These appeals bring up questions concerning the proper method of computing corporate excess profits taxes for the years 1942 and 1943 and of interest upon asserted deficiencies in their payment. The taxpayers' appeals in both the actions before us present the interesting contention that a corporation may accept a higher tax than the commissioner allows in order thus to take advantage of a higher refund under the statutory postwar refund of 10 per cent herein applicable. And the government's appeal in the first action, taken for the administratrix of the estate of a deceased collector who is the defendant of record in both actions, asserts that interest must be separately assessed in favor of the taxpayer and the government upon offsetting excesses and deficiencies in income and excess profits taxes.

The Internal Revenue Code, 26 U.S.C., provides alternative methods of computing excess profits taxes covering the years in question. Section 713 allowed an excess profits credit, in the case of a domestic corporation, of 95 per cent of average net income over a base period. Section 714 alternatively provided for such a credit based upon stated percentages of invested capital. Section 712(a) directed that the credit to be allowed shall be an amount computed under § 713 or § 714, 'whichever amount results in the lesser tax under this subchapter * * *.' Section 26(e) adjusted ordinary income tax to the excess profits tax by in substance relieving of ordinary tax the amount of income subject to excess profits tax; and § 710(a)(1)(B) placed an 80 per cent ceiling on the combined income and excess profits taxes.

Taxpayers The Babcock & Wilcox Company and The Babcock & Wilcox Tube Company, plaintiff and plaintiffs below in these actions for tax refund, found themselves in a position where under either calculation the 80 per cent ceiling of § 710(a)(1)(B) would govern to limit the total tax. But in 1942 an addition was made, by I.R.C. § 780, to the provisions for this tax (originally established in 1940) whereby upon the cessation of hostilities of the then war, a refund of 10 per cent would apply. With an eye to a greater refund the taxpayers so calculated their returns as to maximize that portion of their over-all 80 per cent liability allocable to the excess profits tax and to minimize that due on the ordinary income tax. Their claim for credit under § 780 calculated on this maximized basis was disallowed and they brought these actions. The only difference in this regard in the two actions is that, as is permissible, the taxpayer in the first action relied on the average net income method of calculation, and in the second on the invested capital method.

The plaintiffs' quite ingenious argument on the issue of tax computation, an argument which Judge Holtzoff rejected below in granting defendant's motions for summary judgments, presents with some elaboration the contention that Congress, since it was granting these alternative methods of tax computation as a measure of relief to hard pressed taxpayers, must have desired such choice to extend to-- and to be operable in reverse, so to speak, in-- the application of the tax refund later to be made, in order that full measure of relief be accorded. But we agree with Judge Holtzoff that this hardly rises above the level of supposition, however shrewd, and cannot vary the plain language of the statute. We see little, if anything, to be added to his reasoning and, so far as concerns this issue, accordingly affirm on his opinion, D.C.S.D.N.Y., 98 F. Supp. 548-551.

The question of interest requires more extensive consideration. This is presented only in the first action brought by plaintiff, The Babcock & Wilcox Company, and comes before us upon the defendant's appeal from the grant of plaintiff's motion for summary judgment for the refund. It arises as follows: When the collector recalculated the excess profits tax liability of this plaintiff to find an 'overpayment' upon its chosen method of computation, he necessarily found a corresponding 'deficiency' of like amount in its ordinary income tax. In accordance with the interest provisions of §§ 292(a) and 3771(b)(1), set forth below, he allowed interest on the overpayment and assessed interest on the deficiency. Due to the different dates from which the two computations run, he determined and assessed a net interest liability of $52,621.84, which plaintiff paid and for which it now sues.

It is plaintiff's position that the government had in its possession at all times all the funds to which it was entitled. Accordingly the overpayment and deficiency were fictitious bookkeeping transactions and the government was in no way injured. Such interest is designed to compensate for lost use of funds, it cannot here apply. This argument found favor in the district court. D.C.S.D.N.Y., 98 F. Supp. 548, 551.

Against this contention based upon equities, the government vigorously urges the specific mandates of the cited statutes. Section 292(a) provides that deficiency interest shall be computed 'from the date prescribed for the payment of the tax (or, if the tax is paid in installments, from the date prescribed for the payment of the first installment) * * *.' This, for each of the years in question, was the following March 15. Section 3771(b)(1) directs that interest on overpayments shall run 'from the date of the overpayment,' which under § 321 infra and Blair v. U.S. ex rel. Birkenstock, 271 U.S. 348, 46 S. Ct. 506, 70 L. Ed. 983, is the date when the total of installments paid exceeds the total tax liability for the year. In our case this was December 15 following each taxable year.

The district judge did not meet this argument by specific rationale or precedent, but relied generally upon the equities, saying: 'No money changed hands. The Government was not deprived of the use of any money. To charge the taxpayer with additional interest under the circumstances is obviously unfair and inequitable.' D.C.S.D.N.Y., 98 F.Supp. at page 551. But, however appealing, this is somewhat dangerous ground for the testing of taxes. To many persons various parts of the taxing policy seem unfair, e.g., the taxation of capital gains or the entire scheme of progressive taxation. Compare discussion by Randolph Paul in Book Review of Blum & Kalven, The Uneasy Case for Progressive Taxation, in 67 Harv.L.Rev. 725-731. 'The Internal Revenue Code, not general equitable principles, is the mainspring of the Board's jurisdiction.' C.I.R. v. Gooch Milling & Elevator Co., 320 U.S. 418, 422, 64 S. Ct. 184, 186, 88 L. Ed. 139. And so we must turn to the statutes.

To meet the question of statutory authority plaintiff relies on the broad provisions for statutory setoff and credit on tax overpayment. I.R.C. § 321 provides that if the taxpayer has paid as an installment or in full more than the correct amount 'of the tax' the overpayment shall be credited or refunded as provided in § 322; while § 322(a)(1) provides inter alia that overpayment 'of any tax imposed by this chapter' shall be credited 'against any income, war-profits or excess-profits tax or installment thereof then due from the taxpayer.' Plaintiff asserts that this context shows the reference to 'the tax' as necessarily including both forms of taxes and that hence there was here but a single payment due (in installments) to cover both taxes. But the principle of credit allowances here stated is a broad one, so broad in fact as to make clear that it applies after the offsetting amounts are ascertained and not to their ascertainment. Thus it applies as between different forms of tax, Kjar v. United States, 108 Ct. Cl. 119, 69 F. Supp. 406, certiorari denied 332 U.S. 768, 68 S. Ct. 79, 92 L. Ed. 353 (income tax and distilled spirits tax), or as against taxes of different years, Noyes v. United States, 9 Cir., 55 F.2d 870. But it does not and cannot be used to override statutory impositions elsewhere levied, as the cases discussed below (among others) demonstrate. Like considerations apply to the doctrine of equitable recoupment which has been limitedly applied both for and against the taxpayer, Bull v. United States, 295 U.S. 247, 55 S. Ct. 695, 79 L. Ed. 1421; Stone v. White, 301 U.S. 532, 57 S. Ct. 851, 81 L. Ed. 1265; as the later cases strictly restricting application of the doctrine show, it cannot be employed to dissipate specific tax obligations. C.I.R. v. Gooch Milling & Elevator Co., supra, 320 U.S. 418, 64 S. Ct. 184, 88 L. Ed. 139; Rothensies v. Electric Storage Battery Co., 329 U.S. 296, 67 S. Ct. 271, 91 L. Ed. 296.1

Returning therefore to the statutes, we perceive that the issue comes down in essence to the question whether the excess profits tax under the Code may properly be considered as a part of the corporate income tax or whether it is essentially a separate tax. For, once it is determined that the tax is separate, the statutory mandate is clear. The tax each year was due, at least in part, and the deficiency arose on March 15. I.R.C. § 56(a, b). There was no overpayment to apply to it until December 15. I.R.C. § 321; Blair v. U.S. 348, 46 S. Ct. 506, 70 L. Ed. 983. Whether, thereafter, § 322 would serve to apply the overpayment to the deficiency or whether interest arose on both, cancelling each other out, is immaterial. For the interest in dispute arose between March and December when no overpayment existed.

But the point of substantial identity of the two taxes (except of course as to rates) appears to be settled against the contention both by the statutes themselves and by the precedents. The excess profits tax and the ordinary income tax are not one, but are made and levied in separate and distinct manner. They are imposed by separate legislation separately enacted. Thus Subchapter E-- Excess Profits Tax, §§ 710-783, was added to the Code by Act of Oct. 8, 1940, and was repealed (for future years) on Nov. 8, 1945, 59 Stat. 568, without impairment of the regular corporate income tax. The point is made quite clear on a careful analysis of the background of the legislation made by Judge Chase in Superheater Co. v. C.I.R., 2 Cir., 125 F.2d 514, 516, in finding the background of the excess profits tax in the former capital stock tax and in affirming a ruling of the Board of Tax Appeals that it did not have jurisdiction to redetermine the excess profits tax of a petitioner on a petition to review the determination of a deficiency only in the ordinary taxes of the petitioner for the same period.2 Plaintiff objects that this is a matter affecting the jurisdiction of the Board or Tax Court only, and not in point as to an action in the district court; but this is to cite only the ultimate consequence of the decision, and not the persuasive rationale upon which it was based.3

The same reasoning has also been applied in the single other case, where the exact issue before us appears to have been presented. In W. G. Duncan Coal Co. v. Glenn, D.C.W.D.Ky., 120 F. Supp. 948, before Circuit Judge Martin, the issue was, as here, a question of discrepancy amounting to $4,366.29 for 1943 and $8,507.31 for 1944 between the commissioner's separate computation of interest on excess profits tax deficiencies and income tax overpayments as against the taxpayer's treatment of both taxes as one and computation of interest only on the net deficiencies due the government. In upholding the commissioner's position and dismissing the taxpayer's suit for refund, Judge Martin said: 'An analysis of the plaintiff's computation appears to show that income taxes imposed in Chapter 1 and excess-profits taxes imposed by Chapter 2 have been treated as one type of tax. Since the income taxes and the excess-profits taxes are separate and distinct types of taxes there appear to be no provisions of the Internal Revenue Code to warrant the plaintiff's treatment of the two types as one tax for the purpose of computing interest. Interest was properly computed on the several amounts of excess-profits and declared value excess-profits tax deficiencies for each of the fiscal years 1943 and 1944, sought to be recovered. Interest was also computed and allowed on the overpayments of income taxes for the fiscal years 1943 and 1944 and were included in the overassessment determined for the plaintiff.'4

This decision, squarely contrary to the position taken below, seems to us persuasive in its reasoning. The concept of separate taxable units is of course a fairly usual one in the revenue law, as we have had occasion to hold only recently with reference to different taxable years. Rosenthal v. C.I.R., 2 Cir., 205 F.2d 505, 511; and see also Helvering v. Pfeiffer, 302 U.S. 247, 58 S. Ct. 159, 82 L. Ed. 231; Arrowsmith v. C.I.R., 344 U.S. 6, 73 S. Ct. 71, 97 L. Ed. 6. So on a taxpayer's appeal from the commissioner's determination of a deficiency in tax for 1936, the Board was held without jurisdiction to determine the amount of a 1935 overpayment. C.I.R. v. Gooch Milling & Elevator Co., supra, 320 U.S. 418, 64 S. Ct. 184, 88 L. Ed. 139. In Manning v. Seeley Tube & Box Co., 338 U.S. 561, 70 S. Ct. 386, 94 L. Ed. 346, a taxpayer was held not entitled to a refund of the interest assessed on a tax deficiency which was later abated under the carryback provisions of the Code by a net operating loss for a later year. Standard Roofing & Material Co. v. United States, 10 Cir., 199 F.2d 607, is to similar purport. In cases such as these the equities appear as strong as in the present case; but in them-- as here in our view-- the provisions of positive law prevented the abatement. This plaintiff, obviously thoroughly advised, took a calculated risk or tax gamble; having failed, it must pay the collector. Its claim for interest refund must therefore be denied.

On the plaintiffs' appeals in both actions the judgments must be affirmed; on the defendant's appeal in the first action the judgment must be reversed and the action remanded for the entry of judgment for the defendant.

FRANK, Circuit Judge (dissenting in part, i.e. as to the interest).

I agree with my colleagues to this extent: (a) The total tax due, on income and excess profits combined, was precisely, to the penny, what taxpayer reported and paid; (b) the taxpayer in its returns erroneously computed its excess profits, adding to it the precise amount by which it under-stated its income tax. But I disagree with my colleagues that, on account of this error-- utterly harmless to the government-- some $52,000 (of so-called interest) must be exacted from the taxpayer. My reasons follow:

1. To understand this case, it is essential never to forget that the taxpayer, in computing and reporting, as it did, the two items-- income and excess profits-- had this sole aim: to make its excess profits tax larger because the 10% postwar refund, under § 780, applies to the excess profits tax alone, not to the income tax.

To achieve this aim, taxpayer had available either one of two alternative paper plans. Under what I shall call paper Plan I, taxpayer would have computed and reported its income and its excess profits exactly as my colleagues and I hold it should have done (except that perhaps it would have stated in its returns that it believed its reported income was too large in a designated amount, and its reported excess profits too small, in the same amount). It would then have paid a total tax in the exact amount which it did pay. It would then have sued for a refund of that part of its income tax which, according to its interpretation of the Code, it would have overpaid. Had taxpayer won that refund suit, the government, of course, would have re-determined taxpayer's excess profits tax, increasing it by the very amount which taxpayer had been held entitled to recover in its successful refund suit. Thus, through the resultant increase in taxpayer's excess profits tax, the taxpayer would have achieved its sole aim, i.e., increasing its 10% postwar refund of excess profits tax under § 780.

Taxpayer, however, would have lost that refund suit (according to the ruling on which my colleagues and I agree). In that event, it would not have achieved its aim. But-- and this is crucial-- no one would then with a straight face have asserted that, because taxpayer had 'gambled,' and lost, it owed the government some $52,000 for attempting this unsuccessful 'gamble.'

Instead of using paper Plan I, taxpayer used what I call paper Plan Ii-- which consists of doing what it did: In its return, it augmented its excess profits and decreased its income in the same amount; then, when the Commissioner redetermined its income tax, it paid the consequent assessment and sued for a refund of this assessment. Had it won that refund suit, the effect would have been the same as the effect of winning the refund suit under Plan I. In other words, either paper Plan, for taxpayer's purpose, was as good as the other if, under either Plan, taxpayer won its refund suit.

But my colleagues hold that losing the refund suit under Plan II costs taxpayer $52,000-- although (as we have seen) losing such a suit under Plan I would have cost it zero. So, according to my colleagues, a $52,000 liability is imposed on taxpayer for but one reason, namely that, in trying to increase its 10% postwar refund, it resorted to one rather than another paper Plan. The $52,000 thus shows up as a penalty for a mere error in paper-work concerning a single taxable event, an error which in no wise deceived the government and deprived it of nothing whatever. I think that 'tax law' deserves the characterization Dickens' Mr. Bumble gave 'the law' in general-- i.e., 'the law is a ass'-- if the failure of paper Plan II costs taxpayer one nickel more than would the failure of Plan I.

2. I concede, of course, that, if the statutory language directed such a result, we could do nothing to avoid it. But we should not embrace that result unless explicit statutory provisions leave us no escape. I cannot agree that the Code contains such provisions.

My colleagues rest their conclusion on this thesis: For all purposes under the Code, the income tax and the excess profits tax, due on March 15 of any single year, represent two wholly independent and unrelated taxes, so that, if the income tax is underpaid, the government has been deprived of the use of the amount underpaid from March 15 until it is paid, and therefore has a right to interest on that amount for the intervening period. For only by treating the two taxes as wholly unrelated, can it be said that taxpayer owes such interest, since no one denies that, as the taxpayer paid the total amount of tax when due, the government was not deprived of the use of any money for a split second.

Patently, no wall separates these two taxes. On the contrary, these taxes intertwine. Section 710(a)(1)(B) makes this plain: As my colleagues themselves note, it 'placed an 80% ceiling on the combined income and excess profits taxes,' and thereby governs 'to limit the total tax.' Here we have a congressional mandate that the two must be taken together, must be synoptically regarded. Each becomes a function of the other.

3. My colleagues cite decisions where the problem arose whether the statute of limitations barred the setting off of a claim, otherwise valid, made by either taxpayer or the government, against a valid, unbarred claim asserted by the opponent. I think the course of these decisions goes to show that even over the barrier of a limitations statute, the Supreme Court's doctrine calls for avoidance of injustice when (as in the instant case) a single taxable event is in issue: In Bull v. United States, 1935, 295 u.s. 247, 55 S. Ct. 695, 79 L. Ed. 1421, in a suit by the government for income tax, the Court held that taxpayer, by way of recoupment, could successfully assert an overpayment of estate tax, arising from the same transaction, although the statute of limitations would have precluded an independent suit by taxpayer for the refund of the estate tax. In Stone v. White, 1937, 301 U.S. 532, 57 S. Ct. 851, 81 L. Ed. 1265, the question was whether, in a suit by trustees to recover a tax paid on the income of the trust estate, the government could assert as a counterclaim the tax which had not been paid by a beneficiary (who would receive the benefit of a judgment for the trustees) when the government's claim would have been barred by the statute of limitations, had the government brought an independent suit against the beneficiary; the Court sustained the counterclaim. In Commissioner of Internal Revenue v. Gooch Milling & Elevator Co., 1943, 320 U.S. 418, 64 S. Ct. 184, 88 L. Ed. 139, the Court held that, on a taxpayer's appeal to the Board of Tax Appeals from the Commissioner's determination of a deficiency in income tax for 1936, the Board lacked jurisdiction to determine the amount of a 1935 overpayment of income tax, the refund of which was barred by the statute of limitations, and to credit that 1935 overpayment against the 1936 deficiency. The Court, 320 U.S. at page 421, 64 S.Ct.at page 196, distinguished Bull v. United States, supra and Stone v. White, supra, because they related to 'the scope of equitable recoupment when it is asserted in a suit for refund of taxes in tribunals possessing general equity jurisdiction.'

In Rothensies v. Electric Storage Battery Co., 1946, 329 U.S. 296, 67 S. Ct. 271, 91 L. Ed. 296, the facts were these: The taxpayer, from 1919 to 1926, had mistakenly paid excise taxes. In 1926-- when claims for refund of those taxes for 1919 to 1921 inclusive were barred by the statute of limitations-- taxpayer filed suit for refund of the taxes paid between 1922 and 1926. It obtained a judgment, a part of which the government paid in 1935, through a settlement. The Commissioner treated this sum as part of taxpayer's income for 1935, and on this account assessed a deficiency. The taxpayer, having paid this deficiency, sued for a refund. It contended that it had a right to recoup the amount of the excise taxes it had paid in 1919 to 1921 inclusive. The Court, rejecting this contention, distinguished Bull v. United States, supra, and Stone v. White, supra, saying, 329 U.S. 299-300, 67 S. Ct. 272: The doctrine of recoupment in tax cases 'has never been thought to allow one transaction to be offset against another, but only to permit a transaction which is made subject of suit by a plaintiff to be examined in all its aspects, and judgment to be rendered that does justice in view of the one transaction as a whole. The application of this general principle to concrete cases in both of the cited decisions is instructive as to the limited scope given to recoupment in tax litigation. In both cases a single transaction constituted the taxable event claimed upon and the one considered in recoupment. In both, the single transaction or taxable event had been subjected to two taxes on inconsistent legal theories, and what was mistakenly paid was recouped against what was correctly due.'1

In the instant case, relating to a single year (and with no statute of limitations involved), the taxpayer's income and the excess profits constitute together a 'single * * * taxable event'. This 'transaction which is made the subject of suit' should therefore 'be examined in all its aspects, and judgment * * * rendered that does justice in view of the one transaction as a whole.' As the Supreme Court has said, a suit like this is 'equitable in its function.'2 It cannot be equitable to exact a $52,000 penalty based on what is only a bookkeeping rearrangement of figures.

4. It is true, as my colleagues say, that, generally speaking, each taxable year is a separate unit, a doctrine which sometimes yields injustices. But, even that doctrine, designed to avoid difficulties in the administration of the revenue laws, is, as shown above, not without its exceptions designed to prevent major injustices. As I understand my colleagues, they suggest that, because that doctrine often works unfairness to taxpayers, we should not bother about injustice here, where that doctrine has no application.

5. Manning v. Seeley Tube & Box Co., 338 U.S. 561, 70 S. Ct. 386, 94 L. Ed. 346, and Standard Roofing & Material Co. v. United States, 10 Cir., 199 F.2d 607, were cases in which the taxpayer attempted to change the basis of the tax. In the case at bar, nothing similar exists: no question here arises as to the amount of taxpayer's income, the only question being the proportion of taxpayer's income allocable to excess profits tax and income, respectively.3

6. In Superheater Co. v. Commissioner, 2 Cir., 125 F.2d 514, 516, cited by my colleagues, the decision turned on peculiar provisions of the Code under which the Board of Tax Appeals lacked jurisdiction to re-determine excess profits tax of a taxpayer on a review of the Commissioner's determination of a deficiency in income tax. In the instant case, we face no such problem; for this is a suit, brought under 28 U.S.C. § 1346(a)(1)(ii), in the district court which has no such restrictions on its jurisdiction. Our opinion in that case concluded with the statement that 'it seems clear that until the jurisdiction of the Board is further enlarged whatever procedural convenience might be attained by having a formal redetermination of the excess profits taxes in this particular instance must give way to the greater necessity for recognizing and giving effect to the limited statutory jurisdiction of the Board.' Note, too, how in Commissioner of Internal Revenue v. Gooch Milling & Elevator Co., 320 U.S. 418, at page 421, 64 S. Ct. 184, at page 186, 88 L. Ed. 139 (as pointed out above) the Court, holding that the Board lacked jurisdiction, distinguished Bull v. United States, supra, and Stone v. White, supra, on the ground that they related to suits not before the Board but to suits 'for refund of taxes in tribunals possessing general equity jurisdiction.'

In Rosenthal v. Commissioner, 2 Cir., 205 F.2d 505, 512, the decision (so far as here pertinent) turned on the failure of the Commissioner to file a cross-appeal. Moreover, the court referred to 'the settled rule that each taxable year is to be treated as a separate unit for, assessment and decisional purposes', and that rule is irrelevant here.

7. For the foregoing reasons, I think that the income tax and the excess profits tax are not, in a case like this, to be considered wholly distinct from one another. If I am correct in this respect, then, the Code points the way to justice here. Section 321 provides: 'If the taxpayer has paid as an installment of the tax more than the amount determined to be the correct amount of such installment, the overpayment shall be credited against the unpaid installments, if any. If the amount already paid, whether or not on the basis of installments, exceeds the amount determined to be the correct amount of the tax, the overpayments shall be credited or refunded as provided in section 322.' Section 322(a)(1) provides: 'Where there has been an overpayment of any tax imposed by this chapter, the amount of such overpayment shall be credited against any income, war-profits, or excess-profits tax or installment thereof then due from the taxpayer, and any balance shall be refunded immediately to the taxpayer.' And § 729(a) makes applicable to the excess profits tax all provisions applicable to the income tax which are not inconsistent with the specific provisions of the Chapter in respect of the excess profits tax.