Activism and the Move toward Annual Director Elections

The following post comes to us from Tom Nohel, Associate Professor of Finance at Loyola University, and is based on a Conference Board Director Note by Mr. Nohel, Re-Jin Guo of the University of Illinois at Chicago, and Timothy Kruse of Xavier University. Work from the Program on Corporate Governance on staggered boards includes The Powerful Antitakeover Force of Staggered Boards: Theory, Evidence & Policy by Bebchuk, Coates and Subramanian, and Staggered Boards and the Wealth of Shareholders: Evidence from Two Natural Experiments by Bebchuk, Cohen and Wang.

Staggered boards, a structure under which the board is divided into classes, with one class of directors standing for re-election annually, are perhaps the most consequential takeover defense. They also are a favorite target of activist shareholders and governance experts. The effect of collective pressure to move to annual elections for all directors has been dramatic: the proportion of public companies with staggered boards has fallen from 60 percent in 2002 to less than half in 2011 (and less than one-third among S&P 500 companies). Non-binding shareholder proposals have been an important catalyst in the move away from staggered boards. [1] More recently, activist hedge funds have emerged as an alternate and better-financed vehicle to channel shareholder displeasure. This report documents the extent to which activism of any type continues to push corporations to implement annual director elections and compares the influence different forms of activism have had on this governance practice.

The Power of a Staggered Board

As discussed in Director Notes earlier this year and elsewhere, staggered boards and poison pills continue to figure prominently in the battle for shareholder democracy. [2] No poison pill that is meant to deter a hostile takeover has ever been triggered, and since it is the board of directors, rather than the shareholders, that controls the implementation of most poison pills, the ability to influence the board is paramount. [3] While poison pills and staggered boards are powerful deterrents in isolation, the use of these defenses in tandem creates a veritable fortress against would-be suitors.

A poison pill can often be installed or removed by board decree. However, gaining control of a staggered board to be able to remove the pill requires winning elections at two separate annual shareholders’ meetings, which can be prohibitively expensive and time-consuming. In fact, no prospective suitor has ever gained majority control of a staggered board by voting out incumbent directors. The difficulty of fighting through a staggered board was recently brought to the fore during Air Products & Chemicals Inc.’s closely watched hostile takeover battle for Airgas, Inc. (See p. 6 for more discussion of the implications of this battle and the associated decisions by the Delaware courts.)

The Prevalence of Staggered Boards and the Push to De-Stagger

As one of many responses to the wave of hostile takeovers, firms adopted staggered boards in the 1980s, usually with shareholder approval. More recently, firms have been reluctant to ask shareholders to approve proposals to stagger director terms. One exception is taking advantage of investor excitement by going public with a staggered board already in place. [4] For example, recent initial public offerings (IPOs), such as those of LinkedIn Corp, Zillow, Inc., and Pandora Media, Inc., have staggered boards. Though shareholder activists stepped up their criticism of staggered boards in the early 1990s, only 62 firms with staggered boards decided to reinstate annual director elections between 1988 and 2002. As a result, in 2002, roughly 60 percent of public companies still had staggered boards. However, around that time, the resistance to shareholder activism broke down, as the landscape shifted dramatically toward shareholder empowerment. In particular, the bursting of the internet and telecom bubbles in 2000 and the exposure of criminal misconduct at the likes of Enron, WorldCom, and Tyco International, led shareholders and politicians to demand increased accountability. Legislatively, the response was the Sarbanes-Oxley Act of 2002, which set the tone for the remainder of the decade. This trend toward shareholder empowerment has intensified with the possibility of proxy access for outside shareholders and mandated shareholder say-on-pay votes at all public companies, as dictated by the Dodd-Frank Act in 2010.

While there were shareholder proposals to de-stagger boards throughout the late 1990s, the effectiveness of such proposals gained traction following Sarbanes-Oxley. Chart 1 shows Georgeson voting data on shareholder proposals to de-stagger the board. [5] Georgeson identified 506 shareholder proposals during the period 1996 to 2010, 479 of which went to a vote. Overall, the trend has been steadily increasing support for the proposals. Indeed, 65 percent of the votes cast since the passage of Sarbanes-Oxley have been in favor of de-staggering the board.

The changing impact of shareholder proposals is perhaps best seen in the cases of Bristol-Myers, Host Marriott, Merck, and Proctor & Gamble. Each received shareholder proposals to de-stagger their boards for at least 16 consecutive years, including majority shareholder support for the last several proposals. Even so, it was only after the events of 2000 to 2002 that these firms relented to shareholder pressure, inducing them to de-stagger their boards in 2003 or 2004.

As shown by Chart 2, the corporate response to this call for accountability is rather striking. From 1988 to 2002, 62 firms eliminated their staggered boards. From 2003 to 2010, 467 firms put forth management proposals to institute annual director elections or simply implemented annual elections with a board vote. During the 2011 proxy season, shareholder and management proposals to de-stagger the board garnered considerable shareholder support, topping 70 percent, [6] making it appear likely that this trend will continue.

Perhaps not coincidentally, the last decade has also seen a remarkable increase in activism by hedge funds. These largely unregulated pools of capital not only seek governance and other changes at targeted companies, but they also seek to profit from those changes, given the large stakes that they hold. A previous study of activist hedge funds between 2001 and 2006 reported that, of the 1,059 events studied, activists specifically called for governance changes in more than half (548). [7]

Our previous study found shareholder pressure, in the form of shareholder proposals to de-stagger, present in about 34 percent of the instances preceding a management proposal to de-stagger. [8] We did not consider the role of activist hedge funds, but it is unlikely to have been sizeable until near the end of our sample period. In contrast, of the 467 de-staggering events documented since 2002, 42 percent faced shareholder pressure in the form of non-binding shareholder proposals to de-stagger, reaching a peak of 68 percent in 2009. Moreover, 27 percent of the de-staggering firms had prior Schedule 13D filings from activist hedge funds, two-thirds of which specifically stated objectives not supported by management, and several which specifically targeted the elimination of an existing staggered board. Thus, it would seem that shareholders are ratcheting up the pressure and companies are relenting more frequently, but this is not the full story.

Implementation of the Decision to De-stagger

The move toward annual elections is, in part, driven by a sense that this is the structure preferred by shareholders. The strong votes in support of proposals to de-stagger buttress this point. Many institutional shareholders and proxy advisory services specifically state that directors should be accountable to shareholders on an annual basis. In the current climate of shareholder empowerment, boards that ignore the wishes of shareholders are, at a minimum, likely to face considerable shareholder angst, and potentially an outright revolt, such as a “vote no” campaign or a push to unseat them. [9] Therefore, rather than directly resist calls to de-stagger the board, directors have instead often chosen to drag out the implementation of the move to annual elections.

If there is a push for annual director elections, the entire board could all resign and then put themselves up for election all at once (for one-year terms)—an “immediate” declassification of the board. At the other extreme, boards that have submitted a proposal to de-stagger can still put only the current class of directors up for threeyear terms, with a promise to put all subsequent directors up for annual election as their terms expire—a “delayed phased” implementation—since it will be a full three years until the entire board is forced to submit to annual elections. As shown in Chart 3, as the number of firms choosing to de-stagger has increased, the preference has drifted further toward “delayed phased” implementation. Since there are no legal conditions preventing the immediate move to annual elections, any proposed delay is clearly an indication of reluctance on the part of directors to move to annual elections. A typical reason given for the phased-in approach is to smooth the transition to the new method of electing directors. However, there are few, if any, logistical issues involved in transitioning to annual director elections and, as noted, no legal constraints. We conclude that these instances likely represent foot-dragging on the part of directors.

In 2003 and 2004 (and earlier), only a handful of firms chose to drag out the implementation of annual elections to the fullest, while in 2009 and 2010, more than 60 percent of de-staggering firms chose to do so. [10] A more careful analysis suggests that the delayed implementations of the move to annual elections tend to be cases of a response to a shareholder proposal to de-stagger (or multiple proposals to de-stagger over several years). Some proposals, such as one filed by Gerald Armstrong, a vocal advocate for governance reform, at OGE Energy Corp. ask that “the Board declassification shall be completed in a manner that does not affect the unexpired terms of the previous-elected Directors.” [11] Others call for the immediate declassification of the board.

In contrast, in cases in which activist hedge funds are pushing for governance changes, implementation of annual director elections is usually quick, with 76 percent of firms electing the entire board within one year. In contrast, 66 percent of firms de-staggering following shareholder proposals take two or three years to allow the election of the entire board. In other cases, continued shareholder pressure induced a target to alter its implementation plans. Morgan Stanley shareholders approved a delayed phased proposal at the March 15, 2005, annual meeting. However, following continued pressure on CEO Philip Purcell from dissident shareholders, the company agreed to make a proposal accelerating the de-staggering process at the 2006 meeting. [12] The 2006 proxy also contained proposals to eliminate certain supermajority provisions. In the case of a targeted firm that has agreed to an annual director election, a shareholder proposal to hasten implementation might be excludable from the proxy on the grounds that it is not a material change from existing arrangements.

Pros and Cons of a Staggered Board

As argued earlier, the staggered board is most powerful when combined with a poison pill. This combination of defenses is very effective at deterring frivolous bids or in strengthening the board’s negotiating position in a prospective deal, but unfortunately it is also a powerful arrow in the quiver of entrenched managers and/or directors to enable them to stay in power despite shareholder objections. The academic literature in law and finance has also weighed in at length on this topic. [13] The arguments for and against staggered boards tend to highlight one side of this divide (see “Avista Corp. Arguments for and against Staggered Boards”).

Avista Corp. Arguments for and against Staggered BoardsConsiderations Favoring a Classified Board

  • Classification of the Board tends to balance experience, continuity and stability with the regular opportunity to add valuable, fresh perspectives.
  • It takes several years for a new director to become fully conversant in the complexities of the utility business model.
  • Classification makes it more difficult and time consuming to change majority control of the Board which reduces the vulnerability of the Company to an unsolicited takeover proposal. Thus, classification may encourage persons attempting certain types of transactions that involve an actual or threatened change of control of the Company to first seek to negotiate with the Company and may discourage pursuit of such transactions on a non-negotiated basis.

Considerations Against a Classified Board

  • Classification of the Board could make more difficult or discourage the removal of incumbent directors, through a proxy contest or otherwise, and the assumption of control by a holder of a substantial block of the Company’s common stock, and could thus have the effect of entrenching incumbent management.
  • Classification could have the effect of discouraging a third party from making a tender offer or otherwise attempting to obtain control of the Company, even though such an attempt might be beneficial to the Company and its shareholders.
  • Some institutional shareholders and commentators argue that classification reduces director’s accountability to shareholders, since such a structure does not enable shareholders to express a view on each director’s performance by means of an annual vote. The Board does not agree with this argument.

Sample Characteristics

We identify firms that chose to de-stagger their boards between 2003 and 2010, focusing on the role of shareholder activism in the implementation of the events. We draw on several sources to compile our sample. The initial sample firms are identified using governance databases and reports available from the Investor Responsibility Research Center (IRRC), RiskMetrics, and Georgeson. We supplement the sample by searching the Dow Jones Newswire (Factiva) and Lexis-Nexis with the key words “declassification,” “de-staggering,” “declassify,” “de-stagger,” and “annual election of directors.” The final sample consists of 467 firms and the de-stagger events are distributed relatively evenly over the sample period.

Information regarding the de-staggering proposals is collected for each sample firm from proxy statements filed with the U.S. Securities and Exchange Commission (SEC) in the year of the decision to de-stagger and from press reports. Information is also collected from proxy statements about concurrent management and shareholder proposals regarding other takeover defenses and CEO and officer and director share ownership. In terms of implementation, the sample is almost evenly split between cases in which annual director elections are instituted quickly, and cases in which the process is dragged out. In terms of concurrent proposals, typical management proposals include eliminating supermajority voting provisions (60 cases), allowing shareholders the right to call a special meeting (12 cases), and instituting majority voting for directors (19 cases). In seven cases, management called for the elimination of cumulative voting for directors. This latter proposal can be seen as an attempt to minimize the impact of the de-staggering decision because cumulative voting allows shareholders to concentrate a large number votes on outside candidates in the event of a proxy fight. The most common shareholder proposals usually call for majority voting in director elections (26 cases), elimination of supermajority voting provisions (16 cases), and the elimination of poison pills (9 cases).

Summary statistics on the specific governance features of the sample firms are provided in Table 1. These figures do not suggest anything unusual. On average, CEOs and officers and directors hold 3 percent and 9 percent of their firm’s shares, respectively. The CEO is also the board chair at 62 percent of the sample firms.

Data on the degree of entrenchment of the sample firms and industry firms are collected using the E-index of Bebchuk, Cohen, and Ferrell as the entrenchment index. [14] The E-Index assigns a score of zero to six, based on the presence of six variables:

  • 1. staggered board;
  • 2. poison pill;
  • 3. limits to shareholder bylaw amendments;
  • 4. supermajority requirement for mergers;
  • 5. supermajority requirement for charter amendments; and
  • 6. golden parachutes.

The sample firms typically are well protected with 3.5 to 4 out of the six defenses, which is significantly greater than their industry peers at the 1 percent level (either based on means or medians).

Information is collected on the incidence of shareholder activism at the de-staggering firms. Specifically, data are collected from Georgeson, RiskMetrics, and proxy statements on the incidence of shareholder proposals calling for annual elections in the three years leading up to the de-stagger announcement. Schedule 13Ds and news articles are also examined for evidence of shareholder activism targeted at the sample firms. The SEC requires investors acquiring a stake of 5 percent or greater with an intent to influence management to file a Schedule 13D within 10 days of crossing the 5 percent threshold (if there is no intent to influence management, 5 percent shareholders must file a Schedule 13G).

We are focusing on Item 4 of Schedule 13D, which details the investor’s purpose in buying the securities. In many cases, the language is simply boilerplate: for example, the investor states that it bought the shares because it views them as undervalued, and it might informally contact the target management. In other cases, shareholders take a much more active role including writing letters, attending board meetings, making shareholder proposals, or even running proxy fights for board seats. A frequent outcome of this activism is that the board eventually decides to de-stagger.

A wide array of accounting data is collected for the sample firms. The figures are industry-adjusted by deducting the mean (or median) figure based on the three-digit Standard Industrial Classification (SIC) code. These industry-adjusted figures are reported in Panel E of Table 1. Overall, the sample firms tend to significantly out-perform their industry peers at the 1 percent level in terms of both earnings before interest, taxes, depreciation, and amortization (EBITDA) to assets, and profitability. This superior profitability seems to be driven by higher asset turnover rather than by fatter profit margins. The sample firms are noticeably more leveraged than their industry peers, and appear to be healthier in terms of current and quick ratios. [15]

Recent Trends in the Battle over Staggered Boards

The debate over staggered boards was front and center during the recent battle between rivals Airgas and Air Products & Chemicals. Air Products had been in negotiations to acquire Airgas since late 2009 and finally went public with a hostile offer at $60 per share in February 2010. Airgas held firm, rejecting this and all subsequent offers, protected by its staggered board and poison pill. Air Products even won a proxy fight at the September 2010 annual meeting, which placed three candidates on Airgas’s board and passed a shareholder proposal to move Airgas’s next annual meeting from September 2011 to January 2011. Moving up the annual meeting would have greatly reduced the time that Air Products would have had to wait until potentially gaining control of Airgas’s board, significantly reducing the effectiveness of staggered boards as a powerful takeover defense.

Airgas filed a motion against Air Products regarding the attempt to compress the annual meeting calendar. On October 8, 2010, Chancellor William B. Chandler of the Delaware Chancery Court ruled in favor of Air Products, stating that Airgas’s bylaws and charter were ambiguous regarding whether the next election had to be a full year after the previous shareholder meeting or simply in the next calendar year. [16] However, the Delaware Supreme Court overturned that ruling on November 23, 2010. The full board, including the new directors nominated by Air Products, in December 2010 unanimously voted against Air Products’s final bid of $70 per share, calling it inadequate. In February 2011, the Chancery Court rejected Air Products’s challenge to Airgas’s poison pill. [17] Air Products abandoned its bid for Airgas on February 15, 2011, [18] seeming not to have the patience and/or resources to see the bid through to the next annual meeting, slated for September 2011. In conclusion, Air Products failed to convince the courts to reduce the effectiveness of either part of the staggered board/poison pill combination.

Sensing tremendous shareholder pressure for director accountability, firms hoping to maintain their staggered boards in the face of fierce opposition have sought—and, in certain cases, received—help from state legislatures. Specifically, some states, including Indiana, Iowa, and Oklahoma, have adopted laws meant to shield local firms from potential hostile offers by mandating that firms chartered in their states maintain staggered terms for directors. There is some evidence that firms incorporated in the states adopting mandatory classified boards have been highly influential in the process, most notably Chesapeake Energy in Oklahoma. [19] In response, shareholder activists have submitted proposals requesting the target firm to re-incorporate to a state without a law requiring a classified board. [20]


As part of a general trend toward improving corporate governance following the corporate scandals in the early 2000s, an increasing number of companies have moved to de-stagger their boards, requiring their directors to stand for annual re-election. An analysis of de-staggering events between 2003 and 2010 shows that, overall, 60 percent of the companies that decided to de-stagger their boards following the passage of Sarbanes-Oxley have done so in response to some form of shareholder pressure. In addition, the findings show that, when pressured by activist hedge funds, companies tend to respond by de-staggering their boards far more quickly and, in some cases, immediately, while in cases in which a hedge fund is not the main catalyst, companies move much more slowly to implement annual director elections.


[1] See Re-Jin Guo, Timothy Kruse, and Tom Nohel, “Undoing the Powerful Anti-Takeover Force of Staggered Boards,” Journal of Corporate Finance, 14 2008, pp. 274-288.
(go back)

[2] See Andrew L. Bab and Sean P. Neenan, “Poison Pills in 2011,” Director Notes, Vol. 3, No. 5, March 2011, for a discussion of recent cases testing poison pills. See also Georgeson, Inc’s 2010 Annual Corporate Governance Review, which notes that shareholder proposals to repeal staggered terms for directors and proposals to redeem poison pills are among the most common and most popular with shareholders, each garnering close to 70 percent shareholder support over the period 2006-2010.
(go back)

[3] A potential lone exception is the case of Versata v. Selectica, in which Versata intentionally triggered the activation of Selectica’s pill and then challenged its validity in court. However, Selectica’s pill was an net operating loss shareholder rights plan (“NOL poison pill”) with a 4.99 percent trigger, meant to deter transactions that would result in the limiting of Selectica’s ability to make use of their sizeable amount of NOL carry-forwards. These NOL-based pills are not particularly meant to deter possible suitors interested in acquiring the adoptees of the pills. See also Bab and Neenan, “Poison Pills in 2011.”
(go back)

[4] Robert Daines and Michael Klausner, “Do IPO Charters Maximize Firm Value? Antitakeover Protection in IPOs,” Journal of Law, Economics and Organization, 17, April 2001, pp. 83–120.
(go back)

[5] Georgeson typically surveys annual meetings of 1,500 companies during the first half of each calendar year, when most meetings are held, and provides an overview of the issues and voting results in their “Annual Meeting Season Wrap-Up” (1996 to 2001) and “Annual Corporate Governance Review” (2002 to 2010).
(go back)

[6] Ted Allen, et al., “2011 U.S. Postseason Report,” Institutional Shareholder Services, September 29, 2011.
(go back)

[7] See Alon Brav, et al., “Hedge fund activism, corporate governance, and firm performance,” Journal of Finance, 63 2008, pp. 1729–1775.
(go back)

[8] Guo, Kruse, and Nohel, “Undoing the Powerful Anti-Takeover Force of Staggered Boards,” 2008.
(go back)

[9] See Diane Del Guercio, Laura Wallis, and Tracie Woidtke, “Do Boards Pay Attention When Institutional Investor Activists ‘Just Vote No’?,” Journal of Financial Economics, 90, 2008, pp. 84–103.
(go back)

[10] Moreover, the average period until the full board faces annual elections increased from about one year in 2003 to well over two years in 2009 and 2010.
(go back)

[11] OGE Energy Corp. proxy statement, filed March 31, 2009, p. 37,available at (
(go back)

[12] Joann S. Lublin, Randall Smith, and Ann Davis, “Morgan Stanley Directors Endorse Embattled Purcell,” Wall Street Journal, May 2, 2005, p. C1.
(go back)

[13] See John Wilcox, “Two Cheers for Staggered Boards,” Corporate Governance Advisor, 10, 2002, pp. 1–5; Lucian Ayre Bebchuk, John Coates IV, and Guhan Subramanian, “The Powerful Antitakeover Force of Staggered Boards: Theory, Evidence, and Policy,” Stanford Law Review, 54, 2002, pp. 887-951; Olubunmi Faleye, “Classified Boards and Long-Term Value Creation,” Journal of Financial Economics, 83, 2006, pp. 501–529; Thomas Bates, David Becher, and Michael Lemmon, “Board Classification and Managerial Entrenchment: Evidence from the Market for Corporate Control,” Journal of Financial Economics,” 87, 2008, pp. 656–677; and Lucian Ayre Bebchuk, Alma Cohen, and Charles Wang, “Staggered boards and the wealth of shareholders: Evidence from a natural experiment,” Harvard University working paper, 2011.
(go back)

[14] Lucian Bebchuk, Alma Cohen, and Allen Ferrell, “What Matters in Corporate Governance?” Review of Financial Studies, 2009, pp. 783–827. These data are available for all firms covered by IRRC on Bebchuk’s website (
(go back)

[15] The current ratio is current assets divided by the current liabilities. The quick ratio is current assets minus inventories divided by current liabilities.
(go back)

[16] Airgas, Inc. v. Air Products & Chemicals, Inc. C.A. No. 5817, Del. Ch., October 8, 2010 (
(go back)

[17] Air Products and Chemicals, Inc. v. Airgas, Inc., C.A. No. 5249 Del. Ch., February 15, 2011 (
(go back)

[18] Air Products press release, “Air Products Withdraws Offer for Airgas,” February 15, 2011 (
(go back)

[19] Daniel Gilber, “Oklahoma Board Rule Benefits Chesapeake,” Wall Street Journal, July 11, 2011.
(go back)

[20] Wellpoint Inc.’s April 2, 2010, proxy statement included a shareholder proposal submitted by AFSCME requesting that Wellpoint move its state of incorporation from Indiana to Delaware, citing 2009 amendments to Indiana corporate law requiring Indiana-incorporated firms to have a staggered board. AFSCME filed a similar proposal at WellPoint in 2011.
(go back)

Both comments and trackbacks are currently closed.

2 Trackbacks

  1. […] full article……..via Activism and the Move Toward Annual Director Elections — The Harvard Law School Forum on Corporate…. Share OptionsPrintTwitterEmailMoreFacebookLinkedInStumbleUponRedditDiggTumblrLike this:LikeBe the […]

  2. […] by many companies in the 1980s — usually with shareholder approval, according to a December 2011 report by Aguilar and The Conference […]