Cal. No. 64, Docket 27611.
Argued November 16, 1962.
Decided December 12, 1962.
Richard L. Braunstein, Washington, D.C. (Bernard J. Long, Washington, D.C., on brief), of Dow, Lohnes Albertson, Washington, D.C., for petitioner.
Harold C. Wilkenfeld, Washington, D.C. (Louis F. Oberdorfer, Asst. Atty. Gen., Lee A. Jackson and Meyer Rothwacks, Attorneys, Department of Justice, Washington, D.C.), for respondent.
Before SWAN, WATERMAN and FRIENDLY, Circuit Judges.
This petition for review presents the rare case in which it is as clear as anything ever can be that Congress did not mean what in strict letter it said. The contretemps arose through a cross-reference in a section inserted in the Internal Revenue Code of 1954 during a late stage of its passage. Relying on the letter, petitioner asserts that its wholly owned subsidiary, which sold real estate in 1954, prior to enactment of the Code, for a gain of $1,808,632.05 that would have been taxable to the subsidiary as the law stood at the time, and which thereafter liquidated on the tax-free basis for the parent provided in prior Revenue Acts and continued by the Code, was also freed from tax on the gain from the sale. A unanimous Tax Court disagreed. So do we.
J.C. Penney Building and Realty Corporation, a wholly owned subsidiary of petitioner, owned an office building and warehouse in New York City and two warehouses elsewhere. On February 2, 1954, it entered into a contract to sell the New York warehouse at a substantial gain; it delivered the deed on August 3, 1954. On November 24, 1954, the directors of the parent and the subsidiary adopted a plan of complete liquidation whereby all the subsidiary's assets were to be distributed to the parent in exchange for the cancellation of its stock and in satisfaction of its indebtedness to the parent. On November 30, the assets were distributed; nine days later the subsidiary was dissolved. In its final tax return the subsidiary elected to have § 392(b) of the 1954 Code apply to the sale — with the result, as claimed, that the gain of $1,808,632.05 would not be taxable to it although, by virtue of § 332, the parent would pay no tax on the gain from the liquidation. The Commissioner, holding § 392(b) to be inapplicable in a situation where § 332 gave tax immunity to gain on the liquidation, determined a deficiency of $471,029.11 (along with an over-assessment of $171.17 for the preceding taxable year) in the subsidiary's tax return, and asserted this liability against petitioner as transferee. The Tax Court, in an able opinion by Judge Train reviewed by the full court, sustained the Commissioner without dissent, disagreeing with the contrary ruling in Diversified Services, Inc. v. United States, 192 F. Supp. 571 (S.D.Fla. 1961), now on appeal to the Fifth Circuit.
Petitioner's argument is this. Section 392(b)(1) of the 1954 Code provides:
"(1) Nonrecognition of gain or loss. — If —
"(A) all of the assets of a corporation (less assets retained to meet claims) are distributed before January 1, 1955, in complete liquidation of such corporation; and
"(B) the corporation elects (at such time and in such manner as the Secretary or his delegate may by regulations prescribe) to have this subsection apply,
then no gain or loss shall be recognized to such corporation from the sale or exchange by it of property during the calendar year 1954."
It is not disputed that what J.C. Penney Building and Realty Corporation did fits (A) and (B) of subdivision (1). The fly in petitioner's ointment is a provision in the next subdivision:
"(2) Certain provisions of section 337 made applicable. — For purposes of paragraph (1) —
* * * * * *
"(B) the limitations of section 337(c) shall apply."
Section 337, new in the 1954 Code, was designed to provide a method for escaping the two-fold taxation that had previously occurred when a corporation sold property at a gain to itself and then liquidated at a gain to its stockholders. Subdivision (a) provided:
"(a) General rule. — If —
"(1) a corporation adopts a plan of complete liquidation on or after June 22, 1954, and
"(2) within the 12-month period beginning on the date of the adoption of such plan, all of the assets of the corporation are distributed in complete liquidation, less assets retained to meet claims,
then no gain or loss shall be recognized to such corporation from the sale or exchange by it of property within such 12-month period."
Quite obviously, it was this same regime that § 392(b) adopted, subject to whatever limitations § 337 contained. Since the object of § 337 was to permit avoiding two taxes on what was essentially the same gain, it was altogether natural that Congress should not wish the section to apply when only one tax would be payable in any event — importantly, for present purposes, on the complete liquidation of an at least 80% owned subsidiary, as to which, under provisions of nearly twenty years standing, Revenue Act of 1934, § 112(b)(6), as added by Revenue Act of 1935, 49 Stat. 1020; Internal Revenue Code of 1939, § 112(b)(6); Internal Revenue Code of 1954, § 332, no gain was recognized on the parent's receipt of the proceeds. The problem here arises from the language that Congress inserted into § 337 in order to carry out this purpose; this was in relevant part:
"(c) Limitations. —
* * * * *
"(2) Liquidation to which section 332 applies. — In the case of a sale or exchange following the adoption of a plan of complete liquidation, if section 332 applies with respect to such liquidation, then —
"(A) if the basis of the property of the liquidating corporation in the hands of the distributee is determined under section 334(b)(1), this section shall not apply; * * *."
In the light of this, says petitioner, what seemed the fly in its ointment turns out to be no fly at all. For § 337(c)(2) is prefixed by the words "In the case of a sale or exchange following the adoption of a plan of complete liquidation," and here the sale preceded the adoption of the plan; hence there is no "limitation of section 337(c)" that applies.
Petitioner does not deny that the results of its position are somewhat startling. Corporation A and Corporation B have wholly-owned subsidiaries, A' and B'. A' sells a capital asset on February 1, 1954, at a gain of $1,000,000, pursuant to a previously adopted plan of liquidation; on March 1 it liquidates to A. B' also sells a capital asset on February 1, 1954, at a gain of $1,000,000, but without a previously adopted plan of liquidation; on March 1, having adopted such a plan, it liquidates to B. As the law stood at the time, both A' and B' would be liable for a tax of $250,000 on the sale; neither A nor B would be subject to tax on gain in the liquidation. Months later, on August 16, 1954, the new Revenue Code becomes law. On petitioner's construction B' finds that § 392(b) allows it to relieve itself of tax on the sale by an appropriate election, although § 332 leaves B free of tax on the gain in liquidation. But no such happy consequences are enjoyed by the A corporate family across the street, since the sale by A' happened to be one "following the adoption of a plan of complete liquidation", § 337(c)(2), and "the limitations of section 337(c)" therefore apply. Although the results for the portion of 1954 after enactment of the Code are less capricious as between taxpayers, since the element of surprise is removed, they are equally unfortunate to the revenue. Although the parent remains free of tax on the gain from the liquidation, the subsidiary can avoid taxation on the gain from the sale by the simple expedient of selling before it adopts a plan of liquidation. Neither is there any true balancing effect through the applicability of § 392(b) to losses as well as gains. If the sale will involve a loss useful for tax purposes, the subsidiary can sell before adopting a plan and then not elect to have § 392(b) apply. Only with the end of 1954 is the goal of nonrecognition of gain or loss on the sale in some cases and on the liquidation in other cases, but not on both in the same case, finally attained.
With commendable candor taxpayer's counsel acknowledged, in answering a question at argument, that he could conceive of no reason why Congress should have wished to confer benefits with so lavish and discriminatory a hand. The legislative history suggests none. Congress is free, within constitutional limitations, to legislate eccentrically if it should wish, but courts should not lightly assume that it has done so. When the "plain meaning" of statutory language "has led to absurd or futile results", the Supreme Court "has looked beyond the words to the purpose of the act"; "even when the plain meaning did not produce absurd results but merely an unreasonable one `plainly at variance with the policy of the legislation as a whole'" the Court "has followed that purpose rather than the literal words." United States v. American Trucking Ass'n, 310 U.S. 534, 543, 60 S.Ct. 1059, 1063, 84 L.Ed. 1345 (1940). "Anything that is written may present a problem of meaning, and that is the essence of the business of judges in construing legislation", Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum.L.Rev. 527, 528 (1947). And, as we were told long ago, "Where the mind labors to discover the design of the legislature, it seizes everything from which aid can be derived;" Marshall, C.J., in United States v. Fisher, 2 Cranch (6 U.S.) 358, 386 (1805); see United States v. Dickerson, 310 U.S. 554, 561-562, 60 S.Ct. 1034, 84 L.Ed. 1356 (1940).
The Internal Revenue Code of 1954 was the culmination of a long history; the Chairman of the Ways and Means Committee said in introducing it, "This bill represents a complete overhaul of all our revenue laws, the first since the enactment of the income tax." 100 Cong. Rec. 3420 (March 17, 1954). Even when the words of the Code may seem clear, courts applying it must look to what had gone before, to what was left intact, to what was altered and why. See Harrison v. Northern Trust Co., 317 U.S. 476, 479, 63 S.Ct. 361, 87 L.Ed. 407 (1943). Always "There is need to keep in view * * the structure of the statute, and the relation, physical and logical, between its several parts." Duparquet Co. v. Evans, 297 U.S. 216, 218, 56 S.Ct. 412, 80 L.Ed. 591 (1936). Moreover, the Code did not spring full-blown from the Congressional brow. The final form of the provisions as to corporate liquidations bore little resemblance to the Ways and Means Committee's proposals; the bill's vicissitudes on its voyage through Congress are highly relevant to the problem of this case. To all this we now turn.
Prior to 1935, gain or loss was recognized on the liquidation of a subsidiary just as on other liquidations. See, e.g., Neptune Meter Co. v. Price, 98 F.2d 76 (2 Cir. 1938), judgment on remand aff'd, 110 F.2d 852 (2 Cir. 1940). Special treatment for liquidations of subsidiaries began with § 112(b)(6) of the Internal Revenue Act of 1934, added by the Act of 1935, 49 Stat. 1020, which was continued as § 112(b)(6) of the Internal Revenue Code of 1939. This provided in substance that no gain or loss should be recognized upon the receipt of property distributed in complete liquidation by a corporation owning 80% or more of the stock (other than preferred stock) of the corporation making the distribution, providing certain other specifications were met. This meant in practical effect that only one tax would be payable on gain from the sale of property incident to liquidation of such a subsidiary, and this irrespective of whether the sale was made by the subsidiary before liquidation or by the parent thereafter. If the subsidiary sold before liquidating, it was subject to the same tax as any other seller; if the parent sold after the liquidation, the same tax would normally be payable, since § 113(a) (15) provided that in such cases the basis of the property to the distributee should, generally speaking, remain what it had been to the transferor; and there would be no tax on gain from the liquidation itself. Parent and 80% owned subsidiary were thus freed from one of the two taxes imposed in other cases when sale at a gain preceded liquidation at a gain, and hence from any need to seek means which often were cumbersome and sometimes impractical in an effort to avoid one of them. See C.I.R. v. Court Holding Co., 324 U.S. 331, 65 S.Ct. 707, 89 L.Ed. 981 (1945); United States v. Cumberland Public Service Co., 338 U.S. 451, 70 S.Ct. 280, 94 L. Ed. 251 (1950).
H.R. 8300, as reported by the Ways and Means Committee on March 9, 1954 (83d Cong.2d Sess., H.R.Rep. 1337), contained comprehensive new provisions dealing with corporate liquidations, pp. 4238-4253, The basic approach was "to limit taxation at the time of complete or partial liquidation of the corporation to amounts distributed which are reflected in the basis of the assets in the hands of the liquidating corporation," p. 4238. Because of this, no special provision with respect to the liquidation of a wholly owned subsidiary was needed. To deal with the Court Holding — Cumberland problem the Committee sought to provide "a definitive rule which will eliminate any uncertainty," p. 4244, by stating that no gain should be recognized to a corporation upon the sale of an asset if the sale was made after adoption of a plan of liquidation and the liquidation was completed within the taxable year in which the sale occurred or the succeeding taxable year. The new provisions were to apply to taxable years beginning after 1953, p. 4589. The minority of the Committee objected to these and other changes with respect to corporate distributions and adjustments, p. 4613, believing that they had been prepared without adequate consideration and that "the suggested new provisions, on balance, in contrast to existing law, court decisions, and administrative rules lean heavily in the direction of tax postponement," p. 4616.
Page references to committee and conference reports are to 3 U.S.Code Cong. Adm.News, 83d Cong.2d Sess.
The Senate Finance Committee made its report on June 18, 1954 (S.Rep. No. 1622). Rejecting the radical changes in the liquidation provisions recommended by the House Committee, the Senate Committee "returned to the basic pattern of the present Code with, however, certain changes hereinafter noted," pp. 4674, 4678. The rejection and return, and the attendant rewriting of "Subchapter C — Corporate Distributions and Adjustments," involved restoration of the special provision as to liquidation of 80% owned subsidiaries, p. 4679; this became § 332. On the other hand, the Finance Committee "follow[ed] the House bill in eliminating questions arising as a result of the necessity of determining whether a corporation in process of complete liquidation made a sale of assets or whether the shareholder receiving the assets made the sale," with two taxes in the former case, by providing "that if a corporation adopts a plan of liquidation and liquidates within 12 months thereafter, sales of assets during this 12-month period will not result in tax at the corporate level"; pp. 4679-4680; this became § 337. The Committee's detailed discussion of § 337 described the limitations of that section as follows: "Subsection (c) limits the application of section 337. * * * Paragraph (2) provides (A) that the section shall not apply to a liquidation under section 332 if the basis of the property in the hands of the distributee is determined under section 334(b)(1) (corresponding to section 113(a) (15) of the 1939 Code) * * *." p. 4897. In "Part VI — Effective Date of Subchapter C", the bill contained a provision, § 392, entitled "Effective Date of Part II", which said that the new provisions with respect to liquidation, among them §§ 332 and 337, would apply only if the first distribution occurred on or after June 18, 1954. For the case of a wholly owned subsidiary here presented, the effective date would have been inconsequential since the law under the new bill was to be the same as the old — with no tax on the liquidation under § 332, but with a tax on the sale whenever made since applicability of § 332 made the limitation of § 337 effective.
Meanwhile time was running. On July 1, 1954, Senator Capehart offered as a floor amendment what is now § 392(b). He explained its purpose as being "to make the liquidation rules of H.R. 8300 applicable to distributions in liquidation which occur in 1954. In particular, it is intended to have section 337, relating to sales made by corporations in process of liquidation, apply to such distributions." Section 337, he said, "provides that there shall be only one tax in any event," 100 Cong.Rec. 9451. The amendment was voted by the Senate forthwith, 100 Cong. Rec. 9452, was accepted by the Conference Committee, H.R.Rep. No. 2543, July 26, 1954, and was in the bill that became law on August 16, 1954. The Conference Report noted that the Senate had adopted a completely new subchapter relating to corporate distributions and adjustments, eliminating "certain new concepts contained in the House bill" but "at the same time preserving, to the greatest extent possible, the degree of certainty which was sought in the House bill and which is lacking in existing law", pp. 5293-5294. With respect to § 392(b), the report made the following remarks here relevant, p. 5296:
"(d) Sales during period of complete liquidation. — Section 337 of the Senate amendment provided for nonrecognition of gain or loss to a corporation from the sale or exchange of certain property by it within the 12-month period beginning on the date of the adoption of a plan of liquidation. Section 392(b) of the Senate amendment provided a substantially similar rule applicable (whether or not a plan of liquidation is adopted) with respect to sales or exchanges during the calendar year 1954, but only where the distributions in liquidation are completed before January 1, 1955.
* * * * * *
"Section 392(b)(2)(B) provides that the special rule for sales or exchanges by a corporation being liquidated which is contained in section 392(b) will not be available if the limitations of section 337(c) apply (that is, among other limitations, if the liquidation is one to which sec. 112(b)(6) of the 1939 Code applies, or if the liquidation is one to which sec. 332 of the 1954 Code applies)."
It is to be noted that the Conference Report, made after the inclusion of § 392(b), like the report of the Senate Finance Committee, made before such inclusion, thus omits from the recitation of the "limitations of section 337(c)" any reference to the "following the adoption of a plan" language on which petitioner relies.
In adding § 392(b) to a section relating to effective date, Congress had no purpose to prescribe generally a different and more favorable regime for corporations selling property and liquidating in 1954 than in later years. Its purpose, as Senator Capehart made clear, was rather to make available to such corporations, including those that had taken action before enactment of the Code or without knowledge of it, the same regime of one tax on gain, but only one, which previous law, continued by § 332, had long provided for 80% owned subsidiaries by eliminating tax on the gain from the liquidation, and which the new § 337 was providing for corporations adopting a plan of complete liquidation on or after June 22, 1954, and distributing within 12 months thereafter by eliminating tax on gain from the sale. This purpose could have been accomplished by changing the date in § 337(a)(1) to January 1, 1954, save that such an amendment would have given the benefit of § 337 treatment only to corporations otherwise qualifying that had adopted a plan first and sold property thereafter, and would have denied that benefit to similar corporations which had sold property before adopting a plan, although there was no reason at the time to suppose that a different timetable would bring tax advantages; § 392(b) was designed to prevent such inequality during 1954, see Kent Mfg. Corp. v. C.I.R., 288 F.2d 812, 816 (4 Cir. 1961). Thereafter, it was evidently supposed, tax counsel would have become familiar with § 337 and would follow its procedure when nonrecognition was desired. What apparently was not perceived was that the introductory words of § 337(c)(2) — which had in no way prevented the denial of § 337 treatment to an 80% owned subsidary subject to § 332 — would, if read literally, prevent the denial of the identical § 392(b) treatment to such a subsidiary which, by good luck before enactment of the Code or deliberate decision during the remainder of 1954, sold first and planned later.
This is preeminently a case for applying the principles enumerated in United States v. American Trucking Ass'n, supra. The introductory words of § 337(c)(2) were quite unnecessary. The "General Rule" stated in § 337(a) was that when a corporation "adopts a plan of complete liquidation on or after June 22, 1954" and distributes its assets "within the 12-month period beginning on the date of the adoption of such plan," then "no gain or loss shall be recognized to such corporation from the sale or exchange by it of property within such 12-month period." Hence § 337 could never apply to a sale or exchange not "following the adoption of a plan of complete liquidation." The beginning words in § 337(c)(2) thus had no function save a stylistic one; the meaning would have been precisely the same if § 337(c)(2) had started "If section 332 applies with respect to such liquidation, then". Yet it is these purely formal words of introduction, without substantive purpose when drafted by the Senate Committee, to which petitioner would give the effect, by virtue of Senator Capehart's amendment, of placing the "sell-first plan-later" 1954 transaction not simply on a par with the "plan-first sell-later" one, as was intended, but, when the seller was an 80% owned subsidiary, in a better position — thereby enlarging the Senator's "only one tax in any event" amendment into a "no tax in some event" one. Both Senator Capehart's statement on the floor and the Conference Report are convincing that nothing of the kind was meant.
Petitioner denies that the Conference Report, by referring explicitly to the applicability of § 332 as among "the limitations of Section 337(c)" but omitting any reference to the "following the adoption of a plan" language, indicates that Congress did not regard this language as having any operative effect on § 337(c)'s "limitations." It points out that the Report also omits any reference to the requirement in § 337(c)(2)(A) that the basis of the property of the liquidating corporation in the hands of the distributee be determined under § 334(b)(1) — which requirement, petitioner asserts, is an operative part of the limitation, so that the Report's failure to mention it demonstrates that the Committee was not attempting to put forth a complete statement concerning the "limitations" here involved. But in the typical situation that would arise under § 392(b), such as the one before us, applicability of § 334(b)(1) would follow automatically from applicability of § 332; indeed, the former section explicitly incorporates the latter by reference. Thus by mentioning one the Committee effectively mentioned the other as well. Indeed, § 337(c)(2) could have dispensed with any reference to the basis provisions had it not been for § 334(b) (2), which provides a different basis rule for the case where the stock of the liquidating corporation has been acquired by the distributee less than two years before the adoption of the plan of liquidation. See Kimbell-Diamond Milling Co., 14 T.C. 74 (1950), aff'd, 187 F.2d 718 (5 Cir. 1951); cert. denied, 342 U.S. 827, 72 S.Ct. 50, 96 L.Ed. 626 (1951); Cohen, Geldberg, Surrey, Tarleau and Warren, Corporate Liquidations under the Internal Revenue Code of 1954, 55 Colum.L.Rev. 37, 42-44 (1955). In such cases, § 337 is partially applicable, as § 337(c)(2)(B) provides. The failure of the Conference Committee to cover this case does not detract from the significance of its failure to mention the language of § 337(c)(2) on which petitioner relies.
Petitioner urges upon us various statements of the Supreme Court with respect to following a plain meaning, particularly the recent one in Hanover Bank v. C.I.R., 369 U.S. 672, 687-688, 82 S.Ct. 1080, 8 L.Ed.2d 187 (1962), that courts "are not at liberty, notwithstanding the apparent tax-saving windfall bestowed upon taxpayers, to add to or alter the words employed to effect a purpose which does not appear on the face of the statute." But that case presented no such situation as here, where words that had been nugatory as drafted and as reported by the Senate Finance Committee are now urged to have become invested, by a later amendment, with meaning directly counter to the declared purpose of Congress and the statutory scheme. Whereas the Supreme Court in Hanover found that "the legislative history, too, is persuasive evidence that the statute * * * allowed the deduction refused these taxpayers," here the history negates an intention to confer any such erratic bonanzas as petitioner claims. The plain meaning doctrine "is rather an axiom of experience than a rule of law, and does not preclude consideration of persuasive evidence if it exists," Mr. Justice Holmes in Boston Sand Gravel Co. v. United States, 278 U.S. 41, 48, 49 S.Ct. 52, 53, 73 L.Ed. 170 (1928); here the evidence is not merely persuasive but overwhelming. Although there may be cases where courts must follow the letter even under such circumstances in order to protect citizens who have acted in justifiable reliance upon it, this is not one of them. Petitioner cannot make any such claim, since the contract of sale was made five months before the Senate adopted § 392(b) and the sale was concluded before the Code became law. But even when everything occurred subsequent to enactment of the Code, as apparently was true in Diversified Services, Inc. v. United States, 192 F. Supp. 571 (S.D.Fla. 1961), we should not deem reliance on the letter to be justifiable. The corporate liquidation provisions of the Internal Revenue Code, with their involuted cross-references, are not for reading by him who runs; to the layman they have no meaning, either plain or fancy. They are for reading by lawyers, and primarily tax lawyers at that. Prudent tax counsel could hardly have assured a client that judges would give the words here relied on a construction so contrary to the legislative scheme and purpose, however hard he might later endeavor to persuade a court to do so. We regard the introductory words of § 337(c)(2) as the same useless surplusage with respect to § 392(b) as with respect to § 337 itself; since they would not have made § 337 confer tax immunity on gain from a sale if a subsidiary subject to § 332 had first adopted a plan and then sold, they equally do not produce that result under § 392(b) when such a subsidiary first sells and then adopts a plan. Cf. Arkansas Oak Flooring Co. v. Louisiana Arkansas Ry. Co., 166 F.2d 98, 101 (5 Cir.), cert. denied, 334 U.S. 828, 68 S.Ct. 1338, 92 L.Ed. 1756 (1948); 2 Sutherland, Statutory Construction (3d ed. 1943), § 4926.
What § 392(b) did was make it possible for corporations acting in 1954, either before or after June 22, and otherwise within § 337, to claim or not to claim the non-recognition principle of that section, and this whether the corporation planned first and sold after as § 337 required, or sold first and planned after. In granting this dispensation, in a section of the Code dealing with effective date, Congress had no purpose to depart from its basic concept that the new principle of nonrecognition of gain or loss on sale by a corporation in the course of liquidation, which § 337 and § 392(b) both embodied, should be inapplicable when there was nonrecognition of gain or loss on the liquidation itself. This is not only what Congress meant by the combination of § 392(b)(2)(B) and § 337(c)(2), but what we believe lawyers familiar with the structure and purpose of the Code would think it meant.