The Never-Ending Quest For Shareholder Rights

THE NEVER-ENDING QUEST FOR SHAREHOLDER RIGHTS: SPECIAL MEETINGS AND WRITTEN CONSENT ABSTRACT We study the evolution of shareholders’ rights to call special meetings and act by written consent from a functional and an empirical perspective. From a functional perspective, we show that these powers are most useful when they can be employed to gain a board majority and only useful if they can be employed at least to change the board composition. As a result, shareholders’ ability to act between annual meetings interacts with several other features of a firm’s governance structure: (1) the shareholders’ ability to remove directors without cause; (2) the shareholders’ ability to expand the board; (3) the shareholders’ ability to fill vacancies; and (4) the requirement of supermajority vote to do (1)-(3). From an empirical perspective, we construct a panel that follows firms in the S&P 500 index from 2005 to 2017 and hand-code for multiple features of their governance structure. We document a sizable increase in the number of firms that allow shareholders to act between annual meetings—but for a substantial fraction of the firms that allow shareholders to act between annual meetings, we show shareholders cannot use that power to gain a board majority or change the board composition. We also document that precatory shareholder proposals were key drivers of the evolution of the ability to act between annual meetings among our sample firms and study how proponents select which firms to target. We conclude that proponents follow a “pecking order” strategy: they first push firms to declassify their boards, then push firms to allow shareholders to call special meetings, and only then push firms to allow shareholders to act by written consent. However, we do not find evidence that proponents target those firms where a grant of ability to act between annual meetings would be most productive in light of the firm’s governance structure. Finally, we offer suggestive evidence that firms cannot appease shareholder proponents by caving into their pressure: proponents seem to be emboldened by firms’ earlier grants of shareholder rights to seek additional rights. INTRODUCTION Almost thirty years ago, Chancellor William Allen famously remarked that “a corporation is not a New England town meeting.”1 Perhaps so—but efforts are under way to change this. One of the most sought-after shareholder rights is the right of shareholders to take action not just at annual meetings, the corporate equivalent of regularly scheduled political elections, but in between, at shareholder-convoked special meetings or by written consent, the corporate equivalent of town meetings. Shareholder proposals asking for the right to call a special meeting or to act by written consent, in turn, constitute one of the most common proposal types submitted over the last ten years and companies have increasingly heeded these shareholder requests. At special meetings or by written consent, shareholders unhappy with the present board may be able to elect directors more to their liking. After the near demise of staggered boards among large U.S. companies,2 the move to permit shareholders to act in between annual meetings may thus be seen as the next logical step towards making the board replaceable by shareholders at will—or as critics may say, at the whim of a shareholder majority. But not so fast. The usefulness of having the right to call a special meeting or act by written consent depends, to a much greater extent than most other shareholder powers, on other provisions of state law and the corporate governance structure. Because different shareholder powers are complements or substitutes, the relevant issue is much more complex than whether or not shareholders, can, say, call a special meeting. As a result, shareholder efforts to obtain the power to call a special meeting or act by written consent offer a unique opportunity to study the shareholder proposal mechanism and companies’ responses. This Article examines shareholder rights to call to special meeting or act by written consent from a functional and empirical perspective. On the functional side, it will present a systematic analysis showing how these rights interact with other governance provisions. Depending on these other rights, the ability to call a special meeting or to act by written consent may add little or a lot to the arsenal of shareholder powers. Perhaps most importantly, the analysis identifies three governance provisions that, oddly, have received virtually no attention by shareholder rights activists and are not part of any of the common corporate governance indices3 but that are highly significant in making rights to call a 1 TW Servs., Inc., S’holders Litig. v. SWT Acquisition Corp., No. 10427, 10298, 1989 Del. Ch. LEXIS 19, at *30 n.14 (Del. Ch. Mar. 2, 1989). 2 See Emiliano M. Catan & Michael Klausner, Board Declassifcation and Firm Value: Have Shareholders and Boards Really Destroyed Billions in Value? 14 (N.Y.U. Sch. of Law & Econ. Paper Series, Working Paper No. 17-39, 2017), https://ssrn.com/abstract=2994559 (documenting destaggering). 3 The relevant indices are the E Index, see Lucian Bebchuk, Alma Cohen & Allen Ferrell, What Matters in Corporate Governance?, 22 REV. FIN. STUD. 783, 789 (2009) (creating index to measure six corporate governance provisions: staggered boards, limits to shareholder bylaw amendments, poison pills, golden parachutes, and supermajority requirements for mergers and charter amendments), and the GIM Index, see Paul A. Gompers, Joy L. Ishii & Andrew special meeting or act by written consent effective: whether shareholders have the power to remove directors without cause, whether shareholders have the power to fill vacancies, and whether the board size is set pursuant to the bylaws or by a board resolution. On the empirical side, this Article seeks to shed light on a number of questions. Why have shareholder rights activists focused on rights to call a special meeting and act by written consent and ignored removal, vacancy and board size provisions? Do shareholders seek powers at the companies where these powers are most useful? What factors determine whether boards take steps to change the charter or the bylaws to implement these shareholder proposals? And finally, are shareholders satisfied when a proposal is implemented or does implementation encourage ever-increasing demands for governance changes? Part I presents the functional analysis. Part II documents and develops hypotheses for why shareholder rights activists have focused on obtaining the power to call a special meeting or act by written consent and ignored other elements of the governance structure that make those powers useful. Part III examines the factors that affect where shareholder proposals seeking venue powers are introduced. Part IV distinguishes between the power to call a special meeting and the power to act by written consent. Part V examines whether board can “buy peace”—whether accommodating one shareholder request satisfies proponents or whether doing so just leads to further requests. I. A FUNCTIONAL ANALYSIS FOR VENUE POWERS This Part discusses whether and why it may be important for shareholders to have a “venue” power: either the power to convene a special meeting4 or the power to take actions by written consent without a meeting.5 The analysis shows how a venue power interacts with other elements of the governance structure and how, depending on such other elements, such a power may matter more or less. Section I.A places venue powers into the context of other rules relevant to shareholder voting. This Section categorizes rules pertaining to shareholder votes into rules of initiation, rules of passage, and rules of venue and shows that rules of venue are only relevant if shareholders also have—or can obtain—the power to initiate and the vote required for passage makes adoption of a proposal a realistic prospect. Section I.B discusses the possible uses of venue powers. As explained, the principal significance of venue powers lies in the possibility that shareholders may obtain a board majority in between annual meetings through one of two Metrick, Corporate Governance and Equity Prices, 118 Q.J. ECON. 107, 110 (2003) (constructing index “to proxy for the level of shareholder rights” at a range of large firms). 4 See, e.g., DEL. CODE ANN. tit. 8, § 211(d) (2019) (providing that special meetings of shareholders may be called by board of directors or by such persons as may be authorized by the certificate of incorporation or by the bylaws). 5 See, e.g., id. § 228(a) (providing that unless otherwise provided in the certificate of incorporation, any action which may be taken at any annual or special meeting may be taken without a meeting if the requisite number of shareholders consent in writing). routes: (1) unseating incumbent directors and filling the resulting vacancies, or (2) expanding the board and filling the resulting vacancies. Section I.C compares different state laws to analyze the state’s legal rules that bear on whether shareholders can obtain a board majority in between annual meetings. State laws matter for two reasons: they generally set the “default” rules—the rules that apply to a company incorporated in the state, absent a contrary provision in the company’s charter or bylaws—and specify whether and how the default rules can be varied. Finally, Section I.D compares the two types of venue powers—the power to call a special meeting and the power to act by written consent. A. The Interaction of Shareholder Voting Rights In publicly traded U.S. corporations, shareholders vote on a host of issues. They elect directors,6 approve mergers and charter amendments,7 vote to change bylaws,8 and cast advisory votes on shareholder resolutions9 and executive pay.10 To get a proper understanding of the significance of shareholder voting—and in particular of the significance of shareholder rights to call a special meeting or act by written consent—it is helpful to divide the various rules associated with voting into three sets. The first includes rules of initiation: the rules that specify when and how a measure is presented for a shareholder vote. The second includes rules of venue: the rules that specify when and how a venue at which shareholders cast votes is convened. And the third includes rules of passage: the rules that specify the required vote to adopt a (binding or non-binding) measure. Special meetings and actions by written consent are venues for shareholder votes. Together with annual meetings, they are the only venues at which shareholders cast votes. Annual shareholder meetings have to be held approximately every twelve months.11 If the board fails to call an annual meeting in a timely fashion, shareholders can seek a court order compelling the board to call a meeting.12 Depending on the state, shareholders can commence that process between twelve to eighteen months after the prior annual meeting.13 6 See, e.g., id. § 216 (providing for election of directors by plurality of shareholders). See, e.g., id. §§ 242, 251 (specifiying, respectively, shareholder voting procedures for amending charter and approving merger). 8 See, e.g., id. § 109(a). 9 See SEC General Rules and Regulations, 17 C.F.R. § 240.14a-8 (2018) (requiring shareholder proposal be included in proxy form for shareholder approval or disapproval). 10 See id. § 240.14a-21 (requiring a shareholder advisory vote on executive compensation). 11 See DEL. CODE ANN. tit. 8, § 211(b) (calling for an annual shareholder meeting). 12 See id. § 211(c) (specifying that if no annual meeting is held “the Court of Chancery may summarily order a meeting to be held upon the application of any stockholder or director”). 13 John C. Coates IV, Explaining Variation in Takeover Defenses: Blame the Lawyers, 89 CALIF. L. REV. 1301, 1402 (2001) (noting that state laws “vary considerably . . . in when an annual meeting is required”). Further delay will ensue due to the time gaps between the filing of a complaint and the date of a court ruling, and the date of the court ruling and the date of the meeting. See Walentas v. Builders Transp., Inc., No. 11567, 1990 WL 90939, at *3-4 (Del. 7 Shareholder power to call a special meeting or to act by written consent is thus relevant if shareholders want to take an action without having to wait for the next annual meeting. Because virtually any action that can be taken at a special meeting or by written consent can also be taken at an annual meeting,14 and because an annual meeting has to be held,15 a shareholder right to call a special meeting or to act by written consent is principally about speed: it enables shareholders to act sooner than they would otherwise be able to act. Speed, of course, is sometimes of the essence. Moreover, a shareholder power to convene a venue in between annual meetings is only relevant if shareholders can place an item on the agenda for a vote and shareholders have a realistic prospect of getting the item adopted. This brings us to the rules of initiation and the rules of passage. At annual meetings, the board is required, under state law, to hold an election for directors;16 and, under federal law, to conduct an advisory “say-on-pay” vote on executive compensation and to hold a vote on properly presented shareholder resolutions.17 By contrast, at special meetings or in actions by written consent, there are no items that have to be presented for a shareholder vote. Thus, if shareholders want an item to be voted on at a special meeting or adopted by written consent, shareholders must have the power of initiation with respect to that item. Depending on the state a company is incorporated in and the company’s charter and bylaws, shareholders may have the power to initiate votes as to the following four matters:18 (1) To remove directors without cause.19 (2) To amend the bylaws. Ch. June 26, 1990) (holding that no complaint can be filed prior to thirteen-month anniversary of last annual meeting even if it is clear that corporation will not comply with thirteen-month deadline). 14 In limited respects, shareholders may have a superior initiation power at special meetings and by written consent than they do for annual meetings. The laws of several states provide that directors can only be removed at a meeting called “for the purpose of removing the director.” See infra Table A1. When shareholders have venue powers to call a special meeting or act by written consent, they can also specify the purpose of the meeting. See supra notes 4-5. But annual meetings are called by the board and it is not clear that shareholders could require the board to add removal of a board member to the purposes for which the meeting is called. Thus, shareholders may not be able to initiate a vote on director removal at an annual meeting. Obviously, this lack of initiation power would only be relevant for a director whose term does not expire at the annual meeting and who could otherwise be removed without cause. 15 See DEL. CODE ANN. tit. 8, § 211(b) (requiring annual meeting). 16 See id. (“[A]n annual meeting of stockholders shall be held for the election of directors . . . .”). 17 See supra notes 9-10. 18 See infra Tables 1-3. 19 Shareholders also have the right to remove directors for cause. See, e.g., DEL. CODE ANN. tit. 8, § 141(k) (“Any director or the entire board of directors may be removed, with or without cause . . . .”). Because “for cause” removal is practically unheard of in U.S. public corporations and the standards are strict, we will not discuss this right further. (3) To fill vacancies on the board of directors. (4) To adopt any precatory (non-binding) resolution. Because shareholder initiation powers are basically confined to these four matters,20 the utility of the power to call a special meeting or to act by written consent depends on how significant it is for shareholders to vote on these matters earlier (at a special meeting or by written consent) rather than later (at the next annual meeting) and how difficult it is to obtain the vote required for passage. The required vote for most items as to which shareholders have an initiation power is either a plurality or a regular majority of the votes cast, of the votes present at a meeting, or of the shares entitled to vote.21 Most state laws, however, permit companies to adopt a higher vote threshold by charter or sometimes by bylaw. Even to the extent that shareholders have initiation power, a supermajority requirement—say a 66.7% vote requirement to remove directors without cause—can make it difficult to pass a proposal and render a venue power useless. As a result, the usefulness of venue rights is, in practice, limited to the items as to which shareholders have initiation power and which are not subject to supermajority requirements. B. Shareholder Venue Power As this Article explains, the most important function of a shareholder venue power is to enable shareholders to change the composition of the board between annual meetings. Two of the items as to which shareholders can initiate votes— removal of directors and filling of vacancies—relate directly to board composition, and one other item—bylaw amendments—can relate indirectly to board composition. The terms of incumbent directors expire only at annual meetings.22 If a board is staggered, directors generally have three-year terms, and the terms of about one third of the board members expire at each annual meeting; in an annually elected board, directors have one-year terms, all of which expire at the annual meeting.23 Terms, however, do not expire between annual meetings.24 20 Among the states in our sample, Minnesota and Ohio do not require a prior board recommendation before shareholders can adopt a charter amendment. See MINN. STAT. § 302A.135 (2018) (charter amendment may be proposed by board or shareholder holding 3% of voting power); OHIO REV. CODE ANN. § 1701.71 (West 2018) (shareholders may adopt charter amendment not proposed by board with two-thirds vote). Shareholders of companies in these states thus also have the power to initiate charter amendments, though Minnesota and Ohio companies often require supermajority approval for certain charter amendments. We will for ease of exposition omit further references to this power. In Massachusetts, shareholders also have the unilateral right to opt-out of section 8.06(b) by a two-thirds vote. See MASS. GEN. LAWS ch. 156D, § 8.06(c)(1) (2018). Such opt-out rights are conceptually analogous to bylaw amendments. 21 See infra Table A1. 22 See, e.g., DEL. CODE ANN. tit. 8, § 141(d) (“[T]he term of office of those [directors] of the first class to expire at the first annual meeting . . . .”). 23 See, e.g., id. (providing for option of staggered structure for election of directors). 24 See id. (providing that terms expire at annual meetings). To obtain a board majority between annual meetings, shareholders must thus be able to create vacancies and be able to fill them. Shareholders have two alternative routes to create vacancies. The first route is to remove incumbent directors; the second route is to enlarge the board size.25 To have the effective ability to obtain a majority of the board at a special meeting or by written consent, shareholders must be able either to remove at least a majority of the incumbent board and then fill the vacancies created by the removal (“remove + fill”) or to expand the board to more than twice its present size and then fill the vacancies created by the expansion (“expand + fill”).26 To avail themselves of the first route, shareholders must have both the power to initiate a removal and the power to fill vacancies, and cannot be constrained in their practical ability to exercise these powers by supermajority voting requirements. To avail themselves of the second route, shareholders must have both the power to initiate a sufficient increase in the board size and to fill vacancies, and cannot be constrained in their practical ability to exercise these powers by super-majority voting requirements. We will refer to a venue power where the other applicable rules would permit shareholders to elect a board majority through either “remove + fill” or “expand + fill” at a single special 25 For a comprehensive examination of these routes, see Coates IV, supra note 13, at 1388 app. B (“Every public company has a set of ‘governance terms’ that regulate how easily shareholders can assert control rights over the company.”). Professor John Coates’s insights provide some of the major foundations of our functional analysis. While our analysis in Sections I.B and I.C differs from the one by Coates is some specific details, it follows it in most respects. 26 Cf. Coates IV, supra note 13, at 1399 (noting importance of either having power to remove or power to expand and fill). In some states, if a majority of the board is removed, a shareholder may petition the court to summarily order an election to fill the vacancies. See, e.g., DEL. CODE ANN. tit. 8, § 223(c) (“[T]he Court of Chancery may, upon application of any stockholder or stockholders . . . having the right to vote for such directors, summarily order an election to be held . . . to replace the directors chosen by the directors then in office as aforesaid . . . .”). Coates concluded that the power to remove directors, without the power to fill vacancies, may therefore suffice to effect a “coup.” See Coates IV, supra note 13, at 1399 n.265 (“[I]f shareholders can remove the entire board in one fell swoop, a coup can be mounted even if they do not technically have the power to fill vacancies.”). However, the Delaware Court of Chancery has subsequently held that section 223(c) constitutes only a “limited exception to Section 223(a)’s grant of director authority [to a board minority] to fill board vacancies . . . [and] merely creates a narrow avenue whereby the Court may prevent directors from filling board vacancies where doing so is necessary to avoid some identifiable inequity.” Canmore Consultants Ltd. v. L.O.M. Med. Int’l, Inc., No. 8645-VCG, 2013 WL 5274380, at *3-4 (Del. Ch. Sept. 19, 2013). The court further suggested that the limited exception may be satisfied where a large percentage of the shareholders petition the court, though it is unclear how a court would weigh a charter provision specifying that only directors may fill vacancies in between annual meetings. See id. at *4 (describing case in which fortythree percent of shareholders successfully petitioned court for new election). In any case, further delay would ensue from the time of the petition until the time of a court-ordered election. meeting, or a single solicitation of written consents, with no more than a majority of the shares entitled to vote as an effective venue power.27 Having an effective venue power can be quite important. The board has the right to manage the company and, in some instances, a board may take an action before the next upcoming meeting—embarking on a major acquisition, effecting a spin-off, or selling major assets28—that would be difficult to reverse. Moreover, the ability to obtain a board majority in between annual meetings is significant in the hostile takeover context. A board can generally resist a hostile takeover bid by refusing to redeem a poison pill. As long as a pill is in place, a target board opposed to a bid can use this time to solicit a competing bid, to pursue a defensive transaction, to do nothing but delay the raider—perhaps hoping that economic fundamentals will change or that the raider’s financing will collapse—or to threaten to do any of these in order to negotiate for a better price. If target shareholders can obtain a board majority in between annual meetings, the amount of time for such resistance can be cut short.29 Even if the venue power cannot be used by shareholders to obtain a board majority, the power to remove directors between annual meetings (without being able to fill the vacancies) or the power to expand the board and fill vacancies to obtain minority representation may be significant. To be sure, removal of most of the board or obtaining minority representation would leave in the incumbent board in control. But such actions can send a powerful signal and may well induce a change in policies. In theory, the removal power could even be used to remove, or to threaten to remove, the entire incumbent board.30 A removal of the entire board, leaving no remaining board members to fill the vacancies created and in circumstances where the shareholders also lack the power to fill any vacancy, would be a highly significant act. But such a removal could also create major havoc and disruptions: a court would have to appoint an acting board or order another shareholder meeting to be held and, until then, the company would lack any board capable of taking actions. The adverse effects that may well be generated by a such an interregnum may make it harder to convince the requisite majority of shareholders to vote in favor of a removal under such circumstances and may reduce the credibility of a threat of such removal. 27 For a similar analysis of what constitutes a “no minimum term” board in a takeover, see Lucian Arye Bebchuk, John C. Coates IV & Guhan Subramanian, The Powerful Antitakeover Force of Staggered Boards: Theory, Evidence, and Policy, 54 STAN. L. REV. 887, 910 (2002) (defining “no minimum term” board as one where shareholders (1) have ability to remove directors without cause and petition court to order new election for directors or (2) pack the board by increasing number of directors). 28 See, e.g., Alexandra Stevenson, Activist Hedge Fund Starboard Succeeds in Replacing Darden Board, N.Y. TIMES, Oct. 10, 2014, at B6 (recounting that Darden Restaurants sold Red Lobster over objection of group of activist shareholders and that shareholders had only succeeded to replace board after sale). 29 See Bebchuk, Coates IV & Subramanian, supra note 27, at 916. 30 See Coates IV, supra note 13, at 1399 (observing that ability to remove directors “gives shareholders the ability to mount a ‘coup’ rather than waiting for regularly scheduled elections of directors”). By contrast, having the power to fill vacancies without the power to remove directors or expand the board size would be useless. Generally, a board has no vacancies to fill in between annual meetings. If a vacancy happens to arise (say, because a director resigns), it can usually be filled by the other board members at the next board meeting before shareholders have the opportunity to act.31 Similarly, having the power to expand the board without having the power to fill vacancies is of no use; the vacancies resulting from an increase in the board would just be filled by the other board members or be left unfilled. Aside from using the venue power to change board composition, shareholders could also use it to change the bylaws. Bylaws can contain some significant governance provisions dealing with issues such as staggered boards,32 eligibility requirements to serve on the board,33 board size,34 proxy access,35 reimbursement of proxy contest expenses,36 forum selection in shareholder lawsuits,37 majority voting for the election of directors,38 supermajority requirements for board actions,39 and more. Mostly, however, outside the context of removing and electing directors, it is not hugely significant whether one of these provisions is changed in between annual meetings or at the next upcoming annual meeting. In other words, when it comes to amending bylaws, time is generally not of the essence. Relative to obtaining board representation or removing board members, the generic ability to change the bylaw between annual meetings is thus substantially less important. Adopting precatory—that is, non-binding—resolutions would be even less important. In sum, the main significance of venue powers is to effect a board change in between annual meetings. In this respect, venue powers interact with the power 31 See DEL. CODE ANN. tit. 8, § 223(a)(1) (“Vacancies and newly created directorships resulting from any increase in the authorized number of directors . . . may be filled by a majority of the directors then in office . . . .”). 32 See, e.g., id. § 141(d) (permitting corporation’s bylaws to divide board directors into three classes with staggered terms). 33 See, e.g., id. § 141(b) (“The certificate of incorporation or bylaws may prescribe other qualifications for directors.”). 34 See, e.g., id. (“The number of directors shall be fixed by, or in the manner provided in, the bylaws . . . .”). 35 See, e.g., id. § 112 (allowing bylaws to require that, under certain conditions, individuals nominated by shareholders be included in company’s proxy solicitation materials). 36 See, e.g., id. § 113(a) (“The bylaws may provide for the reimbursement by the corporation of expenses incurred by a stockholder in soliciting proxies . . . .”). 37 See, e.g., id. § 115 (“[T]he bylaws may require . . . that any or all internal corporate claims shall be brought solely and exclusively in any or all of the courts in this State . . . .”). 38 See, e.g., id. § 216. 39 See, e.g., id. § 141(b) (“The vote of the majority of the directors . . . shall be the act of the board of directors unless . . . the bylaws shall require a vote of a greater number.”). Such a provision recently became an issue in the dispute between Shari Redstone and the board of CBS. See Keach Hagey & Joe Flint, Shari Redstone Moves to Defend Family’s Voting Power Over CBS, WALL STREET J. (May 16, 2018, 7:24 PM), https://www.wsj.com/articles/ redstones-call-cbs-maneuver-unprecedented-usurpation-of-voting-power-1526490887 (reporting “change to CBS’s bylaws requiring a supermajority of board members to approve actions such as dividends and amendments to bylaws”). to remove incumbent board members, the power to expand the board, and the power to fill vacancies. The significance of any of these powers cannot be determined in isolation but depends on what other powers shareholders have. A venue power would be most significant if that power could be used by shareholders to obtain a board majority between annual meetings, either because shareholders have the power to “remove + fill” or the power to “expand + fill” a majority of seats. It may also be important, though less so, if that power could be used by shareholders to obtain a board minority between annual meetings or to remove incumbent directors. If it cannot be used for any of these purposes, a venue power is largely insignificant. C. The Powers to “Remove + Fill” and to “Expand + Fill” Whether shareholders have the power to “remove + fill” or the power to “expand + fill” depends on the default corporate law rules of the state where the company is incorporated, the ability under state law to modify the default rules, whether a company has adopted a different rule, and whether a company has entrenched such a rule. Let us start with the default rules of corporate law. Table A1 gives the applicable default standards for removal of directors without cause and for filling of vacancies and the provisions on how these standards can be changed for twenty-nine states.40 As Table A1 shows, state laws differ significantly in their default rules on whether directors can be removed. In some states, shareholders lack that power altogether or require a supermajority to exercise it. In others—among them Delaware—shareholders lack that power for directors of staggered boards, but can remove directors of annually-elected boards without cause. In a third group, the default rule provides that shareholders can remove directors without cause by a regular majority of shares entitled to vote or, in some states, of shares voted. States laws also differ on how the removal standard can be varied. Most often, the removal standard can be varied only by a charter amendment. But in several states, the standard can also be varied by any bylaw provision or by a bylaw provision adopted by shareholders. Finally, Delaware (for directors on nonstaggered boards), California (for all directors) and a few other states permit no alteration of the “without cause” standard. Other than in California, however, even in states where shareholders enjoy an absolute right to remove the directors without cause, the charter can specify that a supermajority vote is needed for such removal. In almost all states, shareholders have the power to fill vacancies (Massachusetts and Iowa being the lone exceptions).41 Again, charter and sometimes bylaw provisions can be used to modify that power. 40 The states included in Table A1 and Table A3 are the states of incorporation of the firms that are part of the sample that we describe below. 41 Some state corporate codes explicitly provide for shareholder power to fill vacancies. See, e.g., CAL. CORP. CODE § 708(c) (West 2007). Since the shareholder power to fill vacancies is part of the common law shareholder rights, see, e.g., Moon v. Moon Motor Car Co., 151 A. 298, 302 (Del. Ch. 1930) (holding that shareholders have inherent power to fill State statutes do not directly set the board size, but provide that the board size may be regulated either in the charter or in the bylaws.42 Charters of publicly traded companies rarely set a specific board size and sometimes do not address board size at all. When they address it, they provide either that the board size is set by or pursuant to the bylaws or that it is determined by a board resolution and, in addition, sometimes set a minimum floor or maximum limit on the board size. Bylaw provisions, in turn, may either set the actual board size or delegate that power to the board, in the latter case sometimes subject to a bylaw-provided minimum or maximum.43 Corporations can, and do, use charter and bylaw provisions not only to modify the default rules provided by state law but also to entrench rules by making them more difficult to change. If a rule supplied by state law can only be changed in the charter or if a corporation has included an applicable rule in its charter, it would require a charter amendment to modify the rule. Charter amendments must generally be first proposed by the board and then adopted by shareholders.44 In effect, therefore, such rules cannot be changed by shareholders (or the board) unilaterally. However, to the extent a rule supplied by state law can be changed in the bylaws or that a corporation included a provision in its bylaws, and the charter is silent on the issue, it merely requires a bylaw amendment to modify the rule or provision. If shareholders have the power to amend the bylaw, they can adopt such an amendment without board approval.45 Consider, for example, a Georgia firm with an annually elected board that has bylaws denying shareholders the powers to remove directors and to fill vacancies and delegating the power to set the board size exclusively to the board. If shareholders had a venue power, they could call a special meeting to vote on: (1) a bylaw amendment to repeal the provision that denies shareholders the power to remove directors, (2) a bylaw amendment to repeal the provision that denies shareholders the power to fill vacancies, (3) a resolution to remove the incumbent directors, and (4) the election of new directors to fill the resulting vacancies—and thereby obtain a board majority through “remove + fill.” Or they could call a special meeting to (1) repeal the bylaw provision that denies shareholders the power to fill vacancies, (2) repeal the bylaw provision that delegates the power to set the board size to the board and to adopt a provision that sets the board size at a level that creates sufficient vacancies, and (3) elect vacancies between annual meetings), we have assumed that shareholders have the power when the corporate codes are silent. 42 See, e.g., DEL. CODE ANN. tit. 8, § 141(b) (“The number of directors shall be fixed by, or in the manner provided in, the bylaws, unless the certificate of incorporation fixes the number of directors . . . .”). 43 For an example of this and the other provisions mentioned in this paragraph, see Pfizer, Inc., Restated Certificate of Incorporation of Pfizer, Inc. (Apr. 12, 2004). 44 But see supra note 20 (discussing Minnesota and Ohio law). 45 By the same token, for corporations with bylaws creating a staggered board in states that have a stricter removal standard for directors on staggered boards, shareholders would be able to relax the removal standard indirectly by eliminating the staggered board through a bylaw amendment. new directors to fill the resulting vacancies—and thereby obtain a board majority through “expand + fill.” There is thus yet another set of provisions that bears on the power to “remove + fill” or “expand + fill”: provisions of state law limiting the shareholder power to adopt bylaw amendments or authorizing supermajority requirements for such amendments and corresponding provisions in charters or bylaws imposing such supermajority requirements. Most state laws contain no limits on shareholders’ powers to adopt bylaws46 but authorize supermajority requirements for such amendments if contained in the charter (or, sometimes, in bylaws adopted by shareholders). Table A2 below identifies states that have a different regime.47 This Article uses the term “latent power” to refer to instances where shareholders have the power to amend the bylaws by a regular majority vote in a way that gives them the immediate power to remove directors, to fill vacancies, or to set the board size. It uses the term “effective power” to describe cases where shareholders, potentially after the exercise of their latent powers, can either “remove + fill” or “expand + fill” by a regular majority vote and thereby obtain a board majority. If shareholders have venue rights and the effective power to “remove + fill” or “expand + fill,” we refer to the venue rights as effective. Finally, it uses the term “partially effective power” to describe cases where shareholders can either (1) remove a majority of the board or (2) create at least one new directorship and fill it, or both, by a regular majority vote and thereby change the board composition, whether or not they can obtain a board majority. D. Special Meetings and Written Consent As the last element of this framework, this Article compares the relative significance of the power to call a special meeting and the power to act by written consent. A few states either do not permit action by written consent for public companies48 or impose conditions or limitations on actions by written consent 46 See Coates IV, supra note 13, at 1345 n.159 (observing that most jurisdictions allow shareholders to amend bylaws). 47 In some states, notably Delaware, it is unclear whether a board can use its power to amend the bylaws to impose a supermajority requirement governing shareholder amendments. Delaware law, in principle, permits such supermajority requirements to be contained in the bylaws. However, Delaware law provides that the grant of power to the board to amend bylaws “shall not divest the stockholders . . . of the power, nor limit their power to adopt, amend or repeal bylaws.” See, e.g., DEL. CODE ANN. tit. 8, § 109(a). 48 See, e.g., infra Table A3 (showing that Indiana, Minnesota, Missouri, North Carolina, Oklahoma, and Washington do not permit action by written consent for public companies). Some of these states permit shareholder action by written consent only if all shareholders consent, a standard that cannot be realistically satisfied in public corporations. See, e.g., ARK. CODE ANN. § 4-26-710(a) (West 2018) (“Any action required . . . to be taken at a meeting of the shareholders of a corporation or any action which may be taken at a meeting of the shareholders may be taken without a meeting if a consent in writing, setting forth the action so taken, shall be signed by all of the shareholders entitled to vote with respect to the subject matter thereof.”). that substantially impede the exercise of that power.49 In most states, however, both powers are available.50 As either power permits shareholders to act in between annual meetings, the two powers are to a significant degree substitutes: once one of the powers is available, the second adds little if anything to the arsenal of shareholders. Although for parts of the analysis, this Article treats the two venue powers as substitutes, other parts take into account some of the subtle respects in which these powers differ. Specifically, acting by written consent has one potential advantage for a shareholder who wants to convene a venue. Any shareholder, regardless how many (or few) shares she owns, can seek to solicit written consents on a proposal. By contrast, the power to call a special meeting may require a two-step process. State laws, and corporate charters and bylaws, afford that power only to a shareholder, or a group of shareholders, who own the requisite fraction of shares, typically between 10% and 50%.51 Thus, a shareholder who does not already own the required percentage must first obtain the support of other shareholders to call a special meeting; and then, once that meeting has been called, the shareholder must distribute proxies asking shareholders to vote on the proposal to be presented at the special meeting. This two-step process can take more time and involve greater expenses than the onestep process of soliciting written consents.52 On the other hand, special meetings have a potential advantage over actions by written consent. Unlike votes at shareholder meetings, where shareholders generally have the option to vote for, vote against, or abstain, written consents enable shareholders only to consent to—in effect, vote for—a proposal.53 The laws of most states therefore provide that in order for a proposal to pass, the proposal must obtain the number of written consents that would be needed to have the proposal pass at a shareholder meeting where all shares entitled to vote are present and voted.54 This rule, in effect, supposes for purposes of determining whether a proposal passes that any shareholder who did not sign the written consent would have voted against the proposal had that shareholder the chance to do so. This difference in the rules of passage, however, is not always relevant. It is significant only when the standard rule of passage (i.e., the rule that would apply at a special or annual meeting) requires that a majority of shares present or a 49 See, e.g., infra Table A3 (Ohio allows only bylaw amendments and Utah does not allow for election of directors). 50 See infra Table A3 (surveying availability of actions by written consent in different states). 51 See infra Table A3. 52 There may also be differences in the degree of board control over matters such as the time, location, and record date for a special meeting. 53 See, e.g., DEL. CODE ANN. tit 8, § 211(b) (2019) (describing procedure for using written consent in lieu of annual meetings). 54 See, e.g., id. § 228(a) (requiring written consent to have “not less than the minimum number of votes that would be necessary to authorize or take such action at a meeting at which all shares entitled to vote thereon were present and voted”). majority of votes cast (a “simple majority”) vote in favor of a proposal. Where the standard rule of passage requires that a specified (regular or super) majority of shares entitled to vote vote in favor of a proposal, the support needed to secure passage would be the same at a special meeting and at an action by written consent. As a result, the practical significance of this advantage is limited. The differential rule of passage thus would only make it easier for shareholders to obtain a board majority in between annual meetings if either (1) the “remove + fill” option is available and the applicable rules require only a simple majority to remove directors and to fill vacancies or (2) the “expand + fill” option is available and the applicable rules require only a simple majority to amend the bylaws and to fill vacancies.55 But most corporations are incorporated in states where the minimum required vote to remove directors is a majority of shares entitled to vote; for those companies, as well for others where the charter requires a majority of shares entitled to vote to remove directors or fill vacancies, “remove + fill” would not be made easier.56 Moreover, even though most states’ laws require only a simple majority to amend bylaws, many companies require a majority (or even a supermajority) of shares entitled to vote to do so, with the result that the “expand + fill” options would also not be made any easier.57 II. WHY FOCUS ON VENUE POWERS? Six principal types of corporate charter and bylaw provisions, in conjunction, determine whether shareholders can “replace + fill” or “expand + fill” the board in between annual meetings: (a) Whether the charter or bylaws directly modify the state default rules on removal. (b) Whether the charter or bylaws establish a staggered board which, in many states, affects the state law default rules on removal. (c) Whether the charter or bylaws modify the right of shareholders to fill vacancies. (d) Whether the charter provides that the size of the board is determined by a board resolution and, if not, whether it imposes a maximum constraint on the board size. 55 See supra text accompanying note 27 (defining “remove + fill” and “expand + fill” options). 56 See infra Table A1 (listing such states with this requirement). 57 Compare DEL CODE ANN. tit. 8, § 109(a) (“After a corporation other than a nonstock corporation has received any payment for any of its stock, the power to adopt, amend or repeal bylaws shall be in the stockholders entitled to vote.”), with DANIEL E. WOLF & MICHAEL P. BRUECK, KIRKLAND & ELLIS LLP, KIRKLAND M&A UPDATE: VOTING STANDARDS ARE NOT THAT STANDARD 1 (Oct. 10, 2016), https://www.kirkland.com/siteFiles/Publications/Voting_ Standards_Are_Not_That_Standard.pdf [https://perma.cc/EZ6Y-VM7S] (describing supermajority approval provisions for bylaw amendments as a “common variation”). Unlike supermajority vote requirements, charter and bylaw provisions that change the vote requirement from a simple majority to a majority of shares entitled to vote have not attracted shareholder opposition. Companies would thus face no pressure in retaining these requirements. (e) Whether the charter or bylaws require a supermajority vote for bylaw amendments. (f) Whether the charter or bylaws grant or remove a venue power. Shareholder rights activists have pushed for changes in the charter and bylaws for three of these six types of provisions: the elimination of staggered boards, the elimination of supermajority requirements to amend the bylaws, and the grant of venue powers. Between 2005 and the end of 2017, to our knowledge, there were 322 shareholder proposals requesting that the company change its governing documents to permit shareholders to call a special meeting, 122 proposals requesting that the company lower the percentage of shareholders needed to call a special meeting, and 244 proposals requesting that shareholders be permitted to act by written consent. These efforts were substantially successful. Between 2005 and 2017, the percentage of companies in our dataset where shareholders have a venue power has increased from 49% to 69%. Similarly, there were 655 proposals to destagger the board, which contributed to a decline in the percentage of companies in our dataset with staggered boards from 49% to 12%, and 372 proposals to eliminate supermajority requirements, which contributed to a decline in the percentage of companies in our dataset with such requirements from 41% to 26%. But, shareholder rights activists have virtually ignored the other three types of provisions. To our knowledge, there have only been two proposals regarding shareholders’ power to fill vacancies;58 three proposals regarding shareholders’ right to remove directors without cause; and no proposals regarding shareholders’ power to set the board size through a bylaw amendment. Of these few proposals, several were made in actual contests, where the proponents sought to remove incumbent boards and replace them with a different set of nominees. By contrast, the proposals on special meetings and written consent, as well as those on the elimination of staggered boards and supermajority provisions, were made by shareholder rights activists with no current plans to use a venue power on their own who presumably wanted to lay the groundwork for the future use of these powers by others. Moreover, shareholder rights activists have sought the elimination of staggered boards because staggered boards, if “effective,” make it impossible for shareholders to replace a board majority at an annual meeting, and not because its elimination, through the effect on the removal standard, would make it possible to “replace + fill” in between annual meetings.59 Likewise, shareholders may have sought to eliminate supermajority requirements to amend 58 There were an additional three proposals intended to eliminate any ambiguity that shareholders already had that power and six proposals divesting the board from its power to fill vacancies. 59 The website for the Shareholder Rights Project, which contributed to board declassifications at about one hundred S&P 500 and Fortune 500 companies, does not even mention the effect of declassification on the removal standard. See 102 Companies Declassified, SHAREHOLDER RTS PROJECT, http://www.srp.law.harvard.edu/declassificatio ns.shtml [https://perma.cc/B87V-5MMT] (last visited Apr. 19, 2019) (discussing companies declassified but not mentioning removal standard). the bylaws for reasons other than their effect on the ability to “replace + fill” or “remove + fill.”60 This selective focus by shareholder rights activists on only one of the four provisions that most directly bear on whether a company offers an effective venue power is peculiar. While venue powers can be important, they are not always so; and by the same token, the power to remove, the power to fill vacancies, and the power to set the board size can be important as well. These powers are relevant in companies where shareholders already have a venue power, but where this power is not effective because it cannot be used either to “remove + fill” or to “expand + fill.” And these powers are also important where they would render an otherwise “effective staggered board”—that is, a staggered board where removal without cause is not permitted and where shareholders could not expand the board size and fill the vacancies—ineffective and hence enable shareholders to elect a majority of directors at an annual meeting. Shareholder rights activists thus seem to have been quite successful in the battle they have waged. But they have not even started to wage other, equally important, battles to make venue powers effective. There are a number of potential explanations for this failure. First, shareholder rights activists may be unsophisticated about, or indifferent to, the effect of venue rights. Beyond knowing what powers shareholders already have, they lack an understanding of how these powers interact and what they achieve. The implication of this explanation (the “unsophisticated proponent hypothesis”) would be that venue proposals are introduced randomly within the set of companies where the venue power at issue is lacking. Second, shareholder rights activists may follow a semi-sophisticated pecking order in pushing for governance changes. They start off with a heuristic notion of what rights are most important under the rules that govern most firms and then push for them roughly in the order of importance—but without paying attention to how these powers interact with rules that govern specific firms. The pecking order is semi-sophisticated in as much as the elimination of a staggered board generally enables shareholders to replace a board majority in a single annual meeting (as opposed to having to do so over two annual meetings) and the grant of a venue power often enables shareholders to obtain a board majority in between annual meetings (as opposed to having to wait for the next annual meeting). But it is only semi-sophisticated in as much as the grant of a venue power (and the elimination of a staggered board) are sought regardless of whether, for the company at issue, the grant would have that result. The implication of this explanation (the “pecking order hypothesis”) would be that activists first seek to destagger boards—the more fundamental change—and 60 Supermajority elimination proposals often seek to remove supermajority requirements generally, without identifying supermajority requirements for bylaw amendments specifically. See Marcel Kahan & Edward Rock, Symbolic Corporate Governance Politics, 94 B.U. L. REV. 1997, 2017 (2014). Eliminating supermajority vote requirements for charter amendments generally confers no latent powers on shareholders as charter amendments must be recommended by the board before they come up for a shareholder vote. See id. at 2016-18 (arguing supermajority requirements often “irrelevant” for this reason). introduce venue proposals in companies that have already destaggered or always had annually elected boards, but do so randomly within that set of companies.61 Third, shareholder rights activists may focus on proposals that are most likely to enjoy shareholder support and be implemented by the board. They may well have a sophisticated understanding of the effect of their proposal, but prefer not to waste their efforts on proposals unlikely to succeed. After all, filing precatory proposals entails both monetary and non-monetary costs: proponents need to obtain letters from their brokers verifying that the proponent has held the shares in the issuer with value in excess of $2,000 over at least one year; they also typically invest time in responding to no-action letters filed by issuers before the Securities and Exchange Commission; finally, proponents need to attend the meeting where the proposal will be voted on, or send a representative.62 For an individual shareholder that files between fifty and one hundred proposals on various topics in a typical a year, these costs are non-negligible. For whatever reason, shareholder rights activists may have concluded that venue proposals are more likely to succeed than proposals to permit shareholders to fill vacancies, to remove directors without cause, or to have the board size be determined by the bylaws, and thus focus on the former. The implication of this explanation (the “in the cards hypothesis”) would be that venue proposals are introduced at companies where grant of a venue power, without any additional changes in the provisions governing removal, vacancies, and board size, would confer effective venue powers on shareholders. Fourth, shareholder rights activists may follow a sophisticated multi-step strategy. For the time being, they focus on obtaining venue powers where shareholders lack venue power regardless of whether that power would be effective. Once they have succeeded on that front, they plan to turn, at firms where the venue power is not yet effective, to seeking additional changes to the power to remove, fill vacancies, or expand the board size that would make the venue power effective. The implication of this explanation (the “one step at a time hypothesis”) would be that venue proposals are introduced at companies where shareholders lack any venue power regardless of whether the grant of a venue power would enable shareholders to “remove + fill” or “expand + fill” between annual meetings. 61 Even prominent corporate law professors have sometimes employed a similar approach. Thus, in their seminal study of IPO charter provisions, Professors Robert Daines and Michael Klausner classify companies in terms of anti-takeover protections into five levels—those with staggered boards and no venue power (or dual class stock), just staggered boards, just no venue powers, venue powers, and no staggered boards (subdivided into two levels depending on the presence of advance notice bylaws)—without regard to whether a venue power is effective. See Robert Daines & Michael Klausner, Do IPO Charters Maximize Firm Value? Antitakeover Protection in IPOs, 17 J.L. ECON. & ORG. 83, 86-88 (2001) (discussing different antitakeover provisions). 62 See SEC General Rules and Regulations, 17 C.F.R. § 240.14a-8 (2018) (describing procedures one must follow to demonstrate eligibility to file precatory proposal). III. WHICH FIRMS ARE TARGETED BY VENUE PROPOSALS? To examine venue powers empirically, we collected data from firms in the S&P 500 index for the period 2005 to 2017. We selected 2005 as the starting data because firms became subject to increasing pressure to change their rules on venue to provide shareholders with a venue power in 2006. Up to 2006, such pressure was virtually absent.63 We excluded firms not incorporated in the United States, firms with a dual-class share structure, REITs, and utilities. The remaining sample comprises over 4,800 firm-year observations and almost 600 unique firms. Two thirds of the firm-year observations correspond to firms incorporated in Delaware. After Delaware, the most popular states of incorporation are New Jersey, New York, and Ohio (with just over 2.6%, 4%, and 2.8% of the firm-years, respectively). For the firms in the data set, we collected data, for each year that the firm was in the data set, on the governance structure under state law; the firm’s charter and its bylaws, including data on venue power, board structure, provisions on removal, filling of vacancies, and board size; and supermajority voting requirements for bylaw amendments. Figure 1 describes the incidence of several governance features across the firms in our sample at the end of each year. As of the beginning of our sample period, 36% of the firms in the sample provided for shareholder rights to call a special meeting, 25% for written consent, and 12% for both venue powers. However, in roughly 40% of the companies with venue powers, venue powers could not be used to gain control of a majority of the board between annual meetings by simple majority voting;64 for about 30% 63 A search in SharkRepellent does not produce a single shareholder venue proposal dated on or before December 31, 2004. See SHARKREPELLENT.NET, https://www.sharkrepellent.net/ [https://perma.cc/EN4S-Z7PS] (last visited Apr. 19, 2019) (offering “in-depth research on US public companies’ takeover defense strategies”). 64 To be sure, by coding for the presence of the nominal ability to use venue powers to influence the board composition, we may be overstating the degree of effectiveness of the shareholders’ ability to use those venues. As the following example illustrates, the small print of the provisions handling the process through which shareholders may call a venue may be more or less demanding. During 2014, the activist hedge fund Pershing Square teamed up with Valeant Pharmaceuticals to try to aquire Allergan Inc, a large pharmaceutical company. See Stephen Davidoff Solomon, Allergan-Valeant Fight Holds Lessons for All Corporate Shareholders, N.Y. TIMES: DEALBOOK (Sept. 18, 2014, 4:05 PM) https://dealbook.nytim es.com/2014/09/18/allergan-valeant-fight-holds-lessons-for-all-corporate-shareholders/ (describing Pershing’s attempted acquisition). Part of the hostile bidders’ strategy involved taking advantage of the fact that Allergan had recently amended its charter to allow their shareholders to call special meetings to promptly unseat a majority of Allergan’s board. See id. (describing Allergan’s decision to amend charter allowing for special meetings). However, the provision in Allergan’s bylaws that spelled out the requirements that shareholders needed to satisfy to call a special meeting was so burdensome that Chancellor Andre Bouchard from the Delaware Court of Chancery referred to that provision in a hearing as “quite a horsechoker.” See id. Although Allergan and the hostile bidders settled the controversy surrounding the requirements for calling a special meeting soon after the Chancellor’s remark, other firms’ bylaws may also contain requirements that render a nominally effective venue power relatively less likely to be used. See David Gelles, Allergan Escapes Valeant’s Pursuit, of the companies with venue powers, venue powers could not be used to change the board composition in any manner. In addition, we collected data on whether a shareholder made a proposal seeking a venue power and if the board adopted or proposed to shareholders the adoption of a venue power. For any shareholder proposal, we determined the identity of the proponent,65 whether the proposal was implemented prior to a shareholder vote, whether the proposal was voted on, the voting outcome, and whether the proposal was implemented subsequent to a vote. The dynamics in the presence of the power to call special meetings reported in Figure 1 is tightly linked to whether firms received a shareholder proposal asking them to give shareholders that power: out of the 114 firms in our sample that granted that power over 2005-2017, 80% had received a precatory proposal. Relatedly, 84% of the unique firms that received at least one shareholder proposal asking for the right to call special meetings had granted their shareholders that right by the end of 2017. Agreeing to Be Bought by Actavis, N.Y. TIMES: DEALBOOK (Nov. 17, 2014, 9:34 AM) https://dealbook.nytimes.com/2014/11/17/allergan-agrees-to-be-sold-to-actavis/. 65 The proposals were almost exclusively filed by individuals (as opposed to pension funds or other institutional investors). Remarkably, close to 90% of the proposals were filed by members of four families (the Chevedden family, the Rossi family, the Steiner family, and the Young-McRitchie family). .2 .4 .6 .8 1 Figure 1. Governance features across sample firms. 2005 2007 2009 Written Consent Annual Board Effective Venue Power 2011 Year 2013 2015 2017 Special Meetings Venue Power Partially Effective Venue Power To examine shareholder proposals to adopt a venue power more closely, we constructed a measure of whether a firm was “at risk” to receive such a proposal in a given year. We treated a firm as “at risk” in any year where the firm lacked the requisite venue and state law permitted the venue for public corporations.66 We then ran regressions to determine whether proposals are filed randomly among firm-years at risk or, if not, what factors correlate with the filing of proposals. The results are reported in Table 1. In all specifications, the dependent variable takes the value of 1 if a firm received a venue proposal in a given year, and 0 otherwise.67 As independent variables, we include Get_Bd_Majority (a dummy equal to 1 if as of the relevant year the shareholders have the power to obtain a board majority via “expand+fill” or “remove+fill”, or both, and 0 otherwise), Existing_Venue_Power (a dummy equal to 1 if as of the relevant year the firm already had a venue power, and 0 otherwise), Create_Eff_Venue_Power (if implementation of the proposal would confer previously-lacking effective venue power, set to 1 if Get_Bd_Majority=1 and Existing_Venue_Power=0, and 0 otherwise) and, in column 2, Annual_Board (a dummy equal to 1 if as of the relevant year the entire board is elected annually, 66 All states permitted the special meeting venue. We treated states that permitted written consent for public corporations only if all shareholders consented as not permitting the written consent venue. 67 For ease of interpretation, all the regression analyses in this Article estimate linear probability models. However, in unreported results, we verify that our results are robust to alternative specifications that employ logit or probit models. and 0 otherwise).68 The estimation sample for columns (1) and (2) consists of all firm-years at risk of receiving a venue proposal. 69 Table 1. What governance features are associated with the filing of precatory venue proposals? Get_Bd_Majority Existing_Venue_Power Create_Eff_Venue_Power (1) 0.018 (0.025) 0.039** (0.017) 0.026 (0.027) Annual_Board Change_Bd_Comp Create_PartEff_Venue _Power Obs. R-squared Year FE 3684 0.038 Yes Received Venue Proposal (2) (3) -0.009 (0.026) 0.008 0.009 (0.016) (0.011) 0.001 (0.026) 0.102*** (0.016) 0.096*** (0.019) -0.009 3684 0.056 Yes (4) -0.017 (0.011) 0.065*** (0.016) 0.068*** (0.021) -0.034* (0.021) 3684 0.055 Yes Standard errors (in parentheses) are clustered at the firm level.*** p<0.01, p<0.1. (0.021) 3684 0.059 Yes ** p<0.05, * In the first specification, only the estimate of the coefficient for Existing_Venue_Power is statistically significant and positive. This would indicate that proposals are more likely to be filed in firm-years where shareholders already have a venue power—a result not predicted by any of the hypotheses. In the second specification, the estimated coefficient for Annual_Board is economically large and highly significant; all the other variables are now insignificant and have low economic magnitude (the coefficient estimates would indicate a 10.2 percentage point increase in the probability of receiving a proposal if the board is annually elected, and changes in the probability of receiving a proposal of less than 1 percentage point for the other variables). 68 For additional details regarding the construction of all the variables used in regression analyses, see infra Table A4. 69 Out of a total of 506 unique firms that were at risk of receiving a venue proposal at some point during our sample period, 135 unique firms were targeted by at least one venue proposal. Out of the 4,084 firm-years at risk of receiving a venue proposal, 349 firm-years were targeted by (at least) one such proposal. The results of these regressions are most consistent with the pecking order hypothesis. Venue proposals are substantially more likely to be filed in firms with annually elected boards. In fact, those results arguably understate the degree to which that pecking order is present: while 31% of the firm-year observations at risk of receiving a venue power precatory proposal had a staggered board, less than 7% of those proposals were filed at firms with staggered boards. Once we control for board structure, all the other explanatory variables are insignificant. Beyond the lack of statistical significance, the 95% confidence interval for Create_Eff_Venue_Power ranges from 5.2 percentage points to negative 5.0 percentage points. Though we cannot reject the possibility that venue proposals are more likely to be introduced where a venue would bestow a previously lacking effective venue power, we can reject the possibility that this likelihood increases by more than 5.2 percentage points. As a robustness check, in columns (3) and (4) of Table 1, we estimate an alternative specification where we use a broader definition of when a venue power would be effective.70 The explanatory variables in those specifications are Change_Bd_Comp (a dummy equal to 1 if as of the relevant year the shareholders can change the board composition either by removing directors or by expanding the board size and filling the vacancies, or both, regardless of whether they could obtain a board majority, and 0 otherwise), Existing_Venue_Power (as defined above), and Create_PartEff_Venue_Power (if implementation of the proposal would confer a previously-lacking partiallyeffective venue power, set to 1 if Change_Bd_Comp=1 and Existing_Venue_Power=0, and 0 otherwise). As in columns (1) and (2) of Table 1, Annual_Board remains significant. However, while the coefficient associated with Get_Bd_Majority was insignificant, the coefficient associated with Change_Bd_Comp is statistically significant and positive and, in column (4), the coefficient associated with Create_PartEff_Venue_Power is statistically significant and negative. These results are consistent with the pecking order hypothesis. They further indicate that the ability to change board composition is positively associated with the filing of a venue proposal. However, peculiarly, this association is stronger for firms where shareholders already have a venue power than for firms where shareholders lack a venue power. The analysis in Part IV sheds further light on this result. IV. SECOND VENUE PROPOSALS Of the 349 venue proposals, 199 proposals were filed at companies that already provided for a venue power (“second venue proposals”). To the extent that the two types of venue powers are perfect substitutes, second venue 70 We also run robustness checks under the assumptions that any firm that implemented a venue power received a shareholder request seeking such venue power and treating all such requests as proposals even if no formal proposal was filed. The results when we use the broader definition of the outcome variable are qualitatively equivalent to those in Table 1. proposals are irreconcilable with proponents having a proper understanding of these proposals. The explanation for second venue proposals must therefore either lie in the differences between special meeting and written consent powers or else indicate that proponents are unsophisticated. We now turn to an analysis that differentiates between proposals seeking special meeting powers and proposals seeking written consent power. As a first step, we estimate linear probability models with the same independent variables as in Table 1, but with separate dependent variables for special meeting and written consent proposals. The results are reported in the two panels of Table 2. For special meeting proposals (Panel A), the estimation sample consists of all firm-years at risk of receiving a special meeting proposal.71 Once we control for board structure, only the estimate for the coefficient associated with Annual_Board is significant.72 We interpret these results as indicating that, after accounting for the pecking order, special meeting proposals are filed indiscriminately of whether a firm provides for the power to act by written consent and whether shareholders could obtain a board majority or otherwise change board composition at a special meeting. For written consent proposals (Panel B), however, the picture changes.73 The estimate of the coefficient for Existing_Venue_Power is significant and positive in the specifications in columns (1) and (2). Moreover, the point estimate is very large (indicating an increase in the likelihood of receiving a written consent proposal of 8-10 percentage points per at-risk year) relative to the baseline likelihood of receiving a written consent proposal for companies that do not grant the power to call a special meeting (about 2% per at-risk year). In the specification in column (3), the estimate of the coefficient for Existing_Venue_Power is insignificant, but the estimate of the coefficient for Change_Bd_Comp is significant and positive and the estimate of the coefficient for Create_PartEff_Venue_Power is significant and negative. Taken together, these estimates suggest that written consent proposals are filed when shareholders have the ability to change board composition and also already have the right to call a special meeting. This interpretation is confirmed by the specifications in columns (4) and (5), which subdivide the sample at risk of receiving a written consent proposal between firm-years where shareholders, respectively, have and lack the ability to change the composition of the board. As reported in column (4), for firm-years where shareholders have the ability to 71 Out of the approximately 2,500 firm-years at risk of receiving a proposal requesting the firm to give shareholders the right to call a special meeting, 165 (6.6%) were targeted by such a proposal. Approximately 30% of the unique firms that did not allow shareholders to call a special meeting at some point during our sample period received at least one special meeting precatory proposal. 72 Indeed, almost 87% of the special meeting precatory proposals were filed before firms with annually elected boards of directors. 73 Out of the approximately 3,600 firm-years at risk of receiving a proposal requesting the firm to give shareholders the right to act by written consent, 187 (5.2%) were targeted by such a proposal. Approximately 20% of the unique firms that did not allow shareholders to act by written consent at some point during our sample period received at least one written consent precatory proposal. change board composition, having the power to call a special meeting is associated with a statistically significant increase of 12.3 percentage points in the likelihood of receiving a written consent precatory proposal.74 To further examine special meeting proposals, we compare firms that actually received a second venue special meeting proposals to those at risk of receiving such a proposal (i.e. firms where shareholders had written consent power but no special meeting power). If second venue proposals are driven by shortcomings in the first venue power, we would expect that second venue special meeting proposals are filed at firms where the venue powers are effective but the required voting threshold at special meetings would be lower voting than by written consent. In unreported regressions, we find that no significant evidence for this hypothesis. Table 2. What governance features are associated with the filing of precatory proposals on special meetings or written consent? Panel A Existing_Venue_Power Received Special Meeting Proposal (1) (2) (3) 0.008 -0.021 0.007 (0.020) (0.023) (0.026) Get_Bd_Majority -0.003 (0.033) Create_Eff_Venue_Power -0.014 (0.035) Annual_Board 0.096*** (0.016) 0.073*** (0.019) Change_Bd_Comp -0.007 (0.038) Create_PartEff_Venue_Power 0.030 (0.035) 2415 0.060 Yes All Obs. R-squared Year FE Sample 2415 0.032 Yes All 2415 0.060 Yes All Standard errors (in parentheses) are clustered at the firm level. *** p<0.01, ** p<0.05, * p<0.1 74 Once again, the results of the regression analyses arguably understate the degree to which shareholder proponents follow a pecking order: while only 44% of the firm-year observations at risk of receiving a written consent precatory proposal allowed shareholders to call special meetings, 86% of those proposals were filed before firms that already allowed shareholders to call a special meeting. Panel B Existing_Venue_Power (1) 0.100*** (0.015) Received Written Consent Proposal (2) (3) (4) (5) 0.081*** 0.005 0.123*** 0.017 (0.018) (0.011) (0.029) (0.012) Get_Bd_Majority 0.017 (0.031) Create_Eff_Venue_Power -0.021 (0.030) Annual_Board 0.039*** (0.009) 0.040*** (0.015) Change_Bd_Comp 0.099*** (0.024) Create_PartEff_Venue _Power 0.120*** (0.023) 3058 0.108 Yes All Obs. R-squared Year FE Sample 3058 0.088 Yes All 3058 0.093 Yes All 0.080*** (0.018) 0.016 (0.020) 1922 0.097 Yes Can Change Bd Comp’n 1136 0.041 Yes Cannot Change Bd Comp’n Standard errors (in parentheses) are clustered at the firm level. *** p<0.01, ** p<0.05, * p<0.1 Next, we examine the converse: firms that actually received a second venue written consent proposal in comparison to firms at risk of receiving such a proposal (i.e. firms with special meeting and no written consent powers in states where shareholders can act by written consent). As discussed in more detail in Section I.D above, written consents have a generic advantage over special meetings: any shareholder can commence a written consent solicitation, while only shareholders holding the requisite percentage of shares can call a special meeting. The importance of this advantage depends on the percentage required to call a special meeting. If proponents have the requisite degree of sophistication, one would expect second venue written consent proposals to be disproportionately filed at firms where the percentage required to call a special meeting is high. Such second venue written consent proposals would thus be functionally equivalent to proposals seeking to lower the percentage of shareholders required to call a special meeting. In Figure 2, we split the set of approximately 1,300 firm-years at risk of receiving a second venue written consent proposal into four groups, depending on the threshold required for shareholders to call a special meeting. Thus, for example, as denoted by the total height of the second bar from the left, 456 firmyears required the support of between 20.1% and 25% of the shares to call a special meeting. The subsection of each bar that is shaded in dark grey denotes the number of firm-years within the relevant column that were targeted by a second venue written consent proposal. As is evident, second venue written consent proposals are not filed at firms that impose a high percentage requirement for calling a special meeting.75 There is another systematic advantage of written consents over special meetings. In all states but Nevada, written consent provisions must be contained in a company’s charter.76 As a result, once shareholders have the power to act by written consent, the power can only be removed with the approval of shareholders. By contrast, in Delaware and several other states, the state default law governing special meetings can be modified in the bylaws.77 For a firm in those states, a shareholder right to call a special meeting may be removable without shareholder approval, via a bylaw amendment approved by the board, unless the firm’s charter provides for a shareholder right to call a special meeting.78 75 We do find, however, that written consent proposals are generally not filed at firms in states that impose limitations or supermajority vote requirements on written consents. Out of 187 written consent proposal, only a single one was at a firm in such a state. 76 See infra Table A3 (summarizing written consent requirements among states studied). 77 See, e.g., DEL. CODE ANN. tit. 8, § 211(d) (2019) (“Special meetings of the stockholders may be called by the board of directors or by such person or persons as may be authorized by the certificate of incorporation or by the bylaws.”). 78 Under Delaware fiduciary duty law, the board may not adopt bylaw amendments, even if it has the legal power to do so, if they are “inequitable” or have the purpose of impeding a shareholder vote. See Schnell v. Chris Craft Indus., Inc., 285 A.2d 437, 439 (Del. 1971) (holding inequitable bylaw impermissible, even if board nonetheless strictly adhered to statutory requirements); Blasius Indus., Inc. v. Atlas Corp., 564 A.2d 651, 652 (Del. Ch. 1988) (finding board’s attempt to prevent shareholder from expanding board and electing majority violated fiduciary duties). These cases arose in circumstances where the board took an action that impeded the effective exercise of these rights once the board knew of a specific shareholder plan. In Delaware and other states following Schnell and Blasius, therefore, the board may have to take actions well in advance of a specific shareholder proposal to make sure they hold up in court. Boards, of course, may be reluctant to take such actions for the very reasons for which they heed shareholder precatory resolutions to start with. 0 100 Number of firm-years 200 300 400 500 Figure 2. Thresholds of shareholder support necessary for calling special meeting across firm-years at risk of being targeted by a second venue written consent proposal. up to 20% 20.1 to 25% Not targeted by written consent proposal 25.1 to 50% over 50% Targeted by written consent proposal This points to another potential explanation for second venue written consent proposals. Perhaps these proposals are filed in firms where the proposal’s implementation would bestow on shareholders a venue power that is secure from unilateral steps by the board that would render that power ineffective.79 To test this hypothesis, in unreported analysis, we augment the specifications estimated in columns (2) and (3) of Panel B of Table 2 by including as a regressor the variable SM_power_secure set to the 1 if as of the relevant year the board cannot eliminate the right to call a special meeting without shareholder approval, and 0 otherwise. We find no significant evidence that second venue proposals are filed to make the existing venue power secure. On the whole, therefore, the evidence on written consent proposals is most consistent with a semi-sophisticated pecking order dynamic. Proponents may understand that written consents generally have advantages over special meetings, in that any shareholder can commence a written consent solicitation and that the board cannot unilaterally remove written consent powers, but do not bother to determine the difficulty of meeting the requirements for calling a 79 In our analysis, we also account for other reasons why an effective venue power may not be secure, including the unilateral ability of a board under Maryland and Massachusetts to opt into antitakeover legislation, the ability to add provisions prohibiting removal without cause or shareholder filling of vacancies to bylaws that cannot be amended by regular majority vote, and the ability to require a supermajority voting requirement for bylaw amendments. special meeting at a specific corporation or whether, in a particular case, written consents would make a venue power more secure. As written consent powers generically go beyond special meeting powers, shareholders tactically first seek special meeting powers. Given this pecking order, proponents also may exhibit some further sophistication: among the firms that grant shareholders the right to call a special meeting, firms where a venue can be used to change board composition are significantly more likely to be targeted (although firms where a venue power can be used to obtain a board majority are not). However, as the next Part shows, this result is not robust. V. CAN BOARDS BUY PEACE? The evidence of a pecking order between destaggering, the power to call special meetings, and the power to act by written consent raises the question of the implications of acceding to one set of shareholder demands on the next set of shareholder demands. In particular, this raises the question of whether boards can “buy peace”—that is, is a board that, say, recently eliminated its staggered board rewarded, compared to other companies with annual boards, by a lower likelihood of being targeted with a proposal to give shareholders the power to call special meetings? To test for the “buy peace” hypothesis, in Panel A of Table 3 we focus on the sample of firm-years at risk of receiving a special meeting proposal and estimate a linear probability model. The outcome variable equals 1 if the firm received a special meeting proposal in the relevant year and zero otherwise. The explanatory variables in the first specification include dummies for whether the board was annually elected as of the relevant year and for whether a board that was annually elected had become so as a result of a board declassification after 1996 (the latter dummy, Had_Destaggered, equals 0 otherwise).80 The second and third specifications augment the first one by including, respectively, the dummies Existing_Venue_Power, Get_Bd_Majority, and Create_Eff_Venue_Power, and the dummies for Existing_Venue_Power, Change_Bd_Comp, and Create_PartEff_Venue_Power (defined above). In Panel B of Table 3, we focus on the set of firm-years at risk of receiving written consent proposals, and estimate linear probability models where the outcome variable equals 1 for firm-years that received a written consent proposal (and zero otherwise). Specifications (1)-(3) of Panel B augment specifications (1)-(3) of Panel B of Table 2 by including as an explanatory variable the dummy Had_Granted_SM. That variable equals 1 only for those firm-year observations corresponding to firms that, as of the relevant year, allowed shareholders to call special meetings as a result of having granted such a right after 2005 and 0 otherwise.81 If boards can buy peace, the grant of a shareholder right would be 80 Because we focus on changes in the governance structure, we treat firms that went public after 1996 with an annually elected board as not having declassified. 81 Again, since we focus on changes in the governance structure, we treat firms that went public after 2005 with a shareholder power to call a special meeting as not having granted a the power to call special meetings after 2005. associated with a lower likelihood of receiving (or a delay in receiving) a proposal further down the pecking order. Table 3. Panel A Received Special Meeting Proposal (1) (2) (3) Annual_Board Had_Destaggered 0.065*** (0.015) 0.047** (0.022) Existing_Venue_Power Get_Bd_Majority Create_Eff_Venue_Power 0.067*** (0.019) 0.046** (0.023) -0.018 (0.023) 0.013 (0.034) -0.016 (0.035) Change_Bd_Comp Create_PartEff_Venue_Power Obs. R-squared Year FE Sample 2404 0.064 Yes All 2404 0.064 Yes All 0.046** (0.023) 0.046** (0.023) 0.007 (0.026) 0.008 (0.040) 0.019 (0.036) 2404 0.065 Yes All Table 3. Panel B Existing_Venue_Power Had_granted_SM Get_Bd_Majority Create_Eff_Venue_Power Annual_Board Change_Bd_Comp Received Written Consent Proposal (1) (2) (3) 0.025* 0.014 -0.009 (0.013) (0.013) (0.008) 0.235*** 0.233*** 0.225*** (0.033) (0.033) (0.035) 0.011 (0.027) -0.006 (0.027) 0.029*** 0.035*** (0.007) (0.010) 0.031* (0.019) -0.040* (0.021) Create_PartEff_Venue_Power Obs. R-squared Year FE Sample 3058 0.177 Yes All 3058 0.181 Yes All 3058 0.182 Yes All We find that having destaggered (as opposed to having had an annual board since 1996 or before) raises the probability of getting a special meeting proposal by 4.6 percentage points (beyond a baseline probability of about 7.7% for firmyears corresponding to companies that always had an annual board). This is the opposite of what the “buy peace” hypothesis predicts. Similarly, we find that having granted special meeting powers (as opposed to having offered such powers since 2005 or before) raises the probability of getting a special meeting proposal by 23 percentage points (beyond a baseline probability of about 3% for firm-years corresponding to companies that always provided for special meeting powers). In other words, in deciding which firms to target for special meeting (written consent) proposals, proponents disproportionately select—from among the firms that have annual boards (special meetings)—those that had more recently adopted annual boards (special meetings) as opposed to those that had instituted these governance arrangements earlier. Figure 3 focuses on firms that granted the right to call special meetings since 2005. It reports the fraction of firms that were targeted by a written consent proposal within one to five years after they instituted the right to call a special meeting. The figure shows that close to 40% of the firms received such a proposal by the end of the calendar year following such institution. Seemingly, a board cannot buy peace. If a board cannot buy peace, the existence of a pecking order has implications on the incentives of a board to resist proposals even if these proposals do not affect substantive shareholder rights. While the board may not care about the proposal at issue, resisting the proposal may delay the introduction of another proposal, further down the pecking order, that the board cares about. Similarly, with a pecking order, implementation of a proposal (or events increasing the likelihood of implementation) would have significance beyond the proposal at issue. It would speed up the time at which proposals further down the pecking order are introduced and, presumably, implemented. This cascading effect may contribute to the surprisingly large value effect associated with the passage of shareholder proposals.82 82 See Vicente Cuñat, Mireia Gine & Maria Guadalupe, The Vote Is Cast: The Effect of Corporate Governance on Shareholder Value, 67 J. FIN. 1943, 1943 (2012) (finding that “passing a proposal leads to significant positive abnormal returns”). Figure 3. Fraction of firms that granted special meetings targeted by a written consent proposal. 0 .2 .4 .6 Fraction of firms that granted SM targeted by a WC proposal 0 1 2 3 Years after the grant of power to call SM 4 5 Note: Sample includes only firms that did not allow WC by the time they granted right to call SM Far from supporting the “buy peace” hypothesis, the regressions in Table 4 indicate that companies that switched from staggered board to annual board and from no special meeting to special meeting (“switchers”) face a substantially increased likelihood of receiving the next proposal down the pecking order. One possible explanation for this pattern is that shareholder rights activists do not own stock in all companies in our sample. If shareholder rights activists are more likely to own stock in “switchers” than in non-switchers, one would expect a higher likelihood of switchers receiving the next proposal in the pecking order even if activists, within the set of companies in which they hold stock, do not disproportionately target switchers with such proposals. Such proponent-level effects almost certainly explain a portion of the higher likelihood of switchers receiving second venue written consent proposals, as 80% of the companies that instituted shareholders’ power to call special meetings did so after receiving a shareholder proposal. To examine the impact of proponent share ownership, we collected information about all types of shareholder proposals that were filed since 2001 by the four proponent groups that filed the largest number of written consent proposals.83 If a proponent group owned shares in a firm at a given point in time, we assume that the group held stock of any such company in all subsequent year in the sample period. However, we assume that the proponent group did not hold stock in such company in any prior year and did not hold stock in any company 83 Those proponents are members of the Chevedden family, the Rossi family, the Steiner family, or the Young-McRitchie family. for which no proposal was filed by the proponent group. We then rerun the regressions in Table 3, panel B including only the subset of firm-years in which at least one of the four proponent groups was deemed to own stock of a company. The results are reported in Table 4. The third and fifth columns of Table 4 respectively report the results of estimating specifications that augment those in columns (2) and (4), by including Had_Granted_SM_Spontaneously, a dummy that is equal to 1 for firm-years corresponding to firms that, as of that year, had granted their shareholders the right to call special meetings, but had done so spontaneously (i.e., without having ever received a special meeting shareholder proposal), and 0 otherwise. Table 4. Received Written Consent Proposal Existing_Venue_Power Had_Granted_SM (1) (2) (3) (4) (5) 0.109** 0.095** 0.091* 0.025 0.020 (0.043) (0.048) (0.048) (0.061) (0.061) 0.116** 0.117** 0.123** 0.100* 0.105* (0.053) (0.053) (0.053) (0.055) (0.056) 0.022 0.026 (0.049) (0.049) Get_Bd_Majority Create_Eff_Venue_Power Annual_Board 0.018 0.017 (0.057) (0.057) 0.062*** 0.056** 0.066 0.062 (0.023) (0.022) (0.064) (0.063) Had_Granted_SM_Spontaneously -0.060 -0.056 (0.054) (0.056) Change_Bd_Comp Create_PartEff_Venue_Power 0.091 0.092 (0.061) (0.060) -0.090 -0.092 (0.061) (0.061) Obs. 1014 1014 1014 1014 1014 R-squared 0.176 0.181 0.183 0.183 0.185 Year FE Yes Yes Yes Yes Yes Sample Held Top4 Proponents Held Top4 Proponents Held Top4 Proponents Held Top4 Proponents Held Top4 Proponents Standard errors (in parentheses) are clustered at the firm level. *** p<0.01, ** p<0.05, * p<0.1 Consistent with the “pecking order” hypothesis, in the first three specifications, non-switchers (companies that had provided for a power to call special meetings prior to 2005, or since the firm became public) face a significantly higher likelihood of receiving a written consent proposal than companies that did not provide such a power. In all specifications, however, switchers face a likelihood that is even higher. In addition, the coefficient for Change_Bd_Comp is no longer significant in any of the specifications in Table 4 (although the magnitude of the estimate is larger than in Panel B of Table 3). We further find that only switchers that had previously faced shareholder pressure (in the form of a known proposal) to institute a right to call a special meeting face such an increased likelihood. For switchers that did so spontaneously, we find that the likelihood of being targeted is statistically indistinguishable from the corresponding likelihood for firms that always allowed their shareholders to call special meetings.84 Moreover, when considering the identity of the proponent of written consent proposals and earlier special meeting proposals, we find that for 47% of the written consent proposals, the proponent had not previously filed a special meeting proposal before the same firm. While we cannot exclude the possibility that the regression results are due to a misspecification in the proponent ownership assumptions and stock ownership by shareholder rights activists may be correlated with each other, these results suggest that only a portion of the higher likelihood of receiving a second venue written proposal faced by switchers is due to proponent effects. That is, they suggest that proponents disproportionately target switchers within the set of companies in which they hold stock.85 Several possible company-level factors may explain such targeting. First, the same underlying reasons that induce a proponent to file a special meeting proposal or that induce a board to implement the proposal may induce the filing of the written consent proposal. That is, perhaps proponents selected companies for special meeting proposals because, in the proponent’s view, those companies would benefit most from enhanced shareholder rights or because those companies have a shareholder base most likely to support proposals—and selected companies for written consent proposals for the same reasons. Second, the very fact that the company acceded to an earlier demand may embolden proponents to make further demands—proponents may have “smelled blood.” Our data do not permit us to distinguish between these explanations. Based on the degree of care and sophistication proponents have shown in their targeting decisions in other respects, however, we consider it more likely that proponents may take a cue from the management’s reaction to earlier proposals rather than engage in a detailed ex ante analysis of companies and their shareholder base to inform their targeting decisions and determine which companies would benefit most from a proposal or which proposals are most likely to receive support. To the extent that the “smell blood hypothesis” is correct, acceding to shareholder proposals—say, to institute a right to call special meetings— becomes, from the perspective of a board that would prefer not to increase shareholder rights, increasingly costly. Acceding to a demand not only directly increases shareholders rights (the right to call special meetings); and not only, 84 The F-statistic for the null hypothesis that the coefficients for Had_Granted_SM and Had_Granted_SM_Spontaneously add up to zero has a p-value of 0.36 and 0.49, respectively. 85 In unreported analysis, we also examine whether, considering only the set of firm-years where we deem proponents to hold stock in companies, second venue written consent proposals disproportionately targeted firms with a high threshold for calling special meetings. Consistent with the results reported in column (7) of Table 3 and Figure 1, we find no evidence that they are so targeted. as a result of the pecking order, advances the date at which shareholder proposals seeking even more rights will be filed (i.e., the right to act by written consent). Beyond that, it may induce shareholder rights activists to target the company with such proposals earlier and more frequently than companies that did not face the initial shareholder proposal because they had for a long time granted the underlying right (to call special meetings). Figure 4. Presence of governance features across firms in the sample: 2005 vs. 2017 snapshots. An nu Sp al b oa ec ia l m rd W ee rit tin te g n co n Ve s nu ent e Ex pow er Re pan d m B ov oa e r w /o d c au Su Fil s pe l rm vac e an aj c or Ex ity ies pa nd by E / la Pa w rti ffec rem all tiv ov s ye e e ve + ffe fil nu ct l e ive po ve we nu r e po w er 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 2005 2017 CONCLUSION There has been a significant change in shareholder governance rights over the last twelve years. Figure 4 below compares the percentage of S&P 500 firms that were in our sample in 2005 to those in 2017 along several metrics. Along each, shareholder powers have increased, sometimes for a substantial fraction of firms. As other scholars have noted, in large firms annually elected boards have become the norm and staggered boards have become virtually extinct.86 But the trend towards venue powers, which began more recently, does not lag far behind: the percentage of firms that grant shareholders some venue power increased from 49% to 69%. Yet, among firms with venue power, the percentage of firms where the venue power was effective barely budged—from 60% to 66%—and this change is a byproduct, and in our view an unintentional one, of the large number of Delaware firms that destaggered and thereby automatically changed 86 See Catan & Klausner, supra note 2, at 8 (describing “massive wave of board destaggering” beginning in 2003); Marcel Kahan & Edward Rock, Embattled CEOs, 88 TEX. L. REV. 987, 1008 (2010) (predicting demise of staggered boards). the removal standard. Thus, shareholders of a majority of firms still lack an effective venue power, and shareholders of almost 40% of the firms still lack even a partially effective venue power. For a shareholder rights activist, this raises the question of where they should focus their energy. Having almost accomplished the goal of dismantling all staggered boards in large firms, should they continue seeking the power to call a special meeting and then the power to act by written consent? Or should they change course? One way to answer this question is to examine how the percentage of firms with effective venue powers would change if all firms in the 2017 sample changed just a single element of the governance structure—adopted an annually elected board or gave shareholders the ability to (1) call a special meeting, (2) act by written consent, (3) expand the board size, (4) remove directors without cause, (5) fill vacancies in the board, or (6) amend the bylaws by simple majority. To determine the effect of such changes, we focus on the set of firms that lacked the relevant governance feature as of the end of 2017. We then ask the question “if all those firms adopted that feature, what fraction of the firms adopting the feature would thereby grant their shareholders the effective (or the partially effective) power to call a venue?” Figure 5 provides the answer. For example, as of the end of 2017, 137 firms in our sample did not allow shareholders to call a special meeting. Out of those firms, granting the ability to call special meetings would only create an effective venue power for 46 firms, and would create a partially effective venue power for 75 firms.87 Thus, if activist shareholders followed a blanket policy of targeting every firm that does not currently allow their shareholders to call a special meeting, and management obliged in every case, only 33.6% (46/137) of those implementations would give shareholders an effective venue power (while 54.7% of the implementations would cause shareholders to gain a partially effective venue power). As one can see in Figure 5, the impact of activists’ pressure on the incidence of an effective (or partially effective) venue power would vary substantially across proposal types. If all firms that currently require supermajority requirements for bylaw amendments removed those requirements, only 9.5% of those firms would grant their shareholders an effective venue power as a result of that. By contrast, if all firms that did not allow their shareholders to fill vacancies gave their shareholders such a right, almost 45% of those firms would, as a result, grant their shareholders an effective venue power. 87 For twenty-six of the remaining firms, shareholders already had the power to call a venue through action by written consent. For sixty-five of the remaining firms, shareholders could not act by written consent nor expand/remove + fill. Figure 5. Fraction of firms lacking a governance feature where the grant of the feature would cause the shareholders to have effective/partially effective venue power. 0.6 0.5 0.4 0.3 0.2 0.1 cia lm ee tin g W rit te n co ns en t Ex pa nd Bo Re ar d m ov e w /o ca us e Fil lv ac Su an pe cie rm s aj or ity by la ws Sp e An nu al b oa rd 0 Effective Partially Effective Series1 Figure 5 indicates that, ceteris paribus, if semi-sophisticated activist shareholders’ ultimate objective is to increase the fraction of firms that give their shareholders an effective venue power, they should focus on proposals asking firms to enable their shareholders to fill board vacancies. Moreover, asking firms to enable shareholders to fill board vacancies would seem like a “bread and butter” issue with appeal to both governance professionals at institutional investors and retail shareholders. Directors are supposed to be elected by shareholders. While it may make practical sense to give the board a concurrent power to fill vacancies to obviate a requirement to hold a costly special meeting if a vacancy occurs midyear, denying shareholders that power would seem like a clear affront to a fundamental shareholder prerogative. On the other hand, if semi-sophisticated activist shareholders’ ultimate objective is to increase the fraction of firms that give their shareholders a partially effective venue power, their present focus on venue proposals makes sense and should be followed by seeking the right to remove directors without cause. But whether or not shareholder rights activists will turn next to removing restrictions on shareholder rights to fill vacancies or to the right to remove directors without cause, we are confident that their quest for more shareholder rights will continue. Appendix Table A1. Standard for Shareholder Action. ArkansasΔ California ConnecticutΔ Delaware Remove Directors w/out cause* Change Removal Std. Fill Vacancies Yes > No, § 4-27-808# Charter Yes, § 4-27810 Maj. Entitled, § 303(a) None Supermajority if staggered bd. None Yes > No, § 33-742 # Charter Maj. Entitled, § 141(k) None (SM by charter) GeorgiaΔ Yes > No, § 607.0808# Maj. Entitled, § 14-2-808 Charter Yes, § 708(c) None Yes, § 33712 Charter Yes, § 216 Charter or bylaw Charter Yes, § 607.0728 Charter Charter or sh. bylaw Yes, § 14-2728 Charter or sh. bylaw No power if staggered bd. Charter FloridaΔ Change Sh. Power to Fill Vac. # No power if staggered bd. Illinois Maj. Entitled, § 805 5/8.35# None (charter if staggered) Yes, § 805 5/7.60 None IndianaΔ Yes > No, § 23-1-33-8 # Charter Yes, § 23-130-9 [Charter or bylaw] LouisianaΔ Maj. Entitled § 12:81# until 2014, 12:1-808# Charter Yes, § 12:81 until 2014, 12:1-810 Charter or bylaw Maryland Board to 2/3 entl. or charter Yes, § 2-404 No power if staggered bd. Charter Maj. Entitled, § 2-406 Sh. opt-out of § 8.06, with 2/3 majority Massachusetts No Power (unless optout), § 8.06 Michigan Maj. Entitled, § 450.1511 Charter MinnesotaS Maj. Entitled, § 302A.223 Charter or sh. bylaw MissouriΔ Nevada New Jersey Only board Yes, § 450.1441 Board opt-in to § 3-804 Sh. optout of § 8.06, with 2/3 majority Charter Charter or sh. bylaw [Charter Yes, Maj. Entitled, § 351.315# Charter or bylaw or ^ § 351.265 bylaw] [Charter Charter to greater % Yes, or 2/3 Entitled, § 78.335 only § 78.330 bylaw. .3 35] Maj. Cast § 14A:6-6 Yes § 14A:6- [None, Charter 5 14A6-5] No power if staggered bd. Yes, § 302A.215 New YorkΔ No Power, § 706 N. CarolinaΔ Yes > No, § 55-8-08# Charter or sh. bylaw, § 706, 614 Maj. Entitled, § 1701.58 OhioS Charter (SM by sh. Yes, § 55-7bylaw) 28 Charter or sh. bylaw No power if staggered bd. None Maj. Entitled, 18 § 1027 Oklahoma Yes, § 705 None (SM by charter) No power if staggered bd. Charter Oregon Yes > No, § 60.324 # Pennsylvania Maj. Cast, § 1757 and Sh. bylaw § 1726 No power if staggered bd. Charter Rhode IslandΔ Yes, § 7-1.2-805 TennesseeΔ Yes > No, § 48-18-108# Charter Charter Charter # Texas Maj. Entitled, § 21.409 Charter or bylaw No power if staggered bd. Charter UtahΔ Yes > No, § 16-10a-808# VirginiaΔ Maj. Entitled, § 13.1-680# Charter Washington Yes > No, § 23B.08.080# Charter WisconsinΔ Yes > No, § 180.0808# Charter or bylaw Charter [No, if created by removal, 705(b)] Charter Yes, § 1701.55 No, if created by removal, 1701.58( E) [Yes], 18 § 1068 [Charter or bylaw] Yes, § 60.241 Charter Yes, § 1758 [Charter or sh. bylaw] [Yes], § 71.2-804 Yes, § 4817-206 Yes, § 21.239 Yes, § 1610a-728 Yes, § 13.1669 Yes, § 23B.08.280 Yes, § 180.0728 [Charter] Charter None, 21.410 Charter Charter Charter Charter * Table provides passage rule for removal of entire board or of majority of board, whichever is lower. # Meeting must be called for purpose of removal. Δ Special rules apply to for companies with cumulative voting. @ Unless company opted out of § 8.06, shareholders cannot increase board size. S Special limitation for actions by written consent. Table A2. State Indiana Oklahoma Massachusetts Maryland Nevada, Texas Pennsylvania Provisions Default rule that shareholders cannot amend bylaws Default rule that shareholders cannot amend bylaws Unless opt out, shareholders cannot increase board size; board can opt-in to supermajority requirement for bylaw amendment Board can opt into provision that prohibits shareholders from increasing board size Board may be able to impose supermajority requirement for bylaw amendments through its power to amend bylaws Bylaw provision authorizing removal does not apply to incumbent terms Table A3. Special Meeting and Written Consent. Special Meetings How Options Changed Arkansas Yes, 4-26Charter or 10% or 701 Bylaw lower California 10%, 600 Charter or Lower % Bylaw only Connecticut 35%** Charter or Lower % Bylaw only Delaware No, 211 Charter or Any Bylaw Florida 10% Charter Any % up (request), to 50% 702 Georgia 25%, 702 Charter or Any Bylaw Illinois 20%, 7.05 Charter or Lower % Bylaw only Indiana No, 29-2 Charter or Any Bylaw Louisiana Yes, 12:73 Charter or 20% or Bylaw lower Maryland 25% Charter or Any % up (request), 2- Bylaw, to 50%; 502 MUTA 50% if by election MUTA88 Massachusetts 40%, 7.02 Charter or Any Bylaw Michigan No, 1403 Bylaws Any 10% (25% Charter or Lower % MinnesotaS if takeover Bylaws only related) (request), .433 Missouri No, 225 Charter or Any Bylaw Nevada No, 310 Charter or Any [?] Bylaw New Jersey No, 5-3 Bylaws Any New York No, 602 Charter or Any Bylaw North No, 7-02 Charter or Any Carolina Bylaw OhioSS 25%, Charter or Any % up 1701.40 Bylaw to 50% Default Written Consent How Options Changed No, 4-26- Mandatory No 710 Yes, Charter Yes or No 603^^ No, 33-698 Charter Yes or No Default Yes, 228 Charter Yes or No Yes, 704 Charter Yes or No No, 704 Charter Yes or No^ Yes, 7.10## Charter Yes or No No, 29-4 Mandatory No No, 12:76 Charter Yes or No No, 2-505 Charter Yes or No No, 704 Charter Yes or No No, 1407 No, 411 Charter Mandatory Yes or No No No, 273 Mandatory No Yes, 315 Yes or No No, 5-6 No, 615 Charter or Bylaw Charter Charter No, 7-04 Mandatory No No, 1701.54 Mandatory, 1701.52 No Yes or No Yes or No 88 The Maryland Unsolicited Takeovers Act (“MUTA”) gives the board the right to elect to become subject to a set of changed rules unless the charter precludes this election. See MD. CODE ANN. CORPS. & ASS’NS § 3-802(c) (West 2018) (“The charter of a corporation may contain a provision or the board of directors may adopt a resolution that prohibits the corporation from electing to be subject to any or all provisions of this subtitle.”). Ohio (bylaw changes) Oklahoma No, 1059 Oregon No, 204 Pennsylvania No, 2521 [1755] Rhode Island No, 7-1.2701 10%, 17202 Charter or Bylaw Charter or Bylaw Texas 10%, 352 Charter or Bylaw Utah 10%, 701 Bylaw Virginia No, 680 Washington 10%; 020 Charter or Bylaw Charter or Bylaw Wisconsin 10%, 702 Tennessee Charter or Bylaw Charter or Bylaw Charter Charter or Bylaw Charter or Bylaw Any Yes (2/3 vote), 1701.11 No, 1073 Any No, 211 Charter Change to any % > 50%. No (since 2010) Yes or No Any; any % > 25 (if after 6/15). Any No, 2524 [1766] Charter Yes or No No, 7-1.2- Charter 707 No, 17-104 Charter Yes or No No, 6-202 Charter Yes or No Yes (for cos. not exist. in 1992) No, 657 Charter Yes or No# Charter Yes or No No, 040 Mandatory No No, 704 Charter Yes or No^ Any % (higher % by charter only) Any % up to 50% (higher % by charter only) Lower % only Any Any % (higher % by charter only) Lower % only Mandatory Yes or No * If cumulative voting, unanimous written consent required for election or removal of directors. ^ If cumulative voting, unanimous written consent required for election of directors. ** 10% for companies with 10% shareholder on 2/1/88. # Not for election of directors. ## Notice for all non-consenting shareholders required before and after written consents obtained. ^^ Filling of vacancies created by removal requires unanimous written consent. section 305(b). ^^^ 25% or lower if to facilitate business combination. ^^^^ Removal requires unanimous written consent. Table A4. Variable Definitions. Variable name Received_Special _Meeting_Proposalit Received_Written _Consent_Proposalit Received_Venue_Proposalit Get_Bd_Majorityit Existing_Venue_Powerit Create_Eff_Venue_Powerit Annual_Boardit Change_Bd_Compositionit Create_Part_Eff _Venue_Powerit Had_Destaggeredit Definition Dummy equal to 1 if in connection with its annual meeting of shareholders for year t firm i received a precatory proposal requesting that shareholders be granted the power to call special meeting (regardless of whether the proposal was voted on or not), and 0 otherwise. Dummy equal to 1 in connection with its annual meeting of shareholders for year t firm i received a precatory proposal requesting that shareholders be granted the power to act by written consent (regardless of whether the proposal was voted on or not), and 0 otherwise. Dummy equal to the maximum between Received_Special_Meeting_Proposalit and Received_Written_Consent_Proposalit. Dummy equal to 1 if as of the moment firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t, firm i’s shareholders had the explicit or latent power to obtain a board majority in a single venue (by “expanding + filling” or “replacing + filling”), and 0 otherwise. Dummy equal to 1 if as of the moment firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t, firm i’s shareholders had the explicit or latent power to call a venue (i.e. either act by written consent, or call a special meeting, or both), and 0 otherwise. Dummy equal to the product between Get_Bd_Majorityit and (1 - Existing_Venue_Powerit). Dummy equal to 1 if one of the following conditions held as of the moment firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t: (1) all the directors of firm i were elected annually; (2) even if the board of directors of firm i was classified, firm i had amended its governing documents to adopt a phased-in board declassification; (3) even if the board of directors of firm i was classified, the shareholders of i had the latent power to declassify the board; and equal to 0 otherwise. Dummy equal to 1 if as of the moment firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t, firm i’s shareholders had the explicit or latent power to change the composition of i’s board of directors (by “expanding + filling,” “removing + filling” or just removing directors), and 0 otherwise. Dummy equal to the product between Change_Bd_Compositionit and (1 Existing_Venue_Powerit). Dummy equal to 1 if by the time firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t firm i had switched from having a classified board structure to having an annual board structure (this includes the case in which, as of t, firm i was in the course of fully implementing a phased-in board declassification), and 0 otherwise. Had_Granted_SMit Had_Granted_SM _Spontaneouslyit Note: for firms that always had an annual board structure, this dummy is equal to zero for all years t. If a firm had an annual board structure for every year since 1996, we assume that it always had an annual board structure. Dummy equal to 1 if by the time firm i distributed its proxy statement for the annual meeting of shareholders corresponding to year t firm i had switched from not allowing its shareholders to call special meetings to allowing its shareholders to call special meetings, and 0 otherwise. Note: for firms that always allowed their shareholders to call special meetings, this dummy is equal to zero for all years t. If a firm allowed its shareholders to call special meetings for every year since 2005, we assume that it always allowed its shareholders to call special meetings. Dummy equal to 1 if Had_Granted_SMit=1 and by the year firm i granted its shareholders the right to call special meetings firm i had not received a precatory proposal asking the firm to grant such a power.