Recent court decisions impacting the valuation discounts for estate and gift taxes as summarized in an article dated July 2009 in the Journal of Accountancy
One purpose of fixing a value on an interest in a closely held business is to determine gift and estate tax liability. CPAs called upon to provide such valuations know that this can be a painstaking task. It is not an exact science but an educated estimate when, as often is the case, there is no identifiable market for the interest. This uncertainty can cause unintended gift or estate tax consequences for transfers between related parties during the transferor’s life and at death.
The difference between what a person transferring an interest in a business believes is its fair market value and any higher amount the IRS determines is its fair market value can result in a greater gift tax liability. Likewise, a redetermination by the IRS of the fair market value of such interests held in an estate can spell an underpayment of estate tax. Fortunately for CPA valuation analysts, there are methods that, while not always yielding uniformly accepted results, are recognized by taxing authorities and courts as providing a valid basis for those estimates. In applying those methods, however, CPAs must take stock of recent court decisions for guidance. This article gives an overview of valuation principles for gift and estate tax purposes, reviews some current trends in determining fair market value for such purposes, and makes suggestions for seeking a qualified appraiser.
WILLING-BUYER / WILLING-SELLER TEST
For gift and estate tax purposes, the fair market value of property transferred to another party is measured on the date of the transfer as “the price at which the property would change hands between a [hypothetical] 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 “willing-buyer, willing-seller test,” Treas. Reg. § 20.2031-1(b)).
For assets traded on an established market or that have a readily ascertainable value, the value for gift and estate tax purposes is their fair market value on the date of the transfer or death. For other assets, valuation must be established by an educated estimate.
METHODS OF VALUING CLOSELY HELD ENTITIES
Three types of valuation methods are generally used in calculating the fair market value of an interest in a closely held entity. The market method (also referred to as the comparable sales method) compares the closely held company with its unknown stock value to similar companies with known stock values. The income (or discounted cash flow) method discounts to present value the anticipated future income of the company whose stock is being valued. The net asset value (or balance sheet) method relies generally on the value of the assets of the company net of its liabilities.
The market method or income method is most often used when the closely held company carries on an active trade or business. The net asset value is most often used when a closely held company holds primarily real estate or investment assets and does not carry on an active trade or business.
TRENDS IN VALUATION DISCOUNTS
The valuation of closely held entities for gift and estate tax purposes has been a hotly contested issue—especially with the proliferation of family limited partnerships and limited liability companies that are implemented primarily for estate planning purposes. In many instances these closely held entities do not carry on an active trade or business.
Court cases reveal that the valuation of closely held entities is a judgment call that relies upon the opinion of experts. Courts have long upheld a premise often reflected in expert opinions—that the value of closely held interests is usually less than the value of similar publicly traded interests. The factors underlying this premise include the inability to quickly convert the property to cash at minimal cost (“lack of marketability”) and the inability, if the interest held is less than a majority interest, to control managerial decisions and other aspects of the entity (“lack of control”).
In many instances, the courts insert their own opinion as to fair market value, siding with neither the taxpayer’s nor the IRS’ valuations and often taking a “splitthe- baby” approach. However, at a lecture in January 2009 at the Heckerling Institute on Estate Planning sponsored by the University of Miami, Judge David Laro of the U.S. Tax Court noted the uncertainty this approach has caused for parties to a sale. Judge Laro stated that the Tax Court is no longer taking such an approach and will insert its opinion only where it believes the valuations of the parties are based on erroneous assumptions.
DISCOUNT FOR LACK OF MARKETABILITY
Two types of empirical studies are commonly used to benchmark discounts for lack of marketability (DLOM)—restricted stock studies and pre-initial public offering (pre-IPO) studies.
Public companies often issue restricted stock (unregistered shares). SEC rules restrict the transferability of such shares by mandating a minimum holding period and by limiting the pool of eligible buyers for such shares. Restricted stock studies compare the price of publicly traded, unrestricted shares of companies with the private market price of restricted shares of the same companies and attribute the difference to the lack of marketability of the restricted shares. Approximately 15 such studies exist, showing discounts ranging from 13% to 45%. The SEC restrictions have become less stringent, and consequently the average discounts in the newer studies are lower than in previous studies.
Pre-IPO studies compare the price at which a stock was sold while its issuer was still closely held (and the shares were unregistered) with the price of the same company’s common stock at the time of an initial public offering. Sources of pre-IPO studies include Willamette Management Associates’ Valuation Advisors’ Lack of Marketability Discount Study and those developed by John D. Emory of Emory & Co. These studies generally show a discount for lack of marketability ranging from 18% to 59%—higher than in restricted stock studies.
Recent court decisions have made it clear that more important than the type of study used to quantify a discount is the analysis done by the appraiser to tie the study to the facts of the specific case. The District Court for the Eastern District of Texas in Temple v. United States (123 F.Supp.2d 605, 622 (2006)) said, “the better method is to analyze the data from the restricted stock studies and relate it to the gifted interests in some manner.”
The failure to tailor the analysis to specific facts can have drastic consequences. For example, one of the issues in Holman v. Commissioner (130 TC no. 12 (2008); see also “Tax Matters: FLPs Revisited,” JofA, Sept. 08, page 88) involved the valuation of limited partnership units in a partnership holding stock in computer maker Dell Inc. Experts for both the taxpayer and the IRS used restricted stock studies to determine the DLOM. The taxpayer’s expert cited 13 restricted stock studies that showed median and mean discounts of 24.8% and 27.4% and then adjusted the DLOM up to 35% based on vague and general observations about the investment quality of the partnership units. The Tax Court faulted him for not building from his observed sample median and mean discounts “by quantitative means.”
The IRS expert compared the restricted stock studies performed prior to 1990—the year the SEC implemented rule 144A that expanded the pool of eligible buyers of restricted stock—to restricted stock studies conducted between 1990 and 1997, the latter year being when the SEC reduced the holding period under rule 144 from two years to one.
The pre-990 studies showed an average discount of 34%, while the 1990–1997 studies showed an average discount of 22%. The IRS expert proposed that the 12% differential reflected the effect of the opening of a limited resale market and thus the portion of a marketability discount related to lack of a liquid market. He considered separately the holding period of restricted stock reflected in the 1990–1997 average discount of 22% and concluded its applicability in this case was negligible, adding, along with other factors, another 0.5%, for a total DLOM of 12.5%.
So why not take the portion of the discount related to the holding period restrictions into account? The IRS expert argued that he could not think of an economic reason why the partners in this situation would not agree to let another partner be bought out. Since the partnership agreement allowed for dissolution by unanimous consent and the sole asset held by the partnership was highly liquid Dell stock, the partners could dissolve the partnership by unanimous agreement, transfer the Dell stock pro rata to the exiting partner, and then reconstitute the partnership with the remaining partners with little economic risk.
Both parts of the decision are troubling— the Tax Court’s acceptance of the argument that the DLOM inherent in restricted stock studies is only 12%, and that the court accepted without much reasoning or computation of the likelihood of liquidation, that the partnership would be dissolved upon the request of a limited partner simply because the dissolution would pose little economic risk to the remaining partners.
The argument that the discounts shown by restricted stock studies contain components other than lack of marketability is not new. Another critic of restricted stock studies (and pre-IPO studies), Mukesh Bajaj, attempted to isolate the DLOM. He performed a study in 2001 with David Denis, Stephen Ferris and Atulya Sarin of registered and unregistered private placements and concluded (albeit controversially) that the average discount attributed exclusively to marketability is only 7.23% (“Firm Value and Marketability Discount,” Journal of Corporation Law, Vol. 27, No. 1).
Adding pressure to the argument for lowering discounts is the argument that the older restricted stock studies are outdated, since the restrictions placed on the securities by the SEC have been relaxed over time. This argument is flawed because, during that period, the inherent limitations faced by private companies have not changed. Nonetheless, the argument has been accepted by many courts, including the Tax Court in Litchfield v. Commissioner (TC Memo 2009-21 (2009)). InLitchfield, the Tax Court rejected the taxpayer’s DLOM of 36% and 29.7% for two companies as reflecting what the court considered outdated restricted stock studies and settled on discounts of 25% and 20%, respectively (the IRS had argued for discounts of 18% and 10%).
DISCOUNT FOR LACK OF CONTROL The discount for lack of control (DLOC—also referred to as a minority discount) is usually quantified by comparing the trading price of shares of publicly traded, closed-end investment funds to the net asset value per share of the same funds. For entities holding real estate, the DLOC is determined by comparing the trading price of shares of a selected sample of registered real estate limited partnerships (RELPs) or real estate investment trusts (REITs) to the net asset value of the respective shares.
Citing mere averages or using generic samples of data is not sufficient. As with the DLOM, the appraiser’s skill in relating the sample of closed-end funds used to not only the asset type but also the size and other attributes of the assets of the entity being valued is critical.
In Holman, experts for the IRS and the taxpayer used closed-end fund data, but the court favored the IRS’ approach to dealing with outliers in the sample data and rejected the taxpayer’s use of seven specialized funds in his sample. Following the methodology suggested by the IRS’ experts and leaving the specialized funds out of the sample, the Tax Court calculated minority interest discounts of 11.32%, 14.34% and 4.63% of the respective gifts made in 1999, 2000 and 2001 (the taxpayer’s expert determined the discounts to be 14.4%, 16.3% and 10%).
In Jelke v. Commissioner (TC Memo 2005-131 (2005)), the taxpayer’s expert applied a 25% DLOC. He initially selected seven funds as comparables (with an average discount of 14.8%) but then rejected some of the funds with lower discounts. He ultimately derived the 25% discount by adjusting for various factors the average discount of just two of the seven funds.
The IRS’ expert started with a benchmark discount of 8.61% that he obtained from an article in theJournal of Economics and reduced it to 5%. The Tax Court said the choice of comparable funds by the taxpayer’s expert was flawed because he gave insufficient justification for eliminating two funds as comparables, and among those he retained in the sample, he ignored significant differences in investment strategy and risk between them and the interest being valued. Without explaining exactly how it determined the figure, the court held that the appropriate lack-ofcontrol discount was 10%.
In Astleford v. Commissioner (TC Memo 2008-128 (2008)), the issue was the value of limited partnership interests in Astleford Family Limited Partnership (AFLP) that were gifted during 1996 and 1997. AFLP held a 50% general partnership interest in another real estate partnership called Pine Bend, along with 14 other real estate investments.
The first question was whether separate discounts should be applied to the AFLP interest and the Pine Bend interest. The IRS’ expert stated that since Pine Bend was an asset of AFLP, no discounts were appropriate in valuing Pine Bend. The Tax Court disagreed with this argument, holding that tiered discounts (that is, discounts at the lower-tier entity level and the uppertier entity level) were appropriate where a taxpayer owned a minority interest in an entity that held a minority interest in another entity.
However, the court further stated that tiered discounts will be rejected when (a) the lower-level interest constituted a significant portion of the parent entity’s assets or (b) where the lower-level interest was the parent entity’s principal operating subsidiary. In this case, the court noted that the Pine Bend interest constituted less than 16% of AFLP’s net asset value and was only one of 15 real estate investments held by AFLP, making the use of tiered discounts appropriate.
The Tax Court specifically stated that it did not find either RELP or REIT data generally superior to the other and that courts have accepted expert valuations that used both. In valuing the Pine Bend interest, the taxpayer’s expert, using a sample of 17 RELPs to derive a lower (22%) and upper limit (46%) for the discount, concluded that the appropriate combined discount for lack of marketability and lack of control was 40% for Pine Bend. The Tax Court modified the sample of RELPs used by the taxpayer’s expert to arrive at a discount of 30% for Pine Bend.
With respect to the AFLP interest, the taxpayer’s expert selected a comparison sample of four RELPs (with discounts ranging from 40% to 47%) and concluded that the appropriate DLOC was 45% in the first year and 40% in the second year. The IRS’ expert, using REIT data, concluded that the lack-of-control discount was approximately 7% in one year and 8% the next year. The Tax Court said the sample of RELPs used by the taxpayer’s expert was not representative of AFLP. Two were five times the size of AFLP, and the other two were highly leveraged, unlike AFLP. The court chose to use the REIT data provided by the IRS’ expert as its starting point but used a higher adjustment, to end up with DLOCs of 16.17% and 17.47% for the respective years.
DISCOUNT FOR BUILT-IN GAINS TAXES While the courts and the IRS have agreed that built-in gains (BIG) tax on a corporation’s appreciated assets should be taken into account in valuing its stock using the net asset valuation method, they have not agreed on the proper method for quantifying the discount.
Besides the DLOC issue discussed earlier in this article, a discount for BIG tax also was argued in Jelke, and on this issue, the taxpayer prevailed. The decedent owned a 6.44% interest in a closely held corporation whose assets consisted primarily of appreciated securities with a date-of-death value of $178 million. The estate argued that the entire BIG tax liability of approximately $51 million should be allowed against the fair market value of the securities in determining the company’s value using the net asset valuation method. The Tax Court rejected this argument and held that the IRS expert’s method of discounting the BIG tax liability over a 16-year period was reasonable because the facts in the case showed that an immediate liquidation of the company was unlikely, given the corporation’s historical asset turnover ratio.
The taxpayer appealed to the Eleventh Circuit (507 F.3d 1317 (2007)). That court, following the Fifth Circuit’s reasoning in Dunn v. Commissioner (301 F.3d 339 (5th Cir. 2002)), reversed, stating that 100% of the BIG tax must be taken into account when using the net asset valuation method (regardless of the likelihood of liquidation) because the threshold assumption of the net asset valuation method is that all assets are liquidated as of the date of valuation. Despite a strong dissent by Judge Ed Carnes, the Eleventh Circuit declined to rehear the issue en banc, and the U.S. Supreme Court denied certiorari. These two appeals court victories give taxpayers a strong position for taking 100% of BIG taxes into account in valuing C corporation stock by the net asset valuation method....