On February 13, 1955, a Sabena Belgian World Airways DC-6 crashed into the side of a mountain northeast of Rome, killing everyone aboard. This suit was brought against Sabena in the District Court for the Southern District of New York on behalf of the estate and children of one of the passengers, Paul LeRoy.
Because LeRoy was on an international flight, Sabena’s liability is governed by the Warsaw Convention, 49 Stat. 3000 (1934). Sabena therefore could avoid liability only by showing that “all necessary measures to avoid the damage” were taken. On the other hand, however, on the facts of this case, the plaintiff’s recovery is limited to $8,300 unless he establishes that Sabena was guilty of wilful misconduct.
Judge Murphy directed a verdict for the plaintiff for $8,300, but the question of wilful misconduct was submitted to a jury, which resolved it in the plaintiff’s favor. After a separate trial, without a jury, on the question of damages, Judge Murphy awarded $205,705. Both parties have appealed.
Sabena contends that the district court erred in admitting certain evidence and in its instructions to the jury. The plaintiff argues that it was error to take account of federal and state income taxes in estimating Paul LeRoy’s future net earnings and that the method used to discount the contributions which LeRoy would have made to his family to their present value was incorrect. We find no error in the rulings of the district court. However, as we find that the district court should have computed the award on somewhat different principles, we remand the case for recomputation of the award.
I. Wilful Misconduct
The Sabena plane crashed on the last leg of a Brussels-to-Rome flight; it should have been flying within a ten-mile wide airway between Florence and Rome. The plaintiff’s case begins with the fact that the site of the crash was over 30 miles east of the airway.
The plaintiff does not contend that the plane was off course as a result of wilful misconduct. Rather, he contends that the Sabena crew deliberately misled the Rome controller as to their position and that the controller therefore authorized a descent which, though it would have been safe within the airway, was fatal over the mountainous country to the east, where the plane was then flying.
The airway and crash site appear in approximate scale on the following diagram. The dashes mark the course taken by the plane under the plaintiff’s theory; the numbered circles indicate its hypothetical positions at one-minute intervals just before the crash.
*269
The plaintiff’s theory depends heavily on the range of the Viterbo radio beacon, which marked the airway, and on the transcript of radio conversations between the Sabena plane (flight ODB) and Rome. The last part of the transcript is the most important.
*270
*271The parenthetical note at 1848 was translated by the plaintiff’s translator as “Rome control is calling Swiss airplane 320 in order to have confirmation on the workings of radio beacon of Viterbo.” “Rumore indefinible,” at 1853, was translated as “undefinable noise.”
The plaintiff’s theory is that the noise at 1853 was produced when the plane’s external antenna struck trees on the mountainside. Since the plane had reported passing Florence, 114 nautical miles away, 24 minutes earlier and the speed of a DC-6 in descent is about 285 knots, the plane must have followed a straight line from Florence to the crash site, and its position from 1848 to 1853 must have been as indicated on the above diagram. This inference receives some support from the line of the swath cut by the plane as it crashed through the trees.
The plane’s hypothetical position at 1850, when its crew reported that it passed the .Viterbo beacon, is more than 30 miles from the beacon. The plaintiff’s witness testified, however, that the maximum range at which a radio compass would home on the beacon was only about 22 miles. It is the plaintiff’s contention that the Sabena crew falsely reported their position — since their report implied at least that they had obtained a radio compass bearing on the beacon1 — in order to avoid the delay which the Rome controller would have required if they had reported that they were uncertain of their position.
The plaintiff’s theory rests on deductions from indirect evidence, and Sabena offered contrary evidence on essential points, particularly as to the range of the Viterbo beacon. But the inferences required are not unreasonable, and it was for the jury to weigh out the conflicting evidence. Its finding must be accepted unless Sabena was prejudiced by the improper admission of evidence or improper instructions.
A. The Radio Transcript
Sabena contends that the radio transcript, designated as exhibit 15, should have been excluded on the ground that it was hearsay and no foundation had been laid for its admission. The transcript is several stages removed from the conversation which is alleged to have taken place on February 13, 1955. The conversation was recorded at the time; a transcript was made from the recording as part of the Italian government’s investigation of the crash and was included as an appendix to its report; exhibit 15 is a copy of that transcript. Analysis of its admissibility requires tracing back this chain:
First, if the appendix to the accident report would have been admissible, was exhibit 15 admissible? Second, if the original recording would have been admissible, would the appendix have been admissible? Third, would the original recording have been admissible? In addition, a further issue arises from the fact that Judge Murphy did not consider the exhibit’s admissibility de novo but based his ruling on a pre-trial ruling by Judge Bonsai that it was admissible.
The first of these questions is the most easily answered. The only objection relevant to this link in the chain is that the exhibit was not certified as required by Rule 44(a) of the Federal Rules of Civil Procedure. But Sabena conceded that there is “no discrepancy” between the exhibit and the appendix to the Italian government’s report. Certification could have done no more. Accordingly, exhibit 15 was admissible if the appendix would have been admissible.
Concessions by Sabena also go far toward answering the second question —whether the appendix would have been *272admissible. Sabena’s answers to pre-trial interrogatories establish that the transcript was indeed an appendix to an accident report and that the report was compiled by the Ministerio Difesa Aeronáu-tica of the Italian government.
That a report has been prepared by a government official or agency does not always require its admissibility. But in most cases in which such reports have been excluded, there has been reason to suspect bias, Hoffman v. Palmer, 129 F.2d 976, 991 (2 Cir. 1942), aff'd, 318 U.S. 109, 63 S.Ct. 477, 87 L.Ed. 645 (1943), or the report has contained conclusions or hearsay, Johnson v. Lutz, 253 N.Y. 124, 170 N.E. 517 (1930), or, in the case of Civil Aeronautics Board accident reports, exclusion has been required by statute, 49 U.S.C. § 1441(e). See generally 153 A.L.R. 163.
Whatever may be in the report as a whole, no conclusions are drawn in the appendix. Nor does Sabena claim that the Italian government was biased in this matter. To the extent that the transcript gives no information not obtainable from the recording, we therefore see no reason for excluding it if the recording itself would have been admissible.
Sabena argues that the transcript does go beyond the recording in certain respects. The most important of these are the notation of the times of transmission; the parenthetical notes, of which one appears at 1848 in the portion quoted above; and the identification of the parties who are transmitting. But the times could be obtained from the recording without difficulty, and the notations do little more than summarize the messages with which they are associated. It is of course possible that the method used to interpolate times was unreliable, but the speculative possibility of error or dishonesty on the part of the government agency in performing such a mechanical task is too minimal to give reason for excluding the document. See Banco de Espana v. Federal Reserve Bank, 114 F.2d 438, 446 (2 Cir. 1940).2
The identity of the parties does present a problem but not one relevant to this second question, where we assume that the recording would have been admissible. It is sufficient here to say that the identity of the Sabena flight is explicit and that of the controller is implicit in the text of the messages.
With respect to the recording itself, we think that it was admissible in view of the provisions of 28 U.S.C. § 1732:
“In any court of the United States and in any court established by Act of Congress, any writing or record, whether in the form of an entry in a book or otherwise, made as a memorandum or record of any act, transaction, occurrence, or event, shall be admissible as evidence of such act, transaction, occurrence, or event, if made in regular course of any .business, and if it was the regular course of such business' to make such memorandum or record at the time of such act, transaction, occurrence, or event or within a reasonable time thereafter.
“All other circumstances of the making of such writing or record, including lack of personal knowledge by the entrant or maker, may be shown to affect its weight, but such circumstances shall not affect its admissibility.
“The term ‘business,’ as used in this section, includes business, profession, occupation, and calling of every kind.”
The archetype of the records to which § 1732 was intended to apply is the typical business accounting record, a routine factual entry normally made without reference to anticipated litigation. Because *273such entries are likely to be accurate and, perhaps more important, because modern business organization makes it impractical to produce in court all the persons directly responsible for the entry, § 1732 and similar state statutes were enacted. See Judge Frank’s extensive discussion of such legislation in Hoffman v. Palmer, 129 F.2d at 983 et seq.
The statute’s rationale is equally applicable to the recording here involved, which was a simultaneous recording made as part of a regular air control procedure. From the deposition and testimony of Captain Pierre Dils of Sabena and the undisputed testimony of the plaintiff’s expert, Captain Richard Hettenbaugh, it appears that major airports were required to keep either a written or a recorded telecommunications log. The latter which automatically records transmissions to and from the controller, was in use in Italy in 1955.
The recording thus was at least the equivalent of a regular written journal kept by the Rome controller 3 *and was a contemporaneous business record. That is all that § 1732 purports to require. Despite the section’s express reference to hearsay, however, courts have been cautious about allowing admissions of business records where the information was obtained only through hearsay. See, e. g., Standard Oil Co. v. Moore, 251 F.2d 188, 212 (9 Cir. 1957), cert. denied, 356 U.S. 975, 78 S.Ct. 1139, 2 L.Ed.2d 1148 (1958); United States v. Grayson, 166 F.2d 863, 869 (2 Cir. 1948) (L. Hand, J.).
The leading case in thus limiting the admission of business records is Johnson v. Lutz, 253 N.Y. 124, 170 N.E. 517 (1930),4 in which the report of a police officer was held inadmissible because it was based in large part on the volunteered statements of bystanders. But the statements involved in this case are not the gratuitous reports of persons not involved in the business. If the plane crews were not required to make such reports, it certainly was at least the regular business practice for them to do so. To such reports Johnson v. Lutz does not apply. See Pekelis v. Transcontinental & Western Air, Inc., 187 F.2d 122, 130-131, 23 A.L.R.2d 1349 (2 Cir.), cert. denied, 341 U.S. 951, 71 S.Ct. 1020, 95 L.Ed. 1374 (1951); United States v. Grayson, supra (dicta); Note, 48 Colum.L.Rev. 920, 926 (1948). Such reports should be as reliable as the typical business record,5 and they surely are just the kind of hearsay which, in the words of the statute,- “may be shown to affect [the record’s] weight, but * * * shall not affect its admissibility.”
Moreover, even if the principle of Johnson v. Lutz is assumed to apply here, the objection based on it can be met on its own terms. Since the objection applies only to “multiple hearsay,” it does not apply to the “rumore indefinible” or to the interpolation of times. Nor does it *274apply to the statements of the Rome controller, including the identification of their source that is implicit in them; the recording is at least as satisfactory as a written record kept by the controller, and his statements and identification should be no less admissible for having been tape recorded rather than written down.
The transmissions from the Swissair flight, on the other hand, are hearsay, but it is difficult to see how Sabena was prejudiced by their admission. They tend to establish, if anything, that the Viterbo beacon was operating. Since one of Sabena’s principal contentions to the jury was that its plane could have picked up the Viterbo beacon at 1850, it could only have been to its advantage to establish that the beacon was operating.
The transmissions from the Sabena plane also are hearsay, but if we can assume that they did in fact come from that plane, they fall within the exception for admissions of an opposing party and therefore do not render the record inadmissible. Missouri Pacific RR v. Austin, 292 F.2d 415, 421 (5 Cir. 1961). There was a tentative identification by Captain Dils, who testified that he had listened to the recording and was “practically sure” that he recognized the voice of the particular crew member. In addition, the context of the transmissions makes it highly unlikely, to say the least, that a transmission purportedly coming from the Sabena plane did not in fact do so. We think that the source of the transmissions was sufficiently identified.
The final issue with respect to admission of exhibit 15 arises from the fact that Judge Murphy appears to have admitted it solely on the basis of Judge Bonsai’s pre-trial ruling. Whether that was a sound basis for admitting the exhibit is a nice question. It is of course customary for a district judge to follow an earlier ruling by one of his brothers in the same litigation, though he is not bound to do so. Dictograph Products Co. v. Sonotone Corp., 230 F.2d 131 (2 Cir.), rehearing denied, 231 F.2d 867, petition for cert, dismissed, 352 U.S. 883, 77 S.Ct. 104, 1 L.Ed.2d 82 (1956). But despite Professor Moore’s flat statement that admissibility may be ruled on at pre-trial, 3 Moore, Federal Practice j[ 16.16, at p. 1122 (1964), the efficacy of such rulings necessarily is limited by the first judge’s inability to foresee the circumstances of the trial which may be relevant to admissibility. Cf. Fidelity and Casualty Co. v. Frank, 227 F.Supp. 948 (D. Conn. 1964).
We need not resolve that issue here, however, for this is not a case in which admissibility was a matter within the district court’s discretion. We conclude that in this case § 1732 required admission of exhibit 15.
The district court’s discretion with respect to § 1732 is a discretion in judging whether the document offered “has an inherent probability of trustworthiness.” Central RR v. Jules S. Sottnek Co., 258 F.2d 85, 88 (2 Cir. 1958). It is therefore a necessary premise for its exercise that the document’s trustworthiness be in doubt. See Shenker v. United States, 322 F.2d 622, 627 (2 Cir. 1963), cert. denied, 376 U.S. 907, 84 S.Ct. 659, 11 L.Ed.2d 606 (1964) (entry made by employee of party); Puggioni v. Luckenbach Steamship Co., 286 F.2d 340 (2 Cir. 1961) (same); Central RR v. Jules S. Sottnek Co., supra (acci dent report based on bystanders’ volunteered statements and containing conclusions as to the accident’s cause). As we have attempted to demonstrate, however, nothing in the nature or content of exhibit 15 gives any reason to doubt its trustworthiness and Sabena in all of its argument has suggested no extrinsic circumstances which would raise such doubt. To exclude such a document as exhibit 15 on the mere speculative possibility of error of fraud would frustrate the evident purpose of the statute.
We therefore conclude that there was no error in receiving exhibit 15.
*275B. Sabena’s Other Objections
Of Sabena’s other objections, only two merit discussion. The first is directed at exhibit 7, a pair of United States Air Force maps which together show the Italian peninsula from Florence to Rome. The Florence-Rome airway and the Viter-bo beacon appear on the maps, and Captain Hettenbaugh added other markings to illustrate the plaintiff’s theory. These consist of a line and numbered circles, as in the above diagram, which indicate the plane’s hypothetical course and pre-crash positions; there also are concentric circles around the Viterbo beacon which mark ranges of 10,15 and 20 miles.
Sabena objected to admission of the exhibit on the grounds that its planes did not use such maps; that the maps’ accuracy was not established; more particularly, that they do not show all of the radio facilities in the area; and that the markings were taken only subject to connection and were never connected.
Since the charts were used only to illustrate distances and positions, the only question concerns their accuracy, and Captain Hettenbaugh testified as to that; it is wholly irrelevant whether Sabena ever used such a map. Captain Hetten-baugh also explained his markings, which are amply connected by exhibit 15 and by testimony as to the speed of a DC-6 and the Viterbo beacon’s range.
There are some discrepancies between exhibit 7 and another map, a radio chart used by Sabena, as to radio facilities existing in the area. However, Sabena made no reference to the discrepancies when it objected to admission of exhibit 7, and it introduced no evidence relating to them after the exhibit was admitted. Moreover, it was for the jury to decide which of the maps was in error. In sum, Sabena has not established any error in the admission of exhibit 7, much less that there was so substantial an error as to require reversal.
Sabena’s final claim of error is that the district court erred in refusing to charge the jury that, “because of the human instinct of self-preservation and the disposition of men to avoid personal harm,” the crew of the Sabena plane “are presumed to have acted with diligence and due care.” This is not an unreasonable presumption, and it is supported by a number of cases. But many, perhaps most, of these cases involve use of the presumption by an appellate court in determining whether a jury verdict is supported by the evidence. E. g., Eastern Air Lines, Inc. v. Union Trust Co., 95 U.S.App.D.C. 189, 221 F.2d 62, rev’d in part per curiam, 350 U.S. 907, 962, 76 S.Ct. 192, 100 L.Ed. 796 (1955); Atchi-son, T. & S. F. Ry. v. Toops, 281 U.S. 351, 356, 50 S.Ct. 281, 74 L.Ed. 896 (1930). Whether Judge Murphy was required to include such common sense in his charge to the jury is another matter. We do not think that he was.
II. Damages
Judge Murphy concluded that New York would apply the law of the place of ; the tort, Italy, to the question of damages. He did so with some reservations in view of recent New York conflict of laws decisions. See Babcock v. Jackson, 12 N.Y.2d 473, 240 N.Y.S.2d 743, 191 N.E.2d 279, 95 A.L.R.2d 1 (1963). Since both parties presented their cases below on the theory that Italian law governs and neither has raised the issue here, we see no reason to review Judge Murphy’s conclusion.
Judge Murphy arrived at the $205,705 award as follows: He found that Paul LeRoy would have earned some $600,000 before retirement in 1993 at age 65. He deducted from this amount 15 per cent for United States and New York income taxes (LeRoy worked in New York) and reduced the balance by 25 per cent to allow for LeRoy’s personal expenses. The resulting figure — the estimated contribution that LeRoy would have made to his family — was $382,000.
To reduce this to present value, Judge Murphy averaged the contributions over 37 years and rounded the result to $10,000 per year. The $205,705 award is the present value of annual payments of $10,000 for 37 years.
*276Although neither party has raised the point, we think it necessary to consistent discussion that a minor arithmetical error in the district court’s computations be corrected. The number of years from 1955 to 1993 is not 37 but 38, and for purposes of this discussion we will modify the plaintiff’s contentions accordingly. Because this error arose only after the total contribution was computed, it affected only the averaging and discounting.
The plaintiff’s first contention is that it was error to deduct estimated federal and state income taxes from Le-Roy’s future earnings. It is the general rule in this Circuit that, where federal law controls or applicable state law is silent, income taxes should not be considered in estimating future net income. The rule applies to both jury and non-jury trials. McWeeney v. New York, N. H. & H. R. R., 282 F.2d 34 (2 Cir.), cert. denied, 364 U.S. 870, 81 S.Ct. 115, 5 L.Ed.2d 216 (1960) (jury trial); Cunningham v. Rederiet Vindeggen, 333 F.2d 309 (2 Cir. 1964) (non-jury trial).
The defendant has not argued that there is an Italian rule in point, and, in any event, there is nothing in the record to support such an argument. Although each party’s Italian law expert testified at some length as to the factors considered in assessing damages, neither made any mention of the effect of income taxes. There was testimony that damage awards under Italian law are intended to compensate for “actual” loss. But this nearly universal principle of damages was expressly held in Cunningham not to be a sufficient indication that income taxes are taken into consideration. 333 F.2d at 313-314.
We- therefore conclude that McWeeney controls. That case did not lay down a flat rule, however. It left open the possibility that adjustment for income taxes would be proper if the income were so high that the award would otherwise be excessive, 282 F.2d at 38, and we subsequently affirmed the existence of this exception in dicta in Montellier v. United States, 315 F.2d 180, 186 (2 Cir. 1963). Although in Montellier we upheld the district court’s refusal to deduct estimated income taxes, we indicated that it could properly have done so on the facts of that case and also that it would have been error not to do so if a sufficiently high income had been at issue.
Whether or hot the district court was required to allow for the effect of income taxes in this case, we think that it was at least a proper exercise of its discretion to do so. The estimated future income on which the award was based averaged $16,000 and reached $25,000 for nine years. This compares with an estimated income of $11,150 in Montellier, D.C., 202 F.Supp. 384, 422, and it contrasts even more sharply with McWeeney ($4,800) and Cunningham ($6,281 — see 60 Adm. 247, S.D.N.Y., April 13, 1962). Certainly the risk that the federal and New York governments will cease to take a substantial portion of a $16,000 income is one of the smaller uncertainties involved in the computations in this case, and the 15 per cent figure adopted by Judge Murphy is a reasonable estimate of what that portion would probably be.
The plaintiff’s second contention is that the district court’s method of discounting the award to present value was not proper and resulted in too low an award. The plaintiff has not fully developed his position in his brief, but it appears to be substantially as follows: Judge Murphy averaged LeRoy’s contributions over the period 1955-1993 (though he used a figure of 37 rather than 38 years), and he discounted the contributions to their 1955 value. The plaintiff’s approach, on the other hand, would take account of the fact that 9 of the 38 years had passed by the time judgment was rendered. In his view, the $382,000 should be averaged over only the remaining 29 years, and the resulting average annual contribution, approximately $13,000, should be discounted only back to 1964. This would give an award of about $234,000.
*277We agree with the plaintiff’s premise that the 88 year period should be treated as having begun in 1955, but we cannot accept the computations which he makes on this basis. It is not that the plaintiff is too generous with himself. On the contrary, this is one of those rare instances in which our view of the award is more favorable to the plaintiff than is the plaintiff’s own.6
The award in this case is intended to be the financial equivalent of the contributions which LeRoy would have made to his family. It should not simply be the sum of these contributions, however. The award may be regarded as having been paid at the time of judgment, 1964,7 while it is assumed that the contributions would have been made from 1955 to 1993. As a matter of economics, therefore, each contribution which would have been made prior to the judgment should be increased by the allowance of prejudgment interest from the date the contribution would have been made, while those which would have been made subsequent to the judgment should be discounted back to the date of judgment.
It is not the law of economics that we are applying, however, but the law of Italy. Judge Murphy found that pre-judgment interest is allowed under Italian law “only when equity would so demand,” and he held that it should not be awarded in this case. The plaintiff has not challenged this holding, and, in any event, we cannot say that Judge Murphy’s finding as to the application of Italian law is clearly erroneous. The contributions for 1955 to 1964 therefore should enter into the award only at their face value.
In Judge Murphy’s calculations, however, these contributions were not merely left at face value; all of the contributions — from 1955 to 1993 — were discounted back to 1955. Since no prejudgment interest was allowed, this had the effect of valuing contributions for 1964, for example, at only about seventy-five cents on the dollar, even though they would have been received at approximately the same time as the award.
If Italian law required such a result, we would of course be bound to accept it. But since there has been no suggestion that it does, we think that it was error to ignore economics and the approach of recent cases. See Conte v. Flota Mercante Del Estado, 277 F.2d 664, 670 (2 Cir. 1960); Montellier v. United States, 202 F.Supp. 384, 424 (E.D.N.Y. 1962), aff’d without discussion of this point, 315 F.2d 180 (2 Cir. 1963); Tweedy v. Esso Standard Oil Co., 190 F.Supp. 437, 442 (S.D.N.Y.1960) (dictum), aff’d per curiam, 290 F.2d 921 (2 Cir. 1961). Contributions which would have been made prior to the date of judgment should enter into the judgment at face value; contributions which would have been made subsequent to the date of judgment should be discounted but only back to the date of judgment.
Judge Murphy discounted the contributions at a rate of 3% per cent. We have held, however, that the discount rate should be at least 4 per cent. Alexander v. Nash-Kelvinator Corp., 271 F.2d 524, 527 (2 Cir. 1959). Since there was no evidence that Italian courts would apply a lower rate — indeed, the plaintiff’s expert testified that Italian courts use a 5 per cent rate for pre-judgment interest — we think that a rate of not less than 4 per cent must be applied to the future contributions in this case. Cf. Conte v. Flota Mercante Del Estado, 277 F.2d 664, 670 (2 Cir. 1960).
The final point relates to the averaging of the contributions which LeRoy would have made over the 38 years. The point was not raised by the parties, but because it is to some extent intertwined *278with the preceding point and should not require introduction of additional evidence on remand, we proceed to consider it.
Since averaging transfers money from the years of high contributions to the years of low contributions, it would have no effect on the award if all years were treated alike. But it is the purpose of discounting to present value to treat different years differently; discounted at the 3% per cent rate used by Judge Murphy, a dollar which would have been contributed at the time of the judgment is worth a dollar, one which would have been contributed 10 years later is worth about 71 cents, and one which would have been contributed 29 years later is worth only about 37 cents. See Am.Jur. 2d Desk Book, Doc,No.l32 (1962).
The result is that the transfer of money backward or forward over the years through averaging increases or decreases the award. The change may not be significant where the term of years is short or the difference between the maximum and minimum contributions is not great or where the high and low years are scattered more or less at random over the term. None of these describe the instant case, however.
Judge Murphy made the following projection of LeRoy’s income:
Averaging the contributions which Le-Roy would have made out of these projected earnings shifted many thousands of dollars from the later years to the earlier years and thus substantially affected the award.
We do not suggest that it is necessary separately to discount the contribution for each year. Contributions for the years prior to the judgment may, as discussed earlier, simply be totalled. Contributions for the subsequent years may be averaged within the income steps indicated in the above projection; it is necessary only that the average not be over so long a term or include so great a range of contributions as seriously to distort the result.
Most of the factual assumptions necessary for recalculation of the award already appear in Judge Murphy’s opinion. However, as there is some uncertainty as to whether Judge Murphy would have made the same allowance for income taxes and personal expenses for all of the years if he had been dealing separately with the years of lower and higher income, we remand for recomputation of the award in the light of this opinion.
To summarize, the award should consist of two parts so far as computation is concerned — one for the years prior to judgment and a second for the years subsequent to judgment. The first is simply the sum of LeRoy’s projected earnings for the pre-judgment years reduced by what Judge Murphy finds to be a proper allowance for income taxes and personal expenses. The second part begins with the projected earnings for the post-judgment years, which likewise must be adjusted for taxes and personal expenses in order to compute the contributions which would have been made. These contributions must then be discounted at 4 per cent to their 1964 value; this discount may be based on the average contributions within each of the income steps projected by Judge Murphy, but averaging the contributions over the full 29 years would not be proper in view of the differences in income over those years.
Accordingly, we remand for recom-putation of the award.
. The beacon’s identifying signal, which consisted of a steady tone interrupted at intervals by a code, apparently could be picked up at a substantially greater distance than that at which a radio compass would home on the beacon, but the jury could properly have found that the position report did carry the implication that the crew had obtained a radio compass bearing and not merely the signal.
. This is in some ways a stronger case for admissibility than existed in Banco de España. The defendant in that case apparently had not had access to the original government record, but Sabena concedes the accuracy of the copy from the appendix to the accident report and Captain Pierre Dils, then chief pilot of Sa-bena’s European Division, heard the original recording.
. Cf. State v. Meyer, 37 Wash.2d 759, 226 P.2d 204 (1951), in which the transcript from a recording of an interview made in a hospital by an attending doctor was admitted under the state business records statute as part of the hospital clinical record. For some purposes a recording might be equated with the testimony under oath of a third-party auditor of the conversations rather than merely a written record. See 2 Wigmore, Evidence § 669, at p. 790 n. 4 (1940); cases cited in Annotation, 58 A.L.R.2d 1024, 1041. But cf. United States v. McKeever, 271 F.2d 669, 676 (2 Cir. 1959). But the distinction is irrelevant to this discussion since the only substantial objection to admission of the recording is that the recorded statements are themselves hearsay. That objection is not met by equating the recording with the testimony of a sworn witness. Cf. People v. Buckowski, 37 Cal.2d 629, 233 P.2d 912 (1951), cert. denied, 343 U.S. 928 (1952) (recording admitted to show fact that statement was made); Paulson v. Scott, 260 Wis. 141, 50 N.W.2d 376, 31 A.L.R.2d 706 (1951) (recording admitted under exception for admissions of an adverse party).
. The New York provision interpreted in Johnson v. Lutz, quoted in 253 N.Y. at 126, 170 N.E. 517, is the predecessor of C.P.L.R. Rule 4518(a) and is substantially similar to § 1732.
. We do not think that the plaintiff’s challenge to the truth of one such statement — the Sabena plane’s position report at 1851 — undercuts the argument for their general reliability.
. Because we reject the plaintiff’s contention with respect to consideration of the federal and state income tax, however, the recovery to which wc find the plaintiff to be entitled is less than the amount which he seeks on this appeal.
. Payment of the judgment may be delayed, for example, pending appeal, but the addition of post-judgment interest makes the ultimate payment economically equivalent to payment at the time of judgment.