Nos. 82-1500, 82-2519.
Argued January 20, 1983.
Decided April 12, 1983. Rehearing and Rehearing En Banc Denied July 7, 1983.
Farley P. Katz, Tax Div., Dept. of Justice, Washington, D.C., for defendant-appellant.
Charles F. Cremer, Jr., Charles F. Cremer, Jr. Co., Indianapolis, Ind., for plaintiff-appellee.
Appeal from the United States District Court for the Southern District of Indiana.
This is a consolidated appeal from the district court's judgment entered on a jury verdict awarding the estate of B.L. Curry a refund of federal estate taxes in the amount of $209,904.59 plus interest. On appeal, the government challenges as error the district court's giving of two jury instructions objected to by the government and the failure to tender to the jury two of the government's proffered instructions. The government also challenges the district court's award of attorney's fees to the estate under the Equal Access to Justice Act, 28 U.S.C. § 2412(d)(1)(A). Because we find that the district court erred in tendering certain instructions and refusing to tender certain other instructions to the jury, we vacate the judgment of the district court, 549 F. Supp. 47, reverse its award of attorney's fees to the estate, and remand for a new trial.
This case arose from a dispute between the estate and the government concerning the proper value, for federal estate tax purposes, of B.L. Curry's stock in B.L. Curry and Sons, Inc., a closely held Indiana corporation engaged in the manufacture of wood veneer. At the time of B.L. Curry's death, the authorized and issued stock of the company was composed of 1500 shares of voting common stock and 4500 shares of Class A non-voting common stock; all 6000 shares were held by B.L. Curry and his children. B.L. Curry himself owned 800 shares of the voting common stock, enough to give him voting control over the company, and in addition 1360 shares of non-voting common stock.
On the decedent's estate tax return, the executor valued B.L. Curry's voting stock in the company at $135,312, or $169.14 per share, and his non-voting stock at $25,554.40, or $18.79 per share. The government disagreed with these valuations after an audit, and instead assigned values of $400 per share and $300 per share to the decedent's voting and non-voting stock, respectively. As a result, the estate was assessed an additional $199,218.84 in taxes. After payment of this increment, the estate's administrative refund efforts were unavailing, and the instant litigation ensued.
At trial, Halsey Sandford, the estate's principal valuation witness, valued the estate's voting and non-voting stock differently, even thought the estate possessed sufficient voting stock to provide it with fifty-three percent voting control over the company. On cross-examination, Sandford explained that this bifurcated analysis assumed that the two blocks of stocks would be sold separately. In order to value the 1360 non-voting shares, Sandford testified, he first posited a hypothetical $125 per share public trading price based on the dual factors of the company's projected earnings as a going concern and the liquidation value of the company's assets, with emphasis given to the former. This figure was then discounted by half to reflect the non-marketability of the company's stock and by an additional ten percent to reflect the existence of stock purchase restrictions in the company's articles of incorporation — discounts which yielded a bottom line figure of $56.25 per share.
In valuing the 800 voting shares, Sandford also averaged their going concern and liquidation values. He arrived at a going concern value of $100 by augmenting each share's intrinsic value by a sixty percent control premium factor; he arrived at a $191 per share liquidation value by discounting net asset value by twenty-five percent to compensate for liquidation expenses. The two resultant figures were averaged to yield a $150 per share value for the 800 voting shares (totalling $120,000), as contrasted with $56.25 per share for the non-voting shares (totalling $76,500).
In contrast to Sandford, the government's main valuation witness assigned an identical value to the decedent's voting and non-voting shares: $290.50 (for a total of $627,480). This figure was derived through an earnings multiplier approach, the multiplier having been based upon a review of market multiples in comparable industries. The resultant figure was in turn adjusted to reflect selling costs, excess working capital and earnings history. The government's witness also testified that the liquidation value proffered by Sandford should be viewed as the minimum value attributable to the decedent's stock.
The court refused to give two of the government's proffered instructions. The first would have required the jury to value the decedent's non-voting stock at the same level as his voting stock, while the second would have prevented the jury from finding a value below that which the estate conceded was realizable upon liquidation. In addition to rejecting these instructions, the district court instructed the jury, over the government's objections, that "[i]f you find plaintiff's corporation was prosperous and the idea of liquidating it was more remote, then you should give stronger emphasis on [sic] earning power and payment of dividends" and that, in valuing the stock, the jury could consider the effect of the stock purchase restrictions contained in the company's articles of incorporation.
The jury accepted the differential values for decedent's voting and non-voting stock testified to by the estate's expert, and the district court accordingly entered judgment for the estate in the amount of $209,904.59 plus interest. Several months later, the district court entered its judgment awarding the estate costs, including attorney's fees, in the amount of $39,851.72 plus statutory costs. On appeal, the government contends that the district court erred in refusing to tender to the jury the government's two proffered instructions, in tendering those to which the government objected, and in awarding attorney's fees to the estate.
II.  A. The Equivalency Instructions
The government first assigns as error the court's refusal to instruct the jury, as the government requested, that "[b]ecause the decedent had voting control of the company, I instruct you that in valuing [decedent's] interest in the company the non-voting stock was worth as much per share as the voting stock." In support of its instruction, the government argues that, for estate tax purposes, the property transferred must be valued as the decedent held it, not in the form it could conceivably take in a subsequent transfer, and that, in the hands of the estate, the absence of voting control appurtenant to some of the shares of stock would not diminish their value, as voting control still resided in the decedent's power. The estate, by contrast, argues that the decedent's stock holdings were more properly split into separate voting and non-voting blocks prior to valuation, and that the non-voting stock could possess a lesser value even if considered as part of a single bundle including its voting counterpart. In our view both the law and common sense compel the conclusion that the fair market value of the non-voting stock in the hands of an estate with sufficient shares of voting stock to ensure the estate's control of a corporation cannot be less than the value of the estate's voting stock. Therefore, we conclude, in rejecting the government's instruction, the district court erred as a matter of law to the substantial prejudice of the rights of the government.
The estate has put forth numerous arguments supporting its contention that the government has "waived" its argument with respect to this and other alleged errors because, inter alia, the government failed to include separate texts of the refused instructions on appeal, failed to cite controlling authority in its brief on appeal, and failed to object with "sufficient clarity" to the offending instructions. We have examined these arguments and find them to be without merit. With respect to the particular instruction examined here, the record reveals that the government's objection alerted the court to each step of its argument: that the stock be valued as part of the estate's entire interest, and that such a perspective requires a finding of equal valuation for both classes of stock. Tr. 465-66.
Section 2001 of the Internal Revenue Code of 1954, 26 U.S.C. § 2031(a) provides that the value of the gross estate of the decedent is determined by including "all property" therein. The corresponding Treasury Regulations provide that the value of includible property is its "fair market value" at the time of decedent's death. Section 20.2031-1(b), Treasury Regulations on Estate Tax (1954 Code). That regulation states further that the "fair market value is the price at which the property would change hands between a willing buyer and a willing seller, neither being under any compulsion to buy or to sell and both having reasonable knowledge of the relevant facts." Id. The first question for our purposes thus becomes whether the property of which the fair market value is to be assessed should be viewed as it exists in the hands of the estate, or as it may exist if fortuitously balkanized through a chain of post-death transactions.
We believe that the first perspective comports more fully with the nature of the estate tax. As the Supreme Court has explained, the estate tax was not conceived as "a tax upon succession and receipt of benefits under the law or the will. It was death duties as distinguished from a legacy or succession tax. What this law taxes is not the interest to which the legatees and devisees succeeded on death, but the interest which ceased by reason of the death." YMCA v. Davis, 264 U.S. 47, 50, 44 S.Ct. 291, 292, 68 L.Ed. 558 (1924). Other courts have emphasized that the resultant "valuation is determined by the interest that passes, and the value of the interest before or after death is pertinent only as it serves to indicate the value at death." United States v. Land, 303 F.2d 170 (5th Cir. 1962) (emphasis in original); see also Estate of Bright v. United States, 658 F.2d 999 (5th Cir. 1981). The interest that passed in this case was the decedent's interest in an 1160-share bundle of stock, the 800 voting shares of which assured complete corporate control. Plainly, then, to meet the mandate of the Code, those shares are to be valued as part and parcel of the interest, not as arbitrarily disaggregated under one possible subsequent transaction scenario.
It is only in rare cases that the event of death will be held to have, for estate tax purposes, any impact upon the fair market value of the interest that passes from decedent, as when, for example, a small business thereby loses the services of a key partner, United States v. Land, 303 F.2d 170, 172, or the will provides for a recapitalization of the stock package owned by the estate, Provident National Bank v. United States, 581 F.2d 1081, 1086-87 (3d Cir. 1978); in such cases, it should be noted, the attendant revaluations occur as a result of conditions internal to the integrated estate and are not based upon any hypothetical piecemeal dissolution of the estate.
Although, surprisingly, this precise issue has not been widely addressed, our interpretation has been embraced by the Ninth Circuit in Ahmanson Foundation v. United States, 674 F.2d 761 (9th Cir. 1981). In Ahmanson the estate argued, as here, that the estate's non-voting stock shares should be valued separately from the estate's sole and controlling voting share. The Ninth Circuit rejected this argument, noting that the estate tax "is a tax on the privilege of passing on property, not a tax on the privilege of receiving property. . . . There is nothing in the statutes or in the case law that suggests that valuation of the gross estate should take into account that the assets will come to rest in several hands rather than one." Ahmanson, 674 F.2d at 768. Likewise, we hold that the interest in the integrated estate forms the only basis for valuation which rationally comports with the purpose of the tax at issue.
Indeed, Ahmanson posed a case even more conducive to the disaggregation principle proffered by the estate here, for the evidence there indicated that the non-voting and voting shares had in fact devolved into separate hands. Ahmanson, 674 F.2d at 768. A fortiori, the principles applied in Ahmanson are applicable here, where only conclusory testimony suggested that the non-voting shares might be sold separately. See note 4, infra.
Additionally, to permit the hypothetical bifurcation of an otherwise integrated bundle of property for valuation purposes would severely undermine the estate tax system and permit abusive manipulation by inviting an executor to invent elaborate scenarios of disaggregated disposition in order to minimize total value. For example, an estate in possession of all shares of a corporation, voting and non-voting, could, under the regime urged by the estate here, arbitrarily slice the voting share block so thinly as to deny attribution of a control premium to any resulting block. Similarly, as the Ahmanson court noted, under the theory professed by the estate here, a testator with two equally valuable pieces of real property could designate equal undivided shares in each to two separate beneficiaries. The resultant valuation of the estate would be diminished by an amount representing the combined discount attributed to undivided shares, even though the two recipients could later exchange their undivided shares and thus reverse the artificial division of the properties, a purely paper maneuver resulting in great loss to the treasury. As the Ahmanson court noted,
Estate planners would implement such a tax-avoidance scheme whenever at least one of the assets in the gross estate had a diminished value if divided among two or more beneficiaries. As there is nothing in either the language of the statute or the underlying theory of the estate tax that requires the existence of this loophole, we shall not impute it to Congress.
Ahmanson, 674 F.2d at 768. We likewise decline to permit such a speculative and manipulable division of property to serve as the basis for valuation.
Thirdly, permitting the estate's argued hypothetical scenario of separate disposition of voting and non-voting shares to form the basis of a proposed valuation would defy common sense and the requirements of the "fair market value" standard, at least in the present case, where the stock of a small, closely held corporation is at issue. The relevant Treasury regulations provide that "fair market value" is the "price at which the property would change hands between a willing buyer and a willing seller . . . both having reasonable knowledge of relevant facts." It is well established that the willing buyer-willing seller rule presumes that the potential transaction is to be analyzed from the viewpoint of a hypothetical buyer whose only goal is to maximize his advantage. See, e.g., Revenue Ruling 59-60, 1959-1 C.B. 237; Estate of Bright v. United States, 658 F.2d 999, 1006 (5th Cir. 1981). And it does not comport with common sense that a willing buyer would be likely to purchase non-voting shares in a small, family-held business, without concomitantly purchasing a controlling voting interest. Such a purchase would put the outside purchaser at the mercy of the voting insiders on matters such as dividend declaration and other important corporate policies, without affording, as in the case of most publicly-traded corporate stock, a ready "exit" remedy of disposing of the purchased stock, or the "voice" remedy of joining with voting non-insiders to protect the minority interest. In applying the willing buyer-willing seller rule, courts may not permit the positing of transactions which are unlikely and plainly contrary to the economic interest of a hypothetical buyer as a basis for the valuation. Thus, even apart from considerations of estate tax policy, there is logical reason to reject the estate's proposed separate fair market valuation of voting and non-voting stock.
On cross-examination, the estate's valuation witness testified in conclusory fashion as to the hypothetical existence of a willing buyer for non-voting shares in the instant case, but he was unresponsive to the government's query as to whether such a peculiarly risky investment could in fact ever attract a willing buyer; instead, the witness asserted woodenly that such a purchase is "exactly what you do when you buy General Motors or any other stock . . .", Tr. 258, thus eliding the crucial differences, noted in the text infra, between a publicly-held and a closely-held corporation.
Once it is determined that the property is to be valued as it exists in the hands of the estate rather than as it might exist if subsequently divided, the court's error in refusing the government's proffered instruction becomes clear. The sole reason for assigning a lesser value to the non-voting shares in this case is their lack of voting rights which would, the estate argues, make them less attractive to a prospective purchaser. But that defect disappears where, as here, the non-voting stock is an integral part of the larger estate which retains a controlling equity interest. Here, when viewed in the hands of the estate, the non-voting stock would simply not be subject to the disadvantages of an isolated non-voting interest.
Holders of non-voting stock in this case also did not enjoy the right to inspect the company's books and records. The estate, however, has at no point argued that this disability could form the basis for a valuation differential.
Like this court, the Ninth Circuit in Ahmanson rejected the argument, made by the estate here, that non-voting shares could be assigned a lesser value than controlling voting shares even where both comprised a single block, noting, "The record simply does not contain support for the proposition that non-voting shares are sold at a discount when sold together in a package with sufficient voting shares to give control." Ahmanson Foundation v. United States, 674 F.2d 761, 769 (9th Cir. 1981). The Ahmanson court went on to note the logical impossibility of such a discount, given the block purchaser's power to fully guard the interests of his non-voting stock through control of corporate policy. Id. In short, when viewed as part of the estate's integrated stock holdings, as they must be, the non-voting shares simply do not suffer any strategic disadvantage. To view the matter otherwise would be to permit an estate to arbitrarily divide even a one hundred percent block of voting stock into units so small as to avoid the attribution of a control premium to any unit. We do not think that the "fair market value" rule may be so absurdly and abstractly parsed.
The estate attempts to distinguish Ahmanson on two grounds. First, it argues, the estate in Ahmanson controlled all of the voting shares, rather than merely a majority, as here. This fact, however, does not affect the basic Ahmanson principle that no discount may properly attach to non-voting shares when considered as part of a package "with sufficient voting shares to give control." Ahmanson, 674 F.2d at 769. (Emphasis added). It is uncontested here that the estate's possession of fifty-three percent of the voting shares was sufficient to "give control."
Second, the estate argues that the Ahmanson court held specifically that the record did not contain support for the proposition that non-voting shares are sold at a discount when part of a package conferring voting control, while in the instant case, the estate's witness testified in support of this proposition. However, the estate offered no direct evidence on this point. Its chief witness on cross-examination expressed support for this theory in entirely conclusory and speculative terms and appeared to premise this support on a misapprehension as to the decedent-focused framework of valuation, despite the government's repeated attempts to educe an answer to the question which respected that framework. See Tr. 254-60, 280. A plaintiff, of course, is not entitled to inferences which rest on mere speculation. Carlson v. American Safety Equipment Corp., 528 F.2d 384, 386 (1st Cir. 1976).
Moreover, the following colloquy suggests that, when guided by the appropriate valuation standard, the estate's valuation witness conceded the logical impossibility of a discount for non-voting shares:
Q. . . . In the hands of Bernard L. Curry, his nonvoting stock cannot be adversely affected for purpose of dividends because he owned voting control, isn't that correct?
A. I didn't appraise the stock in his hands. But to continue your question, I would presume that would be correct, in his hands.
Q. Okay. And in his hands —
A. He wouldn't do anything against himself, what we are saying.
Tr. 254-55 (emphasis added).
Finally, even if the estate had arguably presented some specific record evidence on this point, we would agree with the Ahmanson court's alternative holding that a finding of a value differential between the two classes of shares when situated in a control bundle is illogical and impermissible as a matter of estate tax law and policy. See Ahmanson, 674 F.2d at 761.
The estate cites Korslin v. United States, 31 A.F.T.R.2d (P-H) 1390 (E.D.Wis. 1973), to buttress its contention that a lesser valuation for non-voting shares is permissible, even when those shares are part of a larger block including voting shares. Korslin is, however, distinguishable, for in that case the voting shares comprised only a minority interest. Id. at 1390.
In sum, we hold that the district court erred in refusing to instruct the jury that the non-voting and voting shares held in decedent's estate were of equal value. Since the jury found, without the benefit of this instruction, a nearly $100 differential between the two classes of shares, the court's failure to give the instruction was obviously prejudicial to the government's interest and would alone warrant a new trial.
B. The Liquidation Value Instructions
The government also maintains that the district court erred in refusing to tender its proffered instruction which stated:
You are instructed that the liquidation value of this stock is the minimum value that the decedent's controlling interest in the company could be worth, because it is the amount he could realize for himself if he chose to liquidate the company.
The government argues that a rational seller in possession of a controlling interest, under no compulsion to sell his stock, would in no case consent to sell his stock for an amount less than its liquidation value; instead, the government argues, the rational seller would himself liquidate the stock to realize a value higher than could be obtained through such a sale. The government contends that because here the estate conceded that the liquidation value of the stock was $191 per share, but the jury ultimately found lesser values of $150 and $56.25 for the stock (figures obtained by the estate's averaging of liquidation value and a pessimistic, lower going concern value), the court's failure to instruct the jury as requested caused prejudice to the government's interest.
The chief problem with this argument is its assumption that the controlling seller, or a party to whom he sold his interest, could automatically liquidate the company to realize its asset value, unconstrained by the rigorous fiduciary duties which attach to possession of a controlling equity interest. While it is true that Indiana law permits a majority interest to effect a liquidation of a corporation, see Indiana Code Ann. §§ 23-1-6-3, 23-1-7-1(b)(2) (Burns 1972); Gabhart v. Gabhart, 267 Ind. 370, 370 N.E.2d 345, 355 (1977), it is settled law that the power to cause such extraordinary corporate actions as dissolution or liquidation may not be exercised without scrupulous loyalty to the interests of minority shareholders. LeBold v. Inland Steel, 125 F.2d 369 (7th Cir. 1941); Von Hagke v. United States, 43 A.F.T.R.2d (P-H) 1310, 1316-17 (E.D.Wis. 1979); Henn, Law of Corporations § 240 (2d ed. 1970). Indeed, the fiduciary duty owed to the minority interest is even greater where, as here, the corporation is small, non-public and closely held. Donahue v. Rodd Electrotype Co., 367 Mass. 578, 328 N.E.2d 505 (1975). It would simply have been legal error to forbid the jury from concluding that the company's full liquidation value may not be realizable due to legal constraints on the power to liquidate.
While the estate's assigned liquidation value of $191 per share did reflect a twenty-five percent discount for "expenses of liquidation," its evidence did not suggest that this discount included an adjustment for the fiduciary constraints upon liquidation. In explaining the components of this discount, the estate's lead valuation witness noted, "You have got to run the company for this six months, and you are going to have the overhead items that continue. We made that adjustment." Tr. 225.
At least one other court has followed precisely this reasoning in rejecting arguments that liquidation value serves as a benchmark minimum in valuation of stock for tax purposes. In Von Hagke v. United States, 43 A.F.T.R.2d (P-H) 1310 (E.D.Wis. 1979), the court noted that a liquidation following from redemption might result in differential tax consequences to controlling and minority interests, such as enhanced capital gains liability for low-basis minority shares, and that liquidation effects a "squeeze-out" of minority shareholders — both consequences which could give rise to vexing lawsuits founded on breach of fiduciary duty. 43 A.F.T.R.2d at 1317. As a result, the court rejected the government's argument for a liquidation-value minimum, holding instead that legal constraints are properly considered in valuation of the stock. Likewise, we decline to impose a legal minimum on fair market value which slights practical legal realities.
The government has cited Rothgery v. United States, 475 F.2d 591 (U.S.Ct.Cl. 1973), in support of its argument that liquidation value serves as a legal floor for stock valuation in tax cases. Rothgery, however, held merely that the peculiar evidence in that case showed that liquidation value could have been recognized because the "value of an automobile dealership, such as the corporation, is closely related — and generally corresponds — to the value of its underlying assets." Id. at 594. Significantly, the Rothgery court does not address the possible economic impact of fiduciary constraints upon liquidation, and does not appear to have had the existence of such constraints brought to its attention.
In addition to protesting the omission of its minimum value instruction, the government also argues that the district court erred in instructing the jury that "[i]f you find plaintiff's corporation was prosperous and the idea of liquidating it was remote, then you should give stronger emphasis on [sic] earning power and payment of dividends." We agree that this instruction did not conform to the rule of fair market valuation, and should not be given at the new trial.
The relevant Treasury Regulations define fair market value objectively as "the price at which the property would change hands between a willing buyer and a willing seller, neither being under any compulsion to buy or to sell and both having reasonable knowledge of relevant facts." Treas. Reg. § 20.2031-1(b). This requirement has been interpreted repeatedly by the Internal Revenue Service and the courts as requiring that the willing seller be posited as a "hypothetical" seller and not the particular plaintiff seeking recovery. Revenue Ruling 59-60, 1959-1 C.B. 237, modified by Rev.Rul. 65-193, 1965-2 C.B. 370; Estate of Bright v. United States, 658 F.2d 999, 1006 (5th Cir. 1981); United States v. Simmons, 346 F.2d 213, 217 (5th Cir. 1965); Rothgery v. United States, 475 F.2d 591, 594 (Ct.Cl. 1973). In short, the sole relevant consideration in determining whether the company here should be valued as liquidated or as a going concern is which alternative could be expected to yield the profit-maximizing result. Accordingly, the subjective intention or "idea" of this particular plaintiff corporation with regard to imminent liquidation is irrelevant to that determination. To hold otherwise would be to command future juries to wade into the thicket of personal corporate idiosyncrasies and non-market motives as part of their valuation quest, thus doing great damage to the uniformity, stability, and predictability of tax law administration. Estate of Bright v. United States, 658 F.2d at 1006. It is difficult to determine whether this instruction alone created prejudice to the government's interests sufficient to amount to reversible error in this case, but, as a new trial has been ordered for reasons set forth in II. A., supra, we admonish the district court to omit this instruction at the close of the new proceedings.
C. The Stock Purchase Agreement Instructions
The government finally argues that it was error for the district court to instruct the jury that the existence of the "right of first refusal" type stock purchase restriction contained in the company's articles of incorporation could be considered in valuing the stock. That restriction provided,
No shareholder may sell his stock whether it be voting common stock or Class A non-voting common stock to any person, firm or corporation without first offering such stock to the corporation for purchase at book value and in the event the corporation cannot or does not make such purchase then no shareholder may sell his said stock to any person, firm or corporation without first offering said stock to the other shareholders for purchase prorata according to their respective holdings, or in such proportions as they may agree upon, at book value.
The court instructed the jury,
The unrestricted power to dispose of stock, particularly to sell it to the highest bidder, is a material incident of its ownership. When this power is curtailed as the Curry stock is by the Articles of Incorporation, the market value will ordinarily reflect these restrictions. While there is no set formula for reflecting the existence of sales restrictions, those restrictions are one of the factors you should consider in determining fair market value.
The clear implication of this instruction was that these restrictions could lower the market price of the stock.
The government first argues that this instruction was in error because a hypothetical purchaser of the decedent's controlling stock could simply remove the sale restrictions through the exercise of his majority voting power, and thus they could have no impact on the market price of the stock. However, the government's argument, like its argument for the liquidation value minimum discussed in II. B., supra, is not persuasive, for it wrongly assumes that the possession of a controlling block of stock confers upon the owner the unfettered power to effect such a dramatic change. But such a power, especially where it has a differential impact on the rights of non-controlling shareholders in a close corporation, is significantly hedged by the constraints of fiduciary duty. LeBold v. Inland Steel, 125 F.2d 369 (7th Cir. 1941); Von Hagke v. United States, 43 A.F.T.R.2d (P-H) 1310, 1316-17 (E.D.Wis. 1979); Henn, Law of Corporations, § 240 (2d ed. 1970). Even though such a duty would not necessarily prevent the exercise of voting power to remove the sale restriction, it may require that the classes of shareholders deprived of the full benefits of this restriction be alternatively compensated. It therefore cannot be assumed that, in the typical case, a stock sale restriction could exert no downward influence upon the market price of a stock package in possession of voting control.
The government next argues that, even if the controlling shareholder could not eliminate this stock purchase restriction or could do so only at great expense, the restriction could still have no depressant effect on the market price of the stock because, at the time of trial, the price at which the restriction allowed the stock to be purchased — its book value — exceeded its market value from either the estate's or the government's point of view. Sheer economic logic dictates that there would be no takers for such a book value offer, the government argues, and consequently a hypothetical purchaser of decedent's stock would not take such a contingency into consideration in valuing the stock.
This argument, while arithmetically appealing at first glance, does not survive economic analysis. The fact that, at the time of trial, the book value-exceeded the market value of the stock, thus rendering the threat of exercise of the stock purchase provision temporarily nugatory, does not mean that the existence of the stock purchase provision would never have a practical restrictive effect. A rational purchaser of decedent's stock might consider the possibility that at a future date, the book value of the stock might fall or its market price rise, or both, thus making the corporation's or shareholders' exercise of their rights under the provision economically advantageous. A hypothetical buyer might well, then, discount the value of the stock to account for this future contingency. Therefore, we hold, the jury was properly allowed to consider the contingent impact of the stock purchase restriction in its valuation assessment.
Viewed from the government's perspective, this reversal of the relative size of book and market values could not at trial have been considered at all unlikely. The government placed a value of $290.50 per share on the stock, while book value was just over $300 per share.
D. Attorney's Fees
Because the Equal Access to Justice Act, 28 U.S.C. § 2412(d)(1)(A), provides that a court shall award fees and costs only to a "prevailing party," our decision to vacate the district court judgment also necessitates our reversal of that court's award of fees and costs to the estate.
For the foregoing reasons, the judgment of the district court is hereby vacated, its award of fees and costs to the estate is reversed, and the case is hereby remanded for a new trial.
VACATED AND REVERSED.
The government, having lost on all fronts below, argues on this appeal that the district court erred in giving two instructions offered by the appellee and in refusing to give two others that it itself offered. It also claims that the district court erred in awarding $39,851.72 in attorney's fees to the appellee pursuant to the Equal Access to Justice Act, 28 U.S.C. § 2412.
The entire dispute in this case centers on one simple question: What was the value of the stock that Bernard L. Curry owned in B.L. Curry Sons, Inc., when he died on October 20, 1977 at the age of 85? That Mr. Curry died over five years ago and the question remains unanswered does not speak well for the "system". Because the mandate of this court directs a new trial, it almost certainly assures that the question will remain unanswered for a year or two more. Given the circumstances of this case, the nature of the errors found by the majority and the undesirable prospect of having to replay this entire production one more time, I dissent from the judgment of the court.
In a nutshell, Judge Wood's opinion analyzes the four questioned instructions — two that were given and two that were rejected — and holds:
1. That the district court did not err in refusing to give the government's requested instruction regarding liquidation value of the Curry stock.
2. That the district court did not err when it instructed the jury that the existence of a stock purchase restriction (a "right of first refusal") could be considered in valuing the stock.
3. That the district court erred when it told the jury it should put a stronger emphasis on the company's earning power and ability to pay dividends if it concluded that the business was prosperous and the chances of liquidating it remote.
4. That the district court erred when it refused to instruct the jury that Mr. Curry's voting and non-voting stock were of equal worth because, at the time of his death, he had voting control of the company.
Thus, the majority has concluded that the district court was both right and wrong on the four questioned instructions. It properly refused one, properly gave one, improperly refused one and improperly gave one. As to the two it was wrong on, only one, the failure to give the equivalency instructions, is deemed to be prejudicial error requiring that the case be retried.
I have no quarrel with Judge Wood's eloquent discussion of the four questioned instructions. I part company with him only on the mandate ordering a new trial. His decision, at page 14, notes that "Since the jury found, without the benefit of this instruction, a nearly $100 differential between the two classes of shares, the court's failure to give the instruction was obviously prejudicial to the government's interest and would alone warrant a new trial." I agree that it was error to decline to give the requested equivalency instruction, but I believe the error can be corrected without ordering a new trial.
Abundant evidence in the record supports the estate's contention that the voting stock had a value of $150 per share. Halsey Sandford, the estate's primary valuation witness, testified that since there was no established market for the company's stock (which was not publicly traded), he valued Curry's stock on the basis of two primary factors, i.e. — the projected earning power of the company as a going concern, and the liquidated value of the company's assets — with the emphasis given to the company's potential earning power. Sandford testified that, based upon a comparison of the earnings history of B.L. Curry and Sons, Inc., with those of comparable, publicly traded companies, it was his opinion that the company's stock would sell for approximately $125 per share if publicly traded. He then discounted that figure by 50 percent to reflect the lack of marketability of the company's stock to arrive at an intrinsic value of $62.50 per share.
To the $62.50 per share intrinsic value, Sandford added a 60 percent control premium ($37.50 per share) to arrive at a per-share value for the voting stock of $100 based upon a "going business approach". Next, Sandford analyzed the liquidation value of the voting stock. He testified that the net value of the assets of the company, after allowing for projected liquidation expenses, was $1,531,000 (or about $255 per share) on the date of Curry's death. He further discounted this value by 25 percent to compensate for the risks involved in liquidation. Accordingly, he concluded that the liquidation value of the Curry voting stock was $191 per share. He then averaged the "going concern" and the "liquidation" values computed for the stock to arrive at a final value of $150 per share for Curry's 800 shares of voting stock. I do not believe that the failure to give the requested equivalency instruction impacts on this portion of Sandford's testimony. The only real question, one that I would answer "no", is whether the failure to give the requested instruction caused the jury to value the nonvoting stock at too low a price.
The "nearly $100 differential" that Judge Wood mentions is, in my judgment, as logically traceable to the direct action of the government as it is to the failure to give the requested equivalency instruction. On the tax return filed for Curry's estate, the executor valued the voting stock at $169.14 per share and the nonvoting stock at $18.79 per share. On audit, the IRS determined that the value of Curry's voting stock was $440 per share and the value of his nonvoting stock was $300 per share. Why is it not logical to conclude that the cause of the ". . . nearly $100 differential" is here, in the government's basket, rather than in the instructions? While it is interesting to note that the government's position on audit (different values for voting and nonvoting shares) is not consistent with the argument it now makes, I pass the inconsistency but note that its position may have invited the result that the jury reached.
Rather than order a new trial here, I would find as a matter of law, i.e., Ahmanson, supra, and this court's decision, that the value of Curry's voting and nonvoting stock was the same. On the basis of this record, it is clear that the value of the nonvoting stock should have been found to be $150 per share, the same value found for the voting stock. Accordingly, I would remand the case to the district court with instructions to modify the judgment consistent with this opinion.
Were this approach to be followed, the estate would arguably, although it is a close question, still be the "prevailing party" in the case. Thus, the propriety of the award of attorney's fees would be ripe for review. Were I to review that issue I would assume that the estate was the prevailing party but I would find, as a matter of law, that the government's position was "substantially justified". Accordingly, I would reverse the award of attorney's fees. I would maintain that position because the valuation of stock in a small, closely held corporation is not an issue that can be resolved with arithmetic precision. Many judgmental factors must be brought to bear on the question. It is a point, it goes without saying, upon which reasonable minds can disagree. The government is "substantially justified" in litigating such an issue in most any case including this one.