Delaware Appraisal Litigation: Non-Arm’s-Length Transactions, Arm’s-Length Transactions and the Anna Karenina Principle

In this paper, we explore a variety of issues related to statutory rights of appraisal in Delaware, and the search by which to determine the sometimes elusive concept of fair value. In the course of so doing, we: (i) discuss the statutory definition of fair value and some of the case law doctrines surrounding its application in appraisal litigation; (ii) observe and comment on the fact that, in transactions where independent fairness opinions or valuations are provided, the median premium over the transaction price in appraisals in non-arm’s-length transactions is materially greater than is the case with arm’s-length transactions; (iii) describe how the Delaware Court of Chancery picks and chooses among the methodologies selected by the parties’ experts to arrive at its fair value conclusions and; (iv) conclude with observations concerning what the case law tells us are the principle do’s and don’ts in the preparation of financial analyses and testimony by which to determine fair value.

Fair Value Defined

In relevant part, Section 262(h) of the Delaware General Corporation Law provides that dissenting shareholders in all-cash transactions, and any other mergers in which the shareholders receive any consideration other than listed stock, who have otherwise complied with the statute, shall be entitled to obtain a judicial determination of the “fair value” of their shares in the Court of Chancery. In determining fair value, “the Court shall take into account all relevant factors.” Section 262(h) provides that fair value shall exclude any element of value arising from the accomplishment or expectation of the merger, although the characterization of what value does or does not arise from the accomplishment or expectation of the merger is highly fact-sensitive. Dissenters receive fair value plus interest from the date of the transaction.

Under Weinberger, all valuation methods customarily accepted in the financial community may be employed. In Cavalier, the Delaware Supreme Court ruled that a minority discounts and discounts for lack of marketability were impermissible because “fair value” requires that petitioners receive the value taken away from them, i.e., petitioner’s proportionate interest in a going concern.

In appraisal proceedings, each side has the burden of proof of fair value, recognizing that the “value of a corporation is not a point on a line but a range of reasonable values.” Finally, fair value appraisal proceedings may be coupled with claims for breach of fiduciary duty in which plaintiffs allege the absence of fair process or fair price in the transaction in question. That said, the amount of damages chargeable against individual defendants in such cases typically does not vary from the fair value figure determined in the appraisal portion of the case.

The Anna Karenina Principle

In his classic novel, Anna Karenina, Leo Tolstoy observes that “[h]appy families are all alike; every unhappy family is unhappy in its own way.” In statistics, this principle is used to describe significance tests: there are many ways in which a dataset may violate the null hypothesis and only one in which all the assumptions are satisfied.

The appraisal cases we present illustrate the presence of the Anna Karenina principle, but also many commonalities. The Anna Karenina characteristics include alleged appropriation of corporate opportunities and director exculpation in breach of fiduciary duty cases, as well as questions re collateral estoppel, and whether post-hoc adjustments to management’s projections are permissible.

These Anna Karenina principles notwithstanding, there are far more instances of themes that get repeated through most of the cases. These include the acceptance or rejection of the discounted cash flow (DCF) approach and the distinctly second-fiddle role generally played by other valuation methodologies, as well as common tropes such as whether a transaction price is an element of value to be considered in determining fair value (in general, yes if there has been a “robust” auction with a diverse universe of prospective buyers, no if not). The same squabbles in applying DCF analysis found in Chancery decisions generally are also found in fair value disputes. These include which cash flow projections to use (if any); elements affecting the discount rate including equity risk premium, small stock adjustments, company-specific risk premiums, beta (or its absence when using the build-up method); terminal value calculations (perpetuity versus exit multiple); and adjustments to the subject company’s net cash; and value of non-operating assets.

The Imprecision of DCF

In fair value litigation, the inherent sensitivity of the components of DCF analyses to small changes in projected cash flows, discount rates, and/or terminal values can result in major deviations in appraised values.

As an example, we calculated the DCF values as of December 31, 2017, of a hypothetical company with projected revenues in 2018 of $100 million, an EBITDA margin of 25%, depreciation of $10 million, capital expenditures of 115% of depreciation, working capital of 10% of revenues, 5% annual growth through 2022, annual growth of 3% or 4% thereafter, and no net debt (debt minus excess cash). Table 1 below shows the wide variances brought about by differences in the discount rate or the perpetual growth rate.

The variances increase substantively with leverage. Table 2 shows the calculated values assuming $300 million of net debt. Without leverage. the difference between the calculated value using a 3% growth rate and a 10% discount rate and using a 4% growth rate and an 8% discount rate is 108%.

In this leveraged example, the difference between the calculated value using a 3% growth rate and a 10% discount rate and using a 4% growth rate and an 8% discount rate is 242%!

Moreover, DCF valuations are subject to the vagaries of the underlying projections, which can be material. Cases in which competing experts use the same projections for their DCF calculations are rare.

It appears that the Delaware courts favor DCF because it can yield a precise number. Many decisions arrive at DCF values calculated to four significant figures. Unfortunately, DCF suffers from the fact that it merely gives the illusion of precision.

It’s a fact of life that Court of Chancery judges and others have railed at the often-dramatic differences between the fair value determinations of the dueling experts in a given case. Equally true, experts should want to avoid getting skewered in a deposition or cross examination, or by the court. Notwithstanding, adversarial proceedings being what they are, critics should not be shocked when experts come in with fair value conclusions they believe to be within the range of fair values but appear skewed to one end of the range or the other. This is especially true because, as noted, even small changes to inputs in a DCF analysis can dramatically affect calculated values. Similarly, differences in the components of comparable company or comparable transaction analyses, such as which companies to include or which outliers to exclude, can likewise make major (albeit usually smaller) differences in the conclusions reached.

Characteristics of Arm’s-Length and Non-Arm’s-Length Transactions

Arm’s-length transactions are those in which the buyer is not related to the seller and in which controllers and/or officers of the seller receive the sane consideration as other shareholders (other than pre-existing commitments such as severance payments and stock options).

There are at least four factors that characterize non-arm’s-length transactions:

1. in its most extreme form, where buyer is on both sides of the transaction, where buyer controls both parties to the transaction;
2. situations in which the majority acquires at least 90% of a company and later squeezes out the minority in a short-form merger;
3. management buyouts (MBOs), and;
4. transactions in which controllers and/or officers of the seller receive substantially different consideration than other shareholders, such as generous employment contracts, non-compete agreements, or equity in the surviving entity.

The Data

Exhibit I in the complete paper (available here) contains, among other elements, summaries of the Delaware appraisal decisions in the 25 non-arm’s-length transactions in our survey. We have included all relevant decisions from 1998 through February 2018 in which an independent fairness opinion or valuation was rendered. Exhibit I includes the relevant citation, the appraisal awards compared to the transaction prices, the form of each transaction, the firm rendering an opinion, and whether that firm was engaged by the board of directors, a special committee, or a control party. Exhibit II contains the same data for the 16 arm’s-length transactions with fairness opinions in the same period.

The premiums of the appraisal awards over that the transaction prices in the non-arm’s-length deals include seven with premiums over 100% and three with discounts. The median premium was 29.2% and the trimmed mean (excluding the highest and lowest data points) was 58.5%. The arm’s-length deals have a narrow range with a median of 0.0% and a trimmed mean of minus 3.1%.

Issues in Non-Arm’s-Length Transactions

Exhibit III contains snapshot details of valuation methods used by the Court and by investment banks in the fairness opinions or independent valuations in the 25 non-arm’s-length transactions. Exhibit IV discusses some of the issues discussed by the Court in these transactions. The cases we present illustrate the presence of the Anna Karenina principle, but also many commonalities.

Exhibit III shows Delaware’s overwhelming preference for DCF methodology. In only three of these 25 appraisals was the income approach not employed by the Court. Conversely, in 16 out of the 25, DCF (or another form of the income approach) was the only methodology used. Five used both the income method and the market approach (comparable companies and/or comparable transactions) and one used capitalized earnings and book value.

It is interesting to contrast the methods used by the Court of Chancery with those used by the investment banks. Of the 25 opinions and valuations for which information was available, 17 used both the income method and the market method. Five apparently used the income method only, two used DCF and asset value, and one used a single comparable company (the Court did the same in that case).

Exhibit IV contains case descriptions of Anna Karenina issues as well as more common ones. Anna Karenina issues include collateral estoppel, exculpation of directors, whether the entire fairness doctrine can be invoked absent a controlling shareholder, when a controlling shareholder’s duty to disclose ceases, tax liability of Sub Chapter S corporations, the treatment of excess regulatory capital, whether preferred stock should be valued at its liquidation preference or on an as-converted basis, the dilutive effect of long-term stock options, adjustments for misappropriated corporate opportunities, and whether 9/11-related facts should be used to reduce value.

The more quotidian themes spread over many cases include, inter alia, which valuation methods to employ, the transaction price as evidence of value, cost of capital issues (e.g., debt/equity ratios, beta, quantification of small stock premiums, applicability of company-specific risk premiums), which projections to employ, growth rates in terminal value calculations, and present value of net operating loss carryforward.

Research Results—Premiums or Discounts to Transaction Prices

We observe that of the 25 non-arm’s-length transactions studied, 13 sought appraisal relief only, and 12 combined an appraisal request with an allegation of breach of fiduciary duty. In five cases, the Court found no breach of fiduciary duty. Exhibit V shows the premiums awarded in each of these 18 pure appraisal decisions and in the seven that combined an appraisal with a successful breach of fiduciary duty claim. The median premium in the first group was only 19.7%, but the median was 132.5% where appraisal was coupled with a finding of breach of fiduciary duty. The trimmed means were 28.8% and 127.3%, respectively.

Did the courts’ disapproval of respondents/defendants’ conduct cause it to award higher premiums to transaction prices versus the appraisal-only cases? Although the small sample size in the breach of fiduciary duty cases does not allow for a definitive conclusion on this matter, we note then-Vice Chancellor Strine’s observation in a joint appraisal/fiduciary duty case:

Because of the relationship between the appraisal and equitable actions, I have, at the margins, in fact resolved doubts in favor of the plaintiffs. In other words, the valuation I set forth is more optimistic than is strictly justified.

We question the concept that fair value should be determined differently in breach of fiduciary actions than in pure appraisal cases. Fair value should be the same regardless of either the flaws in the transaction or the behavior of any party. If there are damages from an egregious breach of fiduciary duty, exemplary (punitive) damages over and above fair value may lie.

Issues in Arm’s-Length Transactions

Exhibit VI contains snapshot details of valuation methods used by the Court of Chancery (and, in two cases, by the Supreme Court) and in investment banks’ 20 fairness opinions in the 16 arm’s-length transactions (four of these transactions had two fairness opinions). Exhibit VII discusses some of the issues discussed by the Court in these transactions.

Exhibit VI summarizes valuation approaches employed by fairness opinion providers or appraisers. It shows Delaware’s strong preference for relying on the transaction price in arm’s-length deals. In nine of these 16 appraisals, the Court of Chancery based its appraisal solely or primarily on the transaction price. Six Chancery decisions were based on DCF, but two 2017 Supreme Court rulings reversed lower court decisions that had used DCF and instructed the Court of Chancery to give predominant weight to the transaction price. A February 2018

Chancery decision cited the Supreme Court decisions and based its appraisal on unaffected market price prior to announcement of the transaction. In light of this decision, in future cases counsel may choose to engage experts to perform event studies with respect to stock volume, bid-ask spreads, and other such measures to determine whether the efficient market hypothesis applies in the fact pattern at bar.

The fairness opinions rendered prior to these transactions were each opining as to the fairness of the negotiated price of the proposed transaction. Of the 20 fairness opinions, 18 explicitly considered both DCF and comparable companies; 14 of these also considered comparable transactions. The other two decisions noted that the fairness opinions (which were for private companies) had employed DCF but did not discuss whether or not they had used any other method.

In Exhibit VII, Anna Karenina issues include whether a dissenter can withdraw some but not all its shares from its appraisal demand, the admissibility of valuation treatises not entered into evidence, impact on appraisal of taxes and expenses relating to sale of a business that was contingent on sale of the continuing company, and whether a financial buyer’s target IRR affects whether the price it is willing to pay is a meaningful indication of fair value.

Other issues include, in addition to those in Exhibit IV, the relevance of synergies that were not unique to the buyer, treatment of stock-based compensation in DCF analyses, the use of Barra betas, and whether cash held abroad should be adjusted for taxes payable upon repatriation.

Conclusions

1. The Court of Chancery will almost always favor management’s ex ante projections made in the ordinary course of business for corporate planning purposes over projections made for marketing the company or post ante projections made in connection with litigation.
2. Transaction price is a major factor in appraisal if there has been a robust market check. The absence of an adequate market check in many non-arm’s- length transactions has resulted in appraisal valuations materially higher than the deal price.
3. Delaware seldom awards a premium over the transaction price in appraisals of arm’s-length transactions.
4. The Court of Chancery has shown a strong preference for the income approach (primarily DCF) over the market approach.
   1. The capital asset pricing model (CAPM) is preferred to the build-up method, as the latter is more subjective.
   2. For calculating terminal value, a growth model is preferred, both because it is preferred by academics and because the multiples used by experts (usually EBITDA) are commonly based on current market multiples without regard for what multiples might be at the end of the projection period.
   3. The Court had developed a preference for a supply side equity risk premiums (ERP) rather than an historical ERP.
5. In future cases, experts may be asked to perform event studies to determine whether the efficient market hypothesis applies in the fact pattern at bar.
6. The Anna Karenina principle is alive and well in the Delaware courts.

The complete paper, including footnotes and exhibits, is available for download here.

The authors thank Ira T. Kay, a Managing Partner of Pay Governance, LLC, for his suggestion that an article comparing transaction price to the court-adjudicated value in an appraisal proceeding might be a useful contribution to valuation literature. We also appreciate information supplied by Charles Nathan, a Senior Advisor to the Finsbury subsidiary of WPP plc.