(concurring dubitante):
This case is another example of a federal court’s being compelled by the Congressional grant of diversity jurisdiction to determine a novel and important question of state law on which state decisions do not shed even a glimmer of light.1 The question here, how far awards of damages for disabling personal injury or for death shall attempt to make allowance for future inflation, is of great concern to the states since awards like that made here will further escalate the heavily mounting burden of liability insurance costs. The state decisions and the federal cases endeavoring to ascertain state law are in a stage of uncertainty and flux. So too are the decisions with respect to federal law. Compare Sleeman v. Chesapeake & Ohio Railway Co., 414 F.2d 305 (6 Cir. 1969), with Bach v. Penn Central Transportation Co., 502 F.2d 1117, 1122 (6 Cir. 1974). In a case of federal law, the Fifth Circuit recently granted en banc consideration and by a vote of twelve to three expressly disapproved a district court’s effort, in computing lost future earnings, to take account of possible inflationary trends over a period of several decades on the ground that “the influence on future damages of possible inflation or deflation is too speculative a matter for judicial determination.” Johnson v. Penrod Drilling Co., 510 F.2d 234, 236, 241 (5 Cir. 1975) (en banc), petition for certiorari filed, 43 U.S.L.W. 3684.
Both plaintiff’s expert and the court allowed for inflation not by building cost-of-living increases into future earnings but by applying a rate of only 1.5% in discounting to present value estimated *391lost future earnings and other recoverable values calculated in 1971 dollars. The district court derived this 1.5% figure by comparing rates of return on a number of “risk-free” securities issued by the federal government (plaintiff’s expert examined other types of “risk-free” fixed income investments and reached similar conclusions) since 1940 with rates of inflation during the same period as reflected by annual changes in the Consumer Price Index. The court subtracted the latter from the former on the theory that the latter amounted to that portion of the return representing what investors have historically demanded as protection against inflation. The difference varied from year to year, but the court determined that 1.5% was a representative figure for the period. This was deemed to be “that part of the annual yield which constitutes payment for the use of capital” or “real yield”— presumably the rate of return which investors would be willing to accept in an inflation-free economy; while the rate of inflation might rise and fall, investors could be expected to demand about 1.5% return on safe investments in addition to protection against expected inflation. The court recognized that other courts have been reluctant to take explicit account of the effects of future inflation (see 510 F.2d at 236 n.l) but stated that its approach in fact was “a means to avoid undue speculation” with respect to future inflation and even suggested that when, as in this case, the effects of future inflation have been expressly excluded in the calculation of the amount to be discounted, the “appropriate rate of discount” must necessarily be adjusted so that “the additional interest demanded by the investment market as compensation for investors’ assumption of the risk of inflation” is excluded. The distinction drawn between this method and the one more commonly used — adjusting the amount to be discounted so as to include a sum reflecting assumptions about future inflation — apparently given some weight by the majority — is more apparent than real. Plaintiff’s expert acknowledged that “another approach” or “alternative calculation” for this problem would be to increase estimated lost future compensation and living expenses to take account of the effects of future inflation and not reduce the rate of return used for discounting by any amount reflecting inflation. The outcome of the calculation under either approach would be very nearly identical. Indeed, at one point counsel for plaintiff asked his expert to calculate the present value in 1971 dollars of the deceased’s lost earnings based upon the “speculative” assumption of an inflation rate of 4.5% and a rate of return of 6%. This approach would have reduced the recovery by less than $2,000 out of approximately a quarter of a million dollars.
In any event, plaintiff’s expert came up with a $253,424 present value of Mrs. Feldman’s projected earnings. Using higher starting and ending salary figures and a different percentage deduction for income taxes, the judge arrived at an initial sum of $499,953 to which he added $100,000, admittedly drawn from the atmosphere, “for the destruction of the decedent’s capacity to enjoy life’s activities”, an element recognized as appropriate for consideration by Connecticut law, and from which he subtracted $155,897 as the discounted sum of personal living expenses, yielding a total recovery of $444,056. On the judge’s computations, Mrs. Feldman, who had been earning $10,000 a year at the time of her death in 1971, would be earning $33,757 in 1971 dollars in 2011 as a “legislative analyst” for the National League of Cities and United States Conference of Mayors (NLC/USCM), when she would have attained the age of 65. However, as counsel for Allegheny points out, without dispute from counsel for Mr. Feldman and apparently based upon the testimony of plaintiff’s expert,, about an alternative method of calculation discussed above, a calculation deducting 4.5% for inflation from a 6% interest rate assumed to be attainable on an investment free from risks other than inflation implicitly car*392ríes the prediction that Mrs. Feldman, and also all federal employees in the GS 16 — 7 category (which Mrs. Feldman hypothetically would reach after 40 years under the scheme for predicting merit pay increases adopted by the court), would in fact be earning $122,823 in the year 2011. Similar calculations based upon maintaining the 1.5% differential could yield even more striking results, which are largely veiled by the court’s approach. One point that immediately occurs is why, if Mrs. Feldman’s salary would rise to such a figure, income tax, deductible from damages under Connecticut law, should be computed at only 25%, a rate which the court found would achieve “substantial justice.” It is common knowledge that one effect of inflation is that the same progressive rates of income tax take an ever larger bite out of real income, and it is unlikely in the last degree that, in an era of increasing budgets, due in considerable part to inflation, Congress would make the accommodation needed to prevent this.
Save for this important point not urged by Allegheny and the two corrections made by the majority, I have no reason to question the meticulous calculations of the able district judge. Indeed one could argue that, at a time when the national goal is simply to bring back the golden age of single rather than double digit inflation, without too much question what the single digit should be, the entire interest return on otherwise risk-free investments, today probably in excess of 6%, represents compensation against the risk of inflation; in other words, investors in fixed income securities are willing, for the time being, to forego any return if they can keep the real amount of their investment intact. Indeed, insofar as the return is subject to income tax, they are not even achieving that. Yet common sense suggests that investors will not tolerate such a situation indefinitely.
I doubt whether judges, or anyone else, can peer so far into the future; the district court’s computations suffer from what Mr. Justice Holmes, in another context, called “[t]he dangers of a delusive exactness,” Truax v. Corrigan, 257 U.S. 312, 342, 42 S.Ct. 124, 133, 66 L.Ed. 254 (1921) (dissenting opinion). Instead of recognizing the plethora of uncertainties as the Fifth Circuit has done, see Johnson v. Penrod Drilling Co., supra, 510 F.2d at 236, compare Frankel v. United States, 321 F.Supp. 1331, 1346 (E.D.Pa.1970), aff’d, 466 F.2d 1226 (3 Cir. 1972), the court below endeavored to construct an iron-clad guaranty against the unknown and unknowable future effects of inflation. The estate of a young woman without dependents is hardly an outstanding candidate for a forty-year protection against inflation not enjoyed at all by millions of Americans who depend on pensions or investment income and not fully enjoyed by millions more whose salaries have in no wise kept pace with inflation.
The court necessarily assumed not only continued inflation, which unhappily seems likely in some degree, but continued responsiveness to it by equivalent wage increases. Yet we have seen in recent months that employers, particularly municipalities, simply cannot maintain these. Thousands of New York City’s employees have been dismissed and the rest are being subjected to a wage freeze. Other important cities may not be far behind in having to resort to similar measures. Under such conditions can we be sure that NLC/USCM would continue to grant automatic cost-of-living pay increases for 40 years, as the court assumed? Perhaps so, since the main business of NLC/USCM is seeking to obtain federal funds for cities, which surely is a boom industry if any there be; but perhaps not.
I would also question the likelihood— indeed, the certainty as found by the court — that, despite her ability, determination and apparent good health, Mrs. Feldman would have worked full time for forty years until attaining age 65, except for the eight years she was expected to devote to the bearing and early rearing of two children. Apart from *393the danger of disabling illness, temporary or permanent, there would be many attractions to which the wife of a successful lawyer might yield: devoting herself to various types of community service, badly needed but unpaid, or to political activity; accompanying her husband on business trips — often these days to far-off foreign countries; making pleasure trips for periods and at times of the year inconsistent with the demands of her job; perhaps, as the years went on, simply taking time off for reflection and enjoyment. Granted that in an increasing number of professional households both spouses work full time until retirement age, in more they do not. Surely some discount can and should be applied to the recovery for these reasons.
My guess is also that, even if inflation should be taken into account, neither a Connecticut nor a federal jury would have made an award as large as was made here. I say this despite the $369,-400 jury verdict for another death arising out of the same crash which we sustained in Perry v. Allegheny Airlines, Inc., supra, 489 F.2d 1349, where we did not expressly discuss the inflation question. Even though the existence of dependents is legally irrelevant under the Connecticut survival statute, a jury would hardly have ignored that, whereas Perry was survived by a dependent wife and five children ranging from 6 to 14 years in age, Mrs. Feldman had no dependents. More significant to me is that in Perry’s case the jury awarded only $369,400 as against the $535,000 estimate of Mrs. Perry’s expert for economic loss alone; here the judge was more generous in important respects than plaintiff’s expert.
However, I am loathe to require a busy federal judge to spend still more time on this diversity case, especially when I do not know what instructions to give him about Connecticut law. Some of the questions I have raised are not open for exploitation by the defendant since its own expert made his calculations on the basis that Mrs. Feldman would work until age 65. Although intuition tells me that the Supreme Court of Connecticut would not sustain the award made here, I cannot prove it. I therefore go along with the majority, although with the gravest doubts. I do this on the basis that, as far as I am concerned, the decision will not constitute a precedent on the inflation problem in a case arising under federal law. Judgments like Mr. Feldman’s and Mrs. Perry’s also inevitably raise serious policy questions with respect to damages in airline accident cases beyond those here considered, but these are for Congress and not for the courts.
. In recent years we have had many cases in which we have been obliged largely to improvise state law. See, e. g., Phillips, Nizer, Benjamin, Krim & Ballon v. Rosenstiel, 490 F.2d 509 (2 Cir. 1973); East Hampton Dewitt Corp. v. State Farm Mutual Automobile Ins. Co., 490 F.2d 1234 (2 Cir. 1973); Modave v. Long Island Jewish Medical Center, 501 F.2d 1065 (2 Cir. 1974). Despite the sincere respect that I have for Judge Blumenfeld and his long years of experience as a Connecticut practitioner and federal district judge, I do not believe that he, or anyone else, can make any reliable prediction what the Supreme Court of Connecticut would do with the exceedingly difficult question here discussed.