What is a Business Corporation?

Business is conducted under a variety of legal forms, including classically the sole proprietorship and the general partnership, with the corporate form as a relative newcomer. The corporation should thus not be conflated with the business firm or even with the large managerial firm, because managerialism is commonly used by all large organizations, regardless of legal form. Conversely, the corporate form is enjoyed by many nonbusiness entities, including monasteries, universities, towns, and associations, and was granted to them long before it was granted to business entities (Davis Reference Davis1905). The primary rights of a corporation thus have nothing to do with business per se. They are three in number: (1) the right to own property, make contracts, and sue and be sued, as a unitary entity (a legal “person”); (2) the right to centralized management of this property; and (3) the right to establish and enforce rules within its jurisdiction beyond those of the laws of the land—such as the monastic Regula Benedicti, town ordinances, bylaws, and work rules.Footnote ³

The first right establishes how a corporation relates to outside parties—it relates to them as an independent contracting individual, with property and liability wholly separate from its members. The second and third rights establish a corporation's governing authority—its right, via a central board, to govern the property and people within its jurisdiction. As Max Weber himself emphasized, managers, no less than government officials, are understood to have a right to rule within their jurisdiction, and their subordinates to have a reciprocal duty to obey (Weber Reference Weber1978, 213–14, 948). Corporations thus contravene the liberal dichotomy of market freedom vs. government restraint. They make us more governed, not less.

In essence, all that the for-profit business corporation adds to these three primary corporate rights—and it is a revolutionary addition—is (4) the right to turn this governing authority and property to the pursuit of private profit. The political theory of the corporation thus starts from the understanding that corporations are governing entities first and foremost, to which business rights and capabilities were added.

For analytical purposes, it is useful to distinguish three dimensions of the corporation as a governing entity. First, there is the governance of the corporation—understood in this article as the procedures for collectively deciding how to deploy the corporation's labor and capital (Chhotray and Stoker Reference Chhotray and Stoker2009, 3, 145). This can be distinguished (if not always cleanly) from the task of management, the actual work of deploying, or governing, labor and capital. Finally, there is the legal basis of the corporation's right to govern the labor and capital within its jurisdiction.

This article does not explore methods of management, because there is little to distinguish the managerial techniques used in corporations from those used in large partnerships and proprietorships. A common toolkit is available to all and does not mark one as more public or private than another. What sets the corporation apart as a governing entity, and justifies placing it in a different legal category, is found in the other two dimensions—its governance (examined next) and the legal basis of its right to govern (examined in the penultimate section).

Corporation as Constitutional Republic; Constitutional Republic as Corporation

In its governance, as in many other areas, the corporation is distinct from other business entities, but strikingly analogous to constitutional republics. Indeed, its form of governance is that of a constitutional republic. This is more than coincidence; it is because corporations and constitutional republics share a common history. As Blackstone once observed, corporations are republics writ small (Blackstone [1753] Reference Blackstone1893, 456). Indeed, the further back one goes, the more direct the analogy. The Dutch East India Company, the world's first publicly traded limited liability corporation, was created in the image of the Dutch republic, with a governing board composed of a proportionate balance of representatives from each of the country's provinces, just like the Dutch States-General (Adams Reference Adams2005, 51). The English East India Company (EIC; est. 1600), which set the mold for future business corporations in being a republic of shareholders, was chartered shortly thereafter and soon reorganized along the lines of the House of Commons (Robins Reference Robins2006, 30). It operated under a constitution (in this case, a written constitution, or charter), which authorized a property-owning electorate (the shareholders) to elect a parliament (the board of directors), which elevated one of its own to the position of prime minister (the chairman of the board). Also striking, its electoral principle was not one-share-one-vote as today, but, as with Parliament, one-shareholder-one-vote, with a substantial property threshold for the franchise (in the early 1700s, the minimum was £500 worth of shares; Robins Reference Robins2006, 30).Footnote ⁴ The EIC's powers over its members were also analogous to those of Parliament, with directors enjoying the right, by majority vote, to pass bylaws for the corporation and to dismiss, fine, physically chastise, or imprison those who violated them (Elizabeth [1600] Reference Elizabeth and Shaw1887, 7). The EIC could also mint coin, administer justice in its settlements, and wage war (Robins Reference Robins2006, 28).

Blackstone would have done just as well, however, to say that republics are corporations writ large. The Massachusetts Bay Colony, for example, began as the Massachusetts Bay Company, the charter of which served as the colony's first written constitution. A similar story holds for Virginia and several other American colonies. And the line of development from the corporate charters of Massachusetts and Virginia to their colonial constitutions, to their state constitutions, and to the federal constitution is a direct one (Lutz Reference Lutz1998). Indeed, well into the nineteenth century, “charter” and “constitution” were taken as virtually synonymous. As defined by Francis Lieber in his Encyclopaedia Americana, “a corporation is a political or civil institution . . . conducted according to the laws of its constitution.” “All the American governments,” he added, are “corporations created by charters, viz. their constitutions” (Lieber Reference Lieber1830, 547).

In light of this promiscuous history, it is little surprise that, within its jurisdiction, the business corporation exercises powers analogous to those of government, if more limited, including the right to command, regulate, adjudicate, set rules of cooperation, allocate collective resources, educate, discipline, and punish. In fact, it was long common in debate over the proper distribution of power in one to draw on arguments from analogy with the other. For example, in the early republic, the republican practices of state and federal government were invoked as arguments for “republicanizing” the business corporation—for example, increasing the power of shareholders vis-à-vis management or establishing rotation in office for management (Maier Reference Maier1993, 79). During the Progressive Era, this same analogy would be used to argue for “democratizing” the corporation—that is, instituting worker self-management, or “industrial democracy” (Croly Reference Croly1914, 384–85). Conversely, principles of “strict construction” and the doctrine of “implied powers”—mainstays of American constitutional argument—were drawn directly from judicial interpretive practices as applied to corporate charters (Davis Reference Davis1905, 207). Even the Supreme Court's seemingly unprecedented power of judicial review was likely midwifed by the judicial practice of voiding, as ultra vires (“beyond powers”), corporate acts exceeding the powers granted a corporation by its charter (Maier Reference Maier1993, 79; compare Bilder Reference Bilder2006).

A Problematic Republic

Extensive though the analogy is between citizen and shareholder republics, there are important points of difference as well. Most obviously, shareholder republics are run for profit—the commonweal reduced to dollars and cents. This feature becomes especially noteworthy because of a second difference. Unlike citizens, shareholders—the voting members of a corporation—do not normally fall within its jurisdiction or suffer its government, whereas corporate employees—its nonvoting metics—do. There is therefore no reason to expect a shareholder republic to be “for” the governed, because its “ruling class” is neither “of” the governed nor accountable to the governed. I return to this point in what follows when considering the rights of corporate employees.

Finally, although the end of a constitutional republic is the good of its citizen-members, the end of the corporation, at least originally, was not only the good of its shareholder-members but also the good of the chartering government and its general citizenry. As Henry Carter Adams summarized in his President's Address to the 1896 meeting of the American Economics Association, “A corporation . . . may be defined in the light of history as a body created by law for the purpose of attaining public ends through an appeal to private interests” (Adams Reference Adams1886, 16). How to frame corporate charters, corporate governance, and general laws so as to keep corporate powers oriented toward the public interest or at least consistent with the public interest—that is, how to organize and regulate these shareholder republics so that they serve not only themselves but others—is a problem that no ordinary republican theory has had to address and one that has vexed kings and legislatures throughout the corporation's history. Much of the political theory of the corporation revolves around this question, to which I return when considering the proper end or purpose of the corporation.

How Business Corporations Preserve Contractual Individuality despite Investor Plurality

Contractual individuality is important even for nonbusiness corporations. It separates the property and liabilities of the corporation from its members, giving it a separate and stable legal existence—of great importance to Franciscan monasteries, for instance, whose friars could not own property. Additionally, it eliminates many of the coordinating costs and other transaction costs borne by general partnerships and other associations when dealing with external parties.

At least in a private property economy, a corporation's contractual individuality is a grant of government. A group of friends, for example, having created an informal lending library by pooling their respective book collections, cannot unilaterally decide, no matter the content of the contracts they sign with one another, that the courts will treat the books of the library as library property, allow contracts to be made in the library's name with its books as collateral, or limit liability to the library and its property for harms it might cause (from a tumbling stack of books, say). This alone may be sufficient proof of the dependence of the corporation on government.

However, the business corporation, which is presumed the most private type of corporation, has in actuality a dependence on government much deeper than the nonprofit corporation. What distinguishes the modern business corporation from its nonbusiness cousins is that the business corporation has annexed to it a joint-stock mechanism. This allows it to pool capital from multiple parties, turning the corporation into an investment vehicle with a license to make and disburse a profit. The question is, How does the corporation preserve its contractual individuality even as it acquires investor plurality? In other words, how does it convert investor capital into a single and separate corporate fund? It turns out that this is impossible without violating the standing rules of property, contract, and liability. Government allows the corporation to override them as a legal privilege.

Partnerships pool assets in a simple and straightforward fashion. Partners put in money, which is used to purchase assets for carrying on the business of the partnership. These assets remain bound to the partners, who collectively own them. For this very reason, the partnership falls short of being a separate contracting individual. The partnership does not own its own property; the partners own it. An important consequence is that, if a partner decides to leave the business, she takes her portion of the partnership's equity with her, which normally dissolves the partnership, and at the very least forces a buyout by the other partners, who may need to liquidate firm assets to cover the expense.

In a corporation, in contrast, the normal rules of property are broken. Investments are permanent; the investor cannot directly pull out his contribution. An investor may recoup the monetary value of his investment if he can find another investor to take his place—that is, to buy his “share.” However, the assets that the corporation purchases with his initial investment are locked in, becoming corporate property. They form a separate fund (Blair Reference Blair2003, 392).

The business benefits of this feature are considerable. First, it lowers the corporation's capital costs, because lenders need not fear expropriation by withdrawing investors (Blair Reference Blair2003, 427). Second, it increases firm productivity. It allows the corporation to specialize its assets to the production process, rather than keep them in more liquid form out of fear that investor withdrawal will force a sell-off. This in turn allows the corporation to specialize its workers to its specialized assets (Blair and Stout Reference Blair and Stout1999, 271–87). In other words, it advances the specialization of both capital and labor, the classic means to increased productivity (Smith [1776] Reference Smith1976, chap. 1). Being a contracting individual with a separate fund is thus extra-beneficial when a corporation is a business. It lowers the cost of capital by making assets secure and increases productivity by allowing assets to be specialized.Footnote ⁶ What I call asset lock-in is one of the rules of corporate financing that makes corporate funds separate.

Asset lock-in cannot by itself make the separation complete, however, as can be seen by returning to the example of a general partnership. Imagine that a partner incurs significant credit card debts or debts from some other business. If the partner's personal assets are insufficient to pay off the creditors in question, the creditors may go after partnership assets. Again the firm is put under threat of liquidation. What Henry Hansmann and Reiner Kraakman have dubbed strong entity shielding is a legal privilege that protects corporate assets from this threat (Hansmann, Kraakman and Squire Reference Hansmann, Kraakman and Squire2006, 1336). The personal creditor of a shareholder may take the shareholder's shares, but corporate assets are shielded. It is the conjunction of asset lock-in and entity shielding that provides the protection from liquidation that allows corporations to specialize their assets and that boosts their credit.

One more legal privilege is required, however, to provide for full contractual individuality. This is the privilege of limited liability for shareholders, which shields shareholders from the debts of the corporation. Without it, a corporation's credit would not truly rest on its own assets and future prospects, but, like a partnership, would rest in part on the assets and credit of its investors.

Beyond this, limited liability provides economic benefits of its own to the corporation. Most obviously, it makes the corporation much more attractive than a partnership for small and passive investors. Even a small investment in a partnership entangles the partner in potentially large partnership liabilities. A stockholder, however, is liable only up to the value of the stock held. By attracting small investors, as well as diversifying large investors, the corporation can raise great sums at low cost (Easterbrook and Fischel Reference Easterbrook and Fischel1985, 94–101).

Less obviously, limited liability combines with entity shielding to make shares tradable. Tradability is critical because, without it, the only point at which corporate investors could recoup their investments would be at the dissolution of the firm, because investments are locked in. Few would invest in a long-term enterprise under such a regime, unless dividends were particularly high. Tradability returns liquidity to investors, allowing them to draw out the value of their shares provided they can find buyers for them.

Why are limited liability and entity shielding necessary underpinnings of tradability? Without them, the real value of shares would fluctuate not only with the prospects of the firm but also with the financial condition of all stockholders, because the personal assets of the stockholders would be the ultimate foundation of the firm's credit (as they are in a general partnership). This would make the pricing of shares difficult, to say the least. What is more, it would lead existing shareholders to oppose tradability so as to protect their investments from an influx of financially weak investors, who would have incentive to purchase shares with the object of using firm assets as personal collateral (Hansmann, Kraakman, and Squire Reference Hansmann, Kraakman and Squire2006, 1350).

In sum, asset lock-in, entity-shielding, and limited liability completely disentangle corporate assets and liabilities from investor assets and liabilities, preserving the corporation as a separate contracting individual. Because of these features, the assets of the corporation alone bond the contracts of the corporation alone. This brings great business benefits. With separate property, the corporation has strong liquidation protection. This increases its productivity (by enabling asset and labor specialization) and lowers its capital costs (by lowering the risk and monitoring costs of its creditors and investors). Limited liability then lowers the cost of capital yet further by lowering investor risk (both directly and by helping make shares tradable).

This gives us the fundamentals of an economic theory of the corporation. Diagrammed, it becomes clear that these privileges are not tangential, but are the legal foundation of the corporation's productive power (see Figure 1).

FIGURE 1. An Economic Theory of the Corporation

There are circumstances where the sole proprietorship (with its legal simplicity and frequent tax advantages) and the general partnership (with its greater flexibility in assigning control rights) are preferred business forms. But whenever a large quantity of specialized physical capital is called for, the corporate form becomes the legal form of choice. Indeed, this is why the industrial revolution gave the corporate form a new lease on life.

The Dependence of the Corporation's Contractual Individuality on the State

Having identified the key legal attributes of the business corporation, what are their implications for the governmental provenance of the corporation? I have denominated them “privileges,” implying that they must be provided by government, and it is indeed the case that they are part of a standard package of corporate attributes that every state now provides by statute. Some, however, could in principle be partially established through private contract. For example, asset lock-in could be established, at least for a fixed term, by having each investor contractually agree not to withdraw his or her assets from the firm for the duration of that term. Indeed, such an arrangement has not been unusual among partnerships (Hansmann, Kraakman, and Squire Reference Hansmann, Kraakman and Squire2006, 1342). Likewise, limited liability could be established, at least in part, by adding to every corporate contract the proviso that, in case of firm default, the creditor in question will not levy on investor assets. This arrangement, too, can be found historically (Hansmann and Kraakman Reference Hansmann and Kraakman2000, 429–30).

However, these arrangements fall far short of establishing a corporation contractually. There is, for instance, no contractual way to limit the liability of shareholders for tort claims against the corporation, because the harmed third parties will often be outside those with whom the corporation has contracts. Further, courts have been unwilling to uphold contracts among partners to lock in assets beyond a limited term (Hansmann and Kraakman Reference Hansmann and Kraakman2000, 412). The biggest shortcoming of the contractual view, however, is that one of the two indispensable legal attributes of the corporation—entity shielding—cannot be contracted for at all. Doing so would require every shareholder to contract with every one of his creditors (banks, utility companies, hired help) against their laying claim to firm assets in the case of his personal insolvency. Not only are these transaction costs prohibitive but there is also a problem of moral hazard, because it is in the interest of each shareholder to renege and allow firm assets to back up his personal credit. Because of this, every shareholder would have to monitor every other shareholder to ensure compliance, even as shareholders kept changing (Hansmann, Kraakman and Squire Reference Hansmann, Kraakman and Squire2006, 1340). Given privacy laws, it is unlikely this monitoring would even be legal, let alone feasible.

The inescapable fact is that corporations rely on government to override the normal market rules of property and liability and reordain which assets bond which creditors. Indeed, asset lock-in, entity shielding, and limited liability together create the very distinction between corporate assets and personal assets. These privileges are thus not mere benefits tacked on to a preexisting, contractually constituted corporate form. Rather, they are what make the business corporation a corporation. Stripped of asset lock-in and entity shielding, a business corporation loses its propertied independence and becomes a general partnership. Government intervention in the market is what begets the corporate “person.”

This argument, it is worth noting, is not a mere recapitulation of “legal realist” arguments about the pervasiveness of state action in the economy. In defense of factory legislation and labor laws, legal realists of the early twentieth century argued that the common law rules of property, contract, and liability under which business is conducted are not the natural, neutral principles of justice and reason, but are in the final analysis determined by government authority, and thus should be subject to legislative revision (Horwitz Reference Horwitz1992, 193). Kent Greenfield has argued for viewing corporate law in the same light, for similarly reformist reasons (Greenfield Reference Greenfield2006, chap. 2). That government sets the rules of the market does not mean that all business entities depend on government for their bare existence, however. Proprietorships and partnerships, for example, are legally identified with their owners; and these proprietors and partners do not depend on government for their existence. But corporate entities do—and this will be true whatever the reigning rules of property and liability, because these rules will have to be broken to separate the property and liabilities of the corporation from its members.

Legal realism gives us no reason to place corporations in a category separate from proprietorships and partnerships (which we may fairly denominate “private,” while remaining mindful that their flourishing depends on a variety of state services). In contrast, the argument developed so far provides a double reason to place corporations in a separate category. First, the ontology of corporations is different, being grounded in government fiat rather than natural persons. That is to say, corporations are governmental in provenance. Second, the rules by which they operate are different. Although corporations do buy and sell under the same rules as proprietorships and partnerships, they are constituted and financed under different rules created especially for them. And it is their mode of financing that gives corporations their main advantage, securing for them large amounts of specialized capital.

Corporate Property as Socialized Property

No one ever thinks to ask who owns the lecterns of a college, or the benches in a city park, or the desks in a philanthropic organization. It is understood that they are owned by the nonprofit corporation in question—the college, town, or NGO. Yet, as soon as a firm incorporates and sells shares to generate extra financing, however little, it is asserted—in law schools, business schools, and the popular press—that its shareholders “own” the company. This assertion implies that corporate property is reducible to the private property of individuals, which in turn becomes a powerful argument for assimilating corporations to liberal market economics and the category of the private. But the assertion is plainly false.

The relevant contrast is with the partners in a partnership. A partner really does own a percentage of the partnership, and if she leaves, she removes her portion from the assets of the firm. In contrast, corporate assets are locked in. What a shareholder owns instead is a piece of stock, representing her financial interest in the corporation. If the corporation does well, she may receive dividends, at the discretion of management. She may also sell her share to others, it is hoped at a profit. Owning a share, however, is entirely different from owning corporate assets. If I own something, I can (a) use it, (b) exclude others from it, (c) lend it to others on my terms, (d) borrow on it, (e) alienate it, and (f) profit from it in use or sale. Shareholders have none of these rights over corporate assets, either individually or jointly. Nor are they legally liable for them.

It is true that shareholders collectively have the formal right of electing the board of directors, and in theory they could use this right to elect board members precommitted to a specific policy for the use of corporate property. Still, even on the odd occasion when this right is not completely undermined by managerial control of board member nominations and proxy voting, it is a political right, or voting right, intended to allow shareholders to protect their financial interest, and not a property right. It allows shareholders to influence the policy of management, but it implies no more legal title on the part of shareholders to corporate assets than voting rights imply a legal title of citizens to a country's fighter jets, the property of the state.

Also incorrect is the oft-cited argument that shareholders should be regarded as owners because they are the “residual claimants” of corporate revenue (Easterbrook and Fischel Reference Easterbrook and Fischel1989, 1425). In actuality, all profits go to the corporation, the true residual claimant, and are then allocated at the discretion of management, which rarely directs more than a fraction to dividends (Stout Reference Stout2012, 40–41). Even in bankruptcy, shareholders are not treated as owners. If they were, all assets would come into their legal possession at bankruptcy, and it would then fall to them to pay off corporate creditors. Instead, the opposite happens, at least in the United States. Corporate creditors have first claim, and only what remains goes to shareholders (Wood v. Dummer 1824). So even at this rare moment, shareholders are not in the position of owners, but heirs.

So whose assets are they? As we have seen, asset lock-in, entity shielding, and limited liability partition the assets and liabilities of the corporation from those of its investors (and creditors and managers). This partitioning provides the corporation with a stability and potential scale that cannot be matched by the aggregated personal property of partnerships, which are subject to disruption by partner death, withdrawal, or bankruptcy. It also means, however, that no natural person or group of persons owns the assets of the corporation. The corporation owns corporate assets—uses them to bond its contracts, for example, and is liable for the damages they cause—just as the state owns state assets and the church owns church assets. It is corporate property. The rights that natural persons have over corporate property are control rights, which fall, variously, to office holders in the corporation, just as control rights over government property fall to office holders within government. Such corporate property is “private” only in the sense that it is not public. What it really needs is a category of its own, distinct from both public and private (proprietary) property.

Marx was thus not wrong to see bourgeois (individually owned) property as a fetter on the productive powers of capital that would be burst asunder, to be replaced with socialized property (Marx and Engels [1848] Reference Marx and Engels1933, 13). The Marxist mistake was to assume that the socialization of property would occur at the level of the state. Instead it has occurred at the level of the corporation. The underlying reason is the superior scale and stability of socialized property to the merely aggregated property of restless individuals in partnership.

Ours is neither a public property economy nor a private (bourgeois) property economy, but a government-fostered corporate property economy, with the means of production socialized at the level of the firm. No less than a socialist economy, but through devices more opaque, our economy is dominated by property unowned by natural individuals.

Corporate Irresponsibility and a Liberal Cure Worse than the Disease

That corporate property is socialized property has important implications for how the corporate economy works. A central rationale for private property, going back even to Aristotle, is that the owners of private property bear the consequences of their use of that property and thus have every incentive to mind it well (Aristotle 1988, 1263a). A private property economy is, externalities aside, a system of individual economic responsibility.Footnote ⁷ In contrast, socialized property severs this responsibility: Those who control it do not own it and thus do not bear the direct economic or legal consequences of their control. This is familiar as a critique of socialism (Von Mises Reference Von Mises1944, 62), but the same holds for corporate capitalism. The corporate form separates ownership and control; thus, corporate managers, like socialist managers, do not own the assets they control and do not bear the direct consequences of their control. Without supplemental devices to align incentives, a corporate economy is, literally, an institutionalization of individual economic irresponsibility. Corporate property thus creates the conditions both for greater productivity and greater irresponsibility: The former may only be a boon if the latter can be curtailed. Market liberals have actually been quite sensitive of this responsibility gap. Their strategies for closing it, however, have been counterproductive.

The Foundations of Managerial Authority

Management in a business firm (whether proprietorship, partnership, or corporation) has a dual authority: It has authority over the firm's employees (Ciepley Reference Ciepley2004), and it has authority over the firm's assets (Zingales and Rajan Reference Zingales and Rajan2001).

Management's authority over its employees rests on two foundations. Its proximate foundation is the labor contract, which includes, as a nonnegotiable “implied term,” a duty of obedience on the part of the employee to the employer (Muthuchidambaram Reference Muthuchidambaram1979, 105–6). Unless a boss's directive is illegal, unreasonable, or unethical, one has an obligation to carry it out, however unwise or contrary to one's own interests it may be. Together with accompanying duties of loyalty, care, and confidentiality, the duty of obedience shows that today's labor contract, like the feudal contract between master and servant from which it descends, is a contract that is more than a contract.Footnote ¹⁰ It ushers one into a status category, with non-negotiable asymmetrical obligations of deference that last for the duration of the relationship, although limited to the organizational setting. In an organizational age, status relations recrudesce.

As this shows, even the authority of management in a proprietorship or partnership is governmental in the weak, legal realist sense, being exercised under government-sanctioned, non-neutral background rules of labor control. Nonetheless, the authority of management in the corporation is “of government” in a much more direct sense, as seen momentarily.

Why in a liberal age would anyone subject themselves to the status relations of the labor contract? Why take a job? The answer, of course, is because it may be one's best or only means of access to scarce productive assets, the “means of production.” This brings us to the other, mediate foundation of management's authority over the firm's employees—management's authority over firm assets.

The prototypical ground of authority over assets is ownership. An owner has the right to exclude others from the property or, alternatively, to set the conditions of access to it; for example, conditioning it on the signing of a labor contract (Cohen Reference Cohen1927, 12). In the case of proprietorships and partnerships, ownership is indeed the ground of management's asset control. But what is the ground of management's asset control in a corporation? And for that matter, what is the ground of management's authority over employees, because the labor contract is not between the worker and manager, but between the worker and the corporation?

The prevailing view suggests there is nothing special about the corporation in this regard. Shareholders are taken to own the corporation. As a corollary, it is supposed that the right of management to govern the corporation—to control its assets and manage its employees—derives from these owner-shareholders, who authorize the management as their agent. Milton Friedman, in his famous polemical essay, “The Social Responsibility of Business is to Increase its Profits,” provides a classic statement of this view: “In a free-enterprise, private-property system a corporate executive is an employee of the owners of the business. He has direct responsibility to his employers. That responsibility is to conduct the business in accordance with their desires, which generally will be to make as much money as possible” (Friedman Reference Friedman1970, 17). The corporation, on this neoliberal rendering, is but a partnership. It is the private property of its shareholders, who authorize management to run it on their behalf.

Within political science, Robert Dahl is a well-known critic of the shareholder-oriented corporation, especially on account of its antidemocratic implications. As he emphasizes, “By making ownership the only, or at least primary, source of legitimate control over corporate decisions, the new [corporate] order not only excluded democratic controls in the internal government of the enterprise but placed powerful ideological barriers against the imposition of external controls by a government which, for all its deficiencies, was much more democratic than were the governments of business firms” (Dahl Reference Dahl1977, 8). Clearly implied here is the possibility that there might be sources of legitimate corporate control other than ownership. Dahl then takes another step forward, questioning whether it was justifiable to give shareholders the control rights of owners in the first place. Doing so was, in his words, an “extraordinary ideological sleight of hand,” based on extending the “Lockean” rights of private property from the agrarian farm, where they belong, to corporate stock (Dahl Reference Dahl1977, 8).

Unfortunately, Dahl does not fully develop these intuitions, and he even obscures them. Accepting the conventional view that shareholders “own” the corporation (Dahl Reference Dahl1985, 79), yet persuaded that the modern business firm is an association that ought to be organized along democratic lines, Dahl looks to alternative ownership arrangements that he believes are more compatible with worker self-government—employee ownership through single shares, collective employee ownership, state ownership, and ownership by “society” (Dahl Reference Dahl1985, 140–48). As Dahl puts it, “What is a desirable form of ownership ought to be viewed, at least in part, as subordinate to and dependent on a judgment as to what is a desirable form of control” (Dahl Reference Dahl1977, 16). Addressing the problem this way, however, only reinforces the identification of control rights with ownership rights, even if reversing their usual priority.

The analysis of corporate property developed here vindicates Dahl's initial intuitions and allows them to be pushed further. Because no natural persons own corporate assets, legitimate control of these assets not only might, but must, derive from something other than ownership of them. Furthermore, the reason that it is a sleight of hand to attribute a Lockean right of asset control to owners of stock is that stock ownership is not the same thing as asset ownership. Both points follow from the one ownership arrangement that Dahl does not consider, which is the arrangement that actually obtains within the corporation, in which the firm owns itself. Contrary to Dahl's impression, this arrangement makes the corporation a perfectly suitable vehicle for worker self-government, because it completely removes the issue of ownership from the question of who controls. A corporate charter could make the employees the electors of the board—or place employee representatives on the board, as in Germany (Beal Reference Beal1955)—without violating any of the property rights of shareholders. Indeed, nonvoting stock has long been commonplace in the United States (Mitchell Reference Mitchell2007, 4), with no protest about loss of property rights.

Board composition can be customized in this way because the true basis of management's authority—its authority over both corporate employees and corporate property—is not shareholder ownership, but the corporate charter, along with any enabling corporation law that specifies the rights that charters convey. In other words, it is government that authorizes management's rule within the corporation's jurisdiction. This is not just a theoretical extrapolation from the fact of corporate self-ownership. It is fully supported by the specifics of corporate law. First, the mechanics of incorporation support it. A charter formally ordains a corporation and, as part of this, expressly ordains and authorizes a board of directors, usually with members listed by name, to manage corporate assets, hire employees, define their duties, and generally conduct the corporation's affairs (Corporate Laws Committee 2005, §8.01, §3.02). It also expressly authorizes the board to issue stock up to a specified amount (Corporate Laws Committee 2005, §6.01). All this happens before there are any shares or shareholders. Given this sequence of events, it is clear that the government charter, and not the shareholder, creates and authorizes the board. With the boards of nonprofit corporations, this is even more clear, because they never solicit shareholders. In the case of business corporations, purchasers of common stock do receive a nominal right to elect succeeding directors. This right does not derive from ownership, however; it is authorized by the government charter (Corporate Laws Committee 2005, §7.21). Nor does this nominal right of shareholders to choose the occupant of the office of director alter the fact that the office itself and its authority are established by government via the charter, and not by the shareholders. It is telling that this is formally no different from the arrangement that obtains in that other important corporation, the American town, the members of which choose its officers, but which has its offices and their authority by state charter (Frug Reference Frug1980, 1062–63). If shareholders were truly owner-principals and authorized management, they would have the freedom of establishing whatever offices they saw fit, as do partners in a partnership. Instead, the corporate form includes certain non-negotiable constitutional fundamentals, the centerpiece of which is an elected board with specific powers and duties (Corporate Laws Committee 2005, §8.01, §6.01).

That government, rather than the shareholder, authorizes management's rule is also borne out by consideration of management's legal obligations. If shareholders were authorizing principals and directors their agents, directors would have an agent's duties toward them and accountability to them. Instead, the duties of directors—of loyalty, care, and confidentiality—are duties to the corporate entity, and not to shareholders (Corporate Laws Committee 2005, §8.30). Although holders of common stock nominally elect directors, they cannot fire them nor directly sue them for breach of duty,Footnote ¹¹ any more than town residents can fire or sue their elected mayor for breach of duty. Again, the similarity is not coincidental.

In sum, the shareholders are not management's authorizing principal. In the eyes of corporate law, the principal is the corporation itself, and management is its agent. Management has a duty to act on behalf of the corporation (and not simply on behalf of its shareholders), and, in discharging this duty, is authorized to rule over corporate personnel and property. Both this duty and this authorization come from the state, via a corporate charter. The state is the foundation of managerial authority.

Corporations as Franchises of Government

The surprising picture that emerges from analysis of the foundations of managerial authority is of a formal constitutional structure of government in the United States composed of, not two, but three tiers—a triple layer of constitutional republics: (1) the federal government, authorized by the formally sovereign people; (2) the state and territorial governments, authorized by the federal government—if not one of the original 13, authorized by the Crown; (Biber Reference Biber2004); and (3) the towns and other corporations, including business corporations, authorized by the state governments. At each level, authorization takes the form of a charter, or constitution, approved on authority of the legislature at the level above (or, at the top level, by the people). Also, each level of government is republican in form, enjoying autarky (the ability to select its own leaders) and broad, but not limitless, autonomy (the ability to frame laws within its jurisdiction), the limits to its autonomy being drawn at the level above, in a system of ascending legal supremacy.Footnote ¹²

This way of viewing matters is not novel, merely long submerged. The history of corporations is a history of “sovereign” institutions (such as church or state) authorizing inferior institutions (corporations) to do things that the sovereign institution wants done but is unable or unwilling to do itself (Wood Reference Wood, Abe, Sato and Otsuru1999, 9). Corporations are institutions of delegated government. Their authorization is accompanied by a grant, sufficient to the work to be undertaken, of powers and privileges that governments normally deny natural persons and reserve to themselves. At the extreme, this has meant granting corporations nearly plenipotentiary powers, as with the British East India Company. More modestly, it might mean granting a canal or railroad company the power of eminent domain, as was often done in the early American republic, when corporations were chartered to build the nation's infrastructure (Maier Reference Maier1993, 68). Or it might mean granting the privileges of contractual individuality and centralized management alone, as with banks and insurance companies. A history of state power without a history of corporations is thus radically incomplete. Indeed, the practice of chartering corporations can be thought of historically as state-building at one remove—the building of a “franchise state.”

Whether corporations should still be thought of as constituting a franchise state is difficult to say. Even today, all business corporations depend on state-granted privileges and authorizations and thus remain legally closer to the Tennessee Valley Authority than to proprietorships and partnerships. The difficulty is not that corporations have become genuinely “private,” but that their activities have become almost wholly commercial, which liberals do not associate with the state. Put another way, they are difficult to interpret because they have participated in—indeed, have been a principal conduit for—what Hannah Arendt calls the “rise of the social,” in which the animal needs of the private household, or oikos, come to preoccupy the public sphere (Arendt Reference Arendt1998, 38–49). The chartering of the great trading companies in the early modern period stands at the trailhead of the social—a tapping of public power for traditionally private, money-making pursuits and a liquidation of private property for the sake of social wealth. The stepwise unleashing of corporations into all branches of commercial activity is a marker of the progress of the social. Corporations, which begin as indirect arms of the state, are chartered for the sake of increasingly commercial “public benefits” and end as institutions of delegated government for the pursuit of private profit and private consumption alone. This coopting of public power by private need is only rendered more complete by the fact that public power's role in constituting the corporation has been concealed within liberalism and thus placed beyond challenge. Thus, it remains legally correct to describe corporations as a franchise state, but it is only phenomenologically plausible if one bears in mind liberalism's subordination of the state to the needs of society, the political-economic culmination of the rise of the social.

Ontological versus Consequentialist Transgressions of the Public/Private Divide

It has been pointed out many times before that corporations transgress the public/private divide, perhaps most famously by Theodore Lowi in The End of Liberalism, where he highlights the institutionalized inclusion of business actors (often corporate) in the formulation of public policy (Lowi Reference Lowi1969, chaps. 3–5). The point I am making, however, is not just that the activities of corporations transgress this divide, but that their very being transgresses this divide.

I believe this changes the normative landscape. In a conventional critique, corporations would be categorized as private, and one would raise concerns of the sort raised by Lowi or by Jürgen Habermas in Theory of Communicative Action and Michael Walzer in Spheres of Justice— that corporate power in the private market is spilling over into other spheres of life, colonizing them and perverting their principles and goods (Habermas Reference Habermas1984, chap. 6; Walzer Reference Walzer1983, chap. 4). This is surely true and of concern. The deeper point, however, is that corporations themselves are not creatures of the private market, but governmental colonizers of it. Therefore, a strategy of “returning” them to the private market, with license to operate by the same rules as regular market actors, is both theoretically mistaken and pragmatically inadequate, as argued earlier. A distinct analytical category and corresponding set of legal rules must be developed for them.

Closer to the position developed in this article is that of Adolfe Berle, who also emphasized corporate transgression of the public/private divide and rightly concluded from this the need for something like a political theory of the corporation. In Berle's view, it is not just the corporation's political heft but also its provision of essential goods and services that give it a “political aspect” and turn it into a “quasi-governing agency” (Berle Reference Berle1954, 104–5). In other words, corporations become quasi-public when they have significant public consequences.

Unfortunately, this argument suffers from being overly broad, in that it would turn any powerful entity or individual into a quasi-public body by virtue of its public impact; and it suffers from being vague, because almost any act by any actor may have eventual public consequences, forcing one into ad hoc line drawing and endless, destabilizing reclassifications of particular entities as their public impact varies over time.

In something of an afterthought, Berle adds that corporations can become quasi-public by virtue of receiving public benefits. As an example, Berle mentions corporate use of publicly funded scientific research (111), though benefits can also take the form of tax breaks, or subsidies, or even use of public infrastructure. This argument is also overbroad, however, because it threatens to turn even a private citizen who receives, say, Social Security payments into a quasi-public entity; and it is vague, again involving endless controversial line drawing, because, as legal realists forcefully argued, in modern society, almost any activity can be seen as in some way underwritten by state authority and thus in receipt of a public benefit (Sunstein Reference Sunstein2002). The argument thus threatens to erase the very category of the private, which is something the legal realists rightly avoided. Classifying corporations by virtue of their public impact or receipt of public aid, although intuitively appealing, is in practice a conceptual, legal, and policy morass.

Rather than resting the case for the reclassification of corporations on their sending consequences across the public/private divide or their receiving benefits across this divide, I stress that they are constituted across this divide and straddle it throughout their existence, with private parties providing financing, organization, and initiative, and government providing contractual individuality, a socialized mode of property, and governing rights. This conception clearly identifies all corporations, regardless of size and impact, as governmental, even if not fully public. It makes them proper objects of political theory, and it justifies placing them in a distinct category, subject to rules and norms beyond those applied to properly private business entities. The remainder of this section begins fleshing out the category of the corporate with some of the distinct rules and norms that I believe follow from arguments in previous sections.

Whom Should Corporations Serve? The Shareholder/Stakeholder Debate

That the corporation is co-constituted by government has implications for the corporation's proper ends. It has been strongly asserted in recent decades that the purpose of the corporation is to “maximize shareholder value,” or share price. This doctrine of “shareholder primacy” has two main rationales: that shareholders are the legal owners of the corporation, and that treating them as such is economically efficient. Neither rationale is tenable, however. Shareholders are not the owners; and running a corporation as if they were, by maximizing short-term share price, is not economically efficient—at least not in allocating resources to produce sustained growth (Stout Reference Stout2012, chap. 4).

The alternative to shareholder primacy is corporate social responsibility—a responsibility not just to shareholders but also to other stakeholders and society. Justifications for it vary, from long-term self-interest to the ethical principle that one should act in consideration of the consequences of one's actions for all, avoiding harm and perhaps even providing help (Garriga and Mele Reference Garriga and Mele2004). However, all the standard justifications, such as these, apply to all business actors and not to corporations specifically.

My argument revives an older view—that corporations, as privileged entities, bear heightened responsibilities toward the public. Intervention in the default rules of property and contract is justified only on grounds of public benefit. At least that is the usual understanding—the legal basis for the exercise of state “police powers,” for example—and for more than 250 years, it was the understanding of those who chartered business corporations. Corporations were chartered only if they promised a clear public benefit, and their charters regulated their activities, often heavily, to ensure this promise was fulfilled (Maier Reference Maier1993, 75–78). Corporations were to be instances of “private vices” creating “public benefits,” but with benefits to be secured not through the invisible hand of the market, but through the wise hand of the author of the charter.

So long as corporations were well regulated by charter and restricted to activities with a clear public benefit, it was considered justifiable to grant them legal privileges and to give their shareholders the profits. Profit was their incentive to move the public good forward. Once corporations became general business forms, however, with no limits on their activity but those of general legality, one could well ask, What is the public benefit to justify the special privilege? Why do they get to enjoy rights without corresponding duties? Corporations may offer heightened productivity. This is a public benefit, however, only if the fruits of increased productivity are widely spread. Without charter restrictions assuring public benefits, the practice of catering solely to the interests of shareholders (or managers) loses legitimacy.

How exactly corporations are to be brought to attend to broader stakeholder interests has always been a difficult question.Footnote ¹³ Nevertheless, corporations’ debt to public authority is a strong argument that they ought to, beyond what is expected of genuinely private actors.

The Rights of Corporate Workers

The authorization of corporate management by the state also has implications for worker rights. Without safeguards, the potential for worker exploitation by corporations is quite real. Although business corporations have the external form of constitutional republics, their internal governance is, generally speaking, neither liberal nor democratic. Internally, corporations are for-profit governments whose rulers are, with few exceptions, not accountable to the governed.

One ameliorative arrangement is to make management more accountable to workers, and there is no rights-based reason this cannot be done. As argued before, corporations could be organized as worker republics with no harm to shareholder property rights. That said, I know of no rights-based reason requiring business corporations to be organized as worker republics as opposed to shareholder republics.Footnote ¹⁴ The question comes down to one of efficacy, policy, and preference.Footnote ¹⁵

On the other hand, I do believe that the state's authorization of corporate management argues for more extensive civil liberties in the corporate workplace, whether organized as a republic or an autocracy. At mid-twentieth century, the Supreme Court began requiring state and local government to observe the core provisions of the Bill of Rights, which originally had only applied to the federal government (Ciepley Reference Ciepley2006, 231–51). Because the governing authority of the business corporation is derived from the state, parallel to the way in which municipal corporations derive their governing authority from the state, it would seem that corporate management ought to be required to observe the Bill of Rights as well. Yet the Court exempts corporations, as “private” business concerns. Workers do enjoy the protection of labor laws; the Bill of Rights, however, runs out at the company gate. Thus is created a paradoxical situation in which corporations are extended ever more rights from the Bill of Rights (see the following), whereas their employees, though proper citizens, are denied them.

There are legitimate reasons why civil liberties might be selectively narrowed in the workplace. For example, trade secrets probably justify the curtailing of speech rights for corporate employees just as state secrets do for government employees. It seems doubtful, however, that an employee's duty of loyalty can justify an across-the-board gutting of First Amendment rights in the corporation. If it is accepted that corporations, as publicly privileged entities, should be held to a higher standard of public benefit, then speech protection should be elevated where matters of public interest are at stake. For example, the currently piecemeal protection of corporate whistle-blowers might be expanded into a general protection against adverse action for publicizing corporate actions harmful to public or consumer interests (Blumberg Reference Blumberg1971). This would both expand worker rights and help trim corporations to the public good.

The corporation is a major anomaly in America's civil liberties revolution—an anomaly that political theorists are well placed to address. Political theorists have thought long and hard about institutional bulwarks against state tyranny and about issues of efficiency and fairness in government. At least some of their conclusions should apply to corporate government.Footnote ¹⁶

Reconsidering the Constitutional Rights of Corporations

The argument of this article is that corporations are neither public nor private, but belong in a separate category, with different rights and responsibilities. However, American courts have not agreed. Over the long nineteenth century, and with renewed energy since the 1960s, they have escorted corporations down liberal lanes and into the category of the private, granting them license to act with minimal regard for, or accountability to, the public, and outfitting them with a layer of protection from public scrutiny and public regulation (Horwitz Reference Horwitz1992, 74). A political theory of the corporation should be able to controvert every step down this path, from Dartmouth College v. Woodward (1819) (institutionalizing the distinction between public and private corporations and also redefining corporate charters as contracts immune to legislative revision) to Citizens United (2010). As promissory that it can, the remainder of this section challenges one of the most consequential steps in this assimilation to liberalism—the Court's practice of extending to corporations the constitutional rights of private citizens.