National Carbide Corp. v. Commissioner , 69 S. Ct. 726 ( 1949 )

336 U.S. 422 (1949)

NATIONAL CARBIDE CORP.
v.
COMMISSIONER OF INTERNAL REVENUE.

No. 151.

Supreme Court of United States.

Argued January 6, 1949. Decided March 28, 1949. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT.^[*]

Erwin N. Griswold argued the cause for petitioners. With him on the brief were Boykin C. Wright, Edgar J. Goodrich and John A. Wilson.

Hilbert P. Zarky argued the cause for respondent. With him on the brief were Solicitor General Perlman, Assistant Attorney General Caudle, Stanley M. Silverberg, Ellis N. Slack and Lee A. Jackson.

MR. CHIEF JUSTICE VINSON delivered the opinion of the Court.

Petitioners are three wholly owned subsidiaries of Air Reduction Corporation (Airco). They seek a determination of the question whether deficiencies in income and declared value excess profits taxes for the year 1938 found by the Commissioner of Internal Revenue are properly chargeable to them. Their contention is that they are corporate agents of Airco, that the income from their operations is income of Airco, and that income and excess profits taxes must be determined on that basis.

By a series of combinations and dissolutions of previously acquired subsidiary companies, Airco had, prior to 1938, reduced the number of its subsidiaries to four. All operated strictly in accordance with contracts with Airco.^[1] The subsidiaries were utilized by Airco as operating companies in the four major fields of operation in which it was engaged. Air Reduction Sales Company carried on the manufacture and sale of the gaseous constituents of air; National Carbide Corporation, the manufacture and sale of calcium carbide; Pure Carbonic, Inc., the manufacture and sale of carbon dioxide; and Wilson Welder & Metals Co., the manufacture and sale of welding machines, equipment and supplies.^[2]

The contracts between Airco and its subsidiaries provided, in substance, that the latter were employed as agents to manage and operate plants designed for the production of the products assigned to each, and as agents to sell the output of the plants. Airco was to furnish working capital, executive management and office facilities for its subsidiaries. They in turn agreed to pay Airco all profits in excess of six percent on their outstanding capital stock, which in each case was nominal in amount.^[3] Title to the assets utilized by the subsidiaries was held by them, and amounts advanced by Airco for the purchase of assets and working capital were shown on the books of the subsidiaries as accounts payable to Airco. The value of the assets of each company thus approximated the amount owed to Airco. No interest ran on these accounts.

Airco and its subsidiaries were organized horizontally into six overriding divisions: corporate, operations, sales, financial, distribution, and research. Officers heading each division were, in turn, officers of the subsidiaries. Top officials of Airco held similar positions in the subsidiary companies. Directors of the subsidiaries met only to ratify the actions of the directors and officers of Airco.

Airco considered the profits turned over to it by the subsidiaries pursuant to the contracts as its own income and reported it as such. Petitioners reported as income only the six per cent return on capital that each was entitled to retain. Similarly, in declaring the value of their capital stock for declared value excess profits tax purposes, the subsidiaries reported only the nominal amounts at which the stock was carried on the books of each. The Commissioner notified petitioners of substantial income and excess profits tax deficiencies in their 1938 returns, having taken the position that they are taxable on the income turned over to Airco as well as the nominal amounts retained. The Tax Court held, however, that the income from petitioners' operations in excess of six per cent of their capital stock was income and property of Airco. Three judges dissented. The Court of Appeals for the Second Circuit reversed. 167 F.2d 304. We granted the petition for a writ of certiorari, 335 U.S. 810, because of this conflict of opinion and the disagreement between courts as to the continuing vitality of Southern Pacific Co. v. Lowe, 247 U.S. 330 (1918).

Petitioners' contention is, in substance, that our decision in Moline Properties, Inc. v. Commissioner, 319 U.S. 436 (1943), which held that the tax laws require taxation of the corporate entity if it engages in "business activity," expressly excepted the situation in which the corporation is the agent of its owner; that Southern Pacific Co. v. Lowe, supra, defines the content of "agency" for tax purposes; and that, as the Tax Court found, this Court's characterization of the relationship between the corporations in the Southern Pacific case is "aptly descriptive" of the relationship between Airco and petitioners. It must follow, according to petitioners, that income received by them and transmitted to Airco is taxable only to Airco.

Respondent does not quarrel with the first and third propositions. The collision occurs at the second. The issue as presented by petitioners is, therefore, whether the principal-agent relationship described in the Southern Pacific case — and the similar arrangement between Airco and petitioners — contains the "usual incidents of an agency relationship," as that phrase was used in Moline Properties, Inc. v. Commissioner, supra.

Petitioners' contention that the Southern Pacific case established a concept of agency that has survived our later decisions may be dealt with rather summarily. That case treated income earned by a wholly owned subsidiary before March 1, 1913, the effective date of the Income Tax Act of 1913, as having accrued to its parent prior to that date despite the fact that the actual transfer of funds by declaration of dividends occurred subsequent thereto. The theory of the case was that the two corporations could be treated as identical, for the purposes of the 1913 Act, because of the complete domination and control exercised by the parent over its subsidiary.

By this decision, this Court is said to have "looked beyond the corporate form,"^[4] and ignored "the separate entity of a corporation."^[5] Whatever the dialectics employed, courts and commentators have agreed that parent and subsidiary were treated as one corporation for the purposes of the taxes there in question; transfer of earnings to the parent was merely "a paper transaction." The Southern Pacific case did not, and did not purport to, rest on any principle of agency. The only reference to the subsidiary (Central Pacific) as an agent is made in this context:

". . . the Central Pacific and the Southern Pacific were in substance identical because of the complete ownership and control which the latter possessed over the former, as stockholder and in other capacities. While the two companies were separate legal entities, yet in fact, and for all practical purposes they were merged, the former being but a part of the latter, acting merely as its agent and subject in all things to its proper direction and control." 247 U.S. at 337.

It is thus clear beyond doubt that the subsidiary was not referred to as an agent of the parent in the usual or technical sense. "Agency" and "practical identity," as those words are used in the Southern Pacific case, are unquestionably opposite sides of the same coin.^[6] The close relationship between corporations because of complete ownership and control of one by the other was the basis for the result reached, whatever its articulation.

That basis has been repudiated by subsequent decisions of this Court. Whatever the vitality of Southern Pacific Co. v. Lowe on its special facts, we have held that a corporation formed or operated for business purposes must share the tax burden despite substantial identity, in practical operation, with its owner. Complete ownership of the corporation, and the control primarily dependent upon such ownership — the important ingredients of the Southern Pacific case — are no longer of significance in determining taxability. Moline Properties, Inc. v. Commissioner, supra; Burnet v. Commonwealth Improvement Co., 287 U.S. 415 (1932).^[7]

In both of the cases last cited, the agency argument now urged upon us was made and rejected. In both cases, Southern Pacific Co. v. Lowe, supra, was relied upon by the taxpayers. In both, we found that the contention that the corporation was the agent of its owner was simply the argument that the subsidiary had no corporate identity distinct from its stockholders in a different form. It is true that petitioners here do not ask that they be ignored completely for tax purposes. They are willing to pay taxes on the nominal amounts they retain as Airco's "agents." But this fact serves to emphasize the inapplicability of the Southern Pacific case, upon which they rely. There, as in Commonwealth Improvement Co. and Moline Properties cases, the decision turned upon the question whether the corporate entity was or was not to be completely ignored for tax purposes. If the Central Pacific had been accorded any tax status in the Southern Pacific case, it unquestionably would have been taxed on the entire income it received. In fact, it was so taxed upon all income received after March 1, 1913; only income received prior thereto was considered income of the parent directly.^[8]

We think, therefore, that petitioners' argument is without merit because based on an erroneous interpretation of Southern Pacific Co. v. Lowe, supra. The agency argument, to quote the opinion in Moline Properties, "is basically the same argument of identity in a different form. . . . the question of agency or not depends upon the same legal issues as does the question of identity previously discussed."^[9] Ownership of a corporation and the control incident thereto can have no different tax consequences when clothed in the garb of agency than when worn as a removable corporate veil.

But it is necessary to go farther. The Tax Court did not, as petitioners seem to think, consider the argument that they were agents of Airco as different from or having any greater validity than the argument of identity of Airco and its subsidiaries. The court, in characterizing petitioners as Airco's agents, used that term exactly as it had been used in the Southern Pacific, Commonwealth Improvement Co., and Moline Properties cases. According to the Tax Court's opinion:

"The issue which [was decided] in this proceeding is whether, as the respondent has determined, the income from the operations of the three petitioners belonged not to Airco, the parent, but to the petitioners, and was taxable to them; or whether, as the three petitioners contend, the income from the operations of the petitioners in 1938, exclusive of the small amounts paid to petitioners under the contracts, belonged and was taxable to Airco, the parent company, both because the petitioners were in fact incorporated departments, divisions, or branches of Airco's business and because the petitioners operated pursuant to express contract with Airco."^[10]

The theory upon which the Tax Court expunged the deficiencies apparently was that since the Southern Pacific Co. case was not expressly overruled by Moline Properties, the "business purpose" rule laid down in the latter is not absolute, but that the corporate entity may be disregarded (or the corporation treated as an agent of its owner) for tax purposes when the facts of ownership and control of the corporation approximate those presented by the Southern Pacific case. The Court of Appeals disagreed. It held that under our decisions, when a corporation carries on business activity the fact that the owner retains direction of its affairs down to the minutest detail, provides all of its assets and takes all of its profits can make no difference tax-wise. The court concluded that "Even though Southern Pacific Co. v. Lowe, supra, set up a different test, we regard it as pro tanto no longer controlling."^[11]

The result reached by the Court of Appeals is clearly required by our later decisions. Our reluctance to erase Southern Pacific from the books has been due not to any belief that it lays down a correct rule for tax purposes generally, but to the fact that it concerns "very peculiar facts" which make it clearly distinguishable from later cases involving the tax status of a subsidiary or other wholly owned corporation.^[12] For that reason, we have, instead, held that it lays down no rule for tax purposes. Burnet v. Commonwealth Improvement Co., supra at p. 419; Moline Properties, Inc. v. Commissioner, supra at p. 439. That the concept of identity of the corporation with its owner set out in the Southern Pacific case is incompatible with later decisions of this Court may be demonstrated by a consideration of the facts enumerated and relied upon by the Tax Court, which based such reliance on the emphasis placed upon similar facts in the Southern Pacific case. These facts relate to the ownership, control, and right to income reserved by the parent.

So far as control is concerned, we can see no difference in principle between Airco's control of petitioners and that exercised over Moline Properties, Inc., by its sole stockholder. Undoubtedly the great majority of corporations owned by sole stockholders are "dummies" in the sense that their policies and day-to-day activities are determined not as decisions of the corporation but by their owners acting individually. We can see no significance, therefore, in findings of fact such as, "The Airco board held regular meetings and exercised complete domination and control over the business of Airco and each of the petitioners," and "The chairman, vice chairman, and president of Airco were in charge of the administration and management of the activities of each petitioner and carried out the policies and directives with respect to each petitioner as promulgated by the Airco board."^[13] We reversed the Board of Tax Appeals in Moline Properties in the face of its finding that "Full beneficial ownership was in Thompson [the sole stockholder], who continued to manage and regard the property as his own individually."^[14]

Some stress was placed by the Tax Court, and by petitioners in argument here, upon the form of ownership of assets adopted by Airco and its subsidiaries. Petitioners' capital stock was, as has been stated, nominal in amount. Assets of considerable value, to which title was held by the subsidiaries, were balanced by accounts payable to Airco on the books of each. The Tax Court thought it material that "All assets held by each petitioner were furnished to it by Airco, which paid for them with its own cash or stock. Airco supplied all the working capital of each petitioner."

If Airco had supplied assets to its subsidiaries in return for stock valued at amounts equal to the value of the assets, no question could be raised as to the reality of ownership of the assets by the subsidiaries. Airco would then have been in a position comparable, so far as ownership of the assets of petitioners is concerned, to that of the sole stockholder in Moline Properties. We think that it can make no difference that financing of the subsidiaries was carried out by means of book indebtednesses in lieu of increased book value of the subsidiaries' stock. A corporation must derive its funds from three sources: capital contributions, loans, and profits from operations. The fact that Airco, the sole stockholder, preferred to supply funds to its subsidiaries primarily by the second method, rather than either of the other two,^[15] does not make the income earned by their utilization income to Airco. We need not decide whether the funds supplied to petitioners by Airco were capital contributions rather than loans. It is sufficient to say that the very factors which, as petitioners contend, show that Airco "supplied" and "furnished" their assets also indicate that petitioners were the recipients of capital contributions rather than loans.^[16]

Nor do the contracts between Airco and petitioners by which the latter agreed to pay all profits above a nominal return to the former, on that account, become "agency" contracts within the meaning of our decisions. The Tax Court felt that the fact that Airco was entitled to the profits by contract shows that the income "belonged to Airco" and should not, for that reason, be taxed to petitioners. Our decisions requiring that income be taxed to those who earn it, despite anticipatory agreements designed to prevent vesting of the income in the earners, foreclose this result. Lucas v. Earl, 281 U.S. 111 (1930); Helvering v. Clifford, 309 U.S. 331 (1940); United States v. Joliet & Chicago R. Co., 315 U.S. 44 (1942); Commissioner v. Sunnen, 333 U.S. 591 (1948). Of course one of the duties of a collection agent is to transmit the money he receives to his principal according to their agreement.^[17] But the fact that petitioners were required by contract to turn over the money received by them to Airco, after deducting expenses and nominal profits, is no sure indication that they were mere collection agents. Such an agreement is entirely consistent with the corporation-sole stockholder relationship whether or not any agency exists, and with other relationships as well.^[18]

What we have said does not foreclose a true corporate agent or trustee from handling the property and income of its owner-principal without being taxable therefor. Whether the corporation operates in the name and for the account of the principal, binds the principal by its actions, transmits money received to the principal, and whether receipt of income is attributable to the services of employees of the principal and to assets belonging to the principal^[19] are some of the relevant considerations in determining whether a true agency exists. If the corporation is a true agent, its relations with its principal must not be dependent upon the fact that it is owned by the principal, if such is the case. Its business purpose must be the carrying on of the normal duties of an agent.^[20] Absence of the factors mentioned above, and the essentiality of ownership of the corporation to the existence of any "agency" relationship in the Moline Properties, Commonwealth Improvement Co., and Southern Pacific cases, indicate the fallacy of the agency argument made in those cases.

The same fallacy is apparent in the contention that petitioners are agents of Airco. They claim that they should be taxable on net income aggregating only $1,350, despite the fact that during the tax year (1938) they owned assets worth nearly 20 million dollars, had net sales of approximately 22 million dollars, and earned nearly four and one-half million dollars net. Their employees number in the thousands. We have passed the question whether Airco's interest in these assets is that of owner of the subsidiaries or lender, but whatever the answer, they do not belong to Airco as principal. The entire earnings of petitioners, except for trifling amounts, are turned over to Airco not because the latter could command this income if petitioners were owned by third persons, but because it owns and thus completely dominates the subsidiaries. Airco, for sufficient reasons of its own, wished to avoid the burdens of principalship.^[21] See Moline Properties, Inc. v. Commissioner, supra; Sheldon Building Corp. v. Commissioner, 118 F.2d 835 (1941). Compare Forshay v. Commissioner, 20 B.T.A. 537 (1930). It cannot now escape the tax consequences of that choice, no matter how bona fide its motives or longstanding its arrangements. When we referred to the "usual incidents of an agency relationship" in the Moline Properties case, we meant just that — not the identity of ownership and control disclosed by the facts of this case.

We have considered the other arguments made by petitioners and find them to be without merit. The judgment of the Court of Appeals is

Affirmed.

NOTES

[*] Together with No. 152, Air Reduction Sales Co. v. Commissioner of Internal Revenue, and No. 153, Pure Carbonic, Inc. v. Commissioner of Internal Revenue, also on certiorari to the same court.

[1] The substance of a typical subsidiary-parent contract is as follows: "Airco hereby employs Sales as its agent to manage and operate, during the term of this contract, all plants for the production of oxygen, acetylene and other gases and for the manufacture of apparatus and containers for the utilization and transportation of such gases . . .; and likewise employs Sales as its agent to market and sell, during the term of this contract, the output of all such plants. . . . Airco agrees (1) to give Sales the use of all cylinders, containers, motor trucks, equipment, and shipping facilities, which it now owns or may hereafter acquire; (2) to supply such working capital as Sales may need; (3) to provide such executive management (but not accounting, bookkeeping and clerical service), and office accommodation and facilities, as may be necessary for the proper conduct of Sales business. . . . Sales agrees (1) to manage and operate . . . all of said plants; (2) to maintain the same in first class condition, charging necessary repairs and replacements to operating expense and setting aside and charging to operating expense proper reserves for depreciation . . . (3) to distribute, market and sell, the product manufactured in said plants as efficiently as possible . . . (4) to pay all expenses of such operation, maintenance and selling, and to discharge all expenses or liabilities incurred therein or thereby and to collect all accounts receivable or other proceeds resulting therefrom; (5) to credit monthly on its books to Airco all profits accruing to it from the operation of its entire business over and above an amount equal to six per cent (6%) per annum on its outstanding capital stock, which said amount it is hereby authorized to deduct and retain, and it hereby agrees to accept as full compensation for its services hereunder; and (6) to pay over to Airco upon demand any profits becoming due and credited to Airco as aforesaid."

[2] Wilson Welder had a net deficit during the year here involved and is not a petitioner in this action.

[3] Sales had outstanding 125 shares of stock of $100 par value; Carbide's outstanding capital stock was 50 shares of $100 par value; Carbonic also had 50 shares of $100 par value.

[4] Mertens, Law of Federal Income Taxation (1948 ed.), Vol. 10A, p. 237.

[5] Finkelstein, The Corporate Entity and the Income Tax, 44 Yale L.J. 436, 448.

[6] In Ballantine, Separate Entity of Parent and Subsidiary Corporations, 14 Cal. L. Rev. 12, 18, the writer discusses this use of the agency concept as follows: "What is meant by such terms as ``adjunct,' ``agency,' ``instrumentality,' ``creature' or ``mouthpiece'? What conditions must exist to warrant a court in treating the A corporation as the mere adjunct of the B corporation? The word ``agency' is often used as a synonym of ``adjunct,' whatever that may mean, and as descriptive of a relation variously defined in the cases as ``alter ego,' ``alias,' ``device,' ``dummy,' ``branch,' ``tool,' ``corporate double,' ``business conduit,' ``instrumentality,' etc., but all in the sense of ``means' through which a corporation's own business is actively prosecuted."

[7] The case other than Southern Pacific relied upon by the Tax Court was Munson Steamship Line v. Commissioner, 77 F.2d 849. That case was explained in Moline Properties, Inc. v. Commissioner, supra, as depending upon a particular legislative purpose which justified disregarding the separate entity.

[8] Plaintiff's Exhibit P, No. 452, October Term 1917, is an income tax statement of the subsidiary, Central Pacific Co., showing payment of income taxes on $3,333,846.18, its total net income for 1913 less one-sixth (i.e., making an allowance for the two months before the income tax law went into effect March 1).

[9] 319 U.S. at 440-441.

[10] 8 T.C. 594, 611. After enumerating the facts considered pertinent, the court concluded: "It is true, of course, that, taken separately, some of the foregoing facts would not be sufficient in themselves to make inoperative the general rule that corporations are separate juristic persons and are to be so treated for tax purposes. We think, however, that when all these facts are viewed together they bring petitioners within the rule announced by the Supreme Court in Southern Pacific Co. v. Lowe, supra." Id. at 614.

It should be added that the Court of Appeals, whose opinion was written by its Chief Judge, did not so much as mention the agency argument now made by petitioners. Its only references to agency were isolated quotations from the Tax Court's opinion and the contracts quoted in footnote 1. The court's opinion phrases the question as "when a wholly owned subsidiary of a parent corporation shall be treated for purposes of income taxation as a separate taxable person, and when as merely a part of the corporate activities of the parent." 167 F.2d 304, 305.

[11] Id. at 307.

[12] Two basic distinctions between the Southern Pacific case and subsequent cases (except Gulf Oil Co. v. Lewellyn, 248 U.S. 71 (1918), which followed Southern Pacific) are immediately apparent. First, the Southern Pacific case involved taxation of the parent-owner rather than the subsidiary corporation; second, the question was when the income had been earned, rather than who had earned it. The importance of these distinctions is indicated by the fact that the subsidiary paid income taxes upon income received subsequent to March 1, 1913 (see footnote 8, supra), and that the parent did not dispute its tax liability for dividends from post-1913 earnings of the subsidiary. The decision is based on an interpretation of the Income Tax Act of 1913. The Court felt that it was not the intent of the Act to tax earnings prior to the effective date of the Act, and that the Central Pacific's pre-1913 income had actually accrued to the parent before the effective date of the Act. The opinion states that "The case turns upon its very peculiar facts, and is distinguishable from others in which the question of the identity of a controlling stockholder with his corporation has been raised." 247 U.S. at 338-339. By its very terms, the decision is limited to its precise facts.

[13] Much of the testimony introduced by petitioners had to do with the intercorporate relationship between Airco and its subsidiaries, the use of certain facilities by two or more of the subsidiaries, the duties of various officers who held positions with Airco and its subsidiaries, and the services performed for all of the subsidiaries by certain departments of Airco. So far as this testimony shows the integration of the corporate system and its direction by Airco, it is, as we have indicated, immaterial. So far as it indicates that the subsidiaries received the use of equipment and services for which they were not charged, it is relevant as showing that their income was distorted to that extent, but it does not indicate that the income received "belonged" to Airco at the time of its receipt. The Commissioner made allowance for this distortion by allocating over $400,000 of the expenses reported by Airco to petitioners under the authority given him by § 45 of the Revenue Act of 1938, 26 U.S.C. § 45.

[14] 45 B.T.A. 647, 650.

[15] As a practical matter, a considerable part of the assets of petitioners was supplied out of profits from their operations. Even though assets were purchased directly out of the earnings of a subsidiary, however, the amount withdrawn was entered in the accounts payable by the subsidiary and in the accounts receivable of Airco, since substantially all profits of the subsidiaries were, by contract, payable only to the parent.

[16] Since petitioners were required to pay all profits except very small amounts to Airco each year, it was obviously impossible for them to pay the accounts payable to Airco. See note 15. Mr. C.E. Adams, Chairman of Air Reduction Corporation, testified that the assets of the subsidiaries represented by the accounts payable could be realized by Airco only upon dissolution of the subsidiaries. In other words, there was never any expectation that the accounts would be paid prior to dissolution. Since no interest ran on these accounts, the "loans" were identical, except in name, with contributions of capital. See American Cigar Co. v. Commissioner, 66 F.2d 425; Hoyt v. Commissioner, 145 F.2d 634; Van Clief v. Helvering, 135 F.2d 254; Reading Co. v. Commissioner, 132 F.2d 306. Levy and Simonds, Stockholder Advances to Corporations — Are They Loans or Capital Contributions? 25 Taxes 127, 128, state that "intention to lend and expectation of repayment are necessary to the existence of a valid debt." The fact that no interest ran on these "loans" is, of course, further indication that they are capital contributions. Berry Motor Car Co., B.T.A. Memo. Op. Dkt. 99962, Jan. 25, 1941.

Title to gas cylinders used by petitioners, amounting in value to about $13,000,000, was retained by Airco, but the cylinders were used by the subsidiaries without charge. Whether these, too, were capital contributions we find it unnecessary to decide in this case. Free use of the cylinders by petitioners, if they were merely on loan, may have distorted the subsidiaries' income beyond the allocations made by the Commissioner, but that problem is not before us.

[17] Restatement of Agency, § 427.

[18] In United States v. Joliet & Chicago R. Co., 315 U.S. 44 (1942), a lessee railroad agreed to pay rental payments to the lessor's stockholders directly. The lessor thereafter carried on no active business. It was nevertheless held taxable on the income received by its stockholders, since they received the payments only because they held its stock.

[19] Art. 22 (a)-1 of Treasury Regulations 101, promulgated under the Revenue Act of 1938, provides:

"ART. 22(a)-1. What included in gross income. — Gross income includes in general compensation for personal and professional services, business income, profits from sales of and dealings in property, interest, rent, dividends, and gains, profits, and income derived from any source whatever, unless exempt from tax by law. (See sections 22(b) and 116.) In general, income is the gain derived from capital, from labor, or from both combined, provided it be understood to include profit gained through a sale or conversion of capital assets. . . ." See Eisner v. Macomber, 252 U.S. 189, 207 (1920); Merchants' Loan & Trust Co. v. Smietanka, 255 U.S. 509, 519 (1921).

In the case of a subsidiary who supplies the labor and the capital with which the income is earned, as is true of petitioners here, it can hardly be contended that it did not earn the income.

[20] Of course even a corporation which satisfies the usual tests of agency may be disregarded by the Commissioner if it is a sham or unreal. Higgins v. Smith, 308 U.S. 473 (1940); Gregory v. Helvering, 293 U.S. 465 (1935). Escaping taxation is not a "business" activity. See National Investors Corp. v. Hoey, 144 F.2d 466.

[21] The two main purposes for the adoption by Airco of the corporate subsidiary method of operation, as related by Mr. C.E. Adams, Chairman of Air Reduction Corp., were these:

"Frankly, in 1918 and still, Air Reduction, Inc., was and is a New York corporation. Even at that early date it became evident, as I already said, we were going to have plants scattered all over the United States. We didn't want to domicile the parent company in 48 states of the Union and have us subject to service in all those states, that is, have the parent company subject to service in all those states, and that was distinctly a reason for using this corporate setup in connection with operations to be run as divisions, just as the contract sets forth.

"Now, in addition to that, as a practical matter, out in the field and on the firing line, to have a representative, an officer, we will say, of Pure Carbonic, when trouble arises with a customer, a vice president of Pure Carbonic, who is not an officer of Air Reduction, Inc., at all, who goes in and straightens that out with that customer, increases his cudos [sic], helps him with all his negotiating efforts, with their competitors on the outside."

It is thus apparent that Airco was attempting to avoid the status of principal vis-a-vis its subsidiaries. As principal it would have been subject to service of process through its agents; as owner of the subsidiary it was not. See Peterson v. Chicago, R.I. & P.R. Co., 205 U.S. 364, 391 (1907); Cannon Mfg. Co. v. Cudahy Co., 267 U.S. 333 (1925). The purpose of having officers of subsidiaries who could deal directly with customers does not indicate an agency relationship. On the contrary, the very purpose of the organization adopted was to lead customers to believe that they were dealing with top men in the company actually manufacturing and selling the products they purchased.