FRIENDLY, Circuit Judge:
On July 10, 1968, this court vacated “so much of the Commission’s orders of [1052]*1052November 16, 1967 and March 1, 1968, as finds the acquisition of NH by Penn Central on the terms provided in the Purchase Agreement and the interim loss-sharing and loan arrangements to be just and reasonable” and remanded the cause to the Commission for further consistent proceedings. 289 F.Supp. 418, 448. We urged the Commission and all parties “to proceed with the utmost expedition” and to hold new testimony to a minimum. By order dated August 13, 1968, Judge Anderson, in the District Court for Connecticut, presiding over NH’s reorganization under § 77 of the Bankruptcy Act, made an order similar in tenor. 289 F. Supp. 451. The reorganization court added one highly significant further item —a direction that unless the Commission issued an order requiring Penn Central (PC) “physically to take over and commence operating the New Haven Railroad, not later than January 1,1969, leaving the price and the other terms of the inclusion to be determined later,” that court would consider itself “forced to entertain a motion to dismiss the reorganization proceedings” and “while liquidation will have to await authority to abandon, the trains will stop in January * * 289 F.Supp. at 459. The Commission acted with expedition, rendering its Fourth Supplemental Report (the Remand Report), on December 2,1968. 334 I.C.C. 25. In response to Judge Anderson’s opinion, it ordered PC to take over operation of NH at midnight on December 31, 1968, and to pay the price fixed by the Commission, which would, however, be subject to judicial review. 334 I.C.C. at 74-78. The reorganization court approved this order on December 24, 1968; we denied PC’s application to enjoin its execution; and the take-over was effected as scheduled.
On the other hand, our hope that the proceedings on remand would be of limited scope has not been realized. Both sides introduced much new evidence on subjects we had not expressly included in our remand, the Remand Report contains much new analysis, and we are confronted with almost as many issues as when the case was previously before us. We intend no criticism of the Commission on this score. While an administrative agency may choose to confine itself to the precise issues remanded, it is not bound to do so but may receive and consider such further evidence as it thinks requisite or desirable for the proper discharge of its duties. See FCC v. Potts-ville Broadcasting Co., 309 U.S. 134, 145, 60 S.Ct. 437, 84 L.Ed. 656 (1940); Fly v. Heitmeyer, 309 U.S. 146, 148, 60 S.Ct. 443, 84 L.Ed. 664 (1940). Moreover, with particular respect to the two large downward adjustments made under the heading “Other Pricing Considerations,” 334 I.C.C. at 53-63, we see no reason for refusing to accept the Commission’s explanation, 334 I.C.C. at 54:
In the inclusion report, we found that the consideration Penn Central had agreed to pay would equal liquidation value there shown, and that made it unnecessary for v. to explore other pricing considerations. The liquidation value that results in this reopened proceeding exceeds the agreed price, obliging v. to make a new determination as to whether the price resulting from such a valuation is fair.
Broadly speaking, what the Commission did in the Remand Report was this: On reexamination it found that if liquidation of NH could have been completed by selling cf. the properties individually in a normal market within six years from the date when a determination to liquidate had been made, the estate would have realized a figure having a present worth of $162.7 million, 334 I.C.C. at 30-53. It next argued that it was not constitutionally required to allow NH this or any other estimate of liquidation value, but only “a fair and equitable price under Section 5 of the Interstate Commerce Act,” 334 I.C.C. at 56. However, it found no need to base its decision on this ground, since it also found that
The alleged right to liquidation values derives from an alleged right to abandon; and there are recognized limitations on the right to abandon that in themselves limit the creditors’ [1053]*1053entitlement to the liquidation value we have computed under the court’s instructions. 334 I.C.C. at 57.
By making two further adjustments to liquidation value, the Commission reached a figure which it considered “a fair price on the current record,” and it is this sum which is before v. for review.
The two adjustments were of $15,386 million for a one-year delay before the start of liquidation due to the need for obtaining a certificate of abandonment, 334 I.C.C. at 58-60,1 and of $6,695 million for the lesser amount that would be realized if the properties were sold at one time as a unit rather than piecemeal over a six-year period, 334 I.C.C. at 60-63. These reductions, aggregating $22,081 million, led to a fair purchase price of $140.6 million, to which the Commission added $5 million as PC’s equitable share of NH’s losses from February 1, 1968 (the date of the PC merger) to December 31, 1968, 334 I.C.C. 65-71. The parties are in disagreement not only with respect to the propriety of the two large deductions but also concerning countless items of liquidation value and of consideration. We shall not attempt to canvass all these points but will deal with such, quite sufficient in number, as we deem to merit discussion.
I. The Scope of Review cmd the Governing Standard
We begin by reiterating the severe limitations on the scope of our review. In our last opinion in this case, 289 F.Supp. at 427, we called attention to what we had said on that subject in the N & W Inclusion Case (Erie-Lackawanna R. Co. v. United States) D.C., 279 F.Supp. 316, 337 (1967), aff’d with minor modifications, 389 U.S. 486, 88 S.Ct. 602, 19 L.Ed.2d 723 (1968). We also cited Mr. Justice Cardozo’s pertinent admonition, “An intelligent estimate of probable future values, * * * and even indeed of present ones, is at best an approximation. There is left in every case a reasonable margin of fluctuation and uncertainty.” Dayton Power & Light Co. v. Public Utilities Comm’n, 292 U.S. 290, 310, 54 S.Ct. 647, 656, 78 L.Ed. 1267 (1934). 2 Such considerations are peculiarly applicable to a determination so hypothetical as the liquidation value of the New Haven. What that really is, no one ever has known or ever will know, even approximately. The idea of the states of Connecticut and Rhode Island and of a large part of Massachusetts being without rail service is so unthinkable that its consequences on land values are beyond possibility of accurate calculation. When passing on such an issue, it is peculiarly important for a reviewing court to remember that its function is only to make certain that agency actions “are based upon substantial evidence and to guard against the possibility of gross error or unfairness.” Penn Central Merger and N & W Inclusion Cases, 389 U.S. 486, 524, 88 S.Ct. 602, 621, 19 L.Ed.2d 723 (1968).
When the case was last before us, we noted various faults and fuzzy patches in the Inclusion Report, but the principal ones were what amounted to two arithmetical errors aggregating more than $20 million, 289 F.Supp. at 427-428, 438, and a substantial understatement, stemming from a plain misconception on a matter of law, of the value of NH’s claim to half of the “excess income” from the Grand Central Terminal properties, 289 F.Supp. at 428-430. The arithmetical mistakes could have been corrected without a remand, and it might also have been possible to make our own determination [1054]*1054with respect to the GOT properties. However, since the Commission had justified its approval of the price fixed in the Purchase Agreement on the basis that it reflected the liquidation value standard for which the NH bondholders were contending, without, however, committing itself to the need for going so high, and we found this factual position was not borne out by the record, we remanded in order to give the agency an opportunity either to revise its results in the light of our finding or, if so advised, to explain why no revision was required. '289 F.Supp. 440-441. We also thought it best that the Commission should undertake the complex task of setting a value on the GCT properties under the legal standards we had laid down with respect to the strength of NH’s claims, and, since there was to be a remand in any event, that it should explain certain other obscurities.
On this second review we confront once more the question of the proper standard of valuation, which has now been fully discussed by the Commission. 334 I.C.C. 53-57. This matter apart, no significant appeal is made to our authority to interpret the Constitution, statutes, or contracts. There are again a few alleged arithmetical errors, this time relatively minor. But the overwhelming bulk of the attacks concerns the Commission’s economic judgments. With respect to these we must be on guard lest the familiarity about the facts of the case we have somewhat painfully gained should lead v. simply to substitute our views for those of the agency which Congress has appointed.
We begin with a word on whether the Commission was bound to award liquidation value although, like the Commission, we do not find it necessary to decide this. One can scarcely quarrel with the proposition that in general “A fair and equitable price * * * means a price that is fair both to the buyer and to the seller.” 334 I.C.C. at 56. In the Norfolk & Western Inclusion Case, 330 I.C.C. 780, 801 (1967), the first instance in which the Commission was required to prescribe a price, it said:
In the ordinary ease under section 5(2) (d) the acquired line’s maximum commercial value to the acquiring line cannot exceed a price based upon past and foreseeable earnings of the acquired line plus the value of all benefits the acquired line brings to the acquiring line’s system. Moreover, this price is ordinarily the maximum we should expect the acquiring line to pay. It appears that the price, nevertheless, will generally be below book value, and hence it will ordinarily be the minimum price the acquired line will find tolerable.
The prices the Commission set on this basis were later upheld by the Supreme Court, Penn Central Merger and N & W Inclusion Cases, 389 U.S. 486, 524, 88 S.Ct. 602, 19 L.Ed.2d 723 (1968), not simply as against the acquiring line’s contention that they were too high but as against a claim by one of the lines to be acquired that the price, considerably less than current market quotations, was too low.
Here the “past and foreseeable earnings of the acquired line” are huge and increasing deficits — deficits far larger than in 1966 when the Commission required inclusion of NH as a condition of the Penn Central merger, 327 I.C.C. 475, 526-27 (1966). The only “benefit” inclusion of NH is claimed to bring to PC is that the Commission would not have approved the Penn Central merger without it. However, the Commission has rejected the arguments of the NH bondholders based on adding to the value of NH whatever potential savings from the merger are needed to justify a high inclusion price, saying, “This was not the intent of our decision,” 334 I.C.C. at 66, a view we had previously intimated, 289 F.Supp. at 443. There is thus a decided anomaly in the position that, looking only at the words of the statute, the Commission was required to make PC pay a price for NH which is far beyond anything NH is worth to it in an economic sense and which, by hypothesis, it can never realize.
[1055]*1055A more difficult question is whether the Fifth Amendment forbids application of the statute in such a manner as to produce a price less than liquidation value on the facts here presented. When the case was last before us, we indicated tentative agreement with that position. We cited authorities which established in our view that a railroad with no reasonable hope of future earnings, such as NH, had a “constitutional right * * * to cease operations and liquidate,” 289 F.Supp. 440, since to force operation at a loss “would be to take its property without the just compensation which is a part of due process of law.” Railroad Commission of State of Texas v. Eastern Texas R. R., 264 U.S. 79, 85, 44 S.Ct. 247, 249, 68 L.Ed. 569 (1924); Bullock v. State of Florida ex rel. Railroad Commission, 254 U.S. 513, 41 S.Ct. 193, 65 L.Ed. 380 (1921); Brooks-Seanlon Co. v. Railroad Commission, 251 U.S. 396, 40 S.Ct. 183, 64 L.Ed. 323 (1920). Despite observations in the Commission’s report, 334 I.C.C. at 55-57, we adhere to that view, for reasons fully developed in the recent opinion of the reorganization court, subject only to the qualification that liquidation of NH could not commence until the Commission, acting with appropriate speed under § 1(18) of the Interstate Commerce Act, had issued the certificate of abandonment to which NH would have been constitutionally entitled. We believe this position is also supported by two opinions of Mr. Justice Brandéis which we read as declaring that while Congress may in the public interest subject a secured creditor to reasonable delays in realizing on his security, it may not deprive him altogether of his right to sell the res at public auction. Louisville Joint Stock Land Bank v. Radford, 295 U.S. 555, 55 S.Ct. 854, 79 L.Ed. 1593 (1935); Wright v. Vinton Branch of Mountain Trust Bank, 300 U.S. 440, 57 S.Ct. 556, 81 L.Ed. 736 (1937).
However, we are not so sure now as we formerly were that these precedents necessarily lead to the further conclusion concerning the Commission’s duty here that the NH bondholders would draw from them. The just compensation and due process clauses are triggered only by governmental action. Existence of governmental compulsion was not disputed in any of the cases cited above; it is disputed here, for PC argues that although the bondholders seek compensation for governmental taking of a right to liquidate, they never properly asserted such a right. The NH trustees petitioned for inclusion in PC about three months after the merger application was filed in 1962. The bondholders at first supported this move, and it was not until April 1967, long after the commitment to seeking inclusion was for practical purposes irrevocable, that some of them petitioned to dismiss the reorganization. Judge Anderson denied this, In re New York, N. H. & H. R. R., 281 F.Supp. 65 (D.C.1968), the decision was not appealed, and when immediate inclusion was finally ordered, it was without protest from any party except PC. If we could take all this as indicating that inclusion had in fact been sought solely because the bondholders and the trustees felt it would be more in NH’s interest than abandonment, we would find nothing in § 5(2) (d) or in the Fifth Amendment that would prevent the Commission’s fixing a price considerably short of liquidation value and saying, in effect, “This is all we think it fair for the acquiring road to be required to pay. We see no reason why it should be made to assume the entire cost of keeping alive a carrier whose operations the public will not support. On the other hand, if you [the seller] don’t like it, you are free to refuse and seek the abandonment to which you are entitled.”
On the other hand, the present case may well be one where a volume of history must prevail over a page of logic. Ever since June 1962, when the Trustees petitioned for inclusion of NH if the PC merger was approved, it has become increasingly apparent that this solution was so vastly preferable in the public interest that, as we thought, 289 F.Supp. at 444, and the Commission now has confirmed, 334 I.C.C. at 59, “If the Trustees [1056]*1056had presented abandonment applications as an alternative to sale, the Commission would almost certainly have deferred their consideration for a reasonable period while inclusion was being explored.” The bondholders say, with considerable force, that this known attitude of the Commission was the practical equivalent of the denial of an application for abandonment and entailed an obligation to procure for them what they could have realized on abandonment if inclusion became a reality. They point out also that, during the long period when the PC merger was pending and before inclusion occurred, they have been making a substantial contribution to the public interest through the erosion of their property by losses of cash after any real possibility of a viable service had disappeared. Hence, the proper administration of § 5 (2) (d) may require at least a close approach to what they could have realized in an assumed liquidation even though, because of the absence of an actual taking or other governmental compulsion, the Fifth Amendment does not.
Because of the way the Commission has approached the valuation question, we do not feel called upon finally to decide these intriguing issues. The Commission set out to fix a price that would give NH what it thought the estate could realistically hope to achieve on liquidation. If the Commission made demonstrable errors, it is our duty to correct these, hopefully without a still further remand. But, as indicated, we are not required, or indeed permitted, to reexamine every judgment made by the Commission and to substitute our own whenever we think it better. Our function is to determine whether the agency’s conclusions are “equitable and rational,” not “to second-guess each step in the Commission’s process of deliberation.” Penn Central Merger and N & W Inclusion Cases, 389 U.S. 486, 524, 88 S.Ct. 602, 621, 19 L.Ed.2d 723 (1968).
II. The One-Year Delay for an Abandonment Proceeding
At the initial hearings concerning inclusion, the NH trustees submitted estimates of asset values as of the end of 1965, and PC submitted estimates as of the end of 1966. In the Inclusion Report, the Commission by and large accepted the former figures, then adjusted them to the later date, spread them over a six-year estimated liquidation period, and discounted them back to present value. At that stage the Commission was thus assuming a liquidation starting January 1, 1967. The further adjustment in the Remand Report amounts to assuming that the decision to liquidate would have been taken on that date but that liquidation could not begin until a year later because of the necessity for obtaining a certificate of abandonment under § 1 (18). Passing on the reasonableness of such a determination requires a reviewing court to enter a never-never land. The Commission had to estimate the starting date of a liquidation that never happened, that in the world as we know it scarcely could have happened, and that, if it had happened, could have happened in any one of a number of equally imaginary ways, each based on a different pattern of demands on governmental authority that were never made, and ensuing administrative and judicial proceedings that never occurred.
We find no lack of rationality in the Commission’s view that even if the liquidation decision is assumed to have been taken by the end of 1966, although in fact the bondholders did not apply for dismissal of the reorganization proceedings until the spring of 1967, authority to abandon would not have been granted until a year thereafter, even though we believe, as the Commission may not, that it would have been constitutionally required to issue a certificate of abandonment if inclusion was not a viable alternative. Even in such a case an abandonment proceeding serves the important office of affording persons interested in maintaining the service, here the four states directly concerned, an opportunity to make proposals for acquiring so much of the property as was considered essential. See Palmer v. Massachusetts, 308 U.S. 79, 60 S.Ct. 34, 84 L.Ed. 93 (1939); [1057]*1057Meyers v. Jay St. Connecting R. R., 259 F.2d 532 (2 Cir. 1958). The NH bondholders answer that except for the probability of inclusion in PC they would have insisted on abandonment at a much earlier date and that the states have already had such abundant opportunity to make proposals to take over NH’s operations that no significant further period could properly have been accorded. Despite the forceful exposition of these arguments in the recent opinion of the reorganization court and in Judge Weinfeld’s dissent, we do not find either so compelling that the Commission was bound to accept it. Whatever hindsight may indicate would have been wise, the bondholders did not in fact seek dismissal of the reorganization proceedings until the spring of 1967; indeed they had been contending before the Commission that NH could survive as a freight-only railroad. See 331 I.C.C. at 659 (1967). And legislatures can hardly be expected to volunteer to assume large financial burdens so long as they have a reasonable expectation that part of these will be shouldered by a private corporation. Beyond this, as we pointed out in our previous opinion, 289 F.Supp. at 444, if inclusion of NH in PC had been found impracticable —or if, for that matter, the PC merger had been disapproved by the Commission or after administrative approval had foundered in the courts or the constituent roads had been unwilling to accept conditions that were imposed — “the NH bondholders would have had to shoulder losses between the date of such a determination and the grant of authority to abandon.” The effect of what the Commission has done here is to assume that liquidation could have begun on January 1, 1968, a month before the PC merger occurred and NH’s right to inclusion thereby accrued. We see no inexorable logic that required the Commission to assume that liquidation could have started a year earlier, when the first appeal in the Penn Central merger had yet to be argued in the Supreme Court. In sum, we perceive no lack of reason in the Commission’s assuming that a decision would not have been taken before January 1, 1967, and in postulating á delay of a year from the taking of a decision to liquidate for the time a proceeding under § 1(18) would necessarily entail.3 Viewing the matter in a practical rather than a theoretical frame, what the Commission has done here was to select a fair date for shifting the heavy losses of NH’s operation from the bondholders to PC — a judgment peculiarly in its province. We cannot say it went beyond bounds in assuming January 1,1968, as the starting point of the liquidation, adding to the liquidation value (determined as of the end of 1966) $5,000,000 on account of 1968 losses, and requiring a take-over at the end of 1968. '
We cannot join our brother Weinfeld in reading our previous opinion, 289 F. Supp. at 438, as precluding the Commission from postulating a one-year period after December 31,1966, for an abandonment proceeding. What we rejected was a claim that we should allow for such a delay although the Commission then had not. Moreover, as the Remand Report pointed out, 334 I.C.C. at 58, a principal reason for our ruling was that, as matters then stood, the NH estate would be subject to a portion of operating losses for several years after consummation of the PC merger, whereas that burden has now been limited to eleven months. For whatever importance it may have, we likewise do not read the quotation from p. 5 of the joint brief of the Commission and the United States, cited in footnote 16 of the dissenting opinion, as our brother does. The “directive” was to make “detailed findings regarding the piecemeal liquidation value of the NH assets, as if an abandonment certificate [1058]*1058were in hand on December 31, 1966”; we did not mean thereby to preclude the Commission from also considering whether or not it was reasonable to assume that such a certificate would then be in hand and making the necessary adjustment if it answered that question in the negative —any more than we meant to prevent the Commission from finding on remand, had it so chosen, that ten years rather than six would constitute a reasonable period for completing liquidation. If such a determination, which would diminish liquidation value by some $17.5 million, see 331 I.C.C. at 662, would have been within the Commission’s power as it assuredly would, the instant one is equally so. The matter simply does not lend itself to the certainty that the bondholders assume.
The NH bondholders argue further that even if the Commission was justified in postulating a one-year delay in starting liquidation, its figures concerning the effect of this contain numerous errors. They say that if the date for starting liquidation is pushed forward a year, all the properties will have to be revalued — presumably upward. PC responds that the failure to revalue has been beneficial to the bondholders, and that in any case far longer gaps between date of valuation and presumed date of sale have previously failed to stir objection. For example, the properties were all valued as of December 31, 1966, even though it was assumed they would not be sold then but over the six years following, and the same figures were also used in the study that assumed a ten-year program of liquidation sales. The bondholders argue also that non-operating property could be sold immediately, to which it is answered that the record affords no indication how much of this there is and that the Commission was warranted in assuming that, with their serious cash stringency, the NH trustees had done about all such selling that they could. The bondholders contend that sales contracts could be made on a contingent basis, that only sales scheduled for the first year would be delayed, and that the preservation costs would be less than estimated, particularly since holders of defaulted equipment obligations would repossess. While these arguments are not without persuasiveness, they are offset by the Commission’s failure to deduct a further year’s depreciation of some $3 million, by the conservativeness of the estimate that one year would suffice for an abandonment proceeding of such magnitude and possible judicial stays, and by the Commission’s choice of January 1, 1968 as the date for beginning liquidation instead of imposing on NH the costs of an additional year’s delay in order to arrive at the possibly more logical starting date of February 1, 1969 4
Of the many issues in the case, the hypothetical question here posed is least susceptible to scientific determination. We cannot say that, all things considered, the Commission’s basic assumption was irrational or its detailed findings unsupported by substantial evidence.
III. The Bulk-Sale Discount.
At the hearing on remand PC adduced the testimony of Mr. Robert E. Simon, an experienced real estate operator. The burden of this was that the appraisals of the sale of land in some 1700 parcels over a six-year period, discounted merely by 6%, did not reflect certain elements of risk in such an undertaking, notably the likelihood that the liquidation might take longer than estimated and the possibility of an economic downturn, even though the appraisals had substantially marked down land remaining unsold at the end of the fourth year. Mr. Simon thought the proper measure of the value of the land was the price it would command if sold immediately to someone who would assume these risks. Such a person, he testified, would finance 75% of the purchase price by loans, which could not be obtained at less than 9% and [1059]*1059would demand a 15% profit on his own investment for his entrepreneurial contributions and for undergoing the downside risks, despite the possibility that even even higher prices might be realized. Combining these figures to reach a composite of 10.5%, the Commission concluded that the land prices should be discounted by 4.5%, or $6,695 million, above the 6% previously deducted.
In the course of justifying the deduction, the Commission said, 334 I.C.C at 60, that “the liquidation value urged by the creditors assumes not only the immediate right to abandon * * * but also the right to break v. the railroad and sell the parcels for their highest and best price” and that the latter right “may be restricted when a buyer for the entire bulk of the NH properties appears who will continue the operation of needed services.” We entertain the gravest doubt concerning the legal premise underlying this argument; we would suppose that the constitutional right to abandon operations implies a right to break v. the railroad and sell the parcels for their highest and best prices. Moreover we fail to see the relationship of this argument to Mr. Simon’s testimony, which was not at all concerned with a buyer “who will continue the operation of needed service” but rather with one who would do precisely what the NH estate proposed to do.
However, the Commission also stated, “The bulk-sale discount merely reflects a market appraisal of the risks that the estate avoids, and the bulk buyer assumes,” 334 I.C.C. at 61, and that the discount “fully recognizes, without conceding, NH’s right to abandon and liquidate the estate.” 334 I.C.C. at 63. While the Commission’s language is not pellucid, we read this as saying that the Commission deducted the discount as a proper measure of risks not taken into account by the appraisals and the discount to present value. We cannot find this conclusion to be without rational basis; the Commission was not bound to aecépt the testimony of a witness called by the bondholders that the discount to present worth was all that was needed. In addition to the factors emphasized by Mr. Simon, there is the fact, to which we have referred above, that the appraisals gave no effect to the serious adverse economic consequences from discontinuance of NH’s operations, 331 I.C.C. at 667.5 Corporations issuing securities pay substantial premiums for underwriting against the risk of market downturns even for relatively short periods,6 and public utility commissions have considered the charges for such an underwriting as an element in the cost of capital in determining a fair return. See Bonbright, Principles of Public Utility Regulation 251-52 (1961); Garfield & Lovejoy, Public Utility Economics 124, 128 (1965). If it be thus prudent to guard against the risks of downturns over a period of a few weeks, we can hardly find it irrational for the Commission to have concluded that it would be imprudent for the NH estate to take this risk over a period of six years.
IV. Alleged Errors in Computing the Liquidation Value of NH.
A. Grand Central Terminal Properties.
In our previous opinion we concluded in effect that while no court had ever decided that NH was entitled to receive 50% of the surplus income of the GCT properties regardless of its cessation of operations, its claim to this was “exceedingly strong,” 289 F.Supp. at 431. On remand the Commission made a conscientious effort to appraise the value of this share in surplus income and arrived at a figure of $28.887 million as against its former figure of $13 million and our guess of “at least $25 million,” 289 F. [1060]*1060Supp. at 432-438, which, however, did not take account of the proposed Tower Building. The bondholders contend the figure to be too low; PC says it is too high. Our conclusions parallel those of the reorganization court.
The largest single item selected by the bondholders for criticism is the failure to credit NH with the value, estimated at $5.6 million annually, of NH’s right to have its share of terminal expenses met by income from GCT properties. There is no question that this constitutes a significant benefit to PC as compared to a situation where the NH estate would share in GCT income and PC would be left with all the expenses — a situation we have previously found not to be within the language of the contracts or the probable intent of the parties, 289 F.Supp. at 430-431. But it does not follow that a capitalization of this amount constitutes an element of liquidation value of NH. Unless someone could be found to take over NH’s operations into GCT, the right to have terminal income applied to expenses could not be sold to anyone even if the anti-assignment clause in the contracts was not the obstacle that it may be. While it may well be that the State of New York would not allow all of NH’s commuter services to disappear, there is no assurance in this respect and still less assurance that the State would operate into GCT if it had to pay the NH estate the full capitalized value of NH’s right to set Terminal income cf. against Terminal expenses. Indeed, the State, if it wished to continue using GCT, might well be able, either through its own bargaining position or the power of the ICC, to obtain a transfer of this right without making any payments to NH that would increase the “liquidation value” of this unique type of asset above the sum that could be realized if the State were not a buyer. 289 F.Supp. at 431. While we do not disagree with the bondholders’ position that a state must pay the salvage value of property of a defunct railroad that it wishes to acquire, what is the “salvage value” of property that cannot be sold to anyone else and is of no value in the owner’s hands? It is true, on the other hand, as we previously noted, 289 F.Supp. at 431, that cessation of all NH operations into GCT might affect the liquidation value of NH’s interest by increasing surplus income by reducing total Terminal expense. But the bondholders offered no evidence on remand what this would be, and there are countervailing factors, such as probable decreases in concession income and expectable inflationary increases in Terminal expense over the years.
We are similarly unconvinced that the Commission acted without substantial evidence in estimating future hotel income at the average shown in the Terminal Account for the years 1964-67. There was a downturn from 1964 through 1966, while 1967 came near (although not v. to) the level of 1964. The figures in the Terminal Account are net of capital expenditures, and a chief claim of the bondholders is that these should have been removed and deduction made only for what would be approprial. under depreciation or retirement accounting. The Terminal Account in this respect only reflects the long-standing practice of the parties, which is embodied in the hotel leases themselves. Beyond this, the Commission was justified in concluding from the absence of any trend in Terminal Account earnings from 1960 to 1967 that the annual capital expenditures on these old hotels, however, described from an accounting standpoint, constituted regularly recurring costs necessary to maintain income rather than likely to increase it. 334 I.C.C. at 36-38.
Another bone of contention, not dealt with in our previous decision, concerns a lease recently negotiated by PC of the air rights over Grand Central Station for the construction of an office building. PC stipulated that if this were built, it would bring an annual income of $2.3 million or under “certain favorable circumstances” $3 million, beginning about 4 years after the commencement of construction. Prior to the completion of construction the annual [1061]*1061rental would be $1 million, but a deduction from this figure must be made for the transfer to the lessee by the railroads of concession rights worth $700,000 a year, and for certain other expenses. Although the Commission considered the question “whether the proposed building is so speculative that no value can be assigned to it,” it concluded that “it is probable that the building will eventually be constructed,” but only after long delay. 334 I.C.C. at 34. It assumed accordingly that the full rental of $2.3 million would not be realized until 10 years from the valuation date of December 31, 1966. This resulted in a discounted figure of $1,280,000; capitalization of NH’s 50% share at 8% produced a valuation of $8 million.
The bondholders complain over the failure to take account of any part of the $1 million annual rental during construction and use of the. $2.3 million rather than the $3 million figure thereafter. But, apart from the absence of proof that the latter figure was fairly expect-able, any conservatism on these scores was a reasonable offset for the meagre weight given to the doubt whether the building would be constructed at all. Indeed we consider PC’s grievances with respect to the Tower Building considerably more impressive than the bondholders’. While we would have had no objection if the Commission had handled the matter as our brother Weinfeld ingeniously suggests, we see no basis for compelling this.
The only other item relating to GCT requiring discussion is the Commission’s refusal to require PC to pay NH its share of surplus income for 1967 and 1968. The justifications for this refusal were that to make the payment “would be giving the claim a double value, once as a dollar claim and secondly as the capitalized value of its claim as of December 31,1966”; that the bondholders were in effect seeking to move the valuation date forward; and that in that event the Commission “would also have to make other adjustments, such as depreciation for 1967 and 1968; and such reevaluation of the estate might be quite different from the ultimate liquidation figure we have found.” 334 I.C.C. at 38-39.
Such arguments misconceive the nature of the bondholders’ claim, which is simply that they are entitled to receive whatever their property earned while they still owned it. Only if PC had paid NH for the assets on the valuation date could it properly argue that the bondholders were trying to recover twice by laying claim to the earnings of those assets while they also retained the interest paid on the sum given in exchange for them. When the assets were valued, and what they were valued at, is beside the point.
B. Harlem River and Oak Point Yards.
After careful reexamination the Commission adhered to the $18,090,990 appraisal submitted by the NH Trustees “as representing the value of the yards in the event NH were liquidated and its line of railroad dismantled.” 334 I.C.C; at 41-48. Judge Anderson has sustained the bondholders’ claim for a higher appraisal of $22,650,000, which assumed that all the industries in the two yards would enjoy the present level of rail service. The principal argument for this is founded on the premise that, in the event of liquidation of NH, PC would be obliged to continue service over its Port Morris branch to the Hunts Point Wholesale Market which is located north cf. but not adjacent to, the Oak Point yard. Acceptance of the premise would not necessarily entail the conclusion necessary to compel adoption of the higher figure. The Commission’s evaluation of the probable scope of future rail service to these properties in the event of liquidation of NH lies in an area peculiarly within its expertise, and we decline to disturb it.
C. NH’s interest in the Boston & Providence.
In our previous opinion we held that the liquidation value of NH’s inter[1062]*1062est in the Boston & Providence was $5.8 million, a figure readily derived by subtracting from the $8.1 million appraised net liquidation value of B & P’s physical assets the $2.3 million in cash that NH would have to pay for the publicly held stock of that company. 289 F.Supp. at 434-435. Despite the seeming simplicity and correctness of that calculation, the Commission on remand, while rejecting the bondholders’ contention that a much higher price was justified by current market quotations for a few CCBI’s,7 achieved a somewhat higher one of its own, 334 I.C.C. at 50-52. It started with the $2.3 million NH would concededly have to spend in buying v. the 20,000 publicly held shares of B & P. The Commission then assumed that 20,000 shares it said NH already owned must be worth the same amount. To the resulting $4.6 million it added $5 million for the value of outstanding B & P debentures owned by NH, resulting in a $9.6 million total. Since by hypothesis the value of the B & P securities could not be greater than the value of the physical assets underlying them, which had previously been set at $8.1 million, the Commission felt that the missing $1.5 million must be the value of the CCBI’s, which represented a potential that had not been taken into account in the asset valuation study because it was thought too speculative.
Although we do not question the conceptual soundness of taking market price of securities as a rough guide to asset value, it appears that the $2.3 million figure was not based on the market’s estimate of B & P liquidation value, but on the Commission’s rough guess at it. Boston & Providence R.R. Corp. Reorganization, 327 I.C.C. 10, 20-21 (1966). Furthermore, we cannot find any evidence that the debentures were worth $5 million. In fact, the United States, to which they were pledged during the 1950’s by NH as the sole security for loans that greatly exceeded their face value, agreed to exchange them for NH securities with a face value of $3.5 million, and this plan has been judicially approved as a fair compromise between the face amount of the B & P debentures and the well-supported view that because NH was in fact the primary obligor on the debentures, the obligation was extinguished as soon as NH acquired them in 1954. In re Boston & Providence R.R. Corp., 260 F.Supp. 415, 421-422 (D.Mass. 1966). It also appears that NH in fact did not own the 20,000 shares of B & P stock which the Commission claims for it, but only 16,000, with most of the missing 4,000 being held by the debenture sinking fund. Boston & Providence R.R. Corp. Reorganization, 327 I.C.C. 10, 13 (1966).
If we were to retain the Commission’s approach, while correcting the errors just noted, we would come out with a value for B & P very close to, but slightly below, the $8.1 million of the liquidation study. This study, however, was the result of detailed examination of the B & P properties, its findings have never been seriously disputed, and it would be irrational to make what amounts to a minor adjustment in its findings on the basis of assumed values for stock and securities that are far more speculative than the estimates they would be used to correct. We therefore adhere to our previous determination.
D. Deduction for accrued real estate taxes.
In our last decision we found the Commission had made a double deduction for $16.2 million of accrued real estate taxes on property required to be transferred free of liens by also including these in the assumed expenses of liquidation. 289 F.Supp. at 438. On remand the Commission decided that “the proper treatment of this item is to remove it from the liquidation expenses.” 334 I.C.C. at 40. However, the Commission increased liquidation value by only $14.6 million on the theory that under the NH reorganization plan such taxes are to be paid cf. by notes which the market would discount by 10 to 12 percent. 334 I.C.C. at 41. We see no sufficient [1063]*1063basis for this. There is no assurance that the holders of the tax liens will accept the notes offered in the NH reorganization plan, and they would seem to have little motive to do so now that NH no longer needs to conserve cash in order to keep operating a railroad.8
E. Net Additions and Betterments between December 31, 1966 and December 31, 1968.
The bondholders claim that the purchase price should be increased by net increases in NH’s property accounts since the valuation date, December 31, 1966. The Commission contends that the point was not raised before it; the bondholders deny this. We find it unnecessary to pass upon the issue since we are satisfied with the answer made by the Commission and PC on the merits, namely, that acceptance of the bondholders’ contention would require a complete revaluation of the property. Two examples will suffice to illustrate this. Many of the expenditures were improvements to the right of way which, however necessary for continued operation, might add nothing (or even be a negative factor) if the road were to be scrapped. Another is that if an upward allowance were to be made for net additions to the property accounts since December 31, 1966,9 a downward adjustment would have to be made for depreciation insofar as this would affect liquidation value. Since we have already sustained the Commission’s choice of a valuation date for the reasons given in Part II, we decline to order any adjustments inconsistent with that choice.
F. Failure to reduce discount to present worth to reflect price reductions.
The bondholders contend that the Commission reduced the estimated prices to be obtained for portions of NH’s property, but did not correspondingly reduce the discounts used in determining the present worth of what the NH estate would receive. The logic of this position is apparent, and we shall ask the Commission to take appropriate account of it in the decree to be submitted to us.
V. Loss Sharing
In our previous opinion we accepted “the Commission’s view that it was not bound to require Penn Central to assume all losses of NH subsequent to the Penn Central merger” and held that “the Commission was justified in setting a ceiling on Penn Central’s participation in losses and in providing that it bear only a fractional share.” 289 F.Supp. at 443-444. We disapproved its formula with respect to loss-sharing only because the downward progression of PC’s share in the second and third years might have, and indeed was intended to have, the effect of discouraging the bondholders from pressing legitimate claims, 289 F. Supp. at 444-445. The direction for inclusion of NH on December 31, 1968, eleven months after the PC merger, mooted the question of loss sharing for subsequent years, and the Commission adhered to the portion of its formula requiring PC to absorb all 1968 operating losses v. to a maximum of $5.5 million— or $5 million for the 11 months, 334 I.C.C. at 71-74. In light of the legal [1064]*1064views expressed in our last opinion, which we see no occasion for altering, this was well within the area confided to the Commission’s judgment. This is particularly true in light of the Commission’s decision to advance the burden of actual inclusion to December 31, 1968.
VI. The Consideration Payable by PC
The bondholders make two attacks on the valuation of the consideration payable by PC. One of these is of minor, the other of major significance.
In our previous opinion we recognized the force of the position of the Trustee of the First Mortgage securing the Harlem River Division bonds that these bonds should be accorded special treatment because of their fully secured nature, but considered this to be “a matter better left to the reorganization court.” 289 F.Supp. at 442. That court agreed with the view we had tentatively expressed. 289 F.Supp. at 464. The Commission accordingly directed PC to assume the bonds, and said that the value to NH of having this obligation removed would be $8.9 million. 334 I.C.C. at 70-71. The bondholders say that this figure was derived by simply adding the face value of the principal, which is $6.6 million payable January 1, 1973, to the face value of the accrued interest, which is $2.3 million payable in five equal annual installments between 1969 and 1973, without making any deduction for the delays in payment. We accept this criticism, and direct the Commission to modify its order by calculating an appropriate discount.
A much more important issue is the Commission’s adherence to a valuation of $87.50 per share for PC stock. Although we sustained this in our previous opinion, 289 F.Supp. at 436-437, the market’s evaluation of the stock has drastically changed for the worse since that time. Whereas we noted that while that opinion was being written, PC stock had advanced to $85 per share, and it reached a high of $86.50 on July 14, 1968, it had declined to $69.75 by August 13, 1968, when the reorganization court rendered its decision also sustaining the $87.50 figure, to $63.375 on the dates of the Remand Report and of actual inclusion, and still further thereafter. The decline during the last half of 1968 was the more significant in view of the fact that the Dow-Jones rail average on December 31, 1968, was at approximately the same level as on the date of our decision.
On the heárings on remand the bondholders called an expert witness who testified to a “current investment value” of $65 per share. In a footnote constituting its sole discussion of this important problem, the Commission noted that the witness “was unfamiliar with Penn Central’s operating and financial plans and condition; gave undue weight to extraordinary expenses of the past; and generally neglected the future prospects of the company.” 334 I.C.C. at 68 n. 40.
Our views on this subject parallel those recently expressed by the reorganization court. The NH trustees had asked for inclusion of NH in the PC System. It must have been expected from the outset that a large portion of the purchase price would consist of PC stock and that this would not be immediately sold but rather would be held by the reorganized NH as a major long-term participant in what was expected to be a great and successful railroad system.10 Spot fluctuations are therefore not decisive, especially when, as here, there are reasonable explanations for them, such as the unanticipated difficulties in working out the operating details of the merger and the prospect of PC’s having to shoulder NH losses at an earlier date and in larger amounts than had been anticipated. A reasonable time should thus be accorded to determine what the real value of PC stock is. On the other hand, [1065]*1065in the last analysis the market is the true appraiser and, if anticipations, no matter how well founded, should not be achieved, a proper adjustment should be made. We find Judge Anderson’s solution of this difficult problem to be admirable and direct that the order be modified as provided in his opinion.
We referred in our previous opinion to the “unfortunate” character of “the duplicitous system of review” by this court and the reorganization court. 289 F.Supp. at 424-425 & n. 3. Now this appears all the more unfortunate in light of the different conclusions the two courts have reached on important points. We regret our inability to agree fully with Judge Anderson, for whose views, reflecting years of devoted service to the administration of the NH estate, we entertain the greatest respect. We also regret our inability to accept positions forcefully expressed by our esteemed dissenting colleague. Essentially, we think our disagreements with Judge Anderson and Judge Weinfeld reflect a difference in view concerning how far we are at liberty to substitute our own notions for the decisions the Commission has taken in what we regard as a sincere effort to comply with the tasks both courts assigned it on remand. In any event we must decide the case according to our own lights; reconciliation of divergent views will come ultimately in the nation’s highest court, which alone can speak with authority.
The Commission is directed promptly to prepare and serve a proposed decree which, among other things, shall reflect in dollar amounts the changes directed in this opinion. Increases or decreases in the price shall be in approximately the same ratio as the figures for PC stock, PC bonds and cash in the Remand Report, 334 I.C.C. at 71, bear to each other. Other parties may serve counter-proposals within ten days thereafter, and a further five days will be allowed for comments on such counter-proposals.