Daniel J.H. Greenwood

Home | Previous Page

ESSENTIAL SPEECH: WHY CORPORATE SPEECH IS NOT FREE

83 Iowa Law Review 995 (1998)

Daniel J.H. Greenwood [FN*]

Download printable version:

Copyright © 1999 Daniel J.H. Greenwood

TABLE OF CONTENTS

I. Introduction

A. Point

In a democracy, the citizens are the only legitimate sources of law. It follows inexorably that corporations, not being citizens, cannot be legitimate political actors. Like the government itself, corporations are mere tools of the citizenry, political objects rather than political subjects, to be given just as much respect as the citizens deem useful and no more. To grant a tool a right against the citizens who use it is a form of political idolatry that ought to be abhorrent to any democratic regime. [FN1] Rights are for people, not for their instruments. [FN2]

This is even more true in a mixed economy such as ours, in which businesses are directed to pursue profit and we rely on the government to play the role of the invisible hand, assuring that the pursuit of profit leads to general economic well-being and health and not to the horrors of Dickensian England or of those most free of contemporary markets, the international drug trade or civil war era Beirut. [FN3] If the pursuit of profit must be guided to eliminate unattractive means (violence), unpleasant externalities (pollution), excess (of poverty or exploitation), or products we would rather not have efficiently produced to meet consumer demand (child pornography, kidnapping, crack cocaine), the spheres of government--the regulators, controlled by the people--and of business--the profit pursuers, controlled by a market structured and directed by the regulators--must be strictly separated. [FN4] To allow the regulated to capture the regulators threatens the entire system. [FN5]

Similarly, in a multifaceted culture of manifold and various values, it is inevitable that the pursuit of profit, valuable as it is, will conflict with other important goals. For example, economic efficiency is best promoted by flexibility, rapid adjustments to changing conditions, and a mobile labor market. A sophisticated culture and well-raised children, however, require stable intergenerational families and stable communities. The conflict cannot be avoided: a mobile and flexible labor market requires families without strong community roots, children raised away from their grandparents (or with their parents late at work), and the worry and stress of actual or threatened unemployment.

Important and widely shared values conflict or are self-contradictory.[FN6] Citizens must decide how much of the constant revolution of capitalism they are willing to tolerate, or how much of the stultifying stasis of poverty they can stand; they must decide whether they prefer the excitement and challenge of a constantly changing capitalist economy or prefer the comfort of predictability that comes with stability and decline. In a democratic society, the resolution of these conflicts--or at least their debate--is the central task of politics. [FN7] To allow markets or private bureaucrats with legally determined duties to pursue particular agendas to decide when or to what degree we should be market-driven or legally determined actors is to coopt the central task of democratic politics.

B. Counterpoint

On the other hand, we have long recognized that a democratic society can only survive with the aid of intermediate institutions [FN8] strong enough to act as countervailing powers [FN9] against the sometimes irresistible pressures of mass psychology. [FN10] Moreover, a society in which the government is the only, or the leading, organized force is one that is unlikely to be able to control that government. Democracy, even of the mass variety, is likely quickly to collapse as the governors escape from their supposed servitude to the people. This insight is reflected in our political traditions, particularly our radically decentralized and nonmajoritarian form of democracy: differently defined local majorities can serve as powerful forces to slow down the actions of regional majorities, and vice versa. [FN11] It is similarly fundamental to our First Amendment protection of the press, churches, and the peaceable assembly of citizens: unless citizens can organize, they are unlikely to be able to participate meaningfully in the political process. [FN12]

Where, then, do business corporations fit in this picture? Are they essential components of civil society, closely analogous to the citizen groups protected by the First Amendment and important to protect for the same reasons, as one of the key ways in which the citizenry protects itself against its protector? Or are they similar to the government itself, a tool that constantly threatens to control us instead of being controlled by us, which we should have rights against and not vice versa? [FN13]

American political culture often has seen unrestricted large corporations-- especially, but by no means exclusively, financial corporations--as key threats to the political well-being of the nation. Jeffersonian agrarianism and the Jacksonian anti-bank movement, the Grange and the Progressive attacks on the railroads and the trusts, and the consumerist movement of the last generation and much of the environmentalist movement today have each seen large corporations as threats to key American political values. [FN14] The decentralizing impulses of New Deal financial regulation and the enormous relative subsidies extended to small businesses in the tax code and in virtually every modern regulatory regime stem, presumably, from similar skepticism about large corporate enterprise. [FN15]

But the courts consistently have disagreed. Nearly every case that has considered the question has concluded that particular constitutional rights (with a few exceptions, most significantly the Fifth Amendment right against self-incrimination) may be claimed by corporations, generally without any consideration of legal or policy issues that might differentiate between large or small, public or close, or financial or nonfinancial corporations, or between corporations and individuals. Hundreds of other cases simply assume the similarity of corporate to individual rights without any consideration at all. [FN16]

The courts, in short, treat corporations as if they were individuals--by pretending that the corporation itself is an individual under "entity" theories, by looking "through" the corporation to individuals behind it using "aggregate" theories, or by simply treating the corporation as property owned by an individual. By collapsing corporations into individuals, these theories suggest that we need not think about the ways in which organizations work. This Article is an extended attempt to go behind identification of corporations with individuals and, taking seriously the collective nature of the corporation, to examine the effect of one particular corporate right--the right to free speech--on actual human beings and their relations to one another.

C. A Different View

Political action by business corporations, I contend, implicates no First Amendment values.

Not because the distinction between political speech and economic advertising can be revived--it cannot. The most pressing political issues of the day--any day--involve the inevitable trade-offs between economic activity and other values: between, for example, making a living and raising children. So proposals to enter into economic transactions always have political content.

An advertisement for any product, in addition to whatever else it may be, is a political claim that the product is valuable, that society ought to use it, and that alternative social visions in which resources are differently distributed are less valuable. Thus, for example, advertising for cars also implicitly promotes other aspects of an automobile-based culture, including military defense of the sources of cheap oil, highway-based transportation, and suburbanization and its ramifications. Similarly, any product advertisement is an intensely political criticism of many traditional religious views, including those that consider riches to be a danger to moral integrity, regard covetousness as a mortal sin, or advocate using national wealth for purposes other than private consumption--promoting the arts, constructing public monuments, parks or sacred buildings or studying sacred texts, for example. [FN17]

In a world of finite resources to promote one thing is inevitably to argue that others are worth less. [FN18] Accordingly, commercial advertising, even of the more narrow variety, properly is seen as participating in the "robust debate" the First Amendment seeks to encourage on these issues. [FN19]

The issue is not the nature of the speech but rather the nature of the speaker. Corporate speech is coerced, not free. It is compelled, legally mandated speech, not the result of anyone's autonomous behavior. It does not reflect the views of shareholders, nor, if management is acting in good faith, those of managers or other corporate agents. Instead, corporate speech reflects the hypothetical interests of a creature given reality by the market and the law: the fictional shareholder. [FN20]

Corporations speak by spending money: they hire others to speak for them. Corporate speech is thus an agency problem. How corporate managers and employees understand their duties to the corporation determines what corporations will do. This much has long been recognized and, as a result, defenses of constitutional rights for corporations often have looked through the entity to underlying citizen-shareholders who are claimed to be the "real" speakers. [FN21] The corporate-law agency problem is a peculiar one, however, because the fiduciaries who control the corporation do so on behalf of a principal that is a legal fiction, not a citizen.

While conventional views emphasize the rights of supposedly human shareholders, in fact publicly traded business corporations are legally and practically barred from speaking on behalf of any human being. The shareholders in whose interests corporations must speak are not the human beings who own (or, more often, on whose behalf other institutions own) the shares. Indeed, they are not citizens at all, but rather moments in the market, legal abstractions that have interests quite different from those of real citizens in their full complexity. Unlike real people, the fictional shareholder is an entirely one-sided abstraction; it seeks to increase the value of its shares without regard for any other value. Corporations, then, when they act as they are supposed to, pursue only one goal of the many that are important in a civilized society. Corporate agents, in short, work for a principle, not a principal.

As a consequence, the firm's agents and employees are asked to set aside their own views, while the views of the individuals who own its securities, the human beings behind the shares, are largely irrelevant to the positions a properly run corporation takes. This picture of the corporation acting on behalf of a fictional shareholder leads to the conclusion that corporations are defined by the law and the market in a way that makes them inappropriate participants in the political debate.

The paper proceeds as follows: In Part II, I review the current law regarding corporate speech. In general, the Supreme Court has rarely viewed the corporate status of a petitioner as relevant to a discussion of its constitutional rights. In the speech area, the leading case, First National Bank v. Bellotti, [FN22] explicitly rejects the notion that a corporation might have different speech rights from a human speaker, and insists that it is the speech, not the speaker, that determines constitutional protection. While later cases have modified this position in limited areas concerning campaign finance, it remains the basic law.

Conversely, the Supreme Court has never referred to, let alone cited, the basic law regarding governmental lobbying. For reasons quite similar to those I raise with respect to corporations, it has long been the law that governmental units have no claim to First Amendment protection. The institutions of the people should not use the privilege granted to them in order to expand that privilege--but the Court has never explained why the same reasoning should not apply to corporations.

In Parts III and IV, I explain the theory of corporate speech that results from the theory of the fictional shareholder and distinguish corporations from other groups that participate in our political process. Part III reviews the general problems of group speech and speech by agents; those problems are present in corporations as well, but are neither the distinctive feature nor a reason for denying constitutional protection to group or agent speech in general. Readers familiar with the literature on group speech generally may wish to skip Part III.

Part IV turns to the distinctive character of corporate speech. Corporate speech is compelled speech, not free speech. When a corporation "speaks," it allocates money according to the norms of corporate law to pay a corporate agent to speak on behalf of corporate interests. In sharp contrast to ordinary speech by agents on behalf of voluntary organizations, in the publicly traded corporation, none of the human actors involved is in a position to control the speech. None of them is likely to view it as his or her own speech. As a result, the speech is not properly understood as made on behalf of any one of them. Instead, corporate speech is better understood as the expenditure of money in accordance with dictates of the law and the market on behalf of the imaginary interests of a legal fiction: the fictional shareholder. I review my more detailed discussion of the fictional shareholder published elsewhere to point out the aspects most salient here.

Finally, in Part V, I argue that allowing corporate resources to be used to lobby on behalf of the interests of the fictional shareholder is presumptively illegitimate in a self-governing community and contrary to the political goals of the First Amendment. Both the law and the market force corporate actors to run the corporation on behalf of the interests of fictional shareholders rather than in the interest or views of any citizen. While the fictional shareholder is a useful simplification of the interests of the people behind the shares for many purposes, it is not for others. In particular, the legally created fiction will consistently take one side of a series of hotly contested and difficult issues regarding the economy, market, workplace and regulation, because fictional shareholders, unlike real people, can have no loyalty to particular places, relationships, or ways of doing things and place no weight on any nonmarket value. Fictional shareholders, thus, will sacrifice almost anything in the interests of higher profit (in their not-always intuitive understanding of profit); in contrast, the citizens behind the fiction can be expected to have far more diverse and conflicted opinions on these important political struggles.

Accordingly, extending First Amendment protection to corporate speech forces corporate actors to spend corporate money to advocate a consistent series of positions not defined by the free, autonomous or self-governing choices of any citizen. Corporate money, accordingly, distorts the political process, becoming a political force not necessarily tied to any class, party, national interest or opinion. Social resources aggregated for an entirely different purpose become independent creatures controlling their own creators rather than serving them.

Finally, I distinguish some issues that are different from the ones central to my analysis: how, for example, to distinguish between the corporate lobbying with which I am concerned and individual (or group) authors who use corporate publishers to distribute their speech.

II. The Supreme Court's Law: The Irrelevance of Corporations

Under current constitutional law, a corporation's expenditure of money to promote its financial or political interests by lobbying, political advertising, or other attempts to influence the political process or legislative, judicial or administrative lawmaking, is considered "core First Amendment protected speech." Accordingly, neither the states nor the Federal government may restrict or regulate such lobbying in the absence of a compelling state interest. Furthermore, First Amendment doctrine treats corporations indistinguishably from citizens: with the sole exception of expenditures in connection with campaigns for political office, corporations have the same speech rights as citizens.

The law with respect to campaigns for political office is somewhat confused. The Court has acknowledged an important state interest in preventing corruption or the appearance of corruption through "quid pro quo contributions." [FN23] In Buckley v. Valeo, [FN24] the Court distinguished between campaign contributions and direct expenditures, whether those expenditures were made by the candidate out of personal funds or by independent third parties in support of the candidate. Corruption concerns, the Court reasoned, arise only with campaign contributions, not with direct expenditures, and thus only campaign contributions may be regulated consistent with the First Amendment. [FN25] Later cases have added to this contribution-expenditure distinction an additional corporate-noncorporate distinction: corporate expenditures in connection with a candidate's campaign for office may be restricted more strictly than individual ones, apparently because the Court perceives the accumulation of corporate wealth to be more state-assisted than the accumulation of private wealth. [FN26]

A. Three Strands of Constitutional Doctrine

The constitutionalization of the regulation of corporate lobbying stems from three disparate and largely independent lines of cases.

1. Forcing Corporations into the Public/Private Distinction

First, chronologically if not necessarily logically, is a long line of cases in which the Court has refused to take seriously the entity or organizational character of corporations, treating them instead as if they were analogous in a simple way to individual human beings. [FN27] Instead of thinking about corporate functioning, corporate sociology, or corporate law, and then considering how constitutional norms designed to protect citizens from the power of the state should apply to powerful nonstate organizations, the Court often has seen its task as deciding whether corporations are "persons" entitled to protection from the state, and, in general, has concluded that they are. [FN28]

Our Constitution assumes a great divide between the state--feared as a collective entity able to exercise unwarranted power over individual citizens-- and society, understood as relatively helpless and unorganized citizens. Modern corporate law is quite a bit younger than our Constitution: the notion that limited liability and perpetual legal existence should be available on demand dates back only to the middle of the last century, while the modern structure of corporate law, including the abandonment of minimum capital requirements and stated purpose, the right to hold shares of other corporations, and the location of virtually all ultimate authority in the board, is largely a product of the turn of this century. [FN29] Thus, there is no reason to believe that the constitutional categories--which predate the modern corporation--will fit the problems raised by modern corporate law in any simple way. [FN30]

Despite the variety of doctrinal labels--often appearing within the same case and sometimes in the same opinion--the U.S. Reports are virtually bereft of any serious discussion of where corporations fit within the categories of 18th century liberalism that underlie our Constitution. The modern publicly traded multinational corporation, superficially at least, appears to be as large and well organized, as in control of resources and potential instruments of coercion or power over individuals as are most local governments. [FN31] Furthermore, modern corporate law, by locating the center of corporate authority in a board of directors bound to act in the best interests of the corporation, makes clear that the shareholders--let alone any other group of individuals who might be considered to have an interest in determining how the corporation runs--have no right to run the corporation or determine its goals. [FN32] On its face, then, the large modern corporation appears to be a sizable aggregation of power uncontrolled by any group of people.

Yet there is no discussion in the cases of why the politically unanswerable business corporation is in need of protection from the state, while municipal corporations--city governments, in ordinary English--even if democratically elected, are scary things from which individuals must be protected. [FN33] Similarly, one can search the entire corpus of constitutional corporate law for any hint of the possibility that the aggregation and collective-action problems or the other problems of unchecked governmental action that drive the liberal theory of limited government might also affect our other, and often larger, collective enterprises.

Rather, corporations are seen as no different from individuals, or perhaps groups of individuals, petitioning the state from a position of weakness. Thus, the Court saw no need to consider whether business corporations might be any different from Revolutionary American citizen participants in Committees of Correspondence in Citizens Against Rent Control v. Berkeley, [FN34] or from "groups" of citizens in California Motor Transport Co. v. Trucking Unlimited, [FN35] or the "people" itself in Eastern R.R. Presidents Conference v. Noerr Motor Freight, Inc. [FN36]

Several theories of the corporation have been used to support the classification of the corporation on the private, nongovernmental, individual side of the great liberal divide between state and civil society. The oldest theory--conventionally referred to as the "aggregate" theory--deemed the corporation to be "really" just its "members" and, accordingly, entitled to the same protection the "members" would have. As Justice Marshall succinctly stated:

(Natural persons) can (not) be supposed to be less the objects of constitutional provision, because they are allowed to sue by a corporate name. . . . . (T)he corporate name represents persons who are members of The corporation. [FN37]

A more recent example of this reductionist theory can be seen in Central Hudson Gas & Electric v. Public Service Corp., [FN38] where Justice Stevens characterized the corporation's promotional advertising as "expression by an informed and interested group of persons of their point of view"--the corporation, in this view, is just a "group of persons." Similar disregard for the institutional existence of the corporation can be seen in Justice Brennan's Austin concurrence when he stated, in apparently blind disregard of the applicable corporate law, that funds in the general corporate treasury are "his (a shareholder's) money." [FN39]

A variant of this theory views the corporation as nothing but property, so that protecting the corporation is seen as protecting the property rights and thus freedom of the (presumably human) owners. [FN40]

The complementary "entity" theory of corporate personality treats the corporation not as a group but as an individual itself--the corporate person, a rights-bearing individual with its own interests and goals. [FN41] This seems to underlie the announcement, without any analysis, in Santa Clara County v. Southern Pacific Railroad Co., [FN42] that corporations are entitled to the protection of the Equal Protection Clause of the Fourteenth Amendment, and the barely more reasoned application of the Due Process Clause to railroad corporations in Minneapolis & St. Louis Railway v. Beckwith. [FN43]

As the Fourteenth Amendment grew to incorporate much of the Bill of Rights, the Court has from time to time considered whether rights extended to human beings should also be extended to corporations. In most cases, the Court simply applied doctrines designed for individuals to the corporation, with no discussion or consideration of how the corporation, as a collective entity, would use them or to what end, or of whether protecting corporations was supportive of, compatible with, or antagonistic to protecting human freedom. [FN44] Thus, in United States v. Martin Linen Supply Co., [FN45] the Court extended Fifth Amendment double jeopardy rights to corporations in a discussion that assumed that moral arguments justifying the importance to human beings of freedom from excessive "anxiety," "insecurity" and "personal strain" can be applied to corporations without modification. [FN46] In a few cases, the Court labeled particular rights--most prominently, the right against self-incrimination--"personal" liberties and did not extend protection to corporations. [FN47]

But the analysis remained unsophisticated. Corporate entity theory was invoked to call corporations "artificial persons" that should be treated as if they were private individuals. Aggregate theories were invoked to label them private, not the result of governmental action, and to claim that they are no different from the individuals who own their shares. Property theories were invoked to assimilate corporations to real estate, as if the corporation were a thing owned by a human being whose freedom is protected by protecting his control over his objects. Or the corporation was simply ignored, and the entity treated as if it were a human being or as if it were its shareholder.

Strand one, then, is an ancient tradition of seeing the world as composed of private individuals and governmental entities, with the corporation firmly on the former side of the great divide.

2. The Reification of Speech

The second constitutional strand is the combined effect of the reification of speech in First Amendment doctrine with the simple equation of money and speech. In Buckley v. Valeo, [FN48] the Court equated contributions of money with protected speech: spending money to hire someone else to speak, to publicize a third party's speech, to run an advertisement, to litigate a case, or to lobby an administrative body--all are the same, for constitutional First Amendment purposes, as writing the Great American Novel, investigating a political scandal, publishing an editorial, or for that matter distributing dirty (but not obscene) pictures.

Without this equation of speech and spending money, an issue might have arisen as to whether a corporation even could speak. [FN49] Some view corporations as a pure legal creation with no existence outside the law or our over-heated imaginations: how can a fiction, even a legal fiction, speak? Others, more plausibly, admit that corporations exist; after all, most of us get our paychecks from one. Still, even on this realistic view, [FN50] a corporation is not obviously the sort of thing that can speak: some human must speak on its behalf, and there might have been some issue whether it was the human speaker or the corporate entity that was entitled to the protection, particularly on views of the First Amendment that emphasize the personal creativity or autonomy of the speaker.

In contrast, it is quite a familiar notion that corporations can possess and spend money; indeed, among the strongest evidence that corporations do exist is that they spend quite a bit of money. Furthermore, while one might argue that the human who actually creates the corporation's "speech" is the speaker, corporate money belongs to the corporation and no one else. (Theorists and cases seeking to show that corporations do not exist often refer to the corporate money as shareholder money. However, this characterization is simply wrong as a matter of corporate law: one of the most fundamental characteristics of a corporation is that the shareholders have no right to corporate assets.) [FN51] Thus, the idea that the "corporation" is the spender is a far less troubling one than the idea that the "corporation" is a speaker: while a corporation may have "no soul to damn, no body to kick" [FN52] and no mouth with which to speak, it undeniably has a bank account from which to spend.

The assimilation of spending money to speaking, then, points in the same direction as the general constitutional amalgamation of corporations with individual citizens: both suggest that corporations should be given the same protection as speakers as are citizens.

3. Protecting Speech Instead of Speakers

First Amendment doctrine also points in the same way in its increasing shift from protecting the freedom of people to protecting "speech." The traditional notion that freedom of speech involved freedom of speakers from censorship seemed to lead to some troubling results: citizens might want to hear nonresident foreigners, whose rights presumably are not protected by our Constitution, or at least by our courts. [FN53] One solution to this problem would have been to focus on the limited powers of our governments, which do not include censorship. That, however, would have led to difficult standing issues: censorship injures the entire citizenry.

The court went in a different direction, emphasizing not the duty of the government but the rights of listeners. In Lamont v. Postmaster General, [FN54] the Court unanimously held that Americans who wanted to read foreign magazines--not just the purveyors of the stuff--were entitled to protection. As the concurrence put it, "(i)t would be a barren marketplace of ideas that had only sellers and no buyers." [FN55] Then, in Kleindienst v. Mandel, [FN56] a case involving the attempt of a foreigner to enter the country despite a U.S. blacklist, the Court again focused on the rights of the Americans who wished to hear and debate with Mr. Mandel, assuming that as a nonresident alien Mr. Mandel had no relevant rights of his own.

If the foreign-speech cases emphasized the rights of listeners, the commercial-speech cases quickly seemed to center around the rights of the speech itself or the marketplace for speech without regard for speakers or listeners. Thus, in Virginia State Board of Pharmacy v. Virginia Citizen's Consumer Council, [FN57] the Court held that the First Amendment bars a state from banning pharmaceutical price advertising, stating that the First Amendment's protection "is to the communication, to its source and to its recipients both." [FN58]

Virginia State Board of Pharmacy's transformation of the First Amendment to a protector of "speech" rather than "freedom" again eases the simple extension of First Amendment rights to corporations. Were the First Amendment thought of as protecting personal freedom, an issue would immediately arise as to whose freedom is protected by creating a corporate right to free speech. Is such a right similar to an individual right of free speech, or is it more similar to a right of a governmental agency to spend tax dollars lobbying for more tax dollars? [FN59] The governmental agency's right has rarely been thought to be included in the First Amendment, and indeed generally has been considered a threat to the very values of democratic self-rule the First Amendment is meant to promote; [FN60] corporate lobbying might also have been considered potentially dangerous as well, had the issue been so framed. [FN61] If the issue is the rights of the "speech" rather than the freedom of the speaker, however, the focus shifts to considering whether corporate lobbying is the type of subject matter that might come within First Amendment scrutiny.

Virginia State Board of Pharmacy also made a substantial contribution to the subject-matter issue. Prior cases had suggested that commercial speech might be outside the scope of the First Amendment. Perhaps seeking to distinguish its constitutionalization of a particular theory of free-market economic regulation from the Lochner Court's, [FN62] the Virginia State Board of Pharmacy Court rejected any distinction between political debate and pharmaceutical price advertising: the "free flow of commercial information is indispensable" both to the functioning of a free market economy and " to the formation of intelligent opinions as to how that system ought to be regulated and altered." [FN63]

The third constitutional strand, then, is a broad shift in First Amendment theory from a focus on the freedom of citizens from censorship directed at imposing social conformity or suppressing political dissent, to disembodied notions of the "free flow of information." The speech clauses of the First Amendment were no longer understood as extensions of the religion clauses, protecting dissenters from governmental pressures to conform, but rather as a radically separate, Lochnerian enactment of a free market of information. [FN64]

B. Joining Together in Bellotti: Ignoring the Speaker

These various analyses united in the case that continues to define corporate speech law: First National Bank v. Bellotti, [FN65] which held definitively that corporations are entitled to the same First Amendment protection as individual citizens. Bellotti reached this conclusion without any explicit discussion of corporate theory by focusing on the speech rather than on the speaker. The "speech" in question--corporate expenditures to finance advertisements in a referendum campaign--unquestionably would be protected speech were it not corporate financed, said the Court. [FN66] Accordingly, the issue, as the Court saw it, was whether this "speech" loses its protection because of the identity of the speaker (or, more precisely, the funding source). [FN67]

In keeping with the general tradition of constitutional corporate law, the opinion then simply concluded that corporate financed speech is not relevantly different from human speech. Two assumptions appear to underlie the Court's presentation of this claim as if it were self-evident. First, here as in many First Amendment decisions, the Court is abstracting from the power struggle that is part of the essence of politics. The statute in question did not bar the distribution of factual information: any corporation remained free to respond to press or citizen inquiries regarding its corporate view on the benefits or lack thereof of the tax at issue. Rather, the legislation was an attempt to suppress corporate attempts to persuade by advertising--it was directed not at information so much as at rhetoric.

The distinction between rhetoric and information is itself largely rhetorical in most areas of First Amendment law. If a purpose of the First Amendment is to protect the processes of self-government in a democracy, it is precisely rhetorical persuasion that it must protect most closely, for it is rhetoric, not information, that drives the process of democratic decision making. But rhetoric and information do have critical differences, one key one being that money and the volume it buys make more of a difference in the former sphere than in the latter. By characterizing the issue as one of "informing the public," the Court avoided the difficult question of whether a debate can be distorted by imbalances in wealth. At least on a rhetorical level, more information seems obviously better, even if the information is one sided or repetitive. [FN68] The same cannot be said of more propaganda, more rhetoric, more advertising or more persuasion: the side that has greater access to those tools has an obvious advantage. Thus, the Court trivialized the important issue at stake, which is not the neutral distribution of information but rather a political power struggle in which rhetorical volume is extremely important--and in which corporate agents will view their duty as requiring them to purchase the profit-maximizing volume using money with no clear owner and with only narrow marginal cost considerations providing limitations.

Second, the Court appears to be assuming that the corporation in question is a legitimate and independent participant in the political debate. Indeed, Justice Powell's opinion appears to have relied on an implicit theory of the corporation as an individual citizen with a mind and will of its own, entitled to the deference due to an autonomous, self-governing individual in a liberal democracy. Thus, in footnote 22, he claimed that closely held corporations arguably "might welcome" a shift in the tax burden to impose a greater share on wealthy individuals--presumably including the very owners of these closely held firms. This presupposes an extraordinary degree of independence for the closely held corporation and its managers.

But once it is admitted that a political debate involves more than just information, surely no one would claim that it is self-evident that it is "indispensable" that a democracy allow its political process to be influenced by the paid advertising of, for example, a domestic governmental entity, [FN69] a foreign communist, [FN70] or a paid lobbyist for the Libyan government. We commonly allow the democratic process to protect itself from its own creature by restricting governmental participation in the political process--barring governmental agencies (and sometimes even private grantees) from using tax money to lobby for additional tax money, for example. [FN71] In contrast, many people ultimately will conclude that the well-known problems of censorship outweigh any benefit that might result from limiting foreign propaganda. Whichever way one comes down in the end, however, there is a potential problem if institutions that are meant to be controlled by the political process or outsiders who have no right to participate at all come to have excessive influence on the internal debate. The Bellotti Court assumed, without discussion, that corporations are legitimate participants in our political debate, more like citizens than the government or foreigners.

Similarly, Justice Powell's lack of interest in the claim that corporations may take positions different from their shareholders again suggests that he simply assumed that the corporation as an entity is a legitimate participant in the debate, as if it--not the people who work for it or own its securities-- were a citizen. If no "greater solicitude" [FN72] for the views of shareholders is warranted when the corporation enters into politics than when it makes a more ordinary investment decision, then very little solicitude for those views indeed is warranted: it is black letter corporate law that shareholders have no right at all to determine ordinary business decisions. [FN73] It must follow, then, that Justice Powell's protection of the lobbying rights of the corporation was not based on the rights of the shareholders to join together to amplify their collective voice.

On the other hand, the opinion clearly attempted to avoid the problem of corporate theory altogether. Speech, not speakers, said the Court, is what is protected. Thus, one need not consider the speaker at all: we are relieved of the obligation of thought.

C. The Relevance of Corporate Governance

Finally, the constitutional doctrine remains studiously ignorant of state and federal law regulating corporations. The Court holds that corporations have a right to speak--without regard for who makes them speak, or with what resources they speak, or, most important for our purposes, how the law directs them to speak.

The remainder of this Article seeks to demonstrate why the speaker does matter: speech rights given to business corporations are quite different from speech rights given to human beings and can be expected to distort the political process in ways that are antithetical to any theory of the First Amendment.

The Court's First Amendment jurisprudence requires that states allow corporations to lobby. Ordinary corporate law and market pressures then require management to lobby whenever it is profit maximizing to do so, regardless of whether management--or shareholders--believe that doing so is in the best interests of the system, the economy or even the people who are the ultimate beneficiaries of the shareholders. [FN74] Corporations, as a result, become a massive force for the right to externalize one's costs onto others, [FN75] legally required to distort the political decision making process of the people, without any citizen being in a position to exercise political judgment regarding the desirability or limits of this strange program.

III. Group Speak

As we have seen, constitutional law largely ignores the special character of corporate speech. At most, it treats corporate speech as an instance of ordinary group speech and the corporation as an intermediate institution like those to which the accolades of de Tocqueville are directed. In this Part, I review some of the problems commonly seen in group speech generally; in Part IV I then argue that corporations are distinctively different from the other intermediate organizations of civil society in constitutionally significant ways.

By now, many of the problems of group speech are well known. [FN76] I discuss just a few: the boundary problem, the aggregate problem, and the leadership problem.

A. The Boundary Problem

Any time a group makes a collective decision or an individual purports to speak on behalf of others, questions arise regarding the composition of the group--who is in and who is out, whether the boundaries could have been drawn in a different way and how the boundary drawing decision affects the substantive decision.

When the group is making a collective political decision, the boundary issue arises because defining the group often determines its majority and thus the group decision. Thus, for example, a referendum on the status of Northern Ireland would have predictable, and different, results depending on whether the group entitled to decide is understood as the inhabitants of Ireland, the citizens of Northern Ireland, or the Irish Protestant and Catholic communities taken separately. [FN77] Similarly, the character of American schools and urban/suburban areas appears to be heavily influenced by the boundaries of the groups that fund and govern them; were our schools, zoning, social welfare or transport systems regionally or nationally funded or planned, they surely would be quite different. [FN78]

Even when no formal political voting process is involved, the boundary issue arises. Crenshaw, for example, has criticized Black leaders for speaking on behalf of African-Americans in a way that hides the voice of women and feminists for speaking on behalf of women who do not ever seem to be African-American: she is centrally concerned with the critique and definition of these and other group boundaries by which we define ourselves. [FN79]

In these spheres, then, the boundaries of a group can profoundly affect the legitimacy of speech or other actions on "its" behalf. A group defined too broadly may have permanent minorities within it that have needs or wills that are never met; one defined too narrowly will tend to ignore the interests and desires of those outside it.

B. The Aggregation Problem

The decision of a group, as everyone who has ever participated in a committee knows, can be quite different from the decisions of the members taken individually. Political theorists have long struggled with a version of the aggregation problem as a central part of the theory of legitimacy--how can an individual be free and also subject to the law of the group? They have responded with two general families of solutions: either the laws must reflect the inner nature of a uniform group, as Plato, Rousseau and various nationalists, including the liberal nationalists, have suggested, or the laws must be limited to minimalist goals that all rational beings would agree on, as the social contract tradition has argued since Hobbes. In either case, the aggregation problem drops out because of general agreement on at least the important political issues.

Thus, classic liberal nationalism assumed that democratic government was appropriate only for a group that was fairly uniform and united to start with-- that is, that majority rule is legitimate only if there is substantial agreement on the most important issues. [FN80] The aggregation problem, then, is solved by drawing appropriate boundaries to eliminate significant disagreement. Unfortunately, such boundaries normally cannot be drawn, and even if they are, disunity immediately reappears within the newly defined group. [FN81]

In contrast, the social contract traditions of Hobbes, Locke, Kant (at least in the Rawlsian reading), Rawls, and of the libertarians of the left and right limit the scope of government to goals to which all rational men would agree. [FN82] With general agreement on these goals, neither the boundary nor the aggregation problem will be of major importance--a liberal state exists independent of civil society. Instead, politics in the liberal state is reduced to administration: to discussion over the best means to reach an agreed-upon end. In the liberal state, politics is not about collective ends, national purpose, or achievement of the good life. Instead it is limited to relatively noncontroversial topics such as how to increase efficiency so as to allow private citizens to pursue private wealth or whatever such wealth can bring.

Such discussion, of course, need not offer any particularly active role to the population: it may well be best done by the public-minded elite of the Federalist Papers, the brain trusts of the New Deal, or the judges and economists of the Chicago School, with elections serving mainly as a check on potential drift of the elite away from its agency role (that is, to prevent corruption). Alternatively, politics may continue to exist as a sort of log- rolling allocation of divisible goods in a roughly proportional way. [FN83] In either event, however, politics in the classical sense--as a public discussion of who we are and a public creation of public values--should simply disappear.

Furthermore, even if actual politics suggests that there is no general agreement on the ends of the state, this tradition suggests that the problem is still not one for politics. Legitimate government is that to which rational men would agree, in the subjunctive, not that to which actual people do agree, since the latter agreement is likely to be distorted by the unfair power relations of the historical past. What is needed is a correct understanding of what rational individuals would do in a fair bargaining position. That, again, is not an issue for actual messy politics, but rather for expertise--of philosophers and perhaps judges. [FN84] In the event, however, it has proven difficult to specify a role for government sufficiently limited to satisfy these theoretical requirements yet robust enough to allow a decent society.

Where unity is not available or imposed by boundary drawing or agenda setting, the social choice theorists have pointed out another level of difficulties: the best available group aggregation technique, voting, leads to inconsistent and often perverse results. [FN85] Even if it is self-evident that majorities should prevail over minorities, once procedural decisions, agenda setting and boundary drawing appear to determine the results, democracy loses its legitimacy.

In corporate law, much of the discussion of the fundamental principles that shareholders do not control the corporation, may not initiate even fundamental changes, and may not even vote on ordinary ones centers around analogous problems: when or whether decision making by management on behalf of a presumably diverse group of investors and other corporate participants is legitimate. If corporate speech is to be corporate at all, there must be a clear explanation of how the group decision legitimately can be made. [FN86]

C. The Leadership Problem

Here I refer to Michels's Iron Law of Oligarchy: the activists in any group will always end up speaking and acting on behalf of the group, and will always be different in critical and significant ways from their followers. [FN87] So the leadership of labor unions is significantly more likely to vote Democratic (in the United States) or Labor (abroad) than the rank and file. [FN88] The leadership in religious organizations is likely to take a significantly different view of the religion or religious issues than ordinary church members: at a minimum, in most church organizations the professionals or leaders are likely to view the church as more central to their lives than will those who are less fully involved. The leadership of ethnic organizations systematically values ethnicity differently from the membership. The interests of corporate managers are less than perfectly aligned with those of shareholders. And so on.

D. The "Exit" Solution

The boundary, aggregation and oligarchy problems are problems of all group activities, both of intermediate institutions and of the state itself. Nonetheless, for most voluntary organizations, they are manageable problems, limited in their effects by the ordinary processes of inter-and intragroup politics and competition.

When the NAACP speaks, its lobbying activities must be at least roughly correlated with the views of its contributors--otherwise they would cease to contribute. Similarly, if the problems of aggregation, boundaries or oligarchy cause a political party, a church or other intermediate institution to drift too far from the views of its constituents, those individuals are likely to cease supporting the group or to break off from it to form a new one.

In Hirschman's well known analysis, the problems of boundary setting, aggregation and oligarchy are all problems of "voice." [FN89] Voice, in Hirschman's analysis, refers to the internal political mechanisms by which participants direct the organization--voting, participation, debate and other methods of influencing or changing the leadership of the organization.

But institutional politics or "voice" is not the only method to control a voluntary organization. Where the problems of aggregating individual desires into collective action are not adequately resolved--that is, where participants find they need a "voice" but do not have it--participants will simply "exit" from the organization. That is, they can leave and start a new one that better reflects their goals. [FN90] By transferring their support to organizations that happen to meet their needs, individuals (in the aggregate) can control the relative strength and viability of institutions even when they have no internal, institutional political voice at all. [FN91]

Voice and exit, thus, roughly correlate with political and market methods of control over an organization, or, to use a metaphor from a different field, design and evolution. Voice allows the participants to consciously control and design an organization; exit (when combined with competition) allows them to control the system by market or artificial selection mechanisms.

Assuming that we in fact have a fairly broad selection of organizations with which people can affiliate, and leaving aside the various market failures, it seems safe to assume (in the absence of information to the contrary and for most purposes) that any given organizational leadership continues to represent to a reasonable degree of accuracy the views of those who have chosen to continue to affiliate with the organization. Even if we see strong evidence of problems in "voice," we may assume that the "exit" mechanism of artificial selection will keep the leadership relatively representative.

E. The Limits of "Exit": Market Failures

It is relatively easy, however, to identify instances where this comfortable story of evolutionary pressure solving the design problems of Michels's Iron Law of Oligarchy and Arrow's voting paradox will break down. The special status of state politics in liberal theory is perhaps the best known example. There is a key difference between the state itself and the nonstate groups over which it presides: the natural process of competition, growth, and decline that prevents other organizations from moving too far from the views of their constituents may work differently when the group is the state itself. Most obviously, it is extremely difficult, if not always impossible, for citizens to simply leave the state and join another one, so the "exit" option is less likely to provide any useful control on the leadership. [FN92]

Similarly, other organizations may be able to free themselves from their constituents in other ways. Indeed, some organizations may not have any obvious constituents at all: consider, for example, a fully funded foundation. As a formal matter, a foundation operates under the direction of its trustees, and they operate under the direction or control of no one, other than their (largely unenforceable) fiduciary duties to the foundation's purposes or beneficiaries. Voice may fail as a mechanism to control such an institution, because other than the trustees, no one has any authoritative voice at all. Exit, in this instance, is irrelevant: if trustees or staff members attempt to control the organization by leaving, the likely result will be that they will be replaced by newcomers more in line with the institution's current direction. Donors denied a voice may stop donating (exit), but that may have little effect if the institution is already well-endowed. The institution, then, may be best understood as a self-perpetuating entity relatively independent of any influences outside the board room, controlled only by the trustees and their sense of obligation to the foundation's purposes. [FN93]

Another common "exit failure" arises when an institution that exists and maintains its base of support because of one activity is able to enter into other areas as well. [FN94] Consider a fraternal organization that provides a significant economic benefit to members--financially significant networking, discount insurance or road-side assistance, for example (instances of organizations arguably fitting this model might be the AMA, ABA, AARP, AAA, Rotary, and so on). Even if the organization also engages in significant other activities--lobbying or political advocacy, for example--the enticement of the economic benefits may induce members not to exit if the total package is attractive, even if the advocacy elements are not.

More generally, organizations that offer a diverse package of attractions may be able to maintain large memberships (i.e., avoid exit) even though a significant part of their membership disagrees with various activities of the organization. The inducement to remain need not be financial: many religious organizations, for example, may attract members for reasons that have nothing to do with the political stance the organization takes. Thus, if representation or "voice" mechanisms break down (or, in hierarchical organizations, never existed), the organizational stance may vary quite radically from the views of the members without any significant exit. There is some evidence, for example, that the official views of Jewish communal institutions on Israel-related issues are quite different from the views of ordinary Jews, that the differences between the political positions of the mainstream liberal Protestant organizations and the conservative ones are more extreme than those of their respective congregants, and that the Catholic church's official view on birth control is not shared by most American Catholics. Relatively few people, however, are likely to exit these institutions because of these disagreements, and so the Iron Law of Oligarchy and other representation problems can blossom into their full glory.

This points out another issue--the problem of materiality or bounded rationality. For a member of a church or the AARP, the political activity of the organization may be simply immaterial: the reason why the individual joins is for fellowship, salvation, or discounts, and political action is simply not a large enough part of the package to determine the membership decision. Accordingly, the individual rationally may decide to subsidize political activity with which she or he disagrees--or even to remain ignorant of it--in order to obtain the other benefits of membership.

However, political activity that is immaterial relative to any given member may be quite material in the political sphere. Put crudely, politicians, and political lobbying, are cheap. One percent of the organizational budget will be immaterial to most organization members--indeed, in the corporate world, it would not even appear in public disclosure. But if the budget is large enough, it will generate a very large and powerful lobbying force that is not immaterial in the least in the political sphere.

F. Voice and Exit Summarized

To generalize, then, unconstrained exit will pressure organizations to reflect the views of their constituents to a reasonable degree. Organizations that depend on ongoing fund raising or membership for their support and that engage exclusively or almost exclusively in political, speech and lobbying activities fit this model best. Exit will assure that the organization continues to represent most members most of the time, even if it is not possible for any set of political activities to match the opinions of all members at any time. Thus, when the ACLU supported the rights of Nazis to march in Skokie and its constituents disagreed, it lost a vast proportion of its membership. [FN95] Presumably the remaining members and the leadership then became better aligned. Importantly, then, we can assume to a reasonable degree of confidence that most of the time political organizations of this type will represent their members even if the organization has no internal democratic procedures. The Court was correct, therefore, in allowing the NAACP to assert the free speech rights of its members in the seminal cases that began the expansion of the First Amendment to include corporate entities. [FN96]

At the opposite end of the spectrum, an organization that as its primary activity provides valuable and difficult-to-obtain services for its members may then find its political activities relatively unconstrained by the threat of exit. For example, one might feel obliged to belong to a professional organization regardless of its lobbying activities simply in order to obtain professional insurance at a reasonable rate. In that case, exit will not constrain the organization's political activities, particularly if they are small from the perspective of members relative to its primary focus. An outsider, then, should have no a priori reason to believe that the organizational view on any given issue reflects the views of the membership--a detailed investigation of the internal representative mechanisms is necessary. In light of the problems of bounded rationality, oligarchy and the like, even apparently representative governance schemes may not be enough to assure that the leadership and the membership remain on common ground.

IV. Corporations as Group Speakers

A. The View of Corporations as Groups

The classic critique of corporate speech treats corporations as a group like other groups, subject to the failures of representation described above.

On this analysis, corporations are understood as representing their shareholders. Both voice and exit problems have been identified. The exit problems center around materiality and bounded rationality: shareholders "join" a corporation to make money, and provided that corporate lobbying or other political activity does not have a materially negative effect on profits, they are likely simply to ignore it regardless of whether they agree or disagree.

Other scholars have emphasized, in various contexts, that the voice problems, particularly problems of inconsistency and cycling in aggregating individual shareholder preferences, might lead rational shareholders to give great discretion to managers. Conversely, the great discretion given to managers may place them in a position in which they can abuse their authority to cause the corporation to act in accordance with their own views even when those are clearly in conflict with shareholder views. [FN97]

While I am not aware of any attempt to discuss boundary problems in the specific context of corporate political activity, the issue has been hotly debated elsewhere in corporate law. One trigger for this discussion has been the state law innovation of "constituency statutes." These statutes create limited circumstances, generally restricted to hostile takeovers and similar transactions, in which corporate boards of directors are authorized to invoke the interests of nonshareholder participants to avoid what would otherwise be clear fiduciary duties to shareholders. While none of the statutes actually creates any enforceable rights for these nonshareholder corporate "stakeholders," the very acknowledgment that the corporation might include disenfranchised groups necessarily gives rise to questions regarding the proper boundaries of the corporation and questions regarding whether subgroups may be "silenced." [FN98]

The nexus-of-contracts theory of corporate personality, which reduces the corporation not to a group of shareholders but to a moment in the market, has much the same effect. [FN99] If employees, bondholders, customers, neighbors or other stakeholders are considered part of the corporation, or if the corporation is seen not as the shareholders joined together but as a nexus of contracts in which employees can be thought of as hiring capital just as easily as the other way around, then the boundaries of the corporation are no longer clear. With boundaries that may include nonvoting members, the broader corporation limitation of the franchise to shareholders creates an obvious prima facie problem. [FN100]

In the speech and political activity context, then, the primary boundary problem is this: if a corporation is deemed to include more than merely its shareholders, what mechanisms assure that corporate actions will be representative of those nonshareholder groups?

A second boundary issue has arisen with the increasing awareness that some extremely powerful shareholders pursue interests that appear to be contrary to the interests of the classic fictional shareholder. The much reviled short-term speculators and arbitragers--more properly, all the true portfolio investors--often do not seem to have the interests of the firm at heart and often seem to have the market power to coerce corporations to act according to their views rather than more traditional ones. Indeed, rational portfolio investors acting to maximize the returns to their own fictional shareholders will often act as if they reject the norm of profit maximization at the firm level or even the separateness of different firms: if Bell Atlantic profits at the expense of U.S. West, a diversified shareholder holding both firms loses out to the degree that consumers gain. For the portfolio investor, the only relevant boundary is between publicly traded securities (all of which can be thought of a single firm) and everything else. The obvious conflict between the classic notion of a shareholder as a being with no interests other than its stock in one firm and the market reality of portfolio investors seeking to maximize returns to their own fictional shareholders drives much of modern corporate law. As a boundary issue, it is expressed by questioning whether arbitragers or important portfolio investors ought to be deemed "true" or "real" shareholders. [FN101]

Each of these analyses treats the corporation as essentially similar to other groups and organizations that participate in our political life. Corporations may be well along the spectrum towards the unrepresentative end, due to agency problems or materiality and rational ignorance (although the ease of exit--the ease with which one investment can be replaced with another similar one--might suggest that even small differences may become salient). [FN102] Corporations may have more features of "exit failure" than many other organizations, due to the importance of the profit motive as the primary reason investors participate (although the emotional ties of most shareholders to their shareholdings are quite minimal and shareholder ease of exit, in terms of collateral consequences, is greater than in almost any other institutionalized group). [FN103] Business corporations may be better able to overcome free rider problems. They may have more money (although, of course, there are poor corporations as well as rich ones, and rich individuals as well as poor ones). [FN104]

But regardless of whether they attack or defend corporate speech, these critics fundamentally agree that the problems of corporate speech are simply those of politics itself. All political coalitions and attempts to aggregate individual views through politics face these problems--while business corporations may have the problems in greater or lesser degree than some other organizations, the issues that corporations face as political participants are similar to those that any coalition of any variety faces. These views, in short, agree that business corporations properly are classified as intermediate institutions, similar to the other "factions," interest groups, fraternal organizations, and coalitions that make up civil society in a democratic republic.

B. Corporations Represent Only a Fiction, Not a Group

The fictional shareholder view is significantly different. Corporate political participation is different in kind, not simply in degree, from the participation of other groups or individuals. Any coalition of individuals will suffer from the various problems of voice and exit, the problems of boundaries, aggregation, consistency, saliency and so on. Corporations have an additional and significantly different set of problems: they are legally required to represent not a group of people but a legally defined set of interests--the interests of a fictional creature called a shareholder that has no associations, economic incentives or political views other than a desire to profit from its connection with this particular corporation.

Corporate positions are not determined by debate and struggle among a group of people to define boundaries and manipulate internal procedures, to persuade each other or outmaneuver each other, in the manner of ordinary politics and ordinary intragroup struggle. Instead, corporate positions are determined by fiduciaries who are obligated both to set aside their own views and to ignore the actual views and interests of the other people involved in the corporation. [FN105]

Because corporate policies are not determined by politics, the corporation can become a peculiar type of monster or robot: an organization on automatic pilot with no human in a position to call it back from its logic to the road of moderation. [FN106] Similarly, because corporate decisions are made by fiduciaries on behalf of fictional shareholders, they reflect only a narrow and one-sided aspect of the human beings who stand behind the shares. Of course, as the group theorists from Michels to Crenshaw point out, any group tends to emphasize one aspect of the personalities of the people who make it up and to suppress other aspects of their multifaceted identities. [FN107] But a political party or political pressure group also must be responsive to the conversations, struggles and views of its participants/members: otherwise the members will vote out the leadership or just leave and the organization will wither and die. A business corporation, in contrast, exists independent of the particular individuals who stand behind its shares; individual political views will be filtered out long before they can influence the corporate stance.

C. Fictional Shareholders

State corporate law requires corporate directors to set aside their own personal interests and views and to manage the company in the interests of "the corporation and its shareholders." It is this requirement that corporations be managed in the interests of their shareholders (leaving aside the interesting issue of what the interests of the corporation itself might be) that lends surface credibility to the idea that corporations should be seen as groups of shareholders.

Corporate law is centrally about agency--but a peculiar agency that looks neither to the views nor the actual interests of the alleged principal. Corporate law excludes the actual shareholders from corporate governance, bars managers from considering actual interests of the shareholders and then considers this situation entirely unproblematic. In this section I explore the paradox of agency law without agency problems: in section D, I explain the implications for corporations as speakers.

1. Excluding Actual Politics

First, the business corporation is extraordinarily centralized. As a rule, shareholders are barred from governing the firm directly; state law requires that the directors, not the shareholders, manage the firm. Indeed, federal law has often (though not always) constrained shareholders from using the proxy machinery to mount even advisory campaigns on matters of ordinary business decision making. [FN108]

Thus, shareholders, who under ordinary principles of democratic theory, agency law, or even "best interests" theory might be viewed as the best source of information about shareholder interests, are systematically excluded from the decision-making process. While this exclusion has sometimes been explained on division of labor grounds, it is important to note how far it goes: even to the core decision that every shareholder must make. Shareholders, if they have no other expertise, must at least be specialists in determining whether to buy, hold or sell shares. Yet, the Delaware courts have clearly and consistently stated that a Board must exercise its own judgment in determining whether to sell the firm--it may not delegate even this most shareholder-of-all- shareholder decisions to the shareholders. [FN109]

Similarly, directors--unlike politicians, for example--may not fulfill their duty by doing just what they were elected to do. A director elected on a platform promising a particular course with respect to managing the firm must nonetheless continue to exercise independent judgment; a Contract With American Shareholders requiring the director to follow the expressed views of the voters would be void and a breach of the director's duty. A director elected by a specific faction is required, nonetheless, to represent the interests of all shareholders, not merely the electors, even where there are clear distinctions. [FN110]

In short, then, the law seems to assume that the actual views of the actual individuals who, directly or more often indirectly, stand behind the corporation's shares are legally and practically irrelevant. Shareholders are treated as if they were minor wards of a Dickensian trustee or the proletariat under the guidance of a Leninist vanguard, unable to speak for themselves.

2. Excluding Actual Interests

But that is not all. Not only are the expressed views of the shareholders irrelevant, but their actual interests are as well. Shareholdings in American corporations are concentrated, in the sense that a small number of individuals and institutions control the bulk of the publicly traded shares. But they are quite dispersed in a sense more relevant to my discussion: roughly half of all Americans own stock directly and even more do so indirectly. Institutional shareholders, which together hold over half our outstanding stock, [FN111] for the purposes of this Article should be viewed as themselves corporations (or corporation-like entities) acting on behalf of their own beneficiaries. Those beneficiaries include every American with a pension, a 401(k) plan invested in mutual funds, or an equity mutual fund--a large crowd indeed, and one that is not nearly so dominated by a small segment of the ultrawealthy as is direct stock-holding. Additionally, a significant part of the stock of American corporations is held by foreigners. In short, it is reasonable to assume that both shareholder views and shareholder interests are more or less as diverse as the electorate's (or at least the more affluent half or two thirds of the electorate).

Now, one might say that whatever else shareholders want in the rest of their lives, surely they all want higher share values. [FN112] But human shareholders who are also neighbors or employees or customers or friends may have other commitments beyond an extra nickel in the quarterly dividend. Even on purely economic issues, since shareholdings in this country are not only wide but shallow, many shareholders will find that their basic interests are aligned more with employees, stability or customers than with the highest possible value for their shareholdings: a decrease in your phone bill is likely to be worth more to you than the commensurate drop in the price of the telephone company shares held by your pension fund. Only foreign shareholders with little connection to the American economy or politics beyond their shareholdings approximate this conventional image of a shareholder always interested in higher stock returns. [FN113]

Nonetheless, despite the obvious diversity of shareholders, neither corporate management nor courts ever find it necessary to determine who in fact are the actual human beings behind the shares of a corporation prior to discussing the interests of the shareholders. Rather, corporations are managed as if shareholders were all the same.

Clearly, then, corporate law contemplates a very different sort of politics within the corporation than we know outside of it. In ordinary politics, we rarely assume that the (entire) people has a single interest; rather, assertions of such a unity are generally rightly regarded as dangerous precursors to totalitarianism. Only fascists, Leninists, and perhaps some who think they have privileged communication with God believe that there is an interest that everyone shares and that the political elite is required to pursue without regard to the actual views of the citizenry. Inside the corporation, in contrast, the Vanguard of the Shareholding Proletariat is not merely entitled but required to act in the class interest, regardless of any false consciousness that might lead the actual voters to other views. [FN114]

The paradox of this Leninist politics of the corporation is compounded by the very fact that has led most theorists to dismiss it as a problem. In actual votes, most shareholder elections are won by margins found in ordinary politics only in the plebiscites of Napoleonic, fascist, communist and similarly undemocratic regimes. That is, shareholders appear to act as if they really were all the same, with a Rousseauian general will or a Leninist class consciousness. Within the confines of corporate law, then, discouragement of shareholder participation seems entirely appropriate: shareholder votes, which are enormously expensive, are almost invariably predictable without regard to any information about the actual voters. In the absence of an immediate prospect of a cash payment, management wins. With cash, the side that offers the most cash wins by the same overwhelming margin. In short, not only does the law presume that shareholders are monolithic--shareholders act as if they were. But they are us, and we differ on almost everything important; how can this be?

3. The Poverty of Agency-Cost Theory

For over half a century, since Berle and Means, the central theme of academic corporate law has been agency failure (although the jargon is newer). But the agency problems with which corporate law scholars have concerned themselves are staggeringly primitive: the concern of corporate law is agents who act in bad faith, who steal from their principals or are simply incompetent.

The problem of corporate speech is a problem of role morality. The actual speakers--the lobbyists, advertising copy writers, lawyers, executives, and publicists who speak on behalf of the corporation--speak as agents, not on their own behalf. That is, their roles demand that they set aside their personal views and act as professionals, seeking the most effective means to promote their clients' views. [FN115]

Speech by agents is not unusual. Indeed, it is simply a special case of the representation of another's interests, an inherent aspect of professional service. Many areas of law and professional ethics struggle with the problems of understanding and interpreting clients--these have been the central issues, for example, in the fields of medical ethics (determining how best to care for patients when patient and doctor disagree on the best course, or when it is difficult or impossible to determine what the patient wants); family law (especially in the area of representing children who are considered unable to decide for themselves); poverty law (where much ink has been spilled insisting that attorneys not substitute their own judgment for that of their clients), and so on.

In general, the legal and ethical rules follow a consistent pattern: the professional is restricted from making value judgments for the client, but instead is directed to find the best way to pursue the client's goals. [FN116] Brokers are to take as given a client's risk preferences and financial goals and find the best way to reach them. Lawyers "zealously assert () the client's position": [FN117] the objective to be sought is for the client to determine, not the lawyer. The client, not the lawyer, settles a case or takes a plea. [FN118] Similar distinctions between the goal--which is the client's--and the means--which are for the professional to determine--underlie the informed consent movement in medicine. Judges attempt to interpret statutes, contracts, estate plans and constitutions in accordance with the intent of their authors, setting aside the judge's own preferences to the degree possible.

The problems, similarly, follow a consistent pattern; they fall into two categories. The conceptually more difficult problems are the interpretative and communicative ones: how does the professional figure out what the client's goals are, especially when the client may not know or able to communicate them? [FN119] What does it mean to fulfill contractual intent, when the contracting parties never considered the issue? How do we know what an incompetent person, at the end or beginning of life, would want a doctor to do, if the person were capable of wants at all? [FN120] Does Congress have a collective intent, and if so, how do we identify it? What can a poverty lawyer (or a doctor) do to ensure that it is the client's goals, not the lawyer's (doctor's) preconceived notions of what those goals are likely to be, that drives the litigation (treatment plan)? These are all issues relating to the problem of determining the goals of clients, who often are inconsistent or inarticulate in setting and communicating them. [FN121] The issues are, on an individual level, precisely equivalent to the problem of politics on a collective level: creating an agreed-upon set of goals.

The second set of problems are those of administration. Administrative problems, in turn, also subdivide into two basic categories. The first is the problem of technical expertise: given a goal, what is the best way to reach it? The second is the problem of corruption, [FN122] or in the modern jargon, agency costs: how do we ensure that the professionals in fact are pursuing the goal they are supposed to aim at, rather than, for example, simply lining their own pockets?

The corporate law literature has limited itself to only these administrative parts of the agency problem. Business organizations share all the problems of administration, and the enormous literatures of corporate finance, managerial technique and corporate law are devoted to solving them. But the problem that is central to this Article is the one that remains even when administration is perfect: that of politics, of setting the goal, rather than reaching an already determined one.

There is no discussion in corporate law, unlike family law, professional responsibility or medical ethics, of how the agent acting in good faith is supposed to figure out what the client's interests are, or of how to act when the professional's assessment of the client's interest differs from the client's. "Beyond the Best Interests of the Shareholder" has yet to be written; there is no advocate of shareholder autonomy seeking to "return" control of the large corporation to the actual humans behind its shareholders. Indeed, it is difficult to imagine even what such a theory would look like. A publicly traded firm run by the humans behind the shareholders would be neither firm nor corporate, as shareholders of varying abilities or commitments continually shift in and out of ownership positions.

For corporate law and corporate finance, the interests of the shareholder are the subject of expert articles full of mathematical symbols: logical deductions rather than philosophical meditations or attempts to bridge the communication gap between professional and client. Clients in other areas of the law are different and complex, sometimes even conflicted and inconsistent. The shareholder is not. Agency cost theory thus treats the interests of shareholders as deeply unproblematic and deeply antipolitical. In contrast to the goal debates of politics, corporate administration centers on how to get there, not where we are going. Given the diversity of shareholders and given the broad impact of corporations on our civic life, this is again puzzling and implausible. Why are we suddenly all the same in just this one aspect of life?

D. Replacing Politics with Administration

I have argued that the fictional shareholder is the resolution to the paradox of the unchanging and monolithic firm fronting for a shifting and amorphous mass of shareholders. [FN123]

Politics is suppressed in the corporation. The agents of a corporation speak on behalf not of a group of human principals who more or less consistently consider conflicting goals, apply some more or less imperfect decision-making procedure and reach a collective conclusion about which to pursue, but rather on behalf of a legally mandated principle, one-sided and clear voiced. The cacophony and polynomia of politics [FN124] and individual psychology [FN125] is silenced, replaced by a single, conflict-free, essentially unified, entirely consistent, and therefore utterly inhuman, goal. The humans who stand behind the shares have various and conflicting goals, as all people do: they want their shares to increase in value, of course, but they may also want decent jobs for their kids or neighbors, attractive and safe cities, a clean environment, and other things that, from time to time, conflict with the increase in value of their shares. The corporate law system eliminates all these conflicting goals, leaving the agents with a simple and clear directive: maximize shareholder value. [FN126]

The shareholders on whose behalf the board of directors requires its hired underlings to speak are not human beings in the full multiplicity and variegated splendor of post-identity (or even pre-identity) politics. Indeed, the shareholders whose needs are served by the paid agents of corporate speech are not human at all. Rather, they are legal fictions, created by the law (with the help of the market), whose one-sidedness makes the essentialism of the most essentialist of identity politics look like Foucault in sexy dressed drag. This shareholder, unlike the humans behind it, is an investor, pure and simple: it seeks to maximize the present discounted value of the future returns on its holdings. That is it.

Because the fictional shareholder has only one goal, it need make no value judgments. Politics is simply unnecessary, or, at least, simply a reflection of the problems of imperfect knowledge and limited rationality in the world. Were all (fictional) shareholders fully informed and fully rational, they would all agree--quite unlike ordinary citizens, who would still have to decide what kind of a society they wish to live in even if information were free. When people agree on the goal--as the fictional shareholder does by virtue of its stripped- down essentialism--politics is just administration; just a matter of setting the experts to work and ensuring (as corporate law has tried to do since Berle and Means) that the experts work for the client and not (just) themselves. Where goals are all agreed upon, the only purpose of democracy is to police corruption. Since most agents are not corrupt, most elections will be overwhelming, most clients will defer most of the time to the expert professionals, and corporate law's model makes sense.

The Hobbesian foundation, however, is the underlying agreement that as shareholders, all investors, whatever else they may believe in other areas of their lives, agree that the only important goal is the maximization of the discounted present value of the risk-adjusted cash flows associated with their stock. The law eliminates all other possible purposes. Shareholders, except by unanimous vote, cannot change the goal or values of the corporation. A mere majority of the shareholders simply has no authority under modern corporate law to manage the firm, or, more particularly, to direct the managers to give any significant weight to any value other than shareholder value maximization. This legally imposed unity of purpose is the underpinning for the whole structure. [FN127]

Maximization of shareholder value is (largely) a technical problem of administration, not politics. Accordingly, once this goal is accepted as the decisive directive to which corporate managers must respond, it is entirely appropriate to disempower the actual people behind the fictional shareholder and leave the management to the managers. Technical decisions--unlike value judgments--are best left to the experts.

Corporate law, then, replaces politics with administration. The role of the (real) shareholders is to police the problems of administration: gross incompetence, self-dealing, and the other problems of agency costs.

Because the people behind the shares have no practical authority to vary the single goal on behalf of which corporate managers must act, corporate speakers are agents answerable to a principle, not a principal. The managers' role--like that of any agent--requires them to restrict their attention to means rather than ends. However, unlike other agency relationships, here there is no human principal who can conclude that the time has come to change the balance or modify the goal. Only if corporate agents violate their role obligations will the corporation ever consider any good other than share value maximization.

1. The Inhumanity of the Fictional Shareholder

The fictional shareholder is not completely a modernist fantasy of internal unity. That would be too much to ask even in the political core of the liberal market place. But its variegation is quite limited. For virtually all purposes, there is only one significant variation: sometimes the law (and nearly always the market) views the shareholder as a diversified investor with only a small part of its wealth invested in the securities of the particular firm, but more often the law (and nearly always managers determining their role obligations) views the shareholder as undiversified, all of its wealth (the present discounted value of which it always seeks to maximize) tied up in the security in question.

Sometimes the differences between the diversified and the undiversified shareholder are enormously important--indeed, they drive much of the litigated controversies in corporate law. Diversified and undiversified investors, the theories of corporate finance tell us, have radically different views of risk. [FN128] Managers seeking to act in the interests of their shareholders will think quite differently about their role obligations depending on which of the two visions of the imaginary shareholder they have in mind. Likewise, the cases offer instances of courts confusedly switching between the two essences, using one to justify one result and the other to justify a different one. [FN129]

However, for the purposes of this Article, the two versions of the fictional shareholder are more alike than dissimilar, and both are essentially inhuman. Since the fictional shareholder is just an investor, it is immortal and time indifferent--the market allows any investor to transform future income into present income, short term gains into long terms ones, and so on, simply by applying the correct discount rate. It is context indifferent--since money is perfectly fungible, a pure shareholder is indifferent between money earned in Salt Lake City or Cambridge; Flint or Manilla. It has no commitment to particular enterprises: so long as the investment is on the capital frontier, offering the appropriate risk adjusted rate of return, one project is as good as any other. Tin cans and insurance, news magazines and amusement parks--what the company does is a matter of entire indifference. [FN130] It is universalist in the modernist, not the post-identity, sense: the fictional shareholder recognizes no boundaries, professes no nationality (or, more precisely, will change nationalities at the current or future monetary exchange rate), has no religion, no community, no union, no gender and, oddly enough, no class: the invested funds of the unions are no different from the invested funds of the capitalists against which they struggle. It is, in short, radically uncommitted, cosmopolitan, deracinated, tied to no religion, language, nation or community. [FN131] Perhaps most important for bargaining purposes, the shareholder is fully mobile--able to leap borders (and professions, commitments and projects) at a single bound. [FN132]

This shareholder is a fiction--not in the sense that it is not real, but in the sense that it is a unidimensional story that can be told by another. Any competent professional can imagine and represent shareholder interests without worrying about listening to the human beings behind the shareholder speak for themselves. Indeed, given that most of the human beings at issue are not competent professionals, the professionals generally can articulate and pursue (fictional) shareholder interests better than the (human) shareholders themselves. That, after all, is what they are paid, handsomely, to do.

The fictional shareholder is quite real, at least in its portfolio version. It is Fidelity's Magellan Fund, or CalPERS, or, closer to home for most readers, the TIAA-CREF pension funds, required by law and the market to act in the interests of their own fictional shareholders--namely you but not you, a stripped down version of you and others like you reduced to nothing more than the desire to be as rich as possible in your (never quite arriving) [FN133] retirement. It is on behalf of the interest of this stripped down person that the professionally managed funds, and the professionally managed corporations the shares of which the funds hold, will act. Eternally vigilant, mobile, uncommitted to any investment or relationship, project or community, the professionals search the world over for a higher present discounted value.

But it is a fiction nonetheless, in the classic sense of a legal fiction. When the law demands that corporate directors and managers manage the corporation in the interests of the shareholders and the corporation, it does not mean the interests--let alone the will--of the actual human, multidimensional, conflicted, polymorphic, committed, split-identitied people who are ultimately the legal owners of the shares (or, in our increasingly institutional stock market, the human, multidimensional, conflicted, polymorphic, committed, split-identitied people who are the ultimate beneficiaries of the legal entities that own the stock). The humans are not consulted; they have only primitive, indirect and ineffective means of letting their perceived interests or actual will be known.

2. Enforcing the Fictional Shareholder's Fictional Will: Law, Stock Listings and Immateriality

Indeed, as a general rule, the law explicitly bars directors from consulting shareholders on issues of business policy: business decisions are to be made by the directors in the interests of the corporation and its shareholders, not by the shareholders or in accordance with the will of the shareholders. Nor, in determining the interests of the shareholders, may the directors play the role of a Platonic philosopher king examining the life situations of the actual people out there and determining that, whether they know it or not, their interests require some action or other. Directors are directed to act in the interests of the fictional shareholder, not the humans behind it. [FN134]

The law, however, is not the most important enforcer of the fictional shareholder. More important is the market, mediated through the institutional investors and the problems of limited information and bounded rationality. Markets work best and most efficiently for commodities: uniform items that differ only in a few, easily described features. The stock market is a classic market for a commodity: one share is identical to another share of the same company, and, more importantly, for diversified investors following contemporary investment theory, one company's stock is largely fungible with another's, provided that both have similar risk-reward ratios. Financial investors, accordingly, need look only at a few, relatively easily obtained pieces of information in order to make their decisions.

Fully committed and variegated persons have only a limited amount of time and energy to expend in managing their investments--especially when those investments are a relatively small part of their wealth, let alone their personality. If you have a few thousand dollars to invest, you are far better off putting your energies into your job or your family rather than detailed investigations of what is happening to your money.

Accordingly, reasonable people rely heavily on easily monitored proxies when investing. Investment advisors describe companies in a page or two, and corporate financial statements exclude expenditures that are "immaterial" to an investor interested only in the company's financial future. Most small investors invest via an institution that in turn must compete for funds based only on the most simple information about itself.

When investors limit their view in this way, as they must, the market makes fictional shareholders out of real people. Even an expenditure on political speech, lobbying, or law reform that is large enough to have a dramatic effect on the political outcome--the telephone companies' campaign to influence the new telecommunications order; or corporate efforts to retain depletion allowances, affect revisions in the EPA, or limit liability in mass torts--is highly unlikely to be a significant part of the earnings of a large publicly traded corporation. Investors are unlikely even to learn of the expense. Ordinary accounting principles do not require that immaterial expenses--usually defined as five percent of earnings--be disclosed, even if they are quite material from the perspective of the political or regulatory process. Accordingly, the expenses will be invisible to the investment community. Even were they not, information overload would prevent a diversified investor from paying much attention to the expenditures. In any event, most individual indirect investors would find that their interest--filtered through the institutional layers--is far too small to be worth any attempt to make their will known. [FN135]

Sometimes, of course, corporate political activity is public by its very nature. Lobbying can be done behind the scenes, but law reform often requires persuading the public--think, for example, of the cigarette companies' campaign for "freedom of choice." But even when investors are aware of the company's political activities, as they must be in the cigarette case, they will ignore them.

Consider an investor who, like many Americans, believes that cigarettes are addictive and kill their users, or at least make them reek, and that companies that deliberately attempt to make people into cigarette addicts are immoral, or at least distasteful. Such an investor may well be willing to avoid direct stock holdings of Philip Morris or R.J.R. Nabisco, despite a broker's advice that they have records of outstanding financial performance. But how many such investors would sell each mutual fund that held .5% of its assets in tobacco-related investments? If the investor holds $2000 in the fund, the indirect interest in cigarettes is less than $10, a clearly immaterial amount from the investor's perspective. [FN136]

At the same time that the investor is concluding that avoiding R.J.R. or Philip Morris is not worth the effort, the other participants in the process will be forced to set aside their own personal beliefs. The managers of pension funds or other institutional investors, for example, have a fiduciary duty to maximize their own beneficiary or shareholder value; they would may well view themselves as obligated to invest in R.J.R. so long as it remains profitable and the stock remains undervalued. [FN137] Of course, if individual investors avoid the stock to a significant degree, that act in itself may drive down the stock price, thus forcing institutional investors to purchase it (and counteracting any influence the individual shareholders may have had on the management of the company). [FN138]

Even if the fund managers do not view themselves as morally or legally constrained to set aside their personal views regarding cigarettes, they may well be constrained by the market to do so: funds that avoid profitable investments are highly unlikely to show up well on the simple statistical trackings that are used by investors to track their fund managers. In any event, fund managers have far too much information that they are required to assess to spend much time analyzing corporate lobbying activities that do not affect the value of the investment.

The same story that is true of the fund manager is true of R.J.R. Nabisco's executives. They should view their own beliefs about cigarettes as basically irrelevant to the job they have to do, which is to make money for their shareholders. Accordingly, they may view themselves as obligated to stay in the cigarette business so long as it remains legal and profitable. Those who do not see themselves so obligated likely will quit and be replaced by some who do. [FN139]

Indeed, even if R.J.R. Nabisco were to withdraw from the cigarette industry, the likely result would be that the reduction in competition would make the industry more profitable and attract another, less fastidious, enterprise, with no ultimate change in the availability of the product. Since withdrawal is pointless, even those executives (or entire companies of executives) with qualms are likely to conclude there is not much point in acting upon them. Just as at the investor level, then, the market allows no mechanism for those with moral qualms to affect the actual behavior of the corporation: so long as cigarette smoking remains profitable and legal, the market assures that cigarettes will be produced.

In short, the personal views of all the corporate participants drop out. Most individual shareholders will find it impractical to monitor their investments for political positions they disagree with. Most institutional investors will be barred by law from doing so, or if they are not, they will be bound by market pressures (created by individuals who cannot monitor for political positions) to set aside politics. Most managers will perceive their duty to be to maximize shareholder profits without regard for other values; even those who do not will find that market pressures (created by the institutional investors) will force them to do so. So long as a significant supply of managers is available who are willing to do what we tell them they are supposed to do, and so long as a significant number of investors apply some standard of materiality in policing their portfolios, the individual views and politics of individual investors will disappear, leaving only the unidimensional, monolithic fictional shareholder. [FN140]

D. The Fictional Shareholder Talks

Corporate lobbying is a product just like cigarettes: so long as it remains profitable and legal, the market and the fiduciary duties of corporate directors and managers will assure that it is produced. Regardless of the political beliefs of the corporate participants, corporations will advocate any political position that maximizes shareholder value, even if the human beings behind the shareholders disagree, and even if they will be injured in the simplest economic sense.

If corporate managers are acting as the law and ordinary role morality expect of them, they will set aside their personal politics and cause the company to enter the political fray only in pursuit of profit. The decision to lobby, that is, should be made in just the same way as the decision to build a new factory--by evaluating the risk-adjusted present value of the expected net returns and comparing it to available alternatives. If it is more profitable to convince the legislature or the regulatory apparatus to remove a regulation than to pay the fine, or comply with the law, then the firm should lobby, regardless of the managers' private views about whether the legal change is a good one for the country as a whole. Managers who view their fiduciary responsibility to their shareholders as requiring them to profit maximize to the extent permitted by law will also view themselves as obligated, regardless of personal beliefs, to profit maximize by attempting to change the law.

Institutional money managers should view the issue of corporate lobbying in precisely the same way: they are required to maximize the value of their fund. If the underlying firm's lobbying activities are cost effective, they should support it, regardless of any personal views to the contrary.

Corporations are managed on behalf of the fictional shareholder with no interests other than its investments--not on behalf of real people. When the corporation enters into politics, it does so in the same way--on behalf of an imaginary investor with no other interests. This is not the result of our corporations being run by greedy, selfish semifelons; [FN141] it is the result of ordinary people doing their job precisely as they are supposed to. It is not the result of agency failure but rather of agency success.

But the effect is that corporations cannot speak on behalf of the people behind their fictional shareholders.

V. The Values of Fictional Shareholder Speech

Corporations are strange and unfortunate participants in our politics not because they take the wrong positions (although they often will), but because they must think about their positions in the wrong way. While real people must balance competing values, compare their own needs and those of others important to them, and make difficult choices between various aspects of our too-finite lives, corporations (or the role-constrained managers who decide for them) just maximize shareholder value.

No doubt profit-maximizing is often in the interests of most of the real people behind the shareholders. Often a majority of them would also agree. After all, despite the teachings of some religious leaders, most of us prefer to be richer rather than poorer, and often an indirect shareholder's only, or primary, relationship with a corporation is as shareholder. Not all shareholders live within breathing distance of the polluting smokestack, and some of the distant ones will not care about pollution that does not affect them directly. [FN142]

Indeed, even where there is a clear conflict between shareholder interests and other interests of the person who owns the shares, the shareholder interest often will be the more important: many of us are willing to make quite major sacrifices in other areas of our lives for an incremental bit of individual wealth or collective economic growth. So my complaint is not that business corporations are more likely to end up on the substantively " wrong" side of a dispute than other political participants.

Rather, the issue is that the corporation is barred from making the trade-off at all. Fiduciaries for a fictional shareholder are told that they must not think about the tradeoffs; they must profit maximize without regard for the costs in change, overwork, family stresses, (legal) pollution, and so on (except, of course, to the extent that those factors will have a result on profits). [FN143]

As a rule, this one-sidedness is a good thing--the singular focus of the corporate structure allows corporations to do a singularly good job of the job they do. One of the reasons why business corporations are generally better at producing innovative and inexpensive consumer goods than are government agencies, despite the extra expense of paying the shareholders, is that the business corporation can focus on production as its sole