PowerPoint Assignment - Due 03/19

THE OWNERSHIP MODEL OF BUSINESS ETHICS

DAVID RODIN

Abstract: This essay attempts to develop a new theoretical model for business ethics distinct from the two canonical business-ethics theories, the stakeholder theory and the shareholder value theory. Milton Friedman argued that because managers are agents of the company’s owners, their sole moral responsibility is to maximize owner returns. Thomas Pogge has recently suggested that such a view involves a kind of moral incoherence and that we should reject the efficacy of social arrangements like the principal-agent relationship in altering moral obligations. Both views fail to give proper account of the dispersal of moral responsibilities in business contexts. We must distinguish ‘‘minimal moral obligations’’ (stemming from justice and rights) from ‘‘maximal moral obliga- tions’’ (stemming from all other moral considerations, including duties of aid, beneficence, and the virtues). Minimal obligations apply to all persons, but maximal obligations can be effected by social arrangements like the owner- manager relationship. There may be moral obligations incumbent on owners that do not apply to managers. Understanding this distribution of responsibilities enables us to develop a new and attractive model of business ethics––‘‘the ownership model’’––which places greater emphasis on the rights and responsi- bilities of the owners of business than has been traditional in business ethics.

Keywords: business ethics, corporate social responsibility, stakeholder theory, shareholder value theory, Milton Friedman, Thomas Pogge, corporate govern- ance, ethical investment, beneficence, rights, justice, virtue, responsibility, charity.

Private corporations are today among the most economically powerful institutional actors in the world. In any discussion of institutions and international justice it is therefore natural to consider the role and moral obligations of corporations, their managers, shareholders, and employ- ees. A significant consensus has developed in recent decades among many academics, NGOs, government agencies, and business leaders that private corporations have moral obligations to a set of ‘‘stakeholders’’ beyond their legal owners. These may include employees, customers, suppliers, the community, and even the environment and future generations. Numerous NGOs and some corporate leaders claim that corporations require a moral ‘‘license to operate’’ from the community, which entails being a good ‘‘corporate citizen.’’ A whole academic and advisory

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and 350 Main Street, Malden, MA 02148, USA METAPHILOSOPHY Vol. 36, Nos. 1/2, January 2005 0026-1068

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

industry devoted to ‘‘corporate social responsibility’’ has been estab- lished, as well as high-profile international initiatives, such as Kofi Annan’s Global Compact.

This emerging consensus, however, is fundamentally challenged by an argument expressed in its most pungent form by the Nobel Prize–winning economist Milton Friedman (Friedman 1999). According to Friedman the moral obligation of managers to stakeholder groups other than shareholders is extremely limited. He believes that it is generally not permissible for managers to forgo legally acquired profits for moral reasons. His argument is simple: managers are employed as agents of the owners of the corporation. Their legal and moral obligation is to manage the assets of the shareholders so as to maximize shareholder returns. If they manage assets so as to fulfill ‘‘social responsibilities’’ and thereby fail to maximize returns, they are wrongfully appropriating resources that do not belong to them. As Friedman says, managers ‘‘can do good––but only at their own expense’’ (1999, 252).

1

What is striking about this argument is that it suggests that the institutional relationship between managers and owners profoundly affects the nature and distribution of ordinary moral obligations. Man- agers are not required to doFindeed, they are required not to do–– certain things that would be morally obligatory if they were acting outside the institutional structure, or managing their own assets.

In an intriguing chapter of his book World Poverty and Human Rights, Thomas Pogge presents an argument that (although not presented in this context) can be read as a response to Friedman’s challenge (see Pogge 2002, ch. 3). His aim is to question the moral efficacy of social arrange- ments, such as the manager-owner relationship and the institution of nation- states, in altering the pattern of our moral obligations. Managers, suggests Pogge, cannot escape the normal claims morality makes on us by reference to the fact they are acting on behalf of the asset’s owners. Nor can owners escape such claims by appointing managerial agents.

In contrast to Pogge, I will argue that social arrangements like the owner-manager relationship can appropriately alter the distribution of moral obligations and responsibilities in business. But in contrast to Friedman, I will argue that these relationships do not alter the distribu- tion of moral obligations in such a way as to make ethical obligations irrelevant to corporations. On the contrary, I will sketch what I take to be a promising new model for grounding an account of corporate ethics–– the ownership model of business ethics. This model develops a middle

1 Elaine Sternberg puts the point even more strongly, claiming that managing for social

responsibility is tantamount to theft: ‘‘Managers who employ business funds for anything other than the legitimate business objective are simply embezzling: in using other people’s money for their own purposes, they are depriving owners of their property as surely as if they had dipped their hands into the till’’ (2000, 41).

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN164

way between the two canonical theories of business ethics, the stakeholder and shareholder value theories, and it has important advantages com- pared with these two theories.

Pogge’s argument is rooted in a form of structural critique of moral codes. He claims that a moral code can be shown to be incoherent and in prima facie need of revision if it contains a counterproductive loophole. By this he means roughly that the code provides incentives for an ideal adherent (one who is fully informed and committed to complying with the code) to act in ways that defeat the overall purpose of the code. If a code incentivizes behavior that is regrettable, all things considered, by the lights of the code itself, then other things being equal the code ought to be revised.

One of the examples Pogge uses to illustrate this claim is the agency model of business management. He asks us to imagine that our father owns an apartment block inhabited by elderly tenants who have lived there for many decades and enjoy a strong sense of community. An opportunity arises to increase financial returns on the building by evicting the tenants and converting the apartments into luxury flats. Pogge supposes that the prevalent ethical code endorses the following judg- ments: (1) it is impermissible for the father acting on his own behalf to undertake the conversion; (2) it is permissible for a lawyer appointed by the father to maximize returns on the building to undertake the conver- sion; and (3) it is permissible for the father to appoint the lawyer on these terms.

2

But Pogge argues that a code endorsing this set of judgments is incoherent in that it contains an inherently regrettable loophole. For the father is able to avoid an important moral obligation by effecting what amounts to a merely cosmetic change in social arrangements–– having the lawyer administer the conversion rather than doing it himself. Pogge supposes the conversion to be intrinsically regrettable by the code’s own lights, yet the code provides an ideal incentive to effect this outcome by providing means to do so without color of wrongdoing. Broadening the argument, Pogge concludes: ‘‘We should avoid inherently regrettable ideal incentives toward creating or joining social arrangements by denying altogether the moral significance of social arrangements in matters of common decency and basic justice’’ (2002, 87).

2 In fact Pogge stipulates that this is true of the hypothetical code he presupposes in his

example. But it is clear that he takes this code to be substantially coextensive with the ethical code that is actually prevalent in society, for as he says later in the chapter, he believes his argument has the effect of upsetting ‘‘received moral convictions’’ (2002, 89). I therefore read the argument as engaging with the prevalent ethical code as it actually exists in our society. It should also be noted that ‘‘ethical code’’ here refers to a set of generally accepted ethical norms, but these need not be codified in any formal way.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 165

Against this I will argue that the prevalent moral code governing business activities is a good deal more resilient than this. It does not contain inherently regrettable loopholes in the sense Pogge suggests. There is a coherent code of business ethics that does recognize the significance of social arrangements in apportioning moral responsibility, though spelling out what this amounts to will require a careful analysis of interacting moral responsibilities and will also require a call for some reform of corporate institutional arrangements.

Pogge considers what he sees as ‘‘a weak spot in my entire approach, because, quite generally, the claim to have found a loophole can always be countered by claiming that the relevant ideal incentive is not in fact regrettable on the whole’’ (2002, 79). In other words, though it may be regrettable in some respects that there exists an ideal incentive to hire the lawyer to evict the tenants, there may be some countervailing value embodied in the code that means it is not regrettable, all things considered, for the code to provide this incentive. He notes, rightly, that the key question about such an objection is its plausibility in light of the rationale of the code as a whole. He considers and rejects out of hand a proposal that the code is not regrettable, all things considered, because of the moral importance of hiring lawyers.

But Pogge fails to consider a much more plausible rationale, one that does in fact underlie many aspects of the moral code of private business. This is the moral importance of the benefits that accrue to society as a whole from the practice of managing assets so as to maximize returns. Such a practice helps to achieve the most economically efficient distribu- tion of resources, which benefits society as a whole. For example, in the case of the apartment block, maximizing returns by converting the building releases underlying capital value that may then be used for further investment. This in turn generates additional employment and tax revenue that may be used for health care, education, and other public goods. It is this societal benefit that is the rationale underlying the permission for the lawyer to undertake the conversion, and it provides a strong prima facie reason for believing that the incentive within the code to appoint a lawyer to evict the tenants is not regrettable, all things considered.

I take this observation to be part of a broader account, consequenti- alist in spirit, that underlies the moral rationale for the institution of competitive capitalism as a whole. Of course it could be that such a rationale is based on faulty economic analysis. Alternatively, it may be that the societal benefits in this particular case are not sufficient to outweigh the aggregate harm to evicted tenants, or that tenants have rights against being harmed in certain ways. If this were the case, then the terms of the prevalent code of business would require alteration. But the reasons for the alteration would be internal to the justificatory dynamic of the code itself and would not stem from the existence of

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN166

any kind of incoherence evidenced by the existence of a loophole in Pogge’s sense.

But Pogge’s argument is capable of a deeper formulation. Even if it is not regrettable, all things considered, for the moral code to generate an ideal incentive to evict the tenants, it may still be regrettable, all things considered, actually to evict the tenants. If it is true that performing such an act is impermissible for the owner acting alone but permissible once the owner has appointed an agent, then there is a sense in which moral responsibility has been lost or evaded simply by putting in place a social arrangement that appears to be morally unimportant in itself. Any code that permitted this would clearly be defective.

But I do not believe that the prevalent moral code does allow the evasion of responsibility in this way. To see this, let us first draw an intuitive distinction between what we might call ‘‘minimal’’ and ‘‘max- imal’’ moral obligations.

Minimal moral obligations are claims others have on us stemming from human rights and justice. Minimal obligations tend to be the most stringent form of moral obligation, in that they delineate the minimal standards for human decency––thus the term minimal obligations. If I violate a minimal moral obligation owed to you, then I wrong you (whether or not I harm you). Because of this, the violation of minimal moral obligations provides a ground for complaint by the object of the obligation, and for claims of compensation and redress. Moreover, it can often be appropriate to enforce compliance with minimal obligations through punishment or other forms of sanction.

Maximal moral obligations are those stemming from all other moral considerations, including positive duties of aid, charity, beneficence, commitments to moral ideals, and the virtues. Such considerations provide us with moral reasons for action, but these reasons are in some respects less stringent than those deriving from human rights and justice. Typically, maximal moral obligations are not enforced by punishment, and noncompliance does not give rise to valid claims for compensation. This is because someone who suffers as a result of the violation of a maximal obligation is harmed but is typically not wronged. In certain respects maximal obligations may even have a discretionary element. For example, an agent may, with equal moral honor, choose to structure his life around varying sets of virtues that will make correspondingly different moral claims upon him. Consider, for example, the differing moral demands incumbent on a man who has chosen the life of a dedicated teacher, diligent researcher, and committed family man com- pared with those incumbent on a man who has dedicated his life to delivering medical services in dangerous and war-torn parts of the globe.

Whether an obligation is maximal or minimal can depend on facts about both the subject and the object of the obligation. For instance, providing assistance to a given person may be a maximal obligation for

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 167

you but a minimal obligation for me, if I have made a promise or entered into a contract to assist, or if I am a close family member. On the other hand, human rights, such as the right not to be attacked or unjustifiably deprived of my property, impose minimal obligations on all people.

The distinction between minimal and maximal moral obligations will obviously require a good deal of further clarification and explanation before it can function in a full-fledged moral theory. For example, might it not be the case that some duties of aid are also obligations of justice–– for example, those that arise in situations of extreme need or rescue?

3

My intention here is not to provide such a full analytical clarification but rather to investigate how institutional arrangements affect the distribution and structure of these two classes of moral obligation, intuitively and indicatively understood. Such an investigation has the potential to be valuable however one chooses fully and finally to specify the distinction between minimal and maximal obligations, and it may even provide useful constraints that will assist in the process of drawing the distinction itself.

It is of the greatest importance for the proper interpretation of business ethics cases, such as that of the apartment block, to determine whether we are dealing with minimal or maximal moral obligations. Pogge supposes that the obligation not to evict the tenants is a minimal moral obligation. ‘‘The tenants,’’ he writes, ‘‘are morally entitled to the preservation of their community’’ (2002, 86). But this interpretation is doubtful. It is hard to see how the tenants could have a claim right against the owner not to evict them. What could the source of such a claim be? It is true that the tenants would be harmed by the eviction. But the question, so far as minimal moral obligations are concerned, is not whether the tenants are harmed but whether they are unjustifiably harmed, in other words wronged, by the eviction (consider how local tradesmen and shopkeepers who stand to gain from the conversion would be harmed by the abandonment of the conversion plans). The harm inflicted on the tenants might be unjustified if it violates an explicit or implicit contract, or if it violates an accepted norm (for example, one that judges the eviction of tenants to be unjustly disproportionate to the social benefits achieved from allowing owners to maximize returns on their properties). But there is no reason to suppose that such factors are at work in this case, because if they were then it would be equally impermissible for both the lawyer and the owner to undertake the conversion.

Generalizing this point, it would seem that Pogge’s claim that the ethical code permits the lawyer to evict the tenants is only plausible on the assumption that the obligation to not evict is construed as a maximal, not a minimal moral obligation. Consider how, according to the prevalent moral code, lawyers and managers are not entitled to promote their

3 We will return to this question below.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN168

client’s interests through assault, murder, theft, fraud, enslavement, or other actions that violate clear minimal moral obligations. It would seem to be a general moral principle that obligations that genuinely form part of the moral minimum cannot be evaded by reference to the existence of an agency relationship. Minimal moral obligations cannot even be evaded by the existence of highly coercive command structures such as exist in military organizations (thus the generally accepted claim that ‘‘following orders’’ is not a sufficient exculpation for most serious moral transgres- sions).

Let us take this as our first conclusion about the prevalent code of business: minimal moral obligations apply equally and in all circum- stances to all persons and cannot be evaded or substantially altered by social or institutional arrangements, such as the owner-manager relation- ship. So far as the moral minimum is concerned there is no diminution of responsibility entailed by the owner-manager relationship, for both owners and managers are bound by those requirements.

If the obligation not to evict the tenants is not a minimal moral obligation, then it must be a form of maximal obligation. Indeed, it would seem most plausible to construe it as a maximal moral obligation stemming from considerations of beneficence and virtue. Thus, though the owner would not violate the tenants’ rights or relevant principles of justice by evicting them, it is something he ought not to do, all things considered, because to do so would violate the requirements of compas- sion, generosity, and beneficence. This observation is itself significant, for it shows that though maximal obligations are in a sense less stringent than minimal obligations they are still very significant. Indeed, it may some- times be true––as it seems to be in this case––that a maximal moral obligation (generosity or charity) can override a minimal moral liberty (the fact that the owner has no minimal moral obligation not to evict the tenants).

Suppose I am correct to characterize the duty not to evict the tenants as a maximal moral obligation. Why shouldn’t these obligations be incumbent in precisely the same way on the lawyer acting as a manager? Why should it be permissible for the manager to evict the tenants but not for the owner to do so (as the prevalent code apparently decrees)? There are two ways we may approach this question. The first is by examining constraints on the fulfillment of maximal obligations implicit within the nature of these obligations themselves. The second is by examining the moral relationship that exists between owners and managers.

Regarding the first of these considerations, it would seem to be a general feature of maximal moral obligations that they can only be fulfilled with one’s own resources, not at the expense of others.

4 Obliga-

4 This is one of the intuitive principles invoked by Milton Friedman, though he does not

distinguish carefully between maximal and minimal moral obligations (1999, 252).

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 169

tions of charity, beneficence, generosity, and other virtues can only be properly fulfilled using resources that really are yours to dispose of. Indeed, it is hard to even make sense of what it could mean, in a moral sense, to be generous or beneficent with someone else’s money. If I am an officer of a charitable organization, I might be generous in disbursing the resources of the organization; but the generosity here is the organiza- tion’s, not mine. I am being generous on behalf of the organization, and this will only be morally laudable if I am properly mandated to use the resources in this way.

This constraint, implicit within the structure of maximal obligations themselves, seems to explain why such obligations operate differentially on owners and managers in business-ethics cases. For a manager admin- isters property that does not belong to him, and he is entrusted with generating profits to which he is not entitled. In general it is inappropriate for mangers to use assets and profits that are not their own in order to fulfill maximal moral obligations incumbent on them (we may call this the ‘‘no Robin Hood-ing’’ requirement). This would appear to be the primary explanation for why it is impermissible for the owner to evict the tenants but not impermissible for the lawyer acting as manager to evict them.

One might object to this line of thought by considering the following kind of case: imagine that a company manager can save a person’s life through some very small encroachment on the company’s resources, letting a person use the company car to drive to hospital, for instance. It seems clear that the manager has an obligation to render assistance in such circumstances even though this involves utilizing resources that are not his own. Yet one might think that this obligation, being an obligation of assistance, is maximal rather than minimal.

As I indicated above, obligations of assistance in emergency situations occupy a difficult gray area between obviously minimal and obviously maximal obligations. One might think that rendering significant assis- tance to someone when the costs are very low is in fact a minimal obligation, such that someone who fails to render assistance in these circumstances wrongs the person in need, generating in turn a claim for compensation, redress, or enforcement.

5 On the other hand, one might

hold that such duties of assistance are maximal, albeit of a particularly strong and stringent kind. In this case one will need to provide a more complex analysis of the case. The most natural interpretation is that company owners implicitly authorize managers to use company resources to render assistance in such emergency cases. In which case, to the extent that the act of assistance or generosity involves the use of resources that belong to owners, it is properly described as the owner’s act of assistance or generosity, and it is the owners who have fulfilled the maximal

5 Spain, France, Germany, Greece, and several other jurisdictions impose some form of

legal duty to assist.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN170

obligation. Suppose, however, that the owners do not permit their resources to be used in this way. We still believe that the manager morally ought to render assistance. In this case we will say that the manager has justifiably infringed the owners’ right not to have their resources appro- priated without their consent. If the owners are really so mean as to demand compensation, the manager would be morally obliged to pay it. In this case the manager genuinely has fulfilled the maximal obligation himself, but because he has assumed a moral liability to pay back the resources used or render compensation we can also say that he has fulfilled the obligation using his own resources.

6

The second consideration (which has often been emphasized in discussions of business ethics) refers to the set of obligations implicit within the agent-principal relationship. These stem from the fact that an agent, such as a manager, has contracted or promised to administer the principal’s affairs in accordance with his wishes, and more generally to act in the principal’s best interests. One might be tempted to reason on the basis of such considerations as follows: contractual and promissory obligations are a component of justice and as such they form part of the moral minimum. But because minimal obligations are more stringent than maximal obligations, they trump them in cases of conflict, and it is this fact that explains why the lawyer is not simply morally permitted, but indeed morally required, to evict the tenants.

But we must be very careful with this line of thought. Even if we accept that contractual and promissory obligations are components of justice and therefore part of the moral minimum, clearly not all minimal obligations are of equal strength or importance. As we have already seen, minimal obligations stemming from human rights consistently override whatever obligations one may have as an agent to maximize the interests of the principal. This would only be explicable if some forms of minimal obligation have an importance that can override contractual and promissory obligations.

7 More important, it is not clear that minimal

obligations do always override maximal obligations. We have already

6 This dialectic has obvious relevance to the question of how managers and owners of

corporations ought to respond to radical poverty in markets in which they operate, or how pharmaceutical companies ought to respond to medical emergencies in developing countries. Much will turn on how low the cost of assistance must be before the obligation becomes overriding, and whether directness or proximity to those in need makes a moral difference.

7 There are two ways one might conceive of the relationship between contractual and

promissory obligations and human rights. One may think of human rights as overriding the contractual and promissory obligations stemming from the agency relationship. Alterna- tively, one might think of human rights as placing limits on the nature of the obligations one is empowered to acquire through the process of contracting and promising. For example, it may be that human rights and justice function as an implicit backdrop and constraint to all contractual and promissory relationships (including the agent-principal relationship), such that one simply is not empowered to acquire a valid obligation that contradicts basic human rights and justice.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 171

seen how maximal obligations can in certain circumstances override minimal Hohfeldian liberties. It may be the case that certain maximal obligations are so important that they can override even some weak forms of minimal obligation.

8

So the simple argument that casts contractual obligations as minimal obligations and sees them as overriding all maximal obligations is insufficient in itself. It would seem, in contrast, that these two forms of consideration are best seen as functioning in concert. The fact that maximal obligations can only appropriately be fulfilled with one’s own resources is sufficient to establish a moral liberty for managers to act in ways that would ordinarily be ruled out by maximal considerations. The existence of contractual and promissory obligations (though not necessa- rily sufficient in themselves to defeat maximal obligations genuinely relevant to an agent) functions to establish a moral presumption in favor of preferencing the interests of the principal over those of others, where there is a liberty for the agent to do so.

It is worth pointing out that this account does not imply that maximal moral considerations have no role to play in the moral deliberations of managers acting as agents. It is rather that the role played by maximal considerations is modified by the existence of the agency relationship. For example, in the apartment-block case, if we accept that evicting the tenants genuinely contradicts important maximal obligations by being ungenerous and cruel, then the lawyer thereby assumes the obligation to bring this to the attention of the owner and to lobby the owner not to permit the eviction. At the limit, a manager may be obligated to resign if it is clear that he is being asked to participate in an activity that is morally objectionable for maximal reasons. What a manager cannot do, on this account, is utilize the resources of the business, or the owners of the business, to fulfill maximal obligations as if they were his own. This account would seem to explain why the lawyer acting as an agent is not only permitted but also morally required to evict the tenants, even though it would be impermissible for the owner to do so on his own.

But now it would seem that we are once again saddled with the problem that Pogge identified at the outset. If it is impermissible for maximal moral reasons for the owner to evict the tenants but it is not impermissible for the lawyer to evict the tenants, then it would seem that

8 Warren Quinn gives the example of a lifeguard working for a private client. Quinn is

inclined to suppose that if the lifeguard has to choose between saving his client or five drowning strangers, he should save his client and allow the five strangers to drown (1989, 293). But if we change the example so that the contractual obligation is trivial––buying the client an ice cream––then most would feel that it is permissible to infringe the contractual obligation to save the five, even if one believes the contractual obligation to be morally minimal (implying that the client has been wronged and requiring, for example, the lifeguard to provide compensation or apology to the client for financial loss suffered as a result of his justified infringement).

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN172

the owner could evade his maximal moral obligations without color of wrongdoing by simply appointing the lawyer. It would seem that this involves a regrettable loss of moral responsibility, especially given the moral importance of the maximal obligations concerned.

In reality, however, there is no loss of moral responsibility. To think otherwise is to misunderstand the nature of the relationship between owners and managers, or more generally the nature of the principal-agent relationship. The mistake is to think that because it is permissible for the lawyer to evict the tenants, it follows that it is permissible for the owner to appoint the lawyer on such terms that he would be permitted or obligated to do so.

On the contrary, the principal is responsible for setting morally appropriate terms under which an agent will manage his affairs. Thus if an owner is subject to a moral obligation pertaining to his asset, then he has a further obligation to ensure that any agent managing this asset does so in a way that is consistent with that moral obligation. Moreover, the principal will typically retain ultimate supervisory and executive control over the agent’s decisions and actions. Thus we often say of the agency relationship that the agent acts on behalf of the principal and the principal acts through the agent. Maximal obligations are therefore not lost in the principal-agent relationship, because they are fully retained by the principal.

9

The appropriate image for understanding the nature of a principal’s responsibility is ownership of a machine. Owning a commercial asset, such as a business, is like owning a special kind of machine whose function is to generate outputs of greater value that its inputs. Like any machine, a business brings with it a collection of rights and responsi- bilities for the owner. Rights clearly include first call on the profits generated. But responsibilities come with those rights, and they include the obligation to ensure that the business machine does not cause unjustified harm to others, and that its products are utilized in accord with the requirements of both the moral minimum and the moral maximum. To take a simple example, owning a car brings with it the right to keep the profits generated by its commercial use, but it may also bring with it important obligations and responsibilities. These will include the obligation not to use the car in a way that endangers others, and to

9 This retention of responsibility on the part of the principal is exceptionally robust. It

may even exist when the agent performs acts without specific instruction by the principal. Thus, if I have a troublesome colleague in the department and I hire a thug known to have committed contract killings in the past and tell him to ‘‘just deal with the problem,’’ then I will be guilty in both law and morality if he proceeds to kill the colleague. There are, obviously, limits to the responsibility of principals. If an agent acts in bad faith and without instruction, and moreover the principal has taken sufficiently stringent measures to oversee and supervise the agent’s activities, then the principal may have no responsibility at all for the actions of the agent.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 173

take reasonable steps to ensure that others do not use it dangerously. Ownership may also generate maximal obligations. For example, if my car is the only one in town, then I may acquire an obligation that I would not otherwise possess to drive a sick child to hospital. Ownership of a business generates moral rights and responsibilities in precisely the same way.

Thus Pogge’s example of the apartment block, designed to demon- strate the incoherence of the prevalent moral code, has instead enabled us to elucidate a robust account of that code which accords with our pretheoretical intuitions. It is an account in which moral responsibilities are indeed sensitive to institutional arrangements, such as the agency relationship, but where neither minimal nor maximal obligations are lost or diminished.

10 We might summarize these results as shown in table 1.

This analysis may be viewed as an elementary model of the distribution of moral responsibility in business ethics. Call it the ownership model of business ethics. Note how the model occupies a plausible middle ground between the two dominant theories of business ethics, the shareholder value theory and the stakeholder theory. According to the shareholder value theory, owners and managers are subject to a thin set of minimal moral obligations, but maximal obligations (the bottom line of the matrix) are almost entirely omitted. At best maximal obligations are

limited

Minimal

obligations

Maximal

obligations

ManagersOwners

full full

full

TABLE 1. The Ownership Model of Business Ethics

10 Of course, much of what is most interesting in Pogge’s book as a whole is that he

believes that the moral maximum is much more minimal than we have tended to suppose. Many obligations that have been thought of as more or less discretionary positive duties to assist others are in fact much more stringent moral requirements rooted in human rights and justice. But this argument must stand on its own merits––it can receive no support from the loophole argument. If we are persuaded by Pogge’s account of the moral minimum and come to see the obligation not to evict the tenants in these terms, then we will ipso facto have reason to believe that the lawyer acting as agent is not morally permitted to undertake the conversion.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN174

viewed as discretionary commitments that owners of a business may use company profits or dividends to achieve, but that play no appropriate moral role in the management or governance activities of the company itself.

11

In contrast, stakeholder theory gives prominence to maximal obliga- tions, but its focus is almost exclusively on the obligations of managers rather than owners (the right-hand column of the matrix). According to stakeholder theory, managers are morally required to utilize the resources of the firm to serve the interests of all stakeholder groups, of which the owners are but one (see Evan and Freeman 1999). But viewing business ethics in this way has several counterintuitive consequences. The first is that this view ignores the special rights that owners have to the profits generated by the firm and the special responsibilities that accompany these rights. I have argued that both the rights and the responsibilities are implicit in the moral nature of ownership. The second is that it ignores the constraints on the maximal moral action of managers––the fact that it is morally inappropriate for managers to fulfill maximal obligations with resources that are not their own.

The ownership model suffers from none of these shortcomings. It accounts for our pretheoretical intuitions about the distribution of moral responsibility in institutional business settings, and does so without any regrettable loss of moral responsibility. It would seem to be a robust and promising model for the development of business ethics.

It might be objected, however, that this model has been developed in the context of small businesses where the relationship between owners and mangers is a reasonably direct one. Suppose we wanted to generalize this model into a general theory of business ethics appropriate also to large publicly listed corporations? Would the ownership model still be appropriate?

One difficulty stems from the insertion of the legal corporation as an intermediary between the shareholders and managers of a business. In law a corporation is conceived as an artificial person that is itself the owner of the assets of the business. What shareholders own is a legal entitlement to a share of the corporation’s profits. Because of this it is an open question whether the relationship between managers and shareholders is an agent- principal relationship at all, for on some accounts it is the corporation itself that is the principal.

Notice, however, that my argument for the limited nature of managers’ maximal responsibilities does not depend on their agency relationship

11 Maximal obligations may play a purely instrumental role in management decisions

according to shareholder value theory. Thus, if undertaking activities perceived in society as ‘‘morally good’’ brings net financial benefits to the firm––for example, by generating good will with customers––then managers may appropriately undertake them. But it is clear that such considerations can play no recognizably moral role in management deliberation on such an account.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 175

being with the owners or shareholders. If the principal of the agency relationship is the corporation itself, this will not alter the fact that it is morally inappropriate for managers to fulfill maximal obligations with resources that are not their own. This suggests that the much-discussed question of whether shareholders and managers are really in an agent- principal relationship is not in fact the crucial issue so far as this question is concerned.

A more serious difficulty concerns the status of the shareholders themselves. How appropriate is the ownership model to a large modern corporation where shareholding is highly distributed and the ability of individual shareholders to affect the activities of the company is extremely minimal? Imagine, for example, that instead of owning an apartment block outright, the father owns shares in a large property company that is proposing to convert one of its buildings and evict the elderly community. The father may feel that there are moral reasons prohibiting the owners of the company from undertaking such action, yet he may recognize that as a small shareholder he has no ability to change the company’s policy. Reflecting on the principle that ‘‘ought implies can,’’ he may conclude that a shareholder in his position has no responsibility for the eviction. The situation is even worse for the large number of shareholdings held in mutual funds or other investment vehicles, where shareholding is transi- tory and the ultimate owners may not even know the content of their portfolio at a given time.

Here it would seem we are genuinely faced with a dilution of moral responsibility as a result of institutional arrangements. For the share- holders of modern publicly listed corporations have no effective way of communicating their moral preferences to managers and ensuring that they are effectively implemented, whereas managers, as we have seen, are prohibited from exercising maximal moral obligations using the resources of the corporation without the explicit consent of the owners. It may appear that this is intrinsically morally regrettable, and indeed calls into question the moral viability of the institution of the large public corporation.

Now Friedman argues that there is no diminution of responsibility implicit in the modern corporation because shareholders can always use their company dividends to fulfill individual maximal obligations inde- pendently of the activities of the corporation. But this claim ignores the fact that there are certain maximal moral obligations that only arise, or arise in different forms, in the corporate context and cannot be appro- priately fulfilled by individuals acting alone. For example, there may be certain forms of assistance that corporations are uniquely positioned to provide (think of a drug company holding a patent on AIDS medication), and there are some goods that Microsoft can achieve as a corporation that even an immensely powerful and wealthy shareholder like Bill Gates cannot achieve as an individual. Similarly, corporate activity may create

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN176

harms that cannot be properly compensated or redressed by shareholders acting as individuals.

If one is persuaded by the moral analysis of the ownership model, then it would seem that the tendency of the large modern corporation to erode maximal moral obligations in this way is one of the great unresolved problems of business ethics. I would like to conclude by indicating three potential responses to this problem.

The first seeks to deny that there really is a problem. It takes as its starting point the observation that the social institution of the large publicly listed corporation can hardly be described as a trivial or cosmetic change from the model of the small business owned by an individual or a family, envisioned in our example of the apartment block. The differences are profound, and many have significant moral implications because of the potential benefits they can bring to society. For example, today’s giant corporations can reap huge economies of scale that lower the cost of goods for consumers and create wealth for shareholders. Moreover, there are some products that could literally only have been created by organizations of this scale and structure (think, for example, of the Boeing 747 or an Intel microprocessor). Now it is true that the scale of private corporations can be replicated by state-owned enterprises, but their desirable dispersal of risk over private individuals and access to the deep capital resources of equity markets cannot be so easily replicated. These are all significant goods that there are moral as well as prudential reasons to bring about.

Accordingly, one may attempt to argue that insofar as the institution of the corporation entails the diminution of certain maximal obligations, such as the personal virtues of kindness, compassion, and generosity, this is not morally regrettable, all things considered. For these losses are compensated by important social gains.

12 Indeed, one might suppose that

the loss, insofar as there is one, is not regrettable at all, for it simply represents the transition to a set of institutional virtues more appropriate to the economic sphere: efficiency, rationality, and dedication to max- imizing returns.

It must be doubted, however, how far this line of thought can take us. It is difficult to regard an institutional arrangement that replaces the virtues of kindness and generosity with those of simple profit maximiza- tion as not entailing some profound form of moral loss, even if one accepts that it may lead to utilitarian welfare gains for society as a whole.

A second strategy would be to attempt to claw back some of the manager’s lost competence to act in accordance with maximal moral obligations in large publicly listed corporations. One potential way to do

12 Of course, any such account would have to include a firm insistence on the minimal

obligations of managers and shareholders, in order to prevent the enormous power of large modern corporations being used to inflict unjustified harm.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 177

this would be by permitting, or perhaps even requiring, managers to use what we might call a ‘‘principle of interpretative charity’’ when managing the assets of shareholders. This would specify that in the absence of a reliable way for shareholders to communicate their moral preferences to managers, managers are entitled to administer the corporation in accor- dance with the maximal moral commitments that they can reasonably and honestly attribute to their shareholders. Such an approach is consistent with the call made by many NGOs for managers to use the resources of their corporations to do good irrespective of whether they have a clear mandate from shareholders to do so.

But it must be recognized that there are substantial dangers implicit in this proposal. For corporations are vastly powerful institutions, yet the managers controlling them are largely trained in technical matters and are, as a group, highly conservative: they are still overwhelmingly white, male, and upper-middle class. If they were given a mandate to use the resources of corporations to achieve the maximal moral goods that they may reasonably and honestly attribute to their shareholders, they would be exposed to significant temptations to use those resources to fulfill their own conception of the maximal moral good. Given the narrow social class represented by managers, those conceptions may be idiosyncratic, distasteful, or even dangerous.

13 Managers would require extraordinary

moral imagination and integrity to keep their own conceptions of maximal obligations distinct from those reasonably attributable to their shareholders. Even granting such qualities, they would certainly require stringent oversight and accountability in exercising these powers. It is difficult to see who would be better suited to provide this oversight than the shareholders themselves or those properly authorized to perform the function on their behalf, namely, the board of directors.

This leads naturally to the final response, which is less conceptually innovative than the last, but which instead invites a program of practical institutional reform to corporate governance mechanisms. This approach insists that responsibility for maximal moral obligations rests inalienably with the owners of business assets. The problem is not with the ownership model of business ethics but rather with the governance structures of corporations themselves. The basic goal of such structures should be to align ownership and control, so that managers are accountable to owners not only for the financial performance of the corporation but also for the minimal and maximal moral character of its activities.

13 Of course, something of the same problem arises with all private philanthropy. But

private philanthropists are at least constrained by the fact that they must utilize their own resources for their philanthropic schemes. Moreover, the exercise of maximal obligations by shareholders urged by the ownership model avoids this problem to a significant degree. Because pension funds and mutual funds have replaced wealthy individuals as the most important owners of equity in most markets, the social profile of the ultimate shareholders of large corporations is now much closer to that of society as a whole.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN178

How might this be achieved? There are two broad mechanisms through which shareholders can affect the policies of corporations. The first is through their election of the board of directors; the second is through the buying and selling of shares, which determines market capitalization and indirectly the cost of equity capital, two critical determinants of corporate success. Both mechanisms need to be strengthened and reformed if the moral vision of the ownership model is to be given effect.

There is a vast literature on both topics, and I can only make some very indicative comments here. It is interesting to note, however, that reform of corporate governance is already high on the agenda of shareholders, managers, and government regulators. So the ownership model of business ethics provides further normative motivation for a movement that is already substantially under way. Since the birth of the shareholder activism movement in the 1990s and the corporate scandals of the early years of the new century, numerous corporate governance initiatives have been proposed to protect the interests of shareholders by increasing their control over management behavior. Many of these same reforms may simultaneously be used to empower shareholders to communicate their moral preferences to management and oversee their implementation. But in general the moral dimension of corporate-governance reform needs to be more explicitly recognized. There need to be explicit mechanisms for moral policies to be debated by the board of governors and proposed and voted on by shareholders. Shareholders themselves must become more aware of the moral responsibilities of ownership and be more prepared to become proactively involved in corporate-governance activities. Already many NGOs lobby shareholders directly in an attempt to change a corporation’s policy.

But given the nature of modern shareholding, active and effective participation in corporate-governance activities will never be a realistic possibility for most small investors. For this reason the various ethical- investment initiatives are of the greatest importance. The ethical-invest- ment industry is still in its infancy; it is small and highly fragmented, and many of its products and indices are ethically crude. But it is also the fastest-growing segment of the mutual-fund industry in many markets. If the ethical-investment movement were ever to reach sufficient scale and appropriate structure, it would have the potential to be a powerful mechanism for shareholders to affect the moral activities of corporations by raising the cost of capital for corporations that do not conduct themselves in accordance with the moral preferences of shareholders.

The chief requirement for this to occur (beyond the support of the share buying public) is the existence of accessible, standardized, and impartial data on the ethical aspects of corporate activity. At the moment numerous different ethical standards, investment criteria, and indices compete for investor attention. Most of these are based on a particular conception of what constitutes ‘‘ethical behavior’’ on the part of corpora-

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 179

tions. Yet it is unlikely that any of these particular standards will ever gain universal acceptance among investors because people differ in their moral commitments and, as we have already seen, certain differences in the conception of the maximal moral obligations are entirely appropriate.

What seems to be required is the existence of a third-party rating agency that can assess the performance of corporations along a number of different ethical criteria. It could measure such factors as support of the disadvan- taged, treatment of employees, the manufacture of dangerous products, and so forth. The criteria themselves could be specified at a very fine-grained level involving hundreds of criteria and subcriteria. The data pertaining to these criteria would need to have the following features. First, the data would need to be produced in such a way that shareholders are able to weight the different criteria according to their own moral preferences and commitments to generate a personalized moral profile for their investment activities. For example, one investor may be morally opposed to the manufacture of weapons but less concerned about greenhouse-gas emis- sions, whereas another investor may have the opposite concerns. Second, the data must be capable of being aggregated across a portfolio of shares or a managed fund, so that the portfolio or fund as a whole can be assessed according to an investor’s personal moral profile. Finally, the data must be standardized and easy to use, and the providers must be universally recognized as legitimate and trustworthy. In other words, what is needed is an analogue in the arena of ethical assessment to the standard providers of credit-risk assessment, Moody’s and Standard & Poor’s.

None of these ethical reforms will be quickly or easily achieved, and their ultimate effectiveness is not guaranteed. But what is clear is that appropriate reform will not be achieved without the right theoretical account of the distribution of responsibilities in business ethics––in what ways managers and owners ought to respond to maximal and minimal moral considerations. On the argument of this essay, it would appear that the ownership model of business ethics is a strong contender for providing such an account.

Centre for Applied Philosophy and Public Ethics CPO Box 8260 Australian National University Canberra ACT 2601 Australia [email protected]

References

Evan, William, and Edward Freeman. 1999. ‘‘A Stakeholder Theory of the Modern Corporation: Kantian Capitalism.’’ In An Introduction to

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

DAVID RODIN180

Business Ethics, edited by George Chryssides and John Kaler, 254–66. London: Thomson Business Press.

Friedman, Milton. 1999. ‘‘The Social Responsibility of Business Is to Increase Its Profits.’’ In An Introduction to Business Ethics, edited by George Chryssides and John Kaler, 249–54. London: Thomson Busi- ness Press. First published in 1970.

Pogge, Thomas. 2002. World Poverty and Human Rights. Cambridge, Mass.: Polity.

Quinn, Warren. 1989. ‘‘Actions, Intentions, and Consequences: The Doctrine of Doing and Allowing.’’ Philosophical Review 98, no. 3: 287–312.

Sternberg, Elaine. 2000. Just Business: Business Ethics in Action. Second edition. Oxford: Oxford University Press.

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2005

THE OWNERSHIP MODEL OF BUSINESS ETHICS 181