September 01, 2013

Filed April 8, 2013

In a recent decision, the Tax Court ruled in favor of the IRS and its expert, Francis Burns, in the valuation of a 50.50% interest in an Ohio limited liability company. Referred to as CI LLC, this entity had a net asset value of approximately $318 million at the valuation date, March 3, 2005. The assets were primarily cash and were derived from the sale of the Koons’ family business, Central Investment Corp. (“CIC”), a Pepsi-Cola bottler. The sale was to an affiliate of PepsiCo, Inc., and was the result of several years’ litigation between the business and PepsiCo.

CI LLC was set up in August 2004 in contemplation of the sale of the business and was to be the primary holding entity for the cash from the sale and other non-Pepsi assets. The sale closed on January 10, 2005. The stock purchase agreement had conditions and requirements continuing until January 2012.

The Koons’ four children (all middle-aged adults) held stock in CIC indirectly through the Koons Family Trust set up in 1980. On December 21, 2004, CI LLC issued a letter to each of the children offering to redeem their interests in the trust. By February 27, 2005, all four children had accepted the offer, although during this time, John Koons (the decedent) made several important changes to his Revocable Trust (which was to hold the interest in CI LLC upon his death) and to the governing documents of CI LLC. John Koons transferred his interest in CI LLC to the Revocable Trust (“RT”) on February 10, 2005. Upon completion of the redemption, the Revocable Trust’s interest in CI LLC would be 70.93%. The redemption of the children’s shares had not been finalized at John’s death on March 3, 2005.

Taxpayer’s Valuation Argument

The taxpayers retained Dr. Mukesh Bajaj as their expert. Dr. Bajaj opined that no discount was appropriate for lack of control. He determined a discount for lack of marketability of 31.7%. His discount was determined by use of the regression analysis found in his 2001 study of 88 restricted stock transactions.1 According to Dr. Bajaj, this actually determined a discount of 26.6%. He increased the discount by 7% for special factors related to CI LLC. Dr. Bajaj reasoned that it was not reasonably foreseeable that the redemption would occur; therefore, the appropriate percentage if CI LLC to be valued was 51.5%. (This interest actually consisted of a 46.94% voting interest and a 51.59% nonvoting interest.)

IRS Expert’s Opinion

Francis Burns relied upon the IRS’s opinion that the signed acceptance of the redemption offer was a binding contract. The Court agreed that it was. Based on this opinion, the interest held by the RT rose from 50.5% to 70.93%. The voting interest would rise from 46.94% to 70.42%. Of course the purchase by the RT would transfer significant dollars to the children such that there was essentially no immediate monetary gain by the RT. Disregarding discounts, it would hold the same net monetary amount either way.

What was significant, however, was how the redemption of the children’s interests would confer control upon the holder of the Estate’s 46.94% voting interest. This control is the linchpin of Mr. Burns’ valuation argument. Holding a majority of the voting interest in CI LLC allows the holder to change the provisions of the membership agreement. Specifically, the existing limitations on distributions put in place by John Koons could be abolished. While CI LLC had continuing obligations to PepsiCo to maintain a certain level of cash and assets in the entity, the holder of a 70.42% voting interest would have the ability to immediately distribute a very large amount of cash. The balance could be distributed in seven years. Based on the foregoing, Mr. Burns believed an appropriate discount for lack of marketability would be in the range of 5-10%, or, 7.5%. The Tax Court agreed.


It seems that the appropriate discount applied in determining the Fair Market Value of the subject interest may lie between the opinion of the two experts. Dr. Bajaj’s approach does not seem to be the right methodology for the set of facts present here. However, in fairness to Dr. Bajaj (and the same might be said for Mr. Burns), we do not have his report and we have no other information than that contained in the opinion. Judicial opinions are by their nature condensed versions of the facts and, without becoming encyclopedic, not every bit of relevant information can be included. With this caveat, we still believe that we can form meaningful and informed conclusions about what is in the opinion.

Regarding Dr. Bajaj’s opinions, we cannot help but be struck by the irony of the Tax Court attacking the same analysis from the same expert who provided the Court’s valuation foundation in its opinions in McCord, Lappo, and Peracchio. What the Court rejected about the Bajaj opinion in McCord was the fact that he excluded from his discount the “liquidity” factors (asymmetrical information, monitoring costs, etc.). We argued (and the Court agreed) that, whatever you call these factors, they still exist and they are impediments to value. Here, however, he includes them.

In Koons, the Court says the Bajaj study is inappropriate because it is based on 88 transactions of companies engaged in active business, and CI LLC had but two small operating businesses. This criticism is especially puzzling. The exact same thing can be said about every restricted study used in every case involving an investment entity. The McCord Court went to great lengths to point this difference out and is the primary reason they chose the middle grouping of the Bajaj restricted stock study as its basis for its lack of marketability discount. The Tax Court has followed this precedent almost religiously.

The next criticism of Dr. Bajaj’s methodology was that the Bajaj regression only explained one-third of the variation valuation discounts. This was the same argument this author made in the McCord case and which other experts made in Lappo and Peracchio. The Court was unmoved by these objections in all of these cases.

The Court also cited their agreement with Mr. Burns that the block size (and, no doubt, the issue of control) was not adequately addressed by the Bajaj regression equation. Here, the Court and Mr. Burns would appear to have a valid argument. The possibility of gaining absolute control has to be a very positive characteristic that does not seem to be addressed in the Bajaj analysis.

The court’s wholesale adoption of Mr. Burns’ opinion, in our view, is inappropriate. Mr. Burns’ opinion basically ignores the significant financial risks to the willing buyer here. It seems entirely implausible that a willing and unrelated third-party buyer would, as the Burns opinion would have us accept, so casually step into the middle of the unfinished redemption offer and family dissention over the future control of CI LLC. That the Court can categorically and unreservedly assert that signatures on the redemption offer letters equate to a closed, funded transaction seems unsupportable. The signs of restlessness and unhappiness in the father-children relationships were quite apparent. This is while the domineering father was still alive. One can only imagine what might have surfaced if, after the death of John Koons, the children learned that his interest was to be sold to an outside third party.

With regard to the potentiality for litigation, consider that the offer was first made December 21, 2004. The children were given 90 days to agree to have their interests in CI LLC redeemed. By January 27, 2005, two children had accepted the offer but two had not. Then, it appeared John Koons upped the ante by making the prospect of continuing as a CI LLC member less attractive.

On February 4, 2005, John Koons changed the provisions of the RT by removing the children as beneficiaries. None of the children were trustees of the RT. Under its provisions, a retiring trustee was to be replaced by the action of the remaining trustees. No replacement trustee could be a descendant of Koons.2

Another move, which had the appearance of making even more unattractive the prospect of any child remaining as a member in CI LLC, was seen on February 18. On this day, John Koons eliminated the Board of Advisors – a body made up of the children and with whom the Board of Managers was required to consult. He then eliminated his children as permitted transferees of CI LLC membership interests. Lastly, he limited distributions for the first 15 years of CI LLC’s existence.

This last move must have been seen as particularly ominous as it signaled that the trustees were being charged by John Koons with managing the Company for the long haul. The clear message to the trustees was to invest and operate, not to preserve capital, pay out distributions, and plan for liquidation.

On February 27, the last of four children signed the redemption offer letter. On the part of the children, there seemed no reason not to acquiesce. When one considers the voting power of the other trusts along with the subject interest, the trustees already had majority control of the Company. The children had no means to change this.

On March 3, 2005, John Koons died. The question at hand was: “What would a willing buyer pay for the 50.50% total interest in CI LLC held by the RT?”

At the valuation date, the redemptions were only in the form of a signed letter. The Tax Court decided that these were legally enforceable. But, given the sequence of events, one has to ask: “Had the children been properly represented by independent counsel? Had full disclosure been satisfied? Was the proposed offer fair? Did the actions of John Koons constitute minority oppression?” On the other hand, even if the children were forced into the redemption, the fact that they were going to be cashed out at net asset value (“NAV”) (since most of the assets were in cash), seems to take away any obvious claims of financial harm.

It was not the proposed redemption but the prospect of the control of CI LLC in the hands of the RT trustees that appeared to have the most likelihood of controversy. This concern prompted one son to warn his father in a letter that there would surely be future litigation. His complaint, which we can only presume echoes the sentiments of at least some of his siblings, was that he believed the board of managers of CI LLC would, rather than invest passively in marketable securities, invest in operating businesses or other risky and illiquid investments.

The valuation process requires us to assume that a hypothetical third party will now step in to buy the combined 50.50% interest. This hypothetical situation, of course, never happened. However, if it had, instead of the children acquiescing to the redemption, an entirely new reaction might have been seen.

The valuation premise suggested by the IRS and Mr. Burns and endorsed by the Court was that the third-party buyer could merely step into the shoes of John Koons, pay himself a $140 million dividend, redeem the children, and then liquidate the rest of the business. Just from the information on the record, even if the buyer was presumed to have gained control, Mr.Burns’ value appears a bit overstated.

In the first place, the distribution payable is overstated. If the Company, with a NAV of about $318 million, made a $278 million distribution, it would not have sufficient funds to complete the redemption of the children’s interests and maintain $40 million in the Company as was required by the stock purchase agreement. According to our calculations, this distribution could be no more than $262 million.3 The foregoing analysis is predicated on the Company’s assets being cash or cash equivalents. Of the Company’s total assets of approximately $351 million, $29 million were not liquid.4

Mr. Burns treated the $40 million reserve as if its value was a cash equivalent. In our view, the present value of this reserve, even if untouched at the end of seven years, would be significantly less than its booked value at the valuation date. In addition, since the reserved amount was so high, there must be a reasonably good likelihood that a significant amount would be claimed. The Tax Court implied that the expected claims against this reserve were “far less than $40 million.” It did not imply claims would be zero.

There could be a number of reasons the children would not want an unrelated third party buying their father’s interest. One motivation to oppose the sale by the children might be their concern over the redemption price. The price was to be determined by “CI LLC.” While John Koons was alive, there was no doubt who was in control. If a third party bought Koons’ 46.94% voting interest, no one would control CI LLC. It might be that the children would want to buy the interest and attempt to interfere with the sale to the third party. Without the children’s approval, the third-party buyer might only be able to buy an assignee (non-voting) interest as the operating agreement specified that 75% of the members must approve a transfer of a membership interest to a non-permitted transferee. If that were the case, the vote associated with the transferred interest would most likely reside with the trustees of the RT. They might vote as directed by the buyer or they might not. Other scenarios which might concern a potential buyer can also be envisioned.

If a buyer paid $148.5 million for the interest, the consequences of not gaining control are catastrophic. Without the hypothetical buyer owning a majority in CI LLC, if the buyer pays $148.5 million, he will wind up with a huge loss from a valuation standpoint. Now, instead of receiving an immediate distribution refunding most of the purchase price, he faces the prospect of 15 years and perhaps more of minimal distributions. Furthermore, he has placed an enormous amount of funds in what could be a financial “black hole.” The record provides no evidence of the investment savvy of the board of managers. Even if they were investment experts of the highest caliber, the Company had no known investment strategy and no track record. This was not an investment partnership with a set term and the investment was entirely illiquid. No reasonable investor would place a substantial sum of money into such an uncertain situation.

The value of the investment, with its $148.5 million starting point, would rapidly decline based on the negative annual capital carry. Without control, the willing buyer could expect only minimal distributions for at least 15 years. This could only be recovered by a large gain at some point in the future. At the valuation date, there was no way to assess that this was a reasonable likelihood.

Despite the doomsday scenario described herein, the risks to the buyer of not gaining control should not be overblown. Most arguments would come down on the side of the sale going through, the redemption occurring, and the Company ultimately being liquidated seven years hence. (As we have already stated, however, we believe the distributable amounts to the subject interest were overestimated by Mr. Burns.) But, even if the odds of not gaining control were small, the valuation consequences would be significant.

It is a fact that investment behavior does not follow straight mathematical odds (expected value). According to the “prospect theory” and the utility theory of money, as financial stakes get higher, investors become risk averse.5 At $148.5 million, the investor must be becoming quite risk averse.

There is not enough evidence on the record to assess what the downside case might be for the willing buyer. Indeed, even if one had all available information, there still may be no way to assess this. Given the very high financial stakes and the uncertainty on the downside, any prudent investor would significantly hedge the investment prospect of owning the subject interest.

The decision of the Tax Court cannot be right. It would seem likely that this case will be remanded in order to determine a more reasonable outcome.


1 Bajaj, Mukesh, Denis, David J., Ferris, Stephen P. and Sarin, Atulya, Firm Value and Marketability Discounts (February 26, 2001). Available at SSRN: or

2 It appeared that at the death of John Koons, the Revocable Trust was regarded as irrevocable.

3 The children owned 28.83% of the membership interests so their redemption amount would be about $91.7 million. If a $278 million distribution were paid out to all members, the children would receive $80.1 million, leaving a balance owed them in the redemption of $11.5 million. However, after this was paid to them, the Company would have but $28.5 million remaining – exactly $11.5 million below the required threshold of $40 million. After the distribution of $262 million but before the redemption, a balance of about $56 million is required.

4 Among these assets was a health fitness club.

5 “According to prospect theory, proposed by Kahneman and Tversky (1979), individuals maximize a weighted sum of a ‘value’ function. Decisions are made in terms of gains or losses rather than final wealth, and the ‘value’ of a loss is compensated for by two to three times the ‘value’ of a gain equivalent to that loss; hence, the notion of loss aversion.” Hwang, Soosung and Satchell, Stephen E., How Loss Averse are Investors in Financial Markets? (February 19, 2010). Available at SSRN: or