The Securities and Exchange Commission is reportedly considering whether to adopt a shareholder access proposal, making possible shareholder election of one or more directors to supplant management nominees on corporate boards. Proponents of such proposals argue that granting shareholders these rights fosters “shareholder democracy” or will improve corporate performance. But corporations are institutions with limited economic purposes; they are not like political communities, in which major differences in values and interests are resolved by voting. Corporate law, moreover, requires that directors owe their loyalty to the corporation itself, and does not permit them to represent the specific interests of shareholders the way constituents are represented in a democracy. Finally, and most important, we rely on corporations for the productivity essential to economic growth. Shareholder access proposals are more likely to impair corporate performance than enhance it.
In a late October article in the Wall Street Journal, former Securities and Exchange Commission (SEC) chairman Arthur Levitt Jr. compared voting by shareholders for corporate directors to citizens voting in political elections:
Ever since the recount of 2000, partisans of both parties have paid particular attention to everything from who votes to how they vote and how their preferences are recorded. Counting every vote is not only integral to our political life; it is central to our economic life as well. Shareholder capitalism enables our markets to thrive, our companies to grow, and our economy to remain strong. And central to this system is the principle that shareholders can have a voice in the running of the companies that they own, that their votes will count.
We rely on corporations large and small for the growth and productivity of our economy--hence, their specialized form. The policy question, then, is not whether shareholder voting will vindicate someone’s notion of the ideal social order, but whether giving shareholders a greater voice in the running of corporations will improve corporate performance. If the answer is no, are we being asked--for no better reason than a weak analogy--to apply to corporate governance a concept that has value in a political context for entirely different purposes?
Levitt’s purpose in using this analogy was to persuade the SEC to adopt a rule that would make it easier for shareholders to nominate and elect directors to corporate boards. Such a rule was most recently proposed by former SEC chairman William Donaldson, but was abandoned after widespread corporate opposition. Now the idea is back on the SEC’s agenda because of a court decision that questioned the SEC policy of requiring all contests for board seats to be conducted under the proxy rules. Proxy contests are expensive and are uneconomical for shareholders who want to elect only one or two directors to a corporate board, so institutional investors that might gain special benefits from board representation have sought ways to get their nominees included in the company’s own annual meeting proxy statement. Initiatives of this kind are known as “shareholder access proposals,” and are frequently carried on under the claim that they would be consistent with what is called “shareholder democracy.” The idea that a corporation is a democracy--as a country might be considered a democracy--makes for an attractive slogan, but it misconstrues the nature of both democracies and corporations. If the SEC adopts this notion--permitting shareholder groups to elect directors to corporate boards--it will do serious damage to corporate governance and, over time, serious damage to the public companies that are engines of growth in our economy.
The Purpose of Democracy
In The Constitution of Liberty, F. A. Hayek wrote: “Democracy is a means rather than an end. . . . Whenever it is necessary that one of several conflicting opinions should prevail and when one would have to be made to prevail by force if need be, it is less wasteful to determine which has the stronger support by counting numbers than by fighting. Democracy is the only method of peaceful change that man has yet discovered.” Thus, democracy is a procedure for resolving the serious disputes over values and interests that exist in any political community. Is it, then, applicable or useful in a corporate context?
For institutions in which resolving disputes is neither necessary nor appropriate to the institution’s purpose, democracy--meaning the right of members to vote for their leaders or for specific policies--generally does not exist. Militaries are not democracies. Neither are most charitable, educational, religious, or nonprofit organizations. With rare exceptions, all of these are hierarchical in structure, and for the most part their governing organizations are self-
perpetuating: the members do not vote to elect the trustees or directors that set the organization’s policies. There are two reasons for this. First, when an organization is formed for a particular purpose, those who join it are volunteers; they have accepted its purpose. Second, should any person who is a member of such an organization come to disagree with its purposes or management, he is free to leave. All of these things distinguish these institutions from political communities, which include people who were born within the community’s geographic boundaries, may be coerced into obeying its rules, and for whom leaving the community involves considerably more than resigning a membership or changing a job. Democracy, then, seems efficacious for political communities, but not for organizations in which disputes about purpose either do not occur or do not require a system for peaceful resolution to avoid violence.
Business corporations--that is, corporations organized for production, efficiency, and ultimately profit--are clearly in the same category as single-purpose institutions like armies, churches, or charities. They bear little resemblance to the geographically defined, coercive political communities where democracy is necessary to resolve differences. Thus, business corporations in mutual form--including many insurance companies and some banking organizations--are managed by self-perpetuating groups, with no voting at all by those who are the equivalent of investors. The shareholders of business corporations are volunteers, and if they come to disagree with the organization’s purposes or management they can leave it at any time. Most important, a business corporation has a well-defined purpose--to create value through coordinated productive and financial effort--so there should be no significant disputes concerning interests or values that have to be resolved through democratic means.
So why do some observers believe that corporations should be governed democratically? Under most state corporate laws the shareholders of business corporations are required to vote annually for a board of directors. This bears some superficial resemblance to the structure of a republican form of political democracy, in which citizens vote to elect representatives who will govern them until the next election. But there are two major differences between a political election and the election of a corporate board of directors.
First, unlike a political election, the issues in a board election are not disagreements about corporate purpose. There is a general agreement on the purpose of the corporation, which is to create value for the shareholders, or on the means by which the corporation achieves its purpose. Shareholders are not asked to decide whether the corporation should build a warehouse in Pittsburgh or acquire a trucking company, although they are asked to vote on mergers, major acquisitions, and other transactions that would substantially change the character of their investment. Beyond these special cases, their vote is a referendum on whether the management is adding value to their shares, not the means by which it does so.
Second, the directors of a corporation are not the direct representatives of the shareholders the way elected representatives in a democracy are representatives of their constituents. This is true not only because they have not been elected to perform specific acts--such as selling off the Pittsburgh warehouse--but also because they have a duty as fiduciaries (i.e., trustees) to act in the best interests of the corporation. In effect, shareholders are what lawyers would call third-party beneficiaries of the fiduciary duties that the directors owe to the corporation itself.
This becomes clear when we consider the following hypothetical case. Assume that a majority of the shareholders of a corporation voted to elect Director Jones in order to get the corporation to close the Pittsburgh warehouse. The fact that Jones was elected by a majority of the outstanding shareholders will not protect him from liability if it can be shown that closing the Pittsburgh warehouse was an improvident and foolish thing to do, especially if Jones took the action because he was elected to do so and not through use of his business judgment. Indeed, most courts--under the well-known “business judgment rule”--will protect a director against shareholder suits if he used his business judgment to make a decision. As a director, Jones owes a duty of loyalty to the corporation. For example, the Corporate Directors Guidebook, published by the American Bar Association, states: “The duty of loyalty requires a director’s conduct to be in good faith and in the best interests of the corporation--and not in the director’s own interest or in the interest of another person (such as a family member) or an organization with which the director is associated.” This concept has been followed over many years by state and federal courts.
In other words, even if the shareholders--by majority vote--gave specific directions to the directors they elected to a corporate board, the directors so elected would not be permitted to follow those directions if the result would be to harm the interests of the corporation and hence its shareholders. This is at odds with the democratic process as it works in political communities. With the exception of constitutional proscriptions on certain actions by a majority, democratic majorities are able to elect representatives who carry out their will. Minorities accept this because they know that at some time in the future they may be part of a majority themselves. Accordingly, the relationship between the directors of a corporation and the corporation’s shareholders is much more like the relationship between a trustee and the trust beneficiaries than it is that of a citizen and the government in a democracy.
So if directors are not direct representatives of shareholders, and if they cannot be bound to perform acts or functions that they are directed to perform by a majority of the shareholders, it is misleading to describe a corporation as a shareholder democracy--or to ascribe to shareholders the rights of self-governance enjoyed by citizens in a democracy.
Even if this is true, it is still necessary to ask why state corporate statutes provide for the election of directors and whether shareholder democracy in the form advocated by former SEC chairman Levitt and others will improve the performance by corporations of their specific and necessary purpose.
The Purpose of Shareholder Voting
Corporations are highly specialized organizations, originally formed only by the special act of a legislature. Corporation laws, permitting individuals to form corporations freely by filing an application or a form with a state, made corporations an alternative to a partnership or a joint stock company as a form of business organization. All the partners in a general partnership had and still have unlimited liability, extending to their personal assets, but in a joint stock company the members were able--through a trust arrangement--to separate their personal assets from the assets of the company itself. However, limited liability companies and partnerships both suffered from one significant defect. In the absence of specific agreements to the contrary, the courts allowed the members of both forms of organization to withdraw their funds at any time. Under these circumstances, the members of partnerships or joint stock companies had significant power to discipline those who managed their investments--in both cases they were permitted to withdraw their funds if they were not pleased with the way the organization was managed.
Corporations, on the other hand, offered both limited liability and transferable shares--allowing a separation between ownership and control--but they also locked in the capital contributed by the investors so that it could not be withdrawn. Although we tend to think of corporations today as controlled by their managements, they are legally under the control of their boards of directors. Most corporate laws describe the authority of the board in terms similar to those used in the Delaware law: “The business and affairs of a corporation shall be managed by or under the direction of a board or directors.” The management of the corporation--the chief executive officer and all subordinate officers--has authority to act for the corporation only to the extent that such authority is delegated to them by the board of directors.
Thus, with the advent of the freely available corporate form of organization it became possible for a small group--a board of directors--to manage the assets contributed by investors without the investors having the opportunity to retrieve their contributions to the enterprise at will. This arrangement--highly unusual when it was developed in the mid-1800s--would not have been attractive to investors unless they had some ability to replace the directors in the event that their investments were not being properly or profitably managed. Although the shares in corporations could be freely sold, they would have had little liquidity in the limited securities markets of the time. The compromise appears to be that boards would have unlimited authority to control the disposition of the corporation’s assets, but the shareholders would have some control over the board through the ability to review its actions and reelect the directors annually.
Since the directors’ duty of loyalty runs to the corporation, and since the shareholders cannot influence the directors’ actions in specific cases through the voting process, it seems apparent that the shareholders’ right to vote was intended to allow the shareholders to change the whole board when it was failing to manage the corporation effectively, but not to elect specific directors. This is the way the system has worked in practice, with the possibility of changing the board and management of a company facilitating the market for corporate control, and hence the transfer of corporate assets to more effective hands. Indeed, to make managements more accountable, it ought be easier for shareholders to elect board majorities.
Proxy Rules and Shareholder Access Proposals
Up to now, the SEC’s rules governing corporate elections--although perhaps not designed for this purpose--have been consistent with the concept that when shareholders exercise their voting power, it will be either to reelect an existing board and management or to effect a wholesale change in the management of the corporation. The rules require that a formal document known as a proxy statement be sent to all shareholders whenever there is a challenge to the election of any one of the nominees on management’s slate. Under these circumstances, it makes little economic sense to pay these costs unless one is seeking to change the entire board of directors or otherwise take control of the corporation. Consequently, efforts by shareholder activists to enhance the ability of shareholders to elect non-management nominees--the “shareholder access” idea--have focused on ways to include shareholder nominees in the corporation’s own proxy statement. In this way, changes of one or a few board members can be made without the cost of a proxy statement.
AFSCME v. AIG is an example of how the SEC’s proxy rules can be circumvented so as to permit the election of one or more directors without a wholesale change in the corporation’s management. The case began when the pension plan of the American Federation of State, County and Municipal Employees (AFSCME)--a shareholder of American International Group, Inc. (AIG)--submitted a proposed bylaw amendment for inclusion in AIG’s annual meeting proxy statement. The amendment, if adopted, would have required AIG to include among the nominees for board seats the name or names of persons nominated by certain qualifying shareholders, including (of course) AFSCME. Under the SEC’s proxy rules, companies are required to include in their proxy statements specific proposals submitted by shareholders, even if management disagrees. However, since 1976, the SEC has had a policy of permitting the exclusion of any shareholder proposal that “relates to an election,” and in response to a no-action request from AIG, the SEC staff concluded that the AFSCME bylaw could be excluded. When AIG did so, AFSCME sued to compel its inclusion. The district court upheld the exclusion on the ground that the amendment related to an election, but the court of appeals reversed, concluding that the SEC had been inconsistent over the years in defining when a shareholder proposal related to an election.
It seems likely that the SEC’s policy of excluding any shareholder proposal relating to an election was adopted in order to ensure that any contest for a board seat would be carried on under the proxy rules. Because these rules require that the contending parties each file and distribute proxy statements, this process guarantees that shareholders have the necessary information with which to make a judgment about board nominees. Looked at from this perspective, the AFSCME proposal vitiates the SEC’s purpose by eliminating the requirement for use of the proxy rules, and thus reduces the amount of information available to shareholders in connection with an election. Since the SEC was not party to the case, it is not bound by the court’s decision and it cannot appeal, but the court acknowledged that the SEC has the authority to change its policy--and by implication to affirm its existing policy--as long as it gives its reasons for doing so. Since that policy seems soundly based in ensuring that shareholders have the necessary information for voting on directors, the SEC would be fully justified in reaffirming that the AFSCME bylaw proposal relates to an election and can be excluded.
The SEC’s Decision and Public Policy
It is a separate question, however, whether it would be good policy for the SEC to adopt this method or another for permitting or facilitating shareholder access. Because we rely on corporations for economic growth, a substantial change in the governance of corporations should be avoided unless there is some reason to believe it will enhance corporate performance. If shareholder access would make corporations more efficient or productive, the SEC would be justified in encouraging the practice.
Indeed, the most widely known academic advocate of shareholder access, Lucian Arye Bebchuk, a professor at Harvard Law School, pitches his argument for shareholder access solely on the notion that it will increase corporate performance. He notes: “[I]ncreased shareholder power of participation would be desirable if and only if such a change would improve corporate performance and value.” Bebchuk’s argument is that the possibility of competition from shareholder nominees will give existing independent directors an incentive to perform more effectively. “[T]he prospect of being removed in a proxy contest,” he argues, “is far too remote to provide directors with incentives to serve shareholders.” But if this is the best argument for improving corporate performance through shareholder access, it is weak and unpersuasive.
First, academic studies of the impact of independent directors on corporate performance are mixed, with strong suggestions that supermajorities of independent directors have a negative impact on performance. This is probably true because independent directors have little incentive to take risks and know little about the details of a corporation’s business. It can hardly be expected that one or two shareholder-elected independent directors--likely to be even further removed from the priorities of management--will help the board or management improve performance. Indeed, the likelihood is that these directors will increase controversy and dissonance on the board, further reducing corporate performance. Second, even under the best of circumstances, the shareholders will be electing only one or two members of a corporate board; these new members will supplant existing (probably independent) board members chosen by the nominating committee of the board or by management. Since there is no obvious way for the effectiveness of an existing director to be judged independently of the performance of the board as a whole, a shareholder-elected director is as likely to supplant an effective director as an ineffective one. Finally, the prospect of being removed as a director by an arbitrary shareholder vote will more likely discourage valuable and qualified directors from serving on corporate boards than it will those who only serve as directors for the substantial compensation a directorship can entail.
Thus, allowing shareholders to elect one or two directors to a corporate board--supplanting members of the management slate--will in no way enhance corporate performance. It will have the opposite effect. Indeed, it is something of mystery why the advocates for shareholder access believe that more qualified directors will be elected by shareholders than will be chosen by board nominating committees or management itself. Executive search firms are constantly scouring the worlds of business and finance to find independent directors who have the experience or skills to provide useful advice or perspectives to management in the course of board discussions. There is no reason to suppose--and much reason to doubt--that shareholder groups will be better at finding such candidates than professional search firms. To the extent that shareholder-elected directors are inferior to directors selected by nominating committees or by management, corporate performance will be hampered.
Although oversight of management is one of the principal reasons given for shareholder access proposals, most corporations--and all those that are listed on the New York Stock Exchange and NASDAQ--are already required to have independent majorities on their boards of directors. These majorities frequently meet separately from management and are often chosen by a nominating committee made up of independent directors. Many other corporations have lead directors or independent chairs. Accordingly, there is already a substantial infrastructure for the independent oversight of management, and given the indications that independent directors have an adverse effect on corporate performance, it makes more sense to reduce reliance on this oversight mechanism than to increase it.
Finally, shareholder access proposals that facilitate the election of directors who are not part of the management slate are likely to be abused by activist organizations pursuing agendas other than corporate performance. Principal among these potential sources of activist directors are unions and employee pension managers, who wish to use board memberships to advance the interests of their organizations rather than those of shareholders. Indeed, what could be the objective of a single shareholder-elected director? A single director cannot be expected to change management policies that have anything to do with shareholder value. But a single director who has a special interest, such as pension rights of employees or the company’s position on a public issue, might be able to negotiate an agreement on that single issue in exchange for allowing board meetings to proceed without disruption. As Stephen M. Bainbridge, a law professor at the University of California, Los Angeles, noted in a recent article: “[T]he two classes of institutions most likely to make significant use of [shareholder access opportunities]--union and state and local employee pension funds--are precisely the classes most likely to misuse those powers in the pursuit of private benefits. In other words . . . the shareholders most likely to avail themselves of the newly granted powers would also be those most likely to misuse them.” Bainbridge points out, for example, that “union pension funds tried to remove directors or top managers, or otherwise affect corporate policy, at over 200 corporations in 2004 alone. Union pension funds reportedly have also tried to use shareholder proposals to obtain employee benefits they could not get through bargaining.”
To adopt or facilitate a shareholder-access proposal, the SEC must conclude that business corporations have some purpose other than productivity and the creation of value. Only such a conclusion could justify subjecting business corporations to the distractions and controversy that shareholder access proposals will inevitably bring about.
Peter J. Wallison is a resident fellow at AEI.
AEI research assistant Daniel Geary and editorial assistant Evan Sparks worked with Mr. Wallison to edit and produce this Financial Services Outlook.
1. Arthur Levitt Jr., “Stocks Populi,” Wall Street Journal, October 27, 2006.
2. American Federation of State, County and Municipal Employees [AFSCME] v. American International Group [AIG], 462 F.3d 121 (2nd Cir. 2006), slip opinion.
3. F. A. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), 107.
4. American Bar Association, Section of Business Law, Corporate Director’s Guidebook, 4th ed. (Chicago: American Bar Association, 2004), 14 (emphasis added).
5. See, e.g., Weinberger v. UOP, Inc., 457 A2d 701, 710 (Del. 1983), quoting Guth v. Loft, Inc., 5A2d 503, 510 (Del. 1939): “A public policy existing through the years, and derived from a profound knowledge of human characteristics and motives, has established a rule that demands of a corporate officer or director, peremptorily and inexorably, the most scrupulous observance of his duty, not only affirmatively to protect the interests of the corporation committed to his charge, but to refrain from doing anything that would work injury to the corporation.” See also Zahn v. Transamerica Corporation, 162 F2d 36 (3rd Cir. 1947).
6. See the detailed discussion of this issue in Margaret M. Blair, “Reforming Corporate Governance: What History Can Teach Us,” 1 Berkeley Bus. L. J. 1 (2004).
7. Delaware Code Annotated, §8.141.
8. Lucian Arye Bebchuk, “Shareholder Access to the Ballot,” (faculty discussion paper 428, John M. Olin Center for Law, Economics, and Business, Harvard Law School, Cambridge, MA, 2003), 2 (emphasis added), available at www.law.harvard.edu/programs/olin_center/papers/428_bebchuk.php (accessed November 30, 2006).
9. Ibid., 5
10. The most current survey of the literature on this subject is Sanjai Bhagat and Bernard Black, “The Non-Correlation between Board Composition and Firm Performance,” 27 J. Corp. L. 231 (2002): See also Peter J. Wallison, “All the Rage: Will Independent Directors Produce Good Corporate Governance?” Financial Services Outlook (January 2006); and Sanjai Bhagat and Brian Bolton, “Board Ownership and Corporate Governance Indices” (University of Colorado, Boulder, 2006), 5–6.
11. Stephen M. Bainbridge, “Response to Increasing Shareholder Power: Director Primacy and Shareholder Disempowerment,” 119 Harvard L. Rev. 1735, 1751 (2006).
12. Ibid., at 1755.