Fehrs v. United States

PER CURIAM: This case comes before the court on defendant’s motion, filed January 30, 1980, requesting that the court adopt, as the basis for its judgment in this case, the recommended decision of Trial Judge John P. Wiese, filed November 27, 1979, pursuant to Rule 134(h), no intention to except or exceptions thereto having been filed by the parties and the time for so filing pursuant to the Rules of the court having expired. Upon consideration thereof, without oral argument, since the court agrees with the trial judge’s recommended decision, as hereinafter set forth,* it hereby grants defendant’s said motion and affirms and adopts the recommended decision as the basis for its judgment in this case. Therefore, it is concluded that, in accordance with the trial judge’s recommended decision, plaintiffs are entitled to a partial refund of gift taxes paid, together with interest as provided by law and judgment is entered for plaintiffs to that extent with the amount of their recovery to be determined through further proceedings pursuant to Rule 131(c). In all other respects plaintiffs’ petitions are dismissed. OPINION OF TRIAL JUDGE ** WIESE, Trial Judge: In an earlier tax refund suit involving the present plaintiffs, this court held that the transfer of all of their shares of Fehrs Rental Corporation (a "brother” corporation) to a newly-created Fehrs Finance Company (a "sister” corporation 100 percent owned by the plaintiffs’ daughters) in return for perpetual annuities of a fixed yearly value resulted in a redemption that qualified — 493 — as an "exchange” transaction pursuant to sections 302(b)(3) and 302(c)(2) of the Internal Revenue Code.1 Accordingly, the plaintiffs were deemed entitled to treat as capital gain, rather than as ordinary income, the amounts received pursuant to their annuities. Fehrs v. United States, 214 Ct. Cl. 74, 556 F.2d 1019 (1977). In this consolidated action the same background facts are met once again. The question now, however, is whether the plaintiffs’ transfer of their shares in return for the annuities represented an exchange of economic equivalents or was, as the Government contends, in part a gift by them to the shareholders of the Fehrs Finance Company to the extent of a claimed excess in the market value of the transferred shares. On the basis of the accompanying findings of fact and for the reasons discussed in this opinion, the court concludes that the transaction was in part a gift. I. Background Facts Plaintiffs, husband and wife, were the owners of all 1,380 outstanding shares of the Fehrs Rental Corporation (Rental), a company engaged primarily in the renting, leasing, selling and repairing of heavy road and industrial equipment. On December 21, 1964, plaintiff, Edward J. Fehrs (since deceased2), acting on the basis of tax planning advice given by counsel, implemented the first of a series of steps whose two-fold objectives were: (i) to allow Mr. Fehrs to retire upon a satisfactory income while leaving Rental "in the family” and (ii) to create adequate working capital for a soon-to-be-created Fehrs Finance Company (Finance). Thus, on the indicated date, December 21, 1964, Mr. Fehrs gave 130 shares of Rental stock to his wife and children; subsequently, on January 11, 1965, he gave them an additional 111 shares. As a result of these initial steps, Mr. Fehrs was left owning 982 shares of Rental stock, his wife’s holdings had been increased to 177 shares, and 221 — 494 — shares were now owned by his two daughters and their respective families. In his United States gift tax return for the calendar years 1964 and 1965, Mr. Fehrs valued his gifts of stock at $600 per share. Approximately 2 months later, February 28, 1965, the plaintiffs’ daughters formed the Fehrs Finance Company for the purpose of purchasing the commerical paper generated from the sales of construction equipment by Rental. They, the daughters, were the sole shareholders of Fehrs Finance Company. The following day, March 1, 1965, the final steps of the plan were carried out: a transfer to Finance by Mr. and Mrs. Fehrs of all their remaining shares of Rental in return for lifetime annuities and a reacquisition, by Rental, of nearly all of its stock from Finance in return for cash and an unsecured promissory note. Described in greater detail, the March 1, 1965, transfers involved the following: In return for the remaining 982 shares that it acquired from Mr. Fehrs, Finance promised to pay him $62,000 per year in equal quarterly installments beginning January 1, 1966, and continuing for the remainder of his life; in return for the 177 shares it acquired from Mrs. Fehrs, Finance promised to pay her $8,000 per year— again, in equal quarterly installments beginning January 1, 1966, and also continuing for the remainder of her life. The two annuity agreements were substantially alike. They provided that, upon default in any payment that continued for 30 days, the entire commuted value of the annuity would be immediately due and payable, such commuted value to be computed in accordance with standard actuarial tables and values taking into account the age of the annuitant as of the date of the default. The agreements permitted the annuitants to waive any default; they also provided that neither the agreements nor the payments to be made under them could be negotiated, assigned or anticipated by the annuitants and that Finance would have no obligation to recognize or give effect to any attempted transfer of such sort. On its books and records, and in its corporate income tax return for the taxable year ending November 30, 1965, Finance recorded its liability to Mr. and Mrs. Fehrs, as of — 495 — March 1, 1965, as being in the amount of $725,000 (Mr. Fehrs had been born on December 5, 1894 and Mrs. Fehrs was born on July 9, 1900. Hence, their respective ages as of March 1,1965, were 70 years and 64 years). As to the other transfer of March 1, 1965 — that between Finance and Rental — this was carried out pursuant to a formalized redemption agreement under the terms of which Finance transferred to Rental the 1,159 shares of Rental stock then held by Finance, i.e., the same shares which Finance had received from Mr. and Mrs. Fehrs. In exchange for this stock, Rental transferred to Finance $100,000 in cash and an unsecured negotiable promissory note of Rental in the principal amount of $625,000. The note provided for annual payments of principal and interest. Each principal payment was set at 5 percent of the original principal or 25 percent of Rental’s annual net income after Federal income taxes, whichever amount was greater; the interest rate was set at 7 percent per annum. Annual payments were to be made on January 2 of each year, commencing with 1966. Viewed in terms of their net result, the series of transfers described above resulted in the complete divestiture of the plaintiffs’ ownership of Rental stock, a significant reduction in the outstanding amount of that stock and control of the shares that did remain outstanding being exclusively in the hands of plaintiffs’ children. At the same time, the newly-established Finance company was provided with working capital and it too remained within the exclusive control of the plaintiffs’ daughters. Despite the careful planning that we may assume this realignment of ownership interests to have been based upon, the manner in which the transactions were subsequently reported for tax purposes did not meet with the approval of the Commissioner of Internal Revenue. Finance, for example, for its taxable year ending November 30,1965, had reported neither a gain nor a loss with respect to the proceeds realized from its sale of the Rental stock. However, in the Commissioner’s view, that transaction was seen as giving rise to short-term capital gain that was taxable, in 1965, to the extent of the cash that had then been received, namely, $100,000. Similarly, in regard to the — 496 — plaintiffs (Mr. and Mrs. Fehrs), the Commissioner took issue with the declared amount of their gift tax liability for the year 1965 and, as a separate and later matter, with their treatment of the annuity payments as long-term capital gain rather than as ordinary income. Deficiency assessments followed and then, ultimately, litigation. Finance brought suit in the Tax Court and the decision there went in the Government’s favor. The gist of that court’s decision (of which more shall be said later) was that the sum of the several transfers (as heretofore described) was governed, in the first instance, by section 304 of the Internal Revenue Code (pertaining to redemptions of stock carried out through affiliated corporations) and that, pursuant to that code section, the Rental stock (in Finance’s hands) constituted a "contribution to its capital.” Further, on the particular facts of the case, that stock was deemed to have a zero basis; hence, Finance was held to have realized a gain upon its subsequent disposition of the stock to Rental. With regard to the deficiency determinations asserted against Mr. and Mrs. Fehrs, these matters became the subject of separate actions commenced in this court. As mentioned at the outset, plaintiffs’ position on the treatment of their annuity income as capital gain was subsequently sustained by this court. Left standing, pending the final resolution of that suit, is the question now before the court: the correctness of the Government’s contention that a disparity between the fair market values of the Rental stock and the annuities received for that stock constituted gifts to the shareholders of Finance by Mr. and Mrs. Fehrs. Specifically, the Government’s present position (here modified slightly from the valuations upon which the administrative determinations of deficiency had initially been based) is that, as of March 1, 1965, the 1,159 shares of Rental stock had a fair market value of $754,427.87 or $650.93 per share while the annuities, in turn, had a total value of $529,560.65. Upon these claimed differences in value, the Government bases its case for the assertion of a gift tax liability against the plaintiffs. — 497 — II. Discussion A. The starting point of the discussion is the plaintiffs’ contention that a gift tax liability is foreclosed in this case because the preceding litigation (meaning, plaintiffs’ earlier suit as well as Finance’s suit in the Tax Court) had held the Fehrses’ transfer of shares for annuities to constitute a redemption. Two lines of argument are offered in favor of this position. The first is a legal one, to wit: that a redemption and a transfer in gift signify mutually exclusive concepts. The second is essentially factual in nature. It rests on the Tax Court’s finding that, because of the operation of the constructive ownership provisions of the tax code (i.e., section 318), the plaintiffs retained — in their post-redemption status- — virtually the same degree of corporate control as they had held prior to the transfer of their shares. It is in light of this finding of the Tax Court that plaintiffs now contend that, since there had occurred no cognizable change in stock ownership for income tax purposes, so neither can there now be any recognizable transfer for gift tax purposes. Both arguments lack merit. As to the first, there is no legal basis upon which to sustain the contention that a redemption and a transfer in gift are incompatible concepts that may not be simultaneously applied in characterization of the same' transaction. The argument rests on false notions. For tax purposes, the recognition of a property transfer as a gift does not require a demonstration of donative intent; rather, it is the sufficiency of the consideration received (in money or money’s worth) for the property transferred that underlies the basis of the gift tax. Commissioner v. Wemyss, 324 U.S. 303 (1945). The governing code provision, 26 U.S.C. § 2512, declares that "[w]here property is transferred for less than an adequate and full consideration in money or money’s worth” then the excess in such money value "shall be deemed a gift.” The breadth of this language is plainly sufficient to encompass not only those transfers which accord with the common law concept of gift, but, as Treasury Regulations long have recognized, also embraces "sales, exchanges, and other dispositions of — 498 — property * * * to the extent that the value of the property transferred * * * exceeds the value of the consideration given therefor.”3 The transaction upon which this suit is founded — a corporation’s reacquisition of its stock from its shareholders in return for annuities — is a property transfer and, as such, is plainly within the reach of the gift tax provisions. Decisional law stands squarely in support of such a construction of the regulation. See Epstein v. Commissioner, 53 T.C. 459 (1969) (bargain sale of improved realty to trust established for children held to be a gift to the extent of the excess in value); Small v. Commissioner, 38 T.C.M. (P-H) ¶ 69,211 (1969) (bargain sale of land to children held to be gift in part); and Estate of Bartman v. Commissioner, 10 T.C. 1073 (1948) (taxable gifts to the extent of the excess in land value over annuities given in return). It might also be added, in passing, that the Tax Court too saw no conceptual conflict in the dual characterization of a single transaction as part redemption and part gift. On this subject it had occasion to state in the Fehrs Finance suit: "The issue of whether the transfer was, in part, a gift is not before us, and we cannot decide it because no evidence as to the value of the Rental stock has been presented. Solely for the purposes of this opinion, we shall treat the entire transfer as a redemption.” Fehrs Finance Co. v. Commissioner, 58 T.C. 174, 183 n.3 (1972), aff'd, 487 F.2d 184 (8th Cir. 1973), cert. denied, 416 U.S. 938 (1974). This is not to say, of course, that every transaction of like form is an automatic candidate for gift tax surveillance. Redemptions that are carried out in the ordinary course of business, that is, transactions where the possibility of donative intent can be ruled out on the basis of the arm’s-length nature of the dealings involved and their bona fide business objectives, are not within the intended scope of the gift tax. The regulations so provide, Treas. Reg. § 25.2511-l(g)(l)(1979). Thus, as to such transactions, the requirement for equivalent consideration in money or money’s worth is — 499 — not an essential requirement to preserve the legitimate business character of the property transfer. The tax law does not insist that a bad bargain be turned into a gift by default. The problem here, however, is that we deal entirely with an intra-family transfer, a circumstance which has always prompted special scrutiny by the courts precisely because the genuineness of the transaction cannot reasonably be inferred from any circumstantial assurances of business purpose. Helvering v. Clifford, 309 U.S. 331 (1940). A close look at the transaction is, therefore, unavoidable; the disparities in value that we ultimately come to, when taken together with the plaintiffs’ full control over both ends of the transaction, are inconsistent with any form of property transfer save that of a gift. To put it another way, since the visible aspects of the transaction do not negate the existence of donative intent, there exists no ground upon which to conclude that the transaction was prompted solely by business considerations. Concerning plaintiffs’ second argument — which, as noted, stresses the apparent irreconcilability between the Tax Court’s conclusion that plaintiffs remained in control of the Rental stock even after its redemption (because of attributed ownership) and the present action, which proceeds from the assumption that a total divestiture of such stock actually did take place — the answer is simply that the Tax Court’s decision is not relevant to this proceeding. To explain: the plaintiffs were not parties to the Tax Court litigation and the attribution rules that were there invoked against Finance were subsequently found, in the plaintiffs’ own later suit in this court, to have been satisfactorily overcome. Thus, contrary to the Tax Court, and more importantly, contrary to the very argument now being made, this court held that the redemption did result in a complete termination of the plaintiffs’ interest in Rental. Lest it be misunderstood, it should be explained that this seeming divergence in judicial results was due entirely to the fact that, at the time of the Tax Court litigation, plaintiffs had not accomplished a waiver of the family attribution rules through the filing of an agreement specified for that purpose by section 302(c)(2)(A)(iii) of the Internal Revenue Code. Later, however, upon the conclu — 500 — sion of the Tax Court litigation and several years prior to commencing their income tax refund suit here, such an agreement was filed by them. We held, under the particular facts of the case, that their belated filing was adequately explained by the circumstances and was, therefore, a timely filing. The important point now, of course, is that even as the Tax Court’s earlier result was held not to collaterally estop plaintiffs from arguing for a different outcome based on a changed set of facts, so neither may they now hold the Government to that earlier result. Indeed, at this juncture, it ponders only an academic point to question whether plaintiffs would ever have been entitled to offensively assert the Tax Court’s results in this court had they first litigated here the matter of their gift tax liability rather than their income tax liability. The order of litigation being as it was, and this court having determined that plaintiffs did accomplish a complete termination of their interest in Rental, they are now conclusively bound by that determination. B. With the threshold question thus aside, the central issue in this case comes into focus — whether the fair market value of the Rental stock exceeded the value of the annuities received for that stock. We address first the question of the annuities. The main point of difference between the parties on this question is whether the annuities are to be valued as private annuities, as the Government contends they should be, or, as plaintiffs claim — as commercial annuities. It has been agreed between the parties that, as private annuities, their value as of March 1, 1965, would have been $529,506.65, and, if computed as commercial annuities, their value then would have been $689,123. The Government’s method of valuation is based on the dictates of regulation, Treas. Reg. § 25.2512-5 (1969), and that, as may be surmised, assumes a cost difference depending upon whether the annuity represents the contract obligation of a company regularly engaged in such sales or, as here, the undertaking of a private party. To overcome the presumption of correctness which the law accords to the Government’s regulation, Bowden v. — 501 — Commissioner, 234 F.2d 937 (5th Cir. 1956), the plaintiffs rely on three arguments. Their first, and principal contention, is that the Government’s valuation is wrong because it reflects actuarial values that are drawn from general population tables which, by virtue of their composite character, ignore such individually significant factors as an annuitant’s sex, occupation, income and state of health. These factors, it is said, account for differences in life span and that, in turn, accounts for dollar differences in annuity values. In support of these assertions, plaintiffs point to the decision in Dunigan v. United States, 434 F.2d 892 (5th Cir. 1970), a case in which like criticisms of the Government’s valuation method (the same method as here employed) were considered, along with other evidence bearing upon the taxpayer’s life expectancy, as providing a sufficient eviden-tiary basis upon which to sustain a determination that the Government’s method was arbitrary, unreasonable and erroneous and that the taxpayer was entitled to a valuation reflecting a commercial annuity classification. The result reached in Dunigan may not be duplicated here; the evidence would never permit it. For one thing, there is no proof at all of the very point that plaintiffs make, namely, that the Government’s private annuity tables subsume outmoded statistics. That proposition is present in this case only as an unsupported pronouncement. But even if judicial notice would permit the acceptance here of facts established in another proceeding, a want of proof would yet remain. It would still be necessary for plaintiffs to establish, as the taxpayer in Dunigan did, those personal attributes of wealth and physical condition which, when linked by expert actuarial testimony, would justify a life expectancy exceeding that reflected in the Government’s composite data. Estate of Christ v. Commissioner, 480 F.2d 171, 174 (9th Cir. 1973); Miami Beach First National Bank v. United States, 443 F.2d 116, 119 (5th Cir. 1971). Such proof was not offered and without it, the burden of overcoming the presumptive correctness of the Government’s regulation has not been met. A second argument that plaintiffs raise is based upon the Tax Court’s finding to the effect that, had Finance sought to — 502 — discharge its liabilities to the Fehrses under the annuity agreements by purchasing such annuities from commercial sources, then the total cost to Finance, in 1965, would have been approximately $655,000. From this finding, two related contentions are advanced, the first being that the commercial cost of the plaintiffs’ annuities, i.e., comparable annuities purchased on the open market, represehts the best measure of the actual value of plaintiffs’ annuities; the second is that the Tax Court’s finding represents a prior determination to which the defendant is now bound. Neither argument has merit. The contention that the best measure of value for a private annuity is its cost on the open market is simply another way of saying that there are no differences between the two. The fact is, however, that it is precisely because they are different in terms of their underlying economics that different values apply to each. Dunigan v. United States, supra. The rates charged for commercial annuities are higher because those rates must accommodate business expenses and profit margins, more rigorous restraints on income investment (and therefore lower returns on such income) and, as a self-selective class, wealthier annuitants who are longer lived. Then too, there is the fact that commercial annuities are generally backed by larger assets and are therefore of more worth because they offer greater certainty of payment. For these reasons then, commercial annuities and private annuities are not one and the same either from the standpoint of the issuer or the annuitant. Thus, what the Tax Court had to say about what the cost of the plaintiffs’ annuities would have been had commercial annuities been subsequently substituted in their place is all beside the point; such a substitution never took place. The last point that plaintiffs argue is that their annuities were entitled to a higher valuation than the Government had assigned to them because they actually outlived their life expectancies and thus received payments in excess of the values that were projected as of the valuation date of March 1,1965. There is offered in support of this position, the decision in Estate of Chesterton v. United States, 213 Ct. Cl. 345, 551 F.2d 278, cert. denied, 434 U.S. 835 (1977), a case holding — 503 — that, for estate tax purposes, the value of a former wife’s alimony claim against the decedent’s estate (an allowable deduction under section 2053(a)(3) of the Internal Revenue Code) should be determined by reference to the facts that were known at the time of the estate tax filing rather that on the basis of the earlier facts that prevailed at the time of the decedent’s death. Since the alimony payments in that case were terminated shortly after decedent’s death (because of the wife’s remarriage), the estate was permitted to deduct only the actual amount of those payments rather than the projected (and much larger) amount of such payments as seemed likely on the basis of the facts prevailing at the time of the decedent’s death. What plaintiffs draw from this case is the point first mentioned — that the actuality of subsequent events should displace valuations based upon statistical probabilities. The problem with their argument is that it is much too broad an abstraction. The Chesterton decision was concerned with a claim against an estate and it was only with that particular subject in mind that the court held that the amount of the deduction should be determined in light of those later events which conclusively established its value, i.e., the amount actually paid by the estate in respect to the claim. Here, on the other hand, we deal with an obligation which, by its very nature, is fixed in value at its inception for the rate charged for an annuity, like that charged for a life insurance policy, takes into consideration at the outset the predictable variations in human life span. The market value of an annuity is the value that obtains on the day it is acquired and that value does not change simply because the predictions on which it is based turn out to be different in one case than another. For the several reasons that have been stated, it is concluded that plaintiffs’ annuities are to be valued as private annuities and that value, as the parties have agreed, is $529,506.25. C. There remains for consideration the matter of determining the value, as of March 1, 1965, of the Rental stock that was given by the plaintiffs in return for the annuities. The starting point is the familiar principle that the value of the property is the price at which such property would change hands between a willing buyer and a willing seller, — 504 — neither being under any compulsion to buy or sell and both having reasonable knowledge of relevant facts. With this in mind, the parties have offered two distinct theories of valuation, each the product of expert testimony. On the plaintiffs’ side of the equation is a theory that espouses a "going concern value,” that is to say, a valuation that assumes the continuation of the business on the part of the hypothetical purchaser and hence a valuation that draws its determinants of price largely from the statistics of companies engaged, either in whole or in part, in similar lines of business. The Government’s expert, on the other hand, viewed Rental as being, at best, only a marginally profitable company that was actually worth more in liquidation than as a continuing enterprise. Accordingly, it was his position that, since the interest being valued here was that of a controlling shareholder, the valuation of the Rental stock would have to proceed on the assumption that corporate control would favor (and thus dictate) disposition of the corporate assets rather than the stock in order to reap the advantage of the greater market value inherent in those assets. Although both points of view endorse the proposition that that valuation should obtain which affords the highest realistic appraisal, both, as the discussion will bring out, have their problems. To address first the approach that was taken by plaintiffs’ expert, his analysis was one that relied exclusively upon an application to Rental of certain stock price ratios that were exhibited in the market price of the shares of four publicly-traded companies — Syracuse Supply Company (Syracuse), a New York corporation involved in the distribution of construction equipment, machinery, machine tools and industrial supplies; Hajoca Corporation (Hajoca), a Delaware corporation, with New York sales offices, involved in the distribution of construction equipment, material and supplies; Bluefield Supply Company (Bluefield), a conglomerate enterprise incorporated and located in New York that dealt in contractor equipment, mine and milling equipment, hardware, and electrical heating and ventilating supplies; and, finally, W. W. Williams Company (Williams), an Ohio corporation that — 505 — was a distributor of machinery, repair equipment and supplies. An examination of the published financial statistics concerning these four companies revealed the following: (i) a distribution of price-earnings ratios (i.e, the price per share divided by the earnings per share) ranging from a high multiple of 9.8 to a low multiple of 3.9 with the average and median ratios being respectively 6.3 and 5.8); (ii) a distribution of price-sales ratios (the price per share divided by the sales per share) ranging from a high of .254 to a low of .097 with the average and median ratios being respectively .156 and .137; and (iii) a distribution of price-net worth ratios (the price per share divided by the net worth per share) ranging from a high factor of .838 to a low of .487 with the average and median values being respectively, .67 and .679. From this data, the expert selected three representative values or indicators, the selection for each set being the value falling between the average and the median. Thus, for a representative price-earnings ratio, 6 became the chosen indicator, for a representative price-sales ratio, .14 was chosen and, for the price-net worth factor, the indicator selected was .675. The next step in the analysis involved the application of these values to the corresponding earnings, sales and net worth of Rental. And it is here that trouble sets in. As a threshold criticism, the Government raises the argument that the indicators lack validity because they are being drawn from companies that are not, in any true sense, comparable to Rental. In support of this position, several distinguishing factors are called out, foremost among these being a significant difference in profitability between Rental and the comparison companies.4 This particular factor does indeed bear scrutiny, but, as subsequent discussion will develop, its adverse implications to plaintiffs’ — 506 — analysis relate more to the manner in which the factor was used by the expert than to the issue of comparability per se. The discrepancy in profit margins that the Government has called out was a factor that plaintiffs’ expert took cognizance of from the outset. It was his judgment that, in contrast to industry norms, Rental’s own record of earnings over the 4-year period preceding the stock transfer date (1961-64) had been so poor as to preclude any meaningful use of those earnings as a base from which to project a likely price per share through the application of the price-earnings factor {i.e., the previously identified indicator of 6 times earnings). Over the 4-year period mentioned, Rental’s "normalized” sales (roughly the equivalent of average sales) came to $6.8 million annually but earnings, in any single year during that time, never exceeded $39,000 — a profit-on-sales return barely over V2 percent. In fact, such profits as there were in these years were not only erratic but also included a large loss (in 1962) that, if taken into an average for the period, would leave the company with an overall deficit performance.5 Given this poor showing, the expert concluded that a prudent investor would look to valuing Rental’s stock, not by reference to the earnings as they actually were but rather as they might be given the benefit of a change in management and with it, presumably, an attainment of earnings more in line with that of the industry as a whole.6 To that end then, the expert concluded that, in different hands, Rental’s normalized sales of — 507 — $6,800,000 could be expected to show a pre-tax profit of 4 percent7 or $272,000, which, after allowance for Federal income taxes (measured at 48 percent), would reduce to $141,440. Application of the multiplier (the price-earnings ratio of 6) would indicate an aggregate value of $848,640 (rounded by the expert to $850,000); a figure which was then adjusted downward by 15 percent in order to give recognition to a discount in value that is typically associated with the sale of shares for which there exists no established trading market or other broad-gauged consensus of value. Bader v. United States, 172 F. Supp. 833, 838 (S.D. Ill. 1959). As a net result, the expert concluded that Rental’s shares, when valued by reference to the selected price-earnings factor, would be worth $722,500 (again rounded, in his analysis, to $720,000) or $521.74 per share. As has been noted, the Government maintains that none of the comparison statistics relied upon by plaintiffs’ expert have any validity, least of all the earnings multiplier, for, in its view, those statistics are drawn from companies much different than Rental. The point is debatable but, at the same time, it is one that need not be met head on — there is a problem with plaintiffs’ price-earnings valuation quite apart from the root question of comparability. The problem has to do with the fact that the expert entirely avoided the use of Rental’s own earnings as a basis for the application of the earnings multiplier in favor of a reconstructed earnings base derived wholly from the profitability levels of companies other than Rental. This, as the court sees it, is an approach that proceeds not by comparison but by substitution and, in the last analysis, represents the valuation of a non-existent company. To be sure, prospective earnings are important in the determination of share price. However, a valuation that completely disregards the recognized importance of actual earnings as a factor in the assessment of future expectations, see Central Trust Co. v. United States, 158 Ct. Cl. 504, 523, 305 F.2d 393, 404 (1962), and Rev. Rul. 59-60, 1959-1 — 508 — C.B. 237, 242, must rest on something more than the naked speculation that a dramatic change in corporate fortunes would follow from a change in corporate management. Here, however, we have nothing that would lend even remote support to the premise that Rental’s profit doldrums were of its own making (and implicit in the analysis — instantly correctable) save only the expert’s supposition to such effect. That is not enough to persuade for, like any other judgments, those of an expert can be no better than the soundness of the reasons that stand in support of them. There being no other facts the court might use in place of those offered, it can only be concluded that, insofar as a price-earnings valuation is concerned, plaintiffs have failed to make their case. As to the other branches of the expert’s analysis, here too there are problems of a fundamental kind. For example, in the second of his three valuation techniques, the expert relied upon a price-sales ratio, that is to say, the price of the Rental stock was determined as a percentage of the company’s normalized sales, the percentage, of course, being dictated by the ratio between stock price and sales volume that prevailed among the comparison companies. The problem with this approach to valuation is that it builds upon an essentially meaningless statistic. The prospective purchaser of a going concern is interested in the actual as well as potential earning capacity of the business and while sales volume is certainly an important factor in gauging that capacity, it clearly is not an end unto itself. For the prospective purchaser, the "bottom line” has to be the amount of profit per dollar of sale that can reasonably be expected and any valuation that misses that point — as the instant one surely does — is lacking in economic substance. The case at hand amply demonstrates the point: by the price-sales method of valuation, Rental is assigned a value that, on a relative basis, is the same as the composite of the comparison companies notwithstanding the fact that its pre-tax profit percentages and earnings through the years 1961 to 1964 never came close to matching even the lowest return recorded among the comparison companies. Thus, the approach masks the very point for which it is undertak — 509 — en in the first place, namely, to provide a substantially accurate assessment of corporate worth based upon the determinants of value that prevail in the market place. For the risk-reward appraisal that underlies every sound investment decision, the price-sales ratio is plainly a false indicator to go by. Much the same criticism can be made with respect to the expert’s third approach to valuation — the price-net worth analysis. Where the worth of a business is tied closely to its assets, as in the case of a real estate investment company for example, the net worth approach to valuation commands primary attention. Rev. Rui. 59-60, 1959-1 C.B. 237, 243. However, in the case of a manufacturing or service enterprise that is viewed as a going concern, net worth considerations diminish in importance. Righter v. United States, 194 Ct. Cl. 400, 412, 439 F.2d 1204, 1210 (1971). Market value in such circumstances derives not from liquidation values, but from the assessment that investors make of such elements as managerial skill, product demand, market position and actual and potential competition — in short, all those ingredients that are recognized as having a direct bearing upon a business’ present and future earnings capability. The statistics at hand demonstrate this. What the comparison data shows is that the price-earnings multiples of the four companies stood in reverse order to their returns on equity, i.e., to their profits when measured as a percentage of net worth. For example, Bluefield, the company whose stock traded at 9.8 times earnings (the high end of the multiple range), realized only a 7.42 percent return on its equity while Williams, the company at the opposite end of the multiple range (its stock traded at only 3.9 times earnings), enjoyed the highest return on its equity, namely, 16.3 percent. For the two in-between companies, Hajoca and Syracuse, the relationships followed the same inverse order. Now the point here is not that price-earnings multiples and returns on equity invariably pursue opposite directions; indeed, the fact that they are seen to do so in this instance may perhaps be due only to happenstance. Rather, the point being made is that, if net worth were truly a vital — 510 — determinant of share price, then one could logically anticipate the company showing the largest percentage return on its net worth to find greater favor among investors; in short, share price (as expressed through earnings multiples) would gravitate towards net worth values. It is the absolute absence of that relationship from the comparative data that makes the case for saying that, for a going concern, share value is determined by earnings considerations and not book value concerns. This observation is not intended to disparage the use of price-net worth ratios for all share evaluation purposes. Certainly, in situations where the companies involved stand closer together performance-wise than they do here, it can at least serve as a back-up standard of valuation. But, in this case, where the plaintiffs’ proof has not succeeded in establishing to the court’s satisfaction the validity of the other approaches to valuation, it would leave the matter far too much in doubt to assign to Rental a going-concern value through the use of a single indicator that does not appear to bear any relation to that value. The plaintiffs’ expert too saw deficiencies in the price-to-net worth ratio. His concern was that a prospective purchaser would be hesitant to apply the chosen price-net worth indicator to a book value figure that might not accurately reflect the fair market value of the underlying assets. Thus, his report states: "because [of] our inability to carefully review the individual assets due to this time lapse, the book value of the firm is not a reasonable indicator of the fair market value.” This may well be another good reason for avoiding the use of a price-net worth ratio in this case, but, for the court, the fundamental reason is grounded in the fact that that ratio has not been shown, through the proof made available here, to address any meaningful price determining relationship in the first place. The sum of it all then is that, for the plaintiffs there can be no escape from the Government’s valuation approach for their own has failed at every turn. Although the proof leaves no alternative save to accept the Government’s valuation approach, this does not mean that the court is bound to accept all that the defendant’s expert had to say on the subject. Righter v. United States, — 511 — supra, 194 Ct. Cl. at 435 n.3, 439 F.2d at 1224 n.3 (1971). However, with that approach, there is actually little fault to be found. As mentioned earlier, the defendant’s expert focused on the intrinsic value of Rental’s underlying assets and arrived at the conclusion that those assets, if disposed of within a relatively short time, i.e., over a period of 2 to 3 months, would produce a net liquidation value of $898,283.28 or $650.93 (on the basis of 1,380 shares). The main problem with the analysis — at least from the plaintiffs’ standpoint — has to do with the expert’s treatment of Rental’s accounts receivable. It was his assumption that those accounts (excluding conditional sales contracts which were dealt with separately because they were offset by liabilities of nearly like amount) were of average market quality, i.e., accounts of which no more than 5 to 10 percent would turn out to be uncollectible. On this assumption then, it was concluded that the accounts could be sold on the open market at their face amount less the normal broker discount of 20 percent. Thus, from his tally of the several receivable accounts (which came to $660,172.12) the expert projected a net market value of $528,137.69. The fallacy that plaintiffs see in this approach derives from the fact that Rental’s bad debt and repossession losses were historically much higher than the amount anticipated in defendant’s analysis. It is pointed out, for example, that, in the fiscal year ending September 30, 1964, Rental’s charge against income for bad debts and repossession losses (as shown in its Federal income tax return) was in excess of $425,000 — a figure dramatically different from the 5 to 10 percent loss rate (or $30,000 to $60,000) to which the defendant’s analysis was pegged. The "correction” that plaintiffs would make in the Government’s approach would be to substitute the known amount of bad debt losses in place of the Government’s estimated loss figure. The appeal in this argument lies only on the surface. The point that is completely overlooked is that the defendant’s expert used, as the basis for his calculation, the value of the accounts receivable at year’s end, or, in other words, the amounts reflected in those accounts after deduction of the bad debt and repossession losses that had been recognized during the year. In short, the known loss figure of $425,000 — 512 — had already been taken into account; what was valued by the expert was the sum of the remainder. Thus, the only real question here is whether, as to that remainder, it was appropriate to apply a 20 percent discount. The answer has to be "yes” for Rental itself saw fit to cover the year-end value of its accounts receivable by a bad debt reserve amount that came to about 9.5 percent of the total of those accounts.8 Since the expert’s projection of a 5 to 10 percent default rate was in line with Rental’s own, there is no occasion now to say he was wrong. The other point that plaintiffs challenge concerns the expert’s approach to the problem of valuing Rental’s pledged assets, i.e., its new and used equipment. The expert did not know, and could not ascertain, the true market value of these assets. He therefore made the assumption that the liquidation value of these assets would be equal to the liabilities for which they had been pledged — this based on the premise that the lending institutions would not allow the loan balance to exceed the value of the security. Plaintiffs say that a no less acceptable approach would be to value the assets separately — in which case a 10 percent discount from market value would first have to be recognized — and then, from this discounted value, covering liabilities would be deducted. Accordingly, if this approach were followed, the value of the assets would come to an amount less than the liabilities which they secured. The excess of the liabilities would therefore lead to a reduction in the expert’s overall valuation amount. The logic of plaintiffs’ position is sound — so far as it goes. However, the problem with that position is that it takes as fact what the expert only chose as a compromise solution in the first place, namely, to treat the assets as equal in value to their corresponding liabilities. The whole point of the expert’s position was that, absent actual knowledge of the market value of the assets (a fact no longer ascertainable because of the lapse of time) the best position to take would be to assume an equivalency between assets and liabilities. — 513 — That asset values might, in fact, have actually been greater than the liabilities was as much a possibility as the reverse of that situation. Indeed, looked at strictly in terms of the balance sheet data, such was the case — the assets did exceed their corresponding liabilities by $152,391.44. However, lacking any basis for saying that book value and market value were actually one and the same, the expert adopted the approach that he did. On the basis of the limited information available to him, and the absence of any proof on plaintiffs’ part that pledged assets were actually worth less than the loans they were meant to secure, the expert’s approach must be allowed to stand. The approach was a reasonable one given the problem it had to deal with. Having considered the two approaches to valuation that were offered in this case, the court concludes that the fair market value of the Rental stock is to be determined by, and in accordance with, the values recommended by defendant’s expert, namely, $650.93 per share. III. Conclusion As herein determined, the fair market value of the 1,159 shares of Rental stock which the plaintiffs transferred to Finance on March 1, 1965, was $754,427.87 or $650.93 per share and the value of the annuities received in return was $529,506.65. For the reasons given in the opinion, upon this difference in value, namely, $224,921.22, a gift tax liability is properly founded. Inasmuch as the deficiency assessments which plaintiffs had timely paid were based upon a greater amount, namely, $258,926.16, they are now due a refund to the extent of the excess gift taxes paid together with interest on that excess as provided by law. Determination of the amount of refund is reserved for further proceedings pursuant to Rule 131(c). An order and judgment to these effects is hereby entered in plaintiffs’ favor; in all other respects their petitions are dismissed. — 514 — CONCLUSION OF LAW Upon the trial judge’s findings and the foregoing opinion, which are adopted by the court, the court concludes that plaintiffs are entitled to a partial refund of gift taxes paid, together with interest as provided by law, and in all other respects, their petitions are dismissed. The amount of recovery is to be determined through further proceedings pursuant to Rule 131(c). In accordance with the opinion of the court, a stipulation of the parties, and a memorandum report of the trial judge, it was ordered on June 13,1980 that judgment be entered in favor of plaintiff, Violette E. Fehrs, for $7,925.16, together with statutory interest. Footnotes [*] Although the court adopted the trial judge’s separate findings of fact, which are set forth in his opinion, they are not printed herein since such facts as are necessary to the decision are contained in his opinion. [**] “Edward J. Fehrs (plaintiff in No. 118-72) died subsequent to the commencement of his suit here; accordingly, a motion to substitute Violette E. Fehrs, his surviving wife (plaintiff in No. 119-72), as plaintiff in No. 118-72 will be required prior to the entry of final judgment. [1] Unless otherwise noted, all code references in this opinion are to the Internal Revenue Code of 1954. [2] Mr. Fehrs died in 1973. His jointly-held property passed to his surviving wife, plaintiff herein, and all other assets were transferred under the terms of a revocable trust; no estate of Edward J. Fehrs was subject to formal probate proceedings. [3] The language quoted is taken from Article 8 of Treasury Regulation 79 (1936 ed.) which implemented the gift tax provisions of the Revenue Act of 1932 as amended and supplemented by the Revenue Acts of 1934 and 1935. Except for a minor change in word order, the quoted text has remained in continuous effect since the time of its initial promulgation. The regulation currently appears as Treasury Regulation § 25.2512-8 (1979). [4] In addition to the differences in profitability, other factors which the Government cites as being relevant to the issue of comparability are: (i) that the comparison companies were much larger in size in that their annual sales volume ranged from a minimum of twice that of Rental in the case of Williams to a volume over seven times greater in the case of Hajoca; (ii) that they operated from more than one business location; and (iii) that their product lines were diverse whereas Rental was concerned only with the sale, lease, rental and repair of heavy road and industrial equipment. [5] Rental’s earnings, as shown in its Federal income tax returns for the years in question, were as follows: 1961; $38,357; 1962: ($139,920); 1963: $4,827; 1964: $23,593. Since no income tax was paid by Rental on any of these amounts (presumably, because of tax loss carry forwards), its pre- and post-tax income during these years was the same. [6] In his evaluation of Rental’s earnings, the expert had also given consideration to the profits which Rental recorded for the year ending September 30, 1965. This was error on his part because earnings achieved after the stock transfer date (i.e., March 1,1965), would not have been known to the prospective purchaser — except perhaps by projection — and thus could not logically fit into a valuation analysis that had to be predicated on the facts known as of the date of transfer. Central Trust Co. v. United States, 158 Ct. Cl. 504, 520, 305 F.2d 393, 403 (1962). However, this impermissible viewing of post-transfer data is of little consequence here since it did not affect the expert’s judgment: notwithstanding a substantial profit improvement in 1965, the expert adhered to his position that a prospective purchaser would not view any of Rental’s earnings as indicative of its true value. [7] The comparison companies showed a pre-tax profit on sales ranging from 5.5 percent to 3.2 percent, with the average and median figures being, respectively 4.5 percent and 4.6 percent. Thus, the 4 percent figure which was applied to Rental’s normalized sales was slightly less than the average of the comparison companies. [8] On its balance sheet for the year ending September 30, 1964, Rental showed accounts receivable of $934,098.16 (this included conditional sales contracts of $273,926.04) and a remaining bad debt reserve applicable thereto of $89,000, or, roughly 9.5 percent.