108 F.2d 429 (1939)
HANLIN et al.
Nos. 7069, 7070, 7065, 7074.
COMMISSIONER OF INTERNAL REVENUE, and three other cases.
Circuit Court of Appeals, Third Circuit.
November 27, 1939.
W. A. Seifert, W. W. Booth, Joseph G. Robinson, and Reed, Smith, Shaw & McClay, all of Pittsburgh, Pa., for petitioners at No. 7069 and respondents at No. 7070.
William Wallace Booth and Reed, Smith, Shaw & McClay, all of Pittsburgh, Pa., for taxpayers.
Samuel O. Clark, Jr., Asst. Atty. Gen., and Sewall Key and Newton K. Fox, Sp. Assts. to Atty. Gen., for the Commissioner.
Before BIGGS, MARIS, and CLARK, Circuit Judges.
CLARK, Circuit Judge.
The complex details of the principal case at bar (Hanlin, Nos. 7069, 7070) have been well summarized by the editors of the Harvard Law Review in their comment upon the decision of the Board of Tax Appeals. The writer of this opinion cites it in a spirit of accuracy and loyalty:
"Section 118(a) of the Revenue Act of 1932 provides: `In the case of any loss claimed to have been sustained from any sale * * * of stock or securities where it appears that, within a period beginning 30 days before the date of such sale * * * and ending 30 days after such date, the taxpayer has acquired * * * substantially identical stock or securities, then no deduction for the loss shall be allowed * * *.' 47 Stat. 208 (1932), 26 U.S.C.A. § 118(a). The taxpayer sold $125,000 par value City of Philadelphia bonds, issued in 1918, maturing in May and November, 1948, and on the same dates purchased at the same unit prices $125,000 par value City of Philadelphia bonds, issued in 1919, maturing on March 1, 1949. The taxpayer sold $100,000 par value Federal Land Bank of Omaha bonds maturing July 1, 1953, redeemable after July 1, 1933, and on the same day bought at the same unit price $100,000 par value Federal Land Bank of Omaha bonds maturing January 1, 1956, also redeemable after July 1, 1933. The taxpayer sold $50,000 par value Federal Land Bank of Louisville bonds maturing July 1, 1953, redeemable after July 1, 1933; and on the same day bought at the same unit price $19,000 par value Federal Land Bank of St. Louis bonds maturing January 1, 1954, of which $10,000 par value were redeemable after July 1, 1934, and also $31,000 par value Federal Land Bank of Wichita bonds maturing January 1, 1954, and carrying no advance redemption date. Except to the extent indicated in each case, no provisions of the bonds purchased were shown to be different from those of the bonds sold. The Commissioner disallowed the deduction of losses on the ground that each of the three transactions was a wash sale under sec. 118(a)." 52 Harvard Law Review 530 (note).
In the companion case (Semple Nos. 7065, 7074) New Orleans Land Bank bonds were swapped for bonds of other Land Banks under somewhat similar circumstances.
The same authors have collected the scattered authorities. None of these are directly in point or controlling, nevertheless they warrant the conclusion: "The words `substantially identical' indicate that something less than precise correspondence will suffice to make the transaction a wash sale." 52 Harvard Law Review 530, 531 (note). The legislative history of the statute affords no clue to the nature of that "something less", see H.Rep. No. 350, 67th Cong., 1st Sess., p. 11; Hearings Before the Committee on Finance on H. R. 8245, 67th Cong., 1st Sess., Part 2, p. 387; H. Conference Rep. No. 486, 67th Cong., 1st Sess., p. 1. But it is to be found, we think, in the statute's purpose. When the taxpayer's ability to pay is diminished by the realization of losses, these losses should and do operate to reduce his tax. The wash sales provision is designed to eliminate fictitious losses. As losses are a matter of economics, so the fiction lies in the lack of any change in economic position on the part of the taxpayer. The negative absence of change of position cannot, of course, exist where a new economic factor has come into being which can and has prompted positive economic action. In other words, the "something less" that is required consists of economic correspondence exclusive of differentiations so slight as to be unreflected in the acquisitive and proprietary habits of holders of stocks and securities.
That being so, the specific issues at bar can be appropriately framed within the media of classification adopted with but immaterial variations by all writers on finance. In the words of a leading authority:
"* * * bonds are divided:
"I. According to the character of the issuing corporation
"II. According to the character of the security
"III. According to the purpose or function of the issue
"IV. According to the conditions attending payment of principal or interest."
Chamberlain and Edwards, Principles of Bond Investment (1927) p. 75.
See also, Badger and Guthmann, Investment Principles and Practices, Rev.Ed., 1936, pp. 143-158; Sakolski, Elements of Bond Investment, p. 40; Lagerquist, Investment Analysis p. 153; Lyon, Classification of Investment Bonds, 88 Annals of the American Academy of Political and Social Science 4.
The only diversity in the character of the various issuing corporations occurs in the comparison of the bonds of one federal Land Bank with those of another. We think it clearly lacking in substance. These institutions are all under the supervision of one central authority, formerly the Federal Farm Board, now the Farm Credit Administration, 12 U.S.C.A. §§ 781, 831, Executive Order 6084. They are at least secondarily liable on each other's bonds, 12 U. S.C.A. § 872. The difference between a promise by X to pay if Y cannot, and one by Y to pay if X cannot, may suggest refinements of distinction to the legalist; it is Tweedledum and Tweedledee to the economist. The credit of both is behind each engagement. So the liability of Land Banks is described as "mutual" to the lay investor, Badger and Guthmann, above cited, p. 700.
It cannot be denied, however, that the collateral security underlying the bond issues of different Land Banks, is physically distinct. That collateral, consisting of United States bonds and first mortgages, is held in trust by the farm loan registrar for the issuing bank and the prospective holders of its bonds, 12 U.S.C.A. § 853. The statutes and decisions give no indication that such collateral is pooled indiscriminately behind all the outstanding bonds of all the Land Banks. See 12 U.S.C.A. §§ 881, 891-898; McLucas v. Langworthy, D. C., 7 F.Supp. 457; Holmberg v. Anchell, D.C., 24 F.Supp. 594; Gallagher v. Clark, D.C., 7 F.Supp. 158; Partridge v. St. Louis Joint Stock Land Bank, 8 Cir., 76 F.2d 237; Brusselback v. Chicago Joint Land Bank, 7 Cir., 69 F.2d 598. Consequently, the credit of the United States being what it is, we must, for example, discriminate, if at all, between first mortgages on farms in the vicinity of Louisville, and on those round about Wichita. Those mortgages are, to be sure, selected according to searching and uniform standards, 12 U.S.C.A. §§ 723(a), 751-756. But can it be said that there is no substantial variance in agricultural and financial resource when Louisvillites are contrasted with Wichitavians? Though the average sense of obligation may well be the same, even an economist must recognize, by geographical definition, a salient divergence in, say the type (and marketability) of crops produced — or, perhaps, the likelihood of dust storms. This difference, we think, deprives the bonds of one Land Bank of substantial identity with those of another.
We turn, then, to the conditions attending the payment of principal and interest (unity of purpose and function is not disputed) of bonds having identical obligors and security. Here we are met with a series of chronologically slight, but mathematically concrete, disparities in the maturity dates of the bonds sold and bought. The financial writers we have referred to above, indicate four possible effects of this divergence, two subjective and two objective. Subjectively considered, the longer a bond has to run, the more "acceptable" is its duration from the point of view of freedom from care. See, Chamberlain and Edwards, above cited, p. 23. But the pleasant, yet distant (two decades more or less, barring death) prospect of a few extra months or years of passive coupon clipping is hardly a matter of substance. Almost equally tenuous is the investor's interest in diversification. We quote: "Diversification according to maturity date is still another method of insuring against undue losses. As previously noted the prices of high grade bonds vary inversely with interest rates. That is, high interest rates are accompanied by high bond yields or low bond prices. Conversely, low interest rates are accompanied by low yields and high prices. Let us suppose now that a large proportion of the investments of a specific fund mature at a time when interest rates are extremely low. The reinvestment of such funds can be effected only in disadvantageous terms, because of the general market situation. A contingency such as this can be guarded against, however, by purchasing bonds with varying maturity dates." Badger and Guthmann, Investment Principles and Practices, Rev.Ed., 1936, pp. 122, 123. Here at any rate, there is no evidence that the taxpayer has systematically hedged in the manner indicated, and so buttressed his theoretical ability to pay.
Objectively it is plain that maturity date has a direct arithmetical effect on the yield of a bond, since that all important percentage is a resultant of purchase price, face value, coupon rate (including length of interest periods) and duration. All other things being equal (as here), this effect diminishes according to an abstruse mathematical ratio as the extent of the duration increases. In other words, six months added to a duration of one year is vital — added to a duration of twenty years, negligible. In approximate figures the taxpayer's purchases increased the sixteen year maturity of his Philadelphia bonds by from four to ten months, and the twenty-three year maturity of his Omaha bonds by about two and one-half years. Reference to standard bond tables indicates that the taxpayer thus altered his economic position by reducing the yield on the bonds he sold to the extent of not more than one-twentieth of one per cent. We consider such a deviation insubstantial. In this connection it seems appropriate to observe that ordinary bond tables only go so far as to reflect yields varying one-eighth and/or one-tenth of one per cent, and are computed to durations reckoned semi-annually. See, Rollins, Tables Showing the Net Return from Bonds; Financial Publishing Company, Acme Tables of Bond Values; Financial Publishing Company, Universal Bond Values Tables, and cf. Johnson, Yields of Bonds and Stocks in Per Cent. Moreover, the importance of yield brings about a corresponding triviality in the unproved yet possibly cognizable dissimilarity in the redemption dates of the two sets of Omaha bonds. For they were purchased at a discount, and, consequently, their yield in the economic sense must be calculated from their respective maturity dates. The other repercussions of a variation in the period after which payment may be hoped for (or dreaded) occur in the subjective fields already discussed.
A possible further effect of the disparity in maturity dates lies in the state and generally predicted (but not predictable) future fluctuations of one of the ground swells of capitalism, the market rate of return. To quote again: "During periods of high rates on short time paper, short time bonds will sell at prices which make their yield approximately the same as that found in short time commercial loans. Long term bonds on the other hand, will sell on a somewhat lower yield basis, or at relatively higher prices, for the reason that the investor seeks such issues for the purpose of extending this favorable return for as long time as possible. Conversely when interest rates are especially low long term bonds sell at prices which yield somewhat more than short term bonds. At such periods, short maturity bonds are in greater demand, for investors are reluctant to tie up bonds for long periods at unfavorable interest rates. This stimulates the demand for short term issues and depresses the demand for long term issues." Badger and Guthmann, Investment Principles and Practices, Rev.Ed.1936, pp. 66, 67. That these considerations cannot and do not enter into the case at bar is, of course, established by the identity in unit prices existing between the bonds sold and the bonds purchased.
We are aware of the uncertainties attendant upon the application of the elastic weasel word "substantially" to other and perhaps economically divergent circumstances. However that may be, we feel that a conscientious construction of the statute compels the judgment we are about to render. Other courts have hazarded the possibility of future uncertainty in construing the cognate phrase "substantially all" as employed for the purpose of determining whether two corporations are affiliated, see Handy & Harman v. Burnet, 284 U.S. 136, 52 S.Ct. 51, 76 L.Ed. 207, or whether a transaction is a reorganization, see Nelson Co. v. Helvering, 296 U.S. 374, 56 S.Ct. 273, 80 L.Ed. 821; Western Industries Co. v. Helvering, 65 App.D.C. 205, 82 F.2d 461. Their labors have resulted in a legislative substitution of clear cut percentages in lieu of "substantially", 26 U.S. C.A. § 141(d) (affiliation); 26 U.S.C.A. § 112(g) (reorganization). We may say that the statute with which we are now dealing is capable of similar clarification.
The record in the Semple case discloses a plain error in computing the allowable extent of the taxpayer's deduction for charitable contributions, United States v. Pleasants, 305 U.S. 357, 59 S.Ct. 281, 83 L.Ed. 217. Though excusable, Heinz v. Commissioner, 3 Cir., 94 F.2d 832, it can and should, we think, be corrected.
The decision of the Board of Tax Appeals in the Hanlin case (Nos. 7069, 7070) is affirmed. Its decision in the Semple case (Nos. 7065, 7074) is reversed and remanded for further proceedings with respect to the error specified in this opinion, and in all other respects affirmed.