July 1994 Mr Leigh told Mr Goldstein that he wanted to restructure the company’s share capital. The proposal was to create a new class of share, and to issue shares of that class to the company’s directors, in order to give them an incentive to achieve capital growth. The directors at that time were Mr Leigh himself, Mr Kleiner, Mrs Quinn, Mr Orchard and Mr Smith. Mr Goldstein objected to the proposal and it was ultimately dropped.

In November 1994 Mr Leigh came up with a new proposal. This was to grant options to the directors to buy shares in the company. The rationale for the proposal was the same, namely to incentivise the directors. Mr Goldstein again objected. His objection was based on his perception that the grant of share options would confer excessive and disproportionate benefits on the directors. If exercised, the issue of new shares would also dilute his (and the trustees of his children’s) shareholdings in the company. Despite Mr Goldstein’s objections, the proposal to grant the share options was carried, by a relatively narrow majority, at an extraordinary general meeting of the company on 18 January 1995. In February 1995 the bulk of the options were granted at an exercise price of £47.46 per share.

Mr Goldstein responded by presenting a petition under section 459 of the Companies Act 1985 seeking, amongst other things, the setting aside of the options.

On 19 June 1997 the company terminated Mr Goldstein’s employment with effect from 19 September 1997, on the ground that he was redundant. Although Mr Goldstein claimed that he was unfairly dismissed, an Employment Tribunal rejected his claim on 24 April 1998.

On 4 September 1997 Mr David Epstein, a partner in Levy Gee, prepared a valuation of the entire share capital of the company. No one could remember why that valuation was commissioned. The valuation was based on the company’s audited accounts for the year ended 30 September 1996, with certain adjustments to take account of events subsequent to the balance sheet date. The basis of valuation that Mr Epstein adopted was the net asset value of the company. In calculating the value of the company’s assets, Mr Epstein took as his starting point the value of the properties as shown in the accounts. He then made adjustments to take account of increases in value of three of the properties. The properties were valued at the aggregate of their individual values. Mr Epstein then considered whether shares in a property company traded at a discount to net assets. He concluded that they did, and considered that a discount of 20 per cent was appropriate. He then considered the significance of the fact that the company’s shares were not listed on a stock exchange. He commented that an unlisted company put up for sale as a complete entity is usually more difficult to sell due to greater uncertainties relating to such companies. He concluded that a further discount of 12.5 per cent was appropriate. This gave him a value per share of £44.23.

The termination of Mr Goldstein’s employment triggered the provisions of Article 14 of the company’s articles of association. Under Article 14 (d) (i) Mr Goldstein was deemed to have given the directors a transfer notice on 19 September 1997. The effect of the deeming provision was that under Article 14 (f) the company, the directors and the vendor were to try to agree “the Prescribed Price”. If they did not do so, the company was to refer the determination of the Prescribed Price to “the Valuers”.

“The Prescribed Price” was defined as:

“the price certified … by the Valuers acting as experts and not as arbitrators as being equal to the value of the shares offered valued without discount for minority holding as between a willing vendor and a willing purchaser as at the date of the Transfer Notice. The Valuers shall be entitled to engage surveyors to advise on the value of the Company.”

“The Valuers” were defined as:

“the Company’s Auditors from time to time”

At the relevant time the company’s auditors were Levy Gee. In August 1997 Mr Leigh informed Mr Julian Synett, a partner in that firm, that he would almost certainly be required to carry out a valuation under the deemed transfer provisions of article 14. However, he was not formally instructed until 20 May 1998. Mr Synett was the audit partner at Levy Gee responsible for auditing Marchday’s accounts and was also the responsible client partner.

Mr Synett said that he understood the general principles of valuing companies from his general experience. However, he had not undertaken specific valuations of shares in unlisted property companies before. He did not take any specific steps to acquaint himself with the principles or the procedure appropriate for such a valuation, although he did not think that his valuation was affected by that. Mr Synett was firm in his view that because he was instructed by the company, his client was the company and that his primary duty of care was also to the company.

In his letter to Mr Synett of 27 August 1997, Mr Leigh set out some assumptions on which he assumed the valuers would be asked to value. The second of them was:

“the properties are to be valued at the deemed transfer date for sale as an entire portfolio”

Mr Synett annotated that part of the letter: “disposed in the ordinary course of business – which may be sep. [i.e. separate] or as a whole”. It seems therefore that Mr Synett’s first reaction was to disagree with Mr Leigh’s assumption that the properties were to be valued on the basis of a single transaction. Mr Synett said that he appreciated the difference between a valuation of an entire portfolio on the basis of a single transaction and a valuation on the basis of individual sales. It was put to him that the former would be likely to produce a lower value than the latter. He said that the converse might equally well be true, in that sale of a collection of properties might be worth more than the sum of their individual parts.

There followed correspondence between Mr Leigh and Mr Synett over the terms of the instructions to be given to property valuers in order to value the company’s property assets.

Mr Synett did not keep any detailed notes or working papers. He accepted, in cross-examination, that this was regrettable. Indeed, he went so far as to accept that this was a lapse from acceptable professional conduct. However, no loss arises from this lapse, if lapse it was. But it undoubtedly made Mr Synett’s ability to remember (as opposed to reconstruct) his thought processes more difficult.

Mr Synett’s first draft of the instructions to the valuers was produced in late September 1997. It said that:

“The properties are to be valued individually on an “open market” basis in accordance with the current RICS Appraisal and Valuation Manual (The Red Book).”

Mr Synett sent a copy of this draft to Mr Leigh, at the latter’s request. A manuscript note on the fax cover sheet (in Mr Synett’s writing) reads “include requirement to value as whole”. Mr Synett could not remember how that note came to be made. I infer that it was made during the course of or following a conversation between him and Mr Leigh. In his letter of 29 September 1997 Mr Synett refers to a meeting between himself and Mr Leigh which took place during the preceding week. The conversation could well have taken place at that meeting.

Following that meeting, Mr Synett revised the instructions to the valuers. This draft, the second, contained two instructions. The first was an instruction to value the properties “individually on an open market basis”. The second was a requirement that, having arrived at individual values, the valuer should “indicate what adjustment would be appropriate to the total valuation of the portfolio in order to reflect the sale of the portfolio as a single transaction.” Mr Synett sent a copy of this draft to Mr Leigh. On receipt of the draft Mr Leigh commented:

“We seem to be agreed that the valuation has to be done on a portfolio basis.”

Mr Synett revised the instructions again. The third draft said:

“The properties are to be valued, as a portfolio, on an open market basis (including both positive and negative values) based on your opinion of the best price at which the sale of the portfolio could be completed unconditionally …”

Mr Synett explained in evidence that the second instruction from the second draft had been removed because it had been subsumed in the instruction to value the properties as a portfolio. He amplified this explanation by saying that since the hypothetical buyer of the shares was going to buy the whole of the issued share capital, he was in effect going to buy the whole property portfolio. Hence it was appropriate to value all the properties on the basis that they would be acquired in a single transaction.

In October 1997 Mr Synett invited three firms of valuers to tender for the valuation on the basis of instructions that were included with the invitation to tender. The instructions began:

“The properties are to be valued, as a portfolio, on an open market basis (including both positive and negative values) based on your opinion of the best price at which the sale of the portfolio could be completed unconditionally, for a cash consideration on the date of the valuation” [on various assumptions]

Among the assumptions that the valuer was required to make was that:

“It is to be assumed that the portfolio will be sold on a standard contract requiring completion within twenty-eight days of exchange.”

The successful tenderer was Jones Lang Wootton. In a letter of 24 February 1998 Mr Rod Neal of that firm said:

“I note that the properties are to be valued as a portfolio.”

Jones Lang Wootton produced their valuation on 17 March 1998. It was carried out by Mr Duncan Preston FRICS. It said:

“As instructed, the properties have been valued as a portfolio, on an open market basis …

We have had regard to the general quality of the portfolio, the potential for rental growth, for the redevelopment opportunities and for future liabilities. We have concluded that the portfolio would need to be offered at a discount to the aggregate value of the component parts in order to achieve a sale on a standard contract requiring completion within 28 days of exchange. …

In arriving at the portfolio valuation, we have adopted an overall discount of 10% to reflect its particular characteristics.”

The value that Jones Lang Wootton reported was £26 million. They then set out individual values of different categories of property, divided between trading stock on the one hand and investment properties on the other. In each case the individual property values were given before “portfolio discount”. The individual values amounted, in aggregate, to £29.1 million. In addition there were two leases which had negative values amounting to £235,000.

In early May 1998 Mr Synett began work on the valuation of shares in the company. He produced a first draft on 7 May 1998. Mr Synett’s starting point was the company’s net assets as recorded in its balance sheet. He stripped out the values of the trading and investment properties (which had been directors’ valuations) and substituted Jones Lang Wootton’s figures for the properties (after the application of “portfolio discount”), and added some short leasehold properties, which had not been in the balance sheet, but which had negative values. He made a deduction for the tax that would be payable on a disposal by the company on those properties that it held as trading stock and in a joint venture. He made a further deduction for contingent tax payable on the disposal of those properties held by the company as investments. The contingent tax would be payable on the difference between (a) the value of investment properties as shown in the accounts and (b) the values attributed to those properties by Jones Lang Wootton. He deducted the full amount of that tax. He made an end deduction of 12.5 per cent for “non listed status/costs of realisation etc.” The result of those calculations was a value for each share of £58.95.

Mr Synett produced a second draft on 11 May 1998. That followed the calculations of the first draft down to the calculation of the adjusted net assets of the company. But in the second draft Mr Synett made two further adjustments. The first was to add back the 12.5 per cent discount for the non-listed status of the company. However, this discount was not carried through into the final calculation in this draft. The second was based on the existence of the share options. He assumed that all the share options would be exercised, with the result that the shareholding in the company would increase from 232,550 shares to 824,483 shares and that the assets of the company would increase by the aggregate option price of £27,829,000. This produced a value for each share of £52.76. A note at the end of this draft says “what about 12.5% discount?”

Mr Synett said that in dealing with the share options he had familiarised himself with the terms of the option scheme. He could not remember whether he satisfied himself that the option holders were entitled to exercise their options. However, the options did not expire until 2005, and he said that he was aware of the company’s strategy to grow the business with a view to a sale within that timescale. But he did not make specific inquiries about the likelihood of the options being exercised. Rather, he approached this question from the perspective of a potential buyer of the shares.

On 24 May 1998 Mr Synett produced a third draft. This draft kept in the 12.5 per cent discount for the non-listed status of the company (applied only to the existing shareholding) and also assumed that the share options would be exercised. The result of this draft was to attribute a value to each share of £50.38. This time his treatment of the options was different. He first calculated the adjusted net assets of the company as in the first draft. He then calculated the value of the options. He assumed that 75 per cent of the options would be exercised. He deducted the value of 75 per cent of the options from the adjusted net assets of the company and then divided the result by the number of existing shares. This gave a value per share of £52.52, and a total value of Mr Goldstein’s shareholding of £3,151,496.

On 26 May 1998 Mr Synett sent a copy of his workings and a draft valuation to Mr David Epstein (one of his partners) for comment. But Mr Epstein was away on holiday.

On 29 May 1998 Mr Synett issued his valuation. It certified that the value of Mr Goldstein’s shareholding of 60,000 shares was £3,151,496. It did not set out any of the workings that had been used to arrive at that figure.

Mr Goldstein was very disappointed with that figure. He wrote to Mr Synett asking for answers to a number of questions, but Mr Synett declined to answer.

On 10 September 1998 Mr Goldstein sold his shares at the price fixed by Mr Synett, namely £52.52 per share.

Mr Goldstein now alleges that Mr Synett’s valuation was negligently made. There were four principal complaints:

In support of these complaints Mr Goldstein adduced the evidence of Mr Tony Hindley. Levy Gee’s valuation was defended by Mr Emile Woolf. Mr Hindley’s valuation, taking account of a 25 per cent probability that the share options would be exercised was £75.27 per share, leading to a loss to Mr Goldstein of £1,364,704.

In the course of the evidence the fourth complaint, relating to the treatment of the share options disappeared, and Mr Gatt QC, appearing for Mr Goldstein, did not press it in his closing submissions. I shall say little more about it. By my calculations, a reworking of Mr Hindley’s valuation, allowing for a 75 per cent probability that the share options would be exercised, produces a value per share of £62.96. The value of Mr Goldstein’s shareholding of 60,000 would therefore have been £3,777,600, giving rise to an alleged loss of £626,104. Using the agreed spreadsheet, which differs from Mr. Hindley’s original method, would produce a value per share of £58.95 and an alleged loss of £385.874.

Mr Howe, appearing for Levy Gee, made two preliminary points. First, he submitted that even if all the criticisms of Levy Gee’s valuations were well-founded there could be no loss, and hence no liability, unless the final figure fell outside the range of acceptable valuations. A valuer is engaged to produce an estimation of the value of an asset. All that matters is the final figure at which he arrives. If that figure is one which a reasonably competent and careful valuer could have reached, then there can be no negligence, even if the reasoning process by which the figure was reached is manifestly incompetent, or even non-existent. As Mr Howe put it in argument: a valuation which is within the bracket is by definition right. Every valuation which falls within a reasonable bracket is as valid as every other valuation within the bracket. They are all right valuations.

Mr Gatt QC submitted to the contrary. He said that the essence of the tort of negligence, at least as applied to professional persons, was not concerned with outcomes, but with processes or practices. If a professional makes an incompetent error in the course of his engagement, and that error can be shown to have caused a loss, then that error is negligent. Thus if a valuer makes an incompetent error in the course of his reasoning, and that error leads to a difference in the final figure, then the difference can amount to a recoverable loss, even if the final figure could have been reached by a competent and careful valuer using a different reasoning process.

I was referred to a number of authorities. I find it difficult to reconcile them all, and to extract a coherent and principled approach to the question of valuer’s negligence. It may be that the divergent streams of authority will have to be resolved by a higher court. But that is not a task for me.

In Zubaida v. Hargreaves [1995] 1 EGLR 127 Hoffmann LJ said:

“In an action for negligence against an expert, it is not enough to show that another expert would have given a different answer. Valuation is not an exact science; it involves questions of judgment on which experts may differ without forfeiting their claim to professional competence. The fact that a judge may think one approach better than another is therefore irrelevant… The issue is not whether the expert’s valuation was right, in the sense of being the figure which a judge after hearing the evidence would determine. It is whether he has acted in accordance with practices which are regarded as acceptable by a respectable body of opinion in his profession: see Bolam v. Friern Hospital Management Committee [1957] 1 W.L.R. 582 at p. 587, a well-known citation”

This is in line with the general principle that a professional does not warrant a result. He agrees only to use reasonable skill and care in forming his opinion or giving his advice. In other words it is the process that must be examined rather than simply the end result. That is not to say that the end result is irrelevant. First, the end result may be very far from opinions given by other experts, or may be falsified by some empirical outcome, with the result that one must infer that something has gone wrong with the process. Second, a professional may have a duty, as part of the process, to stand back and look at the end result in the round to see if it accords with his instinctive feel.

In Lion Nathan Ltd v. C-C Bottlers Ltd [1996] 1 W.L.R. 1438 Lord Hoffmann, giving the advice of the Privy Council, said:

“As has been said, a forecast is always the forecaster's estimate of the most probable outcome, the mean figure within the range of foreseeable deviation. The judge appears to have assumed that if a figure would have been within the range of foreseeable deviation from the mean of a properly prepared forecast, it must follow that it would have been proper to put that figure forward as the mean. This proposition has only to be stated to be seen to be fallacious. There is no connection between the range of foreseeable deviation in a given forecast and the question of whether the forecast was properly prepared. Whether a forecast was negligent or not depends upon whether reasonable care was taken in preparing it. It is impossible to say in the abstract that a forecast of a given figure "would not have been negligent." It might have been or it might not have been, depending upon how it was done. Assume, for example, that the vendor had forecast $1.25m. and that the limits of foreseeable deviation would have been regarded as $50,000 either way. Assume that the forecast was unexceptionable in every respect but one: there had been a careless double counting of sales which, if noticed, would have reduced the estimate by $25,000. To that extent, the estimate has not been made with reasonable care. If on account of some compensating deviation the outcome is $1.25m. or more, the purchaser will have suffered no loss and the vendor will incur no liability. But if the outcome is less than $1.25m., their Lordships think that the purchaser is entitled to say that if the estimate had been made with reasonable care, the figure put forward by the vendor as the mean and upon which he relied in fixing the price, would have been $25,000 lower. To this extent, he has suffered loss by reason of the breach of warranty. It is nothing to the point that the outcome is still within what would have been predicted as the limits of foreseeable deviation. His complaint is that the whole range of possible outcomes would have been stated as $25,000 lower. The purchaser has accepted the risk of any deviation attributable to factors which were unforeseeable, unknown or incalculable at the time of the forecast. He has accepted the risk of such deviation whether its true extent would have been foreseeable at the time of the forecast or not. But he has not accepted the risk of any deviation which is attributable to lack of proper care in the preparation of the forecast. The only tolerable forecast is one which, on its facts, was prepared with reasonable care.” (Underlining added)

This passage, as it seems to me also focuses on the process of preparing the valuation or forecast, rather than the end result. Lord Hoffmann says explicitly that there is no connection between the range of foreseeable deviation in a forecast and the question whether it was properly prepared. He also says explicitly that whether a forecast was negligent or not depends upon whether reasonable care was taken in preparing it. Moreover, his example clearly shows that a forecaster (or valuer) may be negligent even though his final figure falls within the limits of foreseeable deviation (i.e. within the “margin of error”). Accepting, as one must, that in valuation cases many questions are matters of judgment, the client is in effect saying to his valuer: “I will accept your judgment on all the questions that will arise in the course of preparing your valuation, provided that you exercise your judgment with reasonable skill and care”. Although a decision of the Privy Council is only persuasive authority, the reasoning is, in my view, very persuasive indeed.

In South Australia Asset Management Corporation v. York Montagu [1997] A.C. 191 (often referred to as SAAMCO or Banque Bruxelles Lambert) the House of Lords considered the measure of damages for negligent valuation. However Lord Hoffmann also considered the basis on which a valuer might be liable at all. He said:

“Before I come to the facts of the individual cases, I must notice an argument advanced by the defendants concerning the calculation of damages. They say that the damage falling within the scope of the duty should not be the loss which flows from the valuation having been in excess of the true value but should be limited to the excess over the highest valuation which would not have been negligent. This seems to me to confuse the standard of care with the question of the damage which falls within the scope of the duty. The valuer is not liable unless he is negligent. In deciding whether or not he has been negligent, the court must bear in mind that valuation is seldom an exact science and that within a band of figures valuers may differ without one of them being negligent. But once the valuer has been found to have been negligent, the loss for which he is responsible is that which has been caused by the valuation being wrong. For this purpose the court must form a view as to what a correct valuation would have been. This means the figure which it considers most likely that a reasonable valuer, using the information available at the relevant date, would have put forward as the amount which the property was most likely to fetch if sold upon the open market. While it is true that there would have been a range of figures which the reasonable valuer might have put forward, the figure most likely to have been put forward would have been the mean figure of that range. There is no basis for calculating damages upon the basis that it would have been a figure at one or other extreme of the range. Either of these would have been less likely than the mean: see Lion Nathan Ltd. v. C. C. Bottlers Ltd., The Times, 16 May 1996.”

Although Lord Hoffmann recognises that within a band of figures valuers may differ without one of them being negligent, it does not seem to me that he was saying that the only way to establish negligence or breach of contract is to prove that the valuer’s particular figure falls outside that band. If that is what he meant, his reference to the Lion Nathan case would surely have been qualified.

Mr Howe submitted that the method of assessing damages laid down by Lord Hoffmann diverged from the normal rule (at least in contract cases) that the party in breach of contract is assumed to have performed the contract in the manner most favourable to himself. That may be so, but the ruling of the House of Lords is clear. Mr Howe also pointed to one possible difficulty of applying the law in the manner postulated by Lord Hoffmann. Suppose that the limits of foreseeable deviation are between £9 million and £11 million. The valuer arrives at a figure of £9.1 million. However, it can be demonstrated that through an obvious and careless error his final figure was lower by £100,000 than it would have been without the error. If that error amounts to negligence, the measure of damages, according to SAAMCO, is the difference between the valuer’s final figure and the mean. Thus the valuer would be liable for £900,000 (i.e. the difference between his actual figure and the mean), even though the effect of his carelessness is only to reduce his figure by £100,000. I agree that this is not a comfortable conclusion.

However, there is a line of authority which focuses on the final figure, rather than the process by which the valuer reached the final figure. That line of authority begins with Singer & Friedlander v. John D Wood & Co [1977] 2 EGLR 84. In that case Watkins J said:

“The valuation of land by trained, competent and careful professional men is a task which rarely, if ever, admits of precise conclusion. Often beyond certain well-founded facts so many imponderables confront the valuer that he is obliged to proceed on the basis of assumptions. Therefore he cannot be faulted for achieving a result which does not admit of some degree of error. Thus, two able and experienced men, each confronted with the same task, might come to different conclusions without anyone being justified in saying that either of them has lacked competence and reasonable care, still less integrity, in doing his work. The permissible margin of error is said by Mr Dean, and agreed by Mr Ross, to be generally 10 per cent either side of a figure which can be said to be the right figure, i.e. so I am informed, not a figure which later, with hindsight, proves to be right, but which at the time of valuation is the figure which a competent, careful and experienced valuer arrives at after making all the necessary inquiries and paying proper regard to the then state of the market. In exceptional circumstances the permissible margin, they say, could be extended to about 15 per cent, or a little more, either way. Any valuation falling outside what I shall call the “bracket” brings into question the competence of the valuer and the sort of care he gave to the task of valuation.”

This is, I think, the first mention, in a reported case, of the “margin of error”. Watkins J, at least in this passage, treats a valuation of property which falls outside the margin of error merely as evidence of lack of care and competence. In the course of his lengthy and careful judgment he examined in detail the process by which the valuer in that case came to his final figure. The deficiencies he identified were deficiencies in the process of valuation, rather than simply a disparity between the final figure and what he called “the right figure”. However, in a later passage in his judgment he said:

“Pinpoint accuracy in the result is not, therefore, to be expected by he who requests the valuation. There is, as I have said a permissible margin of error, the “bracket” as I have called it. What can properly be expected from a competent valuer using reasonable skill and care is that his valuation falls within this bracket.”

This passage does, in my opinion, concentrate on the final figure rather than the process. Moreover, it uses the margin of error in another way, namely to provide the valuer with a defence to a claim of negligence if his figure falls within the bracket, no matter how he arrived at his figure.

In Mount Banking Corporation Ltd v. Brian Cooper & Co [1992] 2 EGLR 142 the plaintiff submitted that where the final valuation figure is within the Bolam principle, an acceptable figure, albeit towards the top end, but where none the less the valuer has erred materially in reaching that figure, the plaintiff can succeed in his claim because of those negligent errors, even though the total valuation figure was not negligent. Mr Robin Stewart QC, sitting as a deputy judge of the Queen’s Bench Division, rejected that submission. He said:

“If the valuation that has been reached cannot be impugned as a total, then, however, erroneous the method or its application by which the valuation has been reached, no loss has been sustained, because, within the Bolam principle, it was a proper valuation.”

Plainly this passage focuses on the end result rather than the process by which the valuer reached the end result. In reaching his conclusion on the facts, Mr Stewart said:

“I conclude, therefore, on this section, that though there was a fault in the process of calculation, none the less a proper and acceptable process could properly have resulted in no, or no perceptible, difference to the end valuation; that is to say that the figure in fact reached by Mr Cohen was acceptable on the Bolam principle.”

In Craneheath Securities v. York Montague Ltd [1996] 1 EGLR 130 at 132 Balcombe LJ (with whom Otton and Aldous LJJ agreed) said:

“Since Craneheath did not establish that the figure of £5.25m was wrong, then I agree with Mr Stow that Craneheath’s action must fail. It would not be enough for Craneheath to show that there have been errors at some stage of the valuation unless they can also show that the final valuation was wrong. If authority be needed for so self-evident a proposition, it can be found in Mount Banking Corporation Ltd v. Brian Cooper & Co [1992] 2 EGLR 142 at pp. 144-5, 149.”

These are the passages from Mr Stewart QC’s judgment that I have just quoted, and in my view are direct approval of his approach by the Court of Appeal. In the same case Otton LJ said:

“In the light of this the plaintiffs faced a formidable task in discharging their burden of proving that the figure of £1.1 m as an assessment of current turnover was erroneous. Without such a finding there could be no finding of negligence.”

It has been argued that it is by no means clear whether Balcombe LJ was using the word “wrong” to denote a figure which fell outside a “margin of error” or whether all that he meant was that if the valuation was in fact correct, it could not be said to be negligent (See Simpson: Professional Negligence and Liability para 8.152). Likewise Otton LJ’s use of the word “erroneous” is open to both interpretations. In the context of the court’s approval of Mr Stewart QC’s judgment, it seems to me to be probable that both expressions were used in the former sense.

In Legal & General Mortgage Services Ltd v. HPC Professional Services [1997] P.N.L.R. 567 the claimant submitted that he was entitled to succeed either by showing that the valuer’s final figure was outside the bracket within which any competent valuer using reasonable skill and care could have valued the property (“the result route”), or that by failing to exercise such skill and care he valued the property at an incorrect figure, albeit a figure within the appropriate bracket (“the method route”). Judge Langan QC rejected that submission. He was referred to the passages in Lion Nathan and SAAMCO to which I have referred and said:

“Both cases were concerned with the quantification of damages and not with liability. Lion Nathan was concerned with breach of a warranty given on the sale of a business, and I feel quite unable to transpose the passages cited by counsel …analogically to the very different area of professional negligence.”

Since Lord Hoffmann in Lion Nathan began by discussing the general nature of a forecast, I am unable to discern the judge’s precise reasons for rejecting the analogy (if that is what it was). It is worth noting that later in his judgment the judge said:

“This is a bracket case. It was so approached by both surveyors. There will be other cases, most probably involving the valuation of premises on which an on-going business is conducted, where the bracket approach will be wholly inappropriate. None of what I have said so far in this judgment is intended to bear upon such cases.”

It is not clear what alternative the judge had in mind for such cases, unless it was the “method route”.

I come now to Merivale Moore plc v. Strutt & Parker [1999] 2 EGLR 171. I find this a difficult case. This was a case in which the valuer was instructed to prepare an appraisal of a proposed purchase. The property was to be acquired for development. The valuer prepared an appraisal which attempted to value the completed investment, in order for the purchaser to decide whether a purchase at the asking price would be a sensible transaction. In order to prepare the valuation, the valuer had to estimate the cost of the development, the rent at which the completed development could be let, and the yield to be applied to that rent in order to arrive at a capital value of the completed development. The task was made more difficult by the fact that the interest on offer was a lease with an unexpired term of 46 years. The valuer took as his starting point a rent for the principal areas of £60 per square foot, and adopted a yield of 7.5 per cent. The trial judge found that taking a rent of £60 per square foot was negligent. He also found that although a yield of 7.5 per cent was “too low” it was not, in itself, negligent. He made an express finding to that effect. However, he found that the yield should have been qualified by a warning about its reliability. The failure to append a qualification was negligent. The valuer appealed to the Court of Appeal. That court, by a majority, reversed the judge’s finding of negligence on the question of the rental value. However, the valuer’s appeal was dismissed, on the ground that the judge’s finding that the yield should have been qualified was one which was open to him. The result of the appeal, therefore, was that although the valuer had adopted a rental value which was not negligent, and a yield which was, in itself, not negligent, he was still liable in negligence because of the failure to qualify the yield with a warning. Simply looking at the result, one might have thought that this was a case in which the process, rather than the end result in figures, was all-important. However, examination of the judgment of Buxton LJ shows that this may not be so.

Buxton LJ began by explaining the structure of the trial judge’s inquiry. He said at page 173:

“In order to determine whether the advice contained in the 12 June assessment was negligent, that is to say, whether the figures set out in the assessment were negligently stated, it was necessary, or if not necessary almost inevitable, that the court should form a view as to what was the correct or true value of the property; that is what would have been the correct figures to include in the assessment. That step has to be taken because a necessary step in determining whether a particular valuation was negligent is to consider the extent to which the valuation diverged from what would have been a correct valuation, and the reasons for that divergence.”

This passage suggests, first, that there is a “correct or true value” of a property, rather than simply a “bracket” and, second, that the court must decide that true value before embarking on the question whether the impugned valuation was negligent. In deciding that question the court must consider both the extent of the divergence from the true value, and the reasons for the divergence. Absent empirical proof of the “true value” of a property on a given day (e.g. an open market sale of that property on that day), the “true” value must mean the figure at which the court values the property, having heard expert evidence. However, if the acid test of liability is whether the impugned end result falls outside a bracket, it is not clear why the court must decide what the “true” value of the property was, rather than simply deciding the bracket. Buxton LJ then reviewed the evidence given to the trial judge. Before coming to his own conclusions, he set out “some indication of the guidance to be obtained from recent authority as to the correct approach to a complaint of negligence against a valuer.”

At 176 Buxton LJ said:

“It has frequently been observed that the process of valuation does not admit of precise conclusions, and thus that the conclusions of competent and careful valuers may differ, perhaps by a substantial margin, without one of them being negligent: see for instance the often quoted judgment of Watkins J in Singer & Friedlander Ltd v. John D Wood & Co [1977] 2 EGLR 84 at p. 85G; and the House of Lords in Banque Bruxelles Lambert S.A. v. Eagle Star Insurance Co Ltd [1977] A.C. 191 at p 221 F-G. That has led to the courts adopting a particular approach to claims of negligence on the part of valuers.

In the general run of actions for negligence against professional men:

“it is not enough to show that another expert would have given a different answer. …the issue is … whether [the defendant] has acted in accordance with practices which are regarded as acceptable by a respectable body of opinion in his profession”: Zubaida v. Hargreaves [1995] 1 EGLR 127 at p 128 A-B per Hoffmann LJ, citing the very well-known passage in Bolam v. Friern Hospital Management Committee [1957] 1 WLR 582 at p 587

However, where the complaint relates to the figures included in a valuation, there is an earlier stage that the court must be taken through before the need arises to address considerations of the Bolam type. Because the valuer cannot be faulted in any event for achieving a result that does not admit of some degree of error, the first question is whether the valuation, as a figure, falls outside the range permitted to a non-negligent valuer. As Watkin J put it in Singer & Friedlander at p 86A:

“There is, as I have said, a permissible margin of error, the “bracket” as I have called it. What can properly be expected from a competent valuer using reasonable skill and care is that his valuation falls within the bracket.”

A valuation that falls outside the permissible margin of error calls into question the valuer’s competence and the care with which he carried out his task. .. But not only if, but only if, the valuation falls outside the permissible margin does that inquiry arise. …

Various further considerations follow. First, the “bracket” is not to be determined in a mechanistic way, divorced from the facts of the instant case. … Second, if it is shown even at the first stage that the valuer did adopt an unprofessional practice or approach, then that may be taken into account in considering whether his valuation contained an unacceptable degree of error. … Third, where the valuation is shown to be outside the acceptable limit, that may be a strong indication that negligence has in fact occurred. … Some caution at least has to be exercised in this respect, because the question must remain, in valuation as in any other professional negligence cases, whether the defendant has fallen foul of the Bolam principle. To find that his valuation fell outside the “bracket” is, as is held by this court in Craneheath and also, I consider, by the House of Lords in Banque Lambert, a necessary condition of liability, but it cannot in itself be sufficient.” (Underlining added)

Buxton LJ contrasts the position in the “general run” of actions against professionals with the particular case of negligent valuers. Yet the passage he quotes to illustrate the test applicable in the “general run” of cases, as contrasted with the “particular approach” in negligent valuation cases, is itself taken from a valuation case. It may be that the “particular approach” in valuation cases is appropriate only where the complaint is a complaint about the “figures included in a valuation” rather than to the method of valuation. This would be consistent with the outcome of the appeal in that case, since the finding of negligence was upheld, not because the yield figure was negligent as a figure, but because it appeared unqualified by a warning. Nevertheless, the complaint in Zubaida was itself a complaint about the figures in the valuation. It is noticeable, moreover, that Lion Nathan was not mentioned in the judgment.

Be that as it may, it is clear that Buxton LJ holds that in cases where the figures are impugned as figures, it is a necessary precondition of liability that the impugned figures fall outside the “bracket”. That being so, it is not easy to see why, at that stage of the inquiry, it matters whether the valuer has adopted an unprofessional practice or approach. Either his final figure falls within the “bracket”, or it does not. If falling outside the “bracket” is a precondition of liability, why should it matter how the valuer arrived at his figure, as long as it is within the “bracket”? There is, I think, another problem. If the “bracket” is the end result, and the end result depends on a number of variables, should the bracket be assessed by arriving at a bracket for each of the variables, or only for those variables that are alleged to have been negligently assessed? Logically, in my opinion, the approach in Merivale Moore leads to the first possibility.

The passage quoted from the judgment of Buxton LJ is, in my view, part of the ratio of the majority of the court. It unequivocally emphasises the decisiveness of the end result, as opposed to the process by which the valuer reached the end result. It also interprets Balcombe LJ’s use of the word “wrong” as meaning “outside the margin of error”, as opposed to “incorrect”. However, it seems to me to be confined to a case where the complaint is about the figures as figures, rather than some other aspect of the valuation.

In Merivale Moore, the Court of Appeal upheld the finding of negligence, even though the figures, as figures, were not negligent. They then went on to consider the quantum of damages. The judge had awarded damages on the basis of his finding about the true value of the property. In reaching his conclusion, he used a rental value of £55 (as opposed to the non-negligent £60 used by the valuer) and a yield of 8.5 per cent (as opposed to the yield of 7.5 per cent used by the valuer). The Court of Appeal upheld his award.

In Currys Group plc v. Martin [1999] 3 EGLR 165 the claimant sued in negligence against a valuer who had conducted a rent review. The submission made on its behalf was that it was sufficient to show that the defendant was negligent in his methodology in a way that was adverse to it, and that damages are recoverable even though the rent determined was one that a reasonably competent surveyor could have determined. That submission was firmly based on Lion Nathan. Merivale Moore had, by then, been decided in the Court of Appeal. The claimant therefore submitted that Merivale Moore was decided per incuriam, on the ground that Lion Nathan did not appear to have been cited. Mr Michael Harvey QC, sitting as a judge of the Queen’s Bench Division, rejected that submission for two reasons. First, he said, it was inconceivable that Buxton LJ overlooked the remarks of Lord Hoffmann in Lion Nathan, because that case had been referred to in an authority cited to the court and to which Buxton LJ referred in his judgment. Second, he said, the doctrine of per incuriam does not apply to decisions of the Privy Council. He therefore held that he was bound to follow the ratio of Merivale Moore.

In Arab Bank plc v. John D Wood Commercial Ltd [2000] 1 W.L.R. 857 the Court of Appeal again considered the question whether it was a precondition of liability in negligence that a valuer’s valuation should fall outside a margin of error. Counsel for the valuers conceded that normally, and on the facts of the actual case, there was such a precondition (para. 20). Mance L.J reviewed the authorities. He referred to the decision of the Court of Appeal in the Merivale Moore case (which I have already quoted). He continued at para 23:

“Where, as in the present case, criticism is addressed to factors such as rental value and yield which bear proportionately on the ultimately assigned value, the issues of the permissible range and of negligence are on any view inseparably linked. The value estimated results from the estimated rental values and yields. Where there is some discrete error, like that postulated in Lion Nathan Ltd v. C-C Bottlers [1996] 1 W.L.R. 1438, it may be appropriate to examine more closely the nature of the valuer’s engagement. Is it simply to produce an end result and to do so within the range of “reasonable foreseeable deviation?”. Or may it be to exercise reasonable skill and care in the circumstances (including whatever instructions may have been given) both informing and in expressing an opinion on value? In Banque Bruxelles Lambert S.A. v. Eagle Star Insurance Co Ltd [1995] Q.B. 375 when it was before this court the judgment given by Sir Thomas Bingham M.R. at pp. 403-404 lends some support to the latter analysis. On that basis, if, as a result of clearly identifiable negligence, a valuer arrives at a figure lower than he would otherwise have put forward, the line of reasoning indicated in the Lion Nathan case [1996] 1 W.L.R. 1438 might still be applicable, although the end figure could not itself be said to fall outside the margin of legitimate valuation by valuers generally. It is however unnecessary to consider this point further on this appeal.”

The decision of the Court of Appeal in the Arab Bank case may indicate that that court would reinstate the possibility of applying the approach displayed in the Lion Nathan case, if the valuer’s engagement is to do more than merely produce a result within the range of reasonably foreseeable deviation. In that event, it would be necessary to explore and define more closely the distinction between an error that is discrete, and an error that is not. However, the fact remains that Merivale Moore is ratio while the passage I have cited from Arab Bank is clearly obiter. In my judgment, consistently with Currys Group, and whatever my own doubts about the coherence of the law on this question, I must follow the ratio of Merivale Moore.

Mr Howe goes on to argue that even if the distinction drawn by Mance LJ is right, this is a case in which all that Levy Gee were required to do was to produce an end result. He says, with force, that since in the normal case a valuer of shares will produce a “non-speaking” award, the reasoning process should not be subject to independent scrutiny. The inquiry should focus on the final figure. The practical difficulty with this submission is that both the experts arrived at their final figures by considering the component parts of the valuation separately. In order to reach a conclusion on the validity of the final figure, it seems to me that I must replicate that process.

Mr Howe’s second pre-emptive strike was his submission that I should pay no attention to the opinions of Mr Hindley. Mr Hindley, he said, was not a chartered accountant or an auditor. On the contrary, he was a specialist share valuer. In Sansom v. Metcalfe Hambleton & Co [1998] 2 EGLR 103 the Court of Appeal warned against finding a professional to have been negligent on the evidence of an expert who was not a member of the same profession. That case concerned a structural survey prepared by a chartered surveyor. Expert evidence for the plaintiff was given, not by a chartered surveyor, but by a structural engineer. Butler-Sloss L.J. said:

“In my judgment, it is clear, from both lines of authority to which I have referred, that a court should be slow to find a professionally qualified man guilty of a breach of duty of skill and care towards a client (or third party) without evidence from those within the same profession as to the standard expected on the facts of the case and the failure of a professionally qualified man to measure up to that standard. It is not an absolute rule, as Sachs L.J. indicated by his example, but, unless it is an obvious case, in the absence of the relevant expert evidence, the claim will not be proved.”

In the result the court held that the evidence of a structural engineer was not admissible evidence on the question whether a chartered surveyor had been negligent.

In support of his submission that an auditor and a specialist share valuer were not the same, Mr Howe referred me to Whiteoak v. Walker (1988) 4 BCC 122. In that case, as in this, the articles of association of a private company provided for shares to be valued by the auditor of the company. The plaintiff transferred shares at a price fixed by the auditor, and subsequently alleged that the valuation was negligently made. One of the issues between the parties was whether the requisite standard of skill and care was that of a reasonably competent chartered accountant who professed specialist skills in valuing unquoted shares (“the specialist standard”) or that of a reasonably competent chartered accountant in general practice acting as an auditor who has agreed to a request to undertake the valuation task (“the auditor standard”). Mr Terence Cullen QC, sitting as a judge of the Chancery Division, held in favour of the latter. He said that the choice facing the members of the company, when agreeing to the terms of the articles of association, was between opting for the specialist skills of a share valuer and the special knowledge of the company’s affairs that the auditor would have. As they were seeking a fair result, as to which a specialist share valuer would not have a special advantage over their own auditor, Mr Cullen concluded that they intended the auditor to apply his skills and not the skills of a specialist share valuer.

Mr Howe did not go so far as to submit that Mr Hindley’s evidence was inadmissible. His submission was that I should attach little weight to it.

In my judgment, part of the skills of a chartered accountant, especially one who is willing to undertake a valuation of shares, is the valuation of shares. While such a person may not have all the skills of a specialist share valuer, I do not consider that the experience and expertise that Mr Hindley possesses is so far removed from the expertise of a chartered accountant willing to value shares as to make his evidence worthless. I do accept that I must be careful not to judge Mr Synett by reference to the standards applicable to a specialist; but neither expert suggested that there was any practical difference between the two on the facts of this case. In my judgment this is a non-point.

As I have said, there are three remaining complaints of negligence against Mr Synett. Both the experts produced valuations which depended on variables for the components in dispute. Neither expressed a view on the end value of the shares based on “market intuition” or “feel”. Each proceeded on the basis that the range of the end result was the product of the range of each variable.

The suggestion was made by Mr Gatt in the course of his cross-examination of Mr Woolf that the variables might be mutually dependent (what, in today’s jargon, he called “joined up thinking”). Mr Woolf agreed that this might be so as regards two of the complaints, namely the discount for lack of marketability and the reduction in value to take account of the probability that the options would be exercised. Apart from these two, I consider that on the basis of the evidence I must approach each variable separately.

I should also mention another point taken by Mr Howe. Based on Mr Preston’s rather grudging agreement that property valuers might be expected to differ by up to 10 per cent, Mr Howe submitted that even if it was negligent for Levy Gee to have adopted a portfolio valuation, that caused no loss, since the property valuers might have valued the properties at a figure which was 10 per cent lower than the figure at which they actually valued them. I reject that submission. No complaint is made about the figures given by the property valuers. What they might have done is, to my mind, beside the point. It was agreed by both experts, that, subject to one matter that I shall mention later, the share valuer does not second guess the property valuer. He puts the figure given to him by the property valuer straight into his spreadsheet.

Before I come to the specific complaints made against Mr Synett, I should say something about the nature of the exercise that Mr Synett was required to perform. The articles of association required a valuation as between a willing buyer and a willing seller. This necessarily postulates a hypothetical transaction. Although the transaction and the parties to it are hypothetical, the market in which they operate is not. All the real circumstances of the case must be taken into account except to the extent that the hypothesis otherwise requires, either expressly or by necessary implication.

The hypothetical seller is a willing seller. He is not an anxious seller, that is one who is willing to sell at any price and on any terms. In the hypothetical negotiations he will make all the points he can in support of his natural desire to get the highest possible price. The hypothetical buyer is likewise willing. He will make all the points he can in favour of obtaining the shares at the lowest possible price. Where the hypothetical seller and the hypothetical buyer will meet is a matter of judgment.

One theme that ran through Mr Synett’s evidence was that he approached the valuation from the perspective of the hypothetical buyer. He took into account all the negative points that could be made on behalf of the buyer, and factored them into his valuation. What he did not do was to consider what ripostes to these negative points could have been made by the hypothetical seller, and where the conflicting aspirations of the buyer and the seller might have met.

Mr Preston made it clear in his oral evidence that he adopted a discount of 10 per cent from the aggregate values of the individual properties solely because he had been instructed to value the properties as a portfolio.

Mr Synett gave two reasons for giving Jones Lang Wootton that instruction. The first was that the hypothetical buyer of the shares was in substance acquiring all the property portfolio in one transaction, and hence it was appropriate to value the portfolio as if it was being acquired directly in one transaction. The second was that the hypothetical buyer would not enter into the transaction at all unless there was a profit in it for him. The second reason seems to me to be manifestly ill-founded. The market price of any commodity is the amount at which both buyer and seller are willing to exchange the commodity for money. By definition both perceive it to be in their interests to do the deal at that price. In addition this explanation is contradicted by Mr Synett’s own evidence that if the property valuation had shown a higher value for the properties valued as a portfolio than valued individually, he would have used the higher figure in his share valuation.

In his draft report Mr Woolf said:

“In my opinion the shares should have been valued on the basis of an orderly sale of the properties assuming the company to be a going concern. [Para. 3.20]

I agree with Mr Hindley’s assertion that the only circumstances in which a property portfolio would be valued on the assumption that it would be sold in its entirety to one purchaser is on the break-up of the company. [Para. 3.23.11]”

Mr Hindley and Mr Woolf met on 17 January 2002. They prepared a joint statement of matters that they agreed. One thing that they agreed was that:

“if the 10% discount is attributable to JLW’s valuation of the properties as a single portfolio to be sold to a single purchaser, the application of such a discount is inappropriate.”

Both at the time of writing his first report and at the date of that meeting, Mr Woolf had been under the impression that there had been a misunderstanding between Mr Synett and Jones Lang Wootton about the basis of the valuation. His conclusion was that although Levy Gee adopted the wrong basis of valuation it was not their fault, and hence they were not negligent. Once Mr Woolf had seen the correspondence between Mr Synett and Mr Leigh, this position became untenable.

In his final report Mr Woolf adopted a radically different approach. He says that in his earlier report he had attributed the adoption of the portfolio approach to a misunderstanding between Mr Synett and Mr Preston. On further consideration Mr Woolf concluded that there had been no misunderstanding, and that the instruction to adopt a portfolio approach was a deliberate decision by Mr Synett. He then posed the question whether the adoption of the portfolio approach was appropriate. He concluded:

“The value of a company is deemed to be the price for which the ownership of the company would be transferred between a willing buyer and a willing seller on an arm’s length basis. Clearly, any willing buyer of Marchday would be buying the company’s property portfolio in its entirety.

On this basis, Levy Gee’s instruction to value the company’s principal asset, namely its property portfolio, “as a portfolio”, seems to me to be entirely appropriate. A valuation of the entirety of the company’s properties, as a single asset, must, after all, be consonant with the objective of valuing the property company’s shares: the shares represent the property assets as a whole. There is no direct relationship between individual shares and individual properties.”

However, this point had not escaped Mr Woolf’s attention when preparing his first report. In paragraph 4.8 of that report, he discussed the consequences to the hypothetical buyer of buying an indirect interest in all the properties, as opposed to selecting only those properties that he wanted. Yet he still came to the conclusion both in that report, and in the first agreed statement, that a valuation on a portfolio basis was incorrect.

Mr Hindley, on the other hand, adhered to the agreed position. He said in his report that if the company is to be valued as a going concern the properties should not be valued on the basis that they would be sold en bloc to a single buyer. The directors would ensure that the properties were disposed of in a manner that would maximise value and they would break the portfolio down into smaller units to achieve higher prices. In paragraph 7.11 of his report he said that a portfolio valuation criterion would be more appropriate in a “break up situation”. He concludes in paragraph 7.12 that:

“the application of a portfolio valuation is incorrect, would not have been requested by a reasonably competent valuer of the shares, and would, of necessity, depress the value of the shares.”

In his oral evidence Mr Hindley accepted that his statement that a portfolio valuation would “of necessity” depress the value of shares was not universally true, but said that it applied to this particular company, whose portfolio was unlikely to have a greater value if valued en bloc. He also pointed to the fact that the company’s balance sheet, signed off by Mr Synett as auditor, and prepared for the purpose of giving a true and fair view of the company’s financial worth, included the properties at £13.3 million for investment properties and £11.6 million for trading stock. Neither figure applied a portfolio discount. Both Mr Synett and Mr Woolf suggested that a portfolio valuation might, at least in theory, enhance the value of the properties, because they might be worth more if sold en bloc. In my judgment that is an unrealistic contention, for three reasons. First, in the case of this particular company, as Mr Leigh made clear, it was obvious that valuation as a portfolio would produce a lower value than a valuation of the properties individually. Second, as Mr Preston said, if a valuer instructed to value properties individually came to the conclusion that they could best be realised by lotting some together, he would report that to his client. This seems to me to be obvious. Third, by the time that Mr Synett came to prepare his own valuation, he had Jones Lang Wootton’s valuation and therefore knew for a fact that the instruction to value on a portfolio basis had decreased the value of the properties by 10 per cent. Mr Woolf’s espousal of this point undermines my confidence in his expressed opinion.

In the course of his oral evidence Mr Woolf suggested that one disadvantage of buying the portfolio that the buyer would perceive was that in disposing of the properties that he did not want, he would incur stamp duty of 2 per cent. This suggestion was ill-founded. In the first place, stamp duty is paid by the buyer not the seller. When the hypothetical buyer of the shares causes the company to dispose of the properties that he does not want to retain, the company will be the seller and not the buyer. So the company will incur no stamp duty. Second, any valuation of property already takes account of the cost to a purchaser of buying the properties to be valued. So stamp duty payable by the ultimate buyer is already factored in to the individual valuations of the properties. In truth, it seems to me that the question of stamp duty points away from the conclusion that a portfolio valuation was appropriate. The hypothetical buyer is not actually buying the properties; he is buying shares in the company that owns the properties. Had he been buying the properties themselves, he would have had to pay stamp duty at 2 per cent. By buying the shares, he pays stamp duty at ½ a per cent. So his saving in stamp duty terms is 1 ½ per cent. This advantage to him in buying shares rather than properties is one that the well-advised seller would surely point out.

I cannot see in Mr Woolf’s description of the new matters that came to light since the preparation of his first report and the agreed statement anything which explains his volte face. It seems to me that, having concluded in his first report that Levy Gee were not negligent for reasons that subsequently became untenable, Mr Woolf was trying to defend Levy Gee by any means available. I prefer Mr Hindley’s opinion on this point, which accords with Mr Woolf’s first thoughts and the agreed statement. It also accords with Mr Synett’s own first thought on the question, as recorded in his manuscript notes on Mr Leigh’s letter of 27 August 1997. Although the suggestion that purchase of the share capital of the company is equivalent to the purchase of the entire property portfolio is superficially attractive, it is, in my judgment unsound. First, it proves too much. If a discount to the individual values of the properties is appropriate at the stage of evaluating the company’s net assets, logically the same would apply to all the other assets of the company, including its motor vehicles, computer equipment, debtors etc. No one suggested that other assets should be similarly discounted. Second, it ignores the distinction between a company and its shareholders. As Mr Glover puts it in Accountants Digest (para. 2-1):

“Although the shareholder is a co-owner of the company, he is not a co-owner of its assets. This is not a legal quibble, but a reality which goes to the very essence of a share and the nature of a company.”

Third, the balance sheet, prepared in order to show the fair value of the company does not discount the property values. Mr Howe suggested that this was because the company’s assets are valued, for balance sheet purposes on the basis that the company is a going concern. So they are, but the shares are to be valued on the same basis. Fourth, Mr Woolf suggested in his oral evidence that on receipt of a valuation showing two figures, one on a portfolio basis and one on the basis of the aggregate of individual values the share valuer could not only choose one or the other, but could “blend them in some other way”. This suggests to me that whether to discount the aggregate values of the individual property assets is not a question for the property valuer, but a question for the share valuer.

In my judgment Mr Synett’s decision to instruct Jones Lang Wootton to prepare a portfolio valuation, and his subsequent decision to include in his share valuation the discounted figure for the portfolio valuation was an error, and fell below the standard to be expected of a competent auditor undertaking a share valuation.

Mr Synett said in evidence that in making his valuation he fed into his spreadsheet the figure which Jones Lang Wootton had supplied to him. He would have fed into it whatever figure they supplied to him, as they were the property valuers. This is, therefore, a discrete error which feeds directly into the bottom line valuation.

In addition both experts agreed that the discount of 10 per cent for the portfolio valuation should either be allowed in full, or ignored completely. There was no halfway house.

In my judgment, there is no permissible range for this item, and the figure to go into the spreadsheet should be the full aggregate values of the properties and not the discounted portfolio value.

Mr Synett deducted from the value of Marchday’s assets the whole of the tax that would have been payable if those assets had been immediately sold. It is agreed that, as regards those properties that were held as trading stock, he was correct to do so. But, as regards those properties that were held as investments, Mr Goldstein’s case is that it was wrong to deduct the whole of the tax that would have become due on an immediate sale of the entirety of the investment portfolio.

The rationale for such a deduction rests on the contrast between acquiring the properties directly and acquiring the shares. If a buyer were to acquire the properties, he would do so at their current values. The seller would pay tax on any capital gain. If the buyer subsequently sells a property, he will pay tax on any capital gain that has accrued during his ownership. But if the buyer acquires the shares instead, and the company subsequently sells a property, the company will pay tax on any capital gain that has accrued during its period of ownership; not merely the gain that has accrued since the change of ownership of the shares. The buyer of the shares could, however, avoid (or at least defer) paying that tax, either by not causing the company to sell the property at all, or by selling his shares rather than the property.

The company’s accounts did not show such a deduction as regards investment properties. This is because, for the purposes of presenting a true and fair view of the company’s finances, a contingent liability to tax is included only if the event which potentially gives rise to the liability to pay tax is likely to occur. If there is no realistic prospect of selling the asset in question, then no deduction for contingent tax liability needs to be made.

In a valuation that he prepared in September 1997, Mr Epstein made a deduction for the whole of the tax contingently payable, but qualified this deduction in a note saying:

“In order to be prudent, for the purposes of this valuation, any potential tax that would be payable on the disposal of any of the company’s assets shown in the balance sheet, should be deducted from the values of those assets. This procedure has been assumed notwithstanding the fact that in open market negotiations, some discount might be given for the deferred element of this taxation liability.”

Mr Synett said in evidence that although he had known cases where the whole of the contingently payable tax was reflected in the share price of a company, that was the exception rather than the rule. He accepted in cross-examination that in practice it was likely that a discount on the percentage would have been negotiated on an open market sale of the company’s shares.

However, he said in evidence that, in making his valuation, he did not consider the possibility that on the hypothetical sale the hypothetical buyer and seller would have negotiated a lesser reduction than 100 per cent of the tax contingently payable. His only stated reason for making the full deduction was that “a prospective purchaser would seek a full discount”. I have no doubt that he is correct in saying that a prospective purchaser would seek a full discount. However, that is not the right question. The question is not what the hypothetical buyer would seek; it is what the hypothetical buyer and seller would agree. My Synett never appears to have asked himself that question. In my judgment a competent auditor, carrying out a share valuation, would have asked himself that question. Since Mr Synett did not ask himself that question, I conclude that he fell below the standard of a competent auditor carrying out a share valuation.

The natural aspiration of the seller would be to achieve no deduction for contingent tax liabilities. The natural aspiration of the buyer would be to achieve 100 per cent deduction. Both the hypothetical buyer and the hypothetical seller are willing. I do not think that anyone suggested any particular reason why one would have a stronger bargaining position than the other. If the parties are of equal bargaining power, it seems to me that they would meet in the middle and agree a deduction of 50 per cent of the contingent tax liabilities.

The authors of Practical Share Valuation (1998) state at page 134 that allowing the full amount of the tax contingently payable on a disposal of a property company’s property assets is only appropriate if the company is being valued on a “break-up” basis. They continue:

“It could normally be appropriate to discount the potential charge to take account both of the fact that it would be over stating the net asset value of the company to ignore the tax charge, but also to recognise the fact that there is no immediate intention to dispose of the properties concerned and therefore no actual crystallisation of the tax charge.”

Apart from a reference in a previous paragraph to the Inland Revenue’s willingness to agree a deduction from asset value of between 10 and 30 per cent of the inherent tax liability, they give no guidance on the appropriate proportion of the contingent liability that should be deducted from asset value.

Mr Hindley’s opinion was that, having regard to Marchday’s historic policy of retaining investment properties rather than selling them, it would have been appropriate to have deducted a figure in the range from nil to 20 per cent. He took a middle figure of 10 per cent. Mr Woolf said that Mr Synett’s deduction of 100 per cent of the liability was “aggressive”. Mr Woolf’s own view was that an appropriate deduction from the asset value of the properties would have been between 50 per cent and 75 per cent of the contingent tax liability. In the light of his own view, I am surprised by his evidence that a deduction of 100 per cent was not unreasonable. Although it is a truism that valuation is an art and not a science, a valuation band in which one competent valuer may legitimately make a deduction that is double the deduction made by another equally competent valuer seems to me to overestimate the difficulties facing the valuer. Moreover, since it seems to be common ground that the likelihood is in practice that a lesser deduction than 100 per cent of the contingent tax would be negotiated, Mr Woolf’s expressed written opinion is at odds with reality.

In the course of his oral evidence Mr Woolf modified his position. He said that he had known of a case in which one of his colleagues, who was a particularly hard negotiator, had secured a deduction of the whole of the contingent tax. But, apart from that, his range would not have gone as high as 100 per cent. I cannot place much reliance on this unidentified transaction. There is no way of knowing what bargaining points were made in the course of the negotiations; and the buyer in a real case has real plans for the future of a company’s assets, whereas in the hypothesis the buyer is himself hypothetical.

Mr Woolf was asked whether a reduction from the full amount of the contingently payable tax would be appropriate in order to take into account the possibility that the investment portfolio would (or might) be disposed of over a number of years, and that the tax would not become payable immediately. He replied that a discounted cash flow approach might be appropriate. Although he had not done such a calculation, his instinct was that, taken over a 10 year period, it would come out at “a lot more than 30 per cent”. He did not, however, suggest that it would come out at anything like 75 per cent.

In the case of a valuation involving a hypothetical buyer, Mr Woolf was inclined to agree with the suggestion that since both buyer and seller were hypothetical, and in the absence of evidence that the property market was at an extreme point in the cycle, the hypothetical buyer and the hypothetical seller would split the difference and agree on a deduction of 50 per cent of the contingently payable tax on investment properties.

Mr Hindley accepted in cross-examination that, at least in theory, the negotiations between the hypothetical buyer and the hypothetical seller could have ended up with agreement on the allocation of the risk of having to pay contingent tax wholly to the seller (in which case it would be deducted in full from the asset value) or wholly to the buyer (in which case there would be no deduction) or anywhere in between. Although he himself preferred a deduction of 10 per cent of the contingent tax liability, he accepted that a deduction of anything up to 35 per cent of that liability would not be unreasonable. The reasons he gave for selecting 10 per cent were twofold. First, he had regard to the history of the company and the level of movement among its investment properties. Second, he said that the liability to pay contingent tax could be postponed indefinitely, although it could not be expunged completely. Even if the hypothetical buyer of the shares wanted to realise his investment, he could do so by selling the shares rather than the properties, in which case the contingent tax liability would not crystallise. I do not find the first of these reasons cogent. Ex hypothesi a sale of the whole of the issued share capital in Marchday would give control of the company to the buyer and it would be he who would thereafter decide what policy the company should adopt towards the disposal of investment properties. In addition history shows that about 10 per cent of the existing investment portfolio was sold annually. That would suggest that the entirety of the existing portfolio would be sold over a period of ten years. The second reason is more cogent, but the hypothetical buyer would no doubt say that a sale of shares restricted his flexibility. If, for instance, he wanted to sell only one of the properties, a sale of a minority shareholding in the company would be in no way equivalent to an outright sale of a single investment property.

Although Mr Hindley was also disposed to accept that a hypothetical buyer and a hypothetical seller would split the difference, he said that they would only split the difference as regards that part of the tax that would actually become due. The difficulty with this is that precisely because the parties are hypothetical no one can tell what the hypothetical buyer’s disposal plans are. I do not consider that Mr Hindley’s suggestion that only the tax that would become due would be split was well-founded.

Where does that leave me? Mr Hindley concedes that 35 per cent is not unreasonable. That, I think, is the lower end of the bracket. Mr Woolf’s bracket is between 50 and 75 per cent. Having regard to the practice of the Inland Revenue (which appears to be a recognised yardstick in share valuation), to Mr Woolf’s instinct about the result of a discounted cash flow exercise and also to the fact that the parties are hypothetical, I think that 75 per cent is too high. In my judgment the highest figure (and hence the upper end of the bracket) is 65 per cent.

I think that the most likely figure that a valuer would have deducted in the case of a hypothetical buyer and a hypothetical seller is (for the reasons I have given) 50 per cent.

In his valuation Mr Synett reduced the value of the company’s net assets by 12½ per cent. Although it was suggested that he did so to reflect the disadvantage of the non-listed status of the shares and the cost of realising the company's assets, I accept his evidence that the reference to the costs of realisation was a typographical error, and that the reason for the deduction was attributable to the non-listed status of the shares alone.

In my judgment there are two possible criticisms of Mr Synett’s calculations. First, he applied the deduction at that stage of his valuation in which he was considering the value of the company, leaving the exercise of the options out of account. When he came to recalculate the value of the company assuming the exercise of the options, he did not apply the same discount to the cash that would have formed part of the company’s net assets following that exercise. Yet a shareholder could only realise the cash by selling his shareholding, and that shareholding would remain unlisted. Logically, therefore, the discount should have been applied, if at all, both before and after the exercise of the options. Both experts agreed that this was so. Application of the discount in this way would tend to reduce the value of the shares. Thus in this respect the error that Mr Synett made was an error in Mr Goldstein’s favour.

Mr Hindley said that a discount for lack of marketability was inappropriate when valuing a sale of the whole of the issued share capital of an unquoted company. Mr Woolf agreed in cross-examination that this was correct. Thus the second criticism of Mr Synett’s approach, which was to apply a discount for lack of marketability in valuing the whole of the issued share capital of the company, is that both experts agreed that it was incorrect. This agreement between the experts is a little surprising. First, as Mr Woolf said, one of the differences between quoted and unquoted shares is that the shareholder can ring up his broker and dispose of quoted shares almost immediately, whereas with unquoted shares he cannot. This seems to me to apply to unquoted shares irrespective of the size of the shareholding. Second, in Accountants Digest Mr Glover deals with the discount for lack of marketability (para 4-4) without confining the discount to a shareholding of less than 100 per cent.

However, Mr Woolf went on to suggest that a discount for lack of marketability was appropriate when valuing the sale of a shareholding of anything less than the whole of the issued share capital. Since the ultimate object of the exercise was to value Mr Goldstein’s shareholding, it was appropriate to apply a discount. The same discount would apply, whatever the size of the shareholding, provided it was less than 100 per cent. Mr Howe supported Mr Woolf’s approach. He said that the ultimate object of the exercise was not to value the company as a whole, but to value Mr Goldstein’s shareholding alone. Even if a minority discount is eliminated, the fact remains that Mr Goldstein’s shares are not the same as the whole of the issued share capital of the company. It follows therefore that if a shareholding would attract a discount for lack of marketability despite being a majority holding, that discount ought to be taken into account.

I think that this is too glib, given Mr Woolf’s agreement that no discount was appropriate for 100 per cent of the share capital. Some shareholdings are for practical purposes the same as the whole of the issued share capital. A 99 per cent holding comes to mind. Even in the case of a 90 per cent holding, the shareholder has the right to acquire the remaining shares in the company. So I do not think that I can accept Mr Woolf’s answer at face value. Mr Woolf suggested that the directors might refuse to register a new shareholder. This seems to me to be unrealistic. If a shareholder has a holding of more than 50 per cent, and wishes to sell his shares, he will surely arrange for the board of directors to be at least sympathetic to his desire. Moreover, if the valuer is to value a shareholding which is not the whole of the issued share capital, he must, as it seems to me, consider what interest in acquiring the shares the other shareholders would have. One shareholder might be particularly anxious to acquire a relatively small shareholding in the company in order to achieve control. It seems to me to be unlikely that the members of the company, when agreeing to a valuation which made no discount for minority holding, would have contemplated such an investigation. I appreciate that there is a possible distinction between a minority holding (i.e. less than 50 per cent) and a holding which is not a minority holding but which is still less than the whole of the issued share capital, but it is unlikely that, in agreeing article 14, the members intended to distinguish between the two. The evidence was that a discount for the size of holding is applied even to majority holdings (at least up to 75 per cent); and it seems to me to be unlikely that the members of the company intended a minority shareholder to be entitled to a greater price per share than a majority shareholder. Yet that would be the result if the valuation in the case of a minority shareholding eliminated the minority discount (because it was a minority holding) but took into account the discount applicable to a majority holding of less than the whole (because it was not technically a minority shareholding). In my judgment the articles require the valuation of the company as a whole and then a pro-rating of that value to the particular shareholding.

I note also that in his book “The Valuation of Unquoted Companies” Mr Glover says that a discount for lack of marketability is “only appropriate where a capitalisation rate has been selected from marketable investments such as quoted securities”. However, neither Mr Synett nor the experts adopted this method valuation. All of them valued the company on a net assets basis.

Mr Howe argued that even if the construction of the articles that I prefer is correct, a competent accountant could have interpreted them in the way that Mr Woolf did, and that he could have applied a discount for lack of marketability for that reason. Mr Woolf’s range of permissible discount under this head was 8 to 15 per cent. However, after cross-examination he accepted that he would move to the lower end of his range. Even so, he was unwilling to accept that Mr Synett’s deduction of 12 ½ per cent was unreasonable.

I accept that a competent accountant (or for that matter a competent lawyer) could have interpreted the articles in the way that Mr Woolf did. But that fact is that Mr Synett did not interpret the articles in that way. He applied the discount at the stage of valuing the company as a whole. Moreover, he did not apply the discount to the whole of the company’s assets, after the exercise of the options, although both experts agreed that if a discount was to be applied at all, he should have done.

This raises, in acute form, the question posed by Merivale Moore. Mr Synett made an error in method which fell below the standard of a competent accountant. But that error will only be negligent if the end result falls outside the permissible range. In my judgment, Mr Synett’s valuation must be assessed on the basis of what a competent accountant could have done. The bracket must therefore take into account the possibility that a competent accountant could have made an appropriate deduction for the lack of marketability of the shares. In addition, it must take into account the possibility that a competent accountant would have applied the discount for lack of marketability at the end of the calculation, where it logically belongs.

Based on the evidence, I consider that a competent accountant could have made a deduction in the range 0 to 12 ½ per cent.

The mean figure, which is the most likely figure, is 6 ¼ per cent.

Both experts agreed that this was the most difficult area of the valuation. Although Mr Synett’s treatment of the share options is no longer relied on as a separate head of negligence, it seems to me that the logic of Merivale Moore dictates that I must assess the permissible range in relation to this item too, in order to arrive at the permissible range for the end result.

In his report, Mr Hindley took a figure which assumed a 25 per cent probability that the options would be exercised. In cross-examination he conceded that the figure of 75 per cent adopted by Levy Gee was a reasonable one. Although Mr Woolf suggested in his report that a competent accountant could have made an assumption that it was a certainty that all the options would be exercised (thus assuming 100 per cent exercise), he modified that in his oral evidence. His view was that the mere fact that the options were time limited meant that a lower percentage than 100 per cent probability was appropriate.

One of the reasons why Mr Hindley adopted a 25 per cent probability of the options being exercised was the existence of the section 459 petition. This he said, meant that there was a risk of the options being set aside. In my judgment, the petition was hopeless, at least once Mr Goldstein became bound to sell his shares. The best that he could have hoped for was an order for some form of compensation from the directors and/or option holders. In my view a competent accountant was entitled to disregard the existence of the petition.

Another reason why Mr Hindley selected a 25 per cent probability was the sheer number of options, and the amount of money that each option holder would have to raise in order to pay the exercise price. But it was common ground that the options would not in practice be exercised unless there was an early exit, by way of flotation or sale of the company.

In my view Mr Hindley overstated the difficulties of exercise. On the other hand, since exercise of the options was dependent on a sale or flotation of the company before the expiry of the options in 2005, a competent accountant could, in my view, have estimated the probability of the exercise of the options at 50 per cent.

I conclude therefore that the permissible range in relation to the options is a probability of exercise between 50 and 75 per cent.

My conclusions thus far may be summarised as follows:

The next part of the exercise is to feed these figures into the agreed spreadsheet in order to see whether Mr Synett’s final figure is within the permissible range.

Mr Howe submits that the way to do this is to take all the figures at the lowest end of the spectrum followed by all the figures at the highest end of the spectrum. Although one may instinctively feel that this stacks the figures in the way most favourable to the valuer, it seems to me that the logic cannot be faulted. That is, therefore, what I propose to do.

The first calculation, to calculate the lower end of the bracket, must therefore use the following variables:

Feeding these figures into the spreadsheet produces an end value for Mr Goldstein’s shares of £3,027,628 (£50.46 per share). The spreadsheet is reproduced as Appendix 1 to this judgment.

The second calculation, to calculate the upper end of the bracket, must therefore use the following variables:

Feeding these variables into the spreadsheet produces an end value for Mr Goldstein’s shares of £3,692,892 (£61.55 per share). The spreadsheet is reproduced as Appendix 2 to this judgment.

Levy Gee’s valuation of £3,151,200 (£52.52 per share) is within the bracket. It is also considerably higher than Mr Epstein’s near contemporaneous valuation. It follows, in accordance with the law as laid down by Merivale Moore, that it was not negligent. Consequently the action must be dismissed.