Who Speaks for the Corporation? A Hobbesian Theory of Managerial Authority and Shareholder Responsibility

Abstract

From where does management acquire its authority to act in the name of the corporation? The orthodoxy that shareholders alone authorise management is frequently criticised for treating the corporation as the property of shareholders, rather than as a distinct legal person in its own right (Ciepley, 2013; Deakin, 2012; Robé, 2011; Stout, 2012). However, Hobbes’s theory of incorporation in Leviathan shows this influential critique of shareholder primacy to rest on a non sequitur. It does not follow from the (correct) observation that the corporation is a legal person to the conclusion that its interests are distinct from those of shareholders. Just as individuals become citizens of a state when they authorise a sovereign, shareholders are incorporated when they authorise a representative assembly to act in their interests. Shareholders thereby form a single corporate person and are ultimately responsible for whatever is done in their corporate name.

How does management¹ receive the authority to speak in the name of the corporation, and from what source? The orthodox belief that shareholders alone authorise management, in virtue of their ownership of corporate assets (Friedman, 1970; Sternberg, 2000) or their residual risk (Boatright, 2006; Jensen & Meckling, 1976; Sundaram & Inkpen, 2004), is frequently criticised for its failure to distinguish the corporate legal person from its shareholders. The widely held assumptions that shareholders are corporate owners and principals who authorise directors to act on their behalf and have a right to residual earnings have been undermined by an emphasis on legal facts: namely, the separateness of corporate assets from shareholder property and the autonomy of directors to pursue a distinct corporate interest (Chassagnon & Hollandts, 2014; Ciepley, 2013; Deakin, 2012; Ireland, 1999, 2003; Robé, 2011, 2012; Sison & Fontrodona, 2013: 617; Stout, 2002, 2012; Veldman & Willmott, 2013). This critique has been influential in business ethics (e.g., Phillips, 2003: 19–20; Sison & Fontrodona, 2013: 617) and is generally thought to support the conclusion that shareholder primacy² is morally untenable.

However, I shall argue that this conclusion is a non sequitur. It does not follow from the fact of the corporation’s separate legal personhood that shareholders do not authorise managers to act primarily in their interests. As an alternative perspective, I introduce Thomas Hobbes’s (1588–1679) theory of corporate authority. In each of his political works³ Hobbes develops an account of the formation of corporations, in which individuals authorise a representative to serve an entity distinct from themselves. In Leviathan (2012a), a separate corporate “person” is created when individuals authorise a common representative. The process of authorisation and representation that creates the state, Hobbes’s social contract theory, also creates subordinate entities within the state, including the early joint-stock corporations. Hobbes argues that members (e.g., citizens or shareholders) own the actions of their authorised representative. The interests of the business corporation therefore are the collective interests of shareholders and the latter are responsible for what is done in their corporate name. The duties of corporate managers are delegated to them by shareholder assemblies, who speak in the name of the corporation. Moreover, because the shareholder assembly speaks with one voice, the shareholders, who are naturally many, can act as one person in law. Hobbes’s theory therefore accounts for the separate legal person and shows why it is the shareholders (or members) who are ultimately responsible for corporate actions.

Surprisingly, this theory is seldom discussed in the business ethics literature, where interest in Hobbes has instead focused on his “realist” or “egoist” ethics (Gustafson, 2018) and its application to debates on the morality of contracting or bargaining (Boatright, 1992; Gibson, 1994), the place of self-interest in morality (Boatright, 1995; Donaldson, 1994; Kavka, 1983; Luetge, Ambrüster, & Müller, 2016; Van Oosterhout, 2006), the moral obligations of multinational enterprises (Cordeiro, 2003; Neiman, 2013; Palmer, 2001; Velasquez, 1992; Windsor, 2004), and how Hobbes’s “state of nature” explains when market competition can excuse a firm from wrongdoing (Silver & Garofalo, 2024). Hobbes’s influence on the tradition of social contract theory is acknowledged in contractarian approaches to business ethics (Francés-Gómez, 2018; Mansell, 2013a). However, with few exceptions (e.g., Claassen, 2021; Palmer, 2001), Hobbes’s writing on the corporation is overlooked in business ethics.

To remedy this oversight, my primary objective is to demonstrate that Hobbes establishes a logical connection between the creation of a corporation and the authorisation of its directors by its shareholders. Hobbes shows how members act together as a single corporate entity and, by authorising a representative, they endorse, own or hold responsibility for corporate actions. By contrast, the emphasis in the critical literature on the “complete separation” of the corporation from its shareholders (and by implication from its other stakeholders too) has tended to leave unanswered the questions of whose interests count as corporate and which stakeholders (if any) are morally responsible for the corporation’s purpose.

My second objective is exegetical and a necessary means to the first: to interpret the development of Hobbes’s theory of the corporation through each of his major political texts. Crucial passages in the English Leviathan receive fuller elaboration in his earlier texts (Elements of Law and On the Citizen) or subsequent clarification when Hobbes translated Leviathan into Latin. Notwithstanding scholarship on the political context of Hobbes’s philosophy (e.g., Malcolm, 2012; Skinner, 1996, 2002), the substantial continuity between the four iterations of his theory of the corporation, despite enormous changes in British politics from 1640 to 1668, suggests his aim in each successive text was to refine and clarify his argument (for which he sometimes employs new terminology without negating his earlier arguments). I therefore focus on his internal logic rather than his rhetorical response to political circumstances, and I treat the four texts as mutually reinforcing rather than as independent lines of argument.

In the following section I explore the context of the seventeenth-century corporation and challenge a claim that Hobbes was familiar only with partnerships and regulated companies of independent merchants, but knew nothing of the joint stock corporation (Ciepley, 2020; Claassen, 2021). After clarifying Hobbes’s use of the term “authority,” I then introduce the critique of shareholder primacy and examine alternative sources of authority to shareholders in the business ethics literature (e.g., Fia & Sacconi, 2019; Freeman, 1994; McMahon, 1994; Sacconi, 2006). I show that the questions of how managers are authorised to serve the interests of a separate legal person, and who should be responsible for its actions, are not satisfactorily answered there either. I then offer my interpretation of Hobbes’s theory of the corporation. At stake here is a disagreement with Claassen (2021), Fleming (2020) and Kubala (2023) over the place of the state in the authorisation process. These authors read Hobbes as an advocate of what is today called the “concession” or “state grant” theory, in which managerial authority descends from the state. Because the source of authority is, for Hobbes, also the location of responsibility, to read him as a concession theorist is to trace corporate responsibility to the state and not, as I suggest, to the shareholders.

These exegetical sections demonstrate the compatibility of Hobbes’s theory with contemporary discussion on the purpose of the modern business corporation, and also fill a gap in Hobbesian scholarship in business ethics, in which his moral philosophy has attracted more attention than his theory of corporations. Finally, I examine the ethical implications of his argument for the responsibilities of shareholders, shareholder primacy, and the accountability of directors. In doing so, I draw upon recent literature in political philosophy on the responsibilities of members for the wrongful actions of their groups (Collins, 2023; Fleming, 2020; Lawford-Smith, 2019; Pasternak, 2021).

1. HOBBES AND THE SEVENTEENTH-CENTURY BUSINESS CORPORATION

Given that Hobbes’s reference points are the states and “companies of merchants” (1997: V.10, 23) of his own time and not the twenty-first century corporation, to appeal to his arguments in the context of shareholder primacy is to invite an obvious charge of anachronism (e.g., Skinner, 1969). For example, Claassen has followed Ciepley (2020) in asserting that Hobbes was writing of “regulated companies” whose members traded on their own accounts, and that “the joint stock company of these days … is not a corporation” but a partnership (2021: 121).⁴ If Hobbes indeed knew nothing of the business corporation then he could not have been writing about it, and my argument would misrepresent his views. But I shall now argue this historical objection is unsustainable.

English trading companies between the late sixteenth and mid-seventeenth centuries may be placed in three categories. Regulated companies were chartered by the sovereign to prescribe rules of trade in foreign markets and to license individuals or partnerships to trade on these terms (Scott, 1910: 88). These were certainly not equivalents of the modern corporation, for the company itself did not trade: it licensed merchants to trade on their own accounts and assume all the financial risks of their own ventures (Ciepley, 2020: 26–27).⁵ A second category was the joint-stock partnership, which was usually formed when individual traders pooled together to make joint purchases and make large capital investments (Scott, 1912: 12).⁶ Because the partnership was unincorporated, members individually were signatories to each contract and could not transfer personal liability by selling their shares (Scott, 1910: 7).

Besides granting a monopoly, a charter might also incorporate the members. The earliest example appears to be the Russia Company, whose charter of 1555 created “one bodie and perpetuall fellowship and communaltie” (Scott, 1912: 37). As Keay (1991: 27) writes of the East India Company, chartered in 1600, because of its “national importance” and “long term, high risk” activities it was accepted that the “Company alone, must itself conduct all business.” Incorporation meant that the members traded as a single entity with perpetual succession and a common seal, and they were “‘able and capax in law’ of holding lands and of suing and being sued under the [corporate] name” (Scott, 1912: 38). Shareholders typically subscribed for a succession of voyages (1912: 52) and received intermittent dividends on the joint stock profits. The Russia Company, for example, “locked up” its capital “for only a few different stocks in the sixty-seven years up to 1620” (1912: 153). Shareholders in this period were not active partners but capitalists who were uninvolved in the management of the business and who could sell their shares to outsiders (1912: 45). They elected the board of directors, who would then manage their capital investment for the duration of the subscription (Scott, 1910: 72).

Of the trading companies familiar to Hobbes, this third category—the incorporated joint-stock company with capital controlled by directors rather than by shareholders—is the nearest relation to the modern corporation and is the type he discusses in Leviathan. In Chapter XXII, he asks why merchants would choose to incorporate when they could otherwise “buy and sell … according to their own discretions” (Hobbes, 2012a: 362). Although “there be few Merchants, that with the Merchandise they buy at home, can fraight a Ship, to export it,” the motive to incorporate cannot simply be to pool capital to offset the costs of trade, for “this is no Body Politique” or corporation (2012a: 362). What makes the company a corporation is the charter that grants a monopoly: “to grant to a Company of Merchants to be a Corporation, or Body Politique, is to grant them a double Monopoly, whereof one is to be sole buyers; another to be sole sellers” (2012a: 362). In other words, an additional rationale is needed (besides setup costs) to explain why a joint-stock partnership would seek a charter of incorporation. The answer is the profits of monopoly: “to make their gaine the greater … by sole buying, and sole selling, both at home, and abroad” (2012a: 362). Hobbes also refers to the reasonable motive of every merchant, or shareholder, to participate in deciding how their investment will be employed. An assembly of all the shareholders is therefore the corporation’s most appropriate representative (2012a: 364). Moreover, Hobbes had first-hand experience to draw upon, for he was from 1622 to 1624 an active shareholder in the Virginia Company (Jessen, 2012: 77; Turner, 2016: 215), whose trading activities were arranged on a joint-stock basis from 1609 (Jessen, 2012: 76; Scott, 1910: 249).

Hobbes criticised the monopolistic character of joint-stock enterprise, writing that “one part is disadvantageous to the people at home, the other to forraigners” because these companies set what price they choose on imports and exports (2012a: XXII, 364), and he compares the concentration of wealth in monopolies to the gathering of “Blood in a Pleurisie” (2012a: XXIX, 516). Moreover, according to Fitzmaurice’s research on democracy in the Virginia Company, Hobbes objected that corporations “created a space in which political discourses could develop a life of their own, discourses such as democracy …” (2022: 326). Although a concern with corporations as “nurseries of democratic thought” (2022: 309) is a criticism rarely levelled at the modern corporation, the entities about which he was writing were certainly trading corporations, such as the Russia Company or the East India Company after 1612. These enterprises had external shareholders, traded in their own name, and held a “common stock” under central management. Therefore, although Hobbes writes of “members” or “merchants,” it is historically acceptable to use the term “shareholder” when expounding his theory of the corporation, and in drawing upon it to rethink shareholder primacy no anachronism is implied.

2. MANAGERIAL AUTHORITY AND CRITIQUES OF SHAREHOLDER PRIMACY

Before comparing Hobbes’s position with alternative perspectives on corporate authority, I first isolate the definition relevant to this debate. One definition of “authority,” its most familiar of seven distinct meanings, is given by the Oxford English Dictionary (OED) as the “Power or right to give orders, make decisions, and enforce obedience” (2014: 2). Likewise, McMahon (1994) and Ciepley (2014) begin their own surveys in the same place: “having a right to direct the actions of someone else” (McMahon, 1994: 47) and “governing authority, the ‘right to rule’” (Ciepley, 2014: 1). However, while the managers of a business corporation indeed hold authority to govern employees and manage the firm’s assets, they do not do so on their own behalf. As noted by McMahon (1994: 49–50) and Ciepley (2014: 2–3), managers are not principals. We might therefore ask how managers collectively are authorised in their role of giving orders, or how they receive their “governing authority.” This would be to understand managerial authority in a different sense, as “power derived from or conferred by another” (OED, 2014: 3). McMahon (1994: 47) notes this usage is characteristic of agency relationships, in which “someone who is authorised to perform a certain action is said to have authority to perform it” and the agent has “authority to act for the principal, in the sense of being authorised by the principal to act on her behalf” (1994: 49). This notion is consistent with a principal-agent model of managerial authority. For Ciepley (2014: 2), authorising (in this sense) means “delegating authority to subordinates or others, authorising them to act in the authority’s name.”

This understanding of authority as authorisation is Hobbes’s starting point. For Hobbes (2012a: XVI, 244), an “author” is one who owns the words and actions of his representative (the “actor”). Authority is defined as the right of “doing any action” so that to act “by authority” implies a right of doing the act, and “done by authority” means “done by commission, or license from him whose right it is” (2012a: 244). If the author owns the actions of the representative, then when the actor makes an agreement “by authority” he binds the author no less than if the author had acted himself (2012a: 246). And because authorisation always implies a transfer of right, control is ceded to the representative, and then the author “is said to be OBLIGED, or BOUND, not to hinder those, to whom such Right is granted” (2012a: 200). Skinner aptly calls it a “duty of non-interference.” As an author “you must leave it to your representative, who is now in possession of your right of action, to exercise it at his discretion when acting in your name” (2002: 185–86). This could, of course, describe the separation of ownership and control (Berle & Means, 1932) in the modern business corporation. But from where do managers receive their right of acting for the corporation?

In the business ethics literature, normative arguments concerning corporate purpose typically involve an explicit rejection of shareholder primacy,⁷ and especially the assumption that shareholders confer authority on management in virtue of their ownership of corporate assets (e.g., Friedman, 1970; Sternberg, 2000: 41). Although I shall later challenge this conclusion, the legal personhood of the corporation is held to be incompatible with the granting of managerial authority by shareholders alone. This argument typically begins from the correct observation that the corporation is a person in law with its own property rights and therefore cannot be owned by shareholders. Phillips (2003: 19), in advancing a normative stakeholder theory of the firm, asks “[w]ho would own the corporation if it bought back its own stock?” The corporation, he answers, “is not coextensive with the shareholders. It is an entity unto itself…” (2003: 19–20). Stout (2012: 37) remarks that “shareholders do not, and cannot, own corporations. Corporations are independent legal entities that own themselves, just as human beings own themselves.” For Deakin (2012: 356), the assets of the firm “vest in the separate legal person of the ‘corporation’ … The corporation, in turn, cannot be owned as a ‘thing’ precisely because (juridically speaking) it is a person.”⁸ And Ciepley (2013: 146) argues that “liberals” make a “fundamental mistake” in treating “shareholders as if they were the owners” because no one “owns the assets of the corporation …. It is corporate property.”⁹

However, the argument that corporate assets cannot legally be owned by individual shareholders does not amount to a refutation of shareholder primacy unless it also implies that shareholders are not principals who authorise directors (or agents) to manage the corporation on their behalf. It could then be concluded that managers have no obligation to prioritise shareholder interests. Although I will argue this conclusion does not follow, for shareholders do indeed authorise directors without having individual rights to corporate assets, it is the step invariably taken in the critical literature. For example, Robé (2011: 3–4) contends that managers are agents of the corporation, not of the shareholders, because shareholders do not own the firm’s assets and shareholding is not co-ownership (2011: 24–29). Therefore, directors are no one’s agents but must instead act in the interests of the corporation (2011: 56). Similarly, for Deakin (2012: 360), shareholders cannot be principals because “company law views directors as the agents of the company, and not of the shareholders.” And Stout (2012: 42) describes it as a “mistaken assumption” that “shareholders are principals and directors are their agents,” in part because the corporation and its directors “must exist prior to, and independent of, the stockholders (the supposed ‘principals’).” She also adds that shareholders do not have a “residual claim” to the corporation’s cash flows because the corporation (a separate legal entity) is its own residual claimant (2012: 38–41).

What these authors would deny is that shareholders establish the corporation by authorising directors. Directors may be elected by a vote of shareholders but, in law, their fiduciary duties are to a separate corporate entity whose proper purpose is altogether distinguishable from shareholders’ interests. Indeed, Delaware law refers in numerous places to “the best interests of the corporation”¹⁰ without implying synonymity with stockholder interests. Similarly, the UK Companies Act 2006 requires a director to “promote the success of the company” (s. 172[1])¹¹ and, though it immediately adds “for the benefit of the members as a whole,” it then allows for “purposes other than the benefit of its members” (s. 172[2]). The critiques of shareholder primacy therefore find support in legal systems that refer to a distinct corporate interest.

Nonetheless, by emphasising the complete separation of the corporation from its shareholders, what these authors leave unanswered is who should be held responsible for what is done in the corporation’s name. Corporations can act only through their boards of directors, who delegate authority to managers and other employees. The corporation itself is not a responsible principal. But directors are not principals either: they are merely agents who act for no one but the corporation (Deakin, 2012: 360; Robé, 2011: 3–4; Stout, 2012: 42). Provided directors comply with their duties in law they do not act for themselves but always for another, and yet that other (the corporation) can do nothing except through the directors. If neither the corporation nor their agent should ultimately be held responsible for corporate actions, we have a responsibility gap.

Although arguments are made for alternative sources of authority to shareholders (e.g., government, the people, stakeholders, or employees), the questions of how managers are authorised to serve the interests of a separate legal person and who is morally responsible for that person’s actions, are not answered there either.¹² In McMahon’s (1994) landmark study Authority and Democracy, legitimate authority must be justified to employees “from the standpoint of reasons that apply to them” (1994: 192). He justifies democratic voting, in this context, by its efficacy in achieving fairness and welfare maximisation for employees (1994: 258). McMahon concludes by calling for a “transfer of all ultimate managerial authority to democratically elected representatives of the employees” (1994: 292). His justifications, therefore, all point to the interests or purposes of employees. He does not analyse the corporation as a distinct concept, except to note that it is a locus of independent legal rights and duties (1994: 8), and therefore does not focus on how managers are authorised to pursue the “best interests of the corporation” or who bears responsibility for these interests.

Another alternative to shareholder primacy is to locate the source of managerial authority in a government concession. Opposing the “liberal, contractual theory” that treats the corporation as a “pure creature of the market” (2013: 139–40), Ciepley argues that although the modern corporation is run on private initiative it is nonetheless governmental in provenance, because the state’s charter “ordains its existence and specifies its rights and obligations.” This is because the “normal market rules of property and liability” must be broken to separate the corporation’s rights and liabilities from the shareholders’ (2013: 145). Not only the legal personhood of the corporation but also managerial authority is “of government” (2013: 149) because the powers granted to directors require the state’s concession of a corporate charter that overrides normal market rules. Notwithstanding the “nominal right” of shareholders to elect directors, “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” (2013: 150–51). Ciepley concludes that “the principal is the corporation itself, and management is its agent” and “[t]he state is the foundation of managerial authority” (2013: 151).

Ciepley convincingly demonstrates the necessity for state involvement in the constitution of corporations as legal persons with specific powers over workers. The modern corporation is not merely a creation of market contracts. However, if the power conferred on managers has an exclusively public source (i.e., the state), then what is left unexplained is how directors could be authorised to pursue the particular interest or “private initiative” (2013: 140) of an individual corporation. Consequently, insofar as directors act as agents for the corporation (the principal) the source of responsibility for their actions is unclear, unless the corporation has another means of acting except through the directors. What is missing is the connection between managerial authority and a corporate interest that is reducible neither to the private initiative of directors nor to the public interest.

3. SOCIAL CONTRACT THEORIES OF THE FIRM

I now argue that this connection is established by the tradition of social contract theory (to which Hobbes belongs): specifically, in application to business ethics, its principle of serving an interest impartially acceptable to all corporate members. For example, Freeman (1994) applies Rawls’s hypothesis of an “original position” to the stakeholders of a firm to elucidate “ground rules” for a “corporate constitution” (1994: 416–17) that all stakeholders (employees, financiers, customers, etc.) would agree to in ignorance of their actual stakes. Embracing “a multiple principal/multiple stakeholder model” (1994: 412), stakeholders “hire managers who are fiduciaries to their interest and the interest of the collective” and “the corporation shall be managed as if it can continue to serve the interests of stakeholders through time” (1994: 417). The principle of impartiality that underpins these “ground rules” means that it is not any particular stakeholder whose interest must be served, but rather the interests of all stakeholders in perpetuity—an abstract interest which could be identified with the purpose of the corporation itself, and for which all stakeholders would share responsibility. Although Freeman (1994) did not take this step, Freeman, Phillips, and Sisodia (2020: 220) note that firms increasingly serve their “own higher purpose,” which is shared among the firm’s stakeholders.

In Sacconi’s extension of Freeman’s (1994) argument, stakeholders arrive at a constitutional choice for the firm (2006: 268–69) through an “impartial” process of rational bargaining. According to Francés-Gómez, the purpose is to imagine “what kind of governance structure would be set up by all relevant stakeholders of a corporation if they had to draft a constitution for the firm without knowledge of the position each was going to occupy” (2020: 289). Through participation in the firm’s social contract, the trust of all stakeholders “constitutes the authority of the firm’s owner and manager” (Sacconi, 2006: 278). It follows that the firm “owes fiduciary duties to all its stakeholders,” which means a responsibility “to exercise authority for the good of those who have granted that authority” (Sacconi, 2006: 263).

Like Freeman, Sacconi holds that the enterprise “does not separately pursue the interest of each of the stakeholders” but instead aims “at solving problems of coordination” by pursuing the bargaining solution “to which these stakeholders would agree” (Fia & Sacconi, 2019: 941). Although these authors do not link authorisation by stakeholders to responsibility for corporate actions, their method succeeds in showing how managers could be authorised to pursue a corporate interest abstracted from the temporary interests of individual stakeholders. The social contract method seems a viable step in meeting the challenge set by the critiques of shareholder primacy: that of explaining how managers could acquire authority to speak in the name of a distinct corporate entity.

Another notable example is that of Blair and Stout (1999). With an explicit reference to Hobbes’s Leviathan (1999: 274–75), they argue that a corporation is created when stakeholders cede control of firm-specific investments to a board of directors whose duty is to prevent rent-seeking and shirking among members of the team (1999: 250–51, 274, 276–77). Because directors owe fiduciary duties to the team as a whole or “corporate coalition” (1999: 283), a separate entity in law, they are authorised to act not as agents for a principal who can control their actions, but rather as independent “trustees for the corporation itself” (1999: 281). “The interests of the corporation,” they write, “can be understood as the joint welfare function of all the individuals who make firm-specific investments” (1999: 288).

Who is ultimately responsible for representing the corporate interest? For Blair and Stout, the corporation cannot act for itself but only “through its human agents” (1999: 293). Directors receive their “broad delegation of authority” (1999: 290) from the team itself, but the latter lacks sufficient unity of purpose to bring directors to account (1999: 313). They argue it may be more efficient for shareholders to represent the corporation (e.g., by bringing derivative suits against directors) if doing so is also in the economic interests of the whole team (1999: 289, 293, 297, 309–13). The collective economic welfare of the team is therefore the criterion for deciding whether directors or shareholders act in the corporate interest (1999: 325).

But does the corporation’s interest not also include external responsibilities? What should be the team’s moral position on, for example, climate change, corporate lobbying, animal welfare, income inequality, or their response to a global pandemic?¹³ And who is to decide these questions if the team cannot speak for itself but only through the board of directors or the shareholders, and both are bound to serve the team? The step Blair and Stout have not taken is to say that those who incorporate by authorising a representative are, by implication, responsible for determining what the interests of that corporation ought to be.

In the following sections (4–7), I shall argue that it is coherent to see shareholders as the group who incorporate into a separate legal person and who are therefore responsible for determining its proper interests. In contrast with how the literature in business ethics and corporate law has handled managerial authority, Hobbes explains how individuals form themselves into a separate corporate entity by authorising (or bestowing a right upon) a representative to speak on their behalf. By doing so, they transfer control to their representative and retain responsibility for everything their agent does in their name.

4. THE CORPORATION IN HOBBES’S SOCIAL CONTRACT THEORY: THE ELEMENTS OF LAW AND ON THE CITIZEN

The early stages of Hobbes’s theory of the corporation appear in his earliest political treatises, The Elements of Law and On the Citizen. The first, written as a response to the threat of civil war, was circulated in manuscript copies from 1640 (Gaskin, 1994). This work is a summary in English of two sections, “Man” and “The Citizen,” of Hobbes’s projected Latin work Elementa Philosophiae (“The Elements of Philosophy”). The final section (On the Citizen or De Cive) was then published for a wider audience in 1647 (Tuck, 1997: xii).

Writing in The Elements of Law of the “necessity and definition of a body politic,” Hobbes argues that “sufficient security for their common peace” cannot be established by the mere “concurrence of many men’s wills to one action”; instead, the only way is “union”—“the involving or including the wills of many in the will of one man” or one council (XIX.6, 106). In this manner, a body politic, civil society, or city is created, in which a “multitude of men” are “united as one person by a common power, for their common peace, defence, and benefit” (XIX.8, 107). Hobbes adds that no action could be attributed to every member of the multitude unless the wills of a number of them or “the will of some one man” were to stand “for the wills of every man. And this done they are united, and a body politic” (XX.3, 110). These are the elements of what has become known as Hobbes’s social contract theory of the state.¹⁴

However, he observes that “common peace” is not the only end for which a “union” may be formed. For amongst a multitude there may be “instituted a subordinate union of certain men, for certain common actions to be done by those men for some common benefit of theirs, or of the whole city” (for example, “trade”) (1994: XIX.9, 107). He continues: “these subordinate bodies politic are usually called CORPORATIONS; and their power such … as the whole city whereof they are members have allowed them” (1994: 107). Here we see the rudiments of a theory of corporate authorisation, in which a number of individuals take the will of one person (or an assembly) for their own, and thereby act together as one, either for a private benefit common to themselves or a public benefit for the city. Hobbes claims originality in reasoning by analogy from the joint-stock company to the state: “though in the charters of subordinate corporations, a corporation be declared to be one person in law, yet the same hath not been taken notice of in the body of a commonwealth or city” (1994: XXVII.7, 167–68). It has therefore been argued persuasively that Hobbes modelled his theory of the state on the corporation and not vice versa (e.g., Claassen, 2021: 113; Jessen, 2012: 73; Kubala, 2023: 8; Turner, 2016: 216–17; Cf. Simendić, 2012: 26–27).

In On the Citizen, Hobbes repeats that a “union” (or “commonwealth,” “civil society,” or “civil person”) requires not merely a “combination of several wills in the same end” but rather a “single will,” and this “can only happen if each man subjects his will to the will of a single other” (1997: V.6, 72). If a crowd (multitudo) “individually agree that the will of some one man or the consenting wills of a majority of themselves is to be taken as the will of [them] all, that number then becomes one person” (1997: VI.1, 76). Here he is more explicit that, since it has one will, the union thus formed may act as one person in law: “it is to be taken as one person; and is to be distinguished … by a unique name from all particular men, having its own rights and its own property” (1997: V.9, 73). Again, not all civil persons are commonwealths: “it may happen that several persons will, with the permission of their commonwealth, unite as one person for the purpose of transacting certain business. These will now be civil persons, as companies of merchants are …” (1997: V.10, 73).

In the context of the critique of shareholder primacy, it can be seen in Hobbes that a corporate “person” with its own rights may be created when its members submit to the will of another who speaks in their corporate name. In the case of the business corporation, as seen in Section 1, Hobbes’s preference is for this representative to be an assembly of all the shareholders. A legitimate corporate interest can reflect the shareholders’ interests in choosing to incorporate, even though no shareholder has an individual right to the corporation’s assets.

However, this reading of Hobbes has not accounted for the role of the state in chartering corporations, an inescapable element in the creation of business corporations in his time. Hobbes in The Elements of Law writes of powers “allowed” to the corporation by the city, and in On the Citizen of several persons subjecting “themselves to the will of the group … only in certain matters defined by the commonwealth” (1997: V.10, 73). The commonwealth or city therefore sets explicit limits on the power the members can exercise, as a group, over themselves as individuals. Does Hobbes provide support for a “concession theory” of the corporation, as Claassen (2021), Fleming (2020: 94–95), and Kubala (2023: 9) have recently argued? A Hobbesian concession theory, for Claassen, does not insist on an exclusively public role for corporations, but would accommodate the “simultaneous presence of private and public purpose, and the reinterpretation of wealth maximization as itself a public purpose” (2021: 103). Nonetheless, his interpretation of Hobbes is that corporations are “created by governmental initiative” (2021: 118) and “represent and are authorised by the public” (2021: 128). Similarly, for Fleming, Hobbes’s corporations are “extensions of the state” (2020: 94).

These authors base their interpretation primarily on Chapter XXII of Leviathan, which I shall consider in Section 6. As I have shown, in Hobbes’s earlier treatises, though permission from the city (or state) is clearly required for the exercise of corporate power, he allows that it might be used for a “common benefit” of the members and not necessarily for the state. This suggests that the members (or shareholders) are integral to the determination of the corporation’s purpose. On the Citizen gives an active role in the corporation’s creation to the members: they “unite as one person” and have “subjected themselves to the will of the group.” Likewise, in Leviathan, merchants “bind themselves up in one Corporation” and “joyn together in one Society” (2012a: XXII, 362). Without these positive actions from the members, in which they choose their own representative, there is no union.

Although it is not until Leviathan that Hobbes writes of the “authority” of a corporate “representative,” a process of authorisation is already apparent in these earlier texts. First, a plurality of individuals acquire permission from the commonwealth to let the will of a representative stand for the will of them all, and thereby to act as one person in law. The permission (or authority) thus granted is subject to limits set by the state, and these limits depend on the purposes for which this permission is granted, that is, “for the purpose of transacting certain business” (1997: V.10, 73). Second, individuals exercise this right and become members of one corporation, into which they are now united by their common act of submission. The corporation stands apart from them, “distinguished … by a unique name” (1997: V.9, 73), but is not altogether independent: it is constituted by its members; and this implies, in the case of the joint-stock corporation, by its shareholders.¹⁵

5. PERSONHOOD, AUTHORITY, AND REPRESENTATION IN LEVIATHAN

In On the Citizen a crowd “becomes one person” when a single will stands for the will of them all. In Leviathan (2012a), Hobbes introduces the vocabulary of “authors” and “representatives.” He notes that persona in Latin “signifies the disguise or outward appearance of a man, counterfeited on the stage” (XVI, 244). He reasons that “a Person, is the same that an Actor is … and to Personate, is to Act, or Represent himself, or an other; and he that acteth another, is said to beare his Person, or act in his name” (XVI, 244). A person, therefore, “is he, whose words or actions are considered either as his own, or as representing the words or actions of an other man, or of any other thing to which they are attributed” (XVI, 244).

He defines a person who speaks in her own name as “natural” and one who represents another as “artificial” (XVI, 244). He here introduces the concept of authority which, as noted in Section 2, denotes ownership of actions. Natural persons may speak for themselves but if an artificial (representative) person speaks in their name, then the representative is the actor (XVI, 244) and “the one represented is called the author, as he is the one by whose authority the actor acts” (Latin Leviathan, 2012a: XVI, 245). Authorisation also implies a transfer of the right to act and a corresponding duty, on the part of the author, not to interfere with or attempt to control the actions of their representative.

Hobbes’s theory of persons has attracted much discussion in the scholarly literature because of the inconsistency with which he uses his new terminology. If all persons are representatives or “spokespersons” (Pettit, 2009: 56), then whom does the state or corporation represent if it too is a “person”? One might assume that the state and corporation both represent their members, that is, citizens or shareholders. However, the members have not collectively authorised the state or corporation for it has no prior existence. A multitude becomes a state or corporation “when they are by one man, or one Person, Represented … For it is the Unity of the Representer, not the Unity of the Represented, that maketh the Person One” (XVI, 248).¹⁶ It therefore appears that the “person” of the state or corporation is not a spokesperson but a represented person—one who cannot speak but can be spoken for. Fleming (2021) notices that Hobbes, in Chapters XLI and XLII of Leviathan, interprets his own definition of “person” from Chapter XVI to mean “he that is Represented, as often as hee is Represented.” For example, “God, who has been Represented (that is, Personated) thrice, may properly enough be said to be three persons” (2012b: XLII, 776). Hobbes’s understanding of persons is therefore double-sided: they can be representatives or representees and “one has to use the context as a guide” (Fleming, 2021: 12).

Hobbes applies his concepts of personhood, authority, and representation to give his famous account of the commonwealth. He writes “the only way to erect such a Common Power” is for individuals:

to conferre all their power and strength upon one Man, or upon one Assembly of men, that may reduce all their Wills … unto one Will: which is as much as to say, to appoint one Man, or Assembly of men, to beare their Person … This is more than Consent, or Concord; it is a reall Unitie of them all, in one and the same Person. (XVII, 260)

Everyone is to “owne, and acknowledge himself to be Author of whatsoever he that so beareth their Person, shall Act” (XVII, 260). A commonwealth is therefore defined as “One Person, of whose Acts a great Multitude, by mutuall Covenants one with another, have made themselves every one the Author, to the end he may use the strength and means of them all, as he shall think expedient, for their Peace and Common Defence” (XVII, 260–262). It is therefore the granting of authority to a single representative person by each member of the multitude that turns them all into one represented person (the state).

6. CORPORATE AUTHORITY IN LEVIATHAN

As indicated in his earlier works, the state is not the only case in which individuals may unite into one person. In Chapter XXII, Hobbes turns to the “parts” of the commonwealth which, by analogy with the “Muscles of a Body naturall,” he names “systems”—“any number of men joined in one Interest, or one Businesse” (348). He distinguishes “irregular” systems, consisting “only in concourse of People,” from “regular,” which are “those, where one Man, or Assembly of men, is constituted Representative [in the Latin, ‘bears the person’] of the whole number” (XXII, 348). Commonwealths are “Absolute, and Independent” whereas other systems are “Dependent, that is to say, Subordinate to some Soveraign Power” (XXII, 348). He had previously labelled the latter “subordinate bodies politic,” “corporations,” and “subordinate civil persons,” and now (interchangeably) as “systemes subordinate” (XXII, 348).

For Hobbes, what each type of corporation or “subordinate system” has in common is subordination to the sovereign power. It is on the nature of this subordination that the question of corporate authority arguably depends. Claasen attempts to trace the business corporation “back to an authorising author” and concludes that “only the state can fulfil this role” (2021: 129). In other words, the state owns and authorises the actions of the corporate representative. Likewise, in Fleming’s reading of Hobbes, “an act of a corporate body is an act of state” (2020: 94). Both authors draw their evidence from the following passages from Chapter XXII of Leviathan. Subordinate “bodies politique” or “persons in law” are “made by authority from the Soveraign Power of the Common-wealth” (348), and their power is limited by “their Writt, or Letters from the Soveraign” and “the Law of the Common-wealth” (350). What the representative does within these limits “is the act of every one: For of the Act of the Soveraign every one is Author … and the act of him that recedes not from the Letters of the Soveraign, is the act of the Soveraign, and therefore every member of the Body is Author of it” (352).

Claassen finds in this text a “two-step process of authorisation” in which “all citizens have authorised the sovereign, and the sovereign has authorised (through the Letters and/or the Law) the specific political body” (2021: 117). Claassen concludes that “as the state itself is authorised by the multitude, all the citizens ‘authorise’, or ‘own’, the acts of all corporations” (2021: 129). On this basis, if corporations act according to their mandates, they “fulfil their publicly mandated purpose” (2021: 129).¹⁷ However, noticing that Hobbes’s criticism of monopoly also implies a positive role for profit under the “right economic conditions” (2021: 126), he infers a public interest “compatibility requirement” (2021: 119) for corporations’ private commercial interests. Corporations therefore serve the public by fulfilling “an economic purpose (the profitability of the Commonwealth)” which implies “the infusion of economic activity with public purpose” (2021: 126).

Although I agree with this position on the state’s role in regulating corporate purpose, I question Claassen’s “process of authorisation” in which the corporation is directed to pursue its purpose by the state’s citizens (indirectly), and by the sovereign (directly), but not at all by its own shareholders. Drawing upon familiar critiques of shareholder primacy, he excludes shareholders as a possible source of managerial authority. He writes, citing Stout (2012) and Ciepley (2020), that “it is dubious to describe shareholders as owners of the corporation. The shareholders hold shares, but this gives them no direct control rights over the corporation’s assets. The corporation … is a legal person not itself owned by anyone” (2021: 114–15). He is also inclined to deny that business corporations even have “members” (2021: 115). Yet his authorisation process does not explain why the “actors” of the corporation have authority to act on its behalf, unless the right to do so is granted directly by the state. As I shall argue in Section 7, significant implications follow for locating responsibility for corporate action. If it could be shown, for example, that the corporation’s shareholders confer this right on their representative, they would be “authors” by Hobbes’s definition of authority and would be the responsible owners of actions taken by directors in the corporation’s name.

In Hobbes’s earlier texts, the initiative of private individuals is integral to the process of authorising a corporate representative. Hobbes has not changed his mind on this point in Leviathan. First, he assumes that a corporation not only has members but is a body comprised of them. He refers to merchants who “bind themselves up in one Corporation” and writes of a “grant to a Company of Merchants to be a Corporation, or Body Politique …” (XXII, 362). He repeatedly describes these incorporated merchants as “bodies” (e.g., “Bodies of Merchants,” XXII, 364) and adds that their “most commodious Representative is an Assembly of all the members: that is to say, such a one, as every one that adventureth his mony, may be present at all the Deliberations … of the Body” (XXII, 362).

If corporations are indeed constituted by their members and their most suitable representative is an assembly of them all (or in today’s terms, a general meeting of shareholders), are Claassen and Fleming correct that corporate actions are authorised by all the citizens of the state? Where Hobbes writes that what is done by the corporate representative within the “Letters, or the Law” is “the act of every one,” he appears to refer to all the citizens. However, in the Latin edition, he immediately adds “of those who compose the body” (2012a: XXII, 352–353). Although the Hobbesian sovereign passes the laws that regulate corporations with authority from the citizens, it is the business corporation’s shareholders (if they are citizens)¹⁸ who authorise all that their representative does according to the laws and their charter. This is because, as citizens, they have authorised the sovereign’s actions in making these laws and granting this charter, and therefore cannot disclaim ownership of actions legitimately taken in their name. Conversely, shareholders are not responsible for what the sovereign has not permitted: “whatsoever [the representative] does … which is not warranted in his Letters, nor by the Lawes, is his own act, and not the act of the Body, nor of any other Member thereof besides himselfe” (2012a: XXII, 352).

Hobbes is concerned here to establish limits on the power to represent a “Body Politique of Subjects” (XXII, 350) when it encompasses part of the citizenry but not the whole. The power of such a representative is “alwais limited” (XXII, 350) and “cannot Represent any man in things unwarranted by their Letters” (XXII, 352). But the sovereign’s laws and letters do not appoint the representative—they do not authorise one spokesperson, rather than another, with the right to represent the shareholders. “The bounds of that Power, which is given to the Representative” (XXII, 350) are not the power itself. Instead, the sovereign grants individuals (who become members upon incorporation) the right to endow a representative with the power to speak for the corporation within the limits of the laws and the charter.¹⁹

Although Claassen (2021: 128) finds in Hobbes the idea that corporations “represent and are authorised by the public,” my reading is that it is the representative that is authorised by members who thereby form a corporation. In the case of a subordinate body politic (e.g., a business corporation) the representative acts for the corporation through specific legal powers granted by the sovereign, but it is not this sovereign concession that makes a particular person representative. Indeed, in the Latin Leviathan, when Hobbes argues that one cannot authorise a representative in the state of nature but only “by civil right,” he uses the phrase “per concessionem Civitatis” to mean (in Malcolm’s translation) “by permission of the Commonwealth” (2012a: 248–49). A state’s concession therefore gives permission to the corporation’s future shareholders to incorporate, but it does not make the state itself the author and owner of the corporation’s actions.

Claassen (2021) convincingly renders the private commercial interest of the corporation compatible, on Hobbesian grounds, with the public interest served by competitive markets. But on whose authority is the “private purpose” (2021: 121) of any corporation to be pursued? If the purpose is private, the author cannot be the sovereign, who is authorised to act only for the state and exclusively for public purposes. I have therefore argued that, for Hobbes, the representative (in whose person the members are united) has authority from the members or shareholders to speak in the corporation’s name, within the “bounds” (2012a: XXII, 350) or limitations of the sovereign’s charter. And for the representative to act for the shareholders together, is for the shareholders be a corporation, provided the law recognises them as such.²⁰

7. IMPLICATIONS FOR SHAREHOLDER RESPONSIBILITY

Who represents, or acts for, the modern business corporation? If shareholders individually are the corporation’s authors then, for Hobbes, their “most commodious representative” is the shareholders acting together in assembly (2012a: XXVI, 362). The equivalent in a modern corporation is the shareholders’ annual meeting. The collective decision-making of shareholders, through processes regulated by corporate law, constitutes the voice of the corporation. In a further act of authorisation, the shareholders elect directors to whom they transfer the right to act for the corporation. Critics of shareholder primacy urge that directors are responsible for serving the interests of the corporation as a distinct entity, although in practice (rightly or wrongly) directors must answer to the shareholders. In Hobbes’s theory, because the collective voice of the shareholders is the voice of the corporation, these amount to the same thing.

Blair and Stout (1999: 290) argue that directors cannot be the agents of shareholders because “an agent owes her principal a ‘duty of obedience’” and corporate directors “are not subject to direct control or supervision by anyone….” However, as described earlier, authorisation for Hobbes entails a transfer of the right of control and a corresponding duty, on the part of the principal, “not to hinder” the agent (2012a, XIV: 200). The principal has no right to control the actions of the agent, precisely because control is what has been ceded. For example, citizens of a modern democratic state typically have no right (individually or collectively) to interfere with the decision-making of democratic legislatures, and individual shareholders cannot control (though they may participate in) the deliberations of shareholders in general meeting. But the democratic legislature or shareholder assembly still has a duty to work for the interests of its individual citizens or shareholders—in other words, to represent them.

On this point, therefore, the critique of the principal-agent relationship between shareholders and directors is mistaken. It is asserted that because they do not own corporate assets, shareholders cannot be principals (Deakin, 2012; Robé, 2011; Stout, 2012), and even if they were, directors would not be their agents because shareholders cannot control them (Blair & Stout, 1999; Stout, 2012: 42). Hobbes accepts both premises: citizens or shareholders do not own the assets of a state or business corporation, and they have bestowed control on a representative who may delegate that authority to a government or board of directors. However, neither premise leads to the conclusion that the representative has no obligation to serve the citizens’ or shareholders’ interests. For both the creation of a corporate person and the transfer of control are implied when individuals incorporate by authorising a representative. And if these individuals become shareholders upon incorporation—as they do in a business corporation with share capital—then the corporation is created to further their interests.

What responsibilities do shareholders have as principals? As noted in Section 2, Hobbes defines an “author” as one who owns the actions of his representative (2012a: XVI, 244). When a representative makes an agreement “by authority” he binds the author and subjects him “to all the consequences of the same” (XVI, 246). Skinner interprets Hobbes to mean that authors have a duty to “own up” and “take responsibility” for their representative’s actions, because these actions count as their own (2002: 185–87). In Pettit’s (2009: 57) reading, authorisation means to own or endorse the actions that are authorised. And for Runciman (2000: 271), those who are truly represented must be capable of acting in their own name, and thus of taking responsibility for what is done on their behalf. In the case of the state, the multitude are “jointly committed to taking responsibility for what the sovereign does” (2000: 273).

In a typical modern business corporation, a “hierarchical chain” of authorised representation (Fleming, 2020: 78) runs from individuals who, with permission from the state, incorporate by authorising a representative assembly of shareholders. Executive directors acquire their authority from these shareholders in general meeting. The directors then hire senior managers who select and supervise lower-ranking employees. At every step in the chain, is each author morally responsible for all the actions of their subordinates? Hobbes’s position is more subtle. If the representative is a (shareholder) assembly, anything it decrees contrary to law is attributed to the whole corporation and is “the act of every one by whose Vote the Decree was made” (2012a, XXII: 352). Similarly, when the corporation owes a debt, every shareholder that voted for the borrowing, or for the deed that incurred the fine, is liable “even of the whole debt” (XXII, 354). It may therefore be expedient for a particular shareholder “to make open protestation against the decrees of the Representative Assembly … because otherwise they may be obliged to pay debts contracted, and be responsible for crimes committed by other men” (XXII, 356).

To explain Hobbes’s reasoning in attributing actions to the whole corporation and to culpable individuals, Fleming (2020: 99) distinguishes several forms of responsibility, including: 1) ownership or “nominal responsibility” for an action, and 2) culpability, “meaning guilt or blame.” Culpability “accrues only to the agent who intended the wrongful action.” That is, although ownership “can be incurred vicariously, culpability cannot be, because intent cannot be transferred from a representative to a representee” (2020: 101). If an employee commits fraud against a customer, the blame is not attributable to the employer unless they are complicit (2020: 101). By implication, although ownership of a decision by the shareholder assembly can be attributed to each of its members, unless every shareholder voted for it, moral blame or culpability cannot.

By the same reasoning, a corporation or state cannot itself be culpable. Hobbes explicitly rejects what is today known as “corporate moral agency” (Fleming, 2021: 19–22). In The Elements of Law, he argues that “to make a particular man unjust, which consisteth of a body and soul natural, there is required a natural and very will.” By contrast, “a body politic, as it is a fictitious body, so are the faculties and will thereof fictitious also” (1994: XXI.4, 120). In On the Citizen (1997: VII.14, 97), only the citizens who vote for a decision contrary to natural law are the offenders and not the state itself, for otherwise “those who voted against the decision would also be offenders.” For Hobbes, moral guilt or blame cannot be attributed to the corporation because only natural persons are bound by moral obligations. For the same reason, an assembly of shareholders should not itself be morally blamed for corporate actions. Responsibility lies with its individual members (Runciman, 2000: 70). According to Hobbes, only the author’s intention is culpable: “When the Actor doth any thing against the Law of Nature by command of the Author, if he be obliged by former Covenant to obey him, not he, but the Author breaketh the Law of Nature,”²¹ for the action is not his but the author’s (2012a: XVI, 246).

What moral difference does ownership or “nominal” responsibility for corporate action make, given it does not imply that every shareholder is blameworthy? Lawford-Smith (2019: 96) distinguishes “culpability” from responsibility that applies to citizens when they are “the best candidates for acting to repair the damage done by states.” Likewise, Pasternak (2021: 17) defines the latter type of (non-culpable) responsibility as “forward-looking remedial duty,” which explains why new shareholders may be responsible for redressing past corporate crimes for which they are blameless. She suggests that group liability, where a corporation or state is responsible for the wrong, is necessarily distributive—that is, if the corporation catches a cold, its shareholders end up sneezing (2021: 28). For Lawford-Smith, “passing on costs to members is part of what it takes to hold collectives responsible, because members … constitute the collective” (2019: 99).

Why is it morally defensible to hold citizens (or shareholders) responsible for past crimes of their state (or corporation) which they did not commit? Although Hobbes does not address this question directly, Lawford-Smith (2019) and Collins (2023) answer that it is justified under certain conditions, none of which imply individual culpability. These include association “between the citizen and the state of the right kind to generate obligations” (2019: 129), which means that citizens are best placed to remedy the wrongful actions of their states, assuming they have not been coerced into associating (2019: 127). And if citizens benefit from wrongs perpetrated by their states, then “the fact of benefiting may justify citizens’ responsibilities to disgorge the benefits of their state’s injustices,” which “might be a means of making reparations for those injustices” (2019: 130). Collins (2023: 160–61) explains that if we cannot always track organisational responsibility to blameworthy members, we must have a moral “shortfall” unless we can find these “sources of liability” (associating, benefiting, etc.) that sit alongside the corporation’s own wrongdoing.

Under these conditions, what responsibilities or “remedial duties” should be met from shareholders’ “common stock” (Hobbes, 2012a: XXII, 364) or the corporation’s retained earnings? In the analogous case of the state, citizens’ remedial duties (met partly through taxes) are argued to include “compensation for economically assessable wrongful harms,” providing rehabilitation for the victims, engaging in “reparative measures … e.g., incurring punishment for the wrongdoings and investing in commemorative projects,” and finally “the obligations of nonrepetition, including implementing legal and institutional reforms” (Pasternak, 2021: 27–28). An illustration of how a corporation’s shareholders collectively face similar duties, discharged not from personal assets but from shareholder equity, is BP’s financial response to the Deepwater Horizon tragedy. Between 2010 and 2016, BP paid an estimated $61.6 billion in costs related to the incident (BP, 2016: 16). These included $500 million to the Gulf of Mexico Research Initiative on what Pasternak terms “nonrepetition”: that is, “the goal of improving society’s ability to … mitigate the impacts of petroleum pollution and related stressors on marine and coastal ecosystems” (Gulf of Mexico Research Initiative, n.d.); $8.8 billion on rehabilitation by enabling state and federal agencies to “develop a science-based, comprehensive restoration plan for the Gulf of Mexico” (National Oceanic and Atmospheric Association, 2017); and $14.8 billion in compensation to private claimants (National Oceanic and Atmospheric Association, 2017). These costs were required of BP without implying that any shareholder was morally culpable for the incident, and without distinguishing those who were shareholders when the accident occurred from those who joined the company later.

Another aspect of the moral significance of shareholders’ nominal responsibility is that directors speak for the corporation only if they hold authority from the shareholders together in assembly. Although the represented corporation cannot speak for itself, the representative assembly can speak with a single voice through majority voting (Fleming, 2021: 21–22). As Hobbes puts it: “An Actor may be Many men made One by Plurality of Voyces … the voyce of the greater number, must be considered as the voyce of them all” (2012a: XVI, 250). The many voices of the shareholder assembly converge into one because shareholders agree, for most purposes, to let the will of the majority stand for them all. In Hobbes’s time, this probably meant a show of hands among shareholders (Scott, 1912: 162–63), whereas in the modern corporation votes are typically proportional to the number of shares held. However the majority is calculated, if the assembly has a method of “aggregating judgements” (List & Pettit, 2011: 43) it may constitute itself as one representative person with a voice of its own.²² Shareholder assemblies, therefore, are active corporate persons (Pettit, 2009: 79–81; Fleming, 2021: 21–22) that can speak on the corporation’s behalf by electing directors and holding them to account. Only decisions for which every shareholder is, by virtue of their membership of the assembly, “nominally responsible” are corporate decisions that bind the directors.

For this reason, shareholders ought to take responsibility for shaping the intentions of directors to deter the latter from the wrong actions, which means participating in the assembly to ensure (if possible) that what is done in their name has their consent. As List and Pettit (2011: 164) put it: “Even if there is little those individuals could have done to stop the group behaving as it did, they may inherit a share in the group’s responsibility to the extent that they continue to be members and explicitly or implicitly endorse the group’s actions.” However, if the representative assembly or the board of directors acts contrary to the authority they have received,²³ then shareholders do not hold responsibility in any sense for their representative’s actions (Hobbes, 2012a: XVI, 250).²⁴

How can shareholders shape the intentions of directors? Besides the ultimate sanction of removing a director, shareholder assemblies can (to use examples from UK law) force directors to convene a meeting of the shareholders and answer questions put to them.²⁵ Shareholders can also participate in the decisions of their annual meetings by sponsoring and voting on proposals that bind directors to implement environmental, social, and governance (ESG) policies, International Labour Organization standards on rights at work, or Global Reporting Initiative standards on environmental impact.²⁶ The UK Corporate Governance Code recommends that “in addition to formal general meetings, the chair should seek regular engagement with major shareholders in order to understand their views on governance and performance…” (Financial Reporting Council, 2018: 4). However, it is questionable whether directors’ accountability to shareholders can be discharged outside formal meetings of the assembly, as individual shareholders cannot speak on behalf of the whole corporation.

To what end, or for what purpose, should directors be held accountable by shareholder assemblies? As noted previously, a Hobbesian solution is suggested by Claassen (2021: 123–28), who argues that “even commercially operating businesses contribute to the public purpose of the maximization of social wealth” (2021: 126). This theory “criticizes corporations as a matter of moral judgement when, in situations of market failure, they put their private gain ahead of social wealth maximization” (2021: 127). Although he describes the corporation’s purpose as a mandate from the state and not from shareholders, if his theory is indeed “a cousin of the market failure theory of business ethics” (2021: 128) then it can be aligned with shareholder primacy. Heath (2006, 2007), for example, distinguishes the “intrafirm” context in which managers owe fiduciary obligations to shareholders (2006: 537–40; 2007: 367) from the “extrafirm” context in which they are constrained by a duty not to exploit market failures, in order to protect the social welfare benefits that arise from competition (2006: 550). Heath (2006: 549) writes that “first and foremost” among managers’ obligations when dealing with relationships inside the firm is their fiduciary duty “as the agents of shareholders.” Indeed, in the intrafirm context, “the market failures approach to business ethics follows the shareholder-focused view quite closely” (2006: 549).

The view that shareholders are inside and other stakeholders outside the corporation aligns with Hobbes’s argument that investors form themselves into a corporation when they authorise a representative. A corporation is the union of the otherwise independent voices of its shareholders. For Heath, however, managers’ extrafirm obligations require them to balance the interests of shareholders against those of external parties: “the question becomes how far one should go, as a manager, in advancing the interests of the principal, and when one should start showing more concern for others…” (2006: 542). Hobbes’s theory of authorisation, by contrast, holds shareholders themselves nominally responsible for the corporation’s actions. And this implies a duty to shape managers’ intentions by holding them accountable for how they have represented the corporation to external parties.

When shareholders exercise their moral judgement regarding the actions of directors, they are faced not merely with “deontic constraints” on market transactions (Heath, 2007: 367–68). An emerging literature recognises that shareholder primacy is compatible with positive duties of beneficence to other stakeholders, enacted by managers on shareholders’ behalf (e.g., Mansell, 2013b; Mejia, 2019, 2021). Mejia argues that “one need not go beyond the paradigm of shareholder primacy to ground a good deal of the duties of beneficence that scholars in the field expect managers to fulfil in their corporate roles” (2021: 445). And this obligation is not in tension with duties to shareholders: “it arises from the fact that [the manager] is acting on their behalf” (2021: 445). Hobbes would agree that when “a manager acts on behalf of a shareholder, she is guided by the fundamental principle … he who acts through another acts himself” (Mejia, 2021: 424). For example, if a corporation has a “duty of rescue” to victims of an earthquake, the manager is required to fulfil the duty “because she is acting on behalf of (moral) shareholders” (2021: 435). He argues that shareholders, as principals, “bear the ultimate responsibility for any moral wrongdoing committed on their behalf. As such, shareholders should be more vocal with managers about the importance of conforming to a more encompassing set of moral restrictions” (Mejia, 2019: 537).

8. CONCLUSION

Hobbes’s theory of the corporation makes two principal contributions to the ongoing debate on shareholder primacy. Critics have shown why the interests of shareholders cannot be preferred to those of other stakeholders on the grounds of individual property ownership. The corporation, they rightly argue, is itself a separate property-owning person (Deakin, 2012; Ireland, 1999; Robé, 2012; Stout, 2012). However, the questions of how individuals can authorise directors to speak for an entity that is distinct from them both, and of how responsibility for corporate action should be attributed, are left unanswered. Hobbes theorises the process by which the separate person of the corporation is created by the shareholders’ decision to incorporate. Hobbes explains, first, the constitution of the corporation by its shareholders and, second, the responsibility of the latter for actions taken on the corporation’s behalf. He therefore shows that the conclusion drawn by critics of shareholder primacy, that because managers are agents of the corporation they cannot be agents of the shareholders, does not follow. Because shareholders are incorporated in the shareholder assembly, they are individually responsible for participating in the assembly to hold directors accountable for actions taken in their name. The moral obligations that bind principals in their agents’ dealings with external parties, for example, duties of beneficence, apply directly to shareholders (Mansell, 2013b; Mejia, 2019, 2021).

By explaining the role of the state in the authorisation process, Hobbes also contributes to the concession theory of the modern corporation. Claassen argues convincingly that Hobbes does not offer a “customary interpretation” (2021: 109) in which corporations are created exclusively for public purposes and private initiative has no role. But nor does it follow that the state is the source of managerial authority. If it were the state, and not the shareholders, who authorised directors to represent the corporation, the sovereign would then own and be responsible for every corporate decision. Hobbes’s theory avoids this implication. The sovereign allows a part of the citizenry to form a corporation by choosing a representative whose actions each member of the body endorses, within the limits of the sovereign’s charter (or corporate law). In this theory of conditional representation it is the members (or shareholders), and not the sovereign or the people as a whole, who own and bear nominal responsibility for corporate actions.

This article also offers a new social contract theory of the corporation. Departing from the common perception that such an approach implies a stakeholder theory of the firm (e.g., Boatright, 1994; Donaldson & Dunfee, 1999; Fia & Sacconi, 2019; Freeman, 1994; Freeman & Evan, 1990; Mansell, 2013a; Sacconi, 2006), I have shown that Hobbes’s theory of the corporation, which rarely features in this literature and yet is arguably the model for the social contact tradition, supports shareholder primacy.

Hobbes’s theory therefore shows that several facts of corporate existence assumed to be incompatible with shareholder primacy can be reconciled with it. In particular, the separate legal personhood of the corporation does not imply that directors are not accountable to shareholders, or that the corporate interest is separate from shareholders’ interests. Although concession theorists are correct to insist upon the necessity of the state to the creation of corporations, a top-down theory of managerial authority does not follow. Finally, a corporation’s purpose can reflect a voluntary agreement (or social contract) among its members without implying that all stakeholders have equal status (Freeman, 1994; Sacconi, 2006). In conclusion, Hobbes demonstrates the intrinsic connection between directors’ authority and their accountability to shareholders, and why corporate purpose therefore depends fundamentally upon shareholder responsibility.