Controlling Controlling Shareholders: New Limits on the Operate, Sale of Control, and Freeze Out Alternatives
Publication Date:January 01, 2003
"University of Pennsylvania Law Review" article coauthored by Professor Ronald Gilson, is among the "Top 10 Corporate and Securities Articles of 2004" according to "The Corporate Practice Commentator"
Full Text of Publication
DOCTRINES AND MARKETS CONTROLLING CONTROLLING SHAREHOLDERS RONALD J. GILSON† & JEFFREY N. GORDON†† INTRODUCTION The rules governing controlling shareholders sit at the intersection of the two facets of the core agency problem in United States public corporations law. The first is the familiar principal-agency problem that arises from the separation of ownership and control. With only this facet in mind, the presence of a large shareholder may better police management than the standard panoply of marketoriented techniques. The second is the agency problem that arises between controlling and non-controlling shareholders, which produces the potential for private benefits of control—benefits to the controlling shareholder not provided to the non-controlling shareholders. There is, however, a point of tangency between these facets. Because there are costs associated with holding a concentrated position and with exercising the monitoring function, some private benefits of control may be necessary to induce a party to play that role. Thus, from the public shareholders’ point of view, the two facets of the agency problem present a tradeoff. The presence of a controlling shareholder reduces the managerial agency problem, but at the cost of the private benefits agency problem. Non-controlling shareholders will prefer the presence of a controlling shareholder so long as the † Charles J. Meyers Professor of Law and Business, Stanford Law School, and Marc and Eva Stern Professor of Law and Business, Columbia Law School. Contact at rgilson@ stanford.edu. †† Alfred W. Bressler Professor of Law, Columbia Law School. Contact at jgordon@ law.columbia.edu. We are grateful to Frank Balotti, Bernard Black, Steven Fraidin, Michael Klausner, Ted Mirvis, Victor Lewkow, Charles Nathan, Adam Pritchard, Mark Roe, Kenneth Scott, Vice Chancellor Leo Strine, and participants at the Symposium on Control Transactions cosponsored by the University of Pennsylvania’s Institute of Law and Economics and the Law Review and at faculty presentations at Stanford and Harvard Law Schools and to the University of Pennsylvania Law Review for helpful comments on an earlier draft of this Article. We are also grateful to Scott Ashton for research assistance. 786 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 benefits from reduction in managerial agency costs are greater than the costs of private benefits of control.1 The terms of this tradeoff are determined by the origami of judicial doctrines that describe the fiduciary obligations of a controlling shareholder. In this Article, we examine the doctrinal limits on the private benefits of control from a particular orientation. As we will show, a controlling shareholder may extract private benefits of control in one of three ways: by taking a disproportionate amount of the corporation’s ongoing earnings, by freezing out the minority, or by selling control. Our thesis is that the limits on these three methods of extraction must be determined simultaneously, or at least consistently, because they are in substantial respects substitutes. We then consider a series of recent Delaware Chancery Court decisions that we argue point in inconsistent directions—on the one hand, reducing the extent to which a controlling shareholder can extract private benefits through selling control and, on the other, increasing the extent to which private benefits can be extracted through freezing out noncontrolling shareholders. While judicial doctrine is too coarse a tool to specify the perfect level of private benefits, we believe these cases get it backwards. The potential for overreaching by controlling shareholders is greater from freeze-outs than from sales of control, so a shift that favors freeze-outs as opposed to sales of control is a move in the wrong direction. In Part I, we develop the simultaneity framework for controlling private benefits of control and describe briefly the general doctrinal structure. In Part II, we review and evaluate recent Delaware case law regarding sale of control and minority freeze-outs. In particular, we argue that the Delaware law of freeze-outs can be best reunified by giving “business judgment rule” protection to a transaction that is approved by a genuinely independent special committee that has the power to say “no” to a freeze-out merger, while also preserving what amounts to a class-based appraisal remedy for transactions that proceed by freeze-out tender offers without special committee approval. Part III concludes that, although some may disagree with our views concerning the appropriate levels of restriction governing techniques for extracting the private benefits of control, the terms of the debate will be more sharply focused if the rules governing these techniques are evaluated simultaneously. 1 RONALD J. GILSON & BERNARD S. BLACK, THE LAW AND FINANCE OF CORPORATE ACQUISITIONS 1229-31 (2d ed. 1995). 2003] CONTROLLING CONTROLLING SHAREHOLDERS 787 I. PRIVATE BENEFITS OF CONTROL: THE LINK BETWEEN EXTRACTING PRIVATE BENEFITS FROM OPERATING, SELLING CONTROL, OR FREEZE-OUTS Imagine that a controlling shareholder can extract benefits from its ongoing operation of the company. For example, the controlling shareholder can take out significant benefits through cost-sharing arrangements that overpay the controlling shareholder for providing central services such as pension, accounting, or the like. Alternatively, the controlling shareholder can benefit through “tunneling”—that is, through contractual dealings with the company, like transfer pricing, that favor the controlling shareholder.2 In either event, the controlling shareholder secures value from its control position that is not received by the non-controlling shareholders. In turn, the controlling shareholder can extract the same value from control by selling it at a premium to the value of the noncontrolling shares. The existence of an ongoing stream of private benefits increases the value of the controlling shares compared to the non-controlling shares by the present value of the future private benefits. A sale of control simply capitalizes the cash flow associated with private benefits of control. The same private benefits can also be secured by freezing out the minority shareholders. In a public corporation, the trading price of shares in a corporation with a controlling shareholder reflects the value of a non-controlling share.3 The price of a non-controlling share will have been discounted by the capitalized value of the controlling shareholder’s private benefits. A freeze-out at the discounted price allows the controlling shareholder to capture the capitalized value of future private benefits. The critical point is that, without more, we should expect doctrinal regimes of equivalent rigor to cover each of the three methods of extracting private benefits. While which technique a controlling shareholder resorts to will depend on the particular circumstances, as yet there is no reason to favor one method over another. In fact, the legal rules that govern the three methods are quite different. One set of legal rules specifies the boundaries for private benefits in the ongoing operation of the corporation.4 A second addresses efforts by a 2 For one description of this strategy, see Simon Johnson et al., Tunneling, AM. ECON. REV., May 2000, at 22, 22. 3 GILSON & BLACK, supra note 1, at 1234. 4 See infra text accompanying notes 7-19 (discussing this particular set of rules). 788 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 controlling shareholder to sell control at a premium not shared with others. A third polices freeze-outs of non-controlling shareholders. As we will see, the rules controlling the level of private benefits from operations are the central determinant of the judicial doctrine that controls controlling shareholders; these rules set the level of private benefits that can be appropriately capitalized through sale of control or a freeze-out. The rules governing a sale of control and those governing a freeze-out of non-controlling shareholders are quite different from one another. There is quite limited judicial intervention in the case of sales of control and quite intensive judicial intervention in the case of minority freeze-outs. In this Part, we argue that this pattern of judicial intervention represents the right relationship: more intense judicial review is appropriate in a freeze-out than in a sale of control. The objective of the legal rules in both the sale of control and freeze-out cases should be identical: to protect the controller’s continuing claim to the permissible level of private benefits while limiting the controller’s take to that level plus an appropriate share of the synergy gains. This is much easier to achieve in a sale of control, where continuing shareholders participate pro rata in synergy gains, than in a freezeout, where the synergy gains must be priced and allocated as part of the freeze-out price. In the next Part, we argue that recent Delaware case law is moving in the wrong direction. Getting it right is not a matter of indifference. A significant body of scholarship links capital market development and public shareholder protection.5 As we will see, legal rules and the enforcement mechanisms for those rules affect the “minority discount”—that is, the value difference between the shares of equivalent cash flow rights held by public shareholders and by controlling shareholders.6 The minority discount in turn affects the feasibility of “equity carve-outs,” transactions in which a parent sells a minority interest in a subsidiary via an 5 The literature is summarized in Rafael La Porta et al., Investor Protection and Corporate Governance, 58 J. FIN. ECON. 3 (2000). See also Rafael La Porta et al., Law and Finance, 106 J. POL. ECON. 1113 (1998) (examining the legal rules protecting corporate shareholders in common law and civil law countries and hypothesizing that small, diversified shareholders are not likely to hold much influence in countries that do not protect their rights); Rafael La Porta et al., Legal Determinants of External Finance, 52 J. FIN. 1131, 1131 (1997) (postulating that countries with poorer investor protections have smaller and narrower capital markets). 6 See Alexander Dyck & Luigi Zingales, Private Benefits of Control: An International Comparison, 59 J. FIN. (forthcoming 2004) (documenting cross-country differences in private benefits); Tatiana Nenova, The Value of Corporate Voting Rights and Control: A Cross-Country Analysis, 68 J. FIN. ECON. 325, 342 (2003) (same). 2003] CONTROLLING CONTROLLING SHAREHOLDERS 789 initial public offering (IPO), and also affects, generally, the value of control transactions where some shares remain in public hands. A. Private Benefits of Control in Operating the Company: The Sinclair Standard The legal rules governing private benefits of control in operating a company set the limits on the price of monitoring by a controlling shareholder. If these limits are effective, the presence of a controlling shareholder benefits the non-controlling shareholders because the reduction in managerial agency costs will exceed the level of private benefits. Two basic legal rules police the level of private benefits that result from ongoing operations. First, if the controlling shareholder is a director, then any contract between the controlling shareholder and the corporation is an interested transaction and must meet the standards of statutes like Delaware General Corporation Law section 144,7 which requires that the transaction be sanitized through either procedural techniques or substantive judicial review.8 If the controlling 7 DEL. CODE ANN. tit. 8, § 144 (2001). 8 Delaware General Corporation Law section 144 provides in pertinent part: (a) No contract or transaction between a corporation and 1 or more of its directors or officers, or between a corporation and any other corporation, partnership, association, or other organization in which 1 or more of its directors or officers, are directors or officers, or have a financial interest, shall be void or voidable solely for this reason, or solely because the director or officer is present at or participates in the meeting of the board or committee which authorizes the contract or transaction, or solely because any such director’s or officer’s votes are counted for such purpose, if: (1) The material facts as to the director’s or officer’s relationship or interest and as to the contract or transaction are disclosed or are known to the board of directors or the committee, and the board or committee in good faith authorizes the contract or transaction by the affirmative votes of a majority of the disinterested directors, even though the disinterested directors be less than a quorum; or (2) The material facts as to the director’s or officer’s relationship or interest and as to the contract or transaction are disclosed or are known to the shareholders entitled to vote thereon, and the contract or transaction is specifically approved in good faith by vote of the shareholders; or (3) The contract or transaction is fair as to the corporation as of the time it is authorized, approved or ratified, by the board of directors, a committee or the shareholders. Id.; see also Oberly v. Kirby, 592 A.2d 445, 466 (Del. 1991) (“[W]here an independent committee is not available [to approve an interested transaction under § 144], the stockholders may either ratify the transaction or challenge its fairness in a judicial forum . . . .”). 790 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 shareholder is not a director, then Sinclair Oil Corp. v. Levien9 applies, which sets out the general standards for the conduct of controlled corporations.10 For this purpose, the Delaware Supreme Court essentially divides sources of private benefits into two categories. The first category concerns the business and strategic decisions of the corporation. In Sinclair, for example, a minority shareholder of Sinven Venezuelan Oil Company, a controlled corporation that operated primarily in Venezuela, claimed that Sinven’s dividend policy favored the controlling shareholder, Sinclair Oil Corporation.11 By paying out as dividends a large percentage of Sinven’s profits, Sinven was alleged to favor the controlling shareholder, which apparently had attractive investment opportunities outside of the controlled corporation, and to disadvantage the non-controlling shareholders, who received equal dividends but lost the opportunity for the controlled corporation to reinvest its earnings.12 The second category concerns the core aspect of private benefits— the controlling shareholder’s direct dealings with the controlled corporation. Here we are in the realm of true self-dealing—unfair transfer pricing, the transfer of assets from the controlled corporation to the controlling shareholder, and the use of the controlled corporation’s assets as collateral for a controlling shareholder’s debt. The standards established for the two categories of private benefits are radically different. In general, courts treat business and strategic decisions that even-handedly affect the controlling and noncontrolling shareholders essentially as business judgments. Thus, the Delaware Supreme Court handled the dividend decision in Sinclair, as well as the related claim that the controlled corporation’s business was limited to the development of oil opportunities in Venezuela (presumably why the controlled subsidiary was in a position to pay such 9 280 A.2d 717 (Del. 1971). 10 See id. at 720 (invoking the intrinsic fairness standard when the fiduciary duty a parent owes its subsidiary “is accompanied by self-dealing—the situation when a parent is on both sides of a transaction with its subsidiary”). 11 Id. at 719-21. 12 The dividend may also have had a differential tax impact on minority shareholders. Depending on whether the subsidiary was part of an affiliated group, at least eighty percent and as much as one hundred percent of the dividends received by the parent would not be taxed. See GILSON & BLACK, supra note 1, at 1239-41 (explaining the impact of dividend-received deductions on the decision to employ a minority freezeout). Minority shareholders would be taxed on dividends received unless they were otherwise exempt. 2003] CONTROLLING CONTROLLING SHAREHOLDERS 791 large dividends), as business judgments, and thereby outside the realm of intrusive judicial review.13 In contrast, core self-dealing is held to a dramatically different standard. If the controlling shareholder appears to benefit at the expense of the controlled corporation (for example, when the controller disparately gains from contract terms or the enforcement of those terms where the two parties are on opposite sides), the intrinsic fairness standard—the most rigorous in corporate law jurisprudence— applies. In that situation, the controlling shareholder bears the burden of proving that the terms of the transaction were intrinsically fair, with the court making a de novo determination.14 These two standards thus allow some range of private benefits of control but, consistent with the minority shareholders’ calculus, at a level that still may make the non-controlling shareholders better off.15 What kind of private benefits remain? At the most benign, maintaining a publicly traded, majority-owned subsidiary may benefit the controlling shareholder by more effectively opening the controlled company’s performance to public scrutiny, thereby assuring more accurate pricing of the controlled corporation’s business than if it was bundled with that of the controlling shareholder. Reciprocally, the controlling shareholder may then make use of market signals to help assess its own and the controlled corporation’s business prospects as well as the performance of the controlled corporation’s management. Additionally, controlling shareholders may use market signals to devise more accurate incentive compensation for the management and employees of both corporations.16 In these circumstances, the 13 280 A.2d at 722. 14 Id. at 720. 15 The efficacy of these standards is offered by Johnson et al., supra note 2, at 26, as an explanation for the absence of pyramidal structures in the United States: “Perhaps the reason that pyramidal group structures are relatively rare in the United States and the United Kingdom [yet ubiquitous elsewhere in Europe] is that many transactions inside a group would be challenged on fairness grounds by minority shareholders of subsidiaries, who would get a receptive hearing in court.” 16 These reasons are commonly offered as explanations for the efficiency of equity carve-outs. See Katherine Schipper & Abbie Smith, A Comparison of Equity Carve-outs and Seasoned Equity Offerings: Share Price Effects and Corporate Restructuring, 15 J. FIN. ECON. 153, 182 (1986) (“[T]he equity carve-out may improve public understanding of the subsidiary’s growth opportunities.”); see also Anand M. Vijh, The Positive Announcement- Period Returns of Equity Carveouts: Asymmetric Information or Divestiture Gains?, 75 J. BUS. 153, 189 (2002) (“[T]he market reacts positively to the announcement of carveouts because it thinks that carveouts create value by divesting unrelated businesses, enabling a complete spinoff or a third-party acquisition, providing new financing, undertaking new investments, and reducing stock complexity.”). Announcement of such 792 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 non-controlling shareholders get more focused monitoring at a relatively low cost.17 Other conduct involving private benefits that does not involve core self-dealing may be more costly. Here we have in mind a variety of business decisions that, while not rising to the level of business opportunities, may provide the controlling shareholder a benefit that would not otherwise be available to it, even if the controlled corporation does not directly bear an offsetting cost. These decisions seem to us to have the character of real options: for example, where the activities of the controlled corporation may keep open a strategy for the controlling corporation.18 Nonetheless, this source of private benefit remains limited, certainly when compared to core self-dealing.19 transactions results in a slightly less than two percent positive abnormal return in the parent company’s stock. See Schipper & Smith, supra, at 153 (“These gains are in contrast to the average abnormal loss associated with announcements of seasoned equity offerings.”); see also Heather M. Hulburt et al., Value Creation from Equity Carve-Outs, 31 FIN. MGMT. 83, 99 (2002) (finding that empirical tests support the divestiture gains hypothesis because “rivals of parent firms exhibit negative stock price reactions to equity carve-out announcements”). Additional explanations for this gain include the signal that the parent company’s stock is undervalued (otherwise the offering would have been of parent stock), see Vikram Nanda, On the Good News in Equity Carve-Outs, 46 J. FIN. 1717, 1733 (1991) (“[T]he choice of financing decision may provide information not just about the subsidiary’s assets in place but also about the value of the assets in place of the rest of the corporation.”), and the increased analyst coverage of both companies’ stock, see Stuart C. Gilson et al., Information Effects of Spin-offs, Equity Carve-outs, and Targeted Stock Offerings 18 (June 1998) (unpublished manuscript) (“Investment bankers . . . often argue that the level and quality of analyst coverage significantly improves following these transactions.”), available at http://papers.ssrn.com/ sol3/papers.cfm?abstract_id=42904. In the end, some controversy remains about the source of abnormal returns. See David Haushalter & Wayne Mikkelson, An Investigation of the Gains from Specialized Equity: Tracking Stock and Minority Carve-Outs 24 (May 29, 2001) (unpublished manuscript) (conjecturing that “the stock price effects do not reflect real benefits of specialized equity arrangements”), available at http:// papers.ssrn.com/sol3/papers.cfm?abstract_id=271691. For a more skeptical view of valuation creation by carve-outs and carve-out stability, see André Annema et al., When Carve-outs Make Sense, MCKINSEY Q. No. 2, at 13, 15 (2002) (finding that, two years after the carve-out, most carve-outs “destroy[ed] shareholder value” and warning that “[c]arving out even small stakes in subsidiaries will likely lead to complete and irreversible separation”). 17 This statement is consistent with empirical evidence indicating that carve-out subsidiary stocks do not underperform stock portfolio benchmarks, contrary to the usual findings of underperformance for IPOs or seasoned equity offerings. Anand M. Vijh, Long-Term Returns from Equity Carveouts, 51 J. FIN. ECON. 273, 275 (1999). 18 For this purpose it is useful to consider two different kinds of controlling shareholders: One group has a unidimensional relation to their portfolio company—that is, the controlling shareholder’s only connection with the company is its shareholdings. A second group, in contrast, has a multidimensional relation to their 2003] CONTROLLING CONTROLLING SHAREHOLDERS 793 Finally, there is a level of private benefits extraction that arises from non-actionable self-dealing. For example, a contract with a controlling shareholder at market prices will still impound market-level rents and will not reflect savings from reduced information asymmetries. The relationship may also entail “micro” self-dealing that in each instance is small but in the aggregate is significant. What’s important is that judicial doctrine effectively puts a ceiling on the private benefits of control associated with operating the corporation. Behavior that has the potential to transfer large amounts of value is subjected to intense judicial scrutiny, which is consistent with our hypothesis that controlling shareholders do not take markedly more from non-controlling shareholders than they provide.20 Thus, the level of private benefits from operations provides a benchmark for assessing the standards governing alternative methods of securing private benefits. B. Sale of Control at a Premium The second method by which a controlling shareholder may extract private benefits of control is by selling its control for a premium that reflects the capitalized value of the private benefits of control available from operating the controlled corporation. Although the holding in Perlman v. Feldmann21—that a controlling shareholder cannot sell control at a premium that is not shared with non-controlling shareholders22—continues to amuse corporate law teachers (both because it provides the basis for an interesting class and because of the Second Circuit’s Fantasia-like view of Indiana law), by the early 1990s, portfolio—that is, in addition to the controlling shareholder’s stock position, it also has an operational relation to the company, for example, as customer or supplier. . . . A unidimensional controlling shareholder has few channels by which to appropriate private benefits [from the controlled corporation]. GILSON & BLACK, supra note 1, at 1233. For the unidimensional controlling shareholder, real options may be the primary source of private benefits. Id. at 1233-34. 19 Id. at 1233-34. 20 John Coates is more pessimistic with respect to the potential size of private benefits that controlling shareholders can secure through operations. 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, 1314-27 (1999) (contending that controllers can readily shift value from controlled subsidiaries). However, much of Coates’s focus is on whether value-reducing operational decisions that affect all shareholders can be transmuted into private benefits in a freeze-out as a result of valuation standards. Id. 21 219 F.2d 173 (2d Cir. 1955). 22 Id. at 178. 794 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 the applicable legal rule was radically different. Whether one looks to Delaware case law23 or to the American Law Institute’s (ALI) Principles of Corporate Governance,24 the rule is clear: in general, a controlling shareholder can sell control at a premium that is not shared with noncontrolling shareholders. Given the limits on private benefits of control from ongoing operations, it seems clear that non-controlling shareholders would prefer a rule that allows controlling shareholders the right to sell their shares at a price that reflects the net present value of the flow of private benefits from operating the company.25 Correspondingly, a buyer of control presumably would not wish to acquire the controlled corporation at a price that reflects the capitalized value of private benefits unless it thought it could increase the value of its purchased interest. Because the amount of private benefits from operating the controlled corporation is capped by the legal rule applicable in that situation, the non-controlling shareholders will share any increase in value resulting from an increase in the common value of the controlled corporation. 26 23 The court in Harris v. Carter, 582 A.2d 222, 234 (Del. Ch. 1990), stated that, “[w]hile Delaware law has not addressed this specific question, one is not left without guidance from our decided cases. . . . [It] is [a] principle [of Delaware law] that a shareholder has a right to sell his or her stock and in the ordinary case owes no duty in that connection to other shareholders when acting in good faith.” See also Thorpe v. CERBCO, Inc., 676 A.2d 436, 442 (Del. 1996) (distinguishing the fiduciary relationship between directors and the corporation from the relationship between controlling shareholders and other shareholders); In re Sea-Land Corp. S’holders Litig., No. 8453, 1987 WL 11283, at *5 (Del. Ch. May 22, 1987) (“A controlling stockholder is generally under no duty to refrain from receiving a premium upon the sale of his controlling stock.”); cf. Robert W. Hamilton, Private Sale of Control Transactions: Where We Stand Today, 36 CASE W. RES. L. REV. 248, 249 (1985) (“It is unlikely that any American court today would reject the general proposition that controlling shareholders may obtain a premium for their shares which they need not share with other shareholders.”). 24 PRINCIPLES OF CORP. GOVERNANCE: ANALYSIS AND RECOMMENDATIONS § 5.16 (1994). 25 This discussion draws on GILSON & BLACK, supra note 1, at 1231-32, which in turn was informed by Frank H. Easterbrook & Daniel R. Fischel, Corporate Control Transactions, 91 YALE L.J. 698, 711 (1982) (arguing that minority shareholders want a rule that increases the market value of shares in the corporation). See also Lucian Arye Bebchuk, Efficient and Inefficient Sales of Corporate Control, 109 Q.J. ECON. 957, 974-81 (1994) (comparing the market rule with the equal opportunity rule in the context of corporate control transfers); Marcel Kahan, Sales of Corporate Control, 9 J.L. ECON. & ORG. 368, 378 (1993) (analyzing the legal rules governing the sale of corporate control). 26 An empirical study of Perlman v. Feldmann showed that the stock of Newport Steel, the controlled corporation, experienced abnormal returns of thirty-four percent during negotiations for sale of control and abnormal returns of seventy-seven percent 2003] CONTROLLING CONTROLLING SHAREHOLDERS 795 Next, assume that the new controller realizes certain synergies from its operation of the controlled corporation. Does this change the legal rule that non-controlling shareholders would choose? We think not. So long as the legal rules governing private benefits of control from operations do not allow all of the synergy to be captured by the controlling shareholder, the non-controlling shareholders will participate in the value increase resulting from the sale of control. This is a plausible assumption given that actually achieving synergy will require direct interaction between the controlling shareholder and the controlled corporation, an interaction that will be subject to Sinclair.27 There are exceptions to the permissive general rule, but these seem to fit well within the present analysis. Section 5.16 of the ALI Principles of Corporate Governance states these exceptions.28 The general rule that a controlling shareholder can sell its shares at a premium is qualified in two circumstances: first, when the controlling shareholder acquires shares from non-controlling shareholders in anticipation of the contemplated sale of control without disclosure and, second, when it is apparent that the purchaser is likely to extract illegal levels of private benefits from operating the controlled corporation.29 (twenty-nine percent on an industry-adjusted basis) over the entire year during which control was sold. Michael J. Barclay & Clifford G. Holderness, The Law and Large-Block Trades, 35 J.L. & ECON. 265, 270-71, 270 n.7 (1992). Because market price measures the value of the public minority shares, the data suggest that the minority shareholders benefited from the sale of control. The experience of Newport’s non-controlling shareholders seems to be a generalizable one. On the other hand, it is also possible to see Perlman v. Feldmann as correctly decided on its own peculiar facts, namely the Korean War price controls that produced a valuation gap between the capped wholesale price of steel and the value of the steel to end-users, whose products were not price-capped. 219 F.2d at 175. To try to capture this difference, Newport had insisted that customers provide advances against future purchases, i.e., interest free loans. Id. at 177. Even if the end-users who acquired control of Newport continued to make these advances on their purchases, if the present value of the interest-rate differential was less than the steel-product valuation gap, then at least part of the control premium can be considered a form of special synergy gain that, because it was not ratably shared with the minority, was properly subject to recovery. 27 See supra notes 9-14 and accompanying text for a discussion of the standards imposed by Sinclair. 28 PRINCIPLES OF CORP. GOVERNANCE: ANALYSIS AND RECOMMENDATIONS § 5.16 (1994). 29 Section 5.16 (a) and (b) restrict a controlling shareholder’s right to sell control at a premium if: (a) The controlling shareholder does not make disclosure concerning the transaction to other shareholders with whom the controlling shareholder deals in connection with the transaction; or 796 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 The first exception operates merely as a form of insider trading regulation. The second backstops the rule, limiting the level of private benefits from operations. In circumstances of looting, the controlling shareholder may be judgment proof. The exceptions provide an alternative source of recovery when the seller of control should have known what was coming. In short, the legal rule governing receipt of private benefits through sale of control fits nicely with the legal rule governing the level of private benefits from ongoing operations of the company. Except when there is reason to believe that the operating rules will be violated following the sale, there is no reason for a more restrictive rule. Put differently, if the stream of private benefits from operations is effectively controlled, there is no need to regulate the transfer of its capitalized value.30 C. Freeze-out of Minority Shareholders The third method by which a controlling shareholder can extract private benefits of control is through freezing out minority shareholders at a market price that reflects a discount equivalent to the private benefits of control available from operating the controlled corporation. In contrast to the simple permissive rules governing the sale of control at a premium, the rules governing minority freeze-outs are both complex and restrictive. The modern law of minority freeze-outs dates to the Delaware Supreme Court’s decision in Weinberger v. UOP, Inc.31 In that case, the (b) It is apparent from the circumstances that the purchaser is likely to violate the duty of fair dealing . . . in such a way as to obtain a significant financial benefit for the purchaser or an associate. Id. § 5.16(a)–(b) (citations omitted). 30 Thus, one way to understand “mandatory bid” systems that are common to takeover practice in the European Union, see infra note 115 (last paragraph), which give public minority shareholders an exit right upon an acquisition of control, is in terms of the differential capacity of legal systems to articulate and enforce minority shareholder rights. 31 457 A.2d 701 (Del. 1983). The present historical account relegates Delaware’s flirtation with a business purpose test as a precondition to a freeze-out—announced in Singer v. Magnavox Co., 380 A.2d 969, 975-76 & n.5, 982 (Del. 1977) (apparently in response to pressure from the federal courts) and overruled in Weinberger, 457 A.2d at 715—to accounts more concerned with the impact of federalist considerations on the development of corporate law. For one such account, see GILSON & BLACK, supra note 1, at 1254-69. See also Mark J. Roe, Delaware’s Competition, 117 HARV. L. REV. 588, 600 (2003) (asserting that Delaware’s race is not with other states, but with the risk of federal preemption). 2003] CONTROLLING CONTROLLING SHAREHOLDERS 797 Signal Companies determined to acquire the 49.5% of UOP that it did not own through a merger in which the UOP shareholders would receive cash for their UOP stock.32 Although Signal was prepared to pay up to $24 per share for the stock, the UOP board agreed to accept $21 per share, an approximately fifty percent premium over the market price of UOP stock.33 The court treated the freeze-out transaction as a simple manifestation of the core self-dealing conduct that requires intensive judicial review of the transaction terms for fairness.34 Because the lower the price Signal paid to UOP shareholders, the better off it was, and because Signal had benefited by its receipt of a feasibility study prepared by Signal’s UOP directors to the detriment of the UOP minority, the transaction simply presented a variation of the typical scenario that triggers heightened review of operating transactions under Sinclair.35 Consistent with the general principle that a controlling shareholder is cut no slack in its dealings with a controlled corporation, the court stressed that Signal-designated directors of UOP should be held to the same standard as non-Signal directors; conflicting loyalties had to be resolved in favor of the controlled corporation. 36 Having established that a freeze-out triggered intensive judicial review of the transaction’s fairness, the court went on to delineate the terms of that review. “Fairness,” the court explained, consists of the process by which the transaction is negotiated: “fair dealing” together 32 Weinberger, 457 A.2d at 705. 33 Id. at 705-06. 34 Id. at 710. 35 “Self-dealing occurs when the parent, by virtue of its domination of the subsidiary, causes the subsidiary to act in such a way that the parent receives something from the subsidiary to the exclusion of, and detriment to, the minority stockholders of the subsidiary.” Sinclair Oil Corp. v. Levien, 280 A.2d 717, 720 (Del. 1971). 36 See Weinberger, 457 A.2d at 710 (relying on “the long-existing principle of Delaware law that these Signal designated directors on UOP’s board still owed UOP and its shareholders an uncompromising duty of loyalty”). In the post-Weinberger evolution of freeze-outs, the inherent tensions in a transaction proposed by a controller, who either has the necessary voting power to accomplish the transaction or, if the transaction is conditioned on a majority of minority approval, will remain in control even if the minority refuses, have led to the imposition of entire fairness review in all such freezeouts; no explicit taking advantage of minority shareholders is required. See Kahn v. Lynch Communications Sys., Inc., 638 A.2d 1110, 1115 (Del. 1994) (requiring entire fairness review when a controlling or dominating shareholder stands on both sides of the transaction); Rosenblatt v. Getty Oil Co., 493 A.2d 929, 937 (Del. 1985) (“‘[T]he requirement of fairness is unflinching . . . where one stands on both sides of a transaction, he has the burden of establishing its entire fairness, sufficient to pass the test of careful scrutiny by the courts.’” (quoting Weinberger, 457 A.2d at 710)). 798 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 with “fair price.”37 With respect to fair dealing, the court stressed both the obligation of candor on the part of the parent38 and the importance of a process that mirrors a real arm’s-length transaction in which each party has the right to say “no.”39 As to fair price, the court adopted for this purpose the liberalized appraisal standard previously adopted by the Delaware legislature.40 Unfortunately, the court provided no real guidance as to how the two elements of fairness interacted. On the one hand, Weinberger can be read as suggesting that, if the parent allowed the subsidiary to establish an independent negotiating committee that had the right to say “no,” the court could then infer that the price resulting from arm’s-length bargaining was also fair. Alternatively, however, the court simultaneously and unhelpfully stressed that “the test for fairness is not a bifurcated one as between fair dealing and price. All aspects of the issue must be examined as a whole since the question is one of entire fairness.”41 The importance of this confusion cannot be overemphasized. For this purpose, it is important to keep in mind what is at stake. Controlled corporation shareholders already have a remedy if they believe the price to be paid in a cash-out merger is too low: an appraisal proceeding with precisely the same measure of value as that adopted by the Weinberger court.42 The difference between the two remedies is technically procedural, but ultimately of enormous substantive consequence. Under the Delaware appraisal procedure, a shareholder must jump through a number of procedural hoops, including not 37 Weinberger, 457 A.2d at 711. 38 Id. 39 In an oft-cited footnote, the court stated that the result here could have been entirely different if UOP had appointed an independent negotiating committee of its outside directors to deal with Signal at arm’s length. Since fairness in this context can be equated to conduct by a theoretical, wholly independent, board of directors acting upon the matter before them, it is unfortunate that this course apparently was neither considered nor pursued. Particularly in a parent-subsidiary context, a showing that the action taken was as though each of the contending parties had in fact exerted its bargaining power against the other at arm’s length is strong evidence that the transaction meets the test of fairness. Id. at 709 n.7 (citations omitted). 40 Id. at 703-04; see also DEL. CODE ANN. tit. 8, § 262 (Supp. 1982) (current version at tit. 8, § 262 (2001)). 41 Weinberger, 457 A.2d at 711. The court’s reasoning is unclear. Suppose the price is entirely fair, but the process is faulty. To what else are shareholders entitled beyond a fair price? 42 Id. at 703-04. 2003] CONTROLLING CONTROLLING SHAREHOLDERS 799 voting for the transaction and not accepting payment, in order to retain the right to bring an appraisal action.43 More importantly, the Delaware corporate statute does not authorize a class appraisal procedure. 44 In contrast, a breach of fiduciary duty claim can be brought on behalf of all subsidiary shareholders regardless of how they voted or whether they accepted payment for their shares.45 Thus, the economics of the litigation process mean that, if a fight about price is limited to appraisal, the controlling shareholder is exposed as to price only with respect to the number of shares for which appraisal rights are perfected, typically a quite small number. Moreover, the controller can manage its potential risk by conditioning its obligation to close the merger on a certain level of shareholder approval. In a class action under the Weinberger standard, however, the price exposure extends to all shares acquired through the freeze-out merger without the need for shareholders to take any action at all.46 Finally, if the freeze-out merger consideration is stock in the controller or stock in any publicly traded corporation, the minority shareholders have no right to appraisal.47 Thus, without a cause of action for breach of fiduciary duty, the minority shareholders in such a transaction may have no remedy at all. What remained open after Weinberger, then, was the procedural key. If the parent adopts an arm’s-length negotiating structure, including an independent negotiating committee with a right to say “no,” and receives the approval of a majority of the minority shareholders, does the standard of review shift to business judgment and therefore relegate shareholders to their appraisal rights as the 43 § 262(a). 44 Id. 45 See Kahn v. Lynch Communication Sys., Inc., 638 A.2d 1110, 1111 (Del. 1994) (noting that the plaintiff brought a class action on behalf of all shareholders of the acquired company whose stock had been procured through the merger). 46 See GILSON & BLACK, supra note 1, at 1266-69 (discussing the differences between appraisal actions and other claims). In Andra v. Blount, 772 A.2d 183, 183-84 (Del. Ch. 2000), Vice Chancellor Strine confronted the critical procedural consequence of a plaintiff’s successfully invoking entire fairness review of a freeze-out merger in the context of applying standing as a barrier to entire fairness review. See also Clements v. Rogers, 790 A.2d 1222, 1225 (Del. Ch. 2001) (confronting a similar procedural consequence with respect to the acquiescence doctrine). 47 § 262(b)(2). If, however, the controller owns at least ninety percent of the target’s stock and uses the short-form merger procedure under DEL. CODE ANN. tit. 8, § 253 (2001), then the minority shareholders have appraisal rights irrespective of the consideration. § 262(b)(3). 800 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 Weinberger court suggested in footnote 7?48 Alternatively, would the appraisal measure of value nonetheless be applied on a class basis because, as the Weinberger court also explained, “the test for fairness is not a bifurcated one as between fair dealing and price”?49 This and related issues were more or less clearly worked out in two Delaware Supreme Court opinions, Kahn v. Lynch Communications Systems, Inc. (Kahn I )50 and Kahn v. Lynch Communications Systems, Inc. (Kahn II ),51 involving Alcatel U.S.A. Corporation’s freeze-out of non-controlling shareholders in Lynch Communication Systems, Inc. Kahn I plainly resolved the issues at stake in structuring the approval process of a freeze-out merger. The court considered a perfectly sensible argument that entire fairness review should not apply, and therefore shareholders would be remitted to an appraisal remedy, if the negotiating structure plausibly protected their interests52—as, for example, where the merger terms met the approval of a fully empowered, independent negotiating committee and the merger was conditioned upon approval by the majority of the disinterested minority.53 Where the procedure approximated an arm’s-length negotiation, no special judicial review would be appropriate, and the business judgment standard would apply. Furthermore, it would follow that the frozen-out shareholders would be held to their decision regarding the pursuit of appraisal. Instead, the Kahn I court held that adopting such a negotiating structure served only to shift the burden of proof to the plaintiff on the issue of the freeze-out’s fairness. 54 The court believed that the controlling shareholder retained the capacity to influence the minority that cannot be procedurally 48 457 A.2d at 709 n.7. See supra note 39 for the footnote quotation. 49 457 A.2d at 711. 50 638 A.2d at 1117. 51 669 A.2d 79, 84 (Del. 1995). 52 See Kahn I, 638 A.2d at 1115 (discussing the possibility that approval of a cashout merger by a committee of interested directors “renders the business judgment rule the applicable standard of judicial review”). 53 In re Trans World Airlines, Inc. S’holders Litig., No. 9844, 1988 WL 111271, at *7 (Del. Ch. Oct. 21, 1988): Both the device of the special negotiating committee of disinterested directors and the device of a merger provision requiring approval by a majority of disinterested shareholders, when properly employed, have the judicial effect of making the substantive law aspect of the business judgment rule applicable and, procedurally, of shifting back to plaintiffs the burden of demonstrating that such a transaction infringes upon rights of minority shareholders. 54 638 A.2d at 1117. 2003] CONTROLLING CONTROLLING SHAREHOLDERS 801 dissipated.55 In effect, the court envisioned an implicit threat that, if the non-controlling shareholders did not approve the freeze-out, the controlling shareholder would exercise its operating discretion to their disadvantage.56 In Kahn I itself, the court found that Alcatel had coerced the independent negotiating committee set up by Lynch Communications by threatening to proceed with a tender offer at a lower price if the committee continued to resist.57 The court remanded the case to the chancery court to determine the transaction’s entire fairness.58 Kahn I left open two important issues. First, what happens if the transaction structure fails this initial fair-dealing inquiry and therefore does not operate to shift the burden of proof? If a transaction has to exhibit both fair dealing and fair price to be entirely fair, then how can the fairness standard ever ultimately be satisfied if, as in Kahn I, the fair-dealing component is not met? Second, why should a controlling shareholder allow the creation of a fully empowered negotiating committee if all it gets in return is a burden shift? Unless the evaluation of price is somehow different—even without the presumptions of business judgment review—as a result of procedural protections, what is in it for the controlling shareholder?59 On remand, the chancery court found that the transaction satisfied both the fair-dealing and fair-price components of the entire fairness review.60 As has been suggested, finding that the fair-dealing component was satisfied, despite the controlling shareholder’s coercion of the independent negotiating committee, required some fasttalking. On appeal, the supreme court’s assessment of fair dealing took an unacknowledged but major shift. While in Kahn I the inquiry 55 Id. at 1116-17. 56 “‘The controlling stockholder relationship has the [sic] potential to influence, however subtly, the vote of [ratifying] minority stockholders in a manner that is not likely to occur in a transaction with a noncontrolling party.’” Id. at 1116 (quoting Citron v. E.I. DuPont de Nemours & Co., 584 A.2d 490, 502 (Del. Ch. 1990)). In making this statement, the court appears unaware that this “inherent coercion” can exist only to the extent that judicial review of the controlling shareholder’s operating decisions fails to control private benefit extraction. 57 See id. at 1118 (stating that the independent negotiating committee had “full knowledge of Alcatel’s demonstrated pattern of domination” while considering Alcatel’s proposal to purchase Lynch Communications). 58 Id. at 1121-22. 59 See In re Cysive, Inc. S’holders Litig., No. 20341, 2003 WL 21961453, at *15-16 (Del. Ch. Aug. 15, 2003) (concluding that burden-shifting, particularly on “fair value,” is not generally material either at the pleading stage or at trial). 60 Kahn II, 669 A.2d at 83. 802 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 was whether the independent negotiating committee had been coerced, in Kahn II the inquiry shifted to whether the non-controlling shareholders voting on the freeze-out merger were coerced.61 Despite the finding that “the specter of coercion” had impaired the functioning of the independent negotiating committee, the court concluded that “[w]here other economic forces are at work and more likely produced the decision to sell,” this coercion still may not be deemed material with respect to the transaction as a whole, and will not prevent a finding of entire fairness. In this case, no shareholder was treated differently . . . nor subjected to a two-tiered or squeeze-out treatment. . . . Clearly there was no coercion exerted which was material to this aspect of the transaction . . . .62 Putting Kahn I and Kahn II together, we are left with something like a two-tiered inquiry concerning the fair-dealing component of the entire fairness standard. With respect to whether the burden of proof on entire fairness has shifted to the plaintiff, the appropriate inquiry assesses the presence and true empowerment of an independent negotiating committee.63 Fairly read, Kahn I holds that the burden of proof does not shift unless the independent negotiating committee has the right to prevent the transaction.64 With respect to the ultimate determination of whether the transactional procedure satisfies the fair-dealing component, the inquiry shifts to whether the inherent coercion and the form of the transaction actually influenced the noncontrolling shareholders’ votes. Characterized somewhat less than sympathetically, is fair dealing satisfied despite an unfair, but not structurally coercive, procedure?65 That leaves the issue of the stakes associated with establishing an empowered special committee. Will the assessment of fair price be influenced by the extent to which the transaction structure attempts 61 See id. at 86 (holding that the coercion found by the court in Kahn I did not have a “material” influence on the shareholders’ decision to sell and, therefore, was not indicative of unfair dealing). 62 Id. (citation omitted). 63 How to design a fully empowered, independent negotiating committee is itself an interesting issue. See GILSON & BLACK, supra note 1, at 1303-05 (speculating as to what negotiating procedures will satisfy the fair-dealing requirement after Kahn I ). 64 See supra text accompanying note 54 (describing the court’s holding that an arm’s-length negotiating structure shifts the burden of proof). 65 By contrast, a violation of the duty of candor does appear to result in a per se failure of the entire fairness standard. See Ince & Co. v. Silgan Corp., No. 10941, 1991 WL 17171, at *5 (Del. Ch. Feb. 7, 1991) (stating that “entire fairness includes the obligation ‘to disclose with entire candor all material facts concerning the merger’” (quoting Sealy Mattress Co. v. Sealy, Inc., 532 A.2d 1324, 1335 (Del. Ch. 1987))). 2003] CONTROLLING CONTROLLING SHAREHOLDERS 803 to dissipate the specter of coercion? The answer to that question remains opaque, largely because of the court’s continued insistence on the “non-bifurcated standard of Weinberger.”66 There is, however, a specter of judicial coercion with respect to the link between procedure and price. Although Weinberger eliminated the free option that arose out of Lynch v. Vickers Energy Corp.,67 which held that failing the entire fairness standard exposed the controlling shareholder to the equitable remedy of the monetary equivalent of rescission,68 and the court in the seemingly endless Cede & Co. v. Technicolor, Inc.69 litigation read the appraisal standard in section 262 to include significant elements of post-transaction value,70 Weinberger continued to hold out the prospect of equitable relief beyond the appraisal standard.71 Thus, the court has left room for a link between procedure and damages, with an appropriate incentive effect. D. Summary The doctrinal origami of the limits on controlling shareholders presents a clear but complex pattern. The rule governing the extraction of private benefits of control limits large wealth transfers from non-controlling to controlling shareholders by imposing rigorous judicial review of self-dealing transactions between the controlling shareholder and the controlled company, while still leaving room for a range of private benefits that may be more beneficial to the controlling shareholder than costly to the controlled company and that 66 Kahn II, 669 A.2d at 90. The court’s reference to “a disciplined balancing approach” following its mention of the non-bifurcated standard in Kahn II, id., appears to contemplate an undisciplined tradeoff between procedure and price that seems to assure the continued pattern of fully litigating every freeze-out transaction. 67 429 A.2d 497 (Del. 1981) (awarding the plaintiff rescissory damages for the controlling shareholder’s breach of fiduciary duty in a tender offer, rather than the appraisal damages typically awarded to minority shareholders injured in sales to controlling parties); see also GILSON & BLACK, supra note 1, at 1269 (discussing the option-like effect of differing damage standards in appraisal and entire fairness proceedings). 68 429 A.2d at 501; see also Weinberger, 457 A.2d 701, 703-04, 714 (Del. 1983) (overruling Lynch by holding that appraisal, and not rescissory damages, is the appropriate remedy for minority shareholders in freeze-out mergers). 69 684 A.2d 289 (Del. 1996). 70 Id. at 299-300. 71 457 A.2d at 714 (“While a plaintiff’s monetary remedy ordinarily should be confined to the more liberalized appraisal proceeding herein established, we do not intend any limitation on the historic powers of the Chancellor to grant such other relief as the facts of a particular case may dictate.”). 804 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 may support the more focused monitoring of the managerial agency problem available to a controlling shareholder. In turn, the rule governing the extraction of the capitalized value of the private benefits from operations through the sale of control is both simple and permissive. Because of the restrictions on the extraction of private benefits from operations (which continue to allow a level of private benefits consistent with focused monitoring), an acquirer of control must ordinarily improve the performance of the controlled corporation in order to profit from its investment. Achieving this improvement requires two inputs: the capabilities of the new controlling shareholder and the existing business of the controlled corporation. Because the non-controlling shareholders remain participants in the controlled corporation, the gain that results from this bilateral monopoly is shared more or less proportionately. Judicial intervention is limited to those circumstances where either there is an observable risk that the purchaser of control will exceed the level of allowed private benefits from operation or there has been fraud in the interaction between the selling controlling shareholder and noncontrolling shareholders in anticipation of the control sale. In contrast, the rules that limit extracting the capitalized value of the private benefits of control through freezing out the noncontrolling shareholders are both complex and restrictive. This difference emerges because, unlike with a sale of control, noncontrolling shareholders will not automatically participate in any value increase as a result of the freeze-out. This discrepancy results in an incentive for controlling shareholders to manipulate the operation of the controlled corporation and the market price of its stock in anticipation of the transaction, subject to the limits imposed by the Sinclair standard.72 It also can leave the non-controlling shareholders with no benefit from the post-transaction increase in value even though an input in which they have an interest is necessary to achieve that increase. Requiring an independent negotiating committee and more rigorous judicial review serves to ensure that the non-controlling shareholders receive some portion of the gain that would result from bargaining in a bilateral monopoly.73 72 Sinclair, 280 A.2d 717, 722 (Del. 1971), would not restrict poor decisions that reduce value generally. To the extent that the effects of such decisions may not be reversible, the potential for manipulation is real. 73 There is no obvious reason to believe that giving all the gain to one side or another in a bilateral monopoly is necessary in order to achieve an efficient level of transactions. From the perspective of either participant, any value above the 2003] CONTROLLING CONTROLLING SHAREHOLDERS 805 II. DISTURBING THE SYMMETRY: THE DIGEX AND SILICONIX LINES OF CASES To this point, we have argued that the Delaware doctrine seeking to control the level of private benefits enjoyed by controlling shareholders reflects a sensible symmetry between the three alternative methods by which these benefits can be extracted: through ongoing operations, by a sale of control, or by a freeze-out. As our discussion of the case law reflects, we do not assert that this symmetry is the result of grand design. Rather, we believe only that, when rules governing one or another alternative get out of line, transaction planners are quick to adjust their strategies to compensate, such that the Delaware Chancery Court sees the implications of its previous decisions quickly and is promptly given the opportunity to adjust the rules and restore balance.74 We also do not assert that the pattern necessarily reflects a fully efficient equilibrium that can be reflected in complex equations. Rather, we suggest only that the pattern reflects a rough but workable solution, not necessarily any worse than the results of an effort to mathematically model the solution to three simultaneous equations under restrictive assumptions. Our recognition of the overall structure’s viability should not, however, imply that we believe the Delaware courts always get it reservation price is a rent. Lucian Bebchuk and Alan Schwartz discuss this issue as it pertains to the takeover context in an interesting, albeit lengthy, debate. For examples of their arguments, see Lucian Arye Bebchuk, The Case for Facilitating Competing Tender Offers: A Last (?) Reply, 2 J.L. ECON. & ORG. 253, 256 (1986); Lucian Arye Bebchuk, The Sole Owner Standard for Takeover Policy, 17 J. LEGAL STUD. 197, 228 (1988); Alan Schwartz, Bebchuk on Minimum Offer Periods, 2 J.L. ECON. & ORG. 271, 271 (1986); Alan Schwartz, Search Theory and the Tender Offer Auction, 2 J.L. ECON. & ORG. 229, 244 (1986); Alan Schwartz, The Sole Owner Standard Reviewed, 17 J. LEGAL STUD. 231, 234 (1988). 74 Ronald Gilson and Bernard Black describe “[t]his drastic telescoping of the common law process” with respect to takeovers in the 1980s: “Each new decision was reflected in the tactics of the next transaction; the Chancery Court often had to confront the ‘next case’ on a motion for preliminary injunction soon after the initial decision.” GILSON & BLACK, supra note 1, at 4. We do not intend this analysis as a strong claim for the efficiency of the common law of corporations. We do, however, think that the claim for efficiency is likely the strongest here, where the rules concern an ongoing pattern of transactions and where professionals view their role as involving continual adjustment of transactional structures to reflect new judicial decisions. This explicit interaction between case law and transaction structure is plainly visible in the latter line of cases we discuss in this Part. See also John C. Coates IV & Guhan Subramanian, A Buy-Side Model of M&A Lockups: Theory and Evidence, 53 STAN. L. REV. 307, 328-37 (2000) (documenting the rapid shift from stock options to breakup fees in friendly mergers following an adverse Delaware Supreme Court ruling on stock options). 806 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 152: 785 right. These are complicated and difficult matters with, as we have tried to show, a lot of moving parts. Moreover, the case law does not acknowledge the simultaneity of the three doctrinal lines, which makes maintaining their symmetry that much harder. Thus, mistakes happen. The role of commentary is to identify these glitches and offer suggestions as to how they can be rectified.75 In this Part, we focus on what we believe to be two such glitches that go to the center of the symmetry developed in Part I. The first, the chancery court decision of In re Digex, Inc. Shareholders Litigation,76 deals with the rules governing private benefit extraction through the sale of control, and the second, the line of cases following In re Siliconix Inc. Shareholders Litigation,77 deals with the rules governing private benefit extraction through freeze-outs. Recognition of the relationship between the three doctrinal areas that control controlling shareholders’ extraction of private benefits suggests that, in each, the chancery court is moving in the wrong direction. More particularly, Digex threatens to interfere with the controller’s right to retain a control premium in the sale of control, and the Siliconix line of cases threatens to reduce minority protections in freeze-out transactions in a way that will enhance the controller’s take beyond the permitted level of private benefits. A. Digex: New Restrictions on Sale of Control at a Premium The controversy in Digex now seems terribly dated. The transaction began with a contest between WorldCom, Inc. and Global Crossing, Ltd. to acquire Intermedia Communications, Inc., a telecommunications company, and/or Digex, Inc., Intermedia’s controlled subsidiary in the web hosting business, said to be “well positioned in one of the hottest segments of the technology sector . . . .”78 World- Com won the contest, having made clear “that WorldCom would outbid anyone for Digex.”79 The legal issues were posed by the conflict between Intermedia and Digex shareholders over which group would receive the Wo