In his notice of deficiency, respondent valued decedent's units at $12,469 a unit. Relying on Mr. Thomson's expert testimony, respondent now argues that the fair market value was at least $10,293 per unit, which we regard as a concession. See Estate of Hinz v. Commissioner, supra (finding that the Commissioner conceded his valuation position as set forth in the notice of deficiency by arguing for a lower fair market value on brief).

In general, a taxpayer bears the burden of proof. Rule 142(a)(1). However, section 7491(a) shifts the burden of proof to the Commissioner in certain situations if the taxpayer raises the issue, introduces credible evidence with respect to any factual issue relevant to ascertaining the proper tax liability, and demonstrates compliance with the applicable requirements of section 7491(a)(2).

Petitioner raised the issue of section 7491(a) in his posttrial brief. We need not address whether section 7491(a) applies because the parties have provided 2011 Tax Ct. Memo LEXIS 150">*160 sufficient evidence for us to find that the value of decedent's units as of the valuation date was $32,601,640. See Estate of Black v. Commissioner, 133 T.C. 340">133 T.C. 340, 133 T.C. 340">359 (2009).

Section 2001(a) imposes a tax on "the transfer of" a decedent's taxable estate, the value of which includes the fair market value "at the time of his death of all property, real or personal, tangible or intangible, wherever situated." Sec. 2031(a); United States v. Cartwright, 411 U.S. 546">411 U.S. 546, 411 U.S. 546">550-551, 93 S. Ct. 1713">93 S. Ct. 1713, 36 L. Ed. 2d 528">36 L. Ed. 2d 528 (1973). Fair market value is defined 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." Sec. 20.2031-1(b), Estate Tax Regs. The willing buyer is a hypothetical person; therefore, an actual buyer's personal characteristics are disregarded. Estate of Newhouse v. Commissioner, 94 T.C. 193">94 T.C. 193, 94 T.C. 193">218 (1990).

If the property to be valued is stock of a closely held corporation, its fair market value is best determined through "arm's-length sales near the valuation date of reasonable amounts of that stock". Estate of Noble v. Commissioner, T.C. Memo. 2005-2; accord Estate of Andrews v. Commissioner, 79 T.C. 938">79 T.C. 938, 79 T.C. 938">940 (1982). 2011 Tax Ct. Memo LEXIS 150">*161 If it is not possible to so value the stock, however, fair market value is calculated by "analyzing the value of publicly traded stock in comparable corporations engaged in the same or a similar line of business, as well as" by considering certain factors that an informed buyer and seller would consider. ⁴Estate of Noble v. Commissioner, supra.

The parties rely principally on expert testimony to establish the fair market value of decedent's units as of the valuation date. In addition to the expert testimony of Mr. Thomson, respondent also called as a witness Mr. Paxton, PMG's former chief financial officer 2011 Tax Ct. Memo LEXIS 150">*162 and current president and chief executive officer. Beginning in 1991, Mr. Paxton, as chief financial officer, authored PMG's annual valuation reports, which reported the fair market value of its stock or units as of that year. Although Mr. Paxton testified as to his July 12, 2004, valuation report, which formed the basis for the value reported in the return, petitioner requests that we adopt Mr. May's appraised value.

Valuation is a question of fact. 94 T.C. 193">Estate of Newhouse v. Commissioner, supra at 217. Courts often consider expert witnesses' opinions in deciding such cases; however, we are not bound by the opinion of any expert witness and may accept or reject such testimony in the exercise of our sound judgment. Helvering v. Natl. Grocery Co., 304 U.S. 282">304 U.S. 282, 304 U.S. 282">295, 58 S. Ct. 932">58 S. Ct. 932, 82 L. Ed. 1346">82 L. Ed. 1346, 1938-1 C.B. 279 (1938); 94 T.C. 193">Estate of Newhouse v. Commissioner, supra at 217. Although we may accept an expert's opinion in its entirety, we may instead select what portions of the opinion, if any, to accept. Parker v. Commissioner, 86 T.C. 547">86 T.C. 547, 86 T.C. 547">562 (1986); see Buffalo Tool & Die Manufacturing Co. v. Commissioner, 74 T.C. 441">74 T.C. 441, 74 T.C. 441">452 (1980). Because valuation involves an approximation, "the figure at which we arrive need not be directly traceable to 2011 Tax Ct. Memo LEXIS 150">*163 specific testimony if it is within the range of values that may be properly derived from consideration of all the evidence." Estate of Heck v. Commissioner, T.C. Memo. 2002-34.

The parties disagree over: (1) The date of financial information relevant to a date-of-death valuation of decedent's units, (2) the appropriate adjustments to PMG's historical financial statements, (3) the propriety of relying on a market-based valuation approach (specifically the guideline company method) in valuing the units, and, if appropriate, the proper manner of applying that method, (4) the application of the income approach (specifically the discounted cashflow valuation method), (5) the appropriate adjustments to PMG's enterprise value, and (6) the proper type and size of applicable discounts. ^{52011 Tax Ct. Memo LEXIS 150">*164 2011 Tax Ct. Memo LEXIS 150">*165}

Property includable in a decedent's gross estate is valued as of the valuation date "on the basis of market conditions and facts available on that date without regard to hindsight." Bergquist v. Commissioner, 131 T.C. 8">131 T.C. 8, 131 T.C. 8">17 (2008) (citing Estate of Gilford v. Commissioner, 88 T.C. 38">88 T.C. 38, 88 T.C. 38">52 (1987)). Subsequent events may be considered, however, "to the extent that such events may shed light upon a fact, circumstance, or factor as it existed on the valuation date." Gross v. Commissioner, T.C. Memo. 1999-254, affd. 272 F.3d 333">272 F.3d 333 (6th Cir. 2001).

Mr. Thomson bases his valuation analysis on data gathered from PMG's internally prepared financial statements ending June 27, 2004, and financial information for comparable public companies as of the quarter ending June 30, 2004. He considered that information to be more accurate than earlier data, despite the quarterly report's publication 1 or 2 months after the valuation date. In contrast, Mr. May's report relies upon financial information for comparable public companies ending March 28, 2004, the latest quarterly data available before the valuation date, and PMG's internally 2011 Tax Ct. Memo LEXIS 150">*166 prepared financial statements ending May 30, 2004, the latest statements published before the valuation date. Mr. May declined to use the June 2004 financial statements, stating that a willing buyer and willing seller would be unaware of the information as of the valuation date, since the statements likely would not have been closed and published by such date.

We agree with Mr. Thomson that the June 2004 financial information should be used in valuing decedent's units. Petitioner argues that the June information was not publicly available as of the valuation date, preventing a willing buyer and seller from relying upon it in determining fair market value. That is not to say, however, that our hypothetical actors could not make inquiries of PMG or of the guideline companies (or of financial analysts), which would have elicited non-publicly available information as to end-of-June conditions. Moreover, we understand Mr. Thomson's testimony to be that the June 2004 financial information accurately depicts the market conditions on the valuation date, not that a willing buyer and seller would have relied upon the data. Importantly, petitioner has not alleged an intervening event between the 2011 Tax Ct. Memo LEXIS 150">*167 valuation date and the publication of the June financial statements that would cause them to be incorrect. See Gross v. Commissioner, supra.Therefore, we shall rely on the June 30 and 27, 2004, financial information for comparable public companies and PMG, respectively.

Before commencing their valuation analyses, both Mr. Thomson and Mr. May subtracted items not expected to recur in the future (nonrecurring items) from PMG's historical financial statements to better represent the company's normal operations. The experts disagree only as to the appropriate number of adjustments.

Mr. Thomson made one adjustment to PMG's income statements, subtracting a $7,895,016 gain on divested newspapers in 2000. Mr. May made, among other adjustments, the following three adjustments to PMG's earnings: (1) Reduced PMG's 2000 earnings before interest, taxes, depreciation, and amortization (EBITDA) by a $7,900,000 gain on divested newspapers, (2) subtracted from both EBITDA and earnings before interest and taxes (EBIT) a 2003 $700,000 gain from a life insurance policy PMG inherited through an acquisition, and (3) subtracted from both EBITDA and EBIT 2011 Tax Ct. Memo LEXIS 150">*168 a 2003 $1,100,000 positive claim experience from PMG's self-insured health insurance.

Respondent objects to those three adjustments. We are unclear as to why respondent objects to the 2000 newspaper divestiture adjustment since his own expert, on whose appraisal he relies, made the same adjustment. We thus consider respondent to have acquiesced to the adjustment. However, we disregard Mr. May's self-insured health insurance and life insurance policy adjustments because he provides no explanation as to why the gains were nonrecurring. See Estate of Jung v. Commissioner, 101 T.C. 412">101 T.C. 412, 101 T.C. 412">450 (1993) (granting little weight to an expert's valuation report because it was "deficient in providing data and support for its conclusions").

Mr. May also made adjustments for the following: Net pension income (or expense), and other income (or expense). In his report, Mr. May stated that PMG had an overfunded defined benefit plan, which, for most years, increased the company's reported net income. He eliminated the pension amounts from PMG's historical financial statements in showing its EBITDA, adding back the "full amount of the overfunding", $11,664,000, to PMG's enterprise value. Mr. May has failed 2011 Tax Ct. Memo LEXIS 150">*169 adequately to explain both his $11,664,000 calculation and the reason for assuming the overfunded plan provided no annual benefit under the discounted cashflow method, thus prompting its exclusion from PMG's financial statements. Because we fail to understand his adjustments, we shall disregard them. See 101 T.C. 412">Estate of Jung v. Commissioner, supra at 450; Parker v. Commissioner, 86 T.C. 547">86 T.C. 562. Mr. May also failed to explain his other income (expense) adjustments, which we similarly disregard.

We must next decide whether the guideline company method is an appropriate method to use in valuing decedent's units of PMG, and if appropriate, the size and nature of applicable discounts and adjustments.

A generally accepted method for valuing stock of a closely held company, the guideline company method is "a market-based valuation approach that estimates the value of the subject company by comparing it to similar public companies." Estate of True v. Commissioner, T.C. Memo. 2001-167, affd. 390 F.3d 1210">390 F.3d 1210 (10th Cir. 2004). Since the method determines fair market value using market data from similar public companies (guideline companies), "the selection of appropriate 2011 Tax Ct. Memo LEXIS 150">*170 comparable companies is of paramount importance." Estate of Hendrickson v. Commissioner, T.C. Memo. 1999-278. Both parties' experts used that method in their analyses but only respondent's expert, Mr. Thomson, relied upon its results in valuing decedent's units. Petitioner argues that such reliance is improper because no companies sufficiently similar to PMG exist to support the method's application. We discuss and evaluate Mr. Thomson's guideline company analysis below.

To identify PMG's guideline companies, Mr. Thomson compiled a list of publicly held companies from 10K Wizard, StockSelector.com, and Yahoo.com, screening out those not operated primarily as newspaper publishing companies. He identified 13 potential companies from the list, ultimately selecting the four most similar in size to PMG: Journal Register Co., Lee Enterprises, Inc., the McClatchy Co., and Pulitzer, Inc., and subsidiaries (Pulitzer, Inc.). He believed those companies to be the most comparable to PMG because their underlying price multiples "generally reflect an investor's assessment of both current and future earnings prospects as well as the business and financial 2011 Tax Ct. Memo LEXIS 150">*171 risks, inherent in the Company's business as of the valuation date."

Mr. Thomson analyzed the four chosen guideline companies based on a MVIC-to-EBITDA ⁶ price multiple value indication. He first determined that the MVIC-to-EBITDA price multiples for Journal Register Co., Lee Enterprises, Inc., the McClatchy Co., and Pulitzer, Inc., as of June 30, 2004, were 11.1, 12.4, 10.9, and 12.6, respectively, with a median multiple of 11.8. He then calculated PMG's price multiple by adjusting the guideline companies' median multiple downward to 10.6 because PMG's "EBITDA growth rates slowed while those same growth rates reported by the guideline companies improved, * * * [PMG] was at a size disadvantage and, all else equal, private companies tend to sell for less than public companies."

Mr. Thomson applied PMG's price multiple to PMG's adjusted EBITDA and subtracted the company's interest-bearing debt as of June 27, 2004, concluding that PMG's marketable minority interest value was $435,000,000 (rounded) at the valuation date. Upon applying a 31-percent marketability 2011 Tax Ct. Memo LEXIS 150">*172 discount "to account for the lack of marketability inherent in a minority interest of a closely held company", he determined that the fair market value of the members' equity in PMG on an aggregate minority interest basis as of the valuation date was $300,000,000 (rounded) under that method.

Mr. Thomson failed to analyze sufficiently comparable publicly held companies to warrant application of the guideline company method herein. Publicly held companies involved in similar, rather than the same, lines of business may act as guideline companies. Sec. 2031(b). However, "As similarity to the company to be valued decreases, the number of required comparables increases in order to minimize the risk that the results will be distorted by attributes unique to each of the guideline companies." Estate of Heck v. Commissioner, T.C. Memo. 2002-34. While the small number of guideline companies chosen by Mr. Thomson are engaged in business ventures similar to PMG's, their differences from PMG prevent a reliable comparison.

Size: As of June 2004, PMG was smaller than any of the four guideline companies in terms of both total assets and revenue. 2011 Tax Ct. Memo LEXIS 150">*173 PMG had approximately one-third of the revenue and approximately one-fourth of the total assets of the guideline companies' respective medians ($163,600,000 versus $557,300,000 and $353,000,000 versus $1,368,200,000, respectively).

Products: While PMG primarily published daily and weekly newspapers, it also published a few specialty publications. In contrast, Lee Enterprises and Pulitzer, Inc., complemented their daily and weekly newspapers with a wide variety of classified, specialty, shoppers, and niche publications. Further, three of the guideline companies heavily integrated Internet news into their business models. The Journal Register Co. operated 152 news websites, intending to expand its business through Internet initiatives, while both Lee Enterprises and Pulitzer, Inc., published their newspapers with associated and integrated online services. PMG's business plan did not include an Internet component.

Other Differences: PMG experienced greater EBITDA and revenue growth than the median growth of the guideline companies during the 5 years before June 2004. As of June 2004, however, PMG shared a similar liquidity ratio with, but was more highly leveraged (as measured by long-term 2011 Tax Ct. Memo LEXIS 150">*174 debt in relation to net worth) than, the guideline companies' median liquidity and leverage ratios.

We find that Mr. Thomson improperly relied on the guideline company method because the four guideline companies alone were not similar enough to PMG to warrant its application. See, e.g., Estate of Hall v. Commissioner, 92 T.C. 312">92 T.C. 325, 341 (finding that the taxpayer's experts "acted reasonably in selecting" six comparable companies where those companies were involved in similar businesses as, and occupied similar positions within those industries as, the subject company); Estate of Zaiger v. Commissioner, 64 T.C. 927">64 T.C. 927, 64 T.C. 927">935, 64 T.C. 927">945 (1975) (finding that the comparable companies the Commissioner's expert used were not sufficiently comparable because of differences in product mix and size operations). Although the McClatchy Co. is arguably of sufficient similarity to PMG, a single comparable company is insufficient on which to base the valuation method. See 92 T.C. 312">Estate of Hall v. Commissioner, supra at 339. Because Mr. Thomson improperly relied on the market-based approach in valuing decedent's units of PMG, it is unnecessary to address the other two areas of disagreement concerning the 2011 Tax Ct. Memo LEXIS 150">*175 approach's application: Appropriate adjustments and discounts.

We next consider the proper application of the discounted cashflow (DCF) method in valuing decedent's units. Given the lack of public companies comparable to PMG, we agree that the DCF method is the most appropriate method under which to value the units. See Estate of Heck v. Commissioner, supra.The DCF method is an income-based approach whereby a company's value is measured by the "present value of future economic income it expects to realize for the benefit of its owners." Estate of True v. Commissioner, T.C. Memo. 2001-167. After analyzing the company's revenue growth, expenses, capital structure, and the industry in which it operates, the company's future cashflows are estimated and their present values are calculated based on "an appropriate risk-adjusted rate of return." Id.

Both Mr. May and Mr. Thomson rely on the method in their appraisals but disagree as to: (1) PMG's projections, (2) whether to tax affect PMG's earnings in calculating the company's value, (3) cashflow adjustments, (4) the amounts to be included in the rate of return, (5) earnings adjustments to PMG's 2011 Tax Ct. Memo LEXIS 150">*176 enterprise value, and (6) the nature and amount of applicable discounts. We discuss both experts' computations in arriving at the PMG units' fair market value.

Both Mr. Thomson and Mr. May valued PMG's units under the DCF method; however, they discounted different economic benefits of the company. Specifically, Mr. Thomson discounted PMG's net EBITDA, beginning his calculations with PMG's revenue for the last 12 months ended June 27, 2004. Mr. May, in contrast, derived the present value of PMG's net free cashflow ⁷ and began his analysis with PMG's operating income during the 5 months ended May 30, 2004. The parties have not distinguished between these two economic benefits against which to apply the DCF method. We shall derive the present value of PMG's net free cashflow (net cashflow). However, the financial statements on which the experts' revenue and operating income numbers are based are not in evidence. Petitioner failed to object to and disprove the amount of PMG's revenue as of June 27, 2004, relied upon by Mr. Thomson; therefore, we deem it to be accurate. Because we determine the appropriate revenue growth rate below, and the experts calculated 2011 Tax Ct. Memo LEXIS 150">*177 PMG's projected operating income as a percentage of PMG's projected revenue, we shall construct our own operating income projections in discounting PMG's net cashflow.

Since the DCF method calculates the present value of a company's future economic income, it is imperative to use a reliable economic forecast in applying the method. Both parties' experts prepared their own economic projections rather than use the February 2004 multiyear forecast Wachovia Bank prepared in conjunction with its CIM.

On brief, respondent argues that Mr. May erred in not using the CIM forecast, which respondent argues best projects PMG's future income and expenses. Respondent does not disavow his own expert's appraisal, however, which does not rely on the CIM forecast. We regard respondent's continued reliance on his own expert's appraisal as a retreat from respondent's position regarding the use of the CIM forecast. We, therefore, need not consider the need to use 2011 Tax Ct. Memo LEXIS 150">*178 it herein.

In projecting PMG's future revenue, Mr. Thomson estimated that the company's revenue would grow by 5.45 percent in year 1, ⁸ 1.5 percent in year 2, and 1 percent annually during years 3, 4, and 5. He arrived at those projections by assuming an annual 1-percent baseline growth rate based on PMG management's statements that, absent acquisitions, the company grew approximately 1 percent each year from 1999 to 2004 and was expected to continue to do so.

He then made two adjustments to this annual rate. For year 1, Mr. Thomson assumed a starting revenue of $85,828,303 (half of the revenue estimated in PMG's 2004 budget), to account for the second half of 2004, adding to that 1 percent of growth for that same amount, to account for the first 6 months of 2005. ⁹ This resulted in 5.45 percent of growth from PMG's revenue for the last 12 months ended June 27, 2004. For year 2, he added 0.5 percent of growth to the presumed 1-percent revenue growth to reflect PMG's acquisition of the remaining 50-percent interest in High Point, resulting in 1.5 percent total growth 2011 Tax Ct. Memo LEXIS 150">*179 for the year. Mr. Thomson calculated this percentage increase after determining that PMG grew 0.8 percent in 2002 as a result of a 100 percent acquisition that year.

In his report, Mr. May projected total revenue growth for PMG of 4.6 percent for its year ending December 31, 2004, and of 2.9 percent, 4.9 percent, 3 percent, 4.9 percent, 3 percent, and 4.8 percent for years 2005, 2006, 2007, 2008, 2009, and 2010, respectively. The total revenue growth rates were composed of: (1) An annual newspaper revenue growth rate of 4 percent, and (2) a vacillating annual television revenue growth rate. Mr. May chose a constant 4-percent newspaper revenue rate based on his assumption that PMG "is going to grow at the same rate as the comparable companies long term", thus aligning the newspaper revenue rate with the guideline companies' 3.9-percent average implied long-term growth rate. He did not include PMG's option to acquire the remaining 50-percent interest in High 2011 Tax Ct. Memo LEXIS 150">*180 Point in the company's future expected cashflow because "it would be neither accretive nor dilutive to shareholder value because the price to be paid was to be fair market value."

We find Mr. Thomson's revenue growth projections to be more persuasive. Mr. May chose a growth rate that he himself acknowledged was "significantly" higher than the company's actual 2002 and 2003 newspaper growth (0.6 percent and 1.1 percent, respectively).

In contrast, Mr. Thomson's baseline projection derives from PMG's historical growth absent acquisitions, a reasonable benchmark given that PMG did not specifically identify to either party's expert future acquisitions as of the valuation date. We also agree with Mr. Thomson that revenue expected from the October 22, 2004, completion of the High Point acquisition should be included in the projections, since the completion of that acquisition was foreseeable as of the valuation date. ¹⁰ See, e.g., Bergquist v. Commissioner, 131 T.C. 8">131 T.C. 17 ("Subsequent events are not considered to fix fair market value, except to the extent that they were reasonably foreseeable at the date of valuation." (internal quotation marks omitted)). Contrary to Mr. May's 2011 Tax Ct. Memo LEXIS 150">*181 position, although PMG acquired the remaining 50-percent interest at fair market value, thus not affecting the company's balance sheet, the expected future revenue affects the company's economic projections, thereby demanding its consideration under the DCF method. Finally, Mr. Thomson based adjustments to the baseline projection on actual past performance and resulting effects to the company. We shall rely on Mr. Thomson's revenue projections in valuing decedent's units of PMG.

Respondent next disputes Mr. May's adjustment to his economic projections to account for higher industry newsprint costs. Because of the "industry-expected rise of newsprint costs for 2004 and 2005", Mr. May estimated a 0.7-percent cost increase in 2004 and an annual 1.3-percent cost increase in years 2005 through 2010. He fails, however, to explain how he arrived at those projected costs. Petitioner provides no further support for the annual adjustment, other than referring to Mr. Paxton's testimony that such an adjustment may be necessary if future cashflows will differ from past cashflows. Because neither 2011 Tax Ct. Memo LEXIS 150">*182 petitioner nor his expert, Mr. May, has convinced us as to the propriety of the adjustment, we shall disregard it.

Mr. Thomson projected PMG's operating margin (operating income) as a percentage of revenue for year 1 through year 5 at 39.5 percent of revenue. He computed operating income by subtracting operating costs (excluding depreciation and amortization of syndicated programming contracts), which he estimated at a constant 60.5 percent of revenue, from revenue. He derived that estimate from PMG's 2004 budget, which estimated total operating costs at 60.8 percent of revenue, and PMG's 2003 total operating costs, which were 60.4 percent of revenue.

Mr. May projected an operating income profit margin of 34.7 percent for 2004 and 34.1 percent for all subsequent years. He based his projection, which shows a decrease in operating income margins, on two assumptions: (1) Projected annual corporate overhead expenses of 2.9 percent of revenue, which he arrived at by averaging PMG's historical corporate overhead expenses between 2000 and 2003, after adjusting for nonrecurring items (namely, a 2003 positive claim experience for self-insured life insurance 2011 Tax Ct. Memo LEXIS 150">*183 and a 2003 gain on life insurance from acquired companies), and (2) an increase in newspaper cost of goods sold margins for expected increases in newsprint costs in 2004 and 2005. Absent these two assumptions, Mr. May stated that "our projected operating income margins would have been higher than any of Paxton's historical operating margins." In projecting operating costs, he also factored in a projected depreciation cost of 3.1 percent of revenue.

We do not have confidence in Mr. May's projection as it is based on improper earnings and newsprint cost adjustments. See supra parts VI. and VIII.C.3. We determine Mr. Thomson's analysis to be reasonable. Therefore, we find that PMG's projected total operating costs (excluding depreciation and amortization of syndicated programming contracts) are 60.5 percent of revenue. Because we are discounting PMG's cashflow, we modify Mr. Thomson's forecasted operating margin to include Mr. May's projected depreciation deduction of 3.1 percent of revenue, which we find to be a reasonable projection. After examining all of the evidence, we conclude that PMG's projected operating margin is 36.4 percent of total revenue.

Both Mr. 2011 Tax Ct. Memo LEXIS 150">*184 Thomson and Mr. May adjusted PMG's operating income to reflect other income (expense) from affiliate company management fees and equity in net income of affiliate company. However, they disagree over whether to also add other net income and net pension income (expense). As stated earlier, supra part VI., we disregard Mr. May's exclusion of these items from PMG's historical financial statements. We adopt Mr. Thomson's projection of other income (expense) at 0.1 percent of revenue, which we consider to be reasonable.

The parties next dispute the propriety of tax affecting PMG's earnings under the DCF method. Tax affecting "is the discounting of estimated future corporate earnings on the basis of assumed future tax burdens imposed on those earnings". Dallas v. Commissioner, T.C. Memo. 2006-212. Since most data on which stock valuation is based is derived from publicly traded C corporations, appraisers may tax affect an S corporation's earnings to reflect its S status in its stock value. ^{112011 Tax Ct. Memo LEXIS 150">*185} See Bogdanski, Federal Tax Valuation, par. 6.03[6][e] (2009 & Supp. 2011).

Mr. May tax affected PMG's earnings by assuming a 39-percent income tax rate in calculating the company's future cashflows, before discounting PMG's future earnings to their present value. He also assumed a 40-percent marginal tax rate in calculating the applicable discount rate. In contrast, Mr. Thomson disregarded shareholder-level taxes in projecting both the company's cashflows and computing the appropriate discount rate.

Mr. May failed to explain his reasons for tax affecting PMG's earnings and discount rate and for employing two different tax rates (39 percent and 40 percent) ¹² in doing so. Absent an argument for tax affecting PMG's projected earnings and discount rate, we decline to do so. As we stated in Gross v. Commissioner, T.C. Memo. 1999-254, the principal benefit enjoyed by S corporation shareholders is the reduction in their total tax burden, a benefit that should be considered when valuing an S corporation. Mr. May has advanced no reason for ignoring such a benefit, and we will not impose 2011 Tax Ct. Memo LEXIS 150">*186 an unjustified fictitious corporate tax rate burden on PMG's future earnings.

Mr. May defines net cashflow generally as net operating income after taxes plus depreciation and amortization expenses and minus working capital and capital expenditures. We have accepted similar definitions, see, e.g., N. Trust Co. v. Commissioner, 87 T.C. 349">87 T.C. 349, 87 T.C. 349">365 n.17 (1986), and accept that definition, but, for the reasons stated supra part VIII.D., without regard to a hypothetical corporate tax. The experts disagree only as to the latter two adjustments, and we discuss these modifications below.

Mr. Thomson estimated PMG's annual capital expenditures at 2.77 percent of revenue for years 1 through 5. He based his projection on PMG's operating history and historical amounts, discussions with PMG's management, and "additions necessary to support the projected future revenue volumes". Mr. May, in contrast, projected that the 2011 Tax Ct. Memo LEXIS 150">*187 company's capital expenditures would steadily increase from $4,190,000, or 2.3 percent of revenue, in 2005 to $7,807,000, or 3.1 percent of revenue, in 2014. Mr. May did not justify his projection.

PMG's historical financial statements show capital expenditures ("purchase of property and equipment"), as a percentage of revenue, fluctuating between 1999 and 2003: 4.6 percent in 1999, 14.2 percent in 2000, 0.9 percent in 2001, 3.6 percent in 2002, and 2.9 percent in 2003. The dramatic increase in capital expenditures in 2000 resulted from a major acquisition that year. The average annual capital expenditures for the 1999-2003 period, excluding 2000, was 2.4 percent of revenue. Thus, we consider Mr. Thomson's projection of 2.77 percent of revenue (2.8 percent, rounded) to be reasonable, and we adopt it. Conversely, not only does Mr. May fail to support his projection, but PMG's financial statements do not justify his estimated increase in capital expenditures.

Mr. Thomson estimated that PMG's debt-free working capital (current assets less current liabilities (excluding current maturities of long-term debt and line of credit)) would remain at -2.5 percent of revenue throughout 2011 Tax Ct. Memo LEXIS 150">*188 his projection, on the basis of PMG's historical data. Mr. May projected that PMG's future investment in working capital would fluctuate; however, he failed to explain how he arrived at those estimates. Mr. May's only explanation consisted of an appendix to his report containing his projected income statement, balance sheet, and cashflow statement for PMG. We shall ignore Mr. May's working capital projections because of their complete lack of support. After analyzing the record, we are persuaded that PMG's historical performance justifies Mr. Thomson's projection of working capital levels, and we find in accordance.

Both parties' experts used PMG's weighted average cost of capital (WACC) ¹³ as the appropriate rate of return with which to discount PMG's expected future cashflow under the DCF method. We have previously held that WACC is an improper analytical tool to value a "small, closely held corporation with little possibility of going public." Estate of Hendrickson v. Commissioner, T.C. Memo. 1999-278; cf. Gross v. Commissioner, supra (allowing the use of WACC when the expert used the subject corporation's actual borrowing costs to calculate the cost of debt capital 2011 Tax Ct. Memo LEXIS 150">*189 component of the WACC formula). Neither party has indicated the likelihood of PMG's becoming a publicly held company; however, because both experts used WACC as the rate of return in their analyses, and neither party otherwise raised the issue, we shall adopt it, although we do not set a general rule in doing so.

Mr. Thomson computed a 10-percent WACC, assuming a zero-percent marginal tax rate, whereas Mr. May calculated a WACC of 12.3 percent, assuming a 40-percent corporate tax rate. As stated earlier, supra part VIII.D., we shall not tax affect PMG's expected cashflows or the discount rate; we assume a zero-percent tax rate in discounting the company's earnings. The remaining disagreements between the parties center around two components of the WACC formula: (1) The method for computing PMG's cost of equity capital, and (2) the percentages of PMG's total capital composed of debt and equity. 2011 Tax Ct. Memo LEXIS 150">*190 Although the parties have not raised as an issue the size of PMG's cost of debt capital, the experts have taken different positions, and we must, therefore, decide the issue.

Mr. May used the capital asset pricing model formula (CAPM) ¹⁴ to derive a 13.5-percent cost of equity capital for PMG; Mr. Thomson, in contrast, calculated a 20-percent cost of equity capital under the buildup method. ^{152011 Tax Ct. Memo LEXIS 150">*191} We agree with Mr. Thomson that the buildup method is the appropriate method by which to compute PMG's cost of equity capital. The special characteristics associated generally with closely held corporate stock make CAPM an inappropriate formula to use in this case. See, e.g., Hoffman v. Commissioner, T.C. Memo. 2001-109 ("The use of CAPM is questionable when valuing small, closely held companies.").

Although we accept Mr. Thomson's use of the buildup method, we are not convinced as to the accuracy of his analysis. To compute a cost of equity capital of 20 percent, Mr. Thomson first identified from Ibbotson Associates an equity risk premium of 11.7 percent, ^{162011 Tax Ct. Memo LEXIS 150">*192} to which he added a risk-free rate of 5.22 percent. Ibbotson Associates, Stocks, Bonds, Bills, and Inflation, 2004 Yearbook 32 (Ibbotson 2004). He then subtracted PMG's projected 1-percent long-term growth rate to arrive at a minority capitalization rate. After converting this result to a minority market multiple, he added a 20-percent premium for control, ultimately arriving at a majority discount rate for PMG of 14 percent. To this number, he added a 4-percent firm specific risk premium and a 2-percent premium to account for PMG's "S" corporate status.

We do not understand why Mr. Thomson included PMG's firm specific risk premium at the final step of his analysis rather than considering it along with the risk-free rate and equity risk premium. We find this to be in error. See Trugman, Understanding Business Valuation: A Practical Guide to Valuing Small to Medium-Sized Businesses 250 (1998). In addition, we modify Mr. Thomson's control premium of 20 percent to 30 percent, as discussed below. After taking into consideration these modifications, we find that PMG's cost of equity capital is 18 percent.

Mr. Thomson estimated PMG's pretax cost of debt at 6.6 percent on the basis of his review of PMG's weighted pretax cost of debt as of December 28, 2003, and December 26, 2004 (2.9 percent and 3.5 percent, respectively), and of the average of Baa corporate bonds as of July 2, 2004 (6.6 percent). He also considered PMG's leverage, its industry, and the low debt financing rates in 2003 and early 2004, which were expected to rise.

In his expert report, Mr. May calculated a 5-percent average cost of debt. The entirety of his reasoning lies in a footnote: "Based on Company's existing costs 2011 Tax Ct. Memo LEXIS 150">*193 of debt and estimated costs of debt, given the companies [sic] financial condition and the current interest rate environment."

Neither expert convinced us as to the accuracy of his analysis. We accept the assumptions upon which Mr. Thomson's calculation is based, which cannot be objectionable to petitioner, since Mr. Thomson's proposed higher cost of debt results in a lower present value of expected cashflow. SeeEstate of Heck v. Commissioner, T.C. Memo. 2002-34. We find that 6.6 percent is a reasonable cost of debt capital for PMG.

As part of his WACC analysis, Mr. Thomson selected 75 percent and 25 percent as PMG's total capital composed of debt and equity, respectively. He derived those percentages after considering both PMG's capital structure, (composed of 73-percent audited book debt and 27-percent audited book equity) and the guideline companies' median capital structure (16-percent debt and 84-percent equity, based on market values). He relied upon PMG's own capital structure rather than on the guideline companies' median because: (1) A minority shareholder cannot change the capital structure of the company, and (2) 2011 Tax Ct. Memo LEXIS 150">*194 Mr. Paxton's statement that PMG would continue to be more leveraged than those companies because of future acquisitions financed with debt. Additionally, Mr. Thomson observed that PMG's total capital composed of debt and equity percentages for 2003, the year immediately preceding the valuation date, were 77.5 percent and 22.5 percent, respectively.

Mr. May determined that percentages of debt and equity in PMG's capital structure were 15 percent and 85 percent, respectively. He provided no analysis as to how he arrived at those percentages, other than to state that the 15 percent is "based on analysis guideline companies' capital structure".

Petitioner argues that Mr. Thomson erroneously used book, rather than market, values in determining the equity and debt ratios. We agree that market values of a company's debt and equity are to be used in estimating its weighted average cost of capital. SeeFurman v. Commissioner, T.C. Memo. 1998-157 (finding that "To compute WACC, it is necessary to know the market value of the firm's debt and equity"); Brealey & Myers, Principles of Corporate Finance 525 (7th ed. 2003). However, because both experts rely on the WACC formula and PMG's own market 2011 Tax Ct. Memo LEXIS 150">*195 values are unknown because of its closely held company status, we must determine the appropriate debt and equity ratios on the basis of the persuasiveness of the experts' analyses.

Mr. Thomson testified that he used the company's own capital structure because decedent could not affect PMG's capital structure as a minority interest unitholder. He explained that that decision required that he use the company's book values as its market values were unknown. In contrast, despite Mr. May's conclusion that the guideline companies were not comparable under the guideline company method, he declared that the same companies were sufficiently comparable under the DCF method to justify using their capital structures to calculate PMG's WACC. We lend no weight to Mr. May's wavering stance and therefore use PMG's own capital structure at book value for purposes of this case. We find that 75-percent total capital composed of debt and 25-percent total capital composed of equity is appropriate.

Petitioner also argues that Mr. Thomson erroneously maintained the 75-percent debt and 25-percent equity ratio throughout his analysis, a ratio that violates the requirement under the CIM that PMG reduce its debt 2011 Tax Ct. Memo LEXIS 150">*196 over time. We agree that WACC is an improper discount rate tool for a company planning to pay down its debt, thereby changing its capital structure. See Brealey & Myers, supra at 536. However, given the parties' reliance on their own experts, both of whom use the WACC formula in their respective analyses, we must adopt the approach. Therefore, we will use Mr. Thomson's constant WACC rate in this case, although we do not intend to establish a general rule in doing so.

We agree with Mr. Thomson that the weighted average cost of capital for PMG is 10 percent (rounded), as our modification to Mr. Thomson's cost of equity capital determination does not materially alter this result.

Both experts agree that, under the DCF method, PMG's long-term debt must be subtracted from the present value of its future economic income stream in order to arrive at the fair market value of PMG's units as of the valuation date. They disagree, however, as to the amount of PMG's debt as of the valuation date. Mr. Thomson determined that PMG had $243,602,413 of interest-bearing debt as of June 27, 2004, on the basis of PMG's comparative balance 2011 Tax Ct. Memo LEXIS 150">*197 sheet for interim period ended June 27, 2004. Mr. May, meanwhile, concluded that PMG had $243,300,000 (rounded) of net debt as of May 30, 2004, relying on PMG's internally prepared financial statements for the 5 months ended May 30, 2004.

Mr. Thomson and Mr. May both arrive at their debt numbers on the basis of balance sheets for interim periods that are not in evidence. Because neither party objects to the other's debt number, we deem the parties to have conceded the numbers as accurate. Therefore, we must determine which expert provided a more reliable long-term debt calculation.

Neither expert supports his conclusion with an explanation. After examining the record, we find that the experts generally included the same categories of obligations in their PMG debt calculations, and the difference apparently is because of the dates of the financial data used. Because we have already concluded that PMG's June 27, 2004, unaudited financial statement provides the most relevant information, we shall adopt Mr. Thomson's long-term debt conclusion of $243,602,413. Moreover, Mr. May identified differing debt amounts throughout his report. We will not rely on a consistently changing number, especially 2011 Tax Ct. Memo LEXIS 150">*198 one that Mr. May fails to justify.

Mr. May adjusted PMG's marketable minority enterprise value by $900,000 (rounded) to reflect PMG's working capital excess (or deficit). He determined that, as of the valuation date, PMG's working capital was underfunded by $900,000 (rounded) after comparing PMG's working capital level with the guideline companies' median ratio of working capital-to-sales. He applied this adjustment to his results under both the guideline company method and the DCF method, justifying its need only under the former. Mr. Thomson did not make a similar adjustment under either valuation method.

We do not find Mr. May's analysis to be persuasive. Mr. May once again failed to explain why the public companies that he deemed to be not comparable to PMG under the guideline company method provide a sufficient comparison upon which to base a working capital adjustment. We lend little weight to his seemingly contradictory positions. In addition, although explaining the need for a working capital adjustment under the guideline companies methodology, he failed to do so under the DCF method despite applying the adjustment to the results under both methods. 2011 Tax Ct. Memo LEXIS 150">*199 For these reasons, we disregard his working capital deficit adjustment.

Mr. May made the following three adjustments to his result under the DCF method to account for certain shareholder benefits associated with PMG's S corporation election: (1) Adding $12,847,000 to account for "S shareholder tax savings on all future projected distributions in excess of tax distributions", (2) adding $44,262,000 to reflect the future value of the company's deductible goodwill, discounted back to the valuation date, and (3) adding $6,693,000 to account for the company's extra marginal debt tax shield. Petitioner argues that such adjustments are proper under Gross v. Commissioner, T.C. Memo. 1999-254, in which we stated that "the principal benefit that shareholders expect from an S corporation election is a reduction in the total tax burden imposed on the enterprise" and that such savings ought not be ignored in valuing an S corporation. Petitioner argues that with the aforementioned three adjustments, Mr. May's valuation of PMG's units reflects the full economic value of PMG's S corporation tax status, thus satisfying our directive in Gross. Mr. 2011 Tax Ct. Memo LEXIS 150">*200 Thomson made no similar adjustments.

Although correctly cited by petitioner, Mr. May erred in implementing our directive in the case. In Gross, one of the taxpayer's experts criticized the Commissioner's expert for not accounting for the "known payment" of taxes in valuing the subject S corporation's earnings under the DCF method because he assumed a zero-percent corporate tax rate. Interpreting the taxpayer's expert as arguing that "the avoided C corporation tax must be taken into account as a hypothetical expense, in addition to the shareholder level taxes actually imposed on the S corporation's shareholders" when valuing an S corporation, we concluded that

[The taxpayer's expert] has not convinced us that such an adjustment is appropriate as a matter of economic theory or that an adjustment equal to a hypothetical corporate tax is an appropriate substitute for certain difficult to quantify disadvantages that he sees attaching to an S corporation election. We believe that the principal benefit that shareholders expect from an S corporation election is a reduction in the total tax burden imposed on the enterprise. The owners expect to save money, and we see no reason why that savings 2011 Tax Ct. Memo LEXIS 150">*201 ought to be ignored as a matter of course in valuing the S corporation.

Both experts acknowledged PMG's stock option program, which was in effect as of the valuation date, in their respective valuation analyses. They disagreed, however, as how to best measure its impact on the fair market value of the company's units. To determine the units' fair market value before discounting for lack of marketability, Mr. May assumed that all of the outstanding options would vest and subtracted the expected proceeds from the option exercise, an in-the-money value of $12,100,000 (rounded), from PMG's enterprise value.

In contrast, Mr. Thomson did not account for the outstanding options' strike price. Rather, and without 2011 Tax Ct. Memo LEXIS 150">*202 explanation, he calculated PMG's per unit fair market value by dividing the fair market value of PMG's members' equity by the total number of issued and outstanding fully diluted units as of the valuation date. After reviewing the evidence, it seems that his calculation rests upon the same "treasury stock method" assumption used in Mr. Paxton's July 12, 2004, valuation report; namely, that an option exercise causes a company to use the resulting proceeds to repurchase shares at the prevailing market price and issue a mix of new and repurchased shares to meet its obligation. Relying on this assumption, Mr. Thomson appears to have concluded that money received from the option exercise would not increase PMG's cashflows; the exercise would result solely in an increased number of outstanding units. Mr. Thomson has not convinced us that the above assumption reflects how PMG operates its stock option program, and Mr. Paxton's report does not provide an answer. Therefore, we shall adopt Mr. May's approach (but not his numbers) in estimating the dilutive impact of the options outstanding.

The parties do not dispute that, in valuing the units, it is appropriate to take account of 2011 Tax Ct. Memo LEXIS 150">*203 both a minority discount and a lack of marketability discount. Indeed, we have accepted both discounts when valuing stock of closely held corporations. SeeEstate of Newhouse v. Commissioner, 94 T.C. 193">94 T.C. 249. The parties differ, however, on the size of and, with respect to the minority discount, the method of application of the discounts.

As a preliminary matter, we note that, although similar, there is a discernible difference between the two discounts, as articulated in Estate of Andrews v. Commissioner, 79 T.C. 938">79 T.C. 953:

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. * * *

Mr. Thomson applied a 17-percent minority discount to his result under 2011 Tax Ct. Memo LEXIS 150">*204 the DCF method on the basis of studies of control premiums, since, he believes, the inverse of a control premium "equates to a minority interest discount." He then applied a 31-percent marketability discount to reach an aggregate minority interest value in PMG of $267,000,000 as of the valuation date. In contrast, Mr. May applied only one discount under his DCF analysis: a 30-percent lack of marketability discount. He found a specific minority discount unnecessary because: "The DCF Methodology is derived based on cashflows that we assume would accrue pro rata to all equity holders, therefore, the resulting firm value is on a minority interest basis and needs no further adjustment to reflect a minority interest value." We discuss each discount below.

Mr. Thomson applied a 17-percent minority interest discount to the fair market value of PMG's members equity on a controlling interest basis that he calculated under the DCF methodology. He applied the discount so as to reflect the decline in value of decedent's units given the lack of control inherent in her minority interest. To calculate the discount size, and concluding that a minority interest discount is 2011 Tax Ct. Memo LEXIS 150">*205 the mathematical inverse of a control premium, ^{172011 Tax Ct. Memo LEXIS 150">*206} he reviewed statistics, compiled in Mergerstat Review 2004, on control premiums paid in mergers and acquisitions between 2002 and 2003. He found the following: (1) The median percent premium paid for all industries in 2002 (based on 326 transactions) and 2003 (based on 371 transactions) was 34.4 percent and 31.6 percent, respectively, (2) the overall median percent premium paid for transactions in 2002 (based on 86 transactions) and 2003 (based on 100 transactions) where the purchase price was between $100,000,000 and $499,900,000 was 30.3 percent and 27.4 percent, respectively, and (3) within the printing and publishing services industry during 2002 and 2003, one transaction had a premium of 45.2 percent, and the average premium of the five additional transactions was 67.5 percent. On the basis of those statistics and general factors that affect a control premium's size, he concluded that a 20-percent control premium was reasonable for a majority investment in PMG, which equated to a 17-percent minority interest discount. Mr. May did not apply a specific minority interest discount under his DCF analysis.

Because we generally follow Mr. Thomson's DCF approach, which derives PMG's members' equity on a controlling basis, we agree that a minority discount is appropriate in valuing decedent's 15-percent minority interest in PMG. Cf. Estate of Jung v. Commissioner, 101 T.C. 412">101 T.C. 439 (if an expert's DCF calculations are on a minority basis, no minority discount should be applied to the resulting values). We disagree, however, with Mr. Thomson's analysis. Mr. Thomson only vaguely supported his chosen control premium with the above-referenced statistics. Mr. Thomson determined a control premium for PMG that is 10 percentage points below the median control premium paid for transactions in all industries and 20 percentage points below control premiums paid in PMG's own industry. He provided no explanation as to why he chose such a comparatively low control premium, besides listing general factors that affect a control premium's size. We cannot justify 2011 Tax Ct. Memo LEXIS 150">*207 Mr. Thomson's control premium and resulting minority interest discount without a more comprehensive explanation. Since the parties are in general agreement that a minority discount is appropriate, we determine a 23-percent minority discount to the equity value of PMG computed on a 30-percent controlling interest basis under the DCF method. ¹⁸

Mr. Thomson determined a 31-percent lack of marketability discount after reviewing seven independent restricted stock studies, ¹⁹ which report average and median lack of marketability discounts in restricted stock transactions. Those studies show an average discount of 32.1 percent. Mr. Thomson chose a 31-percent discount based on, among other things, PMG's established name and reputation, its upward trend in distributions, and its trend of redeeming shares.

Mr. May computed a 30-percent lack of marketability discount based on the same seven studies, 2011 Tax Ct. Memo LEXIS 150">*208 plus four ²⁰ additional restricted stock studies and two pre-IPO studies. ²¹ In selecting a discount size, he observed that the restricted stock studies report a smaller average discount than do the pre-IPO studies. He also noted that PMG's stock was significantly more restricted and more likely to be held for a longer period of time than the studied restricted stock, characteristics leading to a higher lack of marketability discount for PMG's units. He, therefore, chose a 30-percent discount, which fell within the range of average discounts (13 percent - 35.6 percent) found by the restricted stock studies.

We have previously disregarded experts' conclusions as to marketability discounts for stock with holding periods of more than 2 years when based upon the above-referenced studies. SeeFurman v. Commissioner, T.C. Memo. 1998-157 (finding the taxpayer's reliance on the restricted stock studies in calculating a lack of marketability discount to be misplaced 2011 Tax Ct. Memo LEXIS 150">*209 since owners of closely held stock held long term do not share the same marketability concerns as restricted stock owners with a holding period of 2 years). Given both experts' reliance on the studies, however, we shall accept them as setting the benchmark discount size for decedent's units. We find a 31-percent lack of marketability discount to be appropriate.

We shall redetermine a deficiency in Federal estate tax commensurate with our finding that the value of the shares as of the valuation date was $32,601,640. See appendix.

Decision will be entered under Rule 155.