ISSN 1936-5349 (print) ISSN 1936-5357 (online) HARVARD JOHN M. OLIN CENTER FOR LAW, ECONOMICS, AND BUSINESS THE CASE FOR AN UNBIASED TAKEOVER LAW (WITH AN APPLICATION TO THE EUROPEAN UNION) Luca Enriques Ronald J. Gilson Alessio M. Pacces Discussion Paper No. 744 05/2013 Harvard Law School Cambridge, MA 02138 This paper can be downloaded without charge from: The Harvard John M. Olin Discussion Paper Series: http://www.law.harvard.edu/programs/olin_center/ This paper is also a discussion paper of the John M. Olin Center’s Program on Corporate Governance. The Case for an Unbiased Takeover Law (with an Application to the European Union) Luca Enriques () ∗ Ronald J. Gilson (**) Alessio M. Pacces (***) Final Draft May 2013 Abstract: Takeover regulation should neither hamper nor promote takeovers, but instead allow individual companies to decide the contestability of their control. Based on this premise, we advocate a takeover law exclusively made of default and menu rules supporting an effective choice of the takeover regime at the company level. For reasons of political economy bearing on the reform process, we argue that different default rules should apply to newly public companies and companies that are already public when the new regime is introduced. The first group should be governed by default rules crafted against the interest of management and of controlling shareholders, because these are more efficient on average and/or easier to opt out of when they are or become inefficient for the particular company. The second set of companies should instead be governed by default rules matching the status quo even if this favors the incumbents. This regulatory dualism strategy is intended to overcome the resistance of vested interests towards efficient regulatory change. Appropriate menu rules should be available to both groups of companies in order to ease opt-out of unfit defaults. Finally, we argue that European takeover law should be reshaped along these lines. Particularly, the board neutrality rule and the mandatory bid rule should become defaults that only individual companies, rather than member states, can opt out of. The overhauled Takeover Directive should also include menu rules, for instance a poison pill defense and a time-based breakthrough rule. Existing companies would continue to be governed by the status quo until incumbents decide to opt into the new regime. () Harvard Law School, LUISS University, and ECGI. ∗ (**) Columbia Law School, Stanford Law School, and ECGI. (***) Erasmus University Rotterdam (School of Law) and ECGI. We wish to thank Jesse Fried, José Maria Garrido, Henry Hansmann, Klaus Hopt, David Kershaw, Jon Macey, Joe McCahery, Roberta Romano, Jaap Winter, and participants in the conference “European Takeover Law – The Way Forward” at the University of Oxford Faculty of Law, in the conference on “Fiduciary Duties in Corporate Law” at Tilburg University, and in seminars at Harvard, University of Pennsylvania, Vanderbilt, and Yale Law School for their very helpful comments on an earlier draft. Usual disclaimers apply. 1 1. Introduction Takeovers remain the most controversial corporate governance mechanism. Since the early 1960s, the debate has been over whether regulation should promote or impede attempts to acquire control of a corporation by making an unsolicited offer directly to shareholders. The dogged persistence of the debate reflects the internal logic of two conflicting positions.1 According to pro-takeover commentators, takeovers are generally beneficial for corporate governance. Takeovers can displace poorly performing managers.2 Ex ante, the threat of displacement encourages better performance. When management is unable to see past existing industry patterns and recognize the opportunity for strategic change, hostile takeovers, possibly initiated by bidders from outside the industry, can facilitate corporate restructuring.3 From this perspective, regulation should encourage takeovers by restricting the ability of managers to block a takeover bid, leaving to shareholders comprised increasingly of sophisticated institutional investors4 the decision whether the bid will succeed.5 Other observers have a darker view. Those who oppose hostile takeovers argue that they can disrupt well-functioning companies6 and encourage short-termism 1 The stylized presentation that follows of the two standard conflicting views cannot reflect the richness of the debate and the variety of nuanced positions scholars have taken therein. See e.g. Marcel Kahan & Edward B. Rock, Corporate Constitutionalism: Antitakeover Charter Provisions As Precommitment, 152 U. PA. L. REV. 473, 480-89 (2003). For a survey of the economic literature on takeovers see e.g. Marco Becht, Patrick Bolton, & Ailsa Röell, Corporate Law and Governance, in HANDBOOK OF LAW AND ECONOMICS 833, 848-853 & 878-886 (A. Mitchell Polinsky and Steven Shavell eds., 2007). 2 See e.g. FRANK H. EASTERBROOK & DANIEL R. FISCHEL, THE ECONOMIC STRUCTURE OF CORPORATE LAW, 162-174 (1991). 3 See e.g. Ronald J. Gilson, The Political Ecology of Takeovers: Thoughts on Harmonizing the European Corporate Governance Environment, 61 FORDHAM L. REV. 161 (1992); Klaus J. Hopt, Obstacles to corporate restructuring: observations from a European and a German perspective, in PERSPECTIVES IN COMPANY LAW AND FINANCIAL REGULATION – ESSAYS IN HONOUR OF EDDY WYMEERSCH, 373 (Michael Tison et al. eds., 2009). 4 By the late 2000s, for example, institutional investors held over 70 percent of the outstanding shares of the 1000 largest U.S. public corporations and the ten largest institutions owned more than 25 percent of the outstanding shares in many large public corporations. See Ronald J. Gilson & Jeffrey Gordon, The Agency Costs of Agency Capitalism (working paper, Oct. 2012) (forthcoming COLUM. L. REV. 2013), available at http://ssrn.com/abstract=2206391. 5 Some commentators, including one of us, have taken a more moderate position, arguing in favour of allowing directors to expend corporate resources either to persuade shareholders that rejecting a hostile bid would create more value than offered by the bidder, or to provide shareholders a more valuable alternative. See Ronald J. Gilson, Seeking Competitive Bids Versus Pure Passivity in Tender Offer Defense, 35 STAN. L. REV. 51 (1982); Lucian A. Bebchuk, The Case for Facilitating Competing Tender Offers: A Reply and Extension, 35 STAN. L. REV. 23 (1982); Lucian A. Bebchuk, Toward Undistorted Choice and Equal Treatment in Corporate Takeovers, 98 HARV. L. REV. 1693 (1985). 6 See e.g. Martin Lipton, Corporate Governance in the Age of Finance Corporatism, 136 U. PA. L. REV. 1, 18-20 (1987). Still others share the negative view of hostile takeovers but without the pro- 2 as opposed to long-term commitments to shareholder value. From this perspective, managers have better information than shareholders, who will apply a myopically high discount rate to managers’ long-term plans.7 In this article, we reject a categorical pro- or anti-takeover position, instead taking what we will call an “unbiased” stance on the desirability of takeovers. While hostile and friendly takeovers may be efficient in the aggregate,8 individual takeovers and individual companies’ exposure thereto are efficient or inefficient depending on a variety of factors. These factors include the production functions of companies, conditions in the relevant industry, the problems confronting the corporation and the best response to those problems. Because these all may differ from company to company and over time, so also may the appropriate stance to takeovers differ. Consequently, we posit that takeover regulation should sanction individual company efforts to devise a takeover regime appropriate to their own, mutable circumstances. Put differently, we propose a kind of horizontal subsidiarity approach.9 Regulation of takeovers should defer to choices made at the level best suited to a nuanced assessment of particular circumstances and industries: that of the individual (target) company. From this perspective, takeover regulation’s central role is to set the management underpinning. They see managers as disloyal agents who will protect their positions one way or another. If they cannot block a hostile takeover directly, they will block it indirectly by taking substantive actions that will make the corporation less attractive to hostile bidders at a cost to shareholders that exceeds that of blocking a hostile offer. See Jennifer Arlen & Eric Talley, Unregulatable Defenses and the Perils of Shareholder Choice, 152 U. PA. L. REV. 577 (2003). Because regulation cannot effectively constrain self-interested managers, the second best solution is to reduce their incentive to do their worst. Id. These commentators thus favour restricting hostile takeovers, whether by allowing the target board of directors to prevent shareholders from accepting them or by making bids costlier via regulation. 7 For discussions of the argument that public company management overly discounts the future, see, e.g., Michael L. Wachter, Takeover Defense When Financial Markets Are (Only) Relatively Efficient, 151 U. PA. L. REV. 787 (2003). On the claim that stock markets are short-termist, see more recently THE KAY REVIEW OF UK EQUITY MARKETS AND LONG-TERM DECISION MAKING (July 2012), available at http://www.bis.gov.uk/kayreview. 8 See Sandra Betton et al., Corporate Takeovers, in HANDBOOK OF EMPIRICAL CORPORATE FINANCE 291, 356-72 (B. Espen Eckbo ed., 2d ed. 2008) (arguing that the combined returns of target and acquiring companies are positive on average). But see also Klaus Gugler et al., Market Optimism and Merger Waves, 33 MANAGE. DECIS. ECON. 159 (2012) (showing that the long-term wealth effects of takeovers are negative on average). 9 In the European Union jargon, horizontal subsidiarity refers to the preference for self-regulation by private actors over (European or national) regulation from public bodies, as expressed in the CONCLUSIONS OF THE PRESIDENCY, Edinburgh, 11-12 December 1992, Annex 1 to Part A, at 21. See e.g. Edward Best, Alternative Methods and EU Policy-Making: What Does “Co-Regulation” Really Mean?, 2008 EIPASCOPE, Issue 2, 11, 12. To be sure, no one has so far used the term to refer to the relationship between private ordering and mandatory provisions in the context of company law. Vertical subsidiarity (or subsidiarity tout court), as enshrined in Article 5(3) of the Treaty on European Union, refers instead to a preference for national or local legislation in matters in which the European has no exclusive competence. See e.g. Kees Van Kesbergen & Bertjan Verbeek, The Politics of Subsidiarity in the European Union, 32 J. COMMON MKT. STUD. 215 (1994). 3 default rules governing the available responses to a takeover. These rules will be subject to opt-out by individual companies. The likelihood of opt-out will depend on the rules governing it and on whether the constituency that such rules empower – be that management, controlling or minority shareholders – welcomes the change. Accordingly, our policy goal here is to determine the default rules and to address the means by which the default rules can be altered in such a way that choice or inertia at the individual company level are not biased either in favour or against takeovers. Our unbiased approach reflects an effort to move the debate beyond categorical positions in which mandatory rules are insensitive to context. Takeovers are one of a range of governance mechanisms the optimal mix of which, for a particular company, should be endogenous to a significant extent. The conditions a corporation confronts can be expected to change over time, thus reinforcing the endogeneity of the “right” mix of governance techniques rather than a mandatory set specified at any one point in time. And while we make recommendations about the correct default rules, these are made with careful attention to their structure, the method by which they can be changed, and the realities of political economy surrounding takeover regulation. In the debate over takeovers, the pro- and anti-takeover positions are typically framed as a simple question: who should decide whether a hostile takeover goes forward? There are three obvious candidates. The board may be given the power to block an offer. Alternatively, the board may be restricted from acting to hinder a takeover, thereby allocating the decision to shareholders. In this vein, the U.K. City Code on Takeovers and Mergers and English company law more broadly disarm the board by prohibiting actions that prevent shareholders from deciding on a bid.10 Third, the final decision could be left to the courts, applying a standard taking context into account when determining whether the board or the shareholders should decide – in effect, allocating the real authority to a court. Over some of the last 25 years, this has been the Delaware approach.11 10 The City Code restricts the board’s ability to adopt defensive tactics that would frustrate an unwelcome bid, unless shareholders authorize it. Rule 21 of The City Code on Takeover and Merger (The Panel on Takeover and Mergers, 9 ed. 2009). On board neutrality as a more general feature of English company law see David Kershaw, The Illusion of Importance: Reconsidering the UK’s Takeover Defence Prohibition, 56 INT’L & COMP. L.Q. 267 (2007). 11 See Ronald J. Gilson, Unocal Fifteen Years Later (and What We Can Do About It), 26 DEL. J. CORP. L. 491 (2001). 4 We argue that individual companies should be able to decide “who decides.” Deferring to the choices of individual companies, however, implies more than mere advocacy of freedom of contract in takeover law. In a world of positive transaction costs, the selection of the default rules matters greatly as do the procedures by which they can be changed.12 The crucial question for an unbiased takeover law thus becomes the choice of the default rules – the rules that apply unless companies provide otherwise in their governing documents, whether initially or as later amended. Selecting the right default rules based on the conditions for opting out at different points in time facilitates efficient, as opposed to inertial or opportunistic, choices at the company level. Default rules matter both for newly public companies (hereinafter, “IPO companies”) and for those companies already public when a new default regime is introduced (hereinafter “installed-base companies”). In the former, setting the default rules that suit the majority of IPO companies saves transaction costs. As we argue below, this generally speaks in favour of defaults that do not restrict takeovers. More importantly, while a company’s efficient exposure to takeovers may vary across time, defaults tend to be sticky.13 Takeover-unrestrictive defaults are easier to change if they turn out to be inefficient down the road. From an economic perspective, takeover-unrestrictive defaults would be also generally preferable for installed-base companies. However, the matter stands differently from a political economy perspective. In fact, the introduction of takeover-unrestrictive default rules is bound to be opposed by those having an interest in existing takeover-restrictive default rules; similarly, those having an interest in existing mandatory rules will likely resist their becoming defaults. These vested interests may successfully oppose regulatory change, thereby depriving IPO companies of the opportunity to benefit from the establishment of an unbiased takeover law. To address this problem, we suggest to reform takeover law based on a model of regulatory dualism.14 Regulatory dualism implies that reform makes two regimes available. The new, unbiased defaults will apply to IPO 12 See Ian Ayres, Regulating Opt-Out: An Economic Theory of Altering Rules, 121 YALE L.J. 2032 (2012). 13 Default rules are ‘sticky’, particularly in corporate law. See Henry Hansmann, Corporation and Contract, 8 AM. LAW ECON. REV. 1 (2006). 14 See generally Ronald J. Gilson, Henry Hansmann & Mariana Pargendler, Regulatory Dualism as a Development Strategy: Corporate Reform in Brazil, the United States, and the European Union, 63 STAN. L. REV. 475 (2011). 5 companies. Installed-base companies will instead remain subject to the existing regime, unless those having endowments therein agree to opt out of it. This solution seeks to mute incumbent opposition to regulatory change by conferring upon them the right to veto a move away from the status quo at the company level. However, such a move is still possible to the extent that parties are willing to contract upon their endowments.15 To illustrate how an unbiased regime would work, we contrast our approach with the current European Union (EU) takeover regime whose main rules are included in the Takeover Bid Directive (hereinafter the Directive).16 We start with the core policy choice on whether the board or the shareholders should make the final decision on a hostile bid. We argue that, differently from the current regime in all EU member states but Italy, the rule should be that shareholders decide unless IPO companies choose otherwise. We then extend the unbiased approach to the provisions that are more important in the presence of a controlling shareholder, where takeovers are typically friendly. In this context we argue that the rules mainly affecting the takeover decision should be pro-minority shareholders by default, not because we believe that minority shareholders will more often be in need of protection, but because at the IPO stage they can be persuaded to give up protection only if this is efficient. Two prominent applications of this reasoning are a one-share-one-vote default (hereinafter 1S1V) and a rule conferring upon minority shareholders the right to sell their shares at the same price as the outgoing controller. The latter provision, which is takeover restrictive, is reminiscent of the mandatory bid rule imposed by the Directive. In the unbiased regime we advocate, the mandatory bid rule should become a default rule. Finally, we contend that menu rules are an important complement of default rules in an unbiased takeover regime. The reason is two-fold. First, the existence of menu rules facilitates opt-out by those companies for which the default regime is inefficient.17 Second, menu rules established at the EU level would displace, if chosen, incompatible mandatory rules set at the member state level, thereby 15 Current examples of such contracting include the holders of a class of common stock with enhanced voting rights agreeing to give up their extra voting rights. For a description of a current effort by a Canadian public company to eliminate the heightened voting rights of one class of shares in a public company, see Bernard S. Black, Equity Decoupling and Empty Voting: The Telus Zero-Premium Share Swap, M&A LAW. 1, 4 (October 2012). 16 Directive 2004/25/EC on Takeover Bids of 21 April 2004, OJ L 142, 30.4.2004 12-23 (2004). 17 See infra text accompanying notes 58-61. 6 countering the residual opposition to the unbiased regime that can be expected from the vested interests despite regulatory dualism.18 The remainder of this article is structured as follows. Section 2 presents the economic case for an unbiased takeover law. Section 3 discusses the different criteria for choosing default rules for IPO companies and for installed-base companies. Section 4 applies this framework to the EU, illustrating how a review of the Directive could result in the introduction of an unbiased takeover regime. Section 5 concludes. 2. Economics of takeovers: the case for unbiased takeover regulation Takeover law should allow individual companies to decide their degree of takeover exposure. No single rule insensitive to context and strategy will perfectly distinguish between value-increasing and value-reducing takeovers. However, economic analysis counsels in favour of a default rule that assigns decision rights to shareholders. In this section we take up the basis for this default rule and how the default rule should be adjusted to address political economy issues. At the outset, a qualification is in order concerning how we differentiate value-creating from value-reducing takeovers. Whether a particular takeover is efficient depends on whether the winners’ gains exceed the losers’ losses net of transaction costs.19 Importantly, we ignore here the wealth effects of takeovers on non-shareholder constituencies (employees, customers, local communities, the national economy and the like – typically referred to unhelpfully in the aggregate as “stakeholders”) and focus only on the joint gain of the acquiring and the target company measured in terms of shareholder value. Our point is not that non-shareholder gains or losses are either irrelevant or necessarily of limited magnitude. Rather, the point is that takeovers are merely one way in which corporations respond to changes in economic conditions. Competition can force corporations to fire workers, lower wages, or close plants that are important to a community. Takeovers are in this respect just one mechanism though which competition operates and equilibration occurs. Serious issues of political economy frame how a particular country addresses competition’s effect on non-shareholders, 18 See infra text accompanying notes 104-105. 19 This is a Kaldor-Hicks measure of welfare: a move is efficient if those who benefit from it can compensate the losers and still be better off. John R. Hicks, The Foundations of Welfare Economics, 49 ECON. J. 696 (1939); Nicholas Kaldor, Welfare Propositions of Economics and Interpersonal Comparisons of Utility, ECON. J. 549 (1939). 7 including those resulting from (hostile) takeovers.20 For our purposes, the scope and the features of the safety net protecting individuals and communities against the effects of economic and regulatory change are only relevant to the takeover debate if takeover regulation is the best (or the only) protection tool available. Because we have not seen that position carefully presented in the takeover debate, our discussion of takeover regulation in the following does not further consider it. The central problem in crafting takeover rules is that decisions about whether a takeover goes forward are made ex ante based on probabilistic information, while the ultimate efficiency of the takeover is revealed only ex post as the passage of time reduces probability distributions to facts. A decision maker can approve a takeover that existing information suggests will be value reducing because: (1) she is mistaken – she misinterprets the existing facts and so overestimates the transaction’s ex post value; (2) she pursues personal benefits (such as empire building or the potential to loot); or (3) she believes that the market will mistakenly overvalue the transaction ex post, whether due to poor information or myopia. In turn, the decision maker can make ex ante mistakes in rejecting a value- increasing takeover for reasons that parallel those associated with mistaken acceptance. The decision maker may reject a takeover because: (1) she is mistaken – she misinterprets the existing facts and so underestimates the takeover’s ex post joint value; (2) she pursues personal benefits (e.g. by entrenching herself or by sustaining a bad strategy that increases the value of the management’s real option on control);21 (3) she is hyperopic, i.e. she mistakenly applies too low a discount rate to future revenues. Traditionally, takeover law has attempted to answer the “who decides” question on an aggregate basis. The idea is that if the law allocates the decision on whether to proceed with a certain takeover to the right decision maker, good transactions will go through and bad ones will be stopped more often than under alternative regimes (including the option of having no mandatory rule in place at all). The fundamental policy choice then becomes whether, with respect to takeovers of public companies with dispersed ownership, the ultimate decision to let a takeover bid 20 See generally VARIETIES OF CAPITALISM: THE INSTITUTIONAL FOUNDATIONS OF COMPARATIVE ADVANTAGE (Peter Hall & David Soskice eds., 2001). 21 In this formulation, management holds a call option on the benefits of control. The value of this option can be increased by extending the period over which the option can be exercised, even if the company’s continuing the current strategy is a negative net present value investment. 8 go forward is allocated to the board (subject or not to conditions that to some extent constrain the board’s discretion), which typically the incumbent management will push toward resisting the takeover, or to the shareholders, who normally would be limited to public information in assessing the merits of a proposed transaction.22 Similarly, with respect to blockholder-controlled companies, the question is whether controlling shareholders should be limited in their ability to enter into, or to block, a control transaction. These limits mainly include two sets of rules. On the one hand, there are rules aimed to protect minority shareholders from transactions that favour the dominant shareholder, e.g. the mandatory bid rule or the requirement that certain transactions be authorized by a majority of the minority shareholders. On the other hand, there are rules aimed to facilitate takeovers by constraining controlling shareholders’ ability to entrench themselves (for instance, a one-share-one-vote mandate) or to block an offer (for instance via the so-called breakthrough rule devised in the Directive23). The problem with this one-size-fits-all approach is that no single rule allows only value-increasing transactions to go through in either setting. Regardless of which party is accorded discretion with respect to a takeover – management, shareholders, a controlling shareholder or a court – the risk of error remains, whether because of mistake or of self-interest. From this perspective, the most that takeover law can accomplish is to set general screens that minimize the average impact of biases across the companies population. However, there is an alternative to general screens that apply the same rules to all companies and all transactions under all circumstances. Takeover regulation could allow an individual company to select the screen that will minimize the impact of its own decisionmaker’s biases. As we argue below, under certain conditions this approach may fare better than a general, one-size-fits-all screen mandated by regulation. 22 Delaware’s intermediate standard, announced in Unocal, held out the possibility of a third decision maker: the courts. Under the initial framing of this intermediate standard, the court independently would review any defensive action based on the presence of a threat and the proportionality of the response. See e.g. Ronald J. Gilson & Rainier Kraakman, Delaware’s Intermediate Standard for Defensive Tactics: Is There Substance to Proportionality Review, 44 BUS. LAW. 247 (1989). Over time the intermediate standard has largely collapsed into an allocation of decision making to the target board so long as the hostile bidder’s chance of success in a proxy fight to replace the board with members who will redeem the pill was not mathematically or practically impossible. Ronald J. Gilson, Unocal Twenty Years Later, supra note 11. 23 See infra text accompanying notes 120-122. 9