- 26 - of the same shares sold later in an IPO. The study concluded that the sales prices in the nonpublic markets were 40 to 45 percent less than sales prices in the IPO's. Thus, Gasiorowski concluded that the estimated marketable minority value, which he implicitly assumes is equal to an IPO value, should be reduced by 45 percent to reflect the nonpublic market value of the shares. We reject this conclusion for the following reasons. Petitioner offered no evidence that the value of the shares was affected by any change in the market conditions, the constraints of the economy, or the financial condition of Savings between the date of decedent's arm's-length sale of 1,111 shares for $307 per share in the nonpublic market and the valuation date 1 month later. Consequently, if we apply the conclusions of the IPO study to the case at hand, we find that it is more likely that $307 is 40 to 45 percent less, rather than more, than the price at which the same shares would sell in an IPO. Furthermore, Gasiorowski disregards the fact that the actual sales value of the shares is nearly identical to his estimated marketable minority value. The near identity of values indicates that the marketability of the Savings shares in the nonpublic market is essentially equal to that of a minority interest in the public market, in which case no discount for marketability is required for a minority interest in Savings.Page: Previous 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Next
Last modified: May 25, 2011