The Commissioner petitions for review of a decision of the Tax Court. 19 T.C. 1082 (1953). The facts were stipulated and were found accordingly. We will attempt to state them generally so that this opinion does not become a welter of figures.

Respondents' decedent held stock of the Southern Natural Gas Company. In 1947 she and the other shareholders of the gas company received a distribution partly in cash and partly in property. The property consisted of shares of stock of the Southern Production Company, Inc., which had, been held by the gas company. The production company stock had been acquired by the gas company at a cost of $3,199,950. At the time of distribution, its fair market value was $8,983,407.75. The total cash distribution by the gas company was $2,113,722.03, making the total of the cash and property received by the shareholders $11,097,129.78. At the time of distribution, the gas company had available for dividends, earnings or profits totaling $5,674,586.32. The footnote contains a tabular account of the arithmetic of what happened.1 The decedent treated roughly half of what she received from the gas company as ordinary income for the year 1947. The Commissioner's deficiency assessment held her total receipts from the gas company to be taxable as ordinary income. Respondents were successful in the Tax Court, and the case comes here.

Forgetting for the moment the cash distribution, the case amounts to this, when reduced to its essentials. A corporation made a distribution in kind to its shareholders of property whose fair market value at the time of distribution exceeded its cost to the corporation and also exceeded the corporation's earnings or profits available for dividends, when considered apart from the property's appreciation in value. The earnings or profits were, however, sufficient to cover the cost to the corporation of the property distributed. In order to decide the tax effect on the distributees, therefore, we must determine whether a distribution in kind of appreciated property reduces the distributing corporation's earnings or profits by the fair market value of the property, without addition of the increment in value, as the Tax Court held, or whether it reduces earnings or profits only by the cost of the property. The two methods produce substantially different results. If earnings or profits, without addition of the increment caused by the appreciation in value, must be matched against the fair market value of the property, the distribution will be a dividend only to the extent of earnings or profits and the distributees will be taxed on that amount only, with the excess over earnings or profits going first to reduce the basis of the shares held by the distributees and any excess over basis being taxed as a capital gain under Section 115(d).2 This was the Tax Court's view. On the other hand, if earnings or profits, again without addition of the increment in value, must be matched against the cost of the property, the entire distribution will be a dividend and the distributees will be taxed upon the full fair market value as ordinary income.

The problem acquires its setting from the following provisions of the Internal Revenue Code. By Section 22(a) "`Gross income' includes gains, profits, and income derived from * * * dividends * * *."3 But Section 22(a) does not define a "dividend." For that purpose, its subsection (e)4 directs us to Section 115. Section 115(a) defines a dividend as a corporate distribution to its shareholders "out of its earnings or profits. * * *" Section 115(b) provides a conclusive statutory presumption to the effect that every corporate distribution is "out of earnings or profits to the extent thereof. * *" Finally, Section 115(j) values a property dividend by providing that the property received "shall be included in gross income at its fair market value. * * *"5

The Tax Court said that the genesis of the problem was the Supreme Court's decision in General Utilities & Operating Co. v. Helvering, 1935, 296 U.S. 200, 56 S. Ct. 185, 80 L. Ed. 154. There General Utilities distributed to its shareholders property that had appreciated in value over its cost to the corporation. The Court held that General Utilities derived no taxable gain from the distribution. Because of procedural points involved and the sequence of the Commissioner's arguments before the Board of Tax Appeals, the Fourth Circuit, and the Supreme Court, the scope of the holding of the General Utilities case is uncertain.6 There is no doubt, however, that the case has received judicial and administrative acceptance as standing for the proposition that a corporation does not realize income from the distribution of property which has appreciated in value over its cost.7 We will so accept it for purposes of discussion here. Construing the General Utilities case as stated above, the Tax Court held that the so-called rule of that case should be applied in computing the amount of a corporation's earnings or profits available for dividends even where no question of the effect on its taxable net income was involved.8 Therefore, the appreciation in value of the property held and distributed by the gas company here was not added to its earnings or profits. The result was that the distribution of the cash and property was a dividend and, thus, taxable to the distributees as ordinary income, only to the extent of the gas company's earnings or profits.

The Commissioner's contention was that earnings covered the cost of the property distributed and, therefore, the distribution was "out of earnings or profits" under 115(a) and was backed by earnings or profits "to the extent" of cost under 115(b). The distribution was, therefore, a "dividend." Having been classified as a dividend under 115 (a) and (b), 115(j) then comes into operation and imposes a tax upon the distributees to the extent of the then fair market value of the property received. This argument was rejected by the Tax Court because it was said to destroy what was thought to be the proper correlation between taxability to the shareholder-distributees and the effect upon statutory earnings or profits of the corporation and because it was thought that to allow the Commissioner's argument to prevail would be to allow him to do indirectly what the Court in General Utilities prevented him from doing directly, i.e., write up the earnings or profits of the distributing corporation by the amount of the unrealized appreciation of the property distributed. We concede, of course, that the solution is not free from difficulty, but we think the Commissioner's theory has none of the effects attributed to it by the Tax Court and is the better of the two rules.

There is no statutory definition of "earnings or profits." The attempt to ascertain its meaning is usually made in a negative way. That is, there is general agreement that "earnings or profits" are not the same as taxable net income.9 The General Utilities case has been accepted as another negative delineation of earnings or profits, in that it has been extended, as the Tax Court did here, as supporting the proposition that earnings or profits do not include the appreciation in value of the property distributed.10 Furthermore, Section 115(l), which came into the Code in 1940, seems to support the rule that unrealized gains should not increase earnings or profits by providing that a realized but nonrecognized gain or loss shall not increase or decrease earnings or profits.11 That section makes it clear that if a gain is not only nonrecognized but also not even realized, a fortiori, it should not be added to earnings or profits. Cf. Commissioner of Internal Revenue v. Wheeler, 1945, 324 U.S. 542, 65 S. Ct. 799, 89 L. Ed. 1166. But, if the appreciation in value of the property is an unrealized gain and, thus, may not be added to earnings or profits, we think it may not be subtracted either. Prior to distribution, the corporation carries the property on its books at its adjusted cost basis, since, as we have seen, the appreciation in value of the property, in the absence of a closed transaction, does not become realized income. Of course, it is quite true that the increment in value is distributed by the corporation in the sense that it inheres in the property and leaves the corporation when the property does, but its distribution does not decrease earnings or profits because it never became part of earnings or profits. If the distribution of appreciated property were a realization of income by the corporation to the extent of the increment in value, simple arithmetic would require that the distribution would cause a simultaneous increase and decrease in the earnings or profits by the amount of the appreciation. That is, fair market value could be subtracted from earnings and profits only if fair market value (made up of cost, plus appreciation) had theretofore been added to earnings and profits. The result is that cost is the only real variant and, therefore, the only real reduction in earnings or profits is to the extent of cost.

We think the correct approach to a case of a corporate distribution of appreciated property is dictated by the Code. Section 115(a) and (b) define a dividend. For purposes of characterizing a distribution as a dividend vel non, i. e., whether it is out of earnings or profits, the sole relevant test is from the standpoint of the distributing corporation and from that standpoint fair market value is immaterial. The amount of a distribution, once it is determined that the distribution is a dividend, is governed by 115(j) and is measured solely from the standpoint of the distributees and with reference to fair market value, and for that purpose the earnings or profits of the corporation are immaterial.12 Nor do we think this destroys any supposed correlation between taxability to distributees and effect on the corporation's earnings or profits. Indeed, there can be no correlation, in terms of amounts of money, for tax purposes, as between what the corporation distributes and what the shareholders receive, because, by hypothesis, the corporation does not, but the distributees do, realize income to the extent of the appreciation in value of the property distributed. The only correlation possible is that taxability to the distributees correspond with the source of the distributed property in the earnings and profits of the corporation. The Tax Court's insistence that there must always be an exact correlation between the amount of dollars by which a dividend in kind affects the earnings or profits of a corporation and the amount of dollars that are taxable in the hands of the stockholder receiving the dividend renders Section 115(j) meaningless and makes of 115(a) and (b) the sections which not only determine the character and the source of the distribution but which value the distribution in the hands of the stockholder.

The following example, suggested to us by the Commissioner, shows that the result of the Tax Court's decision completely distorts the corporation's earnings or profits and its general financial picture. Assume that a corporation has $100,000 worth of capital and $12,000 of earnings or profits. It owns property which cost $10,000 but has now appreciated to $15,000. The property is distributed in kind to its shareholders. The General Utilities case is said to prevent a realization of income to the corporation of the $5,000 unearned increment and also to prevent increasing earnings or profits by the $5,000 appreciation. Now, if the $15,000 fair market value of the distributed property, which includes the $5,000 appreciation, is charged against earnings or profits, there results a $3,000 deficit in that account or a reduction of capital by that amount. Really, however, the corporation's investment in the property was all that it parted with from a tax standpoint. As a matter of fact, after the distribution, the corporation's capital is not only wholly intact but it has $2,000 of earnings or profits remaining and available for future dividends. If we return to the facts of our case, we find that the Tax Court's solution causes equally unrealistic results in the gas company's earnings or profits and financial picture. After the cash distribution the gas company had $3,560,864.29 of earnings or profits available for dividends. To reduce that amount by the fair market value of the distributed property ($8,983,407.75), not only consumes all of the earnings or profits but results in a deficit in that account in the amount of $5,422,543.46 or a reduction of capital in that amount, which is a book situation altogether at odds with the facts, because in reality capital not only was still intact but there were $360,914.29 of earnings or profits left. The Tax Court's decision not only unduly limits the amount which is taxable to the shareholders as a dividend by reason of the instant distribution but removes $360,914.29 of actual earnings or profits from future taxation as a dividend on any subsequent distribution by the gas company.

We hold that upon a corporate distribution of property that has appreciated over cost but whose cost is exceeded by earnings or profits, considered without addition of the property's appreciation in value, the proper reduction of earnings or profits is to the extent of cost. Tax-wise, this means that in our case the entire distribution was a "dividend" and the distributees are taxable upon the full fair market value received. We think this does no violence either to Section 115(l) of the Code or to the General Utilities case, but, on the contrary, is entirely consistent with both and with what we think is required by 115(a), (b), and (j) in a case of this kind.

It must be conceded that, while we are not alone in our view, almost all the Tax Court cases have gone the other way.13 In R. D. Merrill Co., 1945, 4 T.C. 955, 959-963, a case of a distribution in kind of depreciated property, the Tax Court held that earnings or profits should be reduced by cost rather than the lower, fair market value. See Beach Petroleum Corp., 5 CCH T.C.M. 638 (1946), semble, where fair market value exceeded cost. In Reynolds Spring Co. v. Commissioner of Internal Revenue, 6 Cir., 181 F.2d 638, certiorari denied, 1950, 340 U.S. 821, 71 S. Ct. 53, 95 L. Ed. 603, another depreciated property distribution, cost rather than the lower, fair market value was used to reduce equity invested capital for purposes of the excess profits tax. Cf. Charles Leich & Co. v. United States, 7 Cir., 1954, 210 F.2d 901. The concurrence in Guinness v. United States, 1947, 73 F. Supp. 119, 132-134, 109 Ct. Cl. 84, certiorari denied, 1948, 334 U.S. 819, 68 S. Ct. 1083, 92 L. Ed. 1749, is in agreement with our theory and result in treating a distribution of appreciated property. The majority disposed of the case on the theory of Commissioner of Internal Revenue v. Court Holding Co., 1945, 324 U.S. 331, 65 S. Ct. 707, 89 L. Ed. 981. Commissioner of Internal Revenue v. Wakefield, 6 Cir., 1943, 139 F.2d 280, and Binzel v. Commissioner of Internal Revenue, 2 Cir., 75 F.2d 989, certiorari denied, 1935, 296 U.S. 579, 56 S. Ct. 90, 80 L. Ed. 409, reach our result upon different principles.

No purpose would be served by an extended discussion of the cases since we have set out above the two major theories that have competed for acceptance as the solution to our problem. With variations of language, the decisions espouse one or the other. The Tax Court's opinion in this case presents one view as well as we think it can be presented. For the reasons stated we think the other view is the better one.

After this opinion was prepared but before it was filed, the Second Circuit reached the same result upon the same theory in a case involving the same corporate distribution as is involved here. Commissioner of Internal Revenue v. Hirshon Trust, 2 Cir., 213 F.2d 523.

The judgment of the Tax Court will be reversed and the cause remanded for proceedings in accordance with this opinion.