This appeal by the United States from a judgment for the plaintiff in a suit so recover the amount of a deficiency in federal estate taxes paid under protest presents no issue of fact, all facts having been either settled by the pleadings or stipulated, and only a single question of law. That is whether, for the purpose of determining the amount of a decedent's net estate, a deduction may be taken under 812(d) of the Internal Revenue Code of 1939, quoted in material part in the margin,1 for bequests to admitted charities which will take effect in the event that the testatrix' sister, aged 82 at the decedent's death, predeceases both the decedent's daughter and daughter-in-law, who at the testatrix' death were aged 67 and 68 respectively. Stated in simple terms, the contingency which will defeat the gifts to charity is that an 82-year-old woman will survive two younger women aged 67 and 68.

The parties by stipulation have agreed that on the basis of accepted actuarial tables of mortality 'the probability that charity will take the principal of the trust is 0.91, or stated another way and upon the same basis, the probability that a person aged eighty-two will survive both of two persons aged sixty-eight and sixty-nine2 respectively is 0.09, or approximately one chance out of eleven.' In other words, the statistical chances are about eleven to one that charity will eventually receive the bequest. On the basis of this actuarial computation, and also taking the view that aside from mathematical computations persons in general would regard the possibility that an octogenarian would survive two persons fourteen and fifteen years younger as highly improbable, the District Court considered that the possibility that the charities would not take their bequests was so remote as to be negligible. Therefore, applying § 81.46 of Treasury Regulations 105 quoted in material part below,3 it denied the Government's motion for judgment on the pleadings and entered judgment for the plaintiff in the amount stipulated by the parties as correct in the event that the plaintiff was entitled to judgment.

The Government, placing heavy reliance upon certain statements in the opinion of the Supreme Court in the recent case of Commissioner of Internal Revenue v. Sternberger's Estate, 1955, 348 U.S. 187, 75 S. Ct. 229, 232, contends that the District Court misconstrued § 81.46 of Treasury Regulations 105, supra. Its argument in substance is that § 81.46 as construed by the Supreme Court in the Sternberger case permits a deduction on account of a bequest to charity only when as a practical matter there is no possibility whatever that charity will not eventually take the bequest. We do not agree for the reason that the pertinent section of the regulations does not say that the deduction must be disallowed unless there is no practical possibility that charity will not take; it says instead that 'no deduction is allowable unless the possibility that charity will not take is so remote as to be negligible,' and we do not think the Supreme Court in Sternberger intended to imply that the section does not mean what it says.

The question decided in Commissioner of Internal Revenue v. Sternberger's Estate, was whether, in determining a net estate for federal estate tax purposes, a deduction could be taken on account of a charitable bequest to take effect in the event that the decedent's childless 27-year-old daughter, a divorcee, should die without issue surviving her and her 62-year-old-mother. The Tax Court had answered this question in the affirmative. It found that the probability that charity would eventually enjoy the bequest could be computed by the application of established and accepted actuarial principles to known and reliable data, and, overruling the Commissioner, it held that the decedent's estate was entitled to an immediate tax deduction in the amount of the present value of the fraction of the bequest to charity which corresponded, actuarially, to the chance that charity would benefit. That is to say, it reduced the bequest to charity in proportion to the chance that charity would receive it, and then allowed a deduction in the amount of the present value of the bequest as so reduced computed on the basis of 4% interest compounded annually. The Court of Appeals for the Second Circuit affirmed the decision of the Tax Court per curiam, 207 F.2d 600, on the authority of its decision in Meierhof v. Higgins, 1942, 129 F.2d 1002.

The Supreme Court, however, reversed. In doing so it used some language tending out of context to support the Government's basic contention. But it did not rest its reversal on the broad proposition that charitable bequests which at the decedent's death are not assured but are in fact conditional are never deductible. It rested its reversal on the ground that the actuarial art was unable to cope with the contingency of that particular case. The Court called attention to the fact that the bequest it was considering 'offers to the daughter an inducement of about $2,000,000 to remarry and leave a descendant', and noted that such an inducement would render quite undependable the actuarially computed average probability that a 27-year-old woman would have issue. It then based its reversal on the authority of Humes v. United States, 1928, 276 U.S. 487, 48 S. Ct. 347, 72 L. Ed. 667, in which it was held that no charitable deduction could be taken under § 403(a)(3) of the Revenue Act of 1918 on account of a bequest to charity conditioned on the death of the decedent's 15-year-old niece without issue before reaching the age of 40, because there was no reliable statistical data from which the chance that the charity would take could be actuarially computed.

Neither the Humes nor the Sternberger case rules this one, for here the chance that charity will take does not depend upon the probability of anyone having issue, a matter involving an element of volition. In this case the chance that charity will benefit depends entirely upon the relative longevity of three persons, a matter unaffected by volition or personal inducement, for it would be preposterous to assume that either the daughter or the daughter-in-law, or both, would choose to commit suicide in order to defeat the decedent's bequest to charity. In the present situation, then, statistical data is not subject to distortion by any individual's self-interest, and since mortality tables are generally recognized as reliable and have long been used in the courts in tax and other cases, it follows that recognized principles of the actuarial art can safely be relied upon to cope with the contingency that charity will take in this particular case.

It does not follow, however, from the fact that in this case the chance that charity will take is actuarially calculable that a deduction is to be allowed in the amount of the present value of the fraction of the bequest which corresponds, actuarially, to the chance that charity will benefit, as the Tax Court ruled in the Sternberger case and the Court of Appeals for the Second Circuit held in Meierhof v. Higgins, supra. Off hand it would seem eminently fair and administratively simple to allow such a deduction in cases such as this where the chance that charity will take can be accurately computed actuarially. But the Supreme Court in the Sternberger case rejected the argument that the applicable regulations permit proportional deductions. The Court construed § 81.46 quoted in footnote 3 supra as taking not a proportional but an all or nothing approach to the problem of deductions on account of contingent bequests to charity. Thus the section either denies any deduction at all for a contingent bequest to charity, or else it permits the deduction of the present value of the entire contingent bequest, allowing the latter whenever 'the possibility that charity will not take is so remote as to be negligible.'

Since we hold that in this particular case the chance that charity will take can be accurately computed by the application of recognized principles of the actuarial art to generally accepted statistical data, and since it is stipulated that so computed the chance that charity will take is approximately eleven to one, we come to the question whether the chance that charity will not take is 'so remote as to be negligible.' The District Court thought that it was. We feel constrained to disagree.

In Newton Trust Co. v. Commissioner, 1 Cir., 1947, 160 F.2d 175, 179, an invasion of principal case in which we held that a power to invade for the general "use and benefit" of the decedent's widow afforded no standard 'fixed in fact and capable of being stated in definite terms of money', with the result that no charitable deduction could be taken under the rule of Ithaca Trust Co. v. United States, 1929, 279 U.S. 151, 49 S. Ct. 291, 73 L. Ed. 647, we had occasion in discussing an alternative argument advanced by the taxpayer to consider the question now under discussion.

In the Newton Trust Co. case at page 180 et seq. of 160 F.2d, we held, citing Robinette v. Helvering, 1943, 318 U.S. 184, 188, 63 S. Ct. 540, 87 L. Ed. 700, as we hold here, that the value of a charitable interest contingent upon survivorship could be computed by the application of recognized principles of the actuarial art to reliable data with sufficient accuracy for tax purposes. And we held that § 81.46 of Treasury Regulations 105 was valid and allowed deduction of the present value of a charitable bequest whenever the chance that charity would not take was so remote as to be negligible. Applying these propositions we held that the chance that one person would survive another of the same age, the chance upon which the bequest to charity depended in that case, being 50-50, was definitely not a chance so remote as to be negligible. We did indicate, however, that if the chance that charity would take depended upon a person aged 30 surviving a person aged 90, then it could safely be said that the chance that charity would not take would be so remote as to be negligible and the deduction would be allowable.

The line between those chances which are so remote as to be negligible and those which are not lies somewhere between these extremes. We cannot say exactly where. We can only decide specific cases as they arise using the best judgment we have in placing them on one side or the other of the line. And there is no standard to guide us except our estimate of the extent of the encouragement tax-wise which Congress wished to give testators to make gifts to charity. Our judgment being largely subjective, about all we can say is that we do not think that one chance in eleven can be considered so remote a chance as to be negligible, that is, a chance which persons generally would disregard as so highly improbable that it might be ignored with reasonable safety in undertaking a serious business transaction.

The judgment of the District Court is set aside and the case is remanded to that court for entry of a judgment for the defendant. No costs.