MEMORANDUM FINDINGS OF FACT AND OPINION
WILLIAMS, Judge: The Commissioner determined a deficiency in petitioner's Federal estate tax in the amount of $ 1,194,949.31. The issue we must decide is the value for Federal estate tax purposes of shares of stock in a closely-held corporation. Our decision has turned on how poorly petitioner and respondent have carried their respective burdens of proof.
Respondent's determination of a deficiency in estate tax is presumptively correct. Thus, petitioner has the burden of proving that respondent's valuation of Bell Oil stock is erroneous. Welch v. Helvering,290 U.S. 111 (1933); Rule 142(a), Tax Court Rules of Practice and Procedure. Respondent, on the other hand, must prove the correctness of his increased deficiency claimed in his amended answer. Rule 142(a), Tax Court Rules of Practice and Procedure. Based on the record before us, we conclude that neither petitioner nor respondent has satisfied the burden of proof.
Some of the facts have been stipulated and are so found. Norman L. Bell died on July 12, 1981 a resident of Jefferson County, Texas. Shirley Bell McBride was appointed as executrix.
At the time of his death, decedent owned 16,825 shares of stock in Bell Oil of Texas, Inc. (hereinafter "Bell-Tex"), and 4,510 shares of stock in Bell Oil of Louisiana, Inc. (hereinafter "Bell-La"). No other shares of either company were outstanding. The decedent was survived by a wife and left a community estate. The shares of stock in the Bell Oil corporation were, however, decedent's separate property.
During 1981, the Bell Oil companies owned and operated 21 Bell's convenience stores, including several hardware stores, in eastern Texas and western Louisiana. Some of the stores operated "washaterias," and each of the stores retailed gasoline and oil.
Petitioner filed a Federal estate tax return on April 13, 1982. As reflected on Schedule B of the estate tax return, the 16,825 shares in Bell-Tex were valued at $ 452,592.50; the 4,510 shares in Bell-La were valued at $ 470,167.50. Petitioner's valuation was based on multiplying earnings by assuming a price/earnings ratio of 5:1. Petitioner did not offer evidence to support its use of this price-earnings multiple to value the stock or any discount for lack of marketability of the stock. While it is generally recognized that the value of stock in closely-held corporations is subject to a discount for lack of marketability, petitioner must establish (1) the value to be discounted and (2) the amount of the discount. See, e.g., Ward v. Commissioner,87 T.C. 78 (1986); Estate of Andrews v. Commissioner,79 T.C. 938, 953 (1982). Both points are a matter of proof.
The only evidence offered by petitioner to substantiate its valuation of Bell-Tex and Bell-La was the testimony of Shirley Bell Marshall, the daughter of decedent. While she was a credible witness, the scope of her testimony was limited. The principal subject of her testimony was the effect on operations of the death of Norman L. Bell, founder and key employee. The loss of decedent's services decreased the value of the companies, and a knowledgeable individual who could assume supervisory responsibility over all operations had to be found and hired. Mrs. Marshall also testified that a significant number of improvements had to be made to the stores. While improvements prevented deterioration of the stores, petitioner failed to prove that the stores were dilapidated or in serious need of restoration. Further, this testimony lacked a starting point for valuation. Although we accept this testimony, petitioner's valuation of Bell-Tex at $ 452,592.50 and of Bell-La at $ 470,167.50 was asserted summarily and was not supported by the detail necessary to carry the burden of proof.
Respondent valued the decedent's interest in Bell-Tex at $ 1,000,000.00 and in Bell-La at $ 900,000.00. Respondent amended his answer to claim the value of Bell-Tex was $ 2,100,000.00 and of Bell-La was $ 1,136,000.00, based on a seven volume appraisal report by James T. Johnson. Johnson was called as an expert witness to testify concerning his appraisal of the closely-held Bell Oil corporations.
In his seven volume appraisal of Bell Oil, Johnson utilized four methods of valuation: (1) the capitalization of the income stream method, (2) the excess earnings method, (3) the tangible net worth metod, and (4) a ratio of "value of stock to sales," a ratio concocted by Johnson that was little more than speculation masquerading as empirical analysis.
A primary basis for Johnson's analysis was industry data appearing in an article by John F. Rosco, "The Annual Dollars Per Day Survey of the Small Food Store Industry." From this publication, the expert developed capitalization rates and industry ratios which he used in appraising petitioner's corporations. 1 No testimony or evidence supported the authority of the information in the Rosco article. Indeed, another article identified by petitioner reported different financial information for the same companies for the same periods. The subjects of the articles were all publicly traded companies, but no one compared this conflicting information with annual reports filed with the Securities and Exchange Commission. Moreover, no relationship was shown to exist between the performances of these companies and either Bell-Tex or Bell-La. The expert did not know the differences or similarities between these companies and the Bell Oil companies and thus did not analyze any differences or similarities in his valuation report. This lack of careful analysis undermines our confidence in Johnson's report and in the accuracy of the fundamental information supporting the expert's conclusions.
Johnson's method for choosing a capitalization rate in both the capitalization of income stream method and the excess earnings method was arbitrary. After listing the range and mean average of ratios found in the article of doubtful authority, Johnson chose his capitalization rate of 11 percent for the capitalization of the income stream method without any explanation. In determining the capitalization rate for the excess earnings method, Johnson applied different rates of return to income assumed to be earned by three categories of assets and long-term debt 2 on the balance sheet of each corporation: (1) 12 percent for working capital (2) 12 percent for equipment and improvements, (3) 10 percent for land and 13 percent for long-term debt. He assumed, moreover, that an investor would desire a 26 percent rate of return on "good will." He chose these rates of return without explaining how he arrived at these rates or why the rates vary.
The mean average of U.S. Treasury bonds and notes at the time was 12.33 percent according to the expert. This represents the rate of return an investor would expect on risk-free, intermediate and long-term investments generating taxable income. The expert gave no explanation for why the rates applied in his excess earnings method were lower than the U.S. Treasury bond and note rates despite the comparatively higher risk that investment in Bell companies' stock presented. The expert's arbitrary choice of rates of return was neither accurate nor realistic.
The inherent flaws in Johnson's methodology are emphasized by the way in which the expert arrived at his bottom line valuation under the excess earnings method. For some unexplained reason the expert added the value of excess earnings, determined using unrealistic rates of return, to the value of the Bell Oil companies as determined under the expert's tangible net worth method. The expert gave no rationale for this adjustment. It would appear that this method, as employed by Johnson, results in determining a premium or penalty adjustment to tangible net worth and is not an independent valuation method. If so, no premium exists for either company and a significant penalty should be applied. It is apparent from the actual earnings of Bell-Tex and Bell-La that the corporations were earning a return on their assets that fell well below market yield. Consequently, we believe that there were no "excess" earnings.
Furthermore, the rates of return applied by Johnson in the excess earnings method bore no relationship to the capitalization rate Johnson used in the capitalization of income stream method. We believe his choice of varying rates indicates a result-oriented analysis. An appropriate capitalization rate is determined by the comparable investment yield in the market not by the choice of a valuation method. Johnson made little effort to identify comparable investments.
In the tangible net worth method the appraiser valued each asset as if sold individually. While this valuation approach is of interest we do not find it persuasive of the value of the stock in an ongoing, closely-held business.
In figuring the ratio of value to sales, Johnson, relying on the industry data found in "The Annual Dollars Per Day Survey of the Small Food Industry," divided the gross sales by total stock value of each convenience food store company listed in the article. This produced a list of ratios found on Table III and IV of the expert's summary report from which Johnson somehow derived a multiplier of .20 which he applied to the gross sales of each Bell Oil company. This produced a value of $ 2,100,000.00 for Bell-Tex and $ 760,000.00 for Bell-La.3 Johnson admitted that he "eyeballed" a list of gross sales/stock ratios found in his report to arrive at his multiplier of 20 percent which he applied to the gross sales of each Bell Oil company. Not only was this unique method based on arbitrary choice (which Johnson could not articulately defend), the method itself was not reasoned.
Analyzing the results of the valuation methods used, Johnson concluded that the value of the Texas company was $ 2,100,000.00 and the value of the Louisiana company was $ 1,136,000.00.
Aside from arbitrary and result-oriented analysis, Johnson's report had many errors. During the trial the Court found that a whole column of numbers was misplaced in the report and wrongly used by the expert. The stores owned by Bell Oil companies had to be reconsidered during the trial to determine the correct number of them. The Court also had to refigure the breakdown of gas sales and food sales for the Bell Oil corporations because of the expert's inaccuracy. Much time was used to correct the mistakes in the report. Consequently, we have no confidence in Johnson's report. We conclude, therefore, respondent has failed to carry his burden of proving the increased deficiency in the amendment to his answer.
Accordingly and because of concessions, we hold that respondent's original determination of deficiency must stand. The appropriate value to be applied to the Bell Oil company stock is $ 1,000,000.00 for the Texas company and $ 900,000.00 for the Louisiana company.
Decision will be entered for the deficiency originally determined.