Family holding entities have become popular as a way to diversify and pool assets, manage and steward them long term, introduce the finer points of investment management to younger or less financially savvy members of a family and transfer assets from older to younger generations.
Pass-through entities, such as partnerships or limited liability companies, tend to be the vehicles of choice for such asset pools because of the ability of such entities to avoid double taxation, that is, taxation of dividends made to shareholders from earnings that have themselves already been taxed at the entity level. But, what if a less advantageous entity structure has been chosen, such as a Subchapter C corporation? In such entities, “trapped” capital gains can accumulate over time, giving rise to a substantial capital gains tax liability that would apply on the liquidation and sale of the entity’s assets. And that, in turn, gives rise to a valuation problem. Classic cases, such as Estate of Jelke v. Commissioner1 and Eisenberg v. Comm’r,2 have struggled with the same issue: To what extent should the fair market value (FMV) of an interest in a C corporation with built-in gains be discounted for this tax liability?
This is an area of tax controversy in which consistency among various courts has been sadly lacking. Recently, in Estate of Richmond v. Comm’r,3 the Tax Court contributed anew to this controversy in an opinion that may end up further widening the circuit split among U.S. Courts of Appeal on this issue.
In Estate of Richmond, the court:
1. Considered, but ultimately rejected, the use of the income approach (that is, using a capitalization-of-dividends approach, which is a method whereby the market value of the entity is determined by calculating the present value of a future income stream) for an asset-holding entity;
2. Applied significant valuation discounts for lack of marketability and lack of control; and
3. Assessed significant tax penalties.
The court’s decision is likely to strike fear in the hearts of many taxpayers.
The case revolved around: (1) the valuation of shares in Pearson Holding Co. (PHC), a marketable securities holding company, and (2) the misvaluation penalties that the Internal Revenue Service wanted to impose on the estate.
PHC was founded in 1928 to manage a pool of investments, primarily equities, for the descendants of Frederick Pearson and provide them with a steady stream of income while minimizing taxes and preserving capital. The entity’s investment management had been successful over the years, and its portfolio had grown in value. As of the decedent’s date of death, in December 2005, the assets of the company included more than $50 million of common shares, mostly high-yielding large-cap stocks, and it had near-zero liabilities. The dividend income paid by PHC had also grown, at a rate slightly above 5 percent per year since 1970.4
The decedent, Helen P. Richmond, passed away holding 548 shares in PHC. She was the second largest shareholder, with 23.44 percent of common stock outstanding at that time. Over time, the ownership of the company had become increasingly diffuse, with more than 25 shareholders, whose interests ranged from 0.17 percent to 23.61 percent. As the court noted, the shareholders were:
… heirs and legatees with less and less identification with Frederick Pearson and his philosophy and goals, and with less and less knowledge of and affinity for one another. The mindset of the shareholders as owning a family company would more and more give way to an attitude that regards the PHC shares simply as an investment …5
The court concluded that a purchaser of PHC stock would expect the management of the portfolio to become less passive over time, leading to increased “churn” of its holdings.6 PHC had a long-standing policy of minimizing the turnover of its portfolio, so most of its shares had been held for very long periods of time. As a result, most of the portfolio’s holdings were highly appreciated positions—to such an extent that PHC’s built-in capital gain (BICG) was $45.6 million as of the date of death, or more than 87 percent of its value. This situation, in turn, gave rise to a (stipulated) potential capital gains tax liability for the BICG of $18.1 million. The Internal Revenue Service’s trial expert suggested in his testimony that it would be reasonable to expect a complete turnover of the portfolio in 20 to 30 years, and the court accepted this estimate.7
Because of PHC’s growing ownership diffusion and the passage of time, there had also been several changes in its shareholder composition. From 1971 through 1993, there were nine transactions involving the sale or redemption of PHC stock. The court doesn’t go into detail about these transactions. Perhaps they weren’t arm’s length and, thus, not great indications of FMV. Yet, when the court notes that shareholders, with the passing decades, had “less and less knowledge of and affinity for one another”—doesn’t that indicate that shareholders’ relationship with one another had become increasingly arm’s length? It certainly seems as if the prices paid in those transactions might have provided useful data for this valuation. But, there’s no mention in the court’s opinion of these prices or what discounts from net asset value (NAV) were implicit in them, except for the fact that they appear to have been determined using an income approach.8
The Tax Return
The executors, including at least one CPA, timely filed Form 706. The estate’s interest was valued at $3,149,767 using the capitalization-of-dividends method. The NAV of PHC was stipulated at $52.1 million. The indicated NAV attributable to the estate was, thus, $12.2 million. In other words, the estate valued the interest at effectively a 74 percent discount from NAV. The IRS valued the interest at $9.2 million in its notice of deficiency, and the estate filed its petition shortly thereafter.
The Form 706 value was based on an appraisal by Peter Winnington, an accountant, and he, in turn, based his valuation on information relevant to PHC and the estate’s interest, including: the transactions in PHC stock discussed previously, prior valuations performed for past estates and general corporate and financial information on PHC and its portfolio already known to Peter as PHC’s accountant. Because Peter’s qualifications to perform this appraisal became a point of contention, it’s important to note that he had:
• received a Bachelor of Science degree in accounting and a Masters degree in taxation.
• 20 years of experience as an accountant, including audits, management advisory, litigation support and tax planning and was a CPA who held a certified financial planner designation.
• chaired his firm’s corporate service department and sat on the firm’s executive committee.
• a membership in the American Institute of CPAs, the Delaware Society of Certified Public Accountants and the Wilmington Tax Group.
• appraisal experience, having written 10 to 20 valuation reports and testified in court.
Peter provided an unsigned draft valuation report to the estate but was never asked to finalize his report. It should be noted that using an income approach to value the estate’s interest in PHC overlooks the readily available market prices of the underlying securities. Thus, the income approach is best applied when valuing operating companies.9 Yet, a growing number of advisors also have become advocates of the income approach for holding companies.
In our opinion, the Richmond court is correct when observing:
The estate’s valuation method therefore ignores the most concrete and reliable data of value that are available—i.e., the actual market prices of the publicly traded securities that constituted PHC’s portfolio.10
That’s really the crux of the matter—the market has already considered the income approach! The question for the appraiser with PHC’s stock as his task is simply to consider the discount from the market’s valuation. That too is quite difficult, but doesn’t get easier by trying to reverse engineer the market’s valuation of the portfolio.11
By the time the case went to trial, the parties had modified their positions. The estate now valued its shares at $5 million, based on a new appraisal from a new expert, Robert Schweihs, who also testified. Robert valued the estate’s interest using a capitalization-of-dividends method under the income approach (similar to Peter), as well as the NAV method under the cost approach. Under the NAV method, Robert reduced PHC’s NAV by 100 percent of the BICG tax and applied an 8 percent lack-of-control discount and 35.6 percent lack-of-marketability discount to the estate’s interest. The IRS valued the shares at $7.3 million, but also wanted to apply a 20 percent penalty for substantial valuation misstatement.
The BICG Discount
It’s well established that shares in a C corporation with significant amounts of “trapped” BICG that are subject to corporate taxes are worth less than shares in similar companies that don’t have such a potential tax liability. How much less is the question.
Our court system has a poor track record in consistently resolving this issue. A widening circuit split has opened between two warring camps:
• The U.S. Courts of Appeals for the Fifth and Eleventh Circuits hold that the contingent BICG tax liability should be deducted from the value, dollar-for-dollar, as if the portfolio was to be liquidated in its entirety on the valuation date.
• The Second and Sixth Circuits and the Tax Court all hold that the valuation shouldn’t assume immediate liquidation, but instead discount the value of future taxes due upon liquidation to the present using a discount rate (the present value approach). As justification for this approach, the court considers methods that the entity may use to maximize value by mitigating, or at least deferring, the gain and the payment of the tax.
If a court decides that a dollar-for-dollar discounting is appropriate, it must decide which alternative method to apply: a present value approach or some other method that provides a discount for the BICG, perhaps applying market transaction data from entities with BICGs that suffer from the same or similar tax disadvantages as the subject entity.12 If the present value approach to determine the detriment in value related to BICG taxes is used, how should we estimate the future cash flows? Depending on how future taxes upon liquidation are estimated, the present value approach might yield discounts that are just as high as—or even greater than—the discount from the dollar-for-dollar method.
If the case is ultimately appealable to a “dollar-for-dollar” circuit, the issue is simple. As the Fifth Circuit noted in Dunn:
As the methodology we employ today may well be viewed by some … as unsophisticated, dogmatic, overly simplistic, or just plain wrong, we consciously assume the risk of incurring such criticism from the business appraisal community … we observe that on the end of the methodology spectrum opposite oversimplification lies over-engineering.13
Discounting Value of Future Taxes
If a case is appealable to a circuit, like the Second or Sixth, which discounts the value of future taxes, the task becomes much more difficult. There’s almost no data available to rely on, and valuation theory provides little helpful guidance. The discounts can be determined using two main approaches: market data and discounted cash flow (DCF).
Market data. Two methods have been promoted to discount the value of future taxes based on this approach: a real estate investment trust (REIT)-based method and a closed-end fund-based method. The earliest of these solutions, which wasn’t presented to the Tax Court in Richmond, compares the trading market discounts for REITs and real estate holding C corporations.14 The greatest drawback to this method is the staleness of the data, since REITs no longer report reliable appraised values for their portfolios of properties (and haven’t for some time).15 Another drawback is that there are very significant differences between REITs and real estate holding C corporations, beyond just their relative tax dis/advantages.
The other market data method is an analysis of closed-end funds. This, however, is a blind alley because closed-end funds, typically, aren’t subject to double taxation in the first place.16 Thus, differences in the discounts for different closed-end funds are almost certainly not related to the BICG at all, and the approach has consistently failed to impress triers of fact on this issue.
DCF. The main reason often cited for rejecting the dollar-for-dollar methodology is that the management of a C corporation can defer the payment of BICG taxes for a very long time, which may serve to diminish the tax burden.17 The taxes may be deferred but not avoided altogether—at least not for investment holding entities.18 But, does the DCF method really serve to differentiate the BICG tax discount from the dollar-for-dollar approach? It didn’t in Jensen v. Comm’r,19 in which the DCF method applied by the Tax Court in its own calculation yielded a BICG tax discount that was greater than the one using dollar-for-dollar. In fact, with realistic assumptions, the DCF method ought to yield BICG tax discount numbers that are close to those of the dollar-for-dollar approach. To see why, consider:
1. The value of any investment asset, such as the portfolio held by PHC in Richmond, encapsulates assumptions about future returns—and those future expected returns are generally assumed to be equal to the required rate of return on the investments, which in turns equals their discount rates.20
2. By waiting to take gains, the present value of the taxes due on the gains is reduced due to the time value of money. However, the future value is increased, due to the expected return on the investments held. These offsetting adjustments may turn out to be approximately equal.
Based on this analysis, any reduction in the BICG tax discount from the dollar-for-dollar approach yielded by the DCF method may be due, in great part, to inconsistent or poorly developed assumptions about future growth rates and discount rates, rather than a reflection of the real economics of deferring gains (and taxes). Which may bring us back to the Fifth Circuit’s “simplistic” and “unsophisticated” approach—it’s possible that a more sophisticated approach might not be more accurate.
Richmond’s BICG Solution
The court in Richmond simply noted that its opinion is appealable to the Third Circuit and found:
[T]he seller of the company with the contingent future liability would demand a higher price than the seller of a company with the unconditional current liability. As a result, despite contrary holdings by some courts, we find that a 100% discount would be unreasonable, because it would not reflect the economic realities of PHC’s situation.21
The IRS’ expert developed a discount equal to 43 percent of PHC’s BICG tax liability based on a questionable methodology (analyzing the gains of closed-end funds) that the court didn’t accept. However, the court did consider this substantial discount as an admission on the IRS’ part. The taxpayer’s expert applied a discount of 100 percent of the tax liability.
The court’s own analysis provided a BICG tax discount range that bracketed the IRS’ “admitted” 43 percent; thus, it decided that this was the correct amount. The court noted that PHC, despite its long-standing investment policies, had been advised in the past to diversify more, which would increase portfolio churn. Projecting taxes due on liquidation of the portfolio over those time frames and discounting those tax payments back to the present at a range of discount rates, yield a value range for the liability.22
Marketability and Control Discounts
Both parties used data from closed-end mutual funds when estimating the discount for lack of control, a widely accepted technique. The IRS’ appraiser took the average discount for the funds and “rounded down” to 6 percent. The taxpayer’s expert took the median discount of 8 percent. The court disagreed with both, removed three outliers from the data and took the average of the remaining sample—deciding on a 7.75 percent discount.
Very little analysis appears (from the court’s opinion) to have gone into considering the differences between closed-end funds and PHC. In our opinion, that’s a mistake. First, the sample was 59 funds, which probably could have been narrowed further to match the investment profile of PHC’s portfolio.23 Second, closed-end funds have many advantages that may mitigate the control discount, such as professional management, public oversight, a professional board and Security and Exchange Commission-reviewed financial reporting. Third, PHC is almost certainly smaller than the average fund, which may also increase the discount.
Both parties used published studies of marketability discounts, including restricted stock and pre-initial public offering studies. However, all they considered were the averages published by the studies, rather than the unique characteristics of PHC and how those characteristics would impact the discount—despite the well-established fact that the range of discounts indicated by such studies is vast. The IRS’ expert chose a discount at the very bottom of the range of the study averages, while the taxpayer’s expert chose a discount at the very top of the range. The court was unconvinced by either side and simply applied the average. It’s impossible to fault the fact-finder for doing so, when the analyses presented were so conclusory.
The Penalty Phase
Finally, the Richmond court simply noted that its concluded value of $6.5 million is more than double the value reported on the Form 706 and imposed a
20 percent accuracy-related penalty for the underpayment. Was this an open and shut case for a “substantial” valuation understatement penalty? That depends on whether the taxpayer had reasonable cause for making the error and acted in good faith.
The court agreed with the taxpayer that valuing the estate’s interest in PHC is difficult.24 It follows logically that getting it wrong, in and of itself, can’t be sufficient reason for the penalty—it requires genuinely bad behavior (as has been held in many prior cases). Because the estate relied on competent legal and accounting advisors—and an appraisal—what went wrong here? The court notes that: “to establish good faith, taxpayers cannot rely blindly on advice from advisers or on an appraisal” and continues:
On the record before us, we cannot say that the estate acted with reasonable cause and in good faith in using an unsigned draft report prepared by its accountant as its basis for reporting the value of the decedent’s interest in PHC on the estate tax return. Mr. Winnington is not a certified appraiser. The estate never demonstrated or discussed how Mr. Winnington arrived at the value reported except to say that two prior estate transactions involving PHC stock used the capitalization-of-dividends method for valuation. Furthermore, the estate did not explain—much less excuse—whatever defects in Mr. Winnington’s valuation resulted in that initial $3.1 million value’s being abandoned in favor of the higher $5 million for which the estate contended at trial. Consequently, the value reported on the estate tax return is essentially unexplained. (Emphasis added.)25
As the court concluded, “even by the estate’s lights, the value on the estate tax return needed explaining, but no explanation was given.”26
Both sides in this case were ably assisted by some of the best counsel available and well-known and established valuation advisors. Yet, some aspects of this case, at least on a first read, don’t seem to make a great deal of sense.
The court’s analysis of the misvaluation penalty issue will probably gain the most immediate attention. Since valuation is a tricky issue, experts whose “day job” is something very different (accounting, in this case) probably would be better off not dabbling in the valuation arena, as would their clients. Having said that, this decision seems unduly harsh, and it seems it may have an uphill battle surviving on appeal. Peter simply wasn’t that badly qualified. He may have gotten the valuation wrong and chose poorly when deciding his valuation methods, but if these kinds of errors are enough to make it unreasonable for taxpayers to rely on an appraisal, then there are many taxpayers who should be alarmed today.
The case is also highly instructive on the BICG discount issue and, to a lesser degree, on marketability and control discounts. It’s surprising to see so little effort apparently made on the latter two discounts in particular, when even a few points either way makes a significant difference to the outcome—especially when penalties apply on top.
1. Estate of Jelke, T.C. Memo. 2005-131, filed May 31, 2005 and Estate of Jelke v. Commissioner, Case No. 05-15549 (Nov. 15, 2007).
2. Eisenberg v. Comm’r, T.C. Memo. 1997-483 (Oct. 27, 1997).
3. Estate of Richmond, v. Comm’r, T.C. Memo. 2014-26 (Feb. 11, 2014).
4. Dividends running at about $1 million per year lately, or about 2 percent. Pearson Holding Co.’s (PHC) common stocks comprised 97.2 percent of the portfolio. The common stocks were invested in 10 major industries, with 42.8 percent of the stock concentrated in four large-cap companies. The remainder of PHC’s portfolio consisted of government bonds and notes, preferred stocks, cash and equivalents, receivables and a security deposit.
5. Richmond, supra note 3, at p. 7.
6. For example, the court found evidence that Wilmington Trust Co., PHC’s advisor, had suggested PHC diversify more, which would require a significant portion of the portfolio to be sold.
7. Compare with 70-year historical rate.
8. The court also noted that the stock was valued, presumably for inclusion on an estate tax return, for an estate in 1999, and that valuation also was performed using an income approach. However, the discussion of Peter Winnington’s appraisal refers to “some valuations for prior estates that included PHC stock,” which makes it seem as if more than one estate was involved.
9. The method is also highly sensitive to the assumptions chosen. In fact, the taxpayer’s appraiser had published valuation textbooks stating that the income approach is highly sensitive to small variations in the discount rate.
10. Richmond, supra note 3, at p. 24.
11. In addition to the problems the court identified, the method also ties the value of the holding company almost entirely to its dividend policy—what proportion of the annual return of the portfolio does the company pay out to its shareholders. This “lever” then controls the valuation to a very considerable extent. The approach can yield values that are extremely low.
12. To distinguish, a present value approach used to determine a reduction in value to an entity to account for built-in capital gains (BICG) taxes is different from using an income approach, such as the capitalization-of-dividends method, to value an asset-holding entity as a whole as done by the estate’s appraisers and discussed in the prior section.
13. Dunn v. Comm’r, No. 00-60614 (5th Cir. August 1, 2002) at p. 38.
14. See, for example, Carsten Hoffmann, “Life After Davis: Discounting for Built-In Capital Gains Tax Liabilities,” Taxes (August 1999).
15. Real estate investment trust discounts used to be applied as a methodology, in fact often the principal methodology, for determining control discounts for non-controlling interests in real estate holding entities, including family limited partnerships. This method is now rarely used because of the unreliability of the data.
16. To avoid taxes at the fund level, a closed-end fund must meet Internal Revenue Service requirements for sources of income and diversification of portfolio holdings and must distribute substantially all of its income and capital gains to shareholders annually. (See Investment Company Institute’s “Guide to Closed-End Funds”). Therefore, taxes from the distribution of dividends, income and capital gains essentially “pass through” to closed-end fund shareholders. In other words, since such funds are generally not subject to double-taxation, the market is unlikely to reduce the value of a fund for BICGs. Measuring the difference in discounts between high BICG and low BICG funds, therefore, is unlikely to be helpful for this analysis.
17. Sales that trigger capital gains might be deferred to more opportune time periods when, for example, a shareholder is in a lower bracket or when shareholders have tax losses to offset.
18. Investment holding entities, as opposed to operating entities, are unlikely to qualify for S corporation status. Taxes due on possible BICGs for an operating entity might be avoided altogether if the entity elects to be taxed as an S corporation and waits the required 10 years before liquidating assets and triggering the gains.
19. Jensen v. Comm’r, T.C. Memo. 2010-182 (Aug. 10, 2010).
20. There are exceptions to this rule, for example, when assets are illiquid, but that’s not the issue here. The question analyzed by the discounted cash flow method in these cases is whether there’s a benefit to holding assets, rather than selling and taking gains immediately.
21. Richmond, supra note 3, at p. 32.
22. The range of discount rates was 7 percent to 10.27 percent, based on various methods. The resulting value range was $5.5 million to $9.6 million, as compared with the IRS’ admission of $7.8 million (which is 43 percent of the stipulated $18.1 million BICG tax liability and 15 percent of the entire net asset value of the company).
23. For example, the funds sample may be narrowed down by size, yield and investment strategies (for example, sectors, types of companies) to better match the subject company.
24. The estate notes the wide dispersion of values computed for the interest, from just over $3 million to well above $9 million as evidence for that. However, once more well-established valuation experts started working on the valuation for the trial, the differences narrowed substantially.
25. Richmond, supra note 3, at pp. 49-50.
26. Ibid., at p. 51.
Jennifer Cossette is vice president in Pluris Valuation Advisors LLC’s San Francisco office.
Espen Robak is President and founder of Pluris Valuation Advisors LLC.