Estate of Litchfield
New Tax Court Decision on Discounts and Embedded Taxes, Hinges on Experts with Commentary
from BV Expert Carl Sheeler
Estate of Litchfield v. Comm’r, 2009 WL 211421 (U.S. Tax Court) (Jan. 29, 2009)
With about $26.4 million in assets, the Litchfield estate consisted primarily of minority stock interests in two closely held, family owned companies, Litchfield Realty Co. (LRC) and Litchfield Securities Co. (LSC). The Internal Revenue Service (IRS) and the estate agreed on the net asset values (NAV) of the estate’s interests. However, they aggressively disputed the discounts related to built-in capital gains taxes, lack of control, and lack of marketability.
Marjorie Litchfield died in 2001. At that time, her estate owned a 43.1% interest in LRC, which held Iowa farmland and marketable securities along with a farming subsidiary. LRC earned a marginal profit, but the company was not performing up to management expectations. Its farm holdings, for example, yielded less than 1% net annual income compared to Midwestern farmland generally, which yielded about 4% of NAV annually. Historically, the company had sold portions of its farm holdings to raise cash.
To increase profitability and shareholder returns, LRC converted from a C corporation to an S corporation in January 2000. However, for the ten years following conversion, if the company sold any of its former C Corp assets then it would incur corporate-level tax on the sale (per IRC Sec. 1374). The company also postponed switching from share-lease agreements with its farmers to straight cash leases, which would have been more profitable and produced more taxable corporate income. As of the valuation date, LRC’s total NAV of $33.174 million included $28.762 million in built-in capital gains—or 86.7% of NAV. Just over $19.789 million of the built-in capital gains taxes related to its farm holdings and $8.972 to its marketable securities.
To prepare the estate’s tax return in connection with LRC, its expert appraised the estate’s 43.1% interest in the company at a fair market value of $6.475 million—after application of discounts for the built-in capital gains taxes, lack of control, and lack of marketability. On audit, however, the IRS valued the estate’s interest in LRC at just over $10 million and assessed a deficiency of approximately $3.825 million. The chart below shows the breakdown of the parties’ respective valuations for LRC:
Estate IRS
NAV (43.1%) $14.298M $14.298M
Disc for capital gains tax 17.4%- 2.0%-
DLOC 14.8%- 10.0%-
DLOM 36.0%- 18.0%-
FMV $6.475M $10.069M
The estate also owned some 23% of Litchfield Securities Co. (LSC), a C corporation that held “blue chip” marketable securities as well as partnership and other equity investments for a combined NAV of $52.824 million. Like LRC, none of LSC stock had ever been publicly traded, and its stock transfer policies generally restricted redemptions or sales outside of the Litchfield family. Its investment strategy focused on maximizing cash dividends to its shareholders, and these had increased consistently over the years.
In the late 1990s, however, the directors became concerned that elderly shareholders in both LRC and LSC would not have adequate cash reserves to pay for estate taxes and other obligations on their deaths. Consequently, management contemplated sales of LRC and LSC corporate assets to finance stock redemptions for these shareholders. After the death of Mrs. Litchfield, LRC sold its farm services subsidiary and, due to mergers and acquisitions of the public companies in its portfolio, LSC realized a significant appreciation of its holdings.
As of the valuation date, LSC’s NAV included $38.984 million in built-in capital gains, or 73.8% of its total NAV. Note: The capital gains tax applicable to both companies ranged from 35.5% to 39.1%. The estate’s expert discounted its 22.96% stock interest in LSC by capital gains tax as well as lack of marketability and control, but on audit, the IRS determined a deficiency of over $3.014 million. Their respective valuations broke down as follows:
Estate IRS
NAV (22.96%) $12.133M $12.133M
Disc for capital gains tax 23.6%- 8.0%-
DLOC 11.9%- 5.0%-
DLOM 29.7%- 10.0%-
FMV $5.748M $9.565M
The court considered discounts for capital gains first. At trial, U.S. Tax Court Judge Swift found that the built-in capital gains associated with the total assets of both companies were “substantial.” In particular:
A hypothetical buyer would be willing to pay fair market value for the LRC and LSC stock, which would take into account and…reflect the millions of dollars in untaxed appreciation over the years in the values of LRC's and LSC's underlying assets. Knowledgeable buyers, however, also would negotiate discounts in the price of the stock to estimate, on the basis of current tax laws, the corporate capital gains tax liabilities due on that very same appreciation when the assets are sold or otherwise disposed of by the corporation. In other words, if a valuation of…corporate stock in a hypothetical sale is significantly affected by the untaxed appreciated value of the underlying corporate assets, the stock valuation …should reflect the corporate capital gains tax liabilities that the appreciated assets carry with them and that will be paid by the corporation upon sale or other disposition of the assets. (Had the Judge reviewed and the taxpayer’s expert argued referencing the Estate of Jelke, which provided a dollar for dollar discount for the embedded gains, a greater discount may have been achieved.)
In this context, the court considered the two experts’ valuations in turn:
1. Built-in capital gains taxes. In calculating his discount for built-in capital gains tax related to both entities, the estate’s expert not only reviewed historic asset sales but also board meeting minutes. He also spoke with current management about their plans for future sales. He then estimated asset turnover rates, projecting a 5-year holding period for LRC and 8 years for LSC. After estimating appreciation and applying capital gains tax at the projected sale dates, discounted back to present values, he reached his discounts of 17.4% and 23.6%, respectively.
By contrast, the IRS expert did not talk to management, and he used turnover rates based solely on historical asset sales. For LRC, he used a 1.86% turnover rate to project an asset-holding period of over 53 years. Because LRC had elected S Corp status, the expert did not include any capital gains tax liability beyond 2009 (ten years following conversion). For LSC, he used a 3.45% asset turnover rate, resulting in a 29-year holding period. By applying similar calculations to each entity (multiplying capital gains tax by gains on asset sales as of the valuation date, discounted by a ratable portion over the relevant holding periods), he reachedhis 2% discount for LRC’s NAV and 8% for LSC. (This is ludicrous and a classic move by the Service’s expert to create facts that are not in evidence by applying the argument of deferring the liability. When such tactics are employed by the expert, our counter is to thank them for demonstrating the influencing of the controlling interest holder and protracting the time that an economic benefit to the minority interest holder would be received. This extended holding period ramps up the discount for lack of marketability due to the uncertainty of the market before a cash liquidity event would occur. Once this extends past 12 months, the discounts grow considerably greater than 25%!)
The court’s findings: Given the “highly appreciating non-operating investment assets” that both companies held, the court considered it likely that a hypothetical buyer and seller would negotiate “significant” discounts related to the built-in capital gains tax liability. Further, the assumptions by the estate’s expert related to asset turnover rates were based on more accurate data, especially his conversations with management and review of current sales. The IRS expert, on the other hand, did not account for appreciation during the holding periods and looked only at historic data. For these reasons, the court accepted the estate’s expert’s discounts for built-in capital gains tax, without adjustment.
2. Lack of control. To determine the discount for lack of control (DLOC) for LRC’s securities holdings, the estate’s expert used closed-end funds, observing a median 7.16% discount among the data. He used REITs (real estate investment trusts) and RELPs (real estate limited partnerships) in relation to the firm’s farm holdings and found an average 25.5% discount. (Frankly, the closed-end fund discount sounds particularly low given the level of earnings described. It would be considerably higher, if one knew the size of the other interests held and if the entity was encumbered with debt.)
The estate’s expert then looked at factors particular to LRC and assigned each a value between -1 (poor investor rights) and 1 (excellent rights). For instance, a 43.1% stakeholder in LRC would have some ability to force liquidation and change operating policies, and these factors received a zero (neutral) value. The company’s historical returns were substantially below those of similar investments, and this received a -1 value. Using these factors, weighted for LRC’s combined asset classes, he calculated a 14.8% DLOC for the estate’s interest. (It would have been wiser to reflect the level of return a direct or controlling owner could have expected in alternative investments and adjusted the actual interests accordingly on first a control, then a non-control basis. Such, a financial model provides a degree of irrefutable precision that is seldom successfully defeated by the Service.)
For its interest in LSC, the estate’s expert used closed end funds and observed the mean, median, and standard deviation of the discounts. He then assigned values from -1 to 1 to each of the factors that he also used to assess LRC; for example, because a 22.96% stakeholder would have little influence on operations or liquidation, this factor received a -0.5, but the company’s financial efficiency and historical returns received a 0 value. After determining a DLOC of 12.23%, unweighted by asset class, the expert reduced this to account for LSC’s small percentage of cash and short-term investments, resulting in an 11.9% DLOC.
According to the IRS expert, a DLOC is generally required “only if” the buyer intends to change the entity’s operations. Because LRC’s assets were performing well, a buyer would not expect a large discount for lack of control. Notably, in considering the entity’s securities holdings, the IRS expert did not break down his analysis by asset class, but looked at closed end funds to find an average 3.4% discount. Since the standard deviation was 17%, he “trimmed the mean” (removed the top and bottom 10%), resulting in a “trimmed” 5.2% average discount. Given the estate’s sizeable 43.1% interest in LRC, the IRS expert believed that a DLOC below 5% was appropriate for its marketable securities. (Should the operating agreement require a simple or super majority and with knowledge of the other interest holders block of shares, it would be easier to determine the existence and level of influence on the operating entity, if any. Assuming the controlling interest had operating as well as governance authority and was adhering to its fiduciary duty, it is unlikely the 43.1% interest holder could meaningfully influence management or governance.)
When considering the entity’s farmland holdings in LRC, the IRS expert reviewed a variety of data, including Mergerstat, and noted a range of 17% to 20% DLOC. Nevertheless, because discounts for public takeovers are generally higher than those for “normal” sales activity, LRC’s farming assets merited a DLOC lower than the Mergerstat range, or 15%. Even though the farmland comprised the bulk of the firm’s NAV, he averaged the two findings (5% and 15%) to conclude an overall DLOC for the estate’s 43.1% interest in LRC of 10%. (If anything, the DLOC’s for closely-held entities are greater as adherence to regulated activities and audited financial statements are not required.)
For LSC, the IRS expert once again used the 5.2% “trimmed mean” from the closed-end funds. Because the estate’s 22.96% interest was the single largest block of stock, its returns were good—and a purchaser would not want to change operations, a hypothetical buyer “would place no value on control,” he believed and a “nominal” DLOC of 5% was appropriate.
The court’s findings: The court noted that both experts calculated similar DLOC for LRC’s farming assets (15.7% vs. 15%), and both used lower-than-average discounts for its securities; and both averaged the respective discounts in their overall determination of DLOC. But only the taxpayer’s expert used a weighted average to account for LRC’s more significant holdings of farm property, while the IRS expert used a straight average. “A straight average would have been appropriate if LRC’s farmland and securities holdings were roughly equivalent,” the court said, in determining that the estate’s expert’s 14.8% DLOC was more appropriate.
With regard to the DLOC for the estate’s LSC stock interest, the IRS expert used the same 5% discount as he did for LRC’s marketable securities, without accounting for the estate’s much smaller interest in LSC. The estate’s expert accounted for the size difference, and once again, the court adopted his 11.9% DLOC for the 22.96% interest in LSC.
3. Marketability discount. To calculate LRC’s discount for lack of marketability, the estate’s expert looked at restricted stock studies and observed a range from 10% to 30% for larger, profitable companies and a 30% to 50% range for smaller, riskier companies,. Once again, he assigned average investor values of -1 to 1 to each class of LRC’s assets (e.g., its farmland was -0.5 and its marketable securities -0.125) to reach a DLOM for the estate’s 43.1% interest of 36%. (Shockingly, here’s where we depart company. The taxpayer’s expert should have been focused on expected total return and when such a return could be realized. As nominal earnings were received and a case for holding period could be made, the discount should reflect these factors among another two dozen, so the level of discount is directly correlated with return expectation, instead of a wide range (10% - 50%) of DLOM discount, which carries less credence with the trier of fact.)
He also used the same restricted stock studies for the LSC interest. After assigning average values to its assets classes, such as cash and short term investments (-0.5) and marketable securities (0), he concluded a DLOM for the estate’s 22.96% interest in LSC of 29.7%
The IRS expert looked at restricted stock studies, including three from the 1990s that the estate’s expert did not consider, and observed an average 25% DLOM. Because he believed the data did not account for liquidity and “corporate distress,” he also looked at private placement studies (more indicative of a “true” discount, he said), and observed discounts ranging from 7.23% to 17.6%. He then adjusted for entity-specific factors, such as LRC’s dividend-paying policy, the estate’s sizeable interest, and stock transfer restrictions, to determine an 18% DLOM for the estate’s 43.1% interest.
He reviewed the same studies with reference to LSC, and because these assets were more readily ascertainable and saleable, its earning history was consistent and its management competent, he determined a lower than average discount of 10% for the estate’s 22.96% interest.
The court’s findings: The court considered it appropriate to weigh the assets of both entities by class, but believed that both of the estate’s expert’s DLOMs were too high, particularly when combined with his discounts for lack of control (Admittedly, there is likely some overlap, but the size of discounts are directly related to entity/asset class performance and will be high or low based upon alternative returns as of the date of value. Needless to say, the more market and asset type volatility (futures for example), the greater the likely discounts). Some of the restricted stock data the estate’s expert used were aged, and his discounts were higher than the average benchmark studies that included “all components of a lack of marketability discount,” according to the court (Mixed opinion here, as the level of discounts will differ even if the assets were identical because they’re compared to alternatives as of a given period of time). Finally, the estate’s expert had performed a valuation for the same entities in connection with a 2000 gift estate tax return, in which he determined a “significantly lower” discount for the estate’s interest in LSC (21.4% in 2000 vs. 29.7% in 2001). As a result, and without further discussion, the court concluded DLOM for the estate’s respective interests in LRC and LSC of 25% and 20%.
Overall, the court found that the fair market value of the estate’s 43.1% interest in LRC was $7.546 million, and its 22.96% interest in LSC was worth $6.530 million.