1. Introduction
What makes a shareholder a ‘member’ of a corporation, and what are the responsibilities that membership entails? Should shareholders even be considered members at all? And if so, might other stakeholder groups (e.g. employees) also have a legitimate claim to membership? If managers receive their authority from the corporation's members, then the answers to these questions shape the purposes that managers can legitimately pursue on the corporation's behalf. The extent to which a corporation's purpose can encompass the interests of non-shareholding stakeholders (e.g. employees) would seem to depend on whether such groups should be considered corporate members, and therefore on the criteria for attributing membership status.
Theories of corporate purpose in the business ethics and corporate governance literatures frequently entail the rejection of ‘shareholder primacy’ – the view that managers are accountable exclusively for the pursuit of shareholder interests, and (implicitly) that shareholders alone are ‘members’ (Boatright, Reference Boatright2006; Fama and Jensen, Reference Fama and Jensen1983; Friedman, Reference Friedman1970; Jensen and Meckling, Reference Jensen and Meckling1976; Sternberg, Reference Sternberg2000; Sundaram and Inkpen, Reference Sundaram and Inkpen2004). It is typically argued that this view misunderstands the relationship between a corporation and its shareholders. For example, where Friedman (Reference Friedman1970) claimed that ‘a corporate executive is an employee of the owners of the business’ and that ‘the executive is an agent serving the interests of his principal [the stockholders]’, it is objected that corporate property is owned by the corporation itself rather than by shareholders, and that the latter cannot be a ‘principal’ authorising an ‘agent’. These criticisms have been developed extensively in both the business ethics (Boatright, Reference Boatright1994; Kaler, Reference Kaler2006; Phillips, Reference Phillips2003; Sison and Fontrodona, Reference Sison and Fontrodona2013) and progressive corporate law literatures (Blair and Stout, Reference Blair and Stout1999; Chassagnon and Hollandts, Reference Chassagnon and Hollandts2014; Deakin, Reference Deakin2012; Ireland, Reference Ireland1999, Reference Ireland2003; Robé, Reference Robé2011; Stout, Reference Stout2012).
We agree that shareholder primacy can be challenged along these lines. However, we argue that this focus on ownership is inadequate because it targets only the claims of individual shareholders.Footnote 1 Inspired by investigation of medieval sources and practices, we argue instead that a corporation can be identified with a body of individuals that its management is authorised to serve. A more persuasive challenge to shareholder primacy is therefore to ask whether the modern business corporation is constituted, at least in part, by non-shareholding members. For as Ciepley (Reference Ciepley2015: 8–9) observes, besides arguments from property ownership, a further ‘leg’ propping up the ‘stool’ of shareholder primacy is that shareholders alone are ‘members’, ‘and since it is their club, it should be run in their interest’.
For example, the UK Companies Act 2006 states that a director's duty is to ‘promote the success of the company for the benefit of its members as a whole’ (s. 172(1)). In a company with share capital the ‘members’ are the shareholders (s. 8(1b)), and the corporation is identified with the members (s. 16(2)). And Delaware corporation law appears to view ‘member’ and ‘stockholder’ as equivalent terms.Footnote 2 On this reasoning, a corporation is identified with a body of members that authorises a board of directors to serve its interests: the corporation's purpose therefore stems from the interests of its membership. One might expect to find a rationale for distinguishing members from non-members in arguments for both shareholder primacy and more participative (stakeholder oriented) corporate governance. However, on either side of this debate, the issue has received little attention.
Arguments concerning corporate membership did emerge, however, in the context of a range of medieval corporations: universities, craft and merchant guilds, villages and towns, collegiate churches and the Roman Catholic Church. Despite obvious differences between any of these corporate entities and the modern business corporation (Scruton, Reference Scruton1989: 241–243), we find that concepts relevant to the question of corporate purpose today can also be found in medieval thought. The relationship between a corporation and its members was explored and contested between the 12th and 15th centuries; in the process, connections were made between concepts that are central to the contemporary debate.
Critics of shareholder primacy typically appeal to the separateness of the corporation from its individual participants – a view that follows logically from assumptions about the meaning of ownership and the source of managerial authority. In the medieval sources, we find an alternative possibility of how these concepts fit together, which in turn provides a clearer understanding of how members stand in relation to the corporation. These sources enable us to argue that authority and ownership reside in the members as a ‘plurality in succession’, in which individuals participate through their contribution to a distinct collaborative activity (or ‘common good’). By rethinking the relationship between a corporation and the individuals it comprises, which is our primary aim, we develop a framework in which the normative question of non-shareholding membership can be addressed.
Our task in this paper is conceptual: we do not undertake a comparative evaluation of medieval and modern corporations, nor do we argue that values and practices specific to the medieval period should necessarily be adopted in modern business. Our concern is to find an alternative framing of the concepts that bear on the relationship between a corporation and its members. Although our framing of these concepts differs from recent debate on corporate purpose, it is in fact reinforced by recent work on the common good of the firm (Sison and Fontrodona, Reference Sison and Fontrodona2012, Reference Sison and Fontrodona2013) and on the firm as a real entity (Adelstein, Reference Adelstein2010; Gindis, Reference Gindis2009; Vanberg, Reference Vanberg1992), as we explore in the final section.
2. The critique of shareholder primacy
In criticisms of shareholder primacy, it is emphasised that the corporation is ‘completely separate’ from its shareholders. Phillips (Reference Phillips2003: 19) suggests that shareholder ‘ownership’ be tested ‘through a thought experiment: Who would own the corporation if it bought back its own stock?’ He argues: ‘The corporation is not coextensive with the shareholders … The corporation is meaningfully distinct’ (Reference Phillips2003: 19–20). Ireland (Reference Ireland1999: 41) traces the history of this separation, and observes that from 1860 in the UK shares were ‘forms of property independent of the assets of the company’. He argues that a ‘vital legal space thus emerged between companies – owners of assets – and shareholders – owners of shares’ (ibid.), and that with the 1862 Companies Act, ‘people no longer “formed themselves” into incorporated companies, they “formed” companies … made by them but not of them’ (Reference Ireland1999: 42, original emphasis).
In the context of contemporary corporate law, Robé (Reference Robé2011: 24) argues that without a clear separation between assets and shares, ‘the worlds of structured finance, private equity, leveraged buy-out, securitization, etc. would not exist … If the shareholders were owners of anything else other than shares, the whole edifice would crumble’. Similarly, Deakin (Reference Deakin2012: 356) holds that the assets of the firm ‘vest in the separate legal person of the “corporation”’. And in her popular attack on the ‘shareholder value myth’, Stout (Reference Stout2012: 37) imagines: ‘Nonlawyers may find it hard to wrap their heads around the notion that an intangible and abstract institution like a corporation is a “juridical person”’. Nonetheless: ‘Corporations are independent legal entities that own themselves, just as human beings own themselves’ (ibid., original emphasis).Footnote 3
For these critics, the corporation's autonomy as a separate entity means that shareholders cannot be ‘principals’ who authorise managers to act on their behalf. In law, the responsibilities of managers are to the corporation itself: ‘managers can't be the agents of individuals who do not own the assets they are operating – who therefore cannot be principals in connection with these assets’ (Robé, Reference Robé2011: 32, original emphasis). Deakin contends that it was a ‘fiction’ of the founders of agency theory to view shareholders as principals: ‘company law views directors as the agents of the company, and not of the shareholders’ (Reference Deakin2012: 360). The logic of the argument against shareholder ownership applies also to the ‘members’, whether or not they include other stakeholders: there are no ‘owners’ of a corporation, and therefore none who can authorise management to act in their name. The corporation and its members (if it has any) are, as legal concepts, altogether distinct.
On this view, managers receive their authority directly from the corporation itself. The role of the members as an authorising body would therefore be misconceived, and it would be superfluous to argue against the assumption that shareholders alone are members. There are a few exceptions: Ireland (Reference Ireland1999: 47) notes that ‘the basis for the privileged status of the shareholder as a “member” is, indeed, “obscure”’, because in economic reality there is little difference between shareholders and debenture holders. He draws upon Goyder (Reference Goyder1961) to suggest the extension of membership to employees, consumers and the community (2003: 505). Similarly, for theorists of the ‘common good of the firm’ (Sison and Fontrodona, Reference Sison and Fontrodona2012, Reference Sison and Fontrodona2013), a corporate ‘member’ is one who participates in ‘collaborative work through which goods and services are produced’ (Reference Sison and Fontrodona2013: 612).
By contrast, Ciepley (Reference Ciepley2017) argues that business corporations have no members, because they lack the essential features of true ‘member corporations’ (e.g. medieval universities, towns and guilds). For Ciepley, the source of a business corporation's authority cannot be a body of members; instead, corporations are ‘“franchise governments” – their form and powers are delegated by the state, yet they are run on private initiative’ (Reference Ciepley2013: 140). We respond to his arguments in the final part of section 3 and then, in section 4, we defend a theory of membership that is applicable to the business corporation of today.
At this stage, an objection might be raised. Not only did medieval corporations diverge in the purposes for which they were constituted – individual guilds, churches, towns and universities had distinct objectives, of course – but there are manifest differences between any corporate entity of this period and a modern business corporation. Scruton (Reference Scruton1989), for example, writes of the modern firm as a ‘partnership for gain’ and ‘a means to an end … dissoluble at will’ (241–242). By contrast, other types of corporate association (including the Roman Catholic Church) are characterised by internal purposes: i.e. ‘you do not join a church for the sake of a salvation that could be achieved in some other way’ (ibid., 244).Footnote 4
And yet we can speak of all these associations as corporations because of their capacity to attract legal relations (Gindis, Reference Gindis2016) in their own name. We can ask, of any corporation so defined, whose interests its managers are authorised to serve. If corporate ownership and authority reside in a body of members, then that is where one might find the answer. Critics of shareholder primacy have, of course, denied that shareholders have this status and insisted that ownership lies elsewhere. Yet this very objection illustrates that the relationship between a corporation and its members is central to the question of corporate purpose. The importance of the medieval arguments, in the present context, is that they offer a different perspective on precisely this relationship.
3. Medieval corporation theory
There are two main influences on the development of the medieval understanding of corporations. The major theological source is the Vulgate Bible (Corinthians 1 12:12–27), which invoked the body (corpus) and members (membra) of Christ as a metaphor for the ‘mystical body’ of the Church. Saint Paul states: ‘For as the body is one and hath many members, and all the members of the body, whereas they are many, yet are one body, so also is Christ’ (ibid., 12:12) and ‘you [plural] are the body of Christ …’ (ibid., 12:27). While this metaphor informed the ecclesiology of the church as a corporate entity, it is difficult to find a philosophical source that influenced the medieval conception of corporations generally. Aristotle had said helpfully in Politics (I.I. 1252a18) that the ‘compound’ of a city should be analysed prior to its elements, but his text was not translated into Latin until 1260 (Black, Reference Black1992: 2–3), by which time corporation theory had developed primarily on the basis of legal sources. The canonists (i.e. experts on Church law), for example, were much less interested in the metaphysics of corporations than in their constitutional structure: ‘the actual distribution of authority between head and members’ (Tierney, Reference Tierney1955: 102). The most significant impetus for the analysis of such problems was the rediscovery of Roman law in the late 11th (or early 12th) century, particularly the Digest of Justinian of AD 533.
The Digest provided medieval lawyers with the concept of the universitas – from unum, ‘one’ and vertere, ‘to turn’ (literally ‘a number of individuals turned into one’, Brett, Reference Brett and Brett2005: l) – which we now translate as ‘corporation’ (Ryan, Reference Ryan and Lagerlund2011: 236). According to the Digest (3.4), corporations would be allowed to form ‘only in very few kinds of cases’, but exceptions included miners of gold, silver and salt, and guilds of bakers and ship-owners, who were permitted to have a corpus; i.e. a separate body in law with its own assets and liabilities (Digest 3.4.7; see also Ryan, Reference Ryan and Lagerlund2011: 236). Although the Roman term universitas originally meant a ‘lesser association’, it ‘had by the high Middle Ages become a general term’ for states, cities, monasteries, guilds and other groups (Black, Reference Black1992: 14, original emphasis).
Given the breadth of the era we draw upon (approximately the 12th century to the 15th), we use current scholarship to identify recurrent themes in the thought of the period. Although there were many legal systems operating in medieval Europe, there were common ‘languages’ (Pocock, Reference Pocock2009) in which the features of the corporate form were elucidated. Roman law is the common juridical language for legal analyses of the corporation – even where its concepts were mediated through canon or common law. And the Vulgate Bible provided a theological language in which the relation between parts and wholes could be expressed through the metaphor of bodies (corpora) and their members. These juridical and theological languages could also interact: for example, in the idea of the corporation as a persona mystica, as we shall see in the following section.
The corporation as a legal person
Medieval corporations were legal ‘persons’ that could own assets, incur debt and enter contracts independent of their individual members. As the Digest (3.4.7) puts it, ‘What is owed to the corporation is not owed to the individual members, and what the corporation owes the individual members do not owe’. In line with a standard criticism of shareholder primacy today, membership in a medieval corporation could not have been derived from individual ownership of corporate assets. However, to understand how medieval thinkers did understand membership, we first have to see how they explained the corporation's legal status.
With regard to the medieval Church, the corpus mysticum or ‘mystical body’ of Christ was more than a name for the ‘totality of Christian society’: it acquired ‘legal connotations’ and became identified as a ‘juristic person’ (Kantorowicz, Reference Kantorowicz1957: 209). Where Aquinas had stated that ‘the head and members are as one mystic person’ (Reference Aquinas1981 III: 48.2), his choice of the word ‘person’ (persona) lent theological weight to the prevailing juristic view, following Innocent IV's decretal of 1246, that all corporations were personae fictae (‘fictive persons’) (Kantorowicz, Reference Kantorowicz1957: 202). ‘Fictive’ status in law meant that the medieval universitas, like the modern corporation, had legal capacity distinct from its transient ‘members’. The medieval corporation therefore ‘remained the same legal entity even though the persons of the members changed’ (Tierney, Reference Tierney1982: 19).
This terminology applied not only to the Church but ‘to every size and rank of universitas’ (Kantorowicz, Reference Kantorowicz1957: 209–210). The German craft guild, for example, ‘could will and act as an entity … It could enter into contracts and arrangements of all kinds’ (Gierke, Reference Gierke, Black and Fischer1990: 52). The assets of the guild ‘belonged to the fellowship as such … individuals held no private rights but were only authorised to use and profit from such assets as gild members’ (Reference Gierke, Black and Fischer1990: 56). The guild was therefore distinct from the medieval societas (or ‘partnership’) whose assets remained the property of its members (Aquinas, Reference Aquinas and Dyson2002: 228). Medieval towns were also regarded as ‘having a moral right … to juridical personality, to the ownership of collective property’ (Black, Reference Black2003: 45), though it was not always to the town's advantage. In 13th-century Tuscany, a court was persuaded ‘to put an entire subject town under ban and to compensate the plaintiff from that community's common property, even though the town was now divided between factions …’ (Ryan, Reference Ryan and Lagerlund2011: 236). The town, as a universitas, remained liable (ibid.). Legal capacity was also attributed to the city state or kingdom: 14th-century cities of northern Italy, for example, had ‘legal existence and capacity distinct from those of their members’ (Canning, Reference Canning and Burns1988: 474).
This was the legal orthodoxy, and medieval writers were not especially interested in developing a philosophical account of the corporate form. The canonists were more concerned with ‘the flood of litigation … concerning the authority of ecclesiastical corporations and the rights of their various members’ (Tierney, Reference Tierney1955: 104). However, in addressing the distribution of authority (ibid., 102), the simple fact of the corporation's legal status did not offer a solution. Similarly, the fact that individual shareholders do not own corporate assets does not suggest on whose authority, or on behalf of which stakeholders, a modern corporation should be run. Notwithstanding the corporation's separate legal status, medieval writers approached the problem by asking where corporate ownership (dominium) ultimately resides. By explaining their notion of collective ownership, we can now show that the question of whether a modern corporation should be run in the interests of non-shareholding stakeholders depends on whether the latter are parts of a greater corporate whole (the universitas), and not merely on the specific investments they may happen to make in a corporation (e.g. as creditors or employees).
Collective ownership
In the case of collegiate churches, it was recognised that property ownership resided, not with the bishop, but with ‘the church itself’ – ‘understood either as the clergy of the diocese or even the congregation of believers’ (Ryan, Reference Ryan and Lagerlund2011: 237, Tierney, Reference Tierney1955: 137). In his attempt to refute Pope Boniface VIII's claim to temporal power, John of Paris argued that ecclesiastical property ‘has been given to communities, not to individual persons … it is the community which has the right of lordship over its property’ (c. [1302] 1971: 96). Furthermore, ‘the founders of churches intended to transfer lordship and proprietary right of the properties they were giving principally and directly to the community of the individual church’ (ibid., 99). However, ‘collective ownership’ of corporate assets presents us with a paradox: ownership resides in a body of individuals, while the corporation itself owns the assets as a separate person in law.
A solution can be found in the emergence of a theory that explained the relationship between a corporation and its members. Commentators on Roman law, such as Baldus, saw the city as ‘an abstract unitary entity perceptible only by the intellect’ (Canning, Reference Canning and Burns1988: 473). It was constituted by its physical members and able to act only through them, but it was also a universitas and hence ‘immortal’ and ‘quite distinct from its human components’ (ibid., 474). Innocent IV's definition of the universitas as a ‘fictitious person’ made it possible to treat ‘a plurality of individuals juristically as one person’, while distinguishing that person ‘from every natural person endowed with body and soul’ (Kantorowicz, Reference Kantorowicz1957: 306). For example, Aquinas wrote that the state is both ‘composed of many persons’ and established ‘to endure for all time as the citizens succeed one another’ (Reference Aquinas and Dyson2002: 138).Footnote 5
The explanation for this ‘essential feature’ of all corporations ‘is not that they are ‘a plurality of persons collected in one body’ at the present moment, but that they are that “plurality” in succession’ (Kantorowicz, Reference Kantorowicz1957: 310).Footnote 6 By illustration, the ‘individual and material community of Bologna composed of mutable citizens and perishable buildings’ was distinct from the ‘personified community’ of Bologna, ‘an entity which was both immaterial and invariable’ (ibid., 303). When medieval thinkers held that ‘ownership’ resided in a plurality, they conceived of the members collectively in succession – as an entity abstracted from, and surviving, each individual member. From this perspective, the corporation is identical to its members considered as one through time, which is exactly what the etymology of universitas (‘to turn into one’) implies.
By comparison, it is assumed in the contemporary debate that ‘shareholder primacy’ is undermined when shareholder ‘ownership’ is refuted. Whether or not this is an effective criticism of Friedman (Reference Friedman1970), conceptually it works only for the property of individual shareholders. For medieval writers, the members owned corporate property by forming a distinct, collective entity: the universitas itself. A solution can therefore be found to the contradiction apparent in the owning of assets by a corporation that are also owned by its members.
In the modern debate, the membership status of a given stakeholder would therefore refer to their membership in a collective entity identified as the corporation. However, this would be relevant to the question of corporate purpose only if the corporation's managers have a duty to serve the members’ interests. In the medieval arguments, neither the corporation's legal status, nor the idea of collective ownership, unambiguously clarified why this should be so. For example, while ownership of church property ‘rested in some sense with the local community’ (Tierney, Reference Tierney1955: 140), Innocent IV placed it in the aggregate of all Christians. This implied, conveniently for the Pope, that all church property was at his disposal, for Innocent also held that ‘all the jurisdiction of a corporation [is] concentrated in its head’ (ibid., 141).
What mattered was the source of authority for a corporation's actions. If the right to govern the corporation's affairs and adjudicate the implementation of its rules was bestowed by the members, then the corporation's governing body was believed to be accountable for acting in the members’ interests. Depending on the context, the members might then claim a right to participate in the governance of the corporation. But this was not to be taken for granted: the source of corporate authority was a fiercely contested issue, particularly in the church, from the 13th to the 15th centuries. We now review the competing perspectives that emerged in this controversy, and find they provide further reason to rethink the modern debate.
The source(s) of corporate authority
Authority in the Church
For medieval writers, the problem was to establish whether authority is granted from ‘below’ by the members or from ‘above’ by a superior. Cardinal Hostiensis (a pupil of Innocent IV) argued that the bishop acquires jurisdiction from his election by the members of his corporation, i.e. from his cathedral chapter (Ryan, Reference Ryan and Lagerlund2011: 237). Hostiensis’ position corresponded with Accursius’ (d.1263) standard gloss to the Roman law: a judge receives his jurisdiction from the corporation that elects him, though the approval of a ‘superior officer’ is also required (ibid., 236). Hostiensis thus strengthened the view that had emerged through the first half of the 13th century, and afterwards was seldom challenged, that ‘authority in a corporation was not concentrated in the head alone but resided in all the members’ (Tierney, Reference Tierney1955: 117).
In 1313, the canons of Lincoln drew upon Hostiensis to declare that the chapter is the ‘source’ (fons) of the bishop's authority (Tierney, Reference Tierney1955: 127), which would make the bishop subservient to his chapter. This move was deemed unacceptable by the Church – authority could be transferred to the bishop through election of the chapter, without that authority having originated in the members (ibid.). But how can members bestow authority that originates from another source? In Roman law, corporate authority resides with the members and is delegated to a spokesperson. In the case of the Roman private corporation, the members could modify or revoke the agent's powers (Tierney, Reference Tierney1982: 26–27). However, in canon law, the cathedral chapter could elect the bishop but ‘did not simply delegate its own authority … the bishop entered upon an office with its own inherent dignity, rights and powers’, derived from ‘the ancient constitution of the church’ and ultimately from God (ibid.). On the other hand, the canons – who held land independently of the bishop – could elect an ‘agent’ where their interests were at stake (Reference Tierney1982: 27). For practical purposes, this meant ‘authority was shared between bishop and chapter … A major act involving the welfare of the whole Church required the assent of both parties’ (ibid.). The bishop could not, for example, alienate church property without the chapter's agreement (Tierney, Reference Tierney1955: 120–124).
Authority in guilds
Authority in the guild was less fiercely contested than in the Church, but even here the source of authority is ambiguous. Gierke is clear that medieval fellowships (including craft guilds) were founded on the ‘free will of the members’ (Reference Gierke, Black and Fischer1990: 19), even though the exercise of trade rights required ‘the authorisation of the [town] council … or civic overlord’ (Reference Gierke, Black and Fischer1990: 50). Black (Reference Black2003: 18) adds that guilds ‘needed recognition as “colleges” or corporate legal bodies … in order to establish their position as traders’ or producers’ monopolies’.Footnote 7 By agreement with the town authorities, the ‘guild was allotted a prescribed role in the city's economic life, often designated specifically as a duty or office (officium)’ (ibid.).
Did the medieval guild therefore owe its existence, and powers of self-government, to the will of the town? A parallel suggestion is found in ‘concession’ or ‘state grant’ theories of the modern business corporation (e.g. Ciepley, Reference Ciepley2013). Medieval jurists took a similar view of guilds, but they did not generally believe them to be created by town governments. Although the trade rights enjoyed by each guild were granted by the town, the Roman law itself (not a town council) was thought to authorise the guild's corporate existence. The jurists argued that the list of guilds (collegia) explicitly permitted by the Digest (3.4) – e.g. bakers, shipowners and salt miners – was merely indicative, not exclusive (Black, Reference Black2003: 19). They held that craft guilds generally were legalised by the Digest without the need for further approval (ibid.). For example, the guilds of Magdeburg were granted to exist according to ‘liberty’ and their ‘integral right’ (ibid., 18). Even Innocent IV held that craft guilds (unlike cities and churches) required no special privilege from a superior (ibid., 20) – Roman law was said to bestow direct legitimacy on a guild's corporate existence (ibid.).Footnote 8
Authority in universities
Besides Roman law, a corporation might be authorised by custom or common law (Post, Reference Post1934: 423). The universities of Bologna, Paris, Padua and Oxford effectively enjoyed corporate status decades before specific charters of foundation were granted. Jurists of the 14th century classified them as ‘customary studia’, even after the charters had been received (Leff, Reference Leff1968: 19). However, in a parallel case to that of guilds and towns, the earliest universities had to look to a ‘higher’ authority to protect their privileges (Leff, Reference Leff1968: 24), and might appeal to one authority for protection against another.Footnote 9 For example, while the chancellor of the University of Paris claimed the right to ‘fine, excommunicate and jail’ students and teachers, ‘in the conflict that naturally developed, the teachers’ staunchest allies were popes Innocent III, Honorius III, and Gregory IX’ (Bernstein, Reference Bernstein1978: 291).
It follows that medieval university, church and guild corporations could not look solely to their members for the source of their corporate privileges. Roman law, common law or ‘custom’, a pope, town council or king could legitimise a corporation's legal status, depending on the circumstances.Footnote 10 Corporations were also dependent on external authority for the exercise of some of their rights: for example, the right to elect a superior over themselves. However, rather than excluding the members as a source of authority, this dependence upon superiors was justifiable because lesser corporations and their members were constituent parts of larger corporations. The election of a bishop by his chapter had to be ratified externally because it affected the welfare of the Church as a whole. Likewise, a guild was not only a body in its own right, but also part of the town, representing the broader interests of the citizenry (Gierke, Reference Gierke, Black and Fischer1990: 49). It was therefore an intrinsic component of a common body that extended beyond its own immediate interests.Footnote 11 In the case of the University of Paris, only rarely did the interventions of king and pope come into conflict, because whereas ‘the pope sought to oversee the university from the point of view of the ecclesiastical hierarchy … the king regulated its place within the city and, on occasion, the kingdom’ (Leff, Reference Leff1968: 28–29) These sources of authority arguably reflect the different spheres of human well-being, beyond the good of its own members, to which the university contributed.
Reconsidering the critique of shareholder primacy
The conceptual frameworks we have explored complicate the arguments around shareholder primacy. Medieval and modern corporations share the generic feature of legal personhood distinct from their individual members – the very feature invariably emphasised today as a rebuttal to Friedman (Reference Friedman1970). However, it was not considered by medieval writers to imply a conceptual separation of corporation and members. Instead, the corporation was conceived as a legal person, owning assets that were the collective property of its members. The ‘owner’ or dominus, the corporation itself, could be identified with the members as a universitas.
Furthermore, in the modern context, it is assumed that if shareholders own the assets then they must authorise management (the ‘agent’) to act exclusively in their interests, with the consequence that critics have made this premise their principal target. The medieval debate suggests this assumption is unnecessary. While corporate authority can reside in part with the members collectively, giving them certain rights, it can also originate in a source outside the corporation. For example, Ullmann (1966/Reference Ullmann2010: 219) notes that no town ‘was “free” in the sense of being an independent and autonomous entity’. Paradoxically, however, towns became self-governing entities as a consequence of the royal charter that legalised their self-government. Similarly, merchant guilds acquired ‘self governance, legal rights, and predictable taxes – by purchasing from the king privileges enshrined in written charters’ (Richardson, Reference Richardson2004: 19). Authority did not flow in only one direction. Members could be ‘collective owners’ of corporate assets without unlimited authority to decide the corporation's interests. While decisions of the whole corporation should represent the common interests of its members, they were circumscribed by a broader common good, represented by a ‘higher’ authority, of which the corporation itself was a part.
We have thus far examined medieval thought on the corporation's legal status, ownership and authority, and have drawn on these insights to elucidate the conceptual relationship between a corporation and its members. However, none of the conceptions we have reviewed provides an answer to the ethical question of who should count as a member. Nor do they suggest, in the modern context, whether shareholders have an exclusive claim to membership status or even whether they are members at all. We now look at the concept of the corporation's common good and what it implies for the responsibilities of members. We argue that it is with reference to this common good that a member can be identified.
Participation in the corporate common good
That a legitimate corporation existed to pursue a common good for its members was held to be true almost by definition. In the words of Innocent IV, Roman law legitimised craft guilds because craftsmen came together to pursue a ‘praiseworthy profession’ (Black, Reference Black2003: 21). In the 14th and 15th centuries, that craftsmen ‘come together’ or ‘have much to do together’ for a ‘just cause’ recurs in arguments for the legitimacy of craft guilds (ibid.). Richardson (Reference Richardson2005: 152) finds that the goals of the guild reflected the ‘common interests’ of its members: e.g. higher prices, reputations for quality, stable wages and religious redemption. The practice of a craft was also a public office: the common good of the craft guild, at least in part, was to fulfil a duty to the community of the town (Black, Reference Black2003: 14; Gierke, Reference Gierke, Black and Fischer1990: 25; Richardson, Reference Richardson2004: 17).
These obligations implied that an individual's right to enjoy the benefits of membership entailed a duty to shoulder the requisite burdens (Gierke, Reference Gierke, Black and Fischer1990: 33; Richardson, Reference Richardson2005: 150–153). However, rights and duties did not apply equally to all members. Medieval corporations, like their modern counterparts, allowed for degrees of status within an organisational hierarchy.Footnote 12 An undisputed fact of medieval towns was a division between ‘full’ and ‘passive’ members: ‘monasteries, religious orders, Jewish communities, visitors, citizens of other towns, journeymen, servants, dependants etc., always formed a circle of denizens who did not take an active part in the government of the town’ (Gierke, Reference Gierke, Black and Fischer1990: 40). The entire population of the town (including women, children and visitors) had a passive share in the protection offered by the town (ibid., 36), but those entitled to participate directly in the town's government were a more limited circle of ‘full citizens’. ‘Full’ membership entailed active participation in governing for the common good.
Crucially, a right to enjoy the prosperity and protection of the corporation implied a commensurate duty to serve its common good (ibid., 44). Cities held to ‘the principle that a share in the city's rights must always correspond to a share in civic burdens’ (ibid., 45).Footnote 13 For example, from their service to the town, artisans of the craft guilds deduced the right to question: ‘what should be meant by “the profit, honour and gain of the town”, and the right to work … for the common good – that is to govern’ (ibid., 42). However, the vast majority of citizens had only a passive, not an active, share in the rights of the town, either because they lacked full legal personality (e.g. women, minors, apprentices and servants) or because their legal rights extended beyond the town's jurisdiction (e.g. members of chivalric or monastic orders) (ibid., 44–45). Nonetheless, these groups enjoyed the protection of the town and ‘participation in civic burdens’ was expected of them. Jewish communities, for example, were taxed in proportion to the level of protection afforded to them (ibid., 45).
Of course, an application of these principles to the modern corporation would be misplaced if the latter had no members. Ciepley (Reference Ciepley2017) argues that medieval universities, towns and guilds were archetypical ‘member corporations’ (universitates personarum): their members had one vote each through which they controlled the admission of new members and elected their own governments. He contends that the modern business corporation, as presently constituted, has none of these characteristics. From this it follows that neither shareholders, nor any other stakeholder group, have whatever moral status is attached to membership. However, Ciepley's characterisation of medieval ‘member corporations’ implies formal equality of members in decision-making (i.e. one member, one vote). On the other hand, our sources indicate varying degrees of status among the members – for example, between master craftsmen and apprentices of a guild,Footnote 14 or between ‘active’ and ‘passive’ citizens of a town. Medieval ‘member corporations’ cannot, therefore, be defined by strict equality.
What can be said generally of the medieval corporation is that its members participated in the collective pursuit of its common purpose. Although members may have viewed membership as a means to advance individual interests,Footnote 15 what defined their membership of a given corporation was their commitment to the collaborative activity undertaken by the group.Footnote 16 This activity can be described as the corporation's ‘common good’ – that is, an abstract purpose that would persist through changes of membership.
While this understanding of membership avoids the lacunae we identified both in shareholder theory (Friedman, Reference Friedman1970) and in its critiques, it leaves unexamined the key principle of the ‘common good’ – that which the members participate in. A corporation's continuous pursuit of a common purpose cannot simply refer to the aggregate of its members’ separate interests at any given time. But how are members to be identified with the same corporation as individuals come and go? What would make a fluctuating membership a stable entity over time – that is, a ‘plurality in succession’ (Kantorowicz, Reference Kantorowicz1957: 310)? We argue it is the joint pursuit of a common activity, or ‘common good’, which gives continuity of identity to successive members. We proceed, in the final section, to examine the concept of the ‘common good’ and its implications for membership in the modern business corporation.
4. Corporate membership and the common good
The concept of the ‘common good’ assumes an idea of ‘the good’ that is held in common. The concept has taken a particular shape in the business ethics literature on the ‘common good of the firm’ (e.g. Argandoña, Reference Argandoña1998; Sison and Fontrodona, Reference Sison and Fontrodona2012, Reference Sison and Fontrodona2013), based primarily on Aristotelian virtue ethics. First, we examine the arguments offered from this tradition. We then show that the ‘common good of the firm’ need not imply a substantive concept of ‘the good’ to which all members of the firm are committed. Arguments that employ functionally similar concepts, but without the teleology of Aristotle's ethics, can be found in the institutional economics literature on the concept of the firm as an entity (Adelstein, Reference Adelstein2010; Gindis, Reference Gindis2009; Vanberg, Reference Vanberg1992). In both approaches, the common good can be conceived as the collaborative activity that is specific to a given firm.
In the Aristotelian tradition, to understand what something is, we have to inquire into its reason for being or ‘final cause’. For human beings, the final cause is ‘the good’. This is because the final cause not only satisfies a person's inclinations, but also perfects the person, in her essence, not in any accidental or partial way. This final cause is called ‘common’ when it can be achieved only on the condition that all the individual members of a group achieve it in a non-excludable and non-rivalrous manner (Sison and Fontrodona, Reference Sison and Fontrodona2012: 215). Following Aristotle, flourishing (eudaimonia) is posited as the common good of the political community, and children as a common good of spouses in families. Within each group, the common good is what individuals by nature desire and what perfects them in their particular role, as citizens or as spouses and parents.
What, then, is the common good of the business corporation? In the Aristotelian tradition, the common good of the firmFootnote 17 is defined as collaborative or participatory production (Argandoña, Reference Argandoña1998; Sison and Fontrodona, Reference Sison and Fontrodona2012, Reference Sison and Fontrodona2013). It refers primarily to activities that cannot be carried out individually due to technical and time constraints and, instead, require an organized group of people (Sison and Fontrodona, Reference Sison and Fontrodona2013: 612); it therefore involves what economists refer to as a ‘team production’ situation (Alchian and Demsetz, Reference Alchian and Demsetz1972). For example, although a ship can be moved by an adequate team of sailors, that does not mean that each individual sailor alone can move the ship a proportionate distance. Implied in this example is that the members cooperate to achieve a goal for the group as a whole (i.e. moving the ship). In Albert the Great's (1206–1280) illustration, the common good of an army comprises not so much the individual powers of soldiers in provisioning or fighting, as those individual capacities united in one and the same goal – that of securing victory (Kempshall, Reference Kempshall1999: 31). Although the soldiers may join an army to further their own interests, the common good is the victory of the army per se, which might only coincide with the goods of the individual soldiers per accidens.
To speak of the group per se is not to assume a reified entity that pursues ‘organisational goals’ (Vanberg, Reference Vanberg1992) independent of its members’ judgements. As Albert the Great comments, in a definite period of time, the common good may be understood as the total aggregate of individual goods, and projected indefinitely toward the future, the common good comprehends the unending sequence of individual goods (Kempshall, Reference Kempshall1999: 30). That is, the collaborative activity representing the common good may persist through time with a potentially unending stream of different actors. An abstract purpose can therefore be attributed to a ‘plurality in succession’ that persists through changes in individual membership. The members of the business corporation or firm, therefore, are all those who participate in the collaborative productive activity that defines its specific purpose or good.
Institutional theorists of the firm have employed concepts with a similar function. For example, Adelstein (Reference Adelstein2010: 329) describes the firm as an ‘association of individuals’ bound by a multilateral ‘relational contract’. This contract is a long-term commitment where ‘the circumstances under which it will have to take place cannot be anticipated’, and parties agree ‘instead to procedures for taking decisions and resolving disputes as they arise without terminating the relationship itself’ (Reference Adelstein2010: 335). The firm is a ‘“going concern” that all sides have an interest in preserving beyond the single transaction at hand’ (ibid.). Adelstein adds that ‘the activity of the firm manifests the terms of the agreement among its participants’ (Reference Adelstein2010: 336). Membership of the ‘relational contract’ would therefore entail participation in the collaborative activity of the firm. While medieval jurisprudence conceived of the universitas as a ‘fictive’ legal person, the notion of a ‘relational contract’ points to the real collaborative activity of the succession of individuals who constitute the corporation. And if this activity persists beyond the transactions of individual members, then the universitas (or corporation) functions as the legal representation of this ‘going concern’: it enables a plurality to enter durable agreements as a single legal entity (Gindis, Reference Gindis2009: 39).
Does the existence of a ‘relational contract’ imply a common purpose? The firm has been described as a ‘truce’ or ‘treaty’ that mediates ‘divergent individual interests’ (ibid., 340). On the other hand, individuals can serve several ‘interests’ through their participation in a firm's activity, one of which can be assumed to include the preservation of the firm as a ‘going concern’. Indeed, one of the roles of the entrepreneur is to gain ‘the willing cooperation of owners, managers, and workers with the governing objectives of the enterprise’ (ibid., 339). A shared understanding of these objectives can, of course, be renegotiated through the relational contract, as Gibbons and Henderson (Reference Gibbons and Henderson2012: 1359) illustrate with the example of ‘subjective bonuses’. And as we argued in section 3.4, medieval corporations owed their legitimacy to their members’ pursuit of an ongoing collaborative activity – the corporation's common good – even where the interests of their individual members diverge.
Moreover, it has been argued that the persistence of the firm as a ‘unified body’ presupposes a common purpose: ‘unity is produced by common purpose, concerted action and organization’ (Gindis, Reference Gindis2009: 36). Gindis contends that the ‘motivational glue’ that ensures ‘adherence to common goals’ is a characteristic of the firm that ‘unifies the collective action of its members through time’ (ibid., 40). Furthermore, the firm's persistence allows for the replication of the routines, tacit knowledge and capabilities that emerge through the relational contract (Adelstein, Reference Adelstein2010: 333; Gindis, Reference Gindis2009: 39), which implies that membership involves more than mere adherence to a common purpose. The collaborative activity in which members participate is made possible by the collective knowledge, routines and capabilities the firm develops. That these are irreducible to individual capacities enables us to say that the firm remains the same entity through changes of membership (Adelstein, Reference Adelstein2010: 333; Gindis, Reference Gindis2009: 39). As we showed in section 3, the universitas appears in medieval thought as an ‘abstract’ and ‘immaterial’ entity, and Adelstein (Reference Adelstein2010: 337, original emphasis) describes the firm in similar terms as ‘an idea, a mental construction, not a tangible object’. However, these theorists’ emphasis on firm-specific capabilities and routines points again to the perpetual activity, or common good, that underpins an incorporated firm through time. It brings substance to the idea of the corporation as a ‘personified community’ with a perpetual life.
An indispensable element of the ‘relational contract’ is a constitution that, according to Vanberg's classic account, defines ‘the rules and constraints for an ongoing common enterprise’ (Vanberg, Reference Vanberg1992: 235). The firm's constitution enables decision-making for ‘organized or corporate action’ (ibid.), and is therefore a precondition for the collaborative activity in which members participate. Gindis (Reference Gindis2009: 39) observes that the constitution ‘allows the replication of behavioural patterns and collective routines’, which underpins the firm's ‘economic and legal persistence through time’. Similarly, for Adelstein (Reference Adelstein2010: 336), the relational contract ‘codifies the consensual relationships and obligations of the firm's participants’, and it is from the performance of this contract that behavioural routines and capabilities emerge. He concludes that complex routines enable ‘collective intentionality and emergent, coordinated performance’ (ibid., 340), and that firms are autonomous entities ‘existing apart from their temporal participants’ (ibid., 333). These theories explain why it is not merely the legal capacity of the corporation, but also the underlying collaborative activity (the relational contract with its emergent routines and capabilities) that has an indeterminate lifespan. They explain the organisational cohesion and durability implicit in the medieval intuition that the universitas is not simply an autonomous legal entity: it is also a plurality of individuals succeeding one another in perpetual union.
It is worth noting that Vanberg's constitutional theory leads him to a different definition of membership to the one we have defended. He argues that the ‘essence’ of membership ‘is to give up separate control over certain of one's resources, and to submit them to an organizational decision-making procedure’ (Reference Vanberg1992: 240). He explicitly denies that firms necessarily have ‘organisational goals’: beyond submission to the firm's procedures, ‘the question of whether all members of an organization have certain interests or goals in common is viewed as a purely factual matter’ (ibid., 242). According to Vanberg, membership cannot be defined with reference to a corporate ‘common good’: to do so would be to mistake ‘an empirical issue’ for a ‘definitional attribute of organizations’ (ibid., original emphasis).
However, Vanberg perceives of an ‘organizational goal’ as ‘a notion that can make sense only in an organic, collectivistic framework’ (ibid., 227). Our concept of a ‘common purpose’ is nevertheless compatible with Vanberg's individualist perspective. Of the many reasons that individuals may have for transacting with a firm (or corporation), we have argued that participation in, and shared responsibility for, a distinct collaborative activity (or ‘going concern’) constitutes membership. And this is to imply a common purpose, or shared commitment, among the members to advance that activity.Footnote 18 When an individual submits resources to a firm's procedural rules, she also submits to the managerial control of those resources (Oakeshott, Reference Oakeshott1975: 115). If managers receive authority from the members, then the constitution constrains them to act in the pursuit of purposes that have the members’ approval. The constitution does not govern the members and their relationship with management in all things (as in a Hobbesian state, in which the alternative is death), but only within the limits of their shared objectives. The drawing of those limits depends on the common purpose for which the members choose to pool their resources.Footnote 19
In accordance with the conceptual framework developed in section 3.3, the directors of a corporation (or incorporated firm) receive their authority in part from the members that elect them. Within the freedom granted by ‘superior’ authorities who bestow the privileges of the corporate form, directors act legitimately when they serve the members’ interests. What we can now argue is that those who ought to be considered members, who do in fact authorise management to act on their behalf, are those who chooseFootnote 20 to associate in pursuit of the corporation's common good, which for most ‘active’ members would involve participation in and responsibility for the collaborative activity of the firm.
This is not to imply that the roles of participant, beneficiary and decision-maker necessarily coincide.Footnote 21 Some members may be ‘passive’ beneficiaries, who cannot physically participate. It is arguable that for members capable of playing an active role, channels should be kept open for their participation in decision-making. However, we make no specific claims about the particular distribution of authority members should agree to. As Vanberg (Reference Vanberg1992: 240) puts it, since a ‘great variety of potential solutions to the collective decision problem and the distributional problem are conceivable’, one should expect to find a great diversity of organisational constitutions. Our claim is merely that when directors act for the corporation (or firm), they do so with authority from the members and are therefore obliged to serve their common interests.
From the participation of shareholders and other stakeholders in a firm's common good, certain implications follow for their membership status. Shareholders participate by investing the financial capital with which ‘the firm is able to hire workers and acquire the material resources necessary to carry out production’ (Sison and Fontrodona, Reference Sison and Fontrodona2013: 618). The shareholders’ investment, in other words, ‘renders possible the work of the other members of the firm’ (ibid.).Footnote 22 Regarding employees, Sison and Fontrodona (Reference Sison and Fontrodona2013: 622–623) point to a range of ‘organisational citizenship behaviours’ (e.g. altruism, sportsmanship, courtesy and conscientiousness) that go beyond the employment contract. Through these behaviours, ‘worker participation in the corporate common good is enhanced not only objectively, in terms of improved corporate performance, but also subjectively, that is, through opportunities to develop individual virtues’ (Reference Sison and Fontrodona2013: 623). From this Aristotelian perspective, employee loyalty ‘is not so much the result of locking-in assets as a mutual concern for each other's flourishing and well-being’ (Reference Sison and Fontrodona2013: 622).
Whether suppliers and consumers can claim membership status, and under what circumstances, is less clear. There are various means by which these stakeholders participate in the common good of the firm. Consumers can understand themselves as active participants in their relationship with the firm (Sison and Fontrodona, Reference Sison and Fontrodona2013: 619). For example, clients can engage in co-production to influence ‘the design or delivery of products to better suit their particular needs, enhancing satisfaction and strengthening loyalty to that specific corporate community’ (ibid.). Suppliers could participate in the common good by establishing an ethical supply chain initiative that diminishes exploitative labour practices and harmful working conditions (e.g. in sweatshops) (Reference Sison and Fontrodona2013: 620).
However, these cases illustrate the significance of Oakeshott's (Reference Oakeshott1975) criterion that membership entails a choice to associate, and therefore a shared responsibility for the work the members do together. The participation of consumers and suppliers in the activity of a corporation does not necessarily imply they have become part of the association, unless there is a formal agreement to associate (e.g. by purchasing shares in a firm) or a sharing of responsibility for the common purpose over time (e.g. through a long-term contribution to the corporation's aims).Footnote 23 Equally, one might ask whether shareholders and employees with only a temporary attachment to the corporation should be considered members. Mayer (Reference Mayer2013: 265), for example, argues that voting control (and by implication, membership) should be held only by shareholders who commit to their investments for a specified duration.
5. Conclusion
We began by asking why any particular group or individual should have membership status in a corporation. To ask this question is implicitly to challenge the legal norm that shareholders alone have membership status in the business corporation. However, we found that critiques of shareholder primacy, especially in the legal literature, do not typically question this assumption. To examine the issue, we turned to theories of the corporate form which, though developed in a different historical context, show how concepts relevant to the present debate (authority, ownership and participation) can apply to the corporation. By engaging with arguments from the medieval period, we extrapolated an understanding of the relationship between a corporation and its members, which in turn enabled fresh insights into the membership claims of different stakeholders in the modern corporation.
In summary, the medieval corporation, like its modern counterpart, was a legal ‘person’ with assets and liabilities separate from those of its individual members. Without inconsistency, the corporation could also be seen as the representation of its members as a collective entity. Its members, then, were collective ‘owners’ of the corporation's assets, much as described in the late 19th century by Freund (Reference Freund1897: 34): ‘The law does not connect the [property] right with A or B or C absolutely or unqualifiedly, but only in so far as they remain … within the nexus of a certain association.’ Typically, the corporation's governing body was authorised in part by election of the members, and in part through the approval of a ‘higher’ authority which confirmed the members’ right to act as a corporation.
However, we found this framework left unexplained the nature of the activity that members could participate in together to define their corporate identity over time. To complete the argument, in section 4 we explored the idea of the ‘common good’ of the firm. We found that to think of firms as relational contracts, with constitutive structures that allow for the emergence of capabilities and replication of behavioural routines, can explain the persistence of the collaborative activity that defines a corporation. We argued that this activity is captured in the concept, prevalent in Aristotelian business ethics, of the ‘common good’ of a group. We then considered how different stakeholders can participate in the firm's common good and hold the status of members, where membership is defined by this participation. If members are ‘collective owners’ who jointly authorise management to serve their interests, it follows that the role of management is to pursue the common purpose made possible by the collaborative activity in which various stakeholders (not only shareholders) may participate and share responsibility.
Future research might consider how the legal concept of the company (or members) ‘as a whole’ could reflect the interests of all who share responsibility for the company's common good, thus leading to an expanded concept of membership. For example, in the context of the UK Companies Act, Deeming et al. (Reference Deeming, Glasman, Hayes, Longley and Moore2016) have argued that employees who meet a certain threshold (in length of service and hours of work) should have statutory recognition as members of the company and the right to block takeovers recommended by the board. The concept of the corporation as a succession of members committed to a common good supports the preference for ‘committed shareholders’ (Mayer, Reference Mayer2013: 206) and a long-term perspective in equity markets (Kay, Reference Kay2012). There is also potential to extend the critique of the primacy of short-term shareholder value. Stout (Reference Stout2015: 696) has recently argued that because of its perpetuity and ‘asset lock-in’, the corporation is potentially a ‘time machine’ for long-term investments that benefit future generations.
Finally, if the membership status of shareholders is defined by their responsibility for the corporation's common good, they can be argued to have a duty to exercise their moral judgement, and voting power, with respect to the consequences of decisions taken in their name (Mansell, Reference Mansell2013; Monks et al., Reference Monks, Miller J. and Cook2004). The moral arguments for these corporate governance reforms can be strengthened through a better understanding of membership and the source of corporate authority. We have argued these problems are connected. The medieval analysis of ownership, authority, participation and the common good supports a clearer account of the relationship between a corporation and its members, and of what it means for managers to exercise authority in the corporation's name.
Acknowledgements
The authors would like to thank David Gindis, Philip Roscoe, Francois-Regis Puyou, John Ferguson, Lucy Wishart, Shona Russell, Tom Vine and colleagues at Suffolk Business School, Edzia Carvalho, David Ciepley, Geoff Hodgson and three anonymous reviewers for their helpful comments on earlier versions of this paper.
