In the past decade, taxpayers grew accustomed to increasing valuation discounts thanks to a string of victories over the Internal Revenue Service. But that trend appeared to end in early 2003. Now the IRS, through the Tax Court, has delivered three body blows to the taxpayer in a series of 2003 cases: McCord,1Lappo2 and Peracchio.3
This triple play of cases indicates the IRS will battle tooth and nail for lower discounts — and will achieve some degree of success. But the underlying facts of these cases also expose some weaknesses in the IRS campaign. Some of the Tax Court's most crucial decisions were made based on the testimony of valuation experts that went uncontested. Had the taxpayers in these three cases presented the Tax Court with alternative methodologies to support their positions, they may have received more favorable results.
The first of these skirmishes was McCord. The May 2003 decision was significant both for the size of the discount sought and because all 13 Tax Court judges participated. McCord dealt with a Texas limited partnership comprised of stocks and bonds (66 percent), real estate partnerships (29 percent) and real estate and oil and gas properties (5 percent), for which the taxpayer claimed a 22 percent minority-interest discount and a 35 percent lack-of-marketability discount, resulting in a combined discount of 49.3 percent. The taxpayer's discount far exceeded the IRS claim of a mere 14.8 percent combine discount.
The IRS retained valuation expert Mukesh Bajaj, of the Law & Economics Consulting Group (LECG), LLC in Emeryville, Calif., to support the smaller valuation discount. Bajaj had been the IRS expert in Gross,4 which broke new valuation ground by successfully claiming that the value of S corporation income should not be tax-affected (for example, reduced by C corporation level taxes) when determining the estimated value of the S corporation stock.
The innovation in McCord — subsequently repeated in Lappo and Peracchio — was to find that the minority-interest discount for an investment company should equal the weighted average of separate minority-interest discount factors determined for each type of asset held.
In what may turn out to be a self-inflicted wound that could bleed taxpayers for years to come, the taxpayer agreed with the Service that it was appropriate to use a weighted average discount based on the categories of assets held in the partnership. The Tax Court, following the lead from the testifying appraisers, accepted the methodology of deriving a separate lack-of-control discount for each asset category owned by the partnership.5
McCord derived the lack-of-control discount for the partnership's equity portfolio by referring to publicly traded closed-end equity investment funds. The trading price of a closed-end share was compared to its pro rata share of the net asset value (NAV) of the fund to calculate the discount. The average discount of the selected closed-end equity funds was 9.96 percent, which the Tax Court rounded to 10 percent. A similar exercise with municipal bond closed-end funds was undertaken in estimating the lack-of-control discount for the bond portfolio.
The lack-of-control discount for the real estate partnerships owned by the McCord partnership was the first issue on which the taxpayer and IRS appraisers really parted ways. The taxpayer's appraiser looked to publicly traded real estate companies; the IRS appraiser looked to real estate investment trusts (REITs). The taxpayer's appraiser unsuccessfully argued that a REIT is required to pay a large portion of its income to shareholders, which distorts the lack-of-control discount from NAV. The Tax Court observed that a regulated investment company (such as a closed-end investment fund) is legally required to pay substantially all of its income to maintain tax-favored status. Therefore, the court deemed the REIT-based analysis acceptable. The price-to-NAV discount for the selected REIT sample was 1.3 percent. (In Lappo, the IRS expert used a similar REIT-based analysis, and found a 1.48 percent price-to-NAV “premium.”)
Critical to the Tax Court's decision to use the REIT-based analysis was its recognition that a significant adjustment was required to eliminate or extract the effect of a liquidity premium in REIT shares. Therefore, the Tax Court determined, the price-to-NAV discount exhibited by the REIT sample must be increased to remove the inherent liquidity premium.
In McCord, the Tax Court determined the liquidity premium in REITs by referring to Karen Wruck's 1989 study on equities,6 which reported that average discounts for marketability in private placements of unregistered (restricted) shares exceeded those observed in private placements of registered shares by 17.6 percentage points. The Tax Court rounded that to 18 percent. The court then determined that the applicable liquidity premium, which is the inverse of a liquidity discount (1/[1-18 percent]), was 22 percent. The court also found the effective minority-interest discount to NAV from the REIT-based analysis was 23.3 percent (1.3 percent plus 22 percent), after identifying and removing the liquidity premium. In a similar analysis, the Tax Court in Lappo boosted the REIT-based analysis by 18 percentage points.
McCord leaves unexplained why a liquidity premium is applicable to the REIT-based analysis, but not the closed-end equity fund-based analysis. The implication is that the underlying assets of the REITs are illiquid and those of the closed-end equity funds are not — an implication that may not be entirely valid. Therefore, the closed-end equity fund discounts to NAV exclusively reflect minority-interest attributes.
In valuing the underlying assets, the Tax Court placed a heavy emphasis on using REITs as the benchmark for real estate. However, of the discount applied, 22 percentage points were derived from Wruck's study, which examined corporate equities. It seems that a similar increase in the discount also would be appropriate in the analysis of an illiquid limited partner interest in a marketable securities partnership.
McCord addresses the illiquidity of the limited partner interest in the real estate partnership, but not in the marketable securities partnership. That is, both the REIT share and the closed-end equity fund share are marketable. At the same time in McCord, the market-priced REIT and closed-end equity fund shares traded at widely different discounts to NAV (REITs at 1.3 percent and closed-end equity funds at 9.6 percent). Why? I believe the differential is a function of current income distributed to the investor, not the relative liquidity of the underlying assets.
Partnership Profiles, Inc., a Dallas, Texas-based publisher of limited partnership information,7 has assembled 11 years worth of pricing data on real estate-based limited partnerships that have traded on the secondary market. No matter what time period is selected, partnerships that make annual distributions trade at lesser discounts from NAV than partnerships that are non-distributing.
Distribution practices distinguish these partnerships, not the underlying assets. If partnerships of the same asset class have different discounts based on distribution history (and the current-income yield provided to investors) why is not income yield the variable that controls higher or lower minority-interest discounts? The REIT-based analysis of McCord and Lappo that requires an 18-to-22 percentage point adjustment is unable to explain the Partnership Profiles trading data differences in discounts between distributing and non-distributing real estate partnerships.
It's interesting to compare the difference in real estate-based discounts applied in McCord and Lappo. The McCord valuation date was Jan. 12, 1996. There were two valuation dates in Lappo: April 19, 1996 and July 2, 1996. The Tax Court in Lappo was troubled by the volatility in the taxpayer expert's conclusion that the price-to-NAV discounts observed for real estate companies decreased from 29.3 percent in April to 20.3 percent in July. But this same Tax Court concluded that the minority-interest discount for real estate was 23.3 percent in January (in McCord) and 19.0 percent in April and July of the same year (in Lappo). The 9 percentage point difference (representing a 31 percent drop) presented by the taxpayer's expert for the three months between April and July was too volatile for the Tax Court. Yet the judges' decisions in McCord and Lappo reflect a 4.3 percentage point difference (representing an 18.5 percent drop) for the four-month period between January and April 1996.
The troika of key Tax Court cases in 2003 also address the discount for lack-of-marketability issue from the perspective of a family limited partnership. The McCord court found unreliable the analysis of pre-IPO private-market pricing data relied upon by the taxpayer's expert. One reason: Buyers of shares before an IPO are likely to be insiders who provide services to the firm going public, and are compensated, at least in part, by a bargain price.
In place of the IPO studies, the Tax Court looked to studies of private placements of registered and unregistered shares in public companies. The IRS expert, Bajaj, postulated that discounts observed in private placements of restricted or unregistered stock reflected attributes other than impaired marketability, because in his study of unrestricted or registered stock, such private placements also occurred at discounts.8 He contended that higher assessment and monitoring costs are incurred in unregistered private placements than in registered private placements — therefore the full differential between the observed discounts is not exclusively attributable to marketability differences.
In McCord and Lappo, the testifying IRS experts, who were both from LECG, concluded that the appropriate lack-of-marketability discount was 7.23 percent, based on an assessment of the specific portion of private placement discounts attributable solely to lack-of-marketability. Although in both cases the Tax Court preferred the IRS experts' private-placement analysis, the court's conclusion of the lack-of-marketability discount reflected significantly higher discounts than those presented by the IRS experts. In McCord and Lappo, the Tax Court looked to Bajaj's study of 88 registered and unregistered private placements from 1990 to 1995. Based on the middle 29 observations, the McCord court found that the appropriate point of reference was a 20.36 percent average discount. In Lappo, the 22.21 percent average discount for all 88 observations was the relevant base percentage. Each case referred to a 1993 study by Michael Hertzel and Richard Smith9 and its average discount of 20.14 percent as support for the Tax Court's use of Bajaj's study.
Neither McCord nor Lappo referred to Wruck's study in support of its lack-of-marketability conclusion. Although the Hertzel and Smith work and the Bajaj study are derivative of the earlier Wruck study, the Tax Court's application of the three studies is split between support for lack-of-control and lack-of-marketability discounts. The Wruck study is referenced only for lack-of-control discounts attributable to real estate assets and the other two studies are employed only for lack-of-marketability discounts without reference to type of underlying assets.
In all three of the cases, the discounts selected by the Tax Court represented mean discounts of the respective groups. There was no attempt to make a relative comparison of the specific attributes of the partnership interest at issue with the group average. In the past, the court had been critical of blindly applying averages from general studies to a specific asset without explaining where that asset properly fits within the range of observations comprising the study.
The valuation discounts coming out of the Tax Court last year may have taxpayers fondly recalling the 1990s when it appeared that the sky was the limit for discounts. The irrational exuberance of ever-increasing valuation discounts has now gone the way of nonstop stock market appreciation. The Tax Court has proclaimed an attitude adjustment for taxpayers about the magnitude of valuation discounts associated with family limited partnerships.
But taxpayers can learn some lessons from the cases of 2003.
The Tax Court's insistence on minority-interest discounts on an asset-by-asset basis reflects the uncontested testimony of experts in the key cases rather than a preference among competing approaches. The trend of stratifying minority-interest valuation discounts by asset class is now the stated practice of IRS appellate officers.10 This approach results in the curious conclusion that the minority-interest discount is specific to the underlying assets, but the lack-of-marketability discount is not.
The use of REITs-based analysis requires further scrutiny of its appropriateness, particularly in light of the substantial adjustments the Tax Court views as necessary to valuing real estate-based family limited partnerships. Since 2002, at least 12 new closed-end funds investing in REITs have been formed. In light of the Tax Court's preference for closed-end fund analysis, these funds may provide taxpayers a more appropriate reference source than direct REIT-based analysis. (See “Real Estate Discounts,” page 56.)
If Wruck's liquidity premium adjustment of 18 percentage points or more is appropriate to REIT-based analysis to uncover the imbedded minority-interest discount, might there also be portfolio or diversification premiums applicable to closed-end equity funds that mask a higher minority-interest discount percentage?
Lack-of-marketability studies based on private-placement data comparing restricted and unrestricted equity interests suggest a benchmark base of 20 percent to 25 percent. In Mandelbaum,11 which involved the valuation of a private-business interest, the Tax Court started with an average percentage discount from restricted stock studies to determine an applicable lack-of-marketability discount. From the average discount, the Tax Court separately examined nine attributes of the business interest in a compare-and-contrast methodology against the selected average in its determination of applying a below-average or above-average marketability discount. Analysis of the specific attributes of a partnership interest, in a Mandelbaum-like fashion, is necessary to emphasize distinctions between group averages and an individual family limited partnership investment.
It's probably premature to lament last year's cases as heralding the demise of substantial discounts associated with family limited partnerships. The minority-interest and lack-of-marketability discounts remain economically real and they are supported by empirical data. If anything is clear after the recent cases, it is that taxpayers cannot merely go through the motions in claiming discounts, but need to present compelling and specific support to document the discounts they seek.
- Charles T. McCord, Jr. and Mary S. McCord, Donors v. CIR, 120 T.C. No. 13 (May 14, 2003).
- Clarissa W. Lappo v. CIR, T.C. Memo. 2003-258 (Sept. 3, 2003).
- Peter S. Peracchio v. CIR, T.C. Memo. 2003-280 (Sept. 25, 2003).
- Walter L. Gross, Jr. and Barbara H. Gross v. CIR, T.C. Memo. 1999-254 (July 29, 1999).
- Unlike the appraisers in McCord, I believe that the composition of assets in the partnership is less relevant to the minority-interest discount than the current income yield produced by such assets and cash distributions to the partners. That is, the minority-interest discount for one partnership holding non-dividend paying marketable securities and one holding non-income producing vacant land should not be appreciably different. The McCord court's rationale would find a significantly higher lack-of-control discount for the vacant land partnership. Stated differently, a limited partner in a non-distributing partnership has less control than his counterpart in a distributing partnership regardless of the character of the underlying assets, because, at a minimum, once distributions are made, the limited partner has unfettered control over the cash in hand.
- Karen H. Wruck, “Equity Ownership Concentration and Firm Value: Evidence from Private Equity Financing”, 23 J. Fin. Econ. 3 (1989).
- “The Partnership Spectrum,” Partnership Profiles, Inc. (May/June issue of each year).
- Mukesh Bajaj, David J. Denis, Stephen P. Ferris and Atulya Sarin, “Firm Value and Marketability Discounts,” 27 J. of Corporation Law 89 (Fall 2001).
- Michael Hertzel and Richard L. Smith, “Market Discounts and Shareholder Gains for Placing Private Equity,”48 J. Fin. 459 (1993).
- Presentation of Mary Lou Edelstein, the National Appeals Family Limited Partnership coordinator for the Internal Revenue Service, at the Florida Bar Tax Section Seminar on Oct. 24, 2003.
- Bernard Mandelbaum v. CIR, T.C. Memo. 1995-255, aff. without published opinion, 91 F.3d 124 (3d Cir. 1996).
REAL ESTATE DISCOUNTS
Twelve funds show a range varying from a discount of 9 percent to a 4.7 percent premium
|Fund Name Premium/(Discount)|
|AEW Real Estate Income Fund||(8.10)%|
|AIM Select Real Estate Income Fund||(9.00)|
|Cohen & Steers Advantage Income Realty Fund||(1.40)|
|Cohen & Steers Premium Income Realty Fund||(4.00)|
|Cohen & Steers Quality Income Realty Fund||(3.00)|
|Cohen & Steers Total Return Realty||4.70|
|ING Clarion Real Estate Income Fund||3.20|
|Neuberger Berman Real Estate Income||(8.60)|
|Neuberger Berman Realty Income Fund||(6.70)|
|Nuveen Real Estate Income Fund||(0.60)|
|Salomon Smith Barney Real Estate Income Fund||(6.30)|
|Scudder Real Estate Fund||(7.10)|
|Source: Barron's, Oct. 13, 2003|
In 1939, a year after Superman made his debut in Action Comics, he got his own series. This copy of his first solo flight was sold at a Heritage Comics auction Feb. 6 for $47,150.