OPINION

KUNZIG, Judge.

This income tax refund case comes before the court on appeal from the Trial Division where findings and an opinion were filed November 9, 1973, by Trial Judge David Schwartz, pursuant to Rule 134(h). The court has reviewed his decision on the briefs, exceptions, and oral argument of counsel, finds itself in agreement with major portions of that recommended decision and adopts them, with minor modification, as Parts I-II and IV-X of its opinion. The court also adopts most of the trial judge's findings of fact, but modifies portions deemed proper upon consideration of exceptions by the parties. We limit our departure essentially to Parts III and XI of this opinion.

In Part I we deal with plaintiff's claim that it had a right to "expense" certain property costing less than $500 ("minimum rule" property) in its 1942 tax return. In Part II we consider the 1942 deductibility of plaintiff's payroll taxes paid in 1943. In Part III we treat the issue of whether plaintiff may return $13 million to income based upon 1900-1907 accounting errors. In Part IV we handle plaintiff's assertion that certain stock subscription rights received could be included in income. In Part V we resolve plaintiff's request to include stock in certain leased line subsidiaries in its capital assets. In Part VI we answer plaintiff's demand for interest on a 1948 agreement with the Internal Revenue Service. In Part VII we respond to plaintiff's plea to include certain land sale proceeds in its earnings and profits. In Part VIII we decide defendant's "additional defense" that plaintiff erroneously included bond discounts and expenses in its total assets. In Part IX we rule on another additional defense that plaintiff failed to treat other bond discounts, and call premiums as interest. In Part X we pass on plaintiff's entitlement to include certain donations and grants in its equity. Finally, in Part XI we determine the correct amount of plaintiff's equity invested capital based on the value of its stock issued for its operating assets.

We find that plaintiff is entitled to recover in Parts I, V and in some sections of Part X. Plaintiff does not prevail in Parts II, IV, VI and VII. We hold for defendant on its offset claims in Parts VIII, IX and XI. Finally, we remand the issue in Part III for reconsideration by the trial judge.

The Union Pacific Railroad Company brings suit pursuant to the Tucker Act ( 28 U.S.C. § 1491) and the Internal Revenue Code ( § 7422(a), 1954 Code) for a refund of income and excess profits taxes for 1942. Plaintiff has paid over $30 million in income tax and over $7 million in excess profits tax. It seeks a judgment for $13,409,961.46 (consisting of $9,222,801.78 in statutory interest for 1942 and $4,187,159.68 in income and excess profits tax and assessed interest thereon paid for 1942) or such other amount which may be legally refundable for 1942, together with statutory interest thereon.

The extraordinary span of time from 1942 to the filing of suit is accounted for by the following. The taxpayer filed returns in 1943 and made payments of a total of some $42 million in 1943, 1944 and 1946. Between that time and the conclusion of the audit of the returns in 1957, various timely claims for refunds were filed in amounts varying from $7.7 million to $348,000, and various credits, adjustments and refunds were made in amounts ranging from $694,000 to $196.

The audit, begun in 1945, was completed in 1957. One Internal Revenue agent spent over 5,000 hours, or about 630 working days, on the report, almost 1,000 pages long. The major part of the work was the required preparation of the surplus or accumulated earnings and profits accounts for 25 closely related or subsidiary corporations for the 44 years from 1898, the year of the creation of the taxpayer on the reorganization of the old Union Pacific Railroad. These accounts were directly relevant to the invested capital method chosen by plaintiff for reporting its tax under the World War II excess profits tax act (Title II, § 201, Second Revenue Act of 1940, 54 Stat. 975, 26 U.S.C. § 710-752 (1940 ed.)). The primary issue is the valuation of the original Union Pacific Railroad system. This and other major issues required the study of voluminous documents and underlying data located at various points in the country. Various other factors contributing to the length of the audit period included the pendency until 1954 of suits, brought by the taxpayer for earlier tax years, affecting some of the issues involved in 1942 tax liability.

The returns were meantime kept open by consents on the part of the taxpayer. Plaintiff did not press for an early submission of the agent's report, for reasons connected with a certain agreement with the Commissioner concerning freight "cutbacks" or rate refunds, discussed below. It is agreed that the consents were voluntary and without any pressure or coercion and that at no time did plaintiff complain to the Internal Revenue Service concerning the time required to complete the audit of its returns.

The audit, completed in 1957, was the basis of a determination in 1959 of a large net overassessment of tax for 1942. On September 16, 1960, after approval of the determination by the Joint Congressional Committee on Internal Revenue Taxation, a deficiency in excess profits tax for 1942 was satisfied by credits in income tax and a post-war credit, and a total of $7,793,219.55 was refunded or paid to plaintiff, consisting of income tax and declared value excess profits tax for 1942 and statutory interest thereon.

In 1961 plaintiff filed a timely comprehensive claim for refund of $13,409,961.46, together with interest, and on notice of disallowance, the instant suit was timely brought, on September 14, 1962. The petition as first filed contained 27 counts, to which seven counts were added by an amended petition in 1966. In an answer and four amended answers filed in 1966 and 1968 the Government pleaded 22 separate additional defenses.

Aspects of the case have finally been disposed of as follows. On January 19, 1968, the court pursuant to the Government's additional defense 11 dismissed counts 27 through 30 and 32 through 34 as claims for refund, allowing them to remain only as offsets to defendant's set-offs. 389 F.2d 437, 182 Ct.Cl. 103 (1968), cert. denied, 403 U.S. 931, 91 S.Ct. 2254, 29 L.Ed.2d 710 (1971). Summary judgment dismissing counts 29 and 30 was granted in 1968, thus rendering additional defenses 14 and 15 moot. 401 F.2d 778, 185 Ct.Cl. 393 (1968), cert. denied, 395 U.S. 944, 89 S.Ct. 2017, 23 L.Ed.2d 462 (1969), motion for reconsideration denied, 194 Ct.Cl. 1021. Counts 7 and 31 and the additional defenses thereto, Nos. 2, 3, 4 and 10, were ordered separately tried, after final adjudication of the remaining issues, by then Trial Commissioner (now Judge of the U.S. Customs Court) Herbert N. Maletz.

The parties have through their able and diligent counsel engaged in extensive and fruitful pretrial proceedings and negotiations. A substantial number of counts and affirmative defenses were in the course of these proceedings conceded by one or the other party, by agreements limited to this proceeding.

While counts and additional defenses which have been conceded, by agreements limited to this proceeding, are these: it is conceded that plaintiff is entitled to prevail on counts 1 and 16; it is conceded that plaintiff is not entitled to prevail on counts 3, 10 through 14, and 27 (thus rendering additional defense 5 moot); and it is conceded that the defendant is not entitled to prevail on additional defenses 12 and 13. Count 28 is entirely disposed of by agreement. A number of partial concessions of counts were also made as to counts 17, 20, 21 and 22 and additional defense 20. Various items in counts 19-25 not agreed upon by the parties are disposed of in other counts.

The pretrial proceedings culminated in a stipulation of facts of approximately 500 pages, over 500 exhibits agreed by the parties to be received in evidence or ruled upon in advance of trial and an exchange before trial of the direct testimony of the experts to be called by the parties on the valuation issue. The case was tried between February 26 and March 6, 1969. Testimony consisted of the cross and redirect examination of the expert witnesses, and testimony by three additional witnesses. Proposed findings of fact, objections to proposed findings of fact and briefs were filed between October 13, 1969 and January 5, 1971, and further memoranda were filed in May and June, 1973.

I. "MINIMUM RULE" ACQUISITIONS

The Interstate Commerce Commission's rules for accounting by railroads, effective in 1940-1942, required that items of road and equipment property costing less than $500 be charged to operating expenses rather than to a capital account. Such a rule, known as the "minimum rule," had been in effect for many years; in 1940 the break point was raised from $100 to $500. The Government challenges the effectiveness for tax purposes of the change in the rule. The contention is that (1) items of property costing between $100 and $500, concededly of a capital nature and having a useful life of longer than a year, are "permanent improvements" under section 24(a)(2) of the 1939 Code, must be capitalized, and only depreciation deducted, and (2) the minimum rule does not constitute a method of accounting under section 41 of the Code. Plaintiff deducted $113,717.84 for such items, and it is this deduction which is in dispute in count 2.

A similar case has been considered by the court and decided in favor of the taxpayer in Cincinnati, N. O. Tex. Pac. Ry. v. United States, 424 F.2d 563, 191 Ct.Cl. 572 (1970). The minimum rule items, the court held, "are not of such nature or character in relation to the plaintiff's business to constitute permanent improvements or betterments as is contemplated by section 24(a)(2)"; further, that the minimum rule treatment of the items was "in accordance with generally accepted accounting principles and is not such that it inhibits the ability of plaintiff's financial statements to clearly reflect income for tax purposes"; and, finally, that "the minimum rule constitutes a method of accounting as contemplated by section 41 and Treas. Reg. 111, § 29.41-3." 424 F.2d at 572-73, 191 Ct.Cl. at 587-588.

Defendant seeks to distinguish the decision on the ground that the tax consequences there were de minimis while here they are substantial. Substantiality is attempted to be shown by pointing to such facts as the excess of the deductions taken by plaintiff pursuant to the minimum rule over the depreciation allowed by the Commissioner, amounting to $109,926 in 1942, $147,017 in 1943, $175,682 in 1944, $171,850 in 1945 and $115,551 in 1946. The total of $720,026 for the 5 years is said to be substantial.

Dollar figures, even large ones, are not a showing of substantial tax consequences in a case of this type. Whether an accounting method distorts the reflection of income must depend on a whole picture. As the court said in Cincinnati, N. O. Tex. Pac. Ry. v. United States, supra, "[t]he most convincing evidence that the Commissioner has abused his discretion in prohibiting the plaintiff from treating items [in accordance with the minimum rule] . . . is the statistical analysis . . . which indicates the relationship of the quantum of minimum rule expenses to other substantial income and balance sheet figures." 424 F.2d at 571, 191 Ct.Cl. at 584.

Let us therefore compare the relationships in the instant case with those held by the court not to inhibit the ability of the Cincinnati's financial statements to reflect its income clearly. In 1942 the Cincinnati's total operating expense was $16,291,053; total investment account was $69,391,628; and challenged minimum rule items were $9,688. For plaintiff the comparable figures were $218,307,770, $442,726,752 and $113,718. The ratio of minimum rule items to total investment account for the Cincinnati was thus .00014; for plaintiff it is .00026. The ratio of minimum rule items to total operating expense for the Cincinnati was .00059; for plaintiff it is .00052.

The relationships are, therefore, as minimal in plaintiff's fiscal picture as they were in Cincinnati's. There is accordingly no reason not to follow the court's recent decision and affirm the right of the plaintiff to account for the items in question in 1940-1942, pursuant to the ICC's minimum rule.

Plaintiff is entitled to prevail on count 2.

II. PAYROLL TAXES

The dispute, raised by count 4, is over whether the plaintiff, having been permitted by the Commissioner of Internal Revenue to deduct from its 1942 income, as a business expense, the amount of vacation pay earned by plaintiff's employees in that year and paid to them in 1943, should also be permitted to deduct the payroll taxes applicable to the vacation pay, which were payable and paid in 1943. Plaintiff is on the accrual and calendar year basis. The payroll taxes were those imposed by section 1520 of the Internal Revenue Code of 1939, 26 U.S.C. § 1520 (1940) and Section 8(a) of the Railroad Unemployment Insurance Act, ch. 680, 52 Stat. 1094, 1102 (1938), 45 U.S.C. § 358 (1940).

Plaintiff's contentions were apparently not reflected in its tax returns as first filed. In its return for 1942, plaintiff deducted, as business expenses, $1,103,413.31 in vacation pay earned by its employees in 1941 and paid to them in 1942, and payroll taxes on the vacation pay, in the amount of $64,108.31, paid in 1942. Consistently, on its return for 1943 plaintiff deducted $1,518,624.02 in vacation pay earned by employees in 1942 and paid to them in 1943, and the applicable payroll taxes of $90,358.13, paid in 1943.

During the audit of its 1942 return, the plaintiff claimed, and the Commissioner allowed, a deduction of the entire amount of the vacation pay earned by its employees in 1942, $1,518,624.02. Relying on the allowance, the plaintiff then made a claim for refund on the ground that it was entitled in 1942 to accrue and deduct the payroll taxes on the entire amount of vacation pay whose deduction had been allowed. The claim was actually one for $26,249.82 the difference between $90,358.13, the amount of the payroll tax claimed as a deduction, and the deduction of $64,108.31 taken on the return. The Commissioner denied the claim and count 4 of the petition seeks the $26,249.82 involved.

The question for decision is whether plaintiff may in 1942 accrue and deduct payroll taxes to be paid in 1943 on vacation pay which was earned, accrued and allowed to be deducted in 1942, though not to be paid until 1943. The answer here given is, no. The reasons follow.

A tax not assessed and paid until a following year may nevertheless be deducted in the prior year, if it meets the "all events" test of United States v. Anderson, 269 U.S. 422, 46 S.Ct. 131, 70 L.Ed. 347 (1926). By that test a deduction may be taken, in advance of assessment, in the year when all the events take place determining liability and fixing the amount of the tax. Thus ( id. at 441, 46 S.Ct. at 134):

In a technical legal sense it may be argued that a tax does not accrue until it has been assessed and becomes due; but it is also true that in advance of the assessment of a tax, all the events may occur which fix the amount of the tax and determine the liability of the taxpayer to pay it. In this respect, for purposes of accounting and of ascertaining true income for a given accounting period, the munitions tax here in question did not stand on any different footing than other accrued expenses appearing on appellee's books.

The "all events" test has been restated and applied many times, often in cases similar to the instant case. E. g., United States v. Consolidated Edison Co., 366 U.S. 380, 385, n. 5, 81 S.Ct. 1326, 6 L.Ed.2d 356 (1961); Clevite Corp. v. United States, 386 F.2d 841, 843, 181 Ct.Cl. 652, 658 (1967); Denver Rio Grande Western Railroad Co. v. Commissioner, 38 T.C. 557, 572 (1962); Turtle Wax, Inc. v. Commissioner, 43 T.C. 460, 466-67 (1965). The test is phrased in the current income tax regulations as follows: "Under an accrual method of accounting, an expense is deductible for the taxable year in which all the events have occurred which determine the fact of the liability and the amount thereof can be determined with reasonable accuracy." Treas. Reg. § 1.461-1(a)(2) (1970), promulgated by T.D. 6282, 1958-1 Cum. Bull. 215, 22 F.R. 10686, Dec. 25, 1957, 26 C.F.R. § 1.461-1(a)(2) (1970).

So long as a liability remains contingent or if the liability has attached but the amount cannot be reasonably estimated, a business expense deduction is not allowed. Treas. Reg. § 1.461-1(a)(2), supra; Clevite Corp. v. United States, supra; Texaco-Cities Service Pipe Line Co. v. United States, 170 F. Supp. 644, 645, 145 Ct.Cl. 274 (1959); Denver Rio Grande Western Railroad Co. v. Commissioner, supra; Turtle Wax, Inc. v. Commissioner, supra.

In the instant case, both the fact of liability and the amount of tax were as of December 31, 1942 still in doubt. Uncertainty as to two events as of that date made it impossible then to determine the tax.

One of the facts lacking was knowledge of the total vacation pay which plaintiff's employees would actually receive. Under at least some of plaintiff's labor contracts, the employee forfeited the right to a vacation with pay or to pay in lieu of a vacation, if his employment were terminated, for a reason other than retirement, prior to the time scheduled for his vacation in 1943. Since it could not be known by the end of December of the prior year, 1942, which of plaintiff's employees would remain in its employ until the beginning of their respective vacations in 1943, it could not in 1942 be determined how much vacation pay plaintiff would be required to pay and, therefore, what would be plaintiff's tax liability. The uncertainty would not be resolved until the time in 1943 of the actual payment of vacation wages or allowances. All events fixing liability not yet having occurred, a deduction in advance, in 1942, is not permitted, under the foregoing authorities, and particularly Texaco-Cities Service Pipe Line Co. v. United States, supra, and Turtle Wax, Inc. v. Commissioner, supra. Compare similar rulings with respect to the excise tax on payrolls of employers other than carriers. G.C.M. 19692, 1938-1 Cum.Bull. 148; Rev.Rul. 69-587, 1969-2 Cum.Bull. 108.

Plaintiff urges that in allowing it to deduct vacation pay earned in 1942, the Commissioner has ruled, pursuant to I.T. 3956, 1949-1 Cum.Bull. 78, that the liability for vacation pay is not made contingent for purposes of the "all events" test by the possibility that the employee might forfeit his right to a vacation by leaving the employ prior to his scheduled vacation. The Government's position on I.T. 3956 is that the ruling was mistaken and has been revoked in Rev.Rul. 54-608, 1954-2 Cum.Bull. 8, 9-10; moreover, that it concerned vacation pay and should not be extended to payroll taxes.

In revoking I.T. 3956 the Commissioner ruled "that no accrual of vacation pay can take place until the fact of liability to a specific person has been clearly established and the amount of the liability to each individual is capable of computation with reasonable accuracy." Rev. Rul. 54-608, supra. In reaching this decision, the Commissioner relied on three decisions of the Tax Court: Tennessee Consolidated Coal Co. v. Commissioner, 15 T.C. 424 (1950); Morrisdale Coal Mining Co. v. Commissioner, 19 T.C. 208 (1952); and E. H. Sheldon Co. v. Commissioner, 19 T.C. 481 (1952). In these cases it had been held that liability for payment of vacation pay depended on the condition precedent that the recipient employee be working for the employer-taxpayer on the date required by the contract, and that until that date liability remained uncertain.

The reasoning of Rev.Rul. 54-608 and of the cases it relied upon is preferred over that of I.T. 3956. This court has recently ruled that where "payment of vacation pay * * * is contingent upon employment up to the beginning of the vacation period, a taxpayer cannot accrue expenses for vacation pay before the taxable year in which the payments are made. Until the vacation period begins, the `all events' test * * * has not been satisfied." Clevite Corp. v. United States, supra.

Plaintiff urges that the effective date of Rev.Rul. 54-608 has been repeatedly postponed (most recently in Section 903 of the Tax Reform Act of 1969, P.L. 91-172, 83 Stat. 487, 711), and thus that I.T. 3956 is "still the law." The question at hand, however, is not the present effectiveness of I.T. 3956, as governing deductibility of vacation pay by certain classes of taxpayers, but whether its rationale should be extended to the deductibility of payroll taxes on vacation pay. That question is here answered in the negative.

The nature of the payroll tax itself is the source of the second "event" or fact so unknown or uncertain at the close of 1942 as to make the liability contingent and thereby prevent deduction of payroll taxes in that year. The two payroll tax acts (notes 2 and 3, supra) subject to tax only the first $300 of compensation paid to an employee in any calendar month. Under plaintiff's labor contracts, in the event plaintiff could not release an employee for a vacation in 1943, it was obligated to pay the employee an allowance in lieu of the vacation. Where such an allowance was paid in a calendar month in which the employee had already received $300 or more in other compensation, no tax would be due on the excess of $300.

Plaintiff did not know, as of December 31, 1942, which of its employees it would and which it would not be able to release for a vacation in 1943, and to which, therefore, it would pay allowances in lieu of vacation. Unknown, therefore, was how many would receive more than $300, in combined regular pay and vacation allowance, in one calendar month, and thus to what extend the payment of vacation allowances would be free of tax. Liability for the tax was necessarily contingent, until the time of the scheduled vacation in 1943. Only then would it appear how much of the payments to the employee would be liable to tax.

The effect of the $300 maximum on a taxable monthly compensation is confirmed by the facts of plaintiff's actual payments of vacation pay and tax. Plaintiff paid $1,103,413 in vacation pay in 1942. Six percent of this amount — the tax rate in 1942 (notes 2 and 3, supra) — is $66,205, yet plaintiff paid payroll taxes of only $64,108. In 1943 plaintiff paid vacation pay of $1,518,624. At 6.25 percent, the tax rate in 1943 (notes 2 and 3, supra), it would have paid $94,914, yet plaintiff paid only $90,358. The differences between the amount of the tax payable, if the entire amount of vacation pay were taxable, and the amount of tax actually paid, not explained by plaintiff, can only be attributed to payments of regular pay and vacation allowances in total amounts of over $300 in a month, of which the portion over $300 was free of tax.

Plaintiff contends that the effect of the tax freedom for compensation over $300 in one month is not so great as to disqualify the tax from deductibility because, it is said, plaintiff was able, in 1942, to estimate the amount of tax to be paid with the accuracy required by the "all events" test. Not so. While the amount of a tax may be reasonably estimated and need not be precisely known, liability for the tax must have attached and cannot be approximated or estimated. "[T]he fact that the percentage of items which will be paid can be estimated with reasonable accuracy is not sufficient to support accruals. The individual items must represent fixed liabilities." Denver Rio Grande Western Railroad Co. v. Commissioner, supra. The accruability test is not whether there is certainty of payment, or if a reasonable estimate can be made, but whether there is certainty of liability. Trans-California Oil Co., Ltd., 37 B.T.A. 119, 127 (1938). Here, liability was not certain or fixed. The effect of the $300 provision was to make the liability for tax uncertain, until the time when the employee actually went on vacation or received both pay and vacation pay. Liability for payroll tax was thus under the "all events" test not certain or fixed in 1942. Texaco-Cities Service Pipe Line Co. v. United States, supra; Helvering v. Russian Finance Construction Corporation, 77 F.2d 324, 327 (2d Cir. 1935).

Lastly, plaintiff urges that since it has been allowed a deduction for vacation pay earned in 1942, a comparable deduction for the tax on such vacation pay is appropriate or necessary in order clearly to reflect its income for 1942. The effect on income in any year of the prospective payroll tax may be clearly enough reflected by a reserve for the tax as a contingent liability. The need for such a reserve is not the equivalent of a right to a deduction. Lucas v. American Code Co., 280 U.S. 445, 452, 50 S.Ct. 202, 74 L.Ed. 538 (1930).

Plaintiff is not entitled to prevail on count 4.

III. $13 MILLION DEDUCTIONS

Plaintiff's count 8 also seeks a refund of excess profits taxes based upon an alleged series of accounting errors in 1900-1907. It asserts that certain "betterments and improvements" during these years were "expensed" rather than debited to its investment account. It further contends that in later years, these "betterments" were depreciated and thus "expensed" for a second time. Plaintiff wants to "reverse" this $13 million accounting error by adding it back into accumulated earnings and profits.

The trial judge rejected this claim. He found insufficient proof that the accounting transactions were erroneous, and noted that plaintiff had failed to succeed on this same argument in a case before the Interstate Commerce Commission. 44 ICC Val.Rep. 1, 22 (1933).

The trial judge discredited plaintiff's contention that later depreciation deductions created a "double expensing" of a single item of property. He found that the evidence shows that plaintiff in fact made "substantial unrecorded retirements" rather than later retirement deductions. It is this point which troubles the court. The "unrecorded retirement" issues are still before the trial judge. Plaintiff contends that if the court denies its "$13 million" claim for the sole reason that these accounting entries were made to correct earlier "retirement errors," and if it is later found that its "retirement" practices were correct, plaintiff will have lost the right to alter the "$13 million" entries.

Since the retirement issue is still before the trial judge, the court remands this "$13 million" issue to the trial judge for clarification of the relationship between the retirement issue and the "$13 million" issue. Although it appears that the trial judge rejected plaintiff's claims solely for insufficient proof of a "$13 million" error, we wish to make absolutely certain that his conclusion is not tied to any findings of "erroneous retirement" practices by plaintiff, or that plaintiff was not lured into failure to present proof on the "$13 million" issue because it believed the issue would be tried with the retirement issues.

IV. STOCK SUBSCRIPTION RIGHTS

In each of the years 1922, 1923, and 1925 plaintiff and its wholly owned subsidiary, the Oregon Short Line, both holders of common stock in the Illinois Central Railroad Company, received, as such holders, a distribution of rights to subscribe to Illinois Central convertible preferred stock at $100 per share. The rights were exercised in the year received. At both the time of distribution and exercise, the market value of the stock was higher than the subscription price. The lower of the two aggregate "spreads" between market and distribution prices was some $670,000.

The question presented, determinative in count 15, and determinative in part in other counts, is whether the "spread" at either time was income properly to be included in accumulated earnings and profits of the recipient for excess-profits-tax purposes. The answer, here given in the negative, hinges on the checkered history of the taxation of stock dividends prior to the enactment of the excess profits tax in 1940.

That history begins with the 1918 decision holding the 1913 tax on income inapplicable to a dividend in stock, on the ground that the "proportional interests of each shareholder remains the same" after the receipt of the dividend. Towne v. Eisner, 245 U.S. 418, 426, 38 S.Ct. 158, 159, 62 L.Ed. 372. In 1916, however, Congress had by statute directed that a "stock dividend shall be considered income, to the amount of its cash value." Section 2(a), Revenue Act of 1916, ch. 463, 39 Stat. 757 (1916). This statute the Supreme Court soon held in violation of the Sixteenth Amendment, in Eisner v. Macomber, 252 U.S. 189, 40 S.Ct. 189, 64 L.Ed. 521 (1920), on the ground that no income had been received. Though the two cases had involved simple dividends of common stock to common stockholders, Congress took the decisions as forbidding the taxation as income of any stock dividends, and accordingly provided in the 1921 Act and thereafter, through the 1934 Act, that a "stock dividend shall not be subject to tax."

The premise of this statutory exemption of stock dividends from taxation was upset in 1936 when the Supreme Court decided in Koshland v. Helvering, 298 U.S. 441, 56 S.Ct. 767, 80 L.Ed. 1268, that a dividend in common stock to holders of preferred stock gave rise to income (though not subjected to tax), because the resulting interest of the stockholder was different than before. Now appreciating that Eisner v. Macomber was not an absolute, and that some stock dividends could be taxable, Congress promptly provided, in the 1936 Act, that dividends in stock or in rights should be taxable to the extent constitutionally permissible. Such taxation as was thereby imposed was to be prospective. These provisions were repeated in the 1939 Code. Under these provisions, such stock dividends as gave rise to income, that is, those that created interests in the recipient different than before, were taxed. See Helvering v. Griffiths, 318 U.S. 371, 63 S.Ct. 636, 87 L.Ed. 843 (1943); Bittker Eustice, Federal Income Taxation of Corporations and Shareholders, Sec. 5.60 (2d ed.).

Such briefly had been the prior treatment of stock dividends when the excess profits tax was being considered in 1940. Though the change in the taxation of stock dividends in 1936 did not rake up stock dividends of earlier years, the excess profits tax would do so, by its provision that accumulated earnings and profits be an element of equity invested capital. 26 U.S.C. § 718 (1940 ed.). Computation of equity invested capital would require a review and determination of the earnings and profits account of at least some corporate taxpayers from the beginning (one of the causes of the long time spent in auditing the return of the instant plaintiff). On such a review, in the absence of special statutory provision, at least some long-past stock dividends would now by hindsight be understood as having given rise to income and thus includable in earnings and profits, though the dividend had been exempt from tax under the statutes in force between 1921 and 1936.

Special statutory provision was, however, made. Congress dealt explicitly with the effect upon earnings and profits of past, untaxed stock dividends. The draftsmen added to the excess profits tax law a provision that earnings and profits should not be increased by receipt of a dividend, not taxed, whose only effect had been to cause a reallocation of the basis of the old stock to the old and the new stock. This section, quoted in the note, was section 115( l), 1939 Code, 26 U.S.C. § 115(a)(1) (1940 ed.), added by § 501, Second Revenue Act of 1940, ch. 757, 54 Stat. 1004 (1940). Retroactivity was explicit. Section 501(c), Second Revenue Act of 1940, ch. 757, 54 Stat. 1005.

The precise question for present decision is simply whether the distributions of rights to plaintiff and Oregon Short Line were tax-free dividends subject to the bar of § 115( l) to their inclusion in the earnings and profits of the recipient corporation.

The issues are those of construction of § 115( l) and related sections. A first issue is whether the distribution of rights to subscribe to the preferred stock of Illinois Central, the distributing corporation, was a "dividend." It is agreed, incidentally, that Illinois Central had sufficient income available for dividends in the amounts involved.

Distribution of rights to subscribe to stock in the issuer, such as are presently involved, are for tax purposes the equivalent of stock dividends. It is settled that a distribution of rights to subscribe to stock is governed by the same rules as determine taxability of the stock itself, had it been directly distributed. Miles v. Safe Deposit Co., 259 U.S. 247, 42 S.Ct. 483, 66 L.Ed. 923 (1922); Choate v. Commissioner, 129 F.2d 684 (2d Cir., 1942); Charles M. Cooke, Ltd. v. Commissioner, 2 T.C. 147 (1943). Thus a distribution of rights to subscribe is not subject to tax as a dividend, when the stock itself, had it been distributed directly, would not be subject to tax, whether because of the nature of the stock dividend or because of a statutory exemption from tax. Miles v. Safe Deposit Co., supra; Charles M. Cooke, Ltd. v. Commissioner, supra. And a distribution of rights is taxable when distribution of the stock would be taxable, as it was under the 1936 Act, which taxed stock dividends to the extend constitutionally permissible. Choate v. Commissioner, supra.

Why, then, was the distribution of rights not a dividend? Plaintiff claims that the transactions in which the rights were distributed were not distributions of dividends but offers to sell corporate property to the corporation's stockholders at less than its value, which on acceptance by the exercise of the rights gave rise to income in the amount of the lesser of the spreads between market and subscription price at the times of distribution and exercise.

Palmer v. Commissioner, 302 U.S. 63, 58 S.Ct. 67, 82 L.Ed. 50 (1937) and Commissioner v. Gordon, 391 U.S. 83, 88 S.Ct. 1517, 20 L.Ed.2d 448 (1968), cited by plaintiff, do support the proposition for which they are invoked — that the sale of corporate property to stockholders at less than its value is as much a distribution of profits subject to tax as income as the formal declaration of a dividend in money. In these cases the property distributed to stockholders was stock in a corporation other than the distributing corporation. Such stock is when distributed as a dividend no differently treated than is other property. Dividends consisting of stock in the issuing corporation are, however, another matter, subject, as has been seen, to special statutory treatment. The rule governing dividends by offer to sell property is therefore not relevant to the problem at hand. The aspect of that rule, much emphasized by plaintiff, fixing the realization of income at the time of acceptance of the offer as against the time of distribution of the right to buy is as irrelevant as the rule itself.

Accordingly, the criticisms of treatment of the instant distribution as a dividend are invalid. The distributions are dividends within the meaning of § 115( l), if the section is otherwise applicable. The remaining questions have to do with the two conditions for its application stated in § 115( l) (note 8, supra).

The first of these is whether the distributions involved were free from tax, for § 115( l) by its terms governs only a distribution "which (under the law applicable to the year in which the distribution was made) was not a taxable dividend." Note 9, supra.

The distributions, made in 1922, 1923 and 1925, were in fact not taxed. In the years in question, petitioner and Oregon Short Line, in their consolidated returns, did not include in gross income any amounts as attributable to the receipt of the rights in question. The omission cannot, however, be regarded as a conclusive recognition by the taxpayer that the distributions were not taxable income, for in those years 100 percent of dividends received from domestic corporations were deductible. Section 234(a)(6)(A), Revenue Act of 1921, ch. 136, 42 Stat. 255 (1921); § 234(a)(6)(A), Revenue Act of 1924, ch. 234, 43 Stat. 283 (1924). The text of the applicable statute is, however, clear enough. The law applicable to the distributions is the law in force in the years they took place, 1922, 1923 and 1925. The law in force in each of those years provided that "A stock dividend shall not be subject to tax." Section 201(d) of the 1921 Act and § 201(f) of the 1924 Act, note 5, supra. This explicit exemption from taxation fully satisfies the condition of § 115( l) that the distributions have been tax-free under the law in force at the time they were made.

To dispute this conclusion, plaintiff relies upon Choate v. Commissioner, supra, as holding that the distributions were taxable as income. The reliance is misplaced. Choate was a decision under § 115(f) of the 1936 Act (note 6, supra). Stock dividends were taxed, as far as constitutionally permissible, both by that Act and by its successor law, the 1939 Code (note 6, supra). Both statutes, however, were applicable only prospectively (notes 6, 7, supra), and so neither can govern in determining whether the distributions in the 1920's were tax-free.

How clearly Choate depends on the prospective change made in the law by § 115(f) appears from this excerpt from the opinion ( 129 F.2d at 688):

In Miles v. Safe Deposit Trust Co., 259 U.S. 247, 42 S.Ct. 483, 66 L.Ed. 923, * * * it was said that rights issued to its common stockholders, to subscribe to a company's unissued common stock, are analogous to stock dividends. Such stock dividends were not constitutionally taxable under Eisner v. Macomber, 252 U.S. 189, 40 S.Ct. 189, 64 L.Ed. 521 * * *. But under § 115(f) stock dividends are now taxable so far as such a tax is constitutional, and so are rights to the extent that they are dividends. A stock dividend in preferred stock issued to common stockholders is, therefore, now subject to a valid tax.

The second condition for the application of § 115( l) is that the distributions have had the effect only of reallocating the basis of the stock originally held, as between the old stock and the new stock received in the distribution. Before 1936, when stock dividends were not taxed in the belief they were not constitutionally taxable, they were treated by the Treasury as having the effect only of a reallocation of basis as between the old and new stock. The practice was thereafter codified in § 214(e) of the Revenue Act of 1939, ch. 247, 53 Stat. 874 (1939).

The authority for the proposition that the section is a codification of prior Treasury practice is no less than the House Committee which wrote the section. In reporting with approval what became § 214(e), the Committee said (H.R. Rep. No. 2894, 76th Cong., 3d Sess. 42-43 (1940)):

Tax-free distributions in stock or in rights, whether or not constituting income within the meaning of the sixteenth amendment or exempt to the distributee under section 115(f) of the Revenue Act of 1934 or a corresponding provision of a prior Revenue Act, and tax-free distributions of stock or securities in a corporation a party to a reorganization, have consistently been treated by the Treasury as not resulting upon receipt in an increase in earnings or profits, but as causing the basis of the stock in respect of which the distribution was made to be allocated between such stock and the stock securities received, with the result that earnings or profits are increased, upon the sale of such stock or property, by the entire amount of the recognized gain computed upon the basis so determined by allocation. * * [The section] explicitly states the rules heretofore applied by the Treasury.

Section 214(e), plaintiff says, cannot be applied, for it is by its terms limited to distributions which are not income (and, plaintiff would go on to say, the distributions here did create income). The portion of the section (note 12, supra) relied upon is this: "if the new stock was acquired in a taxable year beginning before January 1, 1936, or acquired in a taxable year beginning after December 31, 1935, and its distribution did not constitute income to the shareholder within the meaning of the Sixteenth Amendment to the Constitution." The particular words invoked are: "and its distribution did not constitute income."

Plaintiff would read the last clause, beginning with "and," as applicable to acquisitions both "before January 1, 1936" and "after December 31, 1935." Such a reading is erroneous in that it overlooks the disjunctive effect of the "or" which insulates the condition "if the new stock was acquired in a taxable year beginning before January 1, 1936" from the remaining words, including the "and" clause, and leaves the category of acquisitions "before January 1, 1936" unaffected by the conditions placed on the acquisitions "after December 31, 1935." The section embodies a purposeful differentiation, which plaintiff would ignore, between the two stated classes of transfers — 1935 and earlier, and 1936 and later. Plaintiff's view, were it accepted, would reduce the clause to a great many unnecessary words and phrases.

The words should rather be taken as if they were punctuated as emphasized in the following: "if the new stock was ( i) acquired in a taxable year beginning before January 1, 1936, or ( ii) acquired in a taxable year beginning after December 31, 1935, and its distribution did not constitute income to the shareholder within the meaning of the Sixteenth Amendment to the Constitution." Punctuation of this type appears in a successor statute, § 113(a)(19)(A) of the 1939 Code, 26 U.S.C. § 113(a)(19)(A), added by § 214(a) of the Revenue Act of 1939, ch. 247, 53 Stat. 872 (1939). Properly read, the language of the section says that an acquisition before January 1, 1936, alone and without regard to the "and" clause, fulfills the condition for applicability of the section. The "and" clause is applicable only to the words, following the word "or," dealing with acquisitions after December 31, 1935.

Section 214(e), thus being applicable to the acquisitions in 1922, 1923 and 1925, serves to satisfy the second and last disputed condition of § 115( l) — that the dividend have an effect only on allocation of basis.

Accordingly, § 115( l) is operative, and directly forbids the inclusion in a recipient's earnings and profits of the value of the rights distributed in 1922, 1923 and 1925. Plaintiff is not entitled to recover on this issue in count 15 and in all of the counts in which it is raised.

V. LEASED LINE SUBSIDIARIES

Plaintiff owns all of the stock of the Oregon Short Line Railroad Company, approximately 99 percent of The St. Joseph Grand Island Railroad Company and (together with Oregon) all of the stock of the Los Angeles Salt Lake Railroad Company. A large part of the stockholdings was acquired soon after plaintiff's reorganization in 1898 and most of it has been owned by plaintiff for many years. The Oregon Short Line was acquired in exchange for shares in plaintiff whose valuation is the subject of count 5; the stock in the latter two roads cost approximately $8 million.

During 1942 and for several of the years earlier, plaintiff operated substantially all of the properties of these subsidiaries under leases. For practical purposes, the arrangement was a consolidation of railroad operations. Plaintiff paid all of the roads' expenses, including the expenses of maintaining their corporate existence and dividends on the publicly-owned 1 percent of the stock of The St. Joseph Grand Island, and recorded in its books and reported in its tax returns all the income and expenses of the operations of the lines of the subsidiaries.

Equity invested capital, the basis on which plaintiff computes its excess profits credit, is under the 1939 Code subject to a reduction by the percentage of "inadmissible assets" among total assets. Sections 715, 720, Internal Revenue Code of 1939, 26 U.S.C. § 715, 720 (1952 ed.). An "inadmissible asset" is by Section 720(a)(1)(A) of the 1939 Code, as amended in 1941, defined to mean "[s]tock in corporations . . . except stock which is not a capital asset." Admissible assets are by section 720(a)(2) "all assets other than inadmissible assets." That is, while corporate stock held by a taxpayer entity is presumptively to be considered as a capital asset and as such "inadmissible" as an asset for purposes of invested capital, where the circumstances are such as make the stock a non-capital-asset, it becomes "admissible."

The ultimate question raised by count 18 is therefore whether or not the plaintiff's stock in these subsidiaries is to be treated as an "admissible asset" under Section 720 of the 1939 Code, and thus not be cause for any reduction of the plaintiff's invested capital. Resolution of the question depends on whether the assets are capital assets, which in turn is determined by whether the plaintiff acquired and holds the stock in the lessor roads for a business or for an investment purpose.

The parties are agreed that the legislative purpose was to make a non-capital-asset admissible, and thereby part of equity invested capital, when the income generated by it was includable in a taxpayer's excess profits net income. The simplest illustration of such an asset, prominent in the minds of the drafters of the relevant amendment of the section, is the corporate stock held by a securities dealer for sale to his customers. When such sales take place, there is generated ordinary excess profits net income, the committee reports said, "no different from any other article held for sale by a dealer." H.R. Rep. No. 146, 77th Cong., 1st Sess., 20 (1941); S.Rep. No. 75, 77th Cong., 1st Sess., 20-21 (1941); 87 Cong.Rec., Part 2, 1638 (1941). The regulations thus provide that the term "inadmissible assets" means "stock in all corporations, domestic or foreign, . . except stock which is not capital asset (such as stock held primarily for sale to customers by a dealer in securities)." Treasury Regulations 112, Sec. 35.720-1.

Litigation has produced illustrations of non-capital-assets other than the stock on the security dealer's shelf. One is stock in a restaurant, a going business, bought not for investment but to conduct a restaurant business by the use of the corporate assets, there having been some doubt as to the assignability of the lease of the restaurant premises. John J. Grier Co. v. United States, 328 F.2d 163 (7th Cir. 1964). Another illustration, one of a number of cases involving stock bought as a source of inventory for regular business, is stock in a distillery, bought by a liquor dealer to obtain rights to purchase whiskey and sold promptly after the rights were exercised. Western Wine Liquor Co. v. Commissioner, 18 T.C. 1090 (1952). A relatively recent case in this court involved stock in a manufacturer of yarn, held to be a non-capital-asset because it was bought by the taxpayer, a yarn sales agency, in order to obtain an extremely valuable source of supply of yarn. Waterman, Largen Co. v. United States, 419 F.2d 845, 189 Ct.Cl. 364 (1969), cert. denied, 400 U.S. 869, 91 S.Ct. 103, 27 L.Ed.2d 109 (1970). United States v. Mississippi Chemical Corp., 405 U.S. 298, 92 S.Ct. 908, 31 L.Ed.2d 217 (1972), does not make Waterman, Largen less authoritative.

The rule that emerges from the cases is that corporate stock which is held for a business purpose, that is, one intimately related to the taxpayer's normal source of business income, is not a capital asset. Stock not so related, and held for investment purposes, is a capital asset.

Plaintiff claims that the stock involved here meets the business purpose test for a non-capital-asset, in that plaintiff acquired and has held the stock in these subsidiaries in pursuance of its railroad operations, and not as an investment or speculation; that the operation of the leased lines was intended to produce excess profits taxable income, to be reported in plaintiff's return.

Plaintiff is upheld, and the issue is decided in favor of non-capital-asset status. The result is, however, not free from doubt. It is quite true, as the Government emphasizes, that the case of the securities dealer is far different from that of a leased railroad subsidiary. It is also true that decision in favor of plaintiff permits the money invested in a subsidiary to be treated as invested capital (and thus reduce excess profits taxes) twice, once as part of plaintiff's invested capital and again as part of the invested capital of the subsidiaries. Neither statute nor regulation, however, limits non-capital-asset status to the stock on the dealer's shelf.

The business-purpose-investment-purpose test is at best imprecise. Nuances of the application of the test could doubtless be discussed at length, and a case made for the Government's view, on the basis of the consideration it emphasizes. The consideration which to my mind tips the scales in favor of plaintiff's position is the historical evidence in this case of the intimate relationship of the subsidiaries to plaintiff's success as a railroad system. The loss of its subsidiaries prior to reorganization was a low point in the decline of the old Union Pacific Railroad, plaintiff's predecessor. The reacquisition of at least the Oregon Short Line was among the first thoughts for the future of the reorganizers and the new management. There can be no doubt that these subsidiary and connecting railroads, led by the Short Line, had great significance for the prosperity of the plaintiff as a transcontinental railroad system, in the years after it came out of reorganization in 1898. This significance appears in the course of the discussion of the valuation of the shares issued in the reorganization for the old road and for the Oregon Short Line, in count 5.

With these origins, the acquisition of the stock cannot be treated as a mere investment unrelated to the business operations of the plaintiff. It was accomplished for an operating, business purpose and the stock was held as part of the operation of plaintiff's business as a railroad. If anything, the leases that followed confirm the conclusion of business and not investment purpose, even assuming that they were entered into only to reduce costs. As different as are a railroad and a restaurant, the case presented is in essence similar to the case of the restaurant stock, John J. Grier Co. v. United States, supra. The plaintiff bought the stock to run the railroad and not to make an investment, and thus the stock is a non-capital-asset "admissible" for purposes of equity invested capital under Sections 715 and 720. Cf. Corn Products Refining Co. v. Commissioner, 350 U.S. 46, 76 S.Ct. 20, 100 L.Ed. 29 (1955); Booth Newspapers, Inc. v. United States, 303 F.2d 916, 157 Ct.Cl. 886 (1962).

Plaintiff is entitled to prevail on count 18.

VI. INTEREST ON 1948 AGREEMENT

Count 26 makes a claim for interest of over $12 million under a certain 1948 "cutback" agreement between plaintiff and the Commissioner relating to plaintiff's tax liabilities for 1942 and subsequent years. "Cutbacks" are refunds of railroad freight charges which have been paid by the Government. There is little or no dispute as to the facts; only a dispute as to their significance in the light of the agreement.

In transporting war material in 1942, when considerations of secrecy or the novelty of the material made it impossible to determine that a preferential, land-grant rate should be applied, plaintiff charged the Government the full commercial tariff. Thereafter, on audit by the General Accounting Office between 1943 and 1957 (delayed because of the great volume of auditing of wartime charges), the correct rates were determined, and plaintiff refunded the overcharges as they were determined. The refunds of overcharges for 1942 amounted to $12.8 million.

Plaintiff had accrued the full charges made in 1942 as income in that year and reported them as such in its tax return for that year. The charges having been received under a claim of right, income for 1942 could not be recomputed to exclude the amount of the refunds or cutbacks. Under normal tax procedures, the cutbacks could be treated only as deductions from taxable income in the years in which the cutbacks were paid over to the Government. See Healy v. Commissioner, 345 U.S. 278, 73 S.Ct. 671, 97 L.Ed. 1007 (1953); United States v. Lewis, 340 U.S. 590, 71 S.Ct. 522, 95 L.Ed. 560 (1951). Large amounts of income, reported in years of high, wartime tax rates, were thus, by virtue of the delays in the audit, about to be reversed by reductions in income in postwar years of lower, peacetime tax rates.

In recognition of the prospective inequity, the Commissioner of Internal Revenue and the plaintiff agreed, on November 29, 1948, that normal practice would not be followed — that plaintiff would be permitted to allocate the amounts of the cutbacks not to the years of their payment but to the years in which the original charges had been included in taxable income. Accrued income for 1942, for instance, would be retroactively reduced by the amount of the cutbacks of rates included in income in that year and the cutbacks would be disallowed as reductions in income in the various years — 1943 through 1957 — in which they were actually made.

The letter from the Commissioner to plaintiff embodying this agreement, called the Cutback Agreement, stated as follows:

In view of the facts and circumstances presented, permission is granted under the authority conferred in section 43 of the Internal Revenue Code to allocate, on the terms and conditions hereinafter stated, repayments heretofore or hereafter made of excessive transportation charges of the class described above to the years in which such charges were included in taxable income. However, the allocation of any such repayments of excessive transportation charges to any year shall be made only to the extent the refund or credit of the overpayment of income and/or excess profits tax, if any, resulting therefrom is not prevented for any reason, and the deficiency if any, of income and/or excess profits tax resulting therefrom may be assessed.

The letter then goes on to set out the "terms and conditions" of the agreement, in numbered paragraphs:

In this connection, it is understood that you agree, as follows:

1. All amounts received by you as transportation charges from the Federal Government Departments and Agencies shall be included in taxable income on the accrual basis.

2. All refunds of transportation charges made by you to the Federal Government shall be allowed as deductions in the year or years in which such transportation charges were included in income, and any deductions claimed in the year or years such refunds were made will be disallowed. * * *

The result of the permitted reduction in income for 1942 and the corresponding increase in income, distributed over 1943-1957, all other things being equal, would be an overpayment of tax for 1942 and underpayments in 1943-1957, and thus a refund for 1942 and deficiencies for 1943-1957. (This would all the more be true if, as seems to have been the case, plaintiff did not in the years following the making of the cutback agreement in 1948 cease its practice, in its tax returns, of reducing its annual income by the respective amounts of cutbacks made in those years.) And interest payable to plaintiff on the refund would exceed interest payable by plaintiff on the deficiencies, by reason of the longer span of time involved in the refund than in the deficiencies.

To relieve the Commissioner of such a net interest liability, the parties further agreed, in the Cutback Agreement, that the maximum interest payable to the plaintiff on a refund caused by the cutbacks should be limited to the amount of interest payable by the plaintiff on the deficiencies caused by the cutbacks, as follows:

3. The amount of interest on refunds of income and excess profits taxes resulting from these adjustments shall be allowed only to the extent of, and limited to, the amount of interest on deficiencies resulting from these adjustments.

Plaintiff's tax returns for 1942 were, as noted above, not finally audited until 1959. The tax years following 1942 are by reason of waivers still open; deficiencies for these years resulting from the cutbacks, have thus not been assessed, although they have been disallowed in revenue agents' reports for those years.

Plaintiff filed an excess profits tax return (Form 1121 under the 1939 Internal Revenue Code) for 1942 showing no liability; there followed certain additional payments by plaintiff in anticipation of a deficiency. (The precise amounts, and other details not here necessary, are set out in the accompanying findings of fact.) On audit, and after making general adjustments, the Commissioner computed a deficiency of $15.9 million unrelated to the cutbacks reduction in income.

The cutbacks reduction in income of $12.8 million, alone, would have resulted in an overassessment of $11.3 million. An adjustment attributable to cutbacks therefore required a deduction of $11.3 million from the deficiency of $15.9 million. When this adjustment was made, there remained a total proposed deficiency of $4.6 million, on which $.3 million in interest was payable, or a total assessed deficiency of $4.9 million.

Plaintiff's income and declared value excess profits tax return (Form 1120) showed a tax of $38.4 million. On audit, the Commissioner determined that liability for income tax was $30.4 million and that plaintiff had no liability for declared value excess profits tax. The consequent overassessment of income tax, with adjustments for certain subsequent assessments, was $7.9 million.

The computations for the two taxes were netted out as follows. The deficiency of $4.9 million in excess profits tax was satisfied by a credit of $4.2 million of the $7.9 million overassessment in income tax, and a certain postwar credit of $.7 million. This left $3.7 million due to plaintiff as the net balance of the $7.9 million overassessment of income tax, on which interest of $4 million was payable. The total, $7.7 million, was then paid to plaintiff.

The cutbacks had played a part only in the excess profits tax; the adjustment for cutbacks, the parties are agreed, had no effect on plaintiff's income tax. While the cutbacks of $12.8 million reduced plaintiff's 1942 income subject to income tax by that amount, the effect of the reduction was completely offset by the reduction of an allowable credit, under Section 26(e) of the Internal Revenue Code of 1939, for income subject to excess profits tax.

There was, also, no connection between the cutbacks and the payment of the deficiency in excess profits tax by credit of a portion of the overassessment of income tax and the interest paid to plaintiff on the refund. It is agreed that the interest of $4.0 million, paid as part of the $7.7 million refunded to plaintiff, was attributable to the overassessment of income tax and was not related to the cutback adjustment or the interest in respect of cutbacks now claimed by plaintiff.

With the apparent complexities created by two tax returns stripped away, and, thereby, the income tax return, the refund of income tax and the interest thereon all set aside, the remaining relevant facts, all related to excess profits tax, are seen to be quite simple. The cutbacks reduced excess profits tax income and had there been no other factors the reduction would have meant an overassessment and a refund of excess profits tax of $11.3 million. Simultaneously, however, other items of income, not in question, increased income by a greater amount, which increase, alone, would have meant a deficiency of $15.9 million. When combined, the two adjustments resulted in a deficiency of $4.6 million. The actual tax liability figures were $18.807 million, on the basis of all adjustments except the cutback adjustment, and $7.484 million on the basis of all adjustments; the difference was the $11.3 million due to cutbacks. This sum was used to reduce the liability otherwise existing. In simplest terms possible, a tentative underassessment of $15.9 million, reduced by a tentative overassessment of $11.3 million due to cutbacks, produced a net actual underassessment and deficiency.

No refund was payable or paid. Since a refund, due to cutbacks or otherwise, was not made, no interest was due or payable. Interest is payable only on a refund. Hence plaintiff can have no claim to interest, at least on the basis of general tax law.

Plaintiff contends, however, that the Cutback Agreement took the payment of interest in connection with the diminution of income by reason of cutbacks "out of the normal process of assessment and collection," made it "subject solely to the provisions" of the Cutback Agreement, and required the payment of interest on the overassessment of $11.3 million resulting from cutbacks.

The Cutback Agreement does no such thing. It displaces normal rules of assessment, collection and liability for interest on refunds only so far as the terms of the agreement go. And the terms of the agreement have a narrow effect upon interest. Paragraph 3, the only relevant provision of the agreement, does not in any wise provide for or require the payment of interest to plaintiff. It rather sharply limits payments of interest on refunds resulting from cutback adjustments. Interest to plaintiff on refunds resulting from cutbacks, it is provided, "shall be allowed only to the extent of, and limited to" the amount of interest payable by plaintiff on the deficiencies from the cutback adjustments in the to-be-assessed years of the actual refunds.

There is no limit put on the interest payable by the plaintiff. That was not the problem deemed needful of any agreement. The problem to which the paragraph is directed is that the large refunds to be expected for the war years, and the long span of time involved, would entail a large liability for interest to be paid by the Commissioner. The parties therefore provided in paragraph 3 that interest should be paid to plaintiff on refunds only in the amount of the interest paid by plaintiff on deficiencies.

The effect is to eliminate any payment of interest to plaintiff on refunds resulting from cutback adjustments. No actual payment of interest could be made to plaintiff until the limit on the payment becomes known, and when the limit — the amount of interest payable by plaintiff on deficiencies in the subsequent years — does appear, the cross-obligations wash out and nothing is payable to plaintiff.

A more fundamental reason for the failure of plaintiff's claims is that no refund was made and thus no interest was due, on general principles, unaffected by paragraph 3. Plaintiff seeks to overcome the fact of no refund by arguing that because the agreement required the Commissioner to refund an overassessment by reason of cutbacks, he should therefore not have wiped out the overassessment of $11.3 million resulting from cutbacks. To this plaintiff adds a complaint that its liability for interest on the deficiencies to be assessed for 1943-57 is still running, and that the Commissioner's action has deprived it of a fund of interest with which to minimize its interest bill on the future deficiencies.

Restated to include the relevant facts, the plaintiff's contention is that the agreement requires that interest be paid, as if on a refund, on the tentative overassessment of $11.3 million by reason of one item of income — the reduction in income by the cutbacks — though it was in the final computation wiped out by a larger underassessment of $15.9 million by reason of other items of income. So stated, the lack of merit becomes clear. Interest is payable only on the net, the refund, and not on individual items merged into the net refund. The overassessment of $11.3 million, by reason of the cutbacks, was required to be "wiped out" by the larger underassessment. The Cutback Agreement did not bear on the normal processes of audit and collection. In the paragraph of the agreement preceding the numbered paragraphs, quoted above, there is explicit recognition that the agreed-upon allocation of cutbacks to the year of the original overcharge might result in either "refund or credit." Here it resulted in a credit.

Had the Commissioner not applied the $11.3 million overassessment of excess profits tax resulting from the cutbacks to the $15.9 million deficiency, plaintiff would have been called upon to pay the $15.9 million deficiency resulting from adjustments other than cutbacks, instead of the actual deficiency of $4.6 million. Had plaintiff been granted its wish that it be paid interest for the years since 1943 on the $11.3 million overassessment, it could not have escaped liability for interest on the $15.9 million deficiency. In economic effect, therefore, plaintiff has received the interest on the $11.3 million which it is now claiming.

Finally, plaintiff complains of the continued running of its liability for interest on the deficiencies yet to be assessed. This is as it must be. The Cutback Agreement shows no sign of an intention to limit plaintiff's liability for interest; paragraph 3 limits interest paid to plaintiff, not interest to be paid by plaintiff. Interest on a deficiency compensates the Government for the withholding of the money. Since the time of the Cutback Agreement in 1948, the burden of continued accrual of interest has been deliberately assumed by plaintiff, by its continuation of the practice of deducting the cutback amount from income in the year of the cutback, though the agreement contemplated deduction in the year of the original charge. Moreover, plaintiff could stop the running of interest by one method or another, all of which would involve payment.

The claim on the Cutback Agreement has no merit and plaintiff is not entitled to recover on count 26.

VII. 1898-1918 LAND SALES

This count 34 is one of those before the court as offsets to defendant's set-offs. 389 F.2d 437, 189 Ct.Cl. 103 (1968). The facts needful to be stated for present purposes are few.

In count 34 a claim is made to include in plaintiff's income, and thus in earnings and profits for excess profits tax purposes, some $23 million received as gross proceeds of sales of land between 1898 and 1918.

The facts are these. Among the properties of its predecessor acquired by plaintiff in the 1898 reorganization were 6,577,000 acres of land in Colorado, Kansas, Nebraska, Wyoming and Utah. By 1919, plaintiff owned only 971,348.64 acres. It cannot be found that the entire difference, 5,604,942.85 acres, was sold, for lack of evidence as to the amount of land acquired in the intervening period, and because the figure for 1919 is one for non-carrier land. A great quantity of land was however sold.

Immediately after the reorganization, plaintiff had transferred the land to the Union Pacific Land Company, which it created for the purpose, in return for the stock and bonds of the Land Company. The bonds — $10 million in par value — were secured by a mortgage on all the Land Company's assets. The Land Company stock and bonds were then pledged by plaintiff as collateral under the first mortgage on plaintiff's own property.

This last mortgage provided that the net proceeds from sales of non-railroad land were to be paid to plaintiff to reimburse it for expenditures for betterments, improvements and equipment, exclusive of expenditures charged to operating expenses. Accordingly, the trustee paid over to plaintiff, between 1898 and 1918, $23,286,091.13, of his own total receipts of $23,392,717.22. On its books, plaintiff credited the sum received to its investment in road equipment account. No credit was made to income or surplus. The receipt of the money seems to have been simply the occasion for a credit to the investment account, that is, a decrease in the amount shown as invested.

No more is known than is set out above. We can only speculate on why the plaintiff's accountants and management omitted to show on the books as income or surplus the receipt from the trustee of what plaintiff now says was income. In any event, plaintiff did not in whole or in part record the receipt of the $23 million as income or surplus or as an addition to earnings or retained earnings and profits.

In its post-trial brief and requested findings of fact, plaintiff urged, from the foregoing facts (supplemented only by taking it as a fact that 5,604,942.85 acres of land were sold), that the original credit to the investment in road equipment account was error, and that the proceeds of $23,286,091.13 from the sale of 5,604,942.85 acres of land should now be included in its accumulated earnings and profits for excess profits tax purposes. No suggestion was made that a sum less than the $23 million might be earnings and profits; no findings were requested concerning a cost or basis for the lands claimed to have been sold for the $23 million.

In condensed fashion, plaintiff's case was this: land was sold for $23 million, and therefore that sum was income which belonged and belongs in accumulated earnings and profits, whatever the contemporaneous accounting treatment. Of course this is not the law. For one thing, gross proceeds of the sale of land are not gain or income; only the difference between sales price and cost or basis is realized income. Gross payments on bonds are not income; only the difference between payments and cost is realized income. Section 111, 1939 Code. And this is what the Government replied.

The Government in its reply also pointed out that plaintiff had failed to mention in its proposed findings or brief that $6,972,025.93 of the $23 million — the portion consisting of interest — had already been included by plaintiff in its accumulated earnings and profits for purposes of computation of its excess profits tax, and allowed by the Commissioner, and that this sum could hardly again be claimed.

On this, plaintiff abandoned the claim for $23 million. Admitting that it was entitled to treat as accumulated earnings and profits only the difference between sales price and basis, and admitting that it was not again entitled to the $6.9 million, it now claimed not $23 million, but $5 million. Specifically, it claimed, in newly requested findings proposing figures for each year from 1898 to 1916, that it had sold, for $9,987,507.94, 5,972,197 acres of land which had cost $6,533,664.62, with a realized gain of the difference, $3,453,843.32, and that the Land Company, a "wholly-owned subsidiary," had sold for $9,900,387.23, 3,620,615 acres of land which had cost $8,312,812.83, with a realized gain of the difference, $1,587,574.40. The request for these findings was filed with plaintiff's post-trial reply brief, 10 months after the date on which proposed findings were due and were submitted.

The newly proposed findings contain minutely detailed and varied contentions as to facts and practices over 50 years old. Entries on many pages of the books of account of the plaintiff, exhibits whose significance for this purpose had not been mentioned before, were cited to establish cost and sales prices for land parcels said to have been acquired and sold in the years 1898 to 1916.

Only one of the fatal defects in these proposed findings is that so far as can be determined plaintiff did not earlier give defendant notice of such contentions, on trial or in pretrial or at any other time in the many years in which the case had been pending.

In the pleadings, plaintiff had alleged and defendant denied that lands acquired on reorganization were carried on the books at no value and had been disposed of between 1898 and 1919 for $23,286,091.13, which had been erroneously credited to investment in road and equipment and not credited to accumulated earnings and profits. Only the acquisition of lands in the reorganization was admitted.

All the pretrial papers exchanged by the parties — statements by each party of its contentions of law and facts, statements by each party of the controverted facts it proposed to prove and a joint statement of the issues which identified the documents and witnesses with respect to each issue — show on the one hand that plaintiff claimed a zero basis for all the land sold, and on the other hand that defendant never retreated from its denial of the claimed zero basis and from its positions that income is realized only to the extent of the difference between gross receipts and cost, and that plaintiff realized no income from the land transactions beyond the sums (the $6.9 million) already included in accumulated earnings and profits.

These pretrial papers were exchanged and filed as part of extraordinarily extensive pretrial proceedings, which took place under pretrial orders providing that facts, issues and witnesses not proposed, stated or disclosed by a party prior to trial would not be permitted to be proved, raised or called on the trial, except as required by the exigencies of the case. The exigencies of the case most emphatically do not require that plaintiff, having failed in pretrial and trial to mention the facts, contentions and issues it now relies upon, be permitted to raise these issues for the first time after trial, in a reply brief, long after the time for requested findings has expired.

Plaintiff has until now been committed to a zero basis, which it apparently intended to prove as a matter of law by the arguments and testimony relating to its accounting practices in the opening years of the century concerning retirements of equipment. These were the arguments and testimony which it relied upon in count 8, concerning $13 million deducted from income in those years and credited to investment account. For this position, plaintiff would now substitute a new theory, on new facts, on newly requested findings. It is too late. The disorder of permitting such a step and the consequent prejudice to defendant are unthinkable.

It may be added that even were plaintiff's proposed findings to be considered, they could not be adopted. As might be expected from extemporized findings, the record does not support what plaintiff sets out to prove — the acreage on hand and its cost, the acreage acquired and its cost, and the acreage sold and its proceeds, for plaintiff and the Land Company, for each year from 1898 to 1916. Figures are lacking for major elements in the purported chain of proof; projections and assumptions, offered to fill in the gaps, are hopelessly insufficient. For instance, the proposed figures for gross proceeds are often only contract sale prices, and the lands acquired are often lands surrendered by deferred payment purchasers. Another type of defect is the inconsistency between the new contentions, which ignore the Land Company's bonds, and the earlier position of plaintiff, already imbedded in its tax returns, that this is a case of income from the holding and redemption of bonds.

On the failure of the newly proposed findings, one short statement of this whole affair could be that a basis for the lands sold has not been shown and thus no income is provable. Another is that plaintiff fails to overcome the force of its own books. The crediting of the proceeds to investment in equipment and the failure to credit income and thus add to surplus or earnings or profits, as inexplicable as they may be on looking backwards, were the deliberate choice of plaintiff's management and accountants. As the Interstate Commerce Commission said in connection with the deductions from income involved in count 8, some purpose must be ascribed to the accountants who made up the books. Plaintiff has the burden of proof when it undertakes to show otherwise than is said in its books. For lack of proof of error or meaninglessness, these entries cannot be undone after they have existed for half a century.

Plaintiff is not entitled to prevail on count 34.

VIII. UNAMORTIZED BOND DISCOUNT AND EXPENSE

When a corporation sells bonds at less than their face value, e. g., 20 year 4 percent bonds with a $1,000,000 face value for $950,000, the difference between the face value and the proceeds, or in the illustration $50,000, is called bond discount. (If the bonds are sold at more than face value the excess is called premium.) The expenses of the issue, say $10,000, which reduce still further the net proceeds of the issue, are called bond expense.

By the customary accounting treatment the $1,000,000 payable at maturity goes on the right or liability side of the corporate balance sheet. The $940,000 net proceeds go into cash, and are recorded on the left or asset side. Where shall the $60,000 in bond discount and expense go? It will have to be paid at maturity, one way or another, and it will have to be defrayed, somehow, from income or capital. One type of treatment is to charge it off immediately against surplus, and this is how plaintiff recorded on its books, presumably in accordance with ICC regulations, the bond discount and expense arising from its sales, in its early years, of bonds at a discount. Another method, the approved income tax treatment, call for a systematic annual charge against income of a ratable portion, over the life of the bonds, and this is what plaintiff did in its tax returns. Helvering v. Union Pacific R. R., 293 U.S. 282, 55 S.Ct. 165, 79 L.Ed. 363 (1934).

The question here presented by additional defense is whether unamortized bond discount and expense, remaining in the years 1940 through 1942 after plaintiff's earlier income tax deductions, was correctly treated by plaintiff as an asset for purposes of the excess profits tax, more specifically, for purposes of the reduction in equity invested capital by the percentage of inadmissible assets among total assets, required by sections 715 and 720 of the 1939 Code as amended, 26 U.S.C. § 715, 720 (1952).

While the Commissioner of Internal Revenue did not disturb plaintiff's treatment of the sum as an asset, the Government here challenges the propriety of such treatment, in an additional defense, No. 9, by way of a setoff. The amount involved for 1940 through 1942, in excess profits tax credit for 1942 and in credit carryovers from prior years, is $11,526,629, of which $10,355,828 is bond discount and $1,170,801 is bond expense.

The Government is here upheld in its position that unamortized bond discount and expense is not an asset for purposes of the cited sections.

Unamortized bond discount has none of the characteristics of an asset, except for its position on the left side of the ledger. It is essentially interest — the cost to the borrower of the borrowed money or the compensation paid for the use of the borrowed money — and it is to be amortized until maturity of the bonds, and then paid. Helvering v. Union Pacific R. R., supra. The unamortized portion, interest not yet paid, is patently not an "asset," whatever be the definition of that seldom construed term.

The origins of bond discount determine its character as a reflection of interest rates — real or market and nominal. The corporation which so chooses can simply sell its bonds or other obligations at the market rate or interest and encounter no discount. When, however, an issuer desires to sell bonds which will bear on their face a lower rate of interest than that which will cause the bonds to be bought, it perforce must set the price of the bond lower than maturity value. The buyer is offered, in addition to the face rate of interest per annum, the difference between face value and sales price in the form of a lump sum at maturity. This difference is called discount. The yield to the buyer takes two forms — the nominal coupon rate and the deferred lump sum.

Both forms, together called effective interest, are costs to the issuer of the borrowing and compensation to the lender for the loan. Both must be found by the issuer and paid to the lender. Coupon interest is paid annually from income. Discount is accumulated annually (the term "accumulation" is, by accounting authorities, preferred to "amortization"), from income, and paid over at maturity. If the coupon rate, payable annually, is interest, then the lump sum payment, the product of annual accumulation, is also interest.

The income tax treatment of discount — accumulation of the total discount by a charge of a ratable portion against income in each year of the life of the bond — is a recognition of the reality that discount is in essence the same as coupon interest, and that each is a cost of the borrowing to be charged annually against income in the years of the borrowing. The unaccumulated, to-be-charged-against-income interest represented by the remaining years' coupons is of course not an asset, and the similarly unaccumulated to-be-charged-portion of effective interest represented by unamortized bond discount is, therefore, equally not an asset. Interest, payable this year or next or to be set aside year by year and paid at a fixed future date, is not an asset.

Judicial, administrative and accounting authority recognize that bond discount is interest. Discount is held to be deferred interest, "likely to arise when the stated rate of interest on the obligation is less than the rate demanded by the market. American Smelting and Refining Co. v. United States, 130 F.2d 883 (3d Cir. 1942) * * * [E]conomic and business reality * * * recognizes that to the issuer bond interest is reflected both by the stated rate of interest and by the amount below or above par received by the issuer when the bonds are originally distributed." Atchison, Topeka and Santa Fe R. Co. v. United States, 443 F.2d 147, 151, 153 (10th Cir. 1971).

An early, leading decision of the Securities and Exchange Commission held discount to represent "additional interest," "part of the cost which the issuer must eventually pay for the use of the funds." The Commission noted that "[a]ccounting authority recognizes that bond discount should be considered as part of the interest cost of the capital obtained." In Re Alleghany Corporation, 6 SEC 960, 962 (1940). The American Institute of Certified Public Accountants, in a 1961 reissue of earlier research bulletins entitled "Unamortized Discount, Issue Cost, and Redemption Premium," is to the same effect.

The attempted characterization of unamortized bond discount as prepaid interest and the likening of it to prepaid insurance or rent is without basis. Discount represents the portion of the face amount of the bonds not received by the debtor, an amount owed and to be paid on maturity. It is more a nonreceipt, an unpaid cost or charge, a liability or a deferred loss, a deficit or an offset to maturity value, than an asset. Prepaid insurance or rent has been paid, could be refunded and will save an expenditure of equal amount in the time to come. Unamortized bond discount has not been paid, is a liability and remains to be expended in the future.

The custom of recording unamortized bond discount on the left side of the balance sheet, even had plaintiff followed it, gives no support to the notion that it is an asset. Expert testimony in the record, made the basis for one of the accompanying findings of fact, is that unamortized bond discount and expense is put on the asset side of the balance sheet "solely to make the balance sheet balance." Double entry bookkeeping and its compelled symmetry may not alter the nature or control the tax treatment of financial realities. Doyle v. Mitchell Brothers Co., 247 U.S. 179, 38 S.Ct. 467, 62 L.Ed. 1054 (1918); United Profit-Sharing Corp. v. United States, 66 Ct.Cl. 171, 182 (1928).

Percipient auditors have long recognized that not everything that appears on the left side of the ledger warrants the caption "asset." Robert H. Montgomery, a leading authority, writing that bond discount was not an asset but a deferred loss, attributed its inclusion among assets to the "curse of balancing." Despairing of the acceptance of a realistically unbalanced balance sheet, he suggested changing the balance sheet heading to "Assets etc." and making discount on bonds an "etc." R. H. Montgomery, The Curse of Balancing, or Theory v. Practice, 63 J. of Accountancy 279 (1937). Professor W. A. Paton, another authority, concurs in the opinion that bond discount is not an asset.

Plaintiff would support its claim that unamortized bond discount is an asset by the charge-off, on its books, of the entire amount of the discount to surplus, at the time the bonds were first issued. Such accounting for bond discount and expense, while doubtless proper under ICC regulations, is in disfavor with the accounting profession and is not proper tax practice. The profession prefers and the tax law requires the systematic charge to income over the life of the bonds. The source of both the professional preference and legal requirement for an annual charge to income is the character of bond discount as interest, a cost of borrowing, and the consequent feeling that both types of interest — coupon interest and the discount variety — should equally be reflected in the income account by an annual expense item, else income is overstated. An immediate write-off against earned or paid-in surplus, in effect a charge to past income or to capital, is an over-statement of income by the annual cost of the borrowing. Plaintiff's charge of the entire discount to surplus was thus inconsistent with the nature of discount and cannot give grounds for characterizing it as an asset.

In a final effort to show that discount is an asset, plaintiff urges, with support in the expert testimony given on trial, that the unamortized portion of bond discount would have value to a purchaser of the business, as a source of future tax deduction. Such value is surely limited. Unamortized discount would have no value to a purchaser of the assets, no weight to a lender considering a loan. It has not been bought or paid for, it produces no income and it could not be sold or assigned. The value it might have to a purchaser of the going business, as a source of income tax deductions, could only be realized if profits were made by the use of the admitted assets of the enterprise. That purchase, moreover, would find the same "value" in the obligation to pay the fixed, coupon rate of interest, which like the annual accumulated share of discount, must each year be a charge on income, Discount, actually a cost to be defrayed in the years to come, has the same type of value, as a source of income tax deductions, as the corporation's contracts with its executives, the lease of its premises, its pension plan, or, for that matter, a large tax loss carried over from past years. All give rise to tax savings. Since such "value" does not make an asset of next year's tax loss, it cannot transform unamortized bond discount from unpaid interest to an asset, at least for purposes of an excess profits tax based on concepts of invested capital and total assets.

There remains the unamortized bond expense. It, too, must be regarded as is unamortized bond discount — as a non-asset.

Bond issuance expense may differ factually from bond issuance discount, in that expenses are actually paid out for services such as printing, legal fees and commissions, and, sometimes or always to an extent, are paid out by the issuer itself as distinguished from the issuer's underwriter. Accordingly, some accounting authorities distinguish between discount and expense, and hold the latter to be a genuine asset.

If the question were open for fresh decision, perhaps the unamortized portion of bond expense paid out by the issuer (as distinguished from expenditures for services by underwriters which merely reduce the proceeds of sale paid over to the issuer) might be held to be an asset. Plaintiff has, however, failed to show that the bond issuance expenses here involved were of any different character than those held in Helvering v. Union Pacific R. R. Co., supra, to be as much interest as bond issuance discount.

In Helvering v. Union Pacific R. R. Co., the Supreme Court held that commissions paid out on the issuance of bonds, one of the customary bond issuance costs, were to an accrual taxpayer (as is the instant plaintiff) to be capitalized, amortized over the life of the bond and the amortized amount deducted annual from income. The decision has been applied in this court to bond issuance costs generally. Chicago, Milwaukee R. Co. v. United States, 404 F.2d 960, 967, 186 Ct.Cl. 250, 262 (1968).

The Supreme Court in its opinion said that discount and expense were both "factors in arriving at the actual amount of interest paid for the use of capital procured by a bond issue" which "must be added to the aggregate coupon payments in order to arrive at the total interest paid." 293 U.S. at 286, 55 S.Ct. at 167. The Court went on as follows ( 293 U.S. at 286-87, 55 S.Ct. at 167):

But even if the commissions, unlike discount, may, as the Government insists, be regarded as a contemporary expense of procuring capital, it is one properly chargeable to capital account. In practice it is taken out of the proceeds of the bonds by the banker. But in any case it must be deducted from the selling price to arrive at the capital realized by the taxpayer from the sale of the bonds, in return for which he must, at maturity, pay the face value of the bonds. The effect of the transaction in reducing the capital realized, whether through the payment of commissions or the allowance of discount, is the same. * * *

Here the commissions, when paid, were properly chargeable against capital, and reduced by their amount the capital realized by the taxpayer from the bond issue. They come out of the pocket of the taxpayer only on payment of the bonds at maturity. But, unlike the purchase and sale of property, the transaction contemplated from the beginning a fixed date, the due date of the bonds, at which the difference between the net amount of capital realized upon the issue and the par value of the bonds must be paid to bondholders by the taxpayer. * * *.

Given the Supreme Court's ruling on the essential sameness of discount and expense, albeit in an income tax case, the decision in the instant case can only be that expense goes with discount, and since discount is surely not an asset, expense, too, is not an asset.

Accordingly the defendant is entitled to prevail on additional defense 9.

IX. DISCOUNT AND PREMIUMS

In additional defense 6, also involving bond discount and expense, the Government contends that plaintiff erroneously failed to treat as "interest," under section 711(a)(2)(B) (26 U.S.C. § 711(a)(2)(B) (1952), amortized bond discount and expense for 1940-42 (excess profits tax credit carryovers for 1940 and 1941 are involved, in addition to 1942 taxes), and call premiums and unamortized bond discount paid out when certain bonds were retired in 1940 at their face value plus a premium, as required by their terms on an early retirement.

Under section 719(b) (26 U.S.C. § 719(b) (1952) only 50 percent of borrowed capital may be included in invested capital and, consistently, when invested capital is the method used (as it is here) to determine the excess profits credit (which in turn is the basis of the tax), interest deductible from excess profits net income is required by section 711(a)(2)(B), supra, to be reduced by 50 percent of the interest on borrowed capital. In other words, since only half of borrowed capital may be included in invested capital for purposes of the credit, only half of the interest paid on borrowed capital is allowed to be deducted from income for purposes of this tax. See Amana Refrigeration, Inc. v. United States, 285 F.2d 770, 772, 152 Ct.Cl. 406, 410 (1961).

Plaintiff and its subsidiaries deducted in full — that is, did not reduce by 50 percent — a certain sum of $10,253,706.34, composed of $546,273.70, the amount of amortized bond discount and expense in the tax years 1940-42, $3,265,051.46, the amount of unamortized bond discount remaining in 1940, the year of the retirement of a certain bond issue, and paid on the retirement, and $6,442,381.18, the call premium paid in that year on the retirement.

In additional defense 6, defendant contends that these deductions were all for "interest" and pursuant to section 711(a)(2)(B) must be halved. Plaintiff maintains, to the contrary, that amortized discount and the call premiums were ordinary and necessary business expenses under section 23(a) of the 1939 Code, and that the unamortized bond discount on the bonds reacquired at face value plus call premiums was a loss under section 23(f) of the 1939 Code.

The regulations provide that on retirement of bonds issued at a discount the excess of the price paid over the issue price plus the amount of discount already deducted is a deductible expense. Treas. Reg. 111, § 29.22(a)-17(3) (1943). The regulations are not helpful in determining whether the amounts are deductible as "interest."

Bond discount has already, in the preceding section of this opinion, been noted to be essentially interest. Helvering v. Union Pacific R. R., 293 U.S. 282, 286, 55 S.Ct. 165, 79 L.Ed. 363 (1934). See Erie Lackawanna R. R. v. United States, 422 F.2d 425, 427-28, 190 Ct.Cl. 682, 686 (1970). To the seller the difference between the discounted purchase price of a promissory note and the higher sale price is held to be ordinary interest income. United States v. Midland-Ross Corp., 381 U.S. 54, 85 S.Ct. 1308, 14 L.Ed.2d 214 (1965). The precise question has been several times decided, in favor of the Government. Discount has repeatedly been held to be interest under section 711(a)(2)(B) or its successor in the Excess Profits Tax Act of 1950, section 433(a)(1)(O), ch. 1199, 64 Stat. 1137, 1143, 26 U.S.C. § 433(a)(1)(O) (1952). Central Stations Signals, Inc. v. Commissioner, 10 T.C. 1015, 1020-21 (1948), affirmed per curiam, 174 F.2d 479 (2d Cir. 1949) (finance charge for factoring contract held interest under section 711(a)(2)(B) and not expense); Warner Co. v. Commissioner, 11 T.C. 419, 432 (1948), affirmed per curiam, 181 F.2d 599 (3d Cir. 1950) (amount of state tax on loans, imposed on bond purchaser but paid by issuer in addition to interest, held interest under section 711(a)(2)(B) and not a tax); L-R Heat Treating Co. v. Commissioner, 28 T.C. 894, 897 (1957) (negotiated bonus for loan withheld by lender from loan proceeds, in addition to 6 percent interest, held interest under 26 U.S.C. § 433(a)(1)(O) (1952 ed.)); Ringmaster, Inc. v. Commissioner, August 6, 1962, T.C. Memo 1962-187, 21 TCM 1024, 1030-32, dismissed per curiam, 319 F.2d 860 (8th Cir. 1963) (amounts paid as commissions for securing loans held interest under § 433(a)(1)(O), not brokerage fees).

The amounts of bond discount already amortized are thus to be deducted as interest. It may well be that an interest expense is also an ordinary and necessary business expense, but it is interest on borrowed capital and thus within the intent and subject to the reach of section 711(a)(2)(B).

Unamortized bond expense on bonds reacquired, and call premiums payable and paid at such retirement prior to maturity, are no different in principle from amortized bond discount. When an issuer sells bonds at a discount, the entire difference between issue price and face value is interest, to be paid at maturity, and to be charged to income each year and accumulated over the life of the bonds. Whether the bonds are early retired, pursuant to the terms of the issuance, by payment of face value plus a call premium, or are retired on maturity by payment of face value alone, there is no distinction to the lender or borrower between the constituent parts of the amount of discount on the loan.

What has been said indicates that call premiums, too, are additional interest, in this case payable for the privilege of converting a longer term loan to a shorter. A payment closely similar to a bond call premium, a charge paid by a borrower when he prepays, before maturity, the principal of a mortgage or promissory note, has been held in this court to be "interest" gross income to an insurance company. In Equitable Life Assurance Society v. United States, 181 F. Supp. 241, 242, 149 Ct.Cl. 316, 319, cert. denied, 364 U.S. 829, 81 S.Ct. 68, 5 L.Ed.2d 56 (1960), the court held:

The precise question before us was considered by the Tax Court in General American Life Insurance Company, 25 T.C. 1265 (1956). That court decided that prepayment charges are in reality an additional fee for the use of the lender's money for a shorter period of time than originally agreed upon, and that this fee represents the generally higher cost of a short-term, as opposed to a long-term loan. The charges are part of the compensation to the lender for the use of money. Deputy v. duPont, 308 U.S. 488, 498, 60 S.Ct. 363, 84 L.Ed. 416; Old Colony R. Co. v. Commissioner, 284 U.S. 552, 560-61, 52 S.Ct. 211, 76 L.Ed. 484. They are thus directly related to the economic cost of borrowing money and are not merely incidental to the loan transaction, but fall within the statutory term "interest."

See also Prudential Insurance Co. of America v. United States, 319 F.2d 161, 166-67, 162 Ct.Cl. 55, 65-66 (1963); United Benefit Life Insurance Co. v. McCrory, 242 F. Supp. 845, 850-51 (D.C.Neb., 1965) (penalty payment for early repayment of mortgage held interest income of life insurance company). Contrary dicta in Central South West Corp. v. Brown, 249 F. Supp. 787 (D.C.Del., 1965), cannot prevail over the cited authorities.

There would be no rational basis, in the light of the essential sameness of character between discount and call premiums, for holding one to be and the other not to be interest under section 711(a)(2)(B). The sums over the original amount of the loan constitute compensation for the use of the money — interest — payable when the loan is repaid, whether earlier than contemplated or on schedule. The early repayment is for the convenience of the borrower and works no change in the nature of the payment as interest. The decision here is, I believe, all but concluded by the decision of this court that difference between agreed redemption price and issue price (in a case of non-interest-bearing debentures redeemed a year after issue) is income "in lieu of the payment of interest," because it constituted "the agreed compensation for the use of the purchaser's money for the prescribed periods." Pattiz v. United States, 311 F.2d 947, 950, 160 Ct.Cl. 121, 127-128 (1963). If part of that difference is a loss, it is a loss which is in the form of additional interest payable for the privilege of early retirement. Cf. Helvering v. Union Pacific R.R., 293 U.S. 282, 286, 55 S.Ct. 165, at 167, 79 L.Ed. 363 (1934).

Against the foregoing wealth of authority, plaintiff puts forward only inferences from the legislative history of a nearby section, section 711(b)(1)(D), added by section 201 of the Second Revenue Act of 1940, ch. 757, 54 Stat. 974, 26 U.S.C. § 711(b)(1)(D) (1952). Section 711(b)(1)(D), not involved in the instant case, disallows certain deductions from taxable income of expenses and losses on the retirement or discharge of bonds, issued at a discount and outstanding for more than 18 months, in computing excess profits income for base period years beginning before January 1, 1940, the computation of which is necessary where the excess profits credit is based on average income rather than invested capital.

The House bill which became section 711(b)(1)(D) read much as does the section in final form, and affected only the computation of excess profits net income for base period years, not here involved. H.R. Rep. No. 2894, 76th Cong., 3d Sess. 13, 14, 19 (1940). The Senate Finance Committee, however, extended the provision to the computation of excess profits net income for the later years as well — the years involved in the instant case — to apply whether income was to be based on either income credit or invested capital method, thereby proposing an amended section 711(a)(2) which had it been enacted would have disallowed "the deduction otherwise allowable under section 23(a) for expenses paid or incurred in connection with such retirement or discharge (including any premium paid upon any such retirement or discharge), the deduction for losses otherwise allowable in such connection, and the deduction otherwise allowable on account of the issuance of the bonds or other evidence of indebtedness at a discount." S.Rep. No. 2114, 76th Cong., 3d Sess. 11-12 (1940). At the same time the committee was in the quoted words speaking of disallowing "expenses" and "losses," plaintiff emphasizes, the committee spoke of "interest" as the subject matter of section 711(a)(2)(B). S.Rep. No. 2114, supra, at 12. The conference committee report eliminated the proposed change for all years, retaining it only in section 711(b)(1)(D) for the base years. H.R. Conf.Rep. No. 3002, 76th Cong., 3d Sess. 46 (1940).

Plaintiff would see in the Senate's bill an understanding of Congress that the deductions for amortized bond discount, unamortized bond discount and call premium were expenses and losses rather than interest. In other words, that proposed change in the law was on the premise that discount and premium were not, as interest, already subject to section 711(a)(2)(B).

The contention aggrandizes an ambiguous legislative incident into the full-blown status of an intent of the whole Congress of material significance on the construction question which is presented. It may be, as plaintiff contends, that the premise and thus the understanding of the framers of the Senate bill was that call premiums and unamortized bond discount were not interest covered by section 711(a)(2)(B). It may also be that though believing that such items, as interest, were subject to section 711(a)(2)(B) and thus to halving only, the draftsmen desired to change the law and achieve a 100 percent disallowance, and thought that section 711(b)(1)(D) was a convenient vehicle because discount and the like were also expenses and losses. We do not know the premise of the Senate bill. We do know that the conference committee rejected the bill as drawn in the Senate, and so whatever was the premise of the Senate bill, it was not Congress' premise.

The conference committee's thinking is equally unknown. Its report gives only the fact of what it did. We do not know if it felt that discount and call premiums were interest, already covered by 711(a)(2)(B), and should not be treated as losses and expenses under section 23, or whether it felt that such items were not interest, but should nevertheless not be disallowed in full. But whatever were the conference committee's views, it cannot be said that it concurred in the version of the Senate's view which squares with the plaintiff's position.

The Senate's view, whatever it was, was surely not the intent of Congress in enacting or construing section 711(a)(2)(B). The understanding of existing law by one house in the course of legislation, at least in such ambiguous circumstances as are here present, cannot be given effect as the intent of Congress. Too many doubts and questions would surround the result, and as has been said, it is the function of legislative history to resolve doubts and not to create them. The indecisive evidence of legislative intent presented by plaintiff may not gainsay the abundant authority, set out above, supporting the conclusion that amortized bond discount, unamortized bond discount and call premiums paid on retirement of bonds are, all, interest under section 711(a)(2)(B).

The Government is entitled to prevail on additional defense 6.

X. DONATIONS AND CREDITS

By count 17 and related counts, plaintiff seeks to add $3,675,562.72 to its equity invested capital for the year 1942 and $9,798,364.02 and $9,891,701.23 to the equity invested capital of its subsidiaries and itself for the respective years 1940 and 1941, when it filed returns consolidated with its subsidiaries. The additions would decrease plaintiff's tax by increasing its excess profits tax credit for 1942 and its unused consolidated excess profits tax credit carryover from 1940 and 1941 to 1942.

These sums are the account balances, as of the close of the years in question, representing facilities which were constructed with or composed of the cash and property transferred by nonstockholders to plaintiff and its subsidiaries in literally thousands of "donations and grants" from nonshareholders during the years 1914 through 1942. No income tax was paid on the receipts. They were recorded as assets in the books of plaintiff and its subsidiaries (henceforth, together, called the plaintiff) and were held in the tax years in question for use in plaintiff's trade or business.

The Commissioner of Internal Revenue with a few exceptions disallowed the inclusion of these manifold donations and grants in plaintiff's equity invested capital. Plaintiff challenges the disallowance. In a defense of setoff, the Government challenges the Commissioner's allowance of the inclusion in equity invested capital of six donations and grants.

The parties have by agreement reduced the thousands of transfers involved to 56, to represent all the transfers, and they have also agreed, in the course of their proposals for findings, upon individual transfers to represent the several classes of transfers. These classes are as follows, the amounts stated being those for transfers made prior to January 1, 1940:

Class 1, by far the largest, accounting for $5,992,110 of the $9,798,364 involved, is governmental transfers to relocate line on account of dams. It is represented by transactions in which the Federal Government paid the plaintiff the cost of relocating or protecting such parts of its line as would be flooded or threatened by a rise in water level by reason of a Government dam about to be built.

Class 2, $251,669, governmental transfers to relocate line on account of highways, is represented by a case in which the City of Long Beach, as part of a plan to build a new highway, by agreement paid the plaintiff $240,000 for the latter's right to operate on certain streets and the right to use, for pedestrian an and vehicular traffic, the plaintiff's drawbridge across a harbor entrance.

Classes 3 and 4, $1,947,206, governmental transfers for highway underpasses and other highway construction, are represented by transfers in which a state or city, in the interest of public convenience and safety, and in many cases using federal funds made available for the purpose, paid the cost of railroad highway crossings such as a new grade crossing or the replacement of a viaduct with a highway subway under the railroad's line. These transfers are governed by United States v. Chicago, Burlington Quincy R.R., 412 U.S. 401, 93 S.Ct. 2169, 37 L.Ed.2d 30 (1973), in which substantially identical transfers were held not to effect contributions to capital.

Class 5, $320,601, governmental miscellaneous transfers, is divisible as follows: (1) 12 percent is represented by transfers substantially identical with classes 3 and 4; (2) 87.2 percent is represented by payments for fences and street lights on land leased from the plaintiff, cables for utility lines, water mains and irrigation waterways under the right-of-way, and (3) 0.8 percent is represented by a transfer to build sanitary facilities on the plaintiff's premises.

Class 6, $891,354, private transfers for spur and other tracks, is represented by several transfers by shippers for the construction of spur tracks to the transferor's plant or installation and two transfers by nonshippers, one identical with those in class 1, supra, except that the transferor was a private power company, and the other a railroad with whom plaintiff maintained a joint installation at a highway crossing.

Class 7, $395,422, private miscellaneous transfers, is represented by transfers involving payments for street lights on leased property paid for by the lessees; a private road and gates across the right-of-way for the use of, and paid for by, the owner of the land on both sides; feed racks and a scale paid for by a stock yard for its use; a culvert needed and paid for by a water company; a storage facility and a power line for the use of an express company, paid for by it; and a retaining wall needed for a spur track to a shipper's plant, paid for by the shipper.

The issue is whether the facilities built with the cash and property transferred may properly be treated as "money" and "property" "paid in" by nonshareholders "as a contribution to capital," and thus includible in equity invested capital under the 1939 Code. See Treasury Regulations 112 § 35.718-1 (1944).

The dispute centers first on classes 1 and 2 of the transfers in question. As noted above, classes 3 and 4 are governed by Chicago, Burlington Quincy R.R., supra, and classes 5-7 are of a miscellaneous nature whose disposition will best be discussed after decision on classes 1 and 2.

In briefs filed before the decision of Chicago, Burlington Quincy R.R., supra, plaintiff contended that the transfers were made to induce the construction and operation of its railroad for the service and safety of the public and were therefore contributions to capital under Edwards v. Cuba R.R., 268 U.S. 628, 45 S.Ct. 614, 69 L.Ed. 1124 (1925). See Texas Pacific Ry. v. United States, 286 U.S. 285, 52 S.Ct. 528, 76 L.Ed. 1108 (1932). Edwards v. Cuba R.R., supra, held that payments of money and property by the Government of Cuba to a railroad corporation, conditioned on the construction of a railroad line — so much per mile — were made as reimbursement for capital expenditures rather than as a gift or to obtain rate concessions and in consequence were not profits or gains taxable as income. The distinctive feature of the transaction was that the Cuban Government was acting deliberately to induce the construction of a railroad, and to promote its success by making a grant towards its capital. The transfers were akin to those made by Congress to the first Union Pacific Railroad, in the Pacific Railway Acts in the last century. Such transfers are utterly unlike those presently in question. Edwards v. Cuba R.R. is therefore not helpful.

The remaining arguments of the parties are based on Detroit Edison Co. v. Commissioner, 319 U.S. 98, 63 S.Ct. 902, 87 L.Ed. 1286 (1943) and Brown Shoe Co. v. Commissioner, 339 U.S. 583, 70 S.Ct. 820, 94 L.Ed. 1081 (1950), from which the Supreme Court in Chicago, Burlington Quincy R.R., supra, recently distilled a number of the characteristics of a contribution to capital under the Code.

A contribution to capital, the court wrote, "must become a permanent part of the transferee's working capital structure"; it "may not be compensation, such as a direct payment for a specific, quantifiable service provided for the transferor by the transferee." It "must be bargained for" (and thus assets granted by a governmental body for replacement of existing facilities, where none otherwise would have been deemed necessary, are not contributions to capital), and the asset transferred "forseeably must result in benefit to the transferee in an amount commensurate with its value." Finally, "the asset ordinarily, if not always, will be employed in or contribute to the production of additional income and its value assured in that respect," and thus assets intended for the safety of the public are not contributions to capital, because they were "peripheral to the road's business" and "did not materially contribute to the production of further income by the railroad." 412 U.S. at 413-414, 93 S.Ct. at 2176.

As already noted, the facts of classes 3 and 4 of the transfers in the instant case are substantially identical with those in Chicago, Burlington Quincy R.R., supra, and thus the decision there decides that classes 3 and 4 here are not contributions to capital.

The other classes may be summarily disposed of in the light of the characteristics of a contribution to capital set out in Chicago, Burlington Quincy R.R., supra.

The transfers in class 1, typified by a transfer by the Federal Government to the plaintiff to replace a portion of the right-of-way to be flooded by a projected dam, were matter-of-fact business transactions in which the parties made an equal exchange, without altruism or donative intent. The closest analogy is a condemnation proceeding; no one would contend that payment of a condemnation judgment or of a sum in settlement to avoid an eminent domain proceeding is a contribution to capital. In the words of the majority opinion in Chicago, Burlington Quincy R.R., supra, the transfers in class 1 "simply replaced existing facilities" and "did not materially contribute to the production of further income by the railroad." 412 U.S. at 414, 93 S.Ct. at 2176.

In Los Angeles S. L. R. R. v. United States, 21 F. Supp. 347, 86 Ct.Cl. 87 (1947), a railroad (actually a subsidiary of the present plaintiff) gave up a portion of its line to a mining company in need of the land for an extension of the mine's tailings dumps and received in return a new line as a replacement. This court said that the new line, though costlier than the old, "was of no more use to the railway company than the old line and would not produce a cent more income" or increase the value of the railroad's assets "by a single dollar." 21 F. Supp. at 353, 86 Ct.Cl. at 99-100.

Class 2 transfers are essentially the same as those in class 1 and are equally with class 1 not contributions to capital. Classes 3 and 4, as already noted, are not contributions to capital on the authority of Chicago, Burlington Quincy R.R., supra; 12 percent of the transfers in class 5, found to be essentially the same as the transfers on classes 3 and 4, are also not contributions to capital.

Of the remaining 88 percent of the transfers in class 5, 87.2 percent are represented by these transfers: (1) a state prison, as required by its lease of a portion of the plaintiff's right-of-way for use as a pasture, paid for a "hog-tight" fence for the prison farm; (2) a state paid its share of a municipal assessment for street lighting on the railroad's grounds at a railroad station, in accordance with the easement contract covering encroachment of the state's highway on the station grounds; (3) a municipal department of power and light paid for the cost of cables for telephone and telegraph lines, to replace open wire lines; the record gives no further details; (4) a town paid for the cost of a water main, under the right-of-way, to the municipal stockyard; (5) and (6) federal irrigation agencies paid for the cost of irrigation waterways under the plaintiff's line.

In all of these it appears that the transfers were essentially an exchange of values or a payment for a specific quid pro quo which left the plaintiff transferee no better off than before and did not materially contribute to the production of further or additional income. Accordingly, under United States v. Chicago, Burlington Quincy R.R., supra, the transfers did not effect contributions to the capital of the plaintiff.

The final subgroup of 0.8 percent of the transfers in class 5 is represented by a transfer in 1938 in which a town in Kansas furnished $129 worth of W.P.A. labor to construct four new sanitary privies, notice having been given the railroad by the town to abate the nuisance of unsanitary privies. The scanty record leaves a net impression that in this transfer the town intended, in the interest of the users of the facilities, to confer a benefit upon the railroad, and that the transfer replaced existing facilities with new and better ones which plaintiff would otherwise have been required to construct out of its capital funds and thus that the transfer resulted in a benefit to the transferee in an amount commensurate with its value in that it enabled the plaintiff to avoid a capital expenditure to the value of the assets transferred. The sum whose expenditure was avoided was employed in the production of further or additional income. This group of transfers therefore meets the Chicago, Burlington Quincy R.R. test for a contribution to capital.

The Government maintains that in any event the thing contributed was services and neither "money" nor "property," as required by § 718 for inclusion in equity invested capital. There is authority that services compensated with stock are neither "money" nor "property" includible in invested capital. Bard-Parker Co. v. Commissioner, 218 F.2d 52 (2d Cir. 1954), cert. denied, 349 U.S. 906, 75 S.Ct. 582, 99 L.Ed. 1242 (1955); Western Maryland Ry. v. United States, 227 F.2d 576 (4th Cir. 1955), cert. denied, 351 U.S. 907, 76 S.Ct. 696, 100 L.Ed. 1443 (1956). The more pointed cases, however, albeit decided under an earlier excess profits tax act, recognize that such one-time services as those of architects and engineers which go directly into the creation of a tangible capital asset are so sufficiently reflected in capital assets that their value is includible in invested capital. Federal Plate Glass Co. v. Commissioner, 6 BTA 351 (1927); Coatesville Boiler Works v. Commissioner, 9 BTA 1242 (1928); see Palomar Laundry v. Commissioner, 7 TC 1300 (1946). By the thinking of those cases, the cost of the labor used in building the privy is, as much as the lumber and roofing used, a capital asset and includible in equity invested capital. Money actually passed to workmen who labored to build a structure which became a capital asset. No case holds the cost of such labor not includible in equity invested capital. No principle or policy requires that it be not includible, for the transaction is wholly realistic, without any possibility of exaggeration or evasion. See Union Pacific R.R. v. United States, 401 F.2d 778, 185 Ct.Cl. 393 (1968), cert. denied, 395 U.S. 944, 89 S.Ct. 2017, 23 L.Ed.2d 462 (1969), rehearing denied, 194 Ct.Cl. 1021, cert. denied, 403 U.S. 931, 91 S.Ct. 2254, 29 L.Ed.2d 710 (1971).

In the nongovernmental transfers, classes 6 and 7 described above, the transfers were substantially identical to those in class 1 and to the transfers comprising 87.2 percent of the transfers in class 5, or were direct payments for specific things or services. The transfers in those two classes, therefore, are under Chicago, Burlington Quincy R.R., supra, not contributions to capital.

The transfers challenged in the plea of setoff, and the decisions thereon, are as follows:

(1) $200,000 for a spur track to a smelter about to be built, paid for in 1901 by the owner of the smelter; held a payment for a specific quantifiable service and therefore not a capital contribution under Chicago, Burlington Quincy R. R., supra.

(2) $1,076 for railroad line from Orchard, Idaho, to Boise, Idaho, to connect with the existing line from Boise to Nampa, to provide through train service for Boise; paid for in 1925 by the Chamber of Commerce of Boise, Idaho; held this transfer was not a payment for a thing or a service, was bargained for, resulted in benefit to the transferee in an amount commensurate with its value, and the assets transferred were employed in the production of further or additional income. It was therefore a contribution to capital under Chicago, Burlington Quincy R.R., supra, Brown Shoe and Edwards v. Cuba R. R., supra. See Federated Department Store v. Commissioner, 426 F.2d 417, 420 (6th Cir. 1970).

(3) $28,338 for land for a new line from Rogerson, Idaho, to Wells, Nevada, provided in 1925 by a citizens right-of-way committee for the purpose of obtaining a more direct outlet to the California market for Southern Idaho agricultural products, and to open up for tonnage shipments numerous copper mining properties adjacent to the new line. This transfer had the same characteristics as the immediately foregoing transfer and is equally a capital contribution.

(4) $100,000 paid by a citizens committee in 1928 to acquire and transfer to plaintiff a small road which had ceased operations, in consideration of plaintiff's promise to operate it permanently. This transfer was essentially similar to the two foregoing transfers and is equally a capital contribution.

(5) $46,125 paid by the Utah Copper Company to reimburse the plaintiff for the additional expenses of operating a relocated line, in the circumstances detailed in Los Angeles S. L. R. R., v. United States, supra. This transfer is held not a capital contribution for the reasons stated above in connection with classes 1 and 2.

(6) $240,421.01 transferred by a citizens committee in 1925. Plaintiff has conceded that defendant is entitled to prevail on this transfer.

XI. EQUITY INVESTED CAPITAL

Plaintiff also seeks a refund of taxes paid on excess profits based on an alleged erroneous determination by the Commissioner of Internal Revenue [Commissioner] of its 1942 equity invested capital credit. Plaintiff contends that the Commissioner undervalued its equity investment and, therefore, unduly restricted its credit. In a most unusual counterattack, defendant also asserts that the Commissioner erred in his equity investment appraisal. However, defendant claims that the Commissioner overvalued plaintiff's equity. It, therefore, seeks to "offset" any other refunds due plaintiff. Significantly, neither party supports the Commissioner's determination.

The Excess Profits Tax Act of 1940 grants taxpayers a "credit" for equity invested capital. The Act defines equity invested capital as money and property "paid in" for stock. Where a corporation is organized by issuing stock for property, the property received (invested capital) is valued with reference to the fair market value of the stock issued.

In 1898 plaintiff received the assets of the old bankrupt Union Pacific Railroad [U.P.]. On January 31, 1898, plaintiff issued 610,000 shares of $100 par common stock and 750,000 shares of $100 par, four percent preferred stock. In return, it received the stock of the old U.P. and cash (hereinafter referred to as the reorganization). Later in 1899, plaintiff acquired control of the Oregon Short Line Railroad [Oregon] by issuing an additional 273,493 shares of $100 par common for the Oregon common and cash (hereinafter referred to as the acquisition).

Our problem, finding the correct amount of U.P.'s 1942 excess profits tax credit, then resolves into a determination of the value of the U.P. stock issued in these nineteenth century transactions.

In its return for 1942, plaintiff used par values to compute the value of its stock issued for these assets ($163.3 million). The Commissioner disputed this valuation and, at first, attached a value of $122.5 million to the shares. He later modified the stock values to $79.4 million, lowering plaintiff's valuation substantially. Plaintiff paid the 1942 deficiency and filed this refund suit claiming that its original return was correct. In addition to the stock issued, plaintiff also asserted that it was entitled to include the cash contributions received in the reorganization ($9.1 million) and the acquisition ($0.8 million) in its equity invested capital.

Defendant claimed that the value of plaintiff's stock was less than even the Commissioner had determined and asserted an offset against other amounts recoverable by plaintiff. Defendant argued in its pleadings that the fair market value of the U.P. stock was only $57.5 million.

During the trial, plaintiff sought to prove an even greater valuation, and defendant continued to support the value that it had alleged in its pleadings. Plaintiff's valuation expert testified that the proper valuation date for the reorganization stock was January 31, 1898. However, he stated that the true value of the stock must take into account the U.P.'s rapid rise in fortunes during the post-reorganization period. Thus he evaluated the U.P. reorganization and acquisition stock by use of 1907 stock market figures and found values of $175-200 per share for the common and $100 per share for the preferred. His total value estimate was $251.7 million. Defendant's trial expert testified that after considering all valuation techniques, the reorganization common was worth $22.50 per share and the reorganization preferred, $40 per share. Since he rated the acquisition common at $10 million, he assessed plaintiff's total equity invested capital at $57.5 million.

Both plaintiff and defendant contend that the Commissioner's valuation was incorrect and both ask the court to find a per share value for the stock. Significantly, there is absolutely no justification in the record for the Commissioner's valuation since neither party supported it.

The threshold question is whether either party's evidence overcomes the Commissioner's presumption of correctness.

A. Presumption of Correctness:

The Commissioner's determination of taxes due is entitled to a presumption of correctness. Helvering v. Taylor, 293 U.S. 507, 55 S.Ct. 287, 79 L.Ed. 623 (1935); Northlich, Stolley, Inc. v. United States, 368 F.2d 272, 177 Ct.Cl. 435, 442 (1966). This presumption applies to excess profits tax credit determinations. Tri-State Realty Co. v. Commissioner, 180 F.2d 593 (5th Cir. 1950). However, in our case an unusual situation is presented because neither party supported the Commissioner's determination. Each attempted to assert its own conclusion for asset valuation.

The court believes that there is sufficient evidence in the record to rebut the presumption. Presumptions are not evidence, and they disappear in the face of substantive evidence tending to disprove them. United Aniline Co. v. Commissioner, 316 F.2d 701, 704 (1st Cir. 1963); Kentucky Trust Co. v. Glenn, 217 F.2d 462, 465 (6th Cir. 1954). Defendant provided sufficient probative evidence of the U.P. value by offering stock market prices for the U.P. stock to rebut the Commissioner's presumption of correctness.

Once we find that the presumption has been rebutted, we must then determine whether either party has met its burden of proof and, therefore, is entitled to a refund or offset.

B. Burden of Proof:

Where plaintiff sues for a refund he has the burden of proving that the refund is "legally due" him. Helvering v. Taylor, supra; Lewis v. Reynolds, 284 U.S. 281, 52 S.Ct. 145, 76 L.Ed. 293 (1932). Where defendant counters with an offset claim, allocation of the burden depends on the nature of the offset. If the offset is based upon the same tax return as plaintiff's refund claim, the ultimate burden of proof remains on plaintiff. However, defendant has the burden of "coming forward" with sufficient facts to show that it "has a reasonable basis in fact or in law for its setoff." Missouri Pacific R.R. Co. v. United States, 338 F.2d 668, 168 Ct.Cl. 86 (1964). In the instant case, plaintiff has the ultimate burden of proving that the Commissioner undervalued its equity. Since its offset claim involves the same tax return, defendant must "come forward" with sufficient evidence to show that there is a reasonable basis for its claim that the Commissioner overvalued the U.P. stock.

Valuation of assets is a question of fact. American Steel Foundries v. United States, 153 Ct.Cl. 234 (1961). The question before the court is the value of the U.P. investment. The court may adopt plaintiff's conclusions, may adopt defendant's or any reasonable value in between. Toronto, Hamilton Buffalo Nav. Co. v. United States, 88 F. Supp. 1016, 1022, 116 Ct.Cl. 184, 207-208 (1950). We find from the stock market prices presented by defendant that it has met its burden of coming forward. There is sufficient evidence to find a reasonable basis for defendant's allegation of overvaluation by the Commissioner. We also find that plaintiff has failed to meet its burden of proving that the Commissioner undervalued the investment. We hold that defendant is entitled to an offset.

C. Equity Valuation:

Valuation of property is at best an inexact science or highly imprecise art, and there are as many approaches to valuation as there are valuation experts. Asset appraisal, therefore, requires a reasonable or rational approximation rather than exactitude. Primary valuation techniques include actual sale prices, actual or original cost, replacement cost, capitalized income, price-earnings ratios, and comparison with similar property. I J. Bonbright, Valuation 113-269 (1937). Each method has its own variations, strengths and weaknesses. We have received evidence of U.P. value based upon stock market prices, not asset values, capitalization of earnings, price-earnings ratios and comparison of the U.P. with similar stock.

While each valuation method has shortcomings, we find sufficient evidence to adduce a reasonably accurate value for U.P.'s reorganization and acquisition stock.

The departure point for our valuation inquiry begins with the proposition that "fair market value is the price at which property would change hands in a transaction between a willing buyer and a willing seller, neither being under a compulsion to buy or sell, and both being reasonably informed as to all relevant facts." Bankers Trust Co. v. United States, 518 F.2d 1210, 1219, 207 Ct.Cl. ___, ___ (1975); Jack Daniel Distillery v. United States, 379 F.2d 569, 574, 180 Ct.Cl. 308, 315-316 (1967). This court has frequently used stock market prices to value stock. As Judge Davis noted in Bankers Trust, "Where stock is freely traded in an open, organized market, stock exchange quotations for the valuation date generally provide the best evidence of value." Bankers Trust Co., supra, at 1219, 207 Ct.Cl. at ___ citing Moore-McCormack Lines, Inc. v. Commissioner, 44 T.C. 745, 759 (1965); Southern Natural Gas Co. v. United States, 412 F.2d 1222, 1252, 188 Ct.Cl. 302, 351-352 (1969); 10 J. Mertens, The Law of Federal Income Taxation §§ 59.13 at 42-43, 59.14 at 47 (1970). The Bankers Trust opinion contains an excellent analysis of the situations in which stock market values are not accurate value determinants. We note that such situations are not present in the instant valuation.

Two factors which give us pause in considering stock market values are the "pessimistic" nature of the market in 1898-99, and the likelihood of reacquisition of the Oregon stock at the time of reorganization. While these factors might possibly cause some doubt as to the reliability of market evaluation, we find that, in the instant case, they do not. The pessimism of investors in the then current economic situation was well-founded. The nation had undergone a severe depression, and recovery was not a foregone conclusion. In addition, the U.P. itself had recently undergone bankruptcy and was in the process of reorganization. It total mileage had shrunk, and it had no West Coast outlet. The only factors which gave the U.P. any value at all were the possibility of economic recovery and reacquisition of branch lines lost during bankruptcy. We conclude that rather than being of doubtful worth in estimating the value of the U.P. stock, the market serves as its most accurate indicator. In short, the stock exchange values produce the most accurate and comprehensive analysis of the U.P. stock possible after a lapse of some 77 years.

Having determined that we will employ stock market values, we must then decide the dates we will use to appraise the stock. The Excess Profits Tax Regulations mandate valuation at the time of issue. Clearly market quotations are more reliable if they are near in time to the event in question. 10 J. Merten, Law of Federal Income Taxation, § 59.16 (1970). In the instant case, the reorganization stock was issued on January 31, 1898. Reorganization preferred first traded on February 2, 1898. Since this date is fairly contemporaneous with the organization, February 2, 1898 becomes the logical date upon which to rate the reorganization preferred. We therefore adopt February 2, 1898. The reorganization common was not traded until February 25, 1898. This is somewhat remote. However, we can derive market values for the reorganization common on February 2, 1898 from the price of U.P. Trust Receipts. Thus we again use the February 2, 1898 valuation date to assess the reorganization common.

Determining valuation dates for the acquisition common presents another problem. The U.P. issued this stock in several blocks over a period of ten years. Here we use average market prices during the time period that the U.P. issued each block of shares.

After establishing valuation dates, we must next settle upon a technique to determine the specific market price on the date in question. Although found in the Estate Tax Regulations, a generally accepted procedure in income tax cases is to use the average between high and low prices on the date in question. We accept this method. The average between high and low market values on February 2, 1898 represents the most accurate value for the reorganization stock. The average between high and low for each date during the several time periods in which the U.P. issued its acquisition stock is again the most accurate value for the acquisition stock.

Our first measurement concerns the value of the 750,000 shares of U.P. preferred issued during the reorganization. The reorganization preferred was first traded on February 2, 1898. The average between high and low for the preferred on this date was $62.50. Thus the court finds that the 750,000 shares of U.P. preferred issued in the reorganization were worth $46.9 million. Note that 122,644 shares of U.P. preferred were traded during the first week of issue, an average of 24,549 per day. Although this is a fairly large block of stock, it is nowhere near the entire amount. Thus there were sufficient shares of U.P. preferred traded at the time of issue to support the validity of the price set by the market, yet not so many shares that the market failed to absorb them.

Evaluation of the reorganization common presents another problem. The first shares of U.P. common were not traded until nearly a month after the reorganization, February 25, 1898. However, U.P. Trust Receipts, exchangeable for one share of U.P. common and 15/100 of a share of preferred were traded on the market at the time of the reorganization. Using February 2, 1898 Trust Receipt values (an average of $36 per share), less 15/100 of the $62.50 value found for the preferred ($9.375), we derived a value of $26.625 for the U.P. common on February 2, 1898. Thus 610,000 shares of new U.P. common were worth $16.2 million. Again we note that approximately 35,321 Trust Receipts per day were traded during the week in question. Again this is sufficient number of shares to find that the market set an accurate trading price.

The U.P. issued the acquisition stock in blocks over a ten-year period. Defendant has presented evidence which permits the court to separate this stock into several blocks and identify an average market price for each block. We adopt this evidence and find the following values for the U.P. acquisition stock.

The reorganization common, the reorganization preferred, and the acquisition common yield total U.P. equity invested capital of $75.9 million. The court believes that this value is neither too pessimistic nor too optimistic. In each case, there are sufficient market sales to bolster the validity of the price placed upon the shares in the marketplace, yet not so many shares traded that the market found itself unable to absorb them.

In addition, we note that other evidence of value in the record supports our determination. First, although unsupported, the Commissioner's value of $79.4 million is close to that found by the court. Second, by using the net asset values placed in the record by defendant, $116.9-$119.9 million, plus a 15 percent estimate for going concern value and goodwill (factors ignored by defendant in its testimony on net asset values), less debt issued by U.P. during reorganization ($81.7 million in bonds then selling at 92 or $75.2 million in debt securities), we derive a value for the U.P. reorganization assets of $59.2-$62.7 million. Adding the value of the Oregon assets obtained in the acquisition, $12.8 million, we derive a reasonable estimate of total equity based upon net assets and goodwill less debt of $72.0 million to $75.5 million. The stock market values found by the court are nearer the $75.5 million figure. Our stock market figures for the reorganization stock appear to be well-founded.

Finally, viewing the acquisition stock alone, various facts in the record again bolster our determination of value. For example, by using an 1899 average of the price-earnings (P/E) ratios of three comparable railroads (9.8), we find that of the $5.12 per share U.P. earnings in 1899, some $1.4 million in earnings were attributable to the acquisition common. (273,493 shares X $5.12 per share.) Multiplying the P/E ratio for the similar stocks by the earnings attributable to the U.P. acquisition common, we find a reasonable estimate for the Oregon of $13.7 million. Certainly the stock market values ($12.8 million) for the acquisition common are conservative estimates of their value.

Thus the value determined by the Commissioner, the value based upon net assets, and price-earnings ratios lend support to our valuation.

In short, using stock market prices to value the equity invested capital of the U.P. for 1942 excess profits tax purposes, the court finds that the Commissioner overvalued the U.P. stock by some $3.5 million. Accordingly, defendant is entitled to a tax offset for the excess equity invested capital credit granted to plaintiff in 1942 based upon this figure.

Plaintiff's other contention is that it is entitled to include cash contributions received in its equity invested capital in addition to the value of the stock issued. The court rejects this assertion. Since new U.P. stock was issued in exchange for old U.P. common, the Oregon common, and cash, a proper measurement of the equity can be achieved by viewing either side of the equation; the value of the U.P. stock issued or the value of stock and cash received. To allow plaintiff to include both the stock issued and the cash received in its equity invested capital is to "add both sides of the bargain." Certainly the cash received by the U.P. was a partial payment for the stock issued. We, therefore, hold that plaintiff cannot include the cash payments received since we value its equity with reference to the value of the stock issued.

In summary, the court finds that there is sufficient probative evidence to establish reasonable per share values for the U.P. stock issued in the reorganization and acquisition. We find that defendant has overcome the Commissioner's presumption of correctness. The court further finds that defendant has met its burden of establishing that the Commissioner overvalued the U.P. stock. We value the stock as near its issue date as possible. The reorganization preferred was worth $62.50 per share, the reorganization common $26.625 per share, and the acquisition common ranged from $44 per share to $175 per share depending upon the date of issue. The total value of the stock issued, plaintiff's equity invested capital for 1942 excess profits tax credit purposes, was $75.9 million.

To recapitulate, we find that plaintiff is entitled to recover on its refund claim based upon its right to "expense" minimum rule property in its 1942 return (Part I). Plaintiff may also recover on its contention that it would include stock of certain leased line subsidiaries in its capital assets (Part V). Finally, with regard to "donations and grants" given to plaintiff to spur its construction, we conclude that some of these grants were includable capital assets and some were not. Therefore, we hold in part for plaintiff on this assertion (Part X).

We further determine that plaintiff is not entitled to recover refunds for other allegations since it cannot deduct 1943 payroll taxes in 1942 (Part II); it cannot include certain stock subscription rights in its earnings and profits (Part IV); and cannot receive interest on a 1948 agreement with the Commissioner of Internal Revenue (Part VI). Plaintiff also may not recover an offset to setoffs claimed by defendant because it cannot include proceeds from certain land sales in its earnings and profits (Part VII).

Defendant is entitled to prevail on its "additional defense" that plaintiff erroneously included unamortized bond discount and expense in its total assets (Part VIII). Further, we find for defendant on another "additional defense" that plaintiff mistakenly failed to treat other bond discounts and expenses as "interest" (Part IX). We also hold that defendant is entitled to recover an offset based upon its claim that the Commissioner overruled plaintiff's equity invested capital (Part XI).

Finally, the court severs and remands for further consideration by the trial judge the question of whether plaintiff is entitled to increase its earnings and profits by $13 million due to certain accounting errors alleged to have occurred in 1900-1907 (Part III).

Accordingly, we find for plaintiff in Parts I, V and in some sections of Part X. We deny plaintiff's claims in Parts II, IV, VI and VII. We further hold for defendant in Parts VIII, IX and XI. The amount of the recoveries will be determined pursuant to Rule 131(c). Finally, we remand Part III for further proceedings consistent with this opinion.

* * * * * *

"(2) With respect to compensation paid to employees for services rendered during the calendar years 1940, 1941, and 1942, the rate shall be 3 per centum;

"(3) With respect to compensation paid to employees for services rendered during the calendar year[s] 1943, 1944, and 1945, the rate shall be 3 1/4 per centum;

* * * * * *

Sec. 8. (a) Every employer shall pay a contribution, with respect to having employees in his service, equal to 3 per centum of so much of the compensation as is not in excess of $300 for any calendar month payable by him to any employee with respect to employment after June 30, 1939: * * *."