Agranoff v. Miller

791 A.2d 880, 2001 WL 536917

• Court: Court of Chancery of Delaware
• Decided: November 8, 2001
• Docket: Civ.A. 16795
• Status: Published

Opinion

OPINION

STRINE, Vice Chancellor.

This opinion addresses the remand order of the Supreme Court that required this court to determine the price at which certain warrants held by Banker’s Trust (the ‘Warrants” or Warrant Shares”) could have been purchased by either plaintiff Citicorp Venture Capital, Ltd. (“CVC”) or EMS Corp. (“EMS”) in 1998. The Warrants were purchased in autumn 1998 by defendant Edward M. Miller as part of Miller’s conspiracy with EMS director and president John Ovens, Jr. to acquire control of EMS through a variety of inequitable and deceptive tactics. These tactics included multiple breaches of EMS’s and CVC’s contractual rights to purchase other EMS shares, fraud on the EMS board by Ovens with Miller’s approval and knowledge, and the illicit use of EMS inside information by Miller that was funneled to him by Ovens without EMS board approval. In the case of the BT Warrants, Miller and Ovens worked together to ensure that BT would sell them to Miller, even though Ovens was supposed to be assisting EMS in negotiating an extension of those Warrants at the time he was instead operating as an agent of Miller’s secret takeover plan. Thus, Miller’s purchase of the Warrants was held to result from the usurpation of a corporate opportunity belonging to EMS. 1

The remand order requires this court to determine the hypothetical price at which EMS and CVC could have purchased the Warrants from BT. 2 In this opinion, I interpret the Supreme Court’s order as requiring me to determine the fair market value of the BT Warrants in accordance with standard valuation techniques, recognizing that Miller’s pattern of illicit conduct makes it impossible to determine what the Warrants would have been sold for in a market untainted by his multiple violations of EMS’s and CVC’s rights. After examining the testimony of the parties’ experts, I conclude that the fair market value of the BT Warrants as of the valuation date of October 1998 was $41.02. In the event that I have misinterpreted the mandate of the Supreme Court and it wished me to come up with a fair value appraisal award, I conclude that the fair value of the BT Warrants as of the valuation date was $51.13.

I. The Business Of EMS

EMS is a transportation delivery service business. A basic description of its status as of 1998 follows.

EMS is a holding company for 62% of the stock of Express Messenger Systems, Inc. (“Express”). The remaining 38% of Express is owned by Air Canada.

Express has three major business operations. The first, Express Messenger, is an express delivery and courier business operating in several western American cities. The second, Express Mail, operates as an international “remailer.” This means that Express Mail picks up bulk mail by messenger, transports it overseas, and injects the mail into local delivery systems or arranges for hand delivery of it. The third and final operation, California Overnight, is an overnight delivery business focusing on business-to-business deliveries in California. California Overnight originally exploited a niche for late pickups of next-day packages in California that was not being filled by national delivery companies, because of time zone issues. That advantage has disappeared, and California Oversight now faces both national and local competition.

Express enjoyed relatively strong revenue growth during the latter part of the 1990s, with revenues expanding from $21.8 million in fiscal year 1995 to $50.3 million in fiscal year 1998. During that same period, California Overnight’s share of Express’s revenues grew from 35.8% to 65.2%. This shift was intentional and reflected Express’s decision to use California Overnight as its principal engine for growth.

While Express’s sales growth was strong during this period, its ability to derive profits from its revenues was less impressive. Although Express was generally profitable, its net profits fluctuated and its net profit margins required great improvement. Moreover, management had a less than impressive track record of meeting its profit projections. While management had focussed on revenue growth, it recognized that the company had to increase its ability to turn those revenues into profits and that the company faced vigorous competition that limited its ability to do that. Nonetheless, as of 1998, the company’s management believed that Express had the potential for profitable growth.

II. Procedural Background

This case was brought under 8 Del. C. § 225 by plaintiffs L. David Callaway, III, Stuart Agranoff, and CVC against defendants Edward M. Miller and William A. DeLorenzo. The inspiration for the action was the November, 1998 ouster of Calla-way and Agranoff from the board of EMS at the behest of Miller, who claimed to have bought majority control of EMS legitimately. At the time they were removed, Callaway and Agranoff comprised two members of EMS’s three member board. Miller’s co-conspirator John Ovens was the third member of the board, as well as EMS’s president. At that time, Calla-way was EMS’s chief executive officer and Ovens’s management superior. Before Miller came on the scene, CVC controlled over 49% of EMS’s then outstanding shares, and nearly 35% of EMS’s equity on a fully diluted basis that assumed conversion of the BT Warrants into EMS shares.

Miller’s conduct obviously upset CVC, which had controlled EMS before Miller’s attempt to remove Callaway and Agranoff. More relevantly as a legal matter, CVC was party to a shareholders’ agreement (the “first refusal contract”) that gave EMS and then CVC rights of first refusal to purchase the shares of EMS held by other stockholders. 3 The evidence at trial revealed that Miller had tortiously interfered with the first refusal contract in buying his non-BT EMS Shares.

Perhaps most disturbingly, Miller had done so in league with Ovens, and EMS’s chief financial officer, Peter Simpson. The three of them engaged in a concerted plan to help Miller acquire control of EMS through covert activities concealed from Agranoff and Callaway. Ovens and Simpson provided Miller with access to confidential company information without board authorization. Ovens and Simpson also actively concealed Miller’s activity from Agranoff and Callaway. Miller, Ovens, and Simpson started this activity after learning that Callaway was battling life-threatening cancer.

As a result of this concealed conduct, Miller was able to purchase 23,060 shares from minority stockholders of EMS at prices ranging from $24 to $30 a share during the period June 11,1998 to September 15, 1998. 4 This gave him approximately 14% of EMS’s shares on a fully diluted basis, and 21% before the exercise of the BT Warrants.

To obtain a safe majority of EMS’s shares, Miller had to purchase many more shares. Two other sources of shares existed. One was a group of stockholders who collectively held over 15% of EMS’s shares on a fully diluted basis. The bulk of these shares, some 21,550 5 shares, were held by Dexin Holdings, an entity affiliated with Morley Chandler, a former manager at EMS. The remaining shares were held in two separate blocks of 1,905 6 shares by: (i) Chrysalis, an investment group, which had acquired these shares from another former EMS employee Norbert Park, in apparent violation of the first refusal contract; and (ii) Dana Hyatt, a former EMS employee. All of these stockholders were affiliated with a competitor of Express called Golden State Express, which is run by Chandler, Hyatt, and other former managers at EMS. Golden State harbored some idea of making a run at control of EMS itself, using Chrysalis as its instrument, and had approached BT about the possibility of purchasing the Warrants. But there is no record evidence that suggests that the Golden State group ever made a serious offer for the BT Warrants.

The other source was BT, whose Warrants were convertible into 57,591 EMS shares. Once exercised, the BT Warrants translated into shares equaling 35.284% of EMS’s equity on a fully diluted basis, an amount slightly in excess of what CVC controlled. During 1998, EMS was engaged in negotiations with BT to extend the Warrants, which were set to expire later that year. These negotiations were complicated by BT’s displeasure about EMS’s failure to share information on a more regular basis.

When Miller decided to approach BT, EMS was at a sensitive point in its negotiations with BT. It had already exacted from BT a pledge not to sell to Chrysalis without first talking with EMS. Unfortunately for EMS, its principal contact with EMS, Ovens, was trading for Miller, not EMS. Instead of assuring BT that it would be dealt with fairly by EMS, Ovens led BT to believe that EMS was going to dilute BT’s interests. Ovens said that there was a way for BT to avoid this fate: it could sell to a management friendly buyer that Ovens knew — Miller.

That is what resulted. Miller negotiated with BT and on October 15, 1998 obtained an option to purchase their Warrants at $60.37 per Warrant, a price more than the double the highest price he paid to the small shareholders. At the time he negotiated for the Warrants, Miller knew he was buying a control block and that was why he paid a higher price. BT also knew Miller was seeking to buy a control block secretly when it negotiated with him.

Having purchased the BT Warrants, Miller was still slightly short of the absolute majority he needed to execute a consent removing the EMS board. Thus, he went to Dexin, Chrysalis, and Hyatt and bought their shares at $50 apiece. At that time, Miller was again making a purchase that was necessary for him to gain control and those holders knew that.

Having consummated all these purchases, Miller was the putative holder of 65.05% of EMS’s equity on a fully diluted basis, having purchased those units as follows:

[[Image here]]

Meanwhile, CVC and its affiliates held the remaining 57,142 shares, or 34.95% of EMS’s fully diluted equity. 7 Having accumulated control of a majority of EMS’s fully diluted equity, Miller then acted to seize control.

After trial, this court issued an opinion finding that Miller’s control of his EMS shares had been procured by a variety of wrongful acts. These included: violating the rights belonging to EMS, CVC and other parties to the first refusal contract; tortiously interfering with the first refusal contract; aiding and abetting breaches of fiduciary duty resulting in the usurpation of a corporate opportunity belonging to EMS; and misusing confidential information of EMS in aid of his illicit purchases.

As a remedy for Miller’s wrongful acts, this court invalidated his removal of Calla-way and Agranoff and held that those of Miller’s EMS shares that were not derived from the BT Warrants could not be voted until Miller offered those shares first to EMS and (if EMS declined) to CVC in the order, in the blocks, and at the price he purchased them. 8 This remedy tracked the contractual first refusal rights that Miller had violated and principles of restitution as the court understood them. 9 Un der the first refusal contract, EMS and (if EMS declined) CVC would have had the option to purchase shares from holders in the blocks they held them at the price they were offered by a third-party. Neither EMS nor CVC had a duty to buy any particular block or to buy all blocks offered to them. They could pick and choose.

As to the BT Warrants, this court ruled that Miller was not permitted to vote the shares resulting from exercise of the Warrants until he offered the resulting shares only to EMS. The reason for this distinct treatment was that EMS was the only entity to which the corporate opportunity belonged. While the court had decided that the non-BT Shares were covered by the first refusal contract, it had not ruled on that contract’s applicability to the BT Warrants. Indeed, Miller had argued that the first refusal contract did not cover the BT Warrants, and the plaintiffs had dropped any contrary argument to that effect. As a result, the court’s order did not give CVC the right, nor the duty to purchase the BT Warrants. It simply gave EMS — the party with the corporate opportunity that had been wrongfully usurped — the option to buy the BT Warrants at the price Miller paid for them.

As an overall matter, therefore, the court’s order did not require CVC or EMS to purchase any of the non-BT Shares Miller purchased; it gave them that option. Likewise, the court did not require EMS to purchase the BT Warrant Shares; it gave EMS that option. This remedy was designed to be precisely proportionate to Miller’s breaches and to the rights he thereby impaired. Moreover, the remedy did not seem punitive in any manner. For any shares EMS or CVC purchased, Miller would be made whole by receiving his full purchase price plus interest. Any shares that EMS or CVC did not purchase would be retained by Miller and he would be free to exercise all the rights inherent in those shares, including the right to vote them. Thus, Miller did not suffer a forfeiture of any legitimate economic rights; he simply lost his wrongfully secured control position.

Miller was not pleased by this court’s decision and appealed. On appeal, the Delaware Supreme Court affirmed this court’s findings of fact and conclusions of law in all respects except one. The Supreme Court’s order altered this court’s remedy by requiring EMS or (in the event EMS declined) CVC to purchase all of the non-BT Shares if it wished to purchase any of them. It further required EMS to purchase all of the BT Warrant Shares if it wished to purchase any of them. 10 The order did not articulate why this alteration was made. 11

Miller was still not satisfied. By the terms of the Supreme Court order, CVC could simply repurchase all of the non-BT Shares at the prices Miller paid, thus restoring itself to majority control. The average cost for CVC to purchase all the non-BT Shares was well below the $60.87 per share price Miller paid for the BT Warrants. Miller thus feared that he would be stuck as a minority stockholder holding 35% of EMS’s equity — that is, he did not want to be in the position he had attempted to foist on CVC.

Miller therefore moved for reargument, seeking a further amendment of this court’s remedy. Without hearing from the plaintiffs, the Supreme Court granted Miller’s motion and amended this court’s remedy order further. In its second order, the Supreme Court stated that this court must “modify its final judgment to make clear that Appellant Edward M. Miller must offer all EMS shares owned or controlled by him (including the EMS shares acquired by Miller as a result of the exercise of the BT Warrants) and that EMS [or CVC] ... must purchase all such shares, if one or more of them elects to purchase any such shares,” 12 Again, the Supreme Court’s second order did not contain an explanation of the reasoning underlying this alteration. 13 The alteration had the effect of ensuring that Miller, rather than those he injured, would at worst be put back in the position he would have occupied but for his own wrongdoing.

The plaintiffs moved to withdraw the mandate and requested an opportunity to be heard regarding the Supreme Court’s modification of its July 28, 1999 order. The Supreme Court denied that request without explanation, with Justice Berger dissenting.

On remand to this court, Miller presented a proposed order that provided that he would not be permitted to vote any of his EMS shares until he offered them to EMS and CVC at the price he paid for them, plus interest. This court signed that order on September 29,1999.

Miller then offered the Warrant Shares to EMS at $60.37 per share, plus interest. EMS responded that under the literal terms of the final order, it was required to pay only that amount that Miller had paid for the Warrant Shares — the Warrant exercise price of $1 a share — and not the amount Miller paid for the Warrants themselves — $60.37 a Warrant. The parties brought this dispute to the court to resolve.

The court was unable to resolve that dispute because it had no insight into the reasoning behind the orders entered by the Supreme Court. Rather than speculate, this court certified the following question to the Supreme Court, which had been crafted by mutual agreement of the parties:

What is the price EMS, CVC or Air Canada is required to pay for the Warrant Shares: $57,691 plus interest (representing the price of $1 per share paid to EMS to exercise the BT Warrants) or $3,482,691 plus interest (representing the price Miller paid for the BT Warrants plus the $1 per share exercise price paid to EMS)? 14

On March 9, 2000, the Supreme Court issued an order (the “Final Remand Order”) resolving the certified question as follows:

2. Although the Court of Chancery asked about two specific prices, we read the certified question as requesting clarification of this Court’s Order dated August 25, 1999. Accordingly, we are not constrained to select one of the two prices suggested in the certified question.

3. The amount to be paid for the Warrant Shares must be determined in accordance with the principle that, “a fiduciary who purchases for himself individually property which he should purchase for the beneficiary holds the property upon a constructive trust for the beneficiary....” [quoting Restatement of Restitution § 194 cmt. b (1937).] The fiduciary must surrender that property if the beneficiary “restores to the fiduciary the amount of the purchase price, or the price for which he could have obtained it for the beneficiary, whichever is less.” [M]

4. Miller paid a higher price to obtain the Warrant Shares than he did for the [non-BT Shares] because, for Miller, the Warrant Shares represented a control block. It appears that the Warrant Shares would not have commanded a control premium, however, if the purchases had been EMS, CVC or Air Canada. Miller must be paid the lesser of his purchase or the price at which EMS, CVC or Air Canada could have obtained the Warrant Shares, that being the value of the benefit to them. That hypothetical purchase price, which we assume will be a “fair value” price equivalent to that determined through appraisal, is a factual determination, which must be made by the Court of Chancery.

NOW, THEREFORE, IT IS ORDERED that the certified question is answered as provided herein, and this matter is REMANDED for further action in accordance with this Order. Jurisdiction is not retained. 15

After the remand, the parties engaged valuation experts and participated in discovery. On September 13, 2000 and February 16, 2001, trial was held to grapple with the issues posed by the Final Remand Order. The gap in the trial dates was attributable to the fact that the parties vigorously disputed the meaning of the Final Remand Order. In order to attempt to comply with the Final Remand Order and to present the Supreme Court with a full record upon which to decide any subsequent appeal, the parties were given time after the first trial day to develop the record further with the aid of their valuation experts.

III. The Dispute Over The Meaning Of The Final Remand Order

The differences between the parties over the meaning of the Final Remand Order are important. For his part, Miller reads the Final Remand Order as clear: it requires this court to appraise the BT Warrant Shares in accordance with the standards applicable under 8 Del. C. § 262. Those standards require the court to correct for any minority discount inherent in the value of a minority block of shares, and to avoid the application of any discount for lack of marketability. In practical terms, this means that the Warrant Shares would not be valued based on the fair market value principles that would be used in non- § 262 contexts, but on the appraisal-unique “fair value” standard.

The plaintiffs retort that this hyper-literal reading of the Final Remand Order cannot be correct because correction for a minority discount requires the addition of a premium that spreads the value of control over all shares equally. As a result, it would contradict the obvious “spirit” of the Final Remand Order for the court to construe its mandate as to render a decision as to the strict “appraisal” value of the Warrant Shares. 16 Instead, the plaintiffs contend that the Supreme Court obviously recognized the impossibility of determining precisely what would have happened if Miller and his co-conspirators had not engaged in a lengthy and disruptive course of improper conduct. As such, the Supreme Court took a practical approach that directed this court to come up with a fair market value of the Warrant Shares using traditional valuation techniques.

Recognizing that the plaintiffs’ argument has substantial logical appeal and is a highly plausible interpretation of the Supreme Court’s intent, Miller' says that if this court is going to deviate from a hyper-literal approach, it must make a factual determination of the price CVC would have had to pay to BT for the Warrant Shares in 1998 based on a speculative “game theory” approach. In this regard, it contends that the best evidence of the price that CVC would have had to pay is the price that Miller actually paid. Because Miller bargained at arm’s length with BT, the $60.37 a share bargain he struck with BT is reliable evidence of the value that any other purchaser would have had to pay. In support of that argument, Miller advances “what if’ scenarios posited on CVC’s supposed need to purchase the Warrant Shares in order to retain control, against supposed threats from Golden State and Miller himself.

For the following reasons, I reject Miller’s reading of the Final Remand Order and embrace that advanced by the plaintiffs.

I start with the first question, which is whether the court must enter an order in strict compliance with § 262 appraisal standards. Although that is a literal reading of the Final Remand Order, it is a reading that ignores the larger spirit of that Order, which suggests that CVC should not have to pay a control premium to regain control of EMS. As a practical matter, correction of a minority discount requires the court to add back a control premium to the value of the enterprise, and to spread that premium equally across all the enterprise’s shares. The resulting value for a minority share is thus not what would be considered “fair market value” in valuation terms, but an artificial value that reflects policy values unique to the appraisal remedy. In simple terms, those values may be said to consist in this proposition: if a majority stockholder wishes to involuntarily squeeze-out the minority, it must share the value of the enterprise with the minority on a pro rata basis. 17 This solicitude has no proper place here; Miller is not an innocent minority stockholder. He has been adjudicated to have engaged in serious misconduct.

More faithful to the overall text of the Final Remand Order, therefore, is an approach that values the Warrant Shares using traditional valuation techniques, devoid of the policy-based adjustments that are unique to the appraisal context under § 262. Because Miller’s reading of the Final Remand Order is also a plausible one, however, I will also determine the § 262 appraisal value of the Warrant Shares. This will permit the Supreme Court to issue a final order itself without a further remand if I have incorrectly discerned its intent. 18

Likewise, I reject Miller’s alternative “game theory” approach to the Final Remand Order. There is no reliable way to determine what CVC would have paid in 1998 to purchase the Warrant Shares from BT. Any such determination would involve pure speculation.

The fact that Miller paid $60.37 to purchase each of the Warrant Shares is not a reliable basis to determine what CVC would have paid in the event that Miller and his co-conspirators had not engaged in their misconduct. This point is critical.

Marketplace transactions are dependent on many variables. Miller would have me assess the price CVC would have had to pay based on the inequitable and improper posture CVC found itself in during autumn 1998. By this point, Miller had already snapped up a large number of minority shares without CVC’s knowledge. By this point, Miller’s co-conspirator Ovens had already misled his fellow EMS directors about Miller’s activities. By this point, Ovens had already tainted BT’s view of EMS’s CEO, when it was his job to negotiate on behalf of the company faithfully.

Had Miller’s co-conspirators acted properly, EMS and CVC would have known what Miller was doing early on in 1998. Given Miller’s preference for a sneak attack, he likely would have skulked away from any fair fight. As a result, it is unlikely CVC would have faced any genuine contest for control from him. And even if it had, CVC could have gone to BT and done a deal to buy its Warrant Shares. “Aha,” says Miller, CVC would have had to deal with BT and pay a healthy price. Perhaps, but Miller ignores that CVC would have had the option of simply dealing with BT and not buying any other shares. It is only because of Miller’s misconduct that CVC is forced to think about buying the other shares to preserve its position. Similarly, how can one tell what BT’s attitude toward EMS would have been had Ovens acted properly. There is reason to believe that BT would have declined to deal with other purchasers had Ovens simply assured BT that EMS would treat it fairly and acted as a director of EMS to ensure it did so. 19

Put bluntly, Miller’s wrongs fundamentally skewed the situation CVC faced. Under our corporation law, the party whose purposeful misconduct creates unresolvable uncertainty is the one that bears the cost, not the innocent victim. 20

Miller’s theory suffers from other defects. He surmises that Golden State was prepared to launch a contest for control, and there is record evidence that supports the notion that Golden State was pondering such an effort. But Golden State in fact never did undertake a takeover bid, and its affiliates sold to Miller at $50 a share knowing Miller needed their shares for control. It is mere speculation that Golden State had what it took to take on CVC in a real fight for control.

Miller’s theory has even more holes. The non-BT Shares sold to Miller by minority stockholders other than Dexin, Hyatt, and Chrysalis equaled over 14%. Taken together with CVC’s shares, this equals over 49% of EMS’s fully diluted equity. As a result, it is not obvious that CVC would have been desperate to pay a large premium to BT, nor that BT would have perceived CVC as being prepared to do so. This is in distinction from BT’s knowledge that Miller needed the Warrants if he was to secure a control block. Likewise, if CVC held over 49% of the shares already, this would have had to make Golden State think hard before entering a bidding contest, especially given the reality that Golden State was a direct competitor and that the EMS board might have been able to take legitimate measures to issue rights to new shares dependent on capital infusions.

Significantly, the shares held by Chrysalis were sold to it in apparent violation of EMS’s and CVC’s first refusal rights by a former management holder. If this is so, EMS and CVC might have had the right to repossess those shares from Chrysalis at the price Chrysalis paid for them, giving CVC a clear shot at a majority without the BT Warrants.

I raise all these issues to make a fundamental point. While I do not pretend to know for certain that CVC would not have had to pay BT the same price that Miller paid, I believe it to be unlikely because CVC would not have been as anxious a buyer as Miller and would not have been perceived as such by BT. Most important, the reason that no one can tell what would have happened is that Miller and his co-conspirators’ improper acts make it impossible to do so. Having done everything he could to put CVC in a bad situation, Miller is in no equitable position to speculate about what CVC might have had to do had it not been victimized by him and his co-conspirators. 21

I read the Final Remand Order as recognizing the impossibility and impracticability of this game theory approach, and as mandating a less speculative exercise based on the use of recognized valuation techniques. The Supreme Court’s approach thus requires CVC to pay Miller a normatively “fair” price for the Warrant Shares, but does not require CVC to pay Miller a “control premium” to reestablish a control position it lost because of Miller’s own improprieties.

Therefore, I intend to determine a fair market value for the Warrant Shares using accepted valuation techniques. This is the proxy that the Supreme Court established as the measure of determining what price CVC would have paid had it bought from BT in a market untainted by Miller’s misconduct.

IV. What Is The Fair Market Value Of The Warrant Shares?

The parties both utilized highly qualified experts in support of their positions regarding valuation. For his part, Miller proffered the testimony of Professor Donald J. Puglisi, the MBNA America Business Professor and Professor of Finance at the University of Delaware. The plaintiffs submitted the testimony of Morton Mark Lee, a recently retired partner of KMPG LLP and now a senior managing director at Sutter Securities, a professional with thirty years of experience in valuing businesses. Both experts provided the court with helpful testimony.

As a frame for valuing EMS 22 however, Puglisi’s analysis is the preferable one. The Puglisi analysis focused on three variations of the comparable companies method of valuation, involving multiples based on EMS’s revenues (the “Revenues Analysis”), earnings before interest and taxes (“EBIT”), and earnings before interest, taxes, depreciation, and amortization (“EBITDA”). 23 The Puglisi analysis was also very user-friendly, and enabled the reader to follow the steps he used in computing his comparable companies valuation. Puglisi’s report also acknowledged that a discounted cash flow (“DCF”) approach would have been viable, had reliable projections of EMS’s performance for the relevant time period been available. But Puglisi considered the projections that EMS had to be wildly unreliable and overly optimistic. Thus, he believed that a reliable DCF valuation was not possible. 24

Lee used a wider variety of valuation methods. Although Lee also believed that EMS’s projections were unreliable, he purported to base a DCF analysis on a substantial negative revision of those projections that he came up with after discussions with EMS managers after the valuation date. 25 That is, Lee discussed the projections for the years following 1998 with managers who knew what the actual results of those later years were. Based on these conversations, Lee developed revised projections that he plugged into a DCF model.

I refuse to give any weight to this technique and therefore to Lee’s DCF analysis. The possibility of hindsight bias and other cognitive distortions seems un-tenably high. Consider this analogy. Suppose there was an interview with Sir George Martin from 1962 in which he opined as to how many number one songs he thought would be released by his new proteges, the Beatles. Could one fast-forward to 1971, interview Martin, and revise Martin’s earlier projection in some reliable way, recognizing that Martin would have known the correct answer as of that date? How could Martin provide information that would not be possibly influenced in some way by his knowledge of the actual success enjoyed by the Beatles and his recollection of his earlier projection? The parties have approached this valuation exercise with the mutual understanding that they could not consider the actual results for EMS past the valuation date of October 1998. Lee’s DCF analysis seems like an unreliable way to have those actual results influence the court’s valuation in an indirect manner that is not susceptible to fair evaluation. Nor have the plaintiffs provided finance literature supporting the acceptance of Lee’s approach to projection modification. Likewise, I also give no weight to Lee’s valuations that are based solely on equity, rather than entity, valuation techniques. These techniques do not consider the different capital structures of corporations. 26

Instead, I choose to focus on the three variations of the comparable companies method of valuation that both Puglisi and Lee agree are appropriate tools to value EMS: analyses based on multiples of Revenues, EBIT and EBITDA. The comparable companies method of valuation determines the equity value of the company by: (1) identifying comparable publicly traded companies; (2) deriving appropriate valuation multiples from the comparable companies; (3) adjusting those multiples to account for the differences from the company being valued and the compa-rables; and (4) applying those multiples to the revenues, earnings, or other values for the company being valued. 27 Comparable companies analyses are frequently calculated on a debt free basis, to derive the fair market value of the company’s market value of invested capital (“MVIC”). The company’s equity value is derived by subtracting the company’s interest bearing debt from the company’s MVIC.

The comparable companies analysis generates an equity value that includes an inherent minority trading discount, because the method depends on comparisons to market multiples derived from trading information for minority blocks of the comparable companies. 28 In a § 262 appraisal, the court must correct this minority trading discount by adding back a premium designed to correct it. 29 As noted previously, I do not believe that such a rigid, formulaic correction is appropriate in this case. For the sake of creating a record that will enable a final resolution on appeal if I am wrong about that, I will nonetheless later determine what premium should be added to make that correction and thereby derive a “fair value” in appraisal terms.

Puglisi and Lee both chose a number of similar comparables and neither believes that the comparables are an ideal comparison to EMS, which is far smaller than the comparison companies and is not traded on any exchange. The largest disagreement between them is over the multiple to be applied in the Revenues Analysis.

EMS’s sales growth is the one feature of the company that is notably impressive. That sales growth, however, is coupled with a poor profit margin. That is, EMS generates a smaller amount of profit for every dollar of its revenues compared to both the experts’ comparison companies. As the plaintiffs point out, a comparable companies analysis based on revenues is less rehable if those revenues are not correlated in a substantial way with earnings. 30

In his comparable companies’ analyses, Puglisi gave the greatest weight to three of the comparison companies. This was conservative on his part, because those companies performed less impressively than the rest of his sample and thus more in fine with Express. Lee did not quibble with this part of Puglisi’s analysis.

What Lee took issue with was Puglisi’s selective approach to coming up with the multiple he used to generate his final valuation number for his Revenues Analysis. In computing his multiples for his EBIT and EBITDA analyses, Puglisi discounted the multiple from the median generated from the three comparison companies upon which he had focussed, 19% in the case of his EBIT analysis and 10% in the case of his EBITDA analysis. 31

By contrast, Puglisi’s selected multiple for his Revenues Analysis used the average multiple derived from the three comparison companies, a figure higher than the median. 32 His rationale for doing so was that sales were EMS’s strength and that its short-term business plan involved a bet that it could gain market share at a rapid rate, and then sustain that share while raising prices and cutting costs to obtain a better profit margin. Puglisi admitted that there was no guarantee that this would happen, and that EMS faced competition that could limit its ability to execute on this strategy.

I believe that Puglisi’s approach to the Revenues Analysis multiple is too optimistic and is inconsistent with his EBIT and EBITDA analyses. EMS’s historical ability to meet its projected EBITDA targets was not impressive, and there was no evidence that it would face less market competition in the years after 1998. While it is appropriate to give some weight to the fact that EMS’s strategy was based on increasing its sales at the expense of short-term profits, that can be more responsibly achieved by using a multiple at the median. This multiple is generous to Miller in that Puglisi discounted the median multiples in his EBIT and EBITDA multiples, and because Lee’s testimony that the multiple should be reduced to the .21 level is not without logical force.

The plaintiffs’ challenge to the other aspects of Puglisi’s comparable companies analysis focuses on his use of a net operating loss (“NOL”) reflected on EMS’s balance sheet as of the valuation date. The plaintiffs’ claim that this NOL figure was incorrect but never provided a correct amount; moreover, the NOL Puglisi used is EMS’s own figure as of the valuation date. In a valuation of EMS as a going concern, it was also appropriate for Puglisi to use the full amount of the NOL, as Lee admitted. 33

With these objections out of the way, the court can display its valuation of EMS— putting aside for a moment the question of whether a premium should be added because the BT Warrants constituted a substantial block of EMS voting power and whether a marketability discount should be subtracted because EMS shares were not traded on public markets. In coming to this intermediate step, I use Puglisi’s approach of giving equal weight to the Revenues, EBIT, and EBITDA approaches and adopt his EBIT and EBIT-DA multiples. This analysis yields a value of $41.02 for the Warrants Shares, computed as follows: 34

[[Image here]]

This brings us to perhaps the most warmly contested part of the valuation question, which is whether adjustments need to be made to the value I determined above. The adjustments contended for cut in different directions. Miller contends that even if CVC would not have had to pay a “control premium” to BT, it would likely still have paid a 20% premium that recognized the importance of the BT voting block, which constituted over 35% of EMS’s voting power on a fully diluted basis. Meanwhile, the plaintiffs contend that it is undisputed that a fair market value analysis would include a discount for the lack of marketability of EMS’s shares, 35 and that Puglisi and Lee agree that the amount of such a discount should be in the 83% range.

There is merit to both parties’ position regarding these issues. Although the financial literature Puglisi relies upon involves the purchase of minority blocks in situations that are diverse and not all of which are fairly comparable to the situation facing CVC, that literature does support the proposition that important blocks of voting shares have value to controlling stockholders that leads them to pay premiums above the unaffected minority trading price. 36 In this case, CVC’s control of EMS was not firmly secured and the purchase of the BT Warrants would have guaranteed that control. Thus, the BT Warrants had significant value to EMS that would have factored in any sales negotiations between CVC and BT. As a result, I think it fair to conclude that some upward adjustment of the $41.02 value would be in order. 37

At the same time, the negotiations between CVC and BT would have also factored in the reality that BT held a large block of a privately held corporation that was not paying dividends, and that BT wanted to get out without making more investments in EMS. Absent the dream come true that was presented by Miller’s unorthodox takeover moves, BT arguably had few options other than the uncertain possibility that Golden State would try to put EMS in play. Furthermore, the wildly different prices Miller paid for his EMS shares — ranging from $24 to $60.37 — are indicative of the lack of a liquid market for EMS shares.

Although valuation exercises are highly dependent on mathematics, the use of math should not obscure the necessarily more subjective exercise in judgment that a valuation exercise requires. 38 In that regard, I decline to adjust the $41.02 valuation by adding the 20% premium Puglisi contends for, and subtracting the 33% marketability discount the plaintiffs press upon me. 39 Instead, I will not adjust the $41.02 at all, believing that the two factors roughly cancel each out. 40 While this approach is perhaps overly generous to Miller, 41 it my best sense of what is fair, considering the objective economic circumstances that would have existed in the absence of Miller’s misconduct. 42 While CVC was not nearly so eager a purchaser as Miller it still had important reasons to pay a healthy price to BT, given the voting power BT’s block conveyed.

V. What Is The Appraisal Value Of The Warrant Shares?

Because of the uncertainty about the meaning of the Final Remand Order, I now will consider what premium would need to be added to the fair market value of the Warrant Shares to correct for the implicit minority discount. This correction is necessarily an imprecise one that is complicated. In order to determine what the implicit minority discount in a comparable companies analysis is, one is forced to look at the prices paid for control blocks. 43 Such prices are frequently paid in connection with a merger or other fundamental transaction. This source of data is therefore problematic, because the premiums arguably reflect value that is not related to the value of the acquired companies as going concerns under their preexisting business plans, such as synergistic values attributable to transactionally-specific factors. As a practical matter, however, it is impossible to make precise determinations about what motivated an acquiror to pay a control premium. Was the premium paid because the acquiror needed jobs for his cousins? Because the market was undervaluing the acquired company? Or because the acquiror was going to fundamentally change the business plan to create higher value, such as by combining the acquired company with another business in a synergistic way? 44

If the policy basis for the correction for a minority discount is to ensure that the minority that is squeezed out receives their proportionate value of the enterprise regardless of their minority status, it is unclear why the court should parse the available data about control premiums paid in an impossible attempt to determine the factors that contributed to the total premiums paid and their relative economic im- portanee. 45 Such data, after all, is not about value derived from the transaction giving rise to appraisal rights — that is, the value that cannot be considered under § 262. 46 Rather, one hopes the data reflects a diversified information set about the typical premiums that acquirors will pay for control positions, which can be compared with some reliability to the price at which minority blocks sell. Given the policy approach taken in Cavalier, it is not apparent why it is critical to ensure that minority stockholders do not receive their proportionate share of the value that would likely be paid by a hypothetical ac-quiror in the event that the acquiror could attain control of the entity under its existing business plan. 47 Whatever the acqui- ror’s motives, that price in fact reflects the increased value the acquiror places on a control block of the company’s shares over the value the market places on a minority block. Whatever the acquiror’s motives, the acquiror (perhaps as a result of hubris) expects to derive value (from the employment of relatives, synergies, or other gains) that will justify the premium paid.

Thus, while control premiums undoubtedly include value related to end-game transactions in which control passes and the company’s business plan changes to some greater or lesser extent, that is a natural consequence of the practical realities of control sales. How is the court supposed to identify “pure” control block sales in which nothing changes except the identity of the majority stockholder? Certainly the record here does not permit it to do so. 48 For this very reason, the cases in which this court has adjusted minority discounts to exclude value that derives from the expectation of a synergistic change in control have used a necessarily rough approach that simply involves shaving some percentage off the top of the available information about control premiums paid. 49

I am in no better position now than this court has faced in prior cases. Puglisi argues for a 40% adjustment using aggregate control premium data for the year 1998 that involves publicly traded companies. He contends that there is no reliable financial literature that suggests that the average premium in takeovers is higher in transactions involving companies in similar businesses, ie., in transactions that are thought to be based on synergy. Indeed, some empirical evidence suggests that “synergy”-based acquisitions involve lower premiums, a result that is contrary to the intuition of economists. 50 While the plaintiffs contend otherwise, their contrary arguments are not supported by proffered empirical research and they have offered no reasoned basis upon which the court is to parse the available premium to subtract “synergy” value.

The plaintiffs do advance other arguments that have more force. They note that Puglisi has derived his 40% premium by reference to all transactions in 1998, a year in which mega-mergers were common. 51 EMS is a relatively small private company and it is by no means apparent that the voracious appetite that generally existed in 1998 for acquisitions extended to small niche providers of delivery services. Lee also points to two transactions in the delivery services industry in the two years before 1998 that involved premiums of only 16.7%. Furthermore, the aggregate data that Puglisi relies upon shows that control premiums in the transportation sector overall tended to be lower on average than that for the entire marketplace. 52 Finally, the plaintiffs note that Puglisi has been less aggressive in adjusting for the minority discount in other recent cases in which he has testified.

As a result, I conclude that if an adjustment is to be made for the implicit minority discount in the comparable companies analysis, that adjustment should be at the level of 30%. This lower figure reflects the fact that EMS is not directly comparable to the companies in Puglisi’s data survey, that the two transactions Lee points to should be given some weight, and the possibility that Puglisi’s average included some elements of excludible value. 53 The adjustment also tracks that made by this court in two recent valuation opinions. 54

Thus, a § 262 value of EMS’s shares of $51.13 can be derived as follows:

[[Image here]]

This so-called “fair value” gives no weight to the lack of marketability of EMS shares, 55 while giving Miller full access to a presumed control premium. It thus treats Miller as if he was an innocent minority stockholder, rather than an adjudicated tortfeasor, and places most of the burden of uncertainty caused by Miller’s wrongful actions on his victims. In my view it is therefore an unfair and inappropriate remedy.

VI. Conclusion

For all the foregoing reasons, I answer the Supreme Court’s mandate by concluding that the price at which CVC could have purchased the BT Warrants in the absence of wrongdoing by Miller was $41.02 per Warrant. In the alternative, the fair value of the BT Warrants under a § 262 appraisal standard is $51.13 per Warrant. The parties shall confer and present a conforming final order within seven days of this opinion.

1 . This pattern of wrongdoing is detailed in Agranoff v. Miller, Del.Ch., C.A. No. 16795, mem. op., Strine, V.C. (Apr. 9, 1999), aff'd & remanded, Del.Supr., 737 A.2d 530 (1999).

2 . Miller v. Agranoff, Del.Supr., No. 486, 1999, 750 A.2d 530 (March 9, 2000) (Order).

3 . The first refusal contract also gave such rights to Air Canada, a shareholder of EMS’s operating subsidiary, Express, in the event neither EMS nor CVC exercised such rights. For the sake of simplicity, I do not refer to Air Canada again.

4 . JE 40, at 11.

5 . This constitutes 13.18% of EMS’s shares on a fully diluted basis, and 20.3% of the shares before the exercise of the BT Warrants.

6 . Taken together, the two blocks equalled 3,810 shares or 3.6% of EMS’s shares on a pre-exercise basis, and 2.33% on a fully diluted basis.

7 . The parties have been unable to agree on whether EMS has 163,503 or 163,403 shares. The foregoing summary, as well as prior calculations in this opinion, are taken from a document prepared by Miller to track his own purchases which gives EMS the higher number of shares. JE 42 (from the original trial, dated 1/6/99). In valuing the Warrants, I use the lower number, which is to Miller’s advantage.

8 . Agranoff v. Miller, Del.Ch., C.A. No. 16795, 1999 WL 219650, Strine, V.C. (April 16, 1999) (Final Order).

9 .See Restatement of Restitution § 194, cmt. b (1937). ("The beneficiary can at his option permit the fiduciary to keep the property or compel him to surrender it; but he can compel the fiduciary to surrender it only if he restores to the fiduciary the amount of the purchase price, or the price for which he could have obtained it for the beneficiary, whichever is less.") (emphasis added).

10 . Miller v. Agranoff, Del.Supr., No. 178, 1999, 737 A.2d 530 (1999) (Order).

11 . Id.

12 . Miller v. Agranoff, Del.Supr., No. 178, 1999 (Aug. 25, 1999) (Order).

13

14 . Miller v. Agranoff, Del.Supr., No. 486, 1999, ¶ 1 (March 9, 2000) (Order) (quoting Court of Chancery question).

15

16 . Gonsalves v. Straight Arrow Publishers, Inc., Del.Supr., 725 A.2d 442, 1999 WL 87280, at **4 (1999) (Order) (holding in this appraisal action that the "spirit of the remand” permitted the Court of Chancery to examine both the rate and the form of interest to be awarded in an appraisal anew); Cinerama, Inc. v. Technicolor, Inc., Del.Ch., 1999 WL 135242, at *4, Chandler, C. (Feb. 25, 1999) (exercising discretion in the "spirit” of the Supreme Court's remand opinion).

17 . In Cavalier Oil Corp. v. Harnett, Del.Supr., 564 A.2d 1137 (1989), the Supreme Court held that this court was correct in refusing to apply marketability and minority discounts in appraising the stock held by a minority stockholder in a small corporation. In so ruling, the Supreme Court embraced the view “that the objective of a section 262 appraisal is 'to value the corporation itself, as distinguished from a specific fraction of its shares as they may exist in the hands of a particular shareholder’ ...” Id. at 1144 (quoting Court of Chancery opinion without citation) (emphasis in original). The Supreme Court explained the policy justification for this approach as follows:

Discounting individual share holdings injects into the appraisal process speculation on the various factors which may dictate the marketability of minority shareholdings. More important, to fail to accord to a minority shareholder the full proportionate value of [the petitioner’s] shares imposes a penalty for lack of control, and unfairly enriches the majority shareholders who may reap a windfall from the appraisal process by cashing out a dissenting shareholder, a clearly undesirable result.

564 A.2d at 1145.

The unique context of this case manifestly does not involve an appraisal of the shares of an innocent, squeezed-out minority stockholder. Rather, the judicial task at hand is to determine what price CVC would have had to pay for a particular block of shares, a task that appropriately should consider block-specific value factors, especially in view of the fact that it was Miller’s improper behavior that causes the need for this intrinsically imprecise inquiry.

18 . Assuming, of course, that I have otherwise not committed reversible error.

19 . See, e.g., Chiate Dep. at 26.

20 . See, e.g., Thorpe v. CERBCO, Inc., Del.Supr., 676 A.2d 436, 445 (1996); Guth v. Loft, Inc., Del.Supr., 5 A.2d 503, 510 (1939).

21 . Cf. Bomarko, Inc. v. International Telecharge, Inc., Del.Ch., C.A. No. 13052, 1999 WL 33472141, at *25 n. 14, Lamb, V.C. (Nov. 4, 1999 revised Nov. 16, 1999), aff'd, Del.Supr., 766 A.2d 437 (2000) (valuing derivative claims in calculating damage award and refusing to engage in "rank speculation” at the defendants' instance about what would have happened to that claim had the defendant complied with, rather than breached, his fiduciary duties).

22 . As a practical matter, all the valuation methods discussed first valued Express, EMS's operating subsidiary. I refer to EMS for the sake of simplicity.

23 . Puglisi filed two reports. The first report gave no weight at all to Miller’s prior purchases of EMS shares in determining value, stating "the more reasonable and responsible course” was to give those transactions "no weight in the fair value determination process.” JE 47, at 7.

In a second report in which Puglisi was given the opportunity to opine as to the "fair market value” of the BT Warrants rather than their appraisal value, Puglisi changed direction and contended that Miller’s actual purchase from BT is the most reliable evidence of what CVC would have had to pay BT for the Warrants. For the reasons that Pugli-si stated in his first report, however, it is impossible to discern what CVC would have paid BT in the absence of wrongdoing by Miller. What a very different purchaser (Miller) paid to BT in circumstances greatly different to those that would have faced CVC if Miller had not improperly entered the scene bears little on the question presented here. Moreover, Puglisi is a distinguished expert in valuation and finance, but has no particular expertise in determining what would have happened in the event that things that did occur in the commercial world did not.

Finally, Puglisi continues to accord no weight to the fact that Miller purchased shares from one of his agents, Carol Spencer, for $30 a share in September 1998. Spencer had a great deal of information about EMS and sold at a price half of that which Miller paid to BT. This illustrates the difference between the value Miller placed on minority shares and that he placed on the shares that would give him majority control, and the strong bargaining power this gave to BT as a seller to him.

For all these reasons, while the court gives considerable weight to Puglisi’s testimony about fair market value using traditional valuation techniques, it gives no weight to his game theory approach.

24 . Puglisi ran a DCF valuation using the available projections and came up with a value per EMS share of over $150 per share. 9/13/00 Tr. 200-202.

25 . Among the managers Lee spoke with was Callaway, who has a keen interest in the outcome of this suit.

26 . Lee’s final valuation number was derived from a subjective exercise in judgment that gave no ascertainable weight to the various techniques he used. Although he later attempted to derive his final valuation figure from a mathematically weighted approach, his trial testimony and report gave the court a final product without the recipe from which it was made.

27 . This court recently deployed this method of valuation in two well-reasoned valuation opinions. See Bomarko, 1999 WL 33472141; Borruso v. Communications Telesystems International, Del.Ch., 753 A.2d 451 (1999).

28 . Borruso, 753 A.2d at 458; Bomarko, 1999 WL 33472141, at *22; Le Beau v. M.G. Bancorporation, Inc., Del.Ch., C.A. No. 13414, 1998 WL 44993, mem. op. at 18-19, Jacobs, V.C. (Jan. 29, 1998) (citing SHANNON P. PRATT, et al., VALUING A BUSINESS 194-95, 210 (3d ed.1996); C.Z. Mercer, Valuing Financial Institutions 198-200 and Chapter 13 (1992)); Kleinwort Benson Ltd. v. Silgan Corp., Del.Ch., C.A. No. 11107, mem. op., 1995 WL 376911, at *3, Chandler, V.C. (June 15, 1995). In an earlier case, it was held that an adjustment of a comparable companies analysis to correct the implicit minority discount had not been shown to be an accepted valuation technique. Salomon Brothers Inc. v. Interstate Bakeries Corp., Del.Ch., C.A. No. 10054, mem. op. 12, 1992 WL 94367, Berger, V.C. (May 1, 1992). In subsequent cases and this case, the record demonstrates that such a correction is an accepted and required technique in a § 262 valuation using the comparable companies method. Some commentators claim that this technique, however, has become accepted solely because of the need to adapt financial valuation methods to legal rules forbidding the valuation of minority shares qua minority shares, and that it is not used in the nonlegal context. See John C. Coates IV, "Fair Value" As An Avoidable Rule Of Corporate Law: Minority Discounts In Conflict Transactions, 147 U. Pa. L.Rev. 1251, 1286 n. 118 (1999).

29 . E.g., Borruso, 753 A.2d at 458.

30 . SHANNON P. PRATT, et al., VALUING A BUSINESS 246 (4th ed. 2000) ("[W]hen considering using a multiple of sales, it is useful to first see whether the guideline company multiples of revenue are well correlated with return on sales. This value measure tends to be more useful for industries where such a correlation is high than for those when it is not. In a way, capitalization of revenues can be considered a short-cut to capitalization or earnings, since generally there is an implicit assumption that a certain level of revenues should be able to generate a specific earnings level in a given type of business.”) (JE 64); see also ASWATH DAMODARAN, INVESTMENT VALUATION 342 (1995) (JE 65).

31 . JE 80.

32 . In its post-trial brief, Miller stresses that Puglisi’s multiple was far less than the median of all the comparison companies referenced in his report. But as his report and his trial testimony indicates, his final valuation actually focussed on only three companies.

33 . 9/13/99 Tr. 156.

34 . Miller speculates that EMS could have reduced its management costs and thus claims Puglisi’s EBITDA Analysis is conservative. That contention is pure speculation based on an ad hominem attack by Miller on Callaway. Miller in fact held out the promise of more lucrative packages to Ovens and Simpson as a reward for their assistance in his secret takeover scheme. At trial, Puglisi himself disavowed reliance on this factor. 9/13/00 Tr. at 241-42. Likewise, I decline to adopt Puglisi’s alternative analysis that looks at Express’s capitalization of operating leases. That analysis was not founded on sufficient information about the comparable companies and Puglisi himself declined to endorse it in a full-bodied way.

35 . Puglisi contends that such a discount should not be applied solely because he believes that the court should look to Miller’s purchases as the basis for determining fair market value. Because those purchases were the product of improper behavior and involved a buyer (Miller) who was hungry to complete his secret bid for control, I find them an unreliable basis upon which to determine the price CVC would have paid for the BT Warrants in the absence of Miller’s misconduct. As noted previously, my uncertainty reflects Puglisi’s original position regarding the unreliability of Miller’s purchases as a guide to the determination of value in this case.

36 . Clifford G. Holderness & Dennis P. Sheehan, What Constrains Managers Who Own Large Blocks of Stock?, Managerial Economics Research Center, William E. Simon Graduate School of Business Admin., Univ. of Rochester, Working Paper Series No. 88-07 (Mar.1988) (JE 50); Michael J. Barclay & Clifford G. Holderness, Negotiated Block Trades and Corporate Control, 46 J. Fin. 861 (1991) (JE 56); Wayne H. Mikkelson & Hailu Regassa, Premiums Paid in Block Transactions, 12 Managerial & Decision Econ. 511 (1991) (JE 54); Michael J. Barclay & Clifford G. Holderness, Private Benefits From Control Of Public Corporations, 25 J. Fin. Econ. 371 (1989) (JE 52); Michael J. Barclay and Clifford G. Holderness, The Law and Large Block Trades, 35 J.L. & Econ. 265 (1992) (JE 58); Sanjai Bhagat and Richard H. Jefferis, Jr., The Causes and Consequences of Takeover Defense: Evidence from Greenmail, J. Corp. Fin., Aug. 1994, at 201 (JE 60); Wayne H. Mikkelson & Richard S. Ruback, Targeted Repurchases and Common Stock Returns, 22 RAND J. Econ. 544 (Winter 1991) (JE 57); James S. Ang & Alan L. Tucker, The Shareholder Wealth Effects of Corporate Greenmail, 11 J. Fin. Res. 265 (Winter 1988) (JE 51); Karen Hopper Wruck, Equity Ownership Concentration and Firm Value, 23 J. Fin. Econ. 3 (1989) (JE 53).

37 . I note that if the BT Warrants did not have significant voting power, they may well have a fair market value that reflects both a “block” and "marketability” discount.

38 . Cf. Jay W. Eisenhofer & John L. Reed, Valuation Litigation, 22 Del. J. Corp. L. 37, 126-131 (1997) (noting the substantial variation calculations of marketability discounts and adjustments for minority discounts in the financial literature); Mukesh Bajaj, et al., Firm Value And Marketability Discounts at 3 (available on SSRN:ID=262198) (Feb. 26, 2001) (noting wide variation in marketability discounts revealed in the financial literature).

39 . I realize that this would be a rational, if not mandated, valuation conclusion. Therefore, for the benefit of the Supreme Court, I note that this approach yields a value of ($41.02 x .67 x 1.2=) $32.98, all other assumptions remaining constant.

40 . A recent working paper by four economists suggests that the marketability discount agreed upon by the experts in this case might be excessive. As important, the working paper points out the highly imprecise nature of the studies to date, which generate median marketability discounts both far less and far in excess of 33%. See generally Bajaj, et al., supra note 38. For example, the paper notes a study that showed that the private domestic companies were acquired at a discount of approximately 20% to the acquisition of comparable companies with the same EBITDA. Id. at 16.

41 . Indeed, as a practical matter, my approach can alternatively be read as adding a healthy 33% shareholder level control premium to the BT Warrants’ value, offset by a 33% marketability discount.

42 . By giving weight to these objective considerations, I do not lapse into the game theory approach advocated by Miller. Instead, I simply consider among other facts that CVC did not have majority control on a fully-diluted basis going into 1998, that the purchase of the BT Warrants would firmly secure CVC’s control, that BT had limited options that, however, included the possibility of a sale to Golden State that EMS wished to avoid, that EMS’s board of directors could possibly have taken certain legitimate steps that would have had the effect of protecting CVC’s control, and that CVC had first refusal rights on the other EMS shares.

43 . Coates, supra note 28, at 1278 (recognizing that a "minority discount” and a "control premium” can be considered the inverse of one another because "the term 'minority discount' is generally used to mean the difference between the value of control shares and the value of a minority share of a public company,” but also that it may be necessary as a matter of legal doctrine to parse the composition of control premium data to exclude certain elements before that data can be used to correct the minority discount).

44 .Professor Coates has attributed the overall composition of control premiums to three basic categories: (1) synergy value — the value derived from a particular combination of economic assets; (2) expropriation value — the value from being able to use control unfairly to usurp value rightly belonging to the minority; and (3) pure control value — the residual value attaching to the authority to control corporate policy on an ethical and fiduciarily-compliant basis. Coates, supra note 28, at 1274-78. Coates argues that all control premiums "should be analyzed initially as reflecting each of these types of value.” Id. at 1277. He also acknowledges that control premiums reflect the value that acquirors place on these factors, a value that is subjective and that may be quite wrong because of many factors, including the acquiror’s own overconfidence. Id.

45 . The case law requires that a § 262 valuation of an entity value wholly-owned, subsidiaries on a basis that includes a control premium recognizing the possibility that the subsidiaries could be sold. See Rapid American Corp. v. Harris, Del.Supr., 603 A.2d 796, 806-07 (1992). The use of this adjustment recognizes that the wholly-owned subsidiary — which was presumably being operated as a going concern — was controlled by the parent being valued and that the parent could choose to use that control to sell to a buyer who would pay a premium for any of the reasons that buyers of control pay premiums.

In correcting for a minority discount in a comparable companies valuation, it is not clear why the same reasoning would not apply. It seems a fine point to conclude that the value of the entity as a going concern includes the potential to sell controlled subsidiaries for a premium but not the potential to sell the entity itself. Cf. Coates, supra note 28, at 1268-72 (contending that the rigid distinction between firm-level and shareholder-level discounts and premium adjustments cannot be rationally explained). This is particularly so when one recognizes that Rapid-American dealt with a situation where 99% of the net sales of the company being valued came from its three wholly-owned subsidiaries, which were valued by adding a control premium. Rapid American, 603 A.2d. at 799.

46 . In this regard, Professor Coates' excellent paper obscures the precise dilemma that courts face in § 262 actions. In arguing that any adjustment to correct for a minority discount must exclude synergy value, Professor Coates contends that including such value would violate the statutory principle that appraised value not include value attributable to the transaction that gave rise to appraisal rights. Coates, supra note 28, at 1350-51. But the court does not look at the premium paid in the transaction giving rise to appraisal rights in order to come up with a corrected comparable companies analysis valuation, it looks to data provided by the parties and the literature about the typical premiums paid for control blocks of shares of corporations comparable to the company being valued. As Professor Coates admits, it is impractical to think that courts can parse this data to find what portion of the premiums in the data set comprised "synergy” or “expropriation” rather than "pure control” value; moreover, he cites no scholarship that demonstrates that economists or law professors are better positioned to do so in a reliable way. Id. For this reason, Professor Coates argues that in a § 262 action the respondent company should bear the burden of uncertainty on such questions and have to show that portions of the petitioners’ suggested adjustment are comprised of excludible elements such as synergies, otherwise the adjustment will be made on a non-discounted basis. Id. at 1351-52. This approach would, if adopted, create a rule of thumb that would likely result in healthy adjustments to minority value that reflect some slight discount to the available data regarding control premiums paid. In practical terms, Professor Coates's approach has already been followed by this court in Borru-so.

47 .In Kleinwort, this court noted:

In prior appraisal actions, this Court has rejected the use of a control premium derived from merger and acquisition data because the control premium incorporates post-merger value.... The acquiror may value the targetcorporation above its going concern value because of potential synergies or because the acquiror believes it will manage the target better. This portion of a control premium cannot be included in the appraisal value of a corporation because it reflects value arising from the accomplishment or expectation of the merger.

1995 WL 376911, at *3 (citing Salomon Brothers, Inc. v. Interstate Bakeries Corp., Del. Ch., C.A. No. 10154, mem. op. at 12-13, Berger, V.C. (May 1, 1992); Cooper v. Pabst Brewing Corp., Del. Ch., C.A. No. 7244, mem. op. at 22, 1993 WL 208763, Hartnett, V.C. (June 8, 1993)).

48 . Lee could never explain the methodology by which he purported to parse the premia data to distill pure control value. 2/16/01 Tr. at 522-539.

49 . See, e.g., Borruso, 753 A.2d at 459 (reducing adjustment by 10% on this basis); Hintman v. Fred Weber, Inc., Del.Ch., C.A. No. 12839 mem. op. at 24, Steele, V.C. (Feb. 17, 1998) (recognizing the imprecision of attempting to parse out excludible value from control premium data but accepting an expert’s conclusion that mean control premiums in data should be reduced from 45% to 20% to determine the adjustment to minority value); Kleinwort, 1995 WL 376911, at *4 (discounting available premium data from 34-48% to 12.5%).

As a thought-provoking study of this subject suggests, our case law has yet to fashion a coherent approach to this difficult question of valuation. Coates, supra note 28, at 1257-1285.

50 .See PATRICK A. GAUGHAN, MERGERS, ACQUISITIONS, AND CORPORATE RESTRUCTURINGS 527 (2d. ed.1999) (citing George R. Roach, "Control Premiums and Strategic Mergers," Business Valuation Review, June 1998, at 42-49 (discussing Roach study of control premiums paid between 1992 and 1997 "failed to find any difference in the control premium for those deals in which the merging companies have the same or different Standard Industry Classification (SIC codes). This implies that strategic focus is not a determinant of merger premiums.”)) (JE 63); (Alexander R. Slusky & Richard E. Caves, Synergy, Agency and the Determinants of Premia Paid in Mergers 39 J. Industrial Econ. 277, 289-93 (1991) (finding that transactions in study sample that would be considered “synergy” deals had an average premium less than non-synergy deals, and that their study revealed "no evidence of real synergies” that could be measured as an explanation of observed premiums)) (JE 55).

51 . JE 29 (41.3% average premium for all 1998 mergers tracked by Mergerstat®).

52 . JE 30.

53 . Eisenhofer & Reed, supra note 38, at 127 n. 362 (citing Mergerstat® study for 1984-1995 that appears to discount total control premiums by 26% to 30.9% to arrive at implied minority discounts).

54 .See Borruso, 753 A.2d at 459; Bomarko, 1999 WL 33472141, at *22.

55 . See generally Bajaj, et al., supra note 38, at 2 ("In general, investors value marketability. Therefore, other things being equal, investors will pay more for an asset that is readily marketable than for an otherwise identical asset that is not readily marketable. The usual valuation methodologies, which utilize cash flows or market transactions, do not explicitly account for the marketability of an asset. Hence, in order to value an asset that is not marketable, the usual approach is to value the asset as if it were marketable, then apply a marketability discount to this estimated value.”).