*258Wolcott, Chancellor, delivering the opinion of the Court:
The assignments of error raise two questions which will be stated and answered as follows:
1. Was the exchange of La France shares for Postum shares a taxable transaction?
This question is to be answered in the light of the income' tax statute of this State as the same existed in 1928. The pertinent provision of the statute is found in Section 1 of Article 1, Chapter 9, 32 Delaware Laws, ap*259proved March 29, 1921, wherein net income is defined. The provision is as follows:
“The phrase ‘net income’ means the aggregate of all gains, profits, salaries, wages, compensation for personal service of whatever kind and in whatever form paid, income derived from professions, vocations, business, trade, commerce, sales or dealings in real or personal property growing out of the ownership or use of or interest in such property, also from interest, dividends, securities or the transaction of any business carried on for gain or profit, or gains or profits and income derived and actually received, into possession by a taxable from any source whatever, and also the share of the profits of any taxable in a co-partnership whether such profits have been divided or otherwise, less the aggregate of the deductions provided for in Section 4, provided, that for the purpose of ascertaining the gain or loss, resulting from the sale or other disposition of property, acquired before January first, 1920, the fair market price or value of such property as of said date shall be the basis for determining the amount of such gain or loss.”
The argument in the court below and in this court has proceeded on the assumption which we accept without examination as sound, viz., that if the exchange in 1928 of La France stock for Postum stock was simply the mechanics of a reorganization of La France, and not a sale in the ordinary sense, then the taxable realized no taxable net income within the meaning of the act.
This assumption requires, then, that we determine whether or not the transaction of 1928 was in legal contemplation a reorganization of La France.
Even where there has been a reorganization of a corporation, yet if a gain is realized by its stockholders as an incident thereto, it does not necessarily follow that the gain is non-taxable. In U. S. v. Phellis, 257 U. S. 156, 42 S. Ct. 63, 66 L. Ed. 180; Rockefeller v. U. S., 257 U. S. 176, 42 S. Ct. 68, 66 L. Ed. 186; Cullinan v. Walker, 262 U. S. 134, 43 S. Ct. 495, 67 L. Ed. 906; and Weiss v. Stearn, 265 U. S. 242, 44 S. Ct. 490, 68 L. Ed. 1001, 33 A. L. R. 520, the gain to the stockholders, though it was incidentally *260realized from what was a reorganization, was nevertheless adjudged to be taxable.
The agreement of 1928 recited that the parties thereto “have agreed upon a plan* of reorganization involving” La France and Postum, “and as a part thereof, the Sellers (stockholders of La France) have agreed to sell, etc.,” to Postum all the capital stock of La France. This recital indicates that the alleged reorganization of the two corporations was an operation having two or more phases. A sale of their stock to Postum by all the stockholders of La France was one part only of the operation. That is what the agreement says.
But the record fails completely to show any thing evidentiary of a reorganization of the two corporations beyond the sale of stock. We must, therefore, conclude that if a reorganization of the two corporations is to be spelled out of the facts shown by the record, it is solely because of the purchase by Postum of all the stock of La France from the holders thereof.
Now, certainly, no one would suggest that such a purchase constituted a reorganization of Postum. It is earnestly argued, however, that it does show a reorganization of La France.
What is the consideration by which the fact of reorganization is to be tested for the purpose of determining whether a stockholder participating therein has received a taxable gain? Certain it is that the denomination by interested parties of a corporate transaction as a reorganization, does not make it so. Weiss v. Steam, supra. The legal result of acts done is not what the parties declare, but what the law adjudges.
The question above put has never arisen in this court. *261Morris, Ex’r, v. State School Tax Dept., 4 W. W. Harr. (34 Del.) 195, 148 A. 97, cited by the taxable has no pertinency. The question has arisen on several occasions and under varying circumstances in the Supreme Court of the United States. Eisner v. Macomber, 252 U. S. 189, 40 S. Ct. 189, 64 L. Ed. 521, 9 A. L. R. 1570; U. S. v. Phellis, 257 U. S. 156, 42 S. Ct. 63, 66 L. Ed. 180; Rockefeller v. U. S., 257 U. S. 176, 42 S. Ct. 68, 66 L. Ed. 186; Cullinan v. Walker, 262 U. S. 134, 43 S. Ct. 495, 67 L. Ed. 906; Marr v. U. S., 268 U. S. 536, 45 S. Ct. 575, 69 L. Ed. 1079. The principal test laid down by those cases seems to be this — does the stockholder have the same identity of interest in the new or reorganized corporation as he had in the old, without a severing from the assets of any part thereof in some realizable form for his individual benefit.
The taxable puts the test as follows: “the thread of differentiation that runs through all the cases * * * is the identical evidence of ownership and sameness of quantum in contradistinction to the receipt of an obligation of the corporation in exchange of its common stock with resultant segregation and distribution of corporate assets.”
It is hazardous to undertake to lay down any test as a universal solvent for all possible fact ingredients. For the purposes of this case, we may take the test to be as phrased by the plaintiff in error himself. By that test it appears clear to us that the result of the agreement of 1928 was not a reorganization of the La France company in which no basis of a taxable gain is assertable. The transaction appears to us in legal contemplation to have been an ordinary sale of stock in consideration for stock.
What the sellers had after the transaction was something entirely different from what they possessed before. It is impossible for us to see how, as the taxable asserts, *262it can be maintained that the interest which La France stockholders had in the assets of their corporation was the same after the transaction as it was before. If so, having a one hundred per cent, interest before, they would have a one hundred per cent, interest after. That of course is quite obviously not true. Indeed after the transaction, La France stockholders lost all their interest in the company’s assets except as they derived their rights indirectly through ownership of stock of Postum which as a sole stockholder of La France was in turn derivatively interested therein. This is so, if Postum did not acquire La France’s assets by purchase or dissolution. On that assumption, it is quite plain that what La France stockholders possessed after the transaction was something quite different from what they possessed before.
But if Postum took over the physical assets of La France and commingled them in a mass with its own assets, it is equally clear that the interest of the old La France stockholders in its former assets was entirely different after the transaction from what it was before. Their former one hundred per cent, interest then became diluted by the admission to a participation therein by all other holders of Postum stock.
It was equally diluted if the assets were not taken over and commingled with Postum’s. Furthermore, old La France stockholders became proportional owners of all of Postum’s assets which were very large.
Thus it appears that what La France stockholders received was something entirely different from what they had given up. Their old business became solely owned by another company which was and for sometime prior thereto had been engaged in the manufacture and sale of foodstuffs and household commodities. It became so to speak a branch *263or department of the other company. La France stockholders, those of them who took Postum stock, embarked upon a new and different venture. They were then in a corporation which not only conducted their old line of business but as well numerous other lines. The case is not one, then, where stockholders of a company revamped its structure but left undisturbed the identity of their business and their relative participations therein. What their proportionate ownership was in the Postum Company, we are unable to say, because the record does not reveal what the capitalization of that company was. Certain it is that whatever their interest was, it was entirely different from what their proportionate interest in La France had been. We are unable to agree with the taxable in his contention that the quantum of interest which La France stockholders possessed after the transaction was the same as that which they possessed before. It was not the same thing differing only in the unimportant circumstance of the character of stock representing it. The transaction was one of sale. It was as much so as if it had been for cash and the proceeds had been invested in Postum stock.
The insistence with which the plaintiff in error presses Weiss v. Stearn, supra, upon our attention, prompts us to observe that we can find nothing in that case which is adverse to the view we take of this one. As we read that case it holds this — that if the stockholders of a corporation sell one-half of its assets for cash, distribute the cash among themselves and retain a one-half ownership in the corporation, the profit realized on the cash side of the transaction is taxable, while the stock they received for the other half is not. That in its last analysis is what the case holds. It hardly seems necessary to say that Weiss v. Steam, in no wise approaches the proposition of this case.
In Lackey v. State School Tax Department, 5 W. W. *264Harr. (35 Del.) 507, 168 A. 194, Judge Rodney .when sitting in the Superior Court, was called upon to examine substantially the same question as we have before us here. We take the same view of the law as he took. Our conclusion under the first head of the case is that the transaction of July, 1928, was not such a reorganization of La France as removes what was received thereunder by the taxable from the reach of the taxing statute if a gain was realized.
2. Being a sale, the question next arises of whether the taxable realized a taxable gain therefrom. The value of the Postum stock which composed the consideration received by the taxable was fixed by the State Tax Commissioner at $62.50 per share. No complaint is made against that valuation.
To ascertain whether or not that price yielded a gain, it was necessary for the State Tax Commissioner to ascertain “the fair market price or value” of the La France stock as of January 1, 1920, the taxable having acquired the same before that date. The statute so provides. The commissioner valued- the stock as of that date at $114.09 per share. On this basis he found a taxable profit of $411,982.50 and assessed the taxable accordingly.
The commissioner arrived at the value of $114.09 per share by the following method. He ascertained the average net earnings of the company for a period of five years prior to January 1, 1920, and capitalized the same at seven per cent. This gave him a figure for the total net value of all the outstanding stock. Having that figure, the value of the taxable’s shares was a simple matter of calculation.
The taxable contends that the method of valuation thus employed, however unobjectionable it might be conceded to be in many instances, under the facts surrounding this particular case is erroneous.
*265In an action to recover taxes alleged to have been illegally collected, the taxpayer has the burden not only of proving that the assessment was unlawfully made, but also of showing the extent of the illegal exaction. U. S. v. Anderson, 269 U. S. 422, 46 S. Ct. 131, 70 L. Ed. 347; Reinecke v. Spalding, 280 U. S. 227, 50 S. Ct. 96, 74 L. Ed. 385. But in another way, this means that the burden is on the taxpayer first to show the fallacy of the theory on which the assessment was made, and second, the theory which should have been employed and what, if anything, was properly due on that theory. It is so in such a case because the plaintiff in error affirmatively seeks a judgment for money due.
But where the proceeding is one in review of an assessment which the tax commissioner is seeking to establish, as here, the burden is on the commissioner who is active in asserting an affirmative to establish the claim. In such case, there is no duty resting on the taxable to show the correct amount due, if any. Helvering v. Taylor, 293 U. S. 507, 55 S. Ct. 287, 79 L. Ed. 623, affirming Taylor v. Com’r of Internal Revenue (C. C. A.), 70 F. (2d) 619. A burden rests on the taxpayer, however, in reviewing the commissioner’s finding to show that there was error in his determination. The cases just cited so hold.
We accept, then, the taxable’s contention that if he has shown that the method of ascertaining the value of La France stock which the Tax Commissioner adopted was erroneously prejudicial to him, and that the gain arrived at was accordingly excessively stated, the taxable is under no further burden to show what method of valuation should have been used and that, if such method had been used, the correct amount due as tax, if any, would have been so much.
We turn then to the question — is'the record such that we are justified in saying that the method of valuation *266employed by the commissioner was erroneous ? There is a presumption that the finding of the Tax Commissioner was correct. Taylor v. Com’r of Internal Revenue, supra. We would not, therefore, be warranted in overthrowing his determination unless it affirmatively appears that it is vitiated by error.
The La France stock is to be valued as of January 1, 1920. It was the duty of the Tax Commissioner, as it is ours, by the use of historical imagination to transplant the problem of value to that date and examine it in the light of the then known facts. In substance this is stated by Holmes at page 523 of his work on Federal Taxes (1923 Ed.) to be the cardinal rule.
Now there was no “fair market price or value” in 1920 for La France stock in the sense that the same was demonstrable by transactions of purchase and sale. The stock was closely held. There were no sales of it. Fail-market price has been said to be “the resultant of the two opposing views of a willing seller and a willing purchaser, the former of whom is not compelled to sell and the latter of whom is not required to buy.” Allied Chemical & Dye Corp. v. Steel & Tube Co., 14 Del. Ch. 64, 69, 122 A. 142, 144. Many courts have stated this thought in varying ways. There being no market transactions from which aid could be obtained in the process of measuring market value, it became necessary for the commissioner to imagine a sale on January 1, 1920, and to ask the question — what would a willing purchaser have then given for the stock? In answering that question some reasonable satisfactory formula had to be resorted to for a determination of its worth to the theoretical purchaser.
The Tax Commissioner used a formula frequently employed in cases such as this where the worth of a com-*267pony and consequently the value of its stock is almost entirely due to the good-will or earning power of the business. The factors of tangibles and book value are of little significance in this case. The question is almost entirely one of the valuation of intangibles that comprise good-will in its broadest sense.
The formula used by the Tax Commissioner has been often applied in practice. The formula is as follows: The average annual net earnings over a period of years prior to the valuation date (five in this case) are ascertained from the books and capitalized at a rate which under the circumstances of the particular case is considered to be fair (in this case seven per cent). It is used in the Federal taxing practice and is there known as Appeal and Review Memorandum 34. The rates used in application of the formula are of course not uniform, varying with the diversity of differing conditions. Use of the formula has received the sanction of judicial approval in appropriate cases. Pfleghar Hardware Specialty Co. v. Blair (C. C. A.), 30 F.(2d) 614. The rate of seven per cent., used in the instant case, if not generous, was certainly not unfavorable to the taxable. When the formula has been used by the Federal tax officials the rate has ranged from eight per cent, to twenty per cent, depending upon the facts in each case. Holmes, Federal Taxes (1923 Ed.), p. 527. In Couzens v. Commissioner, 11 B. T. A. 1040, which was very like in fact analogy to the case at bar, the rate of ten per cent, was used. Of course the lower the rate, the higher is the value. It would seem then that in using the seven per cent, rate the Tax Commissioner was making some concession due to the particular facts before him.
We are at a loss to discover what other possible method of valuation the Commissioner could have used in light of the facts in hand. It will not do to say that subsequent *268years after 1920 showed a rather remarkable development of the business that La France had theretofore enjoyed, yielding greatly increased net earnings. All the Tax Commissioner had before him was the record of its past. It was said in White & Wells Co. v. Commissioner of Internal Revenue (C. C. A.), 50 F. (2d) 120, that while ordinarily the five-year period preceding the valuation date was a fair period to take for ascertainment of the average yearly net earnings, yet if that period included abnormal years such years should be eliminated. That is reasonable. But upon our study of the figures, it was impossible to select any period of years prior to January 1, 1920, which would have made a showing more favorable to the taxable than did the period used by the Commissioner, unless the period of two years (1915 and 1916) alone had been selected. But that would hardly have been justifiable for those two years stand out in comparison with all the other years from 1903 to 1920 as abnormal. The average yearly earnings for the five-year period were found by the Commissioner to be $109,073.53. The net earnings for 1917 were $75,932.91, for 1918, $72,853.07 and for 1919, $102,846.83. Only in 1915 and 1916 did the net earnings exceed the average ascertained by the Commissioner;
If the Commissioner had undertaken to value the stock on a basis other than that shown by its theretofore demonstrated earning power, he would have indulged in pure speculation. We know of no instance where values have been predicated on other than data in hand.
It is quite true that the use of the formula is predicated on the assumption that it is a fair reflection of future earning power, and that in some cases later experience will disclose that the future earning power exceeded in substantial amounts that which the formula forecast. The converse is also true. Certainly, however, in the matter of *269the capital value of future earnings is something that no appraiser has yet been found able to give. Future earnings are dependent on a multitude of varying factors having to do with general economic conditions, costs of labor and materials, possible new competing products, fluctuating market demand, personnel of management, etc., etc. No formula can possibly be devised that will approximate an absolutely accurate reflection of such uncertain kaleidoscopic conditions lying in the future. All that can be meant by the expression that the formula must fairly reflect future earnings is, that it must reflect such earnings in the light of facts presently known as of the date of its application.
In the light of the facts as they appeared on January 1, 1920, we do not see how it can fairly be said that the application of the formula that was used was an unfair and therefore erroneous method of valuing the taxable’s stock.
It appears to us to have been the only method that could have been used if pure speculation was to be avoided. It is the method that was employed in the case of Couzens v. Commissioner, supra, where the problem was one of retrospectively fixing “a fair market price or value” of Ford Motor stock as of March 1, 1913. The only difference in the formula there employed and here used is in two particulars. There the rate of ten per cent, was used as against seven per cent, here (a difference favorable to the taxable in this case) ; and the base in the Couzens Case was current net earnings for the first two months of 1913 projected for the entire year, while here the base was the average annual net earnings for the five year period prior to January 1,1920. There is no room in this case for taking a base on the principle of the Couzens Case, because the date of valuation here was at the commencement of the year. If, however, the valuation date had been March 1, 1920, instead of January 1, 1920, and current earnings for *270the first two months of 1920 had been projected for the entire year and used as the base, as in the Couzens Case, the result would have been far worse for the taxable, because the net earnings of 1920 were only $94,473.93.
When the taxable and other La France stockholders sold their stock to Postum in 1928 they received therefor a price per share based on a capitalization of the net earnings of La France for one year, the last preceding one at the rate of eight per cent. Had the same method of valuation been used in valuing the stock on January 1, 1920, when an hypothetical sale is deemed to have been made, the result would have been more adverse to the taxable than the Tax Commissioner found. This would indicate that the reasoning of the Tax Commissioner applied to the data of January 1, 1920, was more liberal to the taxable than was his own reasoning when applied to himself on the data of July 16, 1928.
The taxable makes a computation intended to demonstrate that the stock was undervalued by at least $21.78 per share on the basis of the Commissioner’s own method of computation. His calculation is clearly erroneous. He takes the net value of the company ascertained by the Commissioner in the manner already stated, that is by capitalizing its average net earnings at seven per cent., adds capital and surplus of $468,345.94, making a total of $2,026,539.24, then deducts goodwill carried on the books at $169,709.12. This gives him a figure of $1,856,830.11 as the net worth of the company, which works out to .$135.87 per share. Clearly, such a method of computation is not permissible. It ignores the fact that the formula used by the Commissioner is calculated to put a value on the entire assets of the company including of course the assets represented by capital and surplus. When the total value of all assets is thus arrived at, it is improper to make a duplica*271tian by adding assets that are already represented in the total. If the line of thought suggested by the taxable’s contention in this particular were properly pursued, he would have to do this. He would have to take the capital and surplus, allow it a yield of a given rate, deduct that yield from the average annual net earnings, then capitalize the balance at a given rate. If the rate of seven per cent, were used for yield on capital and surplus and for capitalizing the balance of the annual earnings, the result would be worse for the taxable than the Commissioner found; and if the suggestion of the United States Treasury as stated by Montgomery in his 1927 Income Tax Procedure, p. 509, were followed of allowing eight or nine per cent, for returns on tangibles and fifteen per cent, for capitalizing the earnings applicable to goodwill, the result to the taxable would be very much worse.
The taxable complains that the per share value of his stock as determined by the Commissioner allows nothing for such intangibles as potential value, future prospective profits, capitálized value of investments in samples, advertising and selling costs, distributor territories that had been built up, trade-marks, trade-names, secret processes and other intangibles. Every one of these items, except advertising expenses, the formula used in the valuation was designed to evaluate. They were treated as having a market worth equal to the capital value of what the experience of the company showed them capable of earning. As to the investment in advertising which looked solely to the future and could not possibly have a present fact experience upon which judgment could be predicated, its value as a future earning producer was highly problematical.
The Chief Justice in the court below correctly observed that “ it is difficult to fix upon the fair value of shares of stock as of a given date by any set rule or formula, and it *272is true also that expectation of future increase in value reasonably entertained may not be ignored. Yet that expectation must rest on a substantial factual basis.” Taking the basis of facts as they appeared on January 1, 1920, we are of the opinion that every element of value reasonably evident was taken into account and that the determination of the Commissioner is not shown to have been attended with error.
The judgment below will be affirmed.