1. Plaintiff argues that we should reverse for error in granting the summary judgment on the ground that there was a triable issue of fact. We would agree, if Sirrine were available as a witness at a trial. For then it would have been error to deny plaintiff the opportunity to examine Sirrine in open court, with his demeanor observable by the trial judge.1 But as *962Sirrine died before the entry of the summary judgment, such examination is now impossible. There is no showing that, on the issue of the deficiency of Sirrine’s award, the testimony of any other witness for either party is needed. Accordingly, we will not reverse for failure to permit a trial of that issue.

2. We think that the record evidence clearly shows that the appraiser very substantially deviated from the standards of the submissions and that the award must therefore he set aside. Whether we consider paragraph (5) of the contract of May 23, 1943, as establishing a formula, or merely as binding instructions, it is plain that, if Sirrine’s award did not comply with the language of the contract, it is not binding.2 The gist of that paragraph is the provision that Woodstock must he valued “as a going concern in the light of the past, present and prospective future earnings and the net worth of said business.”

We have commented before on the difficulty involved in giving a precise meaning to the ambiguous word “value.”3 Methods of valuation vary with the legal contexts in which the valuation problems arise.4 But this, at least, is clear: When an enterprise is to he valued “as a going concern,” the valuer must consider whether it has a “going concern” value, and that value is something other than what results from the mere appraisal value of its assets 5 “The commercial value of property

consists in the expectation of income from it,” Galveston, H. & S. A. Ry. Co. v. Texas, 210 U.S. 217, 226, 28 S.Ct. 638, 639, 52 L. Ed. 1031. “The Supreme Court has several times said that the best test of the value of a going commercial enterprise is its earning capacity.” Dudley v. Mealey, 2 Cir., 147 F.2d 268, 270. Where the valuation of a going concern was required for the purposes of a corporate reorganization, the Supreme Court said that the expectation of income was the proper criterion, and added: “Since its application requires a prediction as to what will occur in the future, an estimate, as distinguished from mathematical certitude, is all that can be made. But that estimate must be based on an informed judgment which embraces all facts relevant to future earning capacity and hence to present worth, including, of course, the nature and condition of the properties, the past earnings record, and all circumstances which indicate whether or not that record is a reliable criterion of future performance.” Consolidated Rock Products Co. v. Du Bois, 312 U.S. 510, 526, 61 S.Ct. 675, 685, 85 L.Ed. 982.6

The appraiser here, in his answers to the interrogatories, maintained that he had estimated the value of Woodstock as a going concern, and had based that value on “what it could be sold for if you had a willing buyer and a willing seller.” This latter basis, it might be noted, is the usual test when an appraisal is made of property seized in a condemnation case.7 We do *963not hold that here the test was wrong as such. But it should have been qualified: it should have been “what a willing buyer would pay a willing seller of a going concern” Obviously, a willing-buyer would be unlikely to purchase an operating cotton mill without considering its operating costs and profits. But Sirrine’s answers show that he came to his conclusion without considering the operating profits or net earnings of Woodstock for the five years of that corporation’s operations under its then management, despite the express language of the contract that the value should be “in the light of past, present and prospective future earnings.” Whether, in the light of war conditions, Sirrine, if he had considered the operating profits, would have adhered to his conclusion that the enterprise had no prospective earnings, we are not in a position to determine. We can say, however, that, without such consideration, he could not have come to “an informed judgment which embraces all facts relevant to future earning capacity and hence to present worth.”8 We hold, therefore, that he did not value Woodstock as a going concern, and thus departed from the requirements of the contract.

We conclude, then, that the summary judgment must be reversed because, from what appears on this record, the appraisal was not in accordance with the agreement. Since we base that conclusion on the failure of the appraiser adequately to consider earning figures, we need not consider the following: (1) his alleged failure to make an adequate, examination of the mill; (2) his alleged improper consideration of losses suffered by former operators of the mill; (3) his alleged improper use of the accountants’ report.9

3. As the designated appraiser is dead, we think the district court should hear evidence and itself determine the price, applying the standards of clause (5) which were to govern the appraiser.10 For here the contract, which constituted a settlement of vexatious litigation, has been fully performed by the plaintiff, who cannot be restored to status quo ante, and the appraisal clause related to but one item of the settlement contract. On those facts the court should “substitute itself” for the deceased appraiser. Gunton v. Carroll, 101 U.S. 426, 430, 432, 25 L.Ed. 985; Texas Co. v. Z. & M. Independent Oil Co., 2 Cir., 156 F.2d 862, 867, 167 A.L.R. 719; Annotation, 167 A.L.R. 727, 743, 759; Cold Metal Process Co. v. United Eng. & Foundry Co., 3 Cir., 107 F.2d 27, 31, 32; Williams v. Cow Gulch Oil Co., 8 Cir., 270 F. 9, 12; Castle Creek Water Co. v. City of Aspen, 8 Cir., 146 F. 8, 11-13, 8 Ann.Cas. 660; Williston, Contracts (Rev. ed.) § 1421; Pomeroy, Specific Performance (3d ed.) 387-389.

Reversed.

(dissenting).

On the opinion’s premises, these two modifications are just and necessary. The first is a retreat from the extreme position of a grant of summary judgment (notwithstanding protestations critical of the practice) upon a factual conclusion directly contrary to that of Judge Knox and myself. With such a disagreement as to the facts, trial and formal findings on the testimony adduced are needed. The second is an acceptance, in part at least, of the arbitrator selected by the parties, in place of the substitution of a new arbitrator — the district judge — imposed by the court. This is desirable practically, as well as legally, in view of the vagueness and ambiguity in which this question of value is now left; at least it may then be passed upon by experienced engineers, rather than by an uninformed district judge. But all this, I think, points up the more the view I originally urged that no legal justification had been shown for upsetting the'studied conclusion of the original arbitration.

(dissenting).

The parties here stipulated for and admittedly obtained the honest and carefully foimulated judgment of an experienced arbitrator. I think it just as improper to trip him up on a hostile examination of his mental processes, and set aside his award on the basis of an incomplete answer removed from context, as it would be to reverse a trial judge after a similar grueling. It seems to me to neutralize and negate the strong judicial admonitions that a party who has accepted this form of adjudication must be content with the results. American Almond Products Co. v. Consolidated Pecan Sales Co., 2 Cir., 144 F.2d 448, 154 A.L.R. 1205, with annotation 1210-1215; Mutual Benefit Health & Accident Ass’n v. United Casualty Co., 1 Cir., 142 F.2d 390, certiorari denied 323 U.S. 729, 65 S.Ct. 65, 89 L.Ed. 585. Indeed, as applied to this case the words of Judge Learned Hand in the case first cited *964take on a fine irony. He said that the parties who have adopted arbitration “must content themselves with looser approximations to the enforcement of their rights than those that the law accords them, when they resort to its machinery.” 144 F.2d at page 451. Judicial valuation is a notoriously loose approximation at best. Cf. United States v. Brooklyn Union Gas Co., 2 Cir., 168 F.2d 391. But now it seems (though our final standard is in fact left distressingly vague) that we are about to force upon an ad hoc valuation by agreement a much tighter mechanical procedure than we would think of requiring of. a court.

As a matter of fact, the plaintiff has had to shop around until he could find a forum to listen to his plea. This is the third attempt he has made. Two previous proceedings in the state court — one going as far as the Appellate Division of the New York Supreme Court — were summarily dismissed without allowing the cross-examination of the arbitrator which he has obtained in the federal court. I have been a sincere advocate of federal discovery, but its reaches here have surely been far. That, in itself, however, I would not criticize. It is the setting aside of the arbitrator’s award upon what really seems to me a quite trivial basis that I find disturbing in its implications. Here the plaintiff resorted to arbitration because he found himself in a bad way with his then extensive litigation. By his agreement to arbitrate he secured the disposal of the pending lawsuits; and now after their threat has been done away with by his submission, he is able coolly to push aside the award which disappoints him and have another try for a prize which up till now has steadily eluded him. Such manipulation of arbitration seems to me destructive of its purposes and sure to bring it into disrepute as a proposed method of settlement of business disputes.

Our result is placed upon the ground that the arbitrator did not follow the formula of valuation of the stock agreed upon in the arbitration contract. The provision of the contract is quoted in full in the statement of facts, but is only sketchily discussed in the opinion itself. When it is examined with care, it is seen to be not a complete manual of valuation, but a caution designed rather to guard against inflated values than to push prices up. Thus the arbitrator '(a) is not bound by book value, but (b) is to determine the fair and reasonable value of the stock interests in the business of the Woodstock Corporation as a going concern (the latter surely, not its stock, is the “going concern”) in the light of past, present, and prospective future earnings and net worth of the business. This is a natural and so far forth admirable attempt to emphasize the present value of the business, as opposed to the initial investment. Then comes the important final provision, a direct warning against inflated valuation, in its requirement that, (c) in considering the value of Woodstock’s buildings, itiachinery, and equipment and its prospective future earnings, “consideration shall be given to what extent, if any, the present values and earnings are due to war conditions.” The agreement was signed on May 21, 1943. This was a textile business, as to which all of us still have vivid recollections of the wartime shortage and the operation of the companies to the top limit of which government regulation would permit. Of course the valuation of the business was not the ultimate conclusion; that was at most a step in finding the value of the stock. But the stock was not to be puffed up upon the basis of either future hopes or wartime profits.

In the light of this background I would not know, for my part, how any trier o-f the facts — judicial or lay — could proceed, except to consider the business background of the company, apart from war inflation, and then to try to decide as a matter of “business judgment” what a willing buyer was likely to pay for the stock. This is the normal rule, certainly not excluded from, if not reasonably incorporated into, the contract, provision. See our recent discussion and conclusion in a case of similar type in eminent-domain proceedings. United States v. Brooklyn Union Gas Co., supra. Since, so far as I can see, this is just what the arbitrator tried to do, I have great difficulty in discovering just what my brethren find wrong in his procedure, or *965what they now expect the district court— whom they are substituting for the agreed arbitrator — to do instead. The arbitrator was a textile engineer with unusual experience in the field — obviously much more than a district judge can have or indeed develop. Actually the parties were willing to accept his firm, J. E. Sirrine & Co., of Greenville, South Carolina, as the sole arbitrator; instead they received his own individual service. It is ironical that when he first sensed a possible dispute as to his award he tried to withdraw, but was induced to continue upon the insistence of plaintiff, at least equally with the defendant. There is not the slightest doubt that he had before him all the possible {factors for a valuation in the most approved amalgam suggested by any court, including all the material concerning the operating profits or net earnings of Woodstock for the five years in question. This we can see for ourselves, since we have before us the very voluminous reports, memoranda, and figures which the parties had previously submitted to him. Of course he valued the corporation as a going concern, i. e., as a business unit in operation. This he says over and over in his answers to the interrogatories. In fact, he had gone to the company plant and thus had seen it. As he says and reiterates, he tried to establish for himself by the exercise of a judgment informed by experience upon the various factors before him a value in exchange based on these factors and discounting the inflated values as the formula bade him do. Merely because in the light of hindsight we as an appellate court may think the arbitrator’s value not sufficiently generous does not justify us in according the plaintiff another chance at further litigation which to date he has been so steadily losing.

I say this even upon the assumptions made by my brethren as to the arbitrator’s course of decision, for under the authorities first cited or referred to above, it seems to me clear that he did substantially carry out the formula and that in any event the' plaintiff cannot complain even if it be assumed that he reached only a “loose” approximation of it. Hence Judge Knox properly awarded summary judgment to the defendant.1 But I do not think they have been fair to the arbitrator. The case is made to turn upon only certain of his answers when very ill, just before his death, he was compelled to submit to this inquisition as to how he had made up his mind more than three years earlier. In trying to make an honest answer as to his men;-1 processes and after continually emphasizing his attempt to reach a value in exchange for the business as a going concern, he did say at one point that he just took into account the final balance sheet to ascertain present quick assets and *966did not capitalize prospective earnings because he did not consider that it had much prospective earnings. Of course in any event he should not have capitalized prospective earnings. That is widely considered an oversimplification and an erroneous form of valuation, as we pointed out in United States v. Brooklyn Union Gas Co., supra. But the idea he was struggling to express was no more than that ¡he did not think the company’s position in the past (the business in fact had had a poor financial record) held out promise for the future, and he was placing his valuation substantially on the present conditions, as shown by its present balance sheet. And that, I submit, was a perfectly sound and valid approach.

In truth, the idea here stated is one he had expressed more succinctly in a letter, in answer to plaintiff’s complaints, of March 6, 1944, which, written when he was well and the matter fresh in his mind, is a fairer statement of his mental processes than this hostile examination. He then wrote: “I took into account what I considered a fair value of the property as of today based on its earning records and earning possibilities, and what other factors it seemed to me would properly bear on the subject. Frankly, I do not think the property has very much earning capacity and that opinion seems to be justified by its records.” In response to previous questions he had guarded himself by stressing the going value of the business, by refusing to put his result on various limited statements, and by pointing out as to particular factors that he “had to lump it in as a going concern,”'that he “did take into account what I thought the mill’s prospects were on any normal market,” and so on. Indeed, were I to be subject to a hostile cross-examination on any judicial act of mine, and particularly on any attempt to expound the mysteries of judicial valuation, I could only wish that I might do as well. Had he realized that he was really on trial or had he been protected by astute counsel, he would have protected himself in every answer, not in just the earlier ones, and he would have continually claimed merely the exercise of a “business judgment” upon a conglomerate of facts which would have included any particular facts about which he was being questioned. As it now stands, we are disavowing a very fair and frank judge who is not here now to defend his judicial conduct.

The difficulties of an appellate review of a valuation hodgepodge are shown by the posture of the case as it now goes back to the newly substituted trier of the facts. I frankly do not see what the district court is now to do in the way of valuation, unless perhaps it reads between the lines and decides we wish an increased award in some form, always discreetly concealed as a conclusion from the exercise of business judgment upon just the same facts and figures as were in fact before the arbitrator. For there is nothing in the opinion which suggests anything new or different from the old formula, of making sure that all sorts of facts and figures are thrown into the hopper of the judicial mind, to come out bearing the label of fair value. Since we have not yet gone to the point of direct cross-examination of the district judge, I suppose that must be regarded as adequate in the stead of what was found to be inadequate when done by an experienced textile engineer. At least the district judge will have before him the object lesson of the disastrous results to his award of a modicum of frankness on •the part of an arbitrator.

On Petition for Rehearing.

In the light of the petition for rehearing, our previous opinion must be modified in two respects:

1. In our former opinion we said, “There is no showing that, on the issue of the deficiency of Sirrine’s award, the testimony of any other witness for either party is needed.” Appellee points out that it may be able to introduce additional- evidence (for example, the correspondence files of Sirrine & Co.) which will show that Sirrine did in fact take earnings into consideration in making his award. We agree that appellee should have the opportunity to do so. We'therefore remand for a trial on the issue whether the arbitrator conformed to the terms of the submission, *967i. e., gave appropriate consideration to earnings.

2. The arbitrator named in the contract was J. E. Sirrine & Company, not Mr. Sirrine as an individual. Appellee now asserts that that company is still in existence, capable of undertaking the duties of an arbitrator. Appellant argues, on the other hand, that the J. E. Sirrine & Company named in the contract was a partnership, which expired on the death of its dominant partner, so that the company now known as J. E. Sirrine & Company is not the arbitrator named in the contract. He argues further that the parties understood and intended that the actual arbitration should be done personally by Mr. Sirrine.1

We think that the questions raised by these arguments must be considered by the court which tries the case. Assuming that the trial shows that the award did not conform to the submission, we reach the following result: Appellee, on its showing that the J. E. Sirrine & Company named in the contract is still in existence, will be entitled to a redetermination made by that company rather than by the District Court, unless appellant (with due regard to the parol evidence rule) shows that the parties meant to designate Sirrine individually.