OPINION
Whitaker, Judge-.
Respondent determined a deficiency of $160,981.67 in the Federal estate tax of petitioner. The sole issue for decision is the date-of-death fair market value of shares of stock held by decedent in four closely held corporations. Since this case is largely factual, we have combined our findings of fact with our opinion.
Some of the facts have been stipulated and are so found. The parties have stipulated that Woodbury G. Andrews (hereinafter referred to as the decedent) was a resident of Excelsior, Minn., when he died testate on May 16, 1975. It has been further stipulated that Woodbury H. Andrews, a son of decedent, was appointed executor of the estate of decedent and resided in Minneapolis, Minn., when the petition in this case was filed.
Among the assets listed in the estate tax return of decedent were his interests in the following four closely held corporations: (1) 54 shares of W.F. & H.H. Andrews Co. (W.F. & H.H.), (2) 100 shares of St. Anthony Holding Co. (St. Anthony), (3) 50 shares of Green Mountain Investment Co. (Green Mountain), and (4) 63 shares of Andrews, Inc. Decedent owned approximately 20 percent of the total outstanding shares of each of the four corporations at the time of his death,1 with the remainder being owned by his four siblings in approximately equal proportions. The stock had been held by these five individuals since their father died in 1945. Decedent worked with these corporations from approximately 1927 until his death in 1975. His two brothers have been involved with the corporations since the early 1930’s. Together, decedent and his two brothers constituted the entire management of all four corporations; the two sisters did not actively participate in management. There is no evidence of any significant internal management disputes or family discord.
As of the date of death, W.F. & H.H. had been in business 73 years; St. Anthony, 71 years; Green Mountain, 66 years; and Andrews, Inc., 53 years. All four corporations have been involved primarily in the ownership, operation, and management of commercial real estate properties, although they also held some liquid assets such as stocks, bonds, and cash.2 The real estate holdings included warehouses, apartment buildings, factories, offices, and retail stores in the Minneapolis - St. Paul metropolitan area. Many of the properties were in rundown urban areas, and most of the buildings were quite old, having been acquired during the early years of the corporations’ operations. Most of the properties were leased to small tenants under leases for periods of less than 5 years.
To handle their management and maintenance responsibilities, the corporations together employed approximately 22 persons in addition to the three Andrews brothers. Fourteen persons were listed as employed and paid by Andrews, Inc.; two by Green Mountain; and three, each, by St. Anthony and W.F. & H.H. However, the record discloses that many of the employees assigned to Andrews, Inc., performed services for all the corporations, which were billed on a monthly basis for their allocable shares of employee payroll and related charges. The types of employees included janitors, parking lot attendants, night watchmen, office personnel, and maintenance workers, who handled plumbing, carpentry, painting, heating, and other maintenance and repair work.
On its estate tax return, petitioner valued decedent’s stock interest in Andrews, Inc., at $56,700; St. Anthony at $45,000; W.F. & H.H. at $12,690; and Green Mountain at $13,000. In his notice of deficiency, respondent determined that petitioner had significantly undervalued these stocks, and valued Andrews, Inc., at $517,608; St. Anthony at $287,400; W.F. & H.H. at $114,264; and Green Mountain at $93,650.
Fair market value has long been defined as the price at which 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. Sec. 20.2031-l(b), Estate Tax Regs.; United States v. Cartwright, 411 U.S. 546, 551 (1973). This is a question of fact, with the trier of fact having the duty to weigh all relevant evidence of value and to draw appropriate inferences. Hamm v. Commissioner, 325 F.2d 934, 938 (8th Cir. 1963), affg. a Memorandum Opinion of this Court.
In determining the value of unlisted stocks, actual arm’s-length sales of such stock in the normal course of business within a reasonable time before or after the valuation date are the best criteria of market value. Duncan Industries, Inc. v. Commissioner, 73 T.C. 266, 276 (1979). However, the stock of these four corporations has never been publicly traded, and there is no evidence of any sales of stock in these corporations at any time near the date of decedent’s death. In the absence of arm’s-length sales, the value of closely held stock must be determined indirectly by weighing the corporation’s net worth, prospective earning power, dividend-paying capacity, and other relevant factors. Estate of Leyman v. Commissioner, 40 T.C. 100, 119 (1963), remanded on other grounds 344 F.2d 763 (6th Cir. 1965); sec. 20.2031-2(f), Estate Tax Regs.3 These factors cannot be applied with mathematical precision. Rather, the weight to he given to each factor must be tailored to account for the particular facts of each case. See Messing v. Commissioner, 48 T.C. 502, 512 (1967).
Both parties relied upon experts’ valuations derived from analyses of intrinsic factors. However, because of fundamental differences in approach between respondent’s and petitioner’s experts, particularly with respect to the weight to be placed upon net asset value as opposed to earnings or dividend-paying capacity, the amounts arrived at in the valuations were extremely far apart. The following chart, which was used on the estate tax return, lists the different values arrived at by petitioner’s primary expert, Sigurd Wendin; petitioner’s second expert, Orville Lefko; and respondent’s expert, Edward Bard:
Andrews, Green
Inc. St. Anthony W.F. & H.H. Mountain
Mr. Wendin $56,700 $45,000 $12,690 $13,000
Mr. Lefko 55,755 46,500 12,204 12,250
Mr. Bard 570,843 260,000 113,832 97,800
Respondent’s Valuations
Respondent’s first witness, James M. McKenzie, performed an appraisal of the assets held by the corporations. He valued each real property using three commonly accepted approaches to valuation — comparable sales, replacement costs, and income-producing capacity. After correlating the values found under each of these approaches, he arrived at the following total values for the assets held by the corporations:
Andrews, Inc. Green St. Anthony W.F. & H.H. Mountain
Real estate $2,780,700 $1,239,800 $677,500 $645,500
Cash, stocks, and bonds 1,405,024 808,230 56,700 151,682
Miscellaneous assets 121,313 174,105 149,388
Total 4,307,037 2,222,135 883,588 797,182
Liabilities 116,493 2,667 696 515
Net asset value 4,190,544 2,219,468 882,892 796,667
Petitioner has not attacked Mr. McKenzie’s valuations of the underlying assets. However, the executor has argued that in arriving at overall net asset value, adjustments should have been made to reflect costs that would have been incurred if the corporations had liquidated all their real estate properties and placed them on the market at one time. The adjustments sought by petitioner are for blockage, capital gains tax to the seller, real estate commissions, and real estate taxes and special assessments constituting a lien against the real estate. We disagree with this argument. Both parties have agreed here that there was no reasonable prospect of liquidation, and the evidence clearly supports this finding. When liquidation is only speculative, the valuation of assets should not take these costs into account because it is unlikely they will ever be incurred. See Estate of Piper v. Commissioner, 72 T.C. 1062, 1087 (1979); Estate of Cruikshank v. Commissioner, 9 T.C. 162, 165 (1947); Estate of Huntington v. Commissioner, 36 B.T.A. 698 (1937).4
Respondent’s second expert, Edward Bard, a financial analyst employed by respondent, initially used three approaches to value the stock of each corporation: Earnings and dividend-paying capacity, linear regression analysis, and net asset valuation. The former two methods produced values far lower than valuation based on net asset values. For instance, the values he computed for Andrews, Inc., based upon capitalization of earnings and dividends ranged from $1.1 million to $1.5 million,5 the value based on linear regression analysis was $1.6 million but the value based on net asset value was $4.1 million.6 He concluded that the former two values should be disregarded because they were unrealistically low, even though his computations did show that the corporations’ yields on investment had consistently been exceptionally low.7 Thus, his starting point for valuing each of these corporations was the net asset values derived from Mr. McKenzie’s appraisals. Because he saw the corporations involved here as holding passive items, he believed they were comparable to closed end mutual funds, which are also corporate entities holding investment properties. He found that shares of mutual funds generally traded at a 25-percent discount from the value of the assets held by such funds, and accordingly, applied a 25-percent discount to the net asset values of the corporations. To this amount, he then applied a lack of marketability discount, varying from 11.9 percent to 20.74 percent to reflect each corporation’s costs of flotation if the stock were to be publicly offered.8
Respondent argues that Mr. Bard was correct in ignoring earnings and dividend-paying capacity because all four corporations should be classified as investment companies not involved in the active operation of a business. He believes that a willing buyer would recognize that the value of his investment would be in his right to share in the value of the corporations’ underlying net assets, regardless of their low earnings and dividend-paying capacity. Petitioner attacks respondent’s characterization of these corporations as investment companies, stressing that the corporations employed 25 employees and actively performed real estate management and maintenance functions.9 Characterizing these corporations as operating companies, petitioner believes greater weight should be placed on earnings and dividend-paying capacity than on net asset values.
We believe, however, the corporations here cannot be characterized for valuation purposes as solely investment companies or solely operating companies. The cases cited by respondent,10 which involved corporations holding only cash, commercial paper, or marketable securities, are readily distinguishable on the ground that the corporations involved here were actively engaged in the real estate management business. But cases dealing with corporations owning factories and other industrial or commercial operations are also not directly on point.11 Unlike many industrial companies, where the value of the manufacturing equipment and plant is tied to the nature of the manufacturing operation, here, the value of the underlying real estate will retain most of its inherent value even if the corporation is not efficient in securing a stream of rental income. It seems reasonable to assume, as we did in Estate of Heckscher v. Commissioner, 63 T.C. 485, 493 (1975), that a—
potential buyer would have to forego some current return on his investment in exchange for an interest in a net worth much more valuable than the price he pays for the stock, and the seller would have to be willing to part with an equity interest in the company for much less than its indicated value in return for something that would produce a greater yield on his investment. * * *
Furthermore, regardless of whether the corporation is seen as primarily an operating company, as opposed to an investment company, courts should not restrict consideration to only one approach to valuation, such as capitalization of earnings or net asset values. See Hamm v. Commissioner, 325 F.2d 934 (8th Cir. 1963), affg. a Memorandum Opinion of this Court; Portland Manufacturing Co. v. Commissioner, 56 T.C. 58, 80 (1971); Estate of Schroeder v. Commissioner, 13 T.C. 259 (1949); Hooper v. Commissioner, 41 B.T.A. 114 (1940).12 Certainly, the degree to which the corporation is actively engaged in producing income rather than merely holding property for investment should influence the weight to be given to the values arrived at under the different approaches but it should not dictate the use of one approach to the exclusion of all others.
The regulations13 call for all relevant factors to be examined and, in a case such as this, we believe values arrived at under all the accepted valuation methods should be considered. We therefore believe respondent’s expert was incorrect to simply reject earnings and dividend-paying capacity valuations because they produced too low a result. Certainly, a prospective buyer would not so reject one particular type of valuation. A buyer of stock in these corporations would necessarily look to the earning capacity for part of his return on his investment. Nevertheless, this would not be the only factor that he would consider. Undoubtedly, he would also give substantial weight to each corporation’s underlying net asset value even though he would have no ability to directly realize this value by forcing liquidation.
Respondent argues that there is a particular reason why only net asset values should be used in this case. Mr. McKenzie’s determination of the values of the real estate properties held by the corporations was arrived at through a computational method that considered, among other factors, the projected earning capacity of the properties. Respondent contends that in valuing each corporation it is therefore appropriate to use only the underlying net asset value, which is the sum of the appraised real estate values plus the values of other assets held by the corporation, such as stocks and bonds, minus all the corporation’s liabilities.14 However, to do so would ignore the corporate entities. If the properties had been held directly by decedent, respondent would undoubtedly be correct, but here we are not faced with the question of finding a value for the properties themselves; rather, we must find the value of shares of the corporations that held the properties. The corporations are businesses, engaged in the maintenance and management of these real estate properties. Thus, some of the value attached to the corporations must be based upon the operating nature of the businesses, with attention paid to their earnings and dividend history, management, and prospects for growth.
Petitioner’s Valuations
We turn now to analyze petitioner’s experts’ valuations, which we also find to be markedly deficient, largely because of their adopting approaches that minimize the importance of the value of the underlying assets.
Petitioner’s first expert, Sigurd Wendin, selected as compa-rables four large publicly traded real estate management corporations, although he recognized that there were differences in size, financing, number of employees, and types of activities carried on. Based on his analysis of these publicly traded corporations, he assigned price earnings multiples to the corporations involved here of five for future earnings, four for current earnings, five for the average of the last 3 years’ earnings, and seven for the average earnings for the period 1970 through 1973. These multiples were somewhat lower than those obtained from the comparable companies because of the difficulty in marketing nonpublicly traded shares. He then applied these multiples to net operating income during the applicable periods, and applied a minority discount for lack of marketability of 40 percent based on his overall experience with the amount of such discounts commonly allowed.
For dividend capitalization purposes, Mr. Wendin assumed a 10-percent rate, based upon the fact that corporate bond yields were around 9 percent to 11 percent. He believed bond yields were better comparables than the yield rate on stocks because the assets involved here were frozen and stock yields had been quite variable during this time period. He applied this 10-percent rate to the last year dividend and to his estimate of the future dividend that could be expected. Again, he discounted these amounts by 40 percent for lack of marketability.
Mr. Wendin also looked at the corporate balance sheets showing net assets and liabilities of the corporations. He did not have Mr. McKenzie’s or any other appraisal of underlying assets so he looked at book values instead of net asset values. He reduced the book values to 80 percent based on the market-to-book-value ratio here found in analyzing his four "compara-bles.”15 To this he applied a 40-percent discount for lack of marketability.
In combining all of these factors to arrive at an overall valuation of each corporation, Mr. Wendin minimized the impact of the adjusted book value per share. He believed this approach was necessary because there were no plans to liquidate and because a minority stockholder would have had no ability to force liquidation. Listed below as an example are the weights given to each factor, the values per share, and the overall average value arrived at for Andrews, Inc., stock:
Amount Per share $
$4,500.00 Future dividend capitalization 900.00 U3
3,083.04 Last year dividend capitalization 770.76
3,501.00 Future earnings capitalization 1,167.00 CO
1,880.68 Last year earnings capitalization 940.34 <N
Per share Weight Amount
963.57 1.5 $1,445.36 Three-year average earnings capitalization
1.497.11 998.07 1970-73 year average earnings capitalization H
2.345.11 2,345.11 Book value H
18,252.30 00 r-I
Average 1,014.02
To this $1,014.02 value per share he applied an additional discount to reflect specific negative factors such as deterioration of areas, strategic location, rising inflation, taxes, future maintenance costs, and the ability to quickly adjust rentals. He ended up with an estimated value for Andrews, Inc., of $900 per share. The values for the other three companies were determined in equivalent manners. In no case was book value given more than a one-ninth weight in determining the current value.
We find Mr. Wendin’s valuations to be seriously flawed because they did not take into account each corporation’s net asset value, i.e., the total value of its assets less liabilities. Although he attempted to approximate net asset values by using adjusted book values,16 which were arrived at by reducing book values by a market-to-book ratio, the figures so computed bore no reasonable relation to actual net asset values. Mr. Wendin used a market-to-book ratio of 80 percent, but a comparison of book values to the appraised values determined by Mr. McKenzie shows the ratio was actually near 300 percent. Moreover, even if the adjusted book values had been close to actual net asset values, Mr. Wendin assured they would be of little importance in his overall computations by assigning insubstantial weights, varying from one-ninth to one-eighteenth, to them. As discussed earlier, a prospective buyer of stock in these corporations would undoubtedly give substantial weight to net asset values.
Another problem with the Wendin valuations was that he applied to his adjusted book values a 40-percent discount for lack of marketability and control. We believe this was improper because lack of control was also reflected (excessively) in the weighting of factors considered in computing an overall valuation.
Petitioner’s second expert, Mr. Orville G. Lefko, was retained to prepare an analysis after the McKenzie appraisals had been made available to petitioner. Mr. Lefko employed an approach somewhat different than Mr. Wendin but still minimized the importance of net asset values and arrived at figures quite close to those computed by Mr. Wendin.
Mr. Lefko felt that both net asset value and earnings capability were not very important in comparison to dividend projections for purposes of valuing corporations such as the ones at issue here. Although the dividend history of each of the corporations had been irregular, Mr. Lefko came to the conclusion that each corporation had the ability to pay a regular dividend. To arrive at a rate of dividend capitalization, he looked at seven equity real estate investment trusts, which he selected as comparables. These "comparables” yielded a 12-percent dividend capitalization rate, which he increased to between 15 and 17 percent because he believed that real estate investment trusts involved lower risks due to factors such as professional management, geographical diversity, and size. After applying the dividend capitalization rate to projected dividends, he applied a 45-percent discount for lack of marketability.
Mr. Lefko did not entirely dismiss the net asset value of the corporations but attempted to factor into his overall value a liquidation probability value, to reflect the buyer’s prospect of realizing on liquidation a share of the underlying asset values. First, he adjusted the appraised value of the corporation’s assets. He decreased the values of real properties by anticipated liquidation costs of stocks and bonds by handling charges, and receivables by a bad debt allowance. This gave him a net realized value per share of each corporation. Next, he projected when a purchaser of shares that had been held by decedent might reasonably expect to obtain his share of the underlying assets through liquidation. Because of the corporation’s long history of continuing operations and no prospect of imminent liquidation, Mr. Lefko believed there was no chance of liquidation within 5 years, a 5-percent chance in 10 years, a 10-percent chance in 15 years, a 35-percent chance in 20 years, and a 50-percent chance in 25 years. He multiplied the net realized value per share by each of these percentages and then discounted the resulting amounts because of the lost time value of money prior to the assumed future liquidation. He thereafter added together all the discounted amounts and applied a 45-percent illiquidity discount. In this manner, he computed a liquidation probability value for each corporation, equal to approximately 3 percent to 5 percent of the fair market value of the underlying assets. This liquidation probability value was added to the previously determined dividend capitalization value of each of the corporations.17
We find the liquidation probability approach to be highly speculative at best. Moreover, we believe the computation improperly reduced asset values on account of lost use of money. This discounting for lost use of money is unrealistic because it fails to recognize that the underlying assets will themselves appreciate, most probably, at a rate similar to that applied as a discount. Another problem with Mr. Lefko’s valuation was his use of real estate investment trusts (REITs) as comparables. These entities are not truly comparable to the corporations involved here because of the requirement that REITs must pass through 90 percent of earnings as dividends and because REITs are generally highly leveraged while the corporations here were virtually debt free.
Petitioner’s third expert, Thomas C. O’Connell, based his valuation solely upon an approximation of cumulative future dividends of all the corporations. He started with the total average dividend for all the corporations over the 3-year period from 1972 through 1974. To this amount he applied a dividend capitalization rate of between 18.75 percent and 25 percent. He explained that this capitalization rate was arrived at based upon his belief that an investor would require a return of 150 to 200 percent of the prime rate because the companies had certain negative factors such as a low return on invested capital, a low rate of cash flow, very conservative management, fragmented management control, and a lack of consistent policy in prior years. He also stressed that there was no public market for the shares, a low return on net worth and a lack of any long-term plan or growth strategy. Applying the dividend capitalization rate to the prior 3-year total average dividend, he computed a value for the four corporations of between $605,000 and $807,000. For the 20 percent of the shares involved here, he estimated the value of $121,000 to $161,000 total.
We have accorded little weight to Mr. O’Connell’s valuation for several reasons. Most importantly, he did not consider earnings and took no account whatsoever of the net asset values of the corporations. Furthermore, in selecting a dividend capitalization rate, he did not examine comparable companies but simply expressed his relatively unsubstantiated opinion that an 18.75- to 25-percent rate of dividend capitalization would be desired by a willing purchaser. In the absence of more facts showing why this specific rate was chosen, it is difficult for the Court to determine whether the rate was simply an educated guess or whether he really had a substantial basis for concluding that it was appropriate. Another problem with Mr. O’Connell’s approach was that in applying the dividend capitalization rate to the average dividends received for the prior 3-year period, he did not differentiate among the corporations but considered them as if they were only one single entity. Thus, there is no way in which his valuation can be used directly to arrive at the value of any particular corporation involved here.
Lack of Control and Marketability Discounts
In Mr. Bard’s valuation report and in computing the amount of deficiency, the net asset values of the four corporations were each reduced first by a 25-percent discount based on a comparison with publicly traded closed end mutual fund shares, and then by discounts ranging from 11.9 percent to 20.74 percent based on Securities Exchange Commission flotation rates. These discounts were designed to reflect the shares’ restricted marketability and lack of control.
Rev. Rui. 81-253,1981-2 C.B. 187, was published by respondent after the statutory notice was issued, and it sets forth respondent’s current position concerning the allowance of minority discounts in valuing stock of closely held family corporations. Based on Rev. Rui. 81-253, respondent now argues that no discounts for lack of control or restricted marketability should be applied, although he has not asserted a deficiency greater than that asserted in the statutory notice. Thus, respondent argues that discounts should not be applied to the extent they would result in reducing the values below those asserted in the notice of deficiency.
Petitioner has questioned whether it is proper for respondent to now argue against allowing a minority discount even though he allowed a discount in computing the deficiency in the statutory notice. However, this is not a case where petitioner was surprised at trial by respondent’s introduction of a new issue. It is clear that petitioner was prepared for the minority discount question being raised at trial, and has devoted large portions of its briefs to rebutting respondent’s position that no minority discount should be allowed. Therefore, we find no merit to petitioner’s claim that the minority discount issue has been improperly raised by respondent. See Llorente v. Commissioner, 74 T.C. 260, 269 (1980), modified on other grounds 649 F.2d 152 (2d Cir. 1981).
Respondent argues that no discounts should be allowed because all the shareholders in the four corporations, including decedent, shared in control. According to respondent’s argument, the hypothetical "willing seller” used in arriving at valuation of the stock must be presumed to be one of the five family members, including decedent, who held stock in the corporations and shared an element of control. Respondent reasons further that such a willing seller would not have sold his shares except as part of the controlling family interest, unless to another family member or the corporations, themselves. If the shares were sold in this way, respondent contends they would retain their control value, and no minority discount would be justified.
In their arguments, neither petitioner nor respondent clearly focuses on the fact that two conceptually distinct discounts are involved here, one for lack of marketability and the other for lack of control.18 The minority shareholder discount is designed to reflect the decreased value of shares that do not convey control of a closely held corporation. The lack of marketability discount, on the other hand, is designed to reflect the fact that there is no ready market for shares in a closely held corporation. Although there may be some overlap between these two discounts in that lack of control may reduce marketability, it should be borne in mind that even controlling shares in a nonpublic corporation suffer from lack of marketability because of the absence of a ready private placement market and the fact that flotation costs would have to be incurred if the corporation were to publicly offer its stock. However, the distinction between the two discounts is not crucial for purposes of this case. Because respondent’s basis for opposing both discounts is the same — that the hypothetical willing buyer must be seen as a family member — our subsequent discussion, like the parties’ briefs, will consider the two discounts together.
The leading case dealing with this question is Estate of Bright v. United States, 658 F.2d 999 (5th Cir. 1981) (en banc), in which the court recognized that a minority interest in a corporation should not be seen as having any control value, even though the family unit had control of the corporation. The case involved whether a minority discount should be allowed in valuing a decedent’s undivided one-half interest in the control block of 55 percent of the stock of a corporation, where the other one-half interest in the control block was owned by the husband of the decedent (who was also executor of the estate and subsequently trustee of the testamentary trust that received the stock in issue). Two major bases for its decision were explained by the court. First, it found that established case law did not support any type of "family attribution” in which the control of the corporation was attributed among family members. The Fifth Circuit noted that the Tax Court, since at least 1940, has uniformly valued a decedent’s stock for estate tax purposes as a minority interest when the decedent, himself, owned less than 50 percent of the stock regardless of whether control of the corporation was in the decedent’s family.19 The court also cited two District Court opinions that took the same position,20 and it stated that it had found no estate tax cases supporting respondent’s position. The court found further that case authority in the analogous gift tax area also supported the taxpayer’s position, although not unanimously.21
The second major reason cited in the Estate of Bright opinion for applying a minority discount was based upon the concept of the hypothetical willing-buyer, willing-seller rule. Section 20.2031-l(b), Estate Tax Regs., sets forth the following rule, which has been universally applied by the courts and respondent:
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 relevant facts. * * *
The Fifth Circuit saw this language from the regulation as indicating that the "willing seller” is a hypothetical seller rather than the particular estate. Thus, the willing seller should not be identified with the decedent, and the decedent’s stock should not be included as part of a family unit for valuation purposes. Nor was it proper, according to the Fifth Circuit, to place any weight upon the identity of the parties that actually received the stock after distribution from the estate. For purposes of valuation, one should construct a hypothetical Sale from a hypothetical willing seller to a similarly hypothetical willing buyer.
In Propstra v. United States, 680 F.2d 1248 (9th Cir. 1982), the Ninth Circuit followed Estate of Bright and discounted the value of an undivided one-half interest in real estate held by the decedent and his wife as community property at the time of death. The Ninth Circuit noted that Congress has explicitly directed that family attribution or unity of ownership principles be applied in other areas of Federal taxation, and felt that in the absence of any legislative directive, it should not judicially require such principles to be applied in the estate tax area. Furthermore, the court emphasized the advantage of using an objective hypothetical willing-buyer, willing-seller standard, instead of a subjective inquiry into the feelings, attitudes, and anticipated behavior of heirs and legatees, which might well be boundless.
Respondent argues that the Fifth Circuit misinterpreted the concept of the hypothetical willing seller and willing buyer. In his briefs, he relies primarily upon three cases to support his proposition that a minority discount should not be applied if family members control a corporation. Richardson v. Commissioner, a Memorandum Opinion of this Court dated November 30, 1943, affd. 151 F.2d 102 (2d Cir. 1945), involved a family holding company whose assets were readily marketable securities. On appeal, the Second Circuit upheld the value established by us but questioned whether we had used the correct standard of valuation. Our Richardson opinion must be read narrowly in view of the long series of subsequent cases in which we have allowed discounts in valuing shares of family corporations that held operating as well as investment assets. Blanchard v. United States, 291 F. Supp. 348 (S.D. Iowa 1968), contains language supportive of respondent’s position. But the court stressed that it was dealing with a unique situation involving a planned sale of all the family’s shares of stock in the corporation. At first blush, Rothgery v. United States, 201 Ct. Cl. 183, 475 F.2d 591 (1973), seems to support respondent. But even though its valuation was phrased in terms of seeing the willing buyer as a particular person (the family member who already held almost 50 percent of the shares of the corporation), the court also explicitly found that there would have been other potential buyers for the decedent’s shares of the stock at the same price if the stock had been offered for sale to persons outside the Rothgery family. Thus, despite analyzing the hypothetical sale in terms of the sale to a particular person, the court evidently attributed no premium to the fact that the family member would assure himself of control by obtaining these shares.
Three cases provide at best weak support for respondent’s position that no discounts should be applied here. Opposed to this meager case law in favor of respondent is the large number of cases allowing such discounts, which were discussed in the Estate of Bright opinion. We see no reason to depart from such established precedent but follow the Fifth Circuit’s well-reasoned and thoroughly researched opinion. Respondent’s approach would have us tailor "hypothetical” so that the willing seller and willing buyer were seen as the particular persons who would most likely undertake the transaction. However, the case law and regulations require a truly hypothetical willing seller and willing buyer. We must assume these hypothetical parties exist even though the reality of the situation may be that the stock will most probably be sold to a particular party or type of person. Certainly, the hypothetical sale should not be constructed in a vacuum isolated from the actual facts that affect the value of the stock in the hands of decedent, but we do not see any actual facts in this case that require the stock to be valued as anything other than minority interests in each corporation.
Overall Valuation
We have pointed out above the defects of the approaches used by both parties’ experts. Having taken diametrically opposed views as to the factors to be weighed in valuing the stock involved in this case, the parties "failed successfully to conclude settlement negotiations — a process clearly more conducive to the proper disposition of disputes such as this.” Messing v. Commissioner, 48 T.C. 502, 512 (1967). Because we have concluded that none of the experts gave appropriate weight to net assets as well as earning and dividend-paying capacity, we must necessarily make our own best judgments of value. While we have considered the valuation reports of Messrs. Wendin, Lefko, and Bard, we have also weighed all other relevant factors such as the extremely conservative business attitudes and practices of the management, the nature of the real estate holdings, the amount of cash and other liquid assets held by the corporations, and the business climate on the valuation date, both overall and for these corporations in particular. We have discounted the values because of the restricted marketability of the shares and the lack of control a hypothetical willing buyer of these minority shares would be able to exercise. Based on all these factors and on the entire record, we conclude and find as a fact the following date-of-death fair market values for the stock held by decedent:
Andrews, Inc . $302,400
St. Anthony . 158,000
Green Mountain . 56,000
W.F. & H.H . 64,800
Decision will be entered under Rule 155.