Will the Real Shareholders Please Stand Up? Principals and Agents in the Sarbanes-Oxley Era

Discussions of corporate governance using a simplified model of “shareholders as principals versus management as agents” fail to reflect today’s reality. The reality is interactions of various kinds of agents, such as “stockholders” who are usually the paid employees of investment intermediaries–in other words, management agents of organizations that are themselves agents. Related issues include the idea of “independence,” comparisons of public versus private company governance, and the relative contributions of financial capital versus human capital.

Since the passage of the Sarbanes-Oxley Act of 2002, we have been deluged by discussions of and pontifications on corporate governance. Virtually all of them take as their assumed starting point the famous “principal-agent” problem or “separation of ownership and control of the firm,” going back to Adolf A. Berle Jr. and Gardiner C. Means’s The Modern Corporation and Private Property, published in 1932.Berle and Means themselves cite Adam Smith’s celebrated judgment on joint stock companies made in The Wealth of Nations in 1776:

The directors of such companies, being the managers rather of other people’s money than of their own, it cannot well be expected that they should watch over it with the same anxious vigilance with which the partners of a private co partnery frequently watch over their own. . . . Negligence and profusion, therefore, must always prevail, more or less, in the management of the affairs of such a company.[1]

This issue has given rise to a large academic literature, but in many current discussions it is treated in an oversimplified–or even simple-minded–way. The actors are divided into two camps: managers and directors on one side as agents, and stockholders on the other side as principals. The question is then posed: “How can we give the principals more say?”

Agency Capitalism

Of course, managers and directors may themselves be significant shareholders. Clifford G. Holderness, Randall S. Kroszner, and Dennis P. Sheehan found, for example, that “[c]ontrary to the received wisdom, managerial ownership of publicly traded firms is higher now than in 1935. The mean percentage of common stock held by a firm’s officers and directors as a group rose from 13 percent in 1935 to 21 percent in 1995,” and that “[i]n terms of real 1995 dollars, insiders’ holdings are on average four times higher . . . and this increase holds across all firm sizes.”[2]

Although many companies have grown extremely large, internationalized, and complex, the fundamental structure of the limited liability business corporation, with its shares of stock, debt instruments, professional management, and board of directors, has not changed much since Berle and Means were writing in 1932. The structure of the corporation, a great innovation for large-scale and efficient investment and production was a nineteenth-century legal, managerial, and financial invention.

But the public capital markets have changed a great deal in their fundamental structure since then. The individual as direct investor has been overshadowed in stock ownership and overwhelmed in stock trading by an efflorescence of intermediary equity investors, including mutual funds; public, private, and union pension funds; hedge funds and funds of hedge funds; and endowments. There are also brokerage firms holding shares in street name for account of individual investors and voting uninstructed proxies.

All of these intermediary investment organizations are themselves agents, not principals. As the great mutual fund entrepreneur John Bogle writes, “These firms are agents for their principals, the underlying owners of the mutual funds they manage and, arguably, the ultimate beneficiaries of the pension plans they manage.”[3] In short, these mutual funds, pension funds, hedge funds, and the like are not shareholders, but agents for the real shareholders. Referring to the latter, Bogle adds: “After all, they own the company.” But they have rather little to say in the corporate govern-ance debates compared to their hired investment agents.

Moreover, the funds that are themselves agents are run by professional managements that represent another level of agents. They and their overseeing boards face the same principal-agent problems that corporate managements have with respect to their companies. In addition, pension fund managers hire third-party investment managers to act on their behalf, as agents for the agents.

A generation ago, Peter F. Drucker discussed what he called “pension fund socialism”:

Instead of the old-line capitalists, in developed countries pension funds increasingly control the supply and allocation of money. . . . The beneficiary owners of the pension funds are, of course, the country’s employees. If Socialism is defined, as Marx defined it, as the ownership of the means of production by the employees, then the United States has become the most “socialist” country around–while remaining the most “capitalist” one as well. Pension funds are run by a new breed of capitalists: the faceless, anonymous, salaried employees, the pension funds’ investment analysts and portfolio managers.[4]

But the employee (and retiree) “owners of the means of production” are seldom or never heard from. The members of the “new breed of capitalists” are not capitalists at all, but rather, as Drucker goes on to say, employees themselves–in short, agents, not principals.

In this modern situation, seventy-five years after Berle and Means were writing, the simplified model of management as agents of stockholders who are principals is inadequate. Instead, management looks at the equity market and finds a plethora of other agents. These agents act in various ways on behalf of the ultimate-but-distant principals, but cannot escape their own agency issues. They often claim–sometimes in a most intrusive way–to be shareholders themselves, speaking as “owners.” But they are not the owners.

Thus most of the voices we hear speaking for “shareholder rights” and “shareholder value” are those of investment intermediary managers. The voice of the California Public Employees Retirement System (Calpers) is a notable example. Regarding Calpers’s agency problems, Thomas Donlan argues that “[w]e hold Calpers in particular in minimum high regard anyway, for it often seems more beholden to its participants’ unions and their political agendas than to sound fiduciary principles.”[5]

Whether or not one agrees with Donlan’s judgment, it raises an interesting agency issue. What is the relationship between the managers of a union pension fund as agents for the pension beneficiaries and the managers of the union, who are themselves agents for the members of the union? What should the principals–the union member beneficiaries–have to say about the compensation of the pension fund managers, for example?

There is an important link to hedge funds here, since hedge funds are increasingly owned by mom-and-pop investors through pension funds. The often voluble managers of the hedge funds represent another layer of agents for these silent ultimate principals. What should the principals have to say about the compensation of the hedge fund managers, who are paid on the “2 and 20” formula? According to Hugo Dixon of breakingviews.com, “[t]he incentive system–the notorious ‘2 and 20′ under which managers get 2% of funds under management as an annual fee and 20% of the profits–favors those who can scoop up the biggest pots of other people’s money. . . . But when things go pear-shaped . . . the managers don’t share in the losses or give back past takings.”[6] This nicely defines a classic agency conflict.

Even more distant from the real shareholders are proxy advisory firms, such as ISS and Glass Lewis, which wield a large and growing influence in corporate governance decisions, since institutional investors are coming to rely on them for recommendations about how to vote proxies. Related to these are corporate governance rating firms, such as The Corporate Library. These firms are not shareholders, or even agents of shareholders, but they claim to speak on the shareholders’ behalf.

As businesses themselves, they have their own financial and political agendas and philosophical commitments. Concerns have been raised that “the quality of the advice could be tainted by the industry’s expansion and the need to generate profits,” and by the objectives of their parent companies, which in the case of Glass Lewis has the Chinese government as a minority investor.[7]

Nonetheless, such firms are gaining influence because institutional investors–agents themselves–feel safer and less vulnerable to criticism by following someone else’s recommendations. Within mutual funds, it appears that portfolio managers–who have the most direct connection to investments–are losing power over how proxies are voted to centralized committees for proxy policy, which look to the positions of the advisory firms. For example: “Today, a highly structured committee of 15 employees from around the globe sets T. Rowe’s voting policies at more than 6,000 companies internationally. Fund managers still have a say in how votes are cast, but the voting policy set by the committee annually is the default position.”[8]

Try to imagine how such a committee fits into a simple principal-agent model. It doesn’t.

Thus, in today’s public capital markets, corporate governance actually confronts a complex interaction of various kinds of agents. The issues of principals and agents are more complicated and fuzzy than typical discussions of “good corporate governance” suggest. This more complex situation is best described as “agency capitalism.”

Will the real shareholders please stand up?

Can the real shareholders stand up with today’s institutional structures?

My AEI colleague Peter J. Wallison has argued that “shareholders are much more like the beneficiaries of a trust than they are like the owners of a piece of property.”[9] This raises the more radical question of how much shareholders should stand up. The opposite (and more frequently heard) view has been expressed by Arthur Levitt Jr. as “the principle that shareholders can have a voice in the running of the companies that they own, that their votes will count.”[10]

Whichever side of this dispute we may take, we must recognize that a great many people who are standing up in corporate governance debates and casting votes in shareholder meetings are not themselves shareholders. They are agents of various sorts, not principals.


In the corporate governance debates of the Sarbanes-Oxley era, now five years old, the idea of “independence” has loomed large. Independence has been seen as a way to address the stockholders-as-principals versus management-as-agents problem. Independent directors have gotten special attention and are required to be majorities of boards of directors under new rules of the stock exchanges. These rules were, however, adopted under pressure from the Securities and Exchange Commission (SEC), without notice, rule-making, or public comment–apparently in violation of the SEC’s own due process requirements.[11]

Key board committees must now be composed of only independent directors. Some can be considered little fiefdoms inside the board. The SEC has twice tried to require that mutual fund boards have independent chairmen–and has twice lost in court–but it is trying again. Great efforts have been made to insure the independence of public accountants. A recent Commodity Futures Trading Commission regulation creates a new class of hyper-independent “public directors” for futures exchanges.

In what seems to me an “independence fetish,” it is nearly always assumed that more independence is always better–but is this true? Independence is important, but it is certainly not the be-all and end-all. It has significant limitations and must be balanced with other factors.

In particular, there is a trade-off between independence and knowledge. Which is more important to business decisions and successful risk-taking? Insiders will always know more than outsiders, although outsiders can bring helpful perspectives. A manager working sixty hours a week cannot help but know more than a director coming to six or eight board meetings a year, no matter how assiduously the director may study the board agenda book–which must, of course, be prepared by the management. (This, in turn, is much more knowledge than the average shareholder-principal has.)

One of my colleagues on a board of directors has described the ideal independent director as somebody you keep locked in a dark room, who cannot talk to anybody else or know anything about the affairs, issues, or plans of the company, but is available to be rolled out whenever an “independent committee” is needed. Allowing for rhetorical license, this is another way of expressing the trade-off between independence and knowledge.

Even the most diligent directors cannot match the detailed, specific knowledge of the operations, risks, and opportunities of an enterprise, which a competent management must have. If they did, they would be the management, not the board. Nor is this a new situation. As Walter Bagehot wrote in 1873:

Not only would a real supervision of a large business by a board of directors require much more time than the board would consent to occupy in meeting, it would also require much more time and much more thought than the individual directors would consent to give. . . . [F]or the most part, directors of a company cannot attend principally and anxiously to the affairs of the company without so far neglecting their own business as to run great risk of ruin.[12]

What makes a board independent in fact as well as in form? It has often been pointed out that Enron’s board was a model of independence in formal structure. On form versus substance, Warren Buffett reflected in an annual letter to shareholders:

Over a span of 40 years, I have been on 19 public company boards (excluding Berkshire’s) and have interacted with perhaps 250 directors. Most of them were “independent” as defined by today’s rules. But . . . their contribution to shareholder well-being was minimal at best and, too often, negative. These people . . . simply did not know enough about business.[13]

Under the independence theory, a board composed entirely of independent directors except for the CEO would be the ideal. In one discussion, I expressed the view that this would not control the “imperial CEO,” but would instead tend to make the CEO stronger. Agreeing with me, a retired CEO of a Fortune 100 industrial company said, “As CEO, with everybody else independent directors, I would completely control the board. I would be the only one who really knew anything and all information to the board would flow through me.”[14]

Should we consider the British model in which one-third of the directors are typically executives of the company? There is no discussion of such an idea in the United States, which indeed is headed in the opposite direction, toward the maximally independent board.

A different aspect of independence in corporate governance has developed in the role of accounting firms. The Sarbanes-Oxley era has produced examples of these firms not wanting to give advice on the application of complicated and arcane accounting rules for fear of compromising their “independence” as auditors. Equally bad have been accounting engagement teams who could not provide advice because of second-guessing by their head office bureaucracies. Either way, this is a reduction of the independence idea to absurdity, causing accountants to fail in their central professional responsibility to make judgments and give advice.

A final perspective on this issue comes from the government:

The Defense Science Board, a 50-year-old institution in the Defense Department that brings the best minds from the private sector to help advise our military on technical matters, stood up a task force to undertake an important study of the health of the U.S. space industry and the impact of export controls on the industry. But just as the study was coming to its conclusion, Defense Department ethics lawyers decided that all the members of this task force could have perceived or potential conflicts of interest, and they dissolved the task force before it could report its conclusions.[15]

Independence preserved, but knowledge lost.

Public Company versus Private Company Boards

William L. Walton, chairman and CEO of Allied Capital Corporation, contrasts the “guidance board” of private equity investments in private companies with the “compliance board” model for public companies in the Sarbanes-Oxley era.[16] The former is overwhelmingly focused on business performance. The latter requires spending much more time with lawyers and consultants dealing with the regulatory issues of being a public company.

The governance of a company with private equity ownership, Walton suggests, typically includes:

  • Small boards of one or two managers, two or three private equity professionals, and two or three industry experts
  • Frequent board meetings, with telephone meetings in between and “brainstorming” with management
  • Focus on cash measurements, not Generally Accepted Accounting Principles earnings
  • Significant personal investment by management with their own money at risk and compensation based on how the investment does
  • A clear investment thesis and a plan to maximize the investors’ capital gain in three to five years

The goal is the application of deep business knowledge and a strict alignment of the interests of management and the investors.

John Plender offers a balanced view of the private versus public company models. On the one hand, “Private equity investors enjoy huge competitive advantages against investors in public markets. Their corporate governance model aligns incentives between management and shareholders in a way that is not possible for investors in public markets.”[17] On the other hand, remember that private equity managers themselves have agency issues: “The paradox of private equity is that it imposes immense discipline on managers of companies, while the discipline on managers of private equity funds is more nebulous. Performance measurement data are suspect, transparency is poor. . . . This encourages the extreme cyclicality of the industry.”

All this suggests an interesting Darwinian competition between the two governance models that we will be able to continue to observe.

Financial Capital versus Human Capital

The famous “separation of ownership and control” emphasized by Berle and Means could also be thought of as the separation of financial capital from human capital or the separation of providing money and providing talent. These factors and the returns to them are combined in the entrepreneur who invests his own money, borrows on his own credit, and puts his personal ideas, knowledge, and hard work into directing the business. As Bagehot put it: “A thorough man of business, employing a fair capital in a trade, which he thoroughly comprehends, not only earns a profit on that capital, but really makes of his professional skill a large income. He has a revenue from talent as well as from money.”[18]

There is no natural or obvious rule for exactly how much of the return should belong to financial versus human capital, to money at risk versus talent. There is only the market transaction or contract between the suppliers of the two to define their respective rewards. In the simplistic model, this is the investor principal hiring the management agent to make productive use of the capital provided.

By contrast, in the contemporary real world of the public corporation, there are many complex relation-ships among multiple agents: mutual funds, hedge funds, pension funds, other institutional investors, the managements of all of these, independent directors, non independent directors, managements, and public accountants. Also involved are a host of other parties, many claiming to speak for investors: proxy advisors, corporate governance consultants, rating agencies, lawyers, and regulators.

Under this system of agency capitalism, the ultimate principals–the real shareholders providing money at risk–have a lot of agents, capital market managers as well as corporate managers, all of whom they pay, directly or indirectly, and all of whom present principal-agent issues.

Alex J. Pollock is a resident fellow at AEI. Research assistant Daniel Geary and editorial assistant Evan Sparks worked with Mr. Pollock to edit and produce this Financial Services Outlook.


1. Adam Smith, The Wealth of Nations (1776, repr., New York: Alfred A. Knopf, 1991), 229, quoted in Adolf A. Berle Jr. and Gardiner C. Means, The Modern Corporation and Private Property (New York: Macmillan, 1932), 346.

2. Clifford G. Holderness, Randall S. Kroszner, and Dennis P. Sheehan, “Were the Good Old Days That Good? Changes in Managerial Stock Ownership Since the Great Depression,” Journal of Finance 54, no. 2 (April 1999): 436.

3. John C. Bogle, letter to the editor, Wall Street Journal, December 7, 2006.

4. Peter F. Drucker, Post-Capitalist Society (New York: HarperCollins, 1993), 6.

5. Thomas G. Donlan, “The Royal CEO, Part II,” Barron’s, January 15, 2007.

6. Hugo Dixon, “The Greed-Fear Imbalance,” Wall Street Journal, January 29, 2007.

7. Kaja Whitehouse and Shefali Anand, “Proxy-Advisory Firms Encounter Concerns on Owners’ Influence,” Wall Street Journal, May 22, 2007.

8. Kaja Whitehouse, “A Changing Story: How Funds Vote Your Shares,” Wall Street Journal, May 7, 2007.

9. Peter J. Wallison, “Should Corporations Be Democracies?” The American, May/June 2007, available at www.aei.org/publication26169/.

10. Arthur Levitt Jr., “Stocks Populi,” Wall Street Journal, October 27, 2006.

11. Peter J. Wallison, conference remarks (AEI, Washington, DC, June 12, 2007), available through www.aei.org/event1527/.

12. Walter Bagehot, Lombard Street: A Description of the Money Market (1873, repr., Westport, CT: Hyperion, 1979), 126.

13. Berkshire Hathaway, Inc., 2002 Annual Report (Omaha: Berkshire Hathaway, 2002), 17, available at www.berkshirehathaway.com/2002ar/2002ar.pdf (accessed July 10, 2007).

14. Private conversation with the author.

15. John J. Hamre, memorandum to Center for Strategic and International Studies trustees, advisers, and friends, April 11, 2007.

16. William L. Walton, conference remarks (AEI, Washington, DC, June 12, 2007), audio and video available through www.aei.org/event1527/. Walton drew the “guidance board” versus “compliance board” model from Tom Perkins, founding partner of the venture capital firm Kleiner, Perkins, Caufield & Byers.

17. John Plender, “The Privileged Existence of Private Equity Funds,” Financial Times, April 24, 2007.

18. Walter Bagehot, Lombard Street, 127.

Originally posted at AEI and reproduced her with the author’s permission.

3 Responses to "Will the Real Shareholders Please Stand Up? Principals and Agents in the Sarbanes-Oxley Era"

  1. Guest   May 22, 2008 at 9:02 am

    Excellent article. Question: can "agency capitalism" work, provide appropriate incentives, allocate efficiently capital and impose proper corporate governance? If not then what do you suggest to do? Government intervention? Something else? This is a serious as repeated credit and asset bubbles seem to be associated with this agency capitalism.

  2. Guest   May 22, 2008 at 1:05 pm

    Ah, information asymmetries. If only we were Borgs and knew what everyone else knew.

  3. 2cents   May 23, 2008 at 12:15 am

    Wow, Alex great post! Although at first it seemed a bit compartmentalized in the way you presented it, I found that this clearly allowed you to pinpoint the major strengths and weaknesses of each argument fairly concisely.I think in the end what we all wish for are smart (businesswise and intellect), honest, visionary and passionate management and governance members. When we don’t get that we start to create ‘agents’ to cover for any inadequacies. That’s the main reason why entrepreneurial startups can go from here to there so fast. They have these attributes embodied in one or a few people. This streamlines the decision process and maximizes the feedback necessary to move forward quickly. Unfortunately, once a business outruns the capacity of the entrepreneurial founders, it enters a critical phase of finding additional talent to pick up the load.As for the agency mess you describe, you can’t regulate it in any way, because it just spawns additional new agencies to cover the intricacies of the just created regulations. Sort of doomed if you do and doomed if you don’t’ situation. No, what you need are the talents that I stated above, then you could work towards de-agencying thus streamlining your shareholder interaction and also maximizing the bottom line.