Sydney Law Review
Current debate about corporate regulation in Australia is dominated by references to corporate governance. This is a slippery term: it is used both in discussions about the role of companies in society (where the reference is to the governance of companies) and also in discussions about the organisation of affairs within companies (where the concern is with corporate hierarchy). In the latter case, it is used sometimes to describe a desired goal (for example, companies should practice good corporate governance), and at other times to refer to the means by which desired goals (for example, satisfied investors) can be achieved. The term has become so malleable that it is tempting to dismiss it as devoid of any real content a public relations slogan which is invoked by image-conscious companies seeking to disassociate themselves from the excesses of the 1980s. But if we judge it less cynically, the idea of corporate governance is at least useful in reminding us that companies are systems of government. Debates about the role, composition and duties of the board of directors; about the role and rights of shareholders, either individually, in groups, or as a whole; and about the ways in which directors and shareholders interact within a company, all raise some fundamental questions about the structures and processes of government within corporations.
Seen this way these debates can take us to longer-standing arguments in political theory about matters such as representation, participation, the separation of powers, and the validity of majority rule. Whether it is due to indifference or resistance, corporate lawyers and others involved in debates about corporate governance have not delved into these underlying ideas very deeply, if at all.
Despite this lack of attention the departure point for this article is my belief that, more than ever, it is necessary for us to redefine our approach to corporate governance in a way that will take notice of issues such as these, and will acknowledge the political dimension of corporate organisation. I say more than ever because the corporation has become one of the dominant if not the dominant organisational forms in the late 20th century. Companies feature in all aspects of social, political, and economic life, whether the activity be profitmaking or not-for-profit, whether it be in the private or public sector, and whether it involves one or two people or a complex corporate group which transcends national boundaries.
It is easy to accept that companies are significant economic actors. For this reason the relevance of an economic analysis to the study of companies and corporate law has been readily accepted. But the overall purpose of this article is to argue for a reconceptualisation of the corporate legal structure in political terms. By paying insufficient attention to political and constitutional theory corporate lawyers have produced a rather one-dimensional picture of, and somewhat awkward responses to, corporate governance issues. I suggest that by using a political perspective we achieve a richer analytical basis for studying and responding to these issues. Over thirty years ago Richard Eells made a similar plea:
The directors of large corporate enterprises are in need of a more substantial doctrine than legal and economic theory has provided as a rationale for the powers they must exercise. Faced as they are with demands for reform of their so-called despotic corporate governments, managers must look to new sources of knowledge about the great collectivities over which they preside. The modern corporation, as a relatively unsurveyed field of social forces, cannot be measured by the old instruments or accounted for alone by old theories inherited from eighteenth- and nineteenth-century legal and economic theory.
I have given this political perspective the label of corporate constitutionalism  The reasons for adopting this label are set out in more detail in Parts 3 and 4 of the article. For the moment, I stress that my use of the word constitutionalism is more rhetorical than technical. It is intended to import a general sense of the political as well, as a sense of institutional structure: constitutions are both political documents and documents which establish structures for institutions. So, I use the term corporate constitutionalism to suggest that there are values and ideas in our public political life which should be considered in the legal regulation of corporate governance. However the qualifier corporate constitutionalism indicates that within the corporate context these values and ideas may have different formulations, applications, and consequences than in other political contexts. With this in mind, I argue that a framework of corporate constitutionalism has three key features: the idea of dual decisionmaking, which recognises the different roles of the board of directors and the general meeting of shareholders in corporate life; the idea of deliberative decisionmaking, which seeks to ensure that corporate decisions are made on the basis of an open and genuine consideration of all relevant issues; and the idea of a separation of powers, which aims to make corporate decision-making power diffuse and accountable. Drawing these three features together, corporate constitutionalism aims to provide a normative framework with which we can improve the quality of corporate decision-making.
The plan of the article is as follows. Part 2 briefly describes and assesses the two frameworks which currently dominate the analysis of corporate governance in Australia what I call legal contractualism and economic contractualism. As these labels suggest, these frameworks are both located within a contractual paradigm. Each attempts to explain, in different ways, how the formal legal model of the company works. For this reason, Part 2 begins with a general description of that model. Part 3 sets out the foundations for corporate constitutionalism. I explain the relevance of political theory to the analysis of corporate governance, and the content and importance of constitutionalism. In Part 4 I develop the framework of corporate constitutionalism in detail. I introduce the ideas of dual decisionmaking, deliberative decision-making, and the separation of powers as key features of a corporate constitutionalist framework. The conclusion, in Part 5 reemphasises the evaluative nature of the corporate constitutionalism and its use as a comparative framework alongside economic contractualism.
In this Part I assess the legal and economic frameworks which are currently used in discussions about corporate governance. Before doing this it is necessary to describe the formal legal model of the company. This is not only because both of the frameworks which I discuss in this Part are constructed by reference to this basic model, but also because it will serve as the reference for the corporate constitutionalist framework which is developed in Part 4.
In Australian corporate law, the two decision-making components of a company are the general meeting of members and the board of directors. With some exceptions (noted below) the law does not prescribe any particular division of decision-making power between these two groups. Formally speaking, this is a matter left to each company. For good practical reasons the majority of companies give full managerial power to the board of directors. Indeed, despite its formal permissiveness, the legal model tends to presume that this will be the case. This grant of general managerial power, usually stated in the companys articles of association will then be supplemented by more specific grants of power (for example, the power to issue shares, to declare dividends or to make calls on unpaid capital). The Corporations Law reserves decisions on certain matters exclusively for the members in general meeting: for example, alterations to the memorandum or articles of association (sections 172 and 176); reductions of share capital (section 195); conversions from public to proprietary status (or vice versa) (section 168); and, in the case of a public company, removal of directors (section 227). In addition, it is common for the articles to specify certain matters to be decided by the general meeting. This usually includes the election of directors and officers, the variation of rights attached to a class of shares, and the removal of directors (in the case of a proprietary company). . Still other powers are given to the members by common law principles most significantly, the power to ratify a breach of duty by directors.
Although the law is not overly prescriptive about the distribution of decisionmaking power within a company, it does restrict the capacity of members to intervene in directors' decision-making once the respective powers have been set out in the articles of association. Since 1906 courts in the United Kingdom and Australia have adhered to the basic view that where a corporate constitution confers general powers of management exclusively on the directors, the general meeting of members has no power to intervene in the day-to-day exercise of that power or to dictate the manner of its exercise. . In this model the board of directors becomes a decision-making site which has relative autonomy from the general meeting of members.
Described in this way, the formal legal model of the company is static. It merely defines who the corporate actors are and presumes, and then reinforces, a certain division of power between them. Australian corporate law scholarship tends to resort to two frameworks to animate this static model, one traditional, the other more recent. Both of these frameworks operate within a contractual paradigm. The traditional framework legal contractualism has developed integrally with the legal model of the company. . It is a positivist framework; as a matter of legal fact each company is said to be based upon a set of contractual agreements between defined parties. In contrast, the more recent framework economic contractualism uses the idea of contract metaphorically as much as factually. . So, despite being anchored in a contractual paradigm, these frameworks say different things about companies and corporate law. Nevertheless, after reviewing them I will argue that they have some common limitations and problems.
In this framework, the company as a legal entity, its directors, and its members are bound together by a contract which is embodied in the companys memorandum and articles of association. This is one of the foundational concepts of Anglo- Australian corporate law, and is enshrined in section 180(1) of the Corporations Law which declares that the memorandum and articles of a company 11 have the effect of a contract under seal between the company and each member, between the company and each eligible officer, and between a member and each other member.
The peculiarities of this corporate contract when compared to classical contract theory have long been noted but tolerated by lawyers. Recently McHugh and Gummow JJ in the High Court noted five features of this unusual type of contract.
 First, the terms of the contract can be altered without the agreement of all parties, since only a three-quarters majority of votes by those members who actually choose to exercise their voting rights is required. Secondly, there is no jurisdiction in a court of Equity to rectify the articles of association even if they do not accord with the concurrent intention of all the signatories thereof at the moment of signature . Thirdly, individual members face considerable obstacles to enforcing the contract against the company. . Fourthly, their Honours referred to Salmond Js observation  that in a company with share capital the contract attaches to the shares rather than to the shareholder. Finally, they note that under present authority a shareholder is precluded from suing the company for damages for breach of contract whilst still a member and without seeking recission of the contract whereby the shares were obtained.
Given that ordinary legal notions of contract are so strained when applied to the memorandum and articles of association, we must wonder about the rationale for the legal contractualist framework. The case law suggests two explanations.
The first explanation is historical. It traces the evolution of modern companies back to the English unincorporated joint stock companies based upon a deed of settlement. These companies evolved as an amalgamation of partnership and trust concepts, the intention being to escape the restrictions of the Bubble Act. . They consisted of large numbers of investors pooling their finances on a joint stock basis. The joint stock was vested in trustees to be managed according to the terms of a deed of settlement which was executed by each member of the company. This practice was recognised by the English Companies Act 1844 (7 and 8 Vic. c. 110). Whilst this was the first English companies statute to introduce incorporation by registration it still clung to the requirement that as a prerequisite to registration the deed of settlement should be executed by all members (sections 7 and 26). The advent of registration by incorporation, followed eleven years later by limited liability on incorporation, meant that the number of large joint stock company registrations began to grow. The requirement for each member to execute the deed of settlement became an inconvenience, and thus the English Joint Stock Companies Act 1856 introduced a provision which deemed execution to have occurred.  This provision was copied into the various companies statutes in the Australian states (for example, section  of the NSW Companies Act 1936). Those State provisions, in turn, form the basis for the current Corporations Law section 180.22
Thus, for most of its history the statutory provision now found in section 180 of the Corporations Law has not had the primary purpose of declaring the contractual status of the memorandum and articles of association. Instead, it was concerned with resolving what was essentially a procedural problem. Nevertheless, this succession of legislative provisions has been given a deeper significance. As McHugh and Gummow JJ have stated recently, the continued inclusion of such sections in companies statutes evinced the intention, as a matter of form, to preserve the link with the old deed of settlement by carrying over contractual notions to support what, in any event, would later come to be seen as incidents of modern company law.
The historical explanation is conservative. It requires us to accept that the company form has not changed in any important way since the mid-nineteenth century. But surely it is trite to observe that in the 150 years (or so) since the joint stock company first emerged, the structure, role and significance of the corporate form has undergone considerable evolution and diversification. While it is important to appreciate the antecedents of our current law, we should not simply assume that the ideologies and practices which informed earlier company law and practice are still relevant. That early company forms could be explained in contractual terms may be conceded; that modern companies can be comprehended solely in terms of legal contractual relations is a claim that requires more than the dictates of history for support.
The second explanation for legal contractualism is that it allows us to define the boundaries of the company by circumscribing the rights of membership. There has been a considerable amount of judicial and academic debate about whether the corporate contract binds the company as a party, whether it binds outsiders who are granted rights by the articles, and whether it binds members in so far as it grants them rights that are not associated with membership. I will not go into these debates here; at any rate, many of them were settled in 1985 when the precursor to section 180 was changed into its present declaratory form. The common theme in these cases is the use of contractualism as a device to define which legal actors fall within the parameters of the company, and which fall outside. Contract has supplied the framework within which issues of company membership, and the enforcement of membership rights and obligations, have been argued.
A recent case illustrates this point. In Andy Kala Pty Ltd v EJ Doherty (Northcote) Pty Ltd, Vincent J had to decide whether a clause in a companys articles of association which stipulated a particular method of resolving disputes between members of the company thereby created a contractual obligation on a member to accept a determination made under that process. Vincent J held that such an article was not enforceable as a term of the statutory contract. His Honour justified this conclusion by arguing that:
Not every dispute which arises between members of a company, even if it may touch upon their respective obligations as members, is encompassed by [section 180] ... The statutory contract contemplated by such sections as [section 180] arises from and is limited to the nature of the relationships which exist between an entity and its members, or between members, with respect to the entity itself.
Here we see that contract is used as a filtering device. It excludes from consideration the interests of non-contracting parties, as well as the nonmembership interests of the contracting parties. Within this contractualist model, [t]he purpose of the memorandum and articles is to define the position of the shareholder as shareholder, not to bind him in his capacity as an individual. By this device the company is reduced to a strictly limited set of legally defined relationships, reinforcing the legal artifice of the company form.
There is one final point to note about legal contractualism. Within its closely defined set of relationships, this framework is predisposed to favour the interests of the wider group over those of the individual member. This predisposition is not clearly articulated, nor is it absolute. Sometimes the interests of the group are defined by reference to actual majority decisions (for example, a special resolution passed at a general meeting), while at other times they are inferred or hypothesised (for example, when the directors take action in the best interests of the company). But in all cases the conclusion is the same all major aspects of traditional corporate law doctrine begin from the position that the interests of the group have a greater significance than those of individual members. Thus, directors owe duties to the members as a whole or to the separate legal entity of the company, not to individual members. If a wrong is done to the company, it is the company which brings the action an individual member cannot usurp the decision of the majority in this regard unless they can bring themselves within a limited number of narrow or vaguely defined exceptions.  It is true that in recent years legislative reforms and judicial activism have had some impact on these traditional doctrines. Members now have a statutory right of action for oppressive or unfair conduct (section 260). They may soon have a statutory derivative action. Directors of small companies have been held to owe duties to individual shareholders in certain situations. These developments, though, have been generated from outside the contractualist model, relying instead on broader notions of fairness, trust, or fundamental ethical rules of right and wrong. They suggest limitations in the contractual framework which this article seeks to explore further.
This framework draws upon contract-based analyses which have been developed by law and economics scholars, mainly in the United States. It has a growing number of adherents in the academic literature on Australian corporate law, and there is some evidence that legislative reformers have begun to take up the call. 
There are, broadly speaking, two strands of law and economics scholarship in the field of corporate law. One is agency theory which, in the corporate context, is concerned principally with addressing problems arising from the separation of ownership and control functions. The other strand is transaction-cost economics which, amongst other concerns, examines why it is that different forms of economic organisation (for example, firms rather than markets) are used to organise productive relations in different settings. Although they have differences in emphasis and analytical approach, these two approaches are mainly complementary. According to Williamson, both work out of a managerialdiscretion setup. They also adopt an efficient-contracting orientation to economic organization. And both argue that the board of directors in the corporation arises endogenously.  A further commonality is that neither approach regards the organisation as a unit of analysis.  Relying on these commonalities, the following synopsis represents an amalgamation of these two strands.
In the economic contractualist world the company is analysed as a type of firm. According to one definition, the firm is simply a set of feasible production plans which buys inputs and sells outputs in well-developed spot markets. Firms are assumed to maximise profits, or adding time and uncertainty, maximise expected net present value of profit.  A firm consists of a series of transactions, or contracts, between investors, managers, employees, creditors and customers. Agency theory sums up this arrangement in the well-known description of the firm as a nexus of contracts. The use of the term contract here does not necessarily correspond to the strict legal idea of contract. Rather, it is intended to draw attention to the voluntary and adaptive nature of the arrangements which are made between the participants.
The economic framework assumes that managers will use their discretionary powers to maximise their own gains in ways which will not necessarily coincide with maximising the firm's profits. Investors in a company will therefore be concerned to ensure that managers their agents will run the firm in a way that maximises profits rather than managerial self-interest. For their part, managers will be concerned to retain their positions, and they will seek to assure investors that their interests will not be harmed. Therefore, contracts between managers and investors will be concerned, amongst other things, to minimise these so-called agency costs (or, in the language of transaction-cost economics, to reduce managerial opportunism). The process of negotiating these contracts is imperfect. Investors will be limited by the amount of time and knowledge they can devote to negotiations (they are said to have bounded rationality). Moreover, in firms with widespread ownership the prospects of collective action amongst investors will be slight. In summary, the prospects of effective monitoring are low, and the transaction costs of negotiating contracts will be high.
In the case of companies, the economic framework points to two mechanisms which can reduce these transaction costs and act as an efficient substitute for investor monitoring of management behaviour. The first mechanism is the free operation of three efficient, competitive markets: the market for corporate securities (whereby information about the company, including the terms which bind the managers, is built into the price of the securities); the market for corporate control (whereby underperforming and inefficient companies will be taken over by those who place a higher value on the company's assets); and the market for corporate managers (which controls the risk of outright abuse of managerial power). The operation of these competitive markets supplies information about companies and constrains the misuse of private economic power.
The second mechanism is provided by a system of corporate law. According to Easterbrook and Fischel: Corporate law is a set of terms available off-the-rack so that participants in corporate ventures can save the costs of contracting ... Corporate law and in particular the fiduciary principle enforced by the courts fills in the blanks and oversights with the terms that people would have bargained for had they anticipated the problems and been able to contract costlessly in advance. On this view corporate law supplements but never displaces actual bargains ...42
The final sentence in this passage emphasises the private and voluntary nature of corporate relationships within this framework. As a consequence corporate law rules ought to be regarded only as default rules, allowing tailor-made contracts to be adopted. Corporators should have the freedom to opt out of statutory or judicially created rules and to construct their own contractual arrangements. Some law and economics writers take a less stringent view, allowing that corporate law ought to prescribe some mandatory rules and impose some duties and responsibilities in order to correct potential problems which arise from the principal-agent relationship.
Finally, the economic contractualist framework adopts the philosophy of methodological individualism and simply denies the existence of an association or organisation as something apart from the individual constituents.  As Easterbrook and Fischel put it, [t]he personhood of a corporation is a matter of convenience rather than reality.  Furthermore, it insists that for analytical purposes the only relevant function of a company is to maximise the returns to individual investors. Thus companies are assessed on the same single dimension as sole traders they are simply different forms of business organisation.
The legal and economic contractualist frameworks operate within the same paradigm and therefore they have common points of reference, but different points of emphasis. Both frameworks offer a promise of illuminating the internal operation of the firm,  but neither completely delivers on that promise. Legal contractualism promises a framework which takes account of the roles of directors, shareholders and the company. What is presented, however, is a disjointed collection of rules in which the interests of the wider group sometimes identified with the members as a whole, at other times with the company as an entity detached from members and managers are presumed to prevail over those of individual corporate actors. Economic contractualism promises a framework that either eschews or plays down consideration of the company as an analytical construct, focusing instead on the roles of managers and shareholders. But despite the attention which is given to the significance of managerial discretion, this framework ultimately reduces intra-corporate activity to one dimension that of the shareholders for whom managers are ciphers. The shortcomings in both of these frameworks can be traced, in different ways, to their contractual origins. Below, I list four limitations or problems for corporate law and corporate theory that result from a preoccupation with the contractual paradigm.
The first limitation is that the contractual paradigm only comprehends relationships between pairs of individual legal or economic actors. In the legal framework those actors are the company as an individual entity, and each director and each member. In the agency strand of economic theory described above, the company is dropped from this list. In legal and economic contractual theory there is a tendency to reduce corporate organisation to a multiplicity of bilateral agreements. Brennan and Buchanans admission that 2 remains the magic number for the economic analyst also holds for the legal framework. As a result, there is no image of the company as an organisation. Issues of corporate government and power are fractured into a succession of narrowly defined, single instance arrangements. My criticism here is sociological. The organisational life of a company is more than the sum of the actions of individual corporate insiders. What individuals are and do within companies involves a complex interplay of power relations, hierarchies, loyalties, and systems of control, which may vary over time and from one decision-making context to another. As Ewick emphasises, the relationship ... is dialectic. Individuals can no more be separated or detached from their organizational affiliations than the organization can be abstracted from its membership. 
The second limitation is that contractualism favours an economic approach over a political approach. The difference between economic and political in this context is explained succinctly by Kukathas and Pettit:
When two or more people seek to make an agreement which affects their interests differently, so that each would most prefer a different arrangement from the other, the agreement may be pursued in either of two ways: one we describe as economic, the other as political. The economic way is for each to calculate what best suits his own interests and then to try to get this ... The political way is for the parties to put aside their own particular interests and to debate about the arrangement that best answers to such considerations usually considerations in some sense to do with the common good as all can equally countenance as relevant. The economic approach is institutionalized in the process of market negotiation, the political at least ideally in the forum of discussion where the parties are blocked, if only by the sanction of social disapproval, from arguing by reference to special as distinct from common concerns. 
In both the legal and economic frameworks contracts are presumed to be the outcome of negotiation and bargaining, whether actual or hypothetical. In other words, contract is the paradigmatic arrangement for economic thought. Within this paradigm the company is simply an economic means to achieve an economic end.
The third limitation follows on from this. The contractualist frameworks tend to be oriented towards end results, and this is because of the attitudes they embody about consent and voluntariness. Liberal legal and economic theory begins with the premise that individuals should be free from interference and constraint imposed by others. The idea of freely-given consent is then invoked to explain how others may legitimately impinge on that freedom. So, in the corporate context, actions such as the exercise of power by the directors (or by a majority), the allocation and removal of corporate rights and benefits, or the incurring of obligations to the company, are all legitimised by the voluntary and informed consensual grant of power from individual members to directors which is embodied in the corporate contract. Section 180 of the Corporations Law thus represents a primal act of consent that legitimate[s] the subsequent exercise of authority by the office holders. Provided that corporate actions and decisions comply with the terms of the contract they can be judged primarily in terms of whether they achieve some desired goal, rather than by reference to their impact on the rights or interests of the persons involved.
Finally, the contractual paradigm reinforces a belief that there are unique and different rule systems which should apply in a world which is divided into private and public spheres of social and economic activity. So, corporate law is seen as a body of rules which supplements a system of private, individual agreements and, consequently, legislators, judges and corporate regulators bear the burden of justifying why there should not be less external regulation. In the same vein, any intrusion of public law-like concepts, such as natural justice, procedural fairness, or equality of opportunity, into the world of corporate law is also judged with suspicion. 
To summarise, the contractual paradigm is limited because its approach is essentially individualistic, economic, ends-oriented, and private. Now, although I describe these ideas as limitations, I do not deny that they have a role in analysing corporate governance. As Bratton has pointed out, we can easily find evidence of discrete contractual relationships in corporations.  My criticism, like Brattons, is directed at the tendency for these ideas to be overstated, and to be used to the exclusion of other approaches. Unlike Bratton, I fear that even a remodelled contractual paradigm offers little scope for moving away from a preoccupation with these ideas. Rather than trying to promote a more flexible notion of contract, what is needed is a complementary framework, drawing upon a different paradigm.
A different vision of the company might draw upon the democratic ideal which inspires the relation of the citizen to the state. The democratic ideal asserts that those who are substantially affected by the decisions made by political and social institutions in our society should be involved in the making of those decisions. 
In this Part I begin to develop a different vision of the company, which I call corporate constitutionalism. As I noted in the Introduction, the purpose behind adopting a broad constitutionalist perspective is to introduce transparently political considerations into our analysis of corporate institutions.  It is necessary, therefore, to explain in more detail why political theory is a useful perspective from which to assess corporate issues, and what it is that the perspective of constitutionalism might offer us that the contract-based models outlined above do not. This is a prelude to the subsequent Part of the article which looks at the substantive features of corporate constitutionalism.
What I mean by political theory is an approach which allows us to evaluate governmental systems. As Philip Pettit explains, this is a normative process:
We are to identify the purposes of government more strictly, the proper purposes of government so that we can decide on which arrangements it is best for a government to foster in a society: which basic constitution it is best to establish and which procedures or outcomes it is best to prescribe in the day-to-day operation of the society. 
To summarise its application to companies, and with apologies to Kukathas and Pettit, political theory invites us to consider questions such as: What is the purpose of a company? How should companies be organised? What claims do companies have over their members? 
The intuitively simple justification for resorting to a political framework to evaluate companies is that they are political entities. There are two complementary ways in which this is true. First, as I noted earlier, companies, particularly listed public companies, are prominent and powerful actors in our public life. They are major employers and taxpayers; their actions, collectively and individually, have considerable impact on our social and economic structures and the environment. These are not new observations, even though they do seem to be suppressed in legal thought. For many years a variety of commentators, from a variety of perspectives, have made the same point.  Maitland noted that the company and the state are two species within a single genus that of more or less permanently organised groups of individual actors, group units to which we attribute actions, intentions, praise and blame.  More recently, Unger has made the similar observation that modern society looks more like a constellation of governments, rather than an association of individuals held together by a single government. 
Ungers observation takes us to the second sense in which companies can be said to be political institutions, and it is this point which is the primary focus of this article. Companies are political institutions not simply because they are players in social power relations, but also because they themselves are systems in which power and authority, rights and obligations, duties and expectations, benefits and disadvantages, are allocated and exercised, whether actively or passively, collectively or individually. Each company is a body politic, a governance system. This is as true, albeit in different ways, for the smallest, closely-held company as it is for the largest public company. Again, this is not a novel observation writers as diverse as Thomas Hobbes, C. Wright Mills, and Adolf Berle Jnr have all made the same point.  Nor is this an incidental or trivial feature of companies all companies are political as much as they are economic entities. 
The choice between an overtly political theory and the more widely accepted economic analysis of corporate law has been explored by several writers. Lynne Dallas, for example, advocates a power model of the firm, which stands in opposition to the efficiency model (the latter being the economic contractualist framework described earlier in this article).  The power model draws upon organisational and management theory. It focuses upon the political nature of decision-making in the large corporation, and sees the firm as an organic institution with its own internal structure and processes that impact on the control of the firm.  Melvin Eisenberg, on the other hand, favours what he calls the Economic Model over the Political Model of the large publicly held corporation.  As he depicts it, the latter model posits the corporation as essentially a political institution whose legitimacy depends upon the extent to which it is governed by principles appropriate to a democratic state.  Those principles, according to Eisenberg, require a system of representative government which allows participation in decision-making by the various constituencies in the company. Eisenberg questions the relevance of this model to large companies. In his assessment, the Economic Model better reflects both accepted conceptions of institutional competence and legitimacy and the functional characteristics of the large public corporation. 
The argument which I present in the remainder of this article differs from these positions in two ways. First, unlike both Dallas and Eisenberg, I do not think that it is useful to advocate one model or framework to the exclusion of all others. Whilst I argue for a broadly political framework, I do so with the intention that it should be considered as a complementary perspective to, rather than a substitute for, legal and economic analysis. Because companies can be categorised as political as much as economic institutions, they should be subject to political as much as economic analysis. Companies can be many things at one and the same time and, as Gareth Morgan reminds us, there can be no privileged or ultimate analytical framework for studying the complexity of corporate organisations.  Different theoretical frameworks will reveal different things about companies and their internal organisation. In particular, we must be prepared to make a comparative evaluation of economic criteria such as efficiency as against constitutionalist criteria such as participation.
The second point is that in studying companies and their regulation, political theory must be adapted, not simply adopted. Eisenberg apparently sees all political theory in terms of public (for instance, state) politics, leading him to dismiss it as inappropriate for the study of companies. In his view, the proposition that all political institutions should be governed by political principles, and that any institution that makes decisions of major importance to society is a political institution  trivialises the critical differences which exist between the state and the company. These differences are important, but acknowledging them does not then require us to abandon the job of tackling the more complex questions of the ways in which, and extent to which, political theory should influence companies. What is needed is a corporate species of political theory; as I explain in Part 4, this is what I mean by the term corporate constitutionalism. Before exploring this idea it is necessary to explain what I mean by constitutionalism.
In the context of this article, the term constitutionalism has rhetorical, methodological, and normative, rather than technical, uses. It is rhetorical in that I use it to import liberal political theory in general rather than any particular species of that theory. It is methodological in that it presumes that in order to discuss individuals one must look first at their communities and their communal relationships.  It is normative in that it provides a benchmark against which to evaluate the structures and processes of particular governance systems.
The concept of constitutionalism in the history of liberal political thought demonstrates a series of oscillations between two models of government direct self-government and representative government. Liberal democratic theory has tended to be oriented towards the second of these models. At the risk of some generalisation, we can say that in this model liberal democratic theory has been concerned primarily with three issues.  The theory begins from the proposition that the systems of direct or self-government are not feasible in modern large-scale societies. At the same time, there is a concern to preserve, as much as is possible, the sense that individuals can participate in the processes of societal decisionmaking. Hence, what is required is a system of representative government, whereby a smaller number of elected individuals can exercise state functions and make decisions that are binding on the entire community.  Representative government, however, is no guarantee against arbitrary rule; there is always the possibility that those who are elected to govern may use their power to ignore the mandate of representation and serve only the interests of the majority faction in society (the tyranny of the majority) or of a smaller elite.  This possibility of abuse of power can be reduced by breaking up concentrations of government power, and by implementing a system of checks and balances on the exercise of each power. In particular, the power to make rules (legislative power) should be separated from the power of enforcement (executive power), and both should be independent of the power of the judiciary. This separation of powers is reinforced by the recognition of constitutionally reinforced rights and liberties. Where these safeguards fail there may then be calls for avenues of political participation that go beyond mere electoral processes.
The concept of constitutionalism which emerges from this theory has been strongly influenced by legal thought. Within legal discourse the idea of constitutionalism is allied with, and often treated as a synonym for, the rule of law. Whatever the degree of linkage between these two ideas the point is the same: constitutionalism is usually defined in terms of procedures and structures such as due process, the separation of powers, and respect for individual rights.  The liberal legal version of constitutionalism focuses on the need to restrain the public exercise of power and on the creation of structures to maintain those restraints. In liberal theory, constitutionalism defines and maintains a formal separation of private spaces from the public space. 
Whilst I acknowledge these aspects of constitutionalism, I also want to define the concept more broadly. This is because all too often, as Schochet notes, constitutionalism supports a rule formalism that maintains the status quo, leaves no room for political discretion, is incapable of sanctioning resolute responses to emergencies and radically new situations, and is indifferent to political substances.  A broader approach recognises that a constitution creates and defines the governance structures of a political institution and in doing so it helps to define the institution itself. Unlike a contract, a constitution is more than just an agreement about the allocation and exercise of power. A constitution both recognises and reinforces the place of individual constituents within the institution, and also constitutes them as a group or collective. Constitutionalism therefore directs our attention to what it is that unites these individuals rather than just what separates them. It allows us to contemplate the relations between the collective and the individuals who comprise the collective. Thus, a constitution has an integrating role. Sheldon Wolin puts it this way: A constitution not only constitutes a structure of power and authority, it constitutes a people in a certain way. It proposes a distinctive identity and envisions a form of politicalness for individuals in their new collective capacity. 
Later he adds:
a constitution is an experiment in the forging of a collective identity. The identity of the collectivity, who it is and what it stands for politically, is made known through the constitutionally sanctioned actions of public officials and the response, or lack thereof, of the collectivity to those actions. Thus, a constitution has a circular nature: it is constituted by the collectivity ... and the actions performed under it, in turn, constitute the collectivity. 
Whilst the origins of liberal constitutionalism lie in a concern to protect the rights of the individual against absolutism or overbearing power, it is not essential to a constitutionalist theory that the interests of the individual should always trump those of the collective. A constitutionalist framework for companies need not presumptively favour one over the other, but it can give us a framework within which we can make an appropriate balance between these interests as the situation demands. Contract-based models do not allow this. As I noted earlier, the economic contract model simply abolishes the idea of the collective from the outset. The legal contract model, on the other hand, gives only partial (and somewhat confused) recognition to the interests of the individual against those of the company as a whole. Of course, these can be difficult decisions. Determining what is in the collective interest or what counts as a legitimate individual interest and then mediating between the two is problematical. In large part this is because there is no neat dichotomy between these interests.  The relationship between the group and the individual emerges from a continuous process of negotiating and defining power relations.
Not only can a constitutionalist framework recognise the complexity of the relationship between individual and group, it can also support the idea of the individual as a citizen or a member. That is, both the public and the private aspects of a persons existence can be recognised. Recognition of citizenship has two aspects recognition of the rights and the responsibilities of citizenship. In the next Part of the article I will elaborate on these two aspects within the corporate context.
Finally, one (hopefully obvious) feature of this argument should be stressed. There is no single type of constitutional organisation. As Schochet notes, no one set of political arrangements is uniquely required by constitutionalism.  Different types of institutions will be constituted in different ways. This applies not only to comparisons between companies and states, but also between different types of companies. Because of this, different applications of political theory, different conceptions of democratic organisation and process, and different political standards (for example, levels of voter participation) will be required for different types of institution.
In this Part I elaborate a framework of corporate constitutionalism, working from the ideas presented in Part 3. I begin by explaining the importance of developing a corporate species of constitutionalism. Then I set out the key features of a corporate constitutionalist framework.
As I have already noted,  constitutionalist ideas must be adapted to the corporate context, not simply adopted by it. This is because companies are political institutions in different ways to the institutions of public governance. There are no strict analogies between public government and the system which is envisaged in the formal legal model of the corporation, although there are some superficial similarities. In so far as it is elected by company members, the board of directors may be said to occupy the same role as that of the legislature. Senior corporate managers occupy a position similar to that of the executive (that is, implementing decisions made by the board), while company members loosely approximate the general electorate. But because of important structural and doctrinal differences it is not useful to pursue these analogies too far. For example: according to legal doctrine company directors do not represent specific groups of shareholders; company membership is voluntary  and shifting; and voting power is usually calculated per share rather than per member. It is therefore safer to assume that the assumptions underlying public governance structures are not necessarily applicable to companies. At the same time, however, we can say that even though different conceptions of constitutionalism will apply, both companies and the state are constitutional institutions. As Pound has observed: The political approach to corporate governance accords with ... values about how major institutions in our society should be governed, emphasising due process, substantive debate, and the use of formal voting referenda. 
Accepting the need for adaptation, the next question concerns our point of reference. The difficulty in constructing a conceptual framework for companies is the great diversity and variation in forms of corporate organisation. I propose to use the formal legal model of the company as the starting point for this exercise. This is not because I think that the legal model is an accurate description of the way companies actually operate, nor because the legal model is an organisational ideal to which companies should necessarily aspire. The formal legal model is useful because, regardless of their structural diversity, companies are legal creations and thus the legal model provides a common reference point and language for all companies. In other words, we can assess any company in terms of how it differs in practice from the legal model. I anticipate that establishing a constitutionalist framework on this basis will also reveal the limitations of the legal model, and thus provide a sound basis for advocating reforms to that basic model.  Initially, however, the purpose of corporate constitutionalism is to change the way in which we think about companies, without at this stage advocating any fundamental transformations or reforms of basic corporate structures or the corporate legal environment. 
With this in mind, the adaptation of constitutionalist language to the corporate context is not as novel as it may first seem. Everyday usage and history each suggest that there is some relevance. There is a tendency in general discussion, reinforced by the Corporations Law itself, to use the term constitution to describe the memorandum and articles of association of a company.  Of course, this is a weak or euphemistic use of the term. It is used simply as a short-hand expression to denote the rules and regulations which govern internal corporate management. History provides another illustration. Harold Berman has shown how the law of corporations played an important role in the development of canon law in Europe during the late eleventh and twelfth centuries. He points out that it was early corporation law which applied the principle of jurisdictional limitations on power and authority to the corporate legal entity of the church. 
It may be objected nevertheless that constitutionalism has no role to play in the modern corporate context because within a company all is private. This objection relies on the idea that constitutionalism is concerned solely with the separation of the private from the public realm, and on the idea that a company is an entity in which members and managers meet each other as private and individual actors in the pursuit of private and individual goals.  Both of these ideas are contestable. It is, of course, true that for the most part companies are formed to further private rights and interests,  but these need not be considered solely in individualistic terms they also have a collective dimension. Within a company we can identify a realm which is analogous to the public realm in social life call it the corporate realm. By necessary implication, we can also define a realm which is personal or private. The distinction, and the analogy, can be illustrated by comparing my dual status as a citizen (my public role) and as a parent (one of my private roles) with that of a person as a member or director of a company (their role in the corporate realm) and as a shareholder (one of their personal or private roles).  While it is true that in some situations these distinctions may become blurred, nevertheless they remain fundamental to the analysis of political institutions. It should be noted that the formal legal model of the company recognises these two areas of corporate life on the one hand it stresses the need to have regard to the interests of the company as a whole,  while on the other hand it allows that members may act to protect their personal rights against intrusions by the majority of members.  One role for constitutionalist theory in a company is to provide a framework for mediating the corporate and the private realms. In other words, corporate constitutionalism can do for companies what a general theory of constitutionalism does for society at large.
In this section I propose what I see as the three key features of corporate constitutionalism: dual decision-making, deliberative decision-making, and the separation of powers. It is important that these should not be read as descriptions of the rich and varied detail of modern corporate life. As I stated earlier,  this is intended as an evaluative exercise in which these key features provide benchmarks against which actual corporate practice and its legal regulation can be assessed.
Each of these features of corporate constitutionalism stems from the idea that decision-making is a central part of corporate life. Decisions must be made about corporate finance, employment and industrial relations, manufacturing and marketing, research and development, corporate structure, and so on. A large part of modern corporate law in Australia is devoted to rules, doctrines and standards aimed at all aspects of corporate decision-making. Thus we have requirements for the production of information relevant to decisions (for example, preparation of annual accounts, takeover documents, and prospectuses), rules relating to the procedures for requisitioning and conducting the meetings at which decisions are made, and standards governing the behaviour of people making decisions (for example, rules relating to duties and standards of care for directors, oppressive conduct against minority shareholders). Corporate constitutionalism is, therefore, primarily concerned with issues of structure and process, rather than with substantive doctrines.
(i) Dual decision-making
As I noted in Part  , the formal legal model of the company recognises only two decision-making elements the general meeting of the members, and the board of directors.  An important function of the corporate constitution is to constitute these two elements as separate decision-making sites, to confer decision-making powers on each of them, to regulate the exercise of those powers, and to guard against encroachment by one upon the other.
It is important to be clear about the status of the corporate constitution in carrying out this function. It is one thing to say that members, via the memorandum and articles, confer powers on the directors, or that the memorandum and articles are an agreement between members and directors about the division of power. This contractualist approach is embodied in Cozens-Hardy LJs judgment in Automatic Self-Cleansing Filter Syndicate Co Ltd v Cunninghame:
The shareholders have by their express contract mutually stipulated that their common affairs should be managed by certain directors to be appointed by the shareholders in the manner described by other articles, such directors being liable to be removed only by special resolution. If you once get a stipulation of that kind in a contract made between the parties, what right is there to interfere with the contract, apart, of course, from any misconduct on the part of the directors? 
It is a different thing to say that the corporate constitution confers powers on each of those two bodies. This formulation puts the corporate constitution above the status of a simple agreement. Elements of this approach can be found in the Privy Council's reasoning in Howard Smith Ltd v Ampol Petroleum Ltd:
Just as it is established that directors, within their management powers, may take decisions against the wishes of the majority of shareholders, and indeed that the majority of shareholders cannot control them in the exercise of these power while they remain in office, so it must be unconstitutional for directors to use their fiduciary powers over the shares in the company purely for the purpose of destroying an existing majority, or creating a new majority which did not previously exist. To do so is to interfere with that element of the company's constitution which is separate from and set against their powers. 
Ultimately this formulation relies upon the idea that the companies (and their constitutions) are derived from the permissive actions of the state in granting incorporation, legislating general corporate laws, and so on. 
The corporate constitution creates a dual decision-making system.  In this system decisions by members in a general meeting are usually not everyday events. In many companies they occur only annually.  It will be recalled that the Corporations Law gives the general meeting exclusive power to make decisions on certain matters, which are fundamental to corporate affairs (for example, alterations to the articles of association, and reductions of share capital). All of these decisions are constrained by a variety of constitutional conditions (though not all decisions are constrained in the same way), including notice requirements and special majority requirements (for example, ss 172, 176, 195, 168). Despite the often routine way in which some of these decisions are made in practice, it is nevertheless true that within the formal legal model decisions made by members have a special significance which is not possessed by directors' decisions. While both the board and the general meeting can speak as the company, in the absence of contrary evidence, a decision of the members is presumed to be a better approximation of the corporate will compared to a decision of the directors.
Decision-making by directors occurs much more frequently. In the great majority of companies, the corporate constitution confers general managerial power on the board of directors, which can then delegate aspects of that power to senior managers in the company. Indeed, if we include the work of managers exercising power delegated by the directors, we see that these decisions are part of the daily life of the company. Borrowing Ackermans phrase, this can be regarded as the normal path for corporate decision-making.  The dual structure recognises that in many companies most shareholders do not have any regular involvement in decision-making. Thus, normal decision-making is, with the authority of the members, largely the business of directors and managers.  However, because the power to alter the articles and the memorandum resides in the general meeting, decisions by directors are constitutionally subordinate to decisions by members. Directors decisions are also constrained by constitutional conditions which are referenced to the interests of the general membership. These constraints include rules in the articles of association about holding board meetings, voting and other procedures at those meetings, as well as more general requirements to account to the membership on a regular basis.
To summarise, the application of corporate constitutionalism to a formal legal model of the company results in a dual system of decision-making. The normal path for corporate decisions is through the directors and managers of the company. The higher path is through the general meeting. It is important to be clear that the labels normal and higher (which I borrow from Ackerman) refer to the processes and structures of decision-making. They do not necessarily describe the substantive decisions themselves. Directors will make many significant decisions which affect the company, while general meetings will often make seemingly mundane decisions.
There are two important and related implications of this dual decisionmaking system. One is symbolic: the system works as a signalling device to corporate insiders and outsiders about which issues, and therefore which interests and values, are regarded as fundamental to the company. Those issues which are allocated to the general meeting are presumably those which are regarded as fundamentally significant to the company. To the extent that a company is able to tailor-make its own constitutional provisions, then different companies can send different signals. The second implication is more practical: this dual process, with the attendant constitutional constraints, affords a method for securing designated rights or interests from easy or overly-expedient interference by controlling interests in a company. Decisions are made on the normal path on the basis of ultimate accountability to the higher path, while those made on the higher path have the protection of being open to scrutiny and direct participation by members.
Not everyone will agree with this dual decision-making model. I can anticipate three objections: one, that the dual model emphasises the roles of directors and members at the expense of other important constituencies; two, that the model gives too much significance to the board as a separate decision-making site; and three, that the dual model gives undue prominence to the members. I will consider each of these arguments in turn.
Stakeholder theorists might object that the dual model defines the relevant corporate interests too narrowly. Indeed some would argue that, given their general passivity as shareholders, members have the least claim to be recognised in comparison with the interests of other stakeholders.  Stakeholder theorists argue that existing corporate legal structures should be reformed so as to cater for a wider range of non-shareholder constituencies, principally employees, but also creditors, customers, suppliers, and local communities.  I have some sympathy with these arguments, but they fall outside the immediate purpose of this article, which seeks to develop a constitutionalist framework within which the formal legal model of the company can be evaluated. The idea of dual decision-making presented here is simply a political interpretation of that formal model. Once the constitutional parameters of the formal legal model have been evaluated, I envisage that corporate constitutionalism might well be used as the basis for considering the claims of non-shareholder interests in the corporate governance process. 
Economic contractualists may raise a different objection, arguing that the dual model fractures what they regard as an agency relationship between directors and owners, and that it gives undue status to decisions of the former group. Agency theory regards all decision-making processes within a company as having the same weight and significance. According to this theory, the members could make all decisions; however it is rational and efficient for them to appoint others to make their decisions for them. For an agency theorist, the only issue of interest that then arises is how to monitor and control the managers in exercising that power. In the final analysis, the principal/agent analysis treats any decisions made by the agent as if they were decisions of the principal. In contrast, the dual model highlights the point that director and member decisions are the product of categorically different processes and are each worthy of specific attention.
A third possible objection to the dual decision-making model might take essentially the opposite stance to the agency theorists, arguing that the model gives undue recognition to the members when what really counts is decision-making by directors. This is based on the idea that provided the directors are properly elected, and provided the corporate constitution gives them full managerial power, then they have plenary decision-making authority in the company. Of course members will retain some decision-making power over those issues specified in the corporate constitution (including the power to elect and remove directors), but this power should be exercised with circumspection. I will deal with this argument in a little more detail because unlike stakeholder theory and economic contractualism, it has not featured prominently in the corporate law literature.
The argument, which has been labelled democratic or competitive elitism,  has its origins in the works of Joseph Schumpeter and Max Weber. It is a procedural, rather than a substantive, theory of democracy. As Schumpeter defines it, democracy is to be found in the process by which voters choose their political leaders. That process implies an electoral system in which there is a competitive struggle for leadership.  The only function of the electorate in this process is to exercise its free vote to produce a government  (this function includes the power to evict that government). This is the limit of the electorates control over the leadership. Using Schumpeters description again:
Voters do not decide issues ... In all normal cases the initiative lies with the candidate who makes a bid for the office of member of parliament and such local leadership as that may imply. Voters confine themselves to accepting this bid in preference to others or refusing to accept it. 
It follows from this that having elected their leaders, voters must refrain from acting as back seat drivers  and from interfering in the processes of government. Schumpeter urges that voters must respect the division of labour between themselves and the politicians they elect.  Weber made the same point in more blunt terms: In a democracy the people choose a leader in whom they trust. Then the chosen leader says, Now shut up and obey me. People and party are no longer free to interfere in his business. 
This functional separation of the elected from the electors is reinforced by two interrelated developments. One is the professionalisation of leadership roles in modern society; the other is the separation of leadership from administration, and the increased specialisation in the latter set of tasks. On the second development, both Weber and Schumpeter argue for the necessity of a well-trained bureaucracy. Weber argues that it is the capitalist market economy which demands that official business of the administration be discharged precisely, unambiguously, continuously, and with as much speed as possible.  So, to summarise the argument, modern political systems are said to be characterised by a three-way division of function: a largely passive electorate, a competitively elected leadership which is supported by a bureaucracy of specialised administrators.
It is easy to see how the competitive elite theory could be accommodated within the formal legal model of the company.  As I noted earlier, although the law does not prescribe any particular division of power between the members and the directors, it does set strict limitations on the behaviour of the members once a division of power has been made. Competitive elitism supplies a non-contractual rationale for those limitations, and for the grant of plenary management powers to the directors.
Competitive elitism has been criticised on a number of grounds, only some of which are noted here, but they provide a sufficient rebuttal for present purposes.  In general, the theory has drawn criticism because of its bleak picture of political life. The theory slides easily from elitism towards oligarchy, in so far as it assumes the dominion of the elected over the electors.  Another limitation of competitive elitism is that it promotes a passive role for citizens after they have exercised their vote. The theory has no room in which to develop the idea that members of a political community can participate as members beyond the ballot box. Finally, competitive elitism favours procedure over substance. We do not concern ourselves with why people vote, nor with the quality of debate and deliberation (if any) which precedes that vote. As long as there is apparent free competition for the vote, democracy is satisfied by a majority vote. The ideal of dual decision-making seeks to counter each of these points. In particular, it recognises the potential role of members after electing the directors, and it sets out a structure for the accountability of decision-making power. It does not, of itself, ensure deliberation; for this we must turn to the second feature of corporate constitutionalism.
(ii) Deliberative decision-making 
Having recognised the relative roles of the two formal decision-making sites in the company, the second feature of corporate constitutionalism seeks to ensure that as far as possible the processes of decision-making in each are open and genuine. By open I mean that decisions ought to be formed by processes of inquiry and a consideration of all relevant arguments. As Sunstein puts it, decisions ought to be the outcome of processes of deliberation and discussion, in which new information and new perspectives are brought to bear.  By genuine I mean that the board room and the general meeting ought to be places where decisions are made, rather than merely ratified. Corporate decision-making should involve processes of deliberation rather than a mere aggregation of a majority of the votes.  In John Deweys words, the counting of heads compels prior recourse to methods of discussion, consultation, and persuasion.  This idea of deliberative decision-making thus stresses the importance of what occurs before a vote is taken and a majority is formed.
The formal legal model of the company already embodies the idea of deliberative decision-making in a number of ways. Both expressly and by implication, the model requires the directors and the general membership to exercise their power collectively through meetings. The Corporations Law and the Table A articles contain detailed procedural requirements for these meetings, covering preliminary processes (such as notice requirements),  and procedures during meetings (such as voting, quorum, proxy and minutes requirements).  These requirements are premised on the idea that the meeting is a forum in which issues will be put, arguments will be heard, and decisions made. In short, the premise is that of deliberation. One aim of deliberative decision-making is to achieve corporate decisions that can rely on a deeper claim to validity than simple majority support. If the process of decision-making is to have this result, it must offer real opportunities for involvement and influence. Amongst other things, this means creating a sense that input into decision-making is important and will be taken into account equally with other contributions.
Having said this, we must acknowledge the practicalities of corporate life. The ways in which member participation is able to be manifested will depend upon factors such as the type of member (for example, institutional v individual), the relative size of shareholding in the company, the nature of shares or membership rights and the membership-management structure of the company (for example, widely v closely held). For example, in smaller companies it is conceivable that participation by members will extend beyond voting indeed, voting may often be regarded as a mere formality for the sake of the required corporate records. Instead, participation will mean involvement in the processes which are set in motion by voting. In companies of this type we can talk of direct democracy, that is, where the difference between the higher and normal decision-making paths is minimal. Political theory suggests that direct democracy is only feasible in organisations where the number of participants is small and limited, the participants do not differ greatly in social or economic position, administration of the organisation is relatively simple and stable, the matters for decision directly affect the participants, and the participants are confident that their input will count. 
In larger companies members frequently have only a limited engagement with corporate affairs.  As law and economics analysis has stressed, gaining even a rudimentary understanding of corporate affairs (for example, which directors to vote for, whether to accept or reject a takeover offer) can take a lot of work, and often requires specialised knowledge. On any given occasion, a member may not participate in decision-making for any one or more of a number of reasons: not understanding the issues; deliberate abstention; not being personally affected by the issues; indecision; plain indifference; and so on. Whilst the pressures towards passivity are understandable, this does not mean that member passivity should become a foundational assumption of our theoretical framework, as is the case with economic analysis. A constitutionalist framework should encourage the creation of legal rules and corporate structures and processes which emphasise the possibilities for participating members rather than passive investors or residual claimants.
One of the normative foundations of corporate constitutionalism is that companies should operate in ways which permit and encourage relevant participation by members and directors. The reference to relevant participation is important. The idea of deliberative decision-making is not so far reaching as to require that all corporate decisions should be the result of broad-based discussion between members. Deliberative decision-making takes place within the dual framework described in the previous section.  Dual decision-making acknowledges that in many companies there are matters which most members will not understand or will not want to become involved in, even though their interests may be affected in the long run. These are matters which are therefore usually left to the directors/managers to decide. On the other hand, there are matters which so directly affect the interests of members that their participation is necessary through the general meeting. 
A commitment to the principle of deliberative decision-making can be tied to other aspirations. One is that the individuals involved will recognise and acknowledge their involvement in a corporate endeavour, and so frame their arguments and contributions in terms of the corporate interest (however that may be defined).  This is not to suggest that individual members must abandon their private interests and concerns. It is undeniable that private interest does and will often supply the main motivation for a members input to a corporate decision. Instead, it is an argument that those private interests and concerns alone should not be a sufficient legitimation for decisions and actions. Nor do I suggest that the corporate interest automatically trumps the personal interests of members in all cases. Instead, I am arguing that considerations of corporate interest should be a necessary ingredient for decision-making. To paraphrase Sunstein, in a deliberative political process, even the most venal or self-interested participants must feel encouraged to invoke corporate justifications in their support, even if they do not actually mean it.  There will, of course, be competing versions of what is in the corporate interest, and different views of the criteria to be used in assessing it. As Buxbaum notes, there is a continuing tension between collective decisionmaking by egoists and collective decisionmaking by altruists.  It is one purpose of deliberative decision-making to encourage this sort of debate.
To conclude, the goal of deliberative decision-making is, in part, to improve the quality of corporate decision-making (by increasing the range of potential viewpoints which are presented), and in part to assist in the protection of personal interests (by improving the monitoring of decisions).  It can also have a broadly educative function, whereby members learn that their personal interests and those of the company are linked and that as a consequence they are corporate as well as private persons. 
Before proceeding to the third key feature of corporate constitutionalism, it is necessary to stress that the dual decision-making structure and the deliberative decision-making process are ideal descriptions. They establish a benchmark against which actual corporate practices may be evaluated. As ideal types, neither the structure nor the process are themselves guarantees against the problems of domination, oppression or unfairness in corporate affairs. There are, for example, many techniques of deliberation and, depending on factors such as the size and composition of the group and the nature of the decision, some will yield better quality decision-making than others. 
(iii) Separation of Powers
The third key feature of the corporate constitutionalist framework is the separation of powers. It, like the two previous features, directs our attention to issues of structure and process in matters of corporate governance.
Liberal legal and political theory demonstrates a deep-seated mistrust of concentrations of decision-making power within political communities. We require that these powers should be exercised non-arbitrarily and that this should be done in ways which are open and accountable. The idea of the separation of powers therefore seeks to achieve the dual aim of the diffusion and accountability of decision-making power. The exercise of decision-making power is to be broken up into different functions so that no single person or agency has full control over that power and so that there are in-built checks and balances on each powerexercising agency.
Lawyers and constitutional lawyers in particular have elevated this idea into a formal doctrine in which a holy trinity of formally distinct public powers is identified: legislative, executive, and judicial.  Modern Australian constitutional writing tends to concede that the first two of these powers have merged in practice, due mainly to the operations of the Cabinet system of government. Hence, in the constitutional law literature, discussion about the doctrine of separation of powers tends to be concerned primarily with the separation of the judicial power.
In the context of corporate constitutionalism I am not concerned with this formal doctrine. In fact I want to make a distinction between that doctrine and the underlying (and more general) idea of the separation of powers. That is, I want to preserve the idea that decision-making power should be diffuse and accountable, but I do not want to be tied to the three-fold classification which underpins the modern public law doctrine. Historically, the idea of separation of powers has developed with reference to public power. The concerns of the key writers in this history for example, Locke, Montesquieu, and Madison were directed at controlling the potential for State power to impinge unduly on the lives of individual citizens.  But as I noted earlier in this article, in late twentieth century society we find that significant sites of social, economic and political power are to be found in the private sector, especially in companies. The idea of a separation of powers must therefore be applied not just to the realm of public systems of government, but crucially to that of private, corporate systems.  In making this reformulation we must expect that corporate governance systems will likely have different taxonomies of power that will require different divisions than is found in public law doctrine.
Of course the idea of dual decision-making, described earlier, has already introduced a particular separation of powers into the corporate constitutionalist framework, distinguishing the normal path of board decisions from the higher path of the general meeting of members. The purpose of further examining the idea of separation is to suggest that these two sites of corporate decision-making power can be further sub-divided, and that they can be supplemented by other sites of power.
I will look at the separation of board power first. Perhaps the most thorough recent judicial examination of the role of the board in Australia is found in AWA Ltd v Daniels.  There, Rogers CJ Comm D set out a three way division of function which serves as a convenient way of organising this discussion. Each division refers to an aspect of the separation between monitoring power and management power. The first division is between the board and the company's senior managers. According to Rogers CJ Comm D:
A Boards functions, apart from statutory ones, are said to be usually four fold:
His Honour went on to note that in a large public company the dictates of business necessity mean that the board cannot be involved in day-to-day management of the company's business. Instead, the directors rely on management to manage the corporation, and thus the functions of management include the following:
(a) to carry out the day to day control of the corporation's business affairs;
(b) to establish proper internal controls, management information systems and accounting records;
(d) implement the policies and strategies adopted by the Board;
(f) prepare proposals and submission[s] for consideration by the Board;
(g) prepare a budget; ... 
There is ample debate in the corporate governance literature about whether this type of separation is workable, sufficient, or desirable.  I do not intend to enter that debate in this article; my purpose is simply to illustrate one obvious way in which corporate lawyers already understand the importance of the separation of corporate decision-making power. Despite the reference to business necessity, this separation of function between the board and management is more than just a pragmatic division of labour it also shows a concern for accountability and monitoring.
The second division of function recognised in the AWA case is that between the non-executive directors and the chief executive officer or managing director.  The wider literature suggests that this division has several dimensions. For example, in terms of a separation of powers, a simple distinction between executive and non-executive directors (NEDs) is not always useful. In many companies NEDs are unable to act as a check on executive management because of past or present ties with management.  Thus it is necessary to further distinguish between affiliated and non-affiliated NEDs, where affiliation is determined by reference to factors such as whether the NED is a substantial shareholder, recent employee, professional adviser, or significant supplier or customer, to or of the company.  Another dimension of this separation relates to the composition and structure of the board. Questions here include whether the board should be comprised entirely of non-affiliated directors, a majority of nonaffiliated directors, or whether a two-tiered structure, with separate management and supervisory boards, is appropriate.  Should board sub-committees be established to deal with potential conflict of interest issues such as executive and directors remuneration, the nomination and selection of board positions, and the auditing of the companys financial, commercial, and environmental performance?  If so, how are those sub-committees to be constituted (for example, wholly comprised of non-affiliated directors)?
Thirdly, in the AWA case Rogers CJ Comm D noted a division of function between the board chairperson and the other directors. In his Honours words, [t]he Chairman is responsible to a greater extent than any other director for the performance of the board as a whole and each member of it.  The call for chairpersons to be drawn from the ranks of the non-executive directors has become an article of corporate governance faith since the 1980s.
A further separation of power, also central in the AWA case, relies on the role of the external auditor of the company as an independent monitor of the company's activities.  This idea may not be as obvious as it first seems. On the one hand there is some lingering scepticism about the quality of auditing practices, in the wake of the 1980s.  In part these problems were due to the control which corporate managers were able to exert over auditors.  More recently there has been some indication that the balance of power between company management and auditors may be shifting. Firms of auditors are reportedly becoming more cautious about the corporate clients they take on, due to the number of large damages claims brought against auditors in recent years.  On the other hand, it can be argued that auditors do too good a job. Michael Power refers to the audit explosion to describe how the audit has become central to our ideas of organisational control. He warns that audits do not contribute automatically to organisational transparency.  Rather than being a vehicle for accountability audits often have more to do with giving credibility to the organisation, so that trust in the fact that an audit is done displaces public preoccupations with what is done and what is discovered.  None of these misgivings about the role of auditors negate their role as one part of a corporate separation of powers. Furthermore, it must be remembered that the separation of powers is only one part of the wider framework of corporate constitutionalism. Flaws in one aspect of that framework (such as poor audit practices) may be detected or remedied by another aspect (such as processes of deliberation at company meetings).
It is important not to be too categorical about these various separations.  As Vile notes, the theory of checks and balances which is part of the separation of powers means that each branch of power must exercise some control over the others by having a limited role in the exercise of the others functions. So, we should not insist on what Vile calls a pure separation of powers but, instead, allow for a partial separation.  That is, we should allow that some decisionmakers may have control over others by their partial participation in the others' functions. We should also allow that sometimes some of these aspects of corporate power might be exercised by agents other than those designated above. It is conceivable, for example, that in some companies and at some times directors might take some hand in the day-to-day management of the company. It is equally conceivable that, in the process of implementing board policies, managers may create further policies. This overlap is particularly apparent in the case of executive appointments to the board. Nor should we insist on finding the precise separations of power outlined above in each and every company. Clearly, different separations will be required in a small proprietary company than in a large public company (for example, not all companies are required to appoint an external auditor). The aim is to ensure a plurality of separations  that is appropriate to the company. In summary, a corporate separation of powers makes use of independent nonexecutive directors, independent chairpersons, external auditors, institutional shareholders and other related elements of corporate organisations. There is nothing new in highlighting the role of these corporate players indeed this is the bread-and-butter of the corporate governance literature. However there is something new, for company lawyers at least, about conceptualising these issues in terms of a separation of powers.
I have argued that corporate constitutionalism has three key features, all of which draw our attention to the structures and processes of corporate decision-making. The first feature emphasises that corporate decision-making has a dual structure. Decisions made by the directors and managers of the company represent the usual (or normal) path for corporate decision-making, while decisions at the general meeting have a higher status. Secondly, the idea of deliberative decision-making seeks to ensure that corporate decisions at either level are open and genuine and are based on deliberation over arguments rather than mere aggregation of votes. Thirdly, corporate constitutionality requires that there be a separation of powers relating to corporate decisions, with the dual aim of diffusing corporate decisionmaking power and at the same time rendering it accountable through a plurality of checks and balances.
Some aspects of these three features can be read fairly easily into the formal legal model of the company (for example, the separation of function between the board and the general meeting). Other aspects (such as the idea of deliberation) require us to make a more conscious reappraisal of that model in political terms. But I suggest that nothing in this framework is foreign to the basic legal structures within which corporate entities must operate. My purpose is to reconceptualise what is already there, rather than to argue for a reformation of the corporate form.
The idea behind this article is that we must, as Mary Stokes has urged, breathe new life into the legal model of the company and its regulation.  In tackling corporate governance issues, Australian courts and corporate regulators have placed too much faith in the instrumental capacity of substantive law reform, while giving insufficient attention to the assumptions which underlie existing rules and to the corporate structures and processes by which those rules are implemented. What is needed is a change in the way in which corporate actors perceive and interpret these legal rules. Some corporate law academics have begun to redress this problem, resorting to contract-based theories drawn from an economic framework.  I have argued that we need a complementary framework which looks at corporate law and corporate organisation through the lens of constitutionalism. The reader, having considered the argument for corporate constitutionalism, may still object that it is irrelevant or, even worse, inimical to the idea of companies as profit-maximising entities. They may be right at least, there are likely to be occasions when the ideas of dual decision-making, deliberation, and the separation of powers conflict with the immediate goals of profitability. But is this a reason to reject corporate constitutionalism as a framework for corporate governance? Only if those values are dismissed as inapplicable or peripheral to the corporate endeavour. In a world where companies threaten to displace individuals and governments as economic and political actors this hardly seems to be a wise move. We should not deny that most (but not all) companies are run for the purpose of returning profit to the corporators, and that many decisions made in general meetings and board rooms are concerned with that purpose. But the argument in this paper is that in making decisions about corporate profit there are issues and factors in addition to efficiency and wealth maximisation which should be considered. I am not arguing that corporate constitutionalism is a replacement or trump for the contractual paradigm. The purpose of the paper is to suggest a framework and a method by which political issues can be considered and weighed against other considerations.
Having said that, I acknowledge that I have used a broad brush. This article sets out only the first stage in the development of a corporate constitutionalist framework. As I have indicated throughout the article, the framework draws attention to issues of corporate structure and process. The next step is to examine the rules and assumptions which define these structures and processes. Nevertheless, whilst this article is only a beginning, I suggest that corporate constitutionalism presents us with significant opportunities. In particular it allows us to contemplate companies as governance systems that is, to take the idea of corporate governance seriously.